Osteoporosis
Methods in Molecular Biology John M. Walker, Series Editor 457. Membrane Trafficking, edited by Ales Vancura, 2008 456. Adipose Tissue Protocols, Second Edition, edited by Kaiping Yang, 2008 455. Osteoporosis, edited by Jennifer J. Westendorf, 2008 454. SARS- and Other Coronaviruses: Laboratory Protocols, edited by Dave Cavanagh, 2008 453. Bioinformatics, Volume 2: Structure, Function, and Applications, edited by Jonathan M. Keith, 2008 452. Bioinformatics, Volume 1: Data, Sequence Analysis, and Evolution, edited by Jonathan M. Keith, 2008 451. Plant Virology Protocols: From Viral Sequence to Protein Function, edited by Gary Foster, Elisabeth Johansen, Yiguo Hong, and Peter Nagy, 2008 450. Germline Stem Cells, edited by Steven X. Hou and Shree Ram Singh, 2008 449. Mesenchymal Stem Cells: Methods and Protocols, edited by Darwin J. Prockop, Donald G. Phinney, and Bruce A. Brunnell, 2008 448. Pharmacogenomics in Drug Discovery and Development, edited by Qing Yan, 2008 447. Alcohol: Methods and Protocols, edited by Laura E. Nagy, 2008 446. Post-translational Modification of Proteins: Tools for Functional Proteomics, Second Edition, edited by Christoph Kannicht, 2008 445. Autophagosome and Phagosome, edited by Vojo Deretic, 2008 444. Prenatal Diagnosis, edited by Sinhue Hahn and Laird G. Jackson, 2008 443. Molecular Modeling of Proteins, edited by Andreas Kukol, 2008 442. RNAi: Design and Application, edited by Sailen Barik, 2008 441. Tissue Proteomics: Pathways, Biomarkers, and Drug Discovery, edited by Brian Liu, 2008
440. Exocytosis and Endocytosis, edited by Andrei I. Ivanov, 2008 439. Genomics Protocols, Second Edition, edited by Mike Starkey and Ramnanth Elaswarapu, 2008 438. Neural Stem Cells: Methods and Protocols, Second Edition, edited by Leslie P. Weiner, 2008 437. Drug Delivery Systems, edited by Kewal K. Jain, 2008 436. Avian Influenza Virus, edited by Erica Spackman, 2008 435. Chromosomal Mutagenesis, edited by Greg Davis and Kevin J. Kayser, 2008 434. Gene Therapy Protocols: Volume 2: Design and Characterization of Gene Transfer Vectors, edited by Joseph M. LeDoux, 2008 433. Gene Therapy Protocols: Volume 1: Production and In Vivo Applications of Gene Transfer Vectors, edited by Joseph M. LeDoux, 2008 432. Organelle Proteomics, edited by Delphine Pflieger and Jean Rossier, 2008 431. Bacterial Pathogenesis: Methods and Protocols, edited by Frank DeLeo and Michael Otto, 2008 430. Hematopoietic Stem Cell Protocols, edited by Kevin D. Bunting, 2008 429. Molecular Beacons: Signalling Nucleic Acid Probes, Methods and Protocols, edited by Andreas Marx and Oliver Seitz, 2008 428. Clinical Proteomics: Methods and Protocols, edited by Antonia Vlahou, 2008 427. Plant Embryogenesis, edited by Maria Fernanda Suarez and Peter Bozhkov, 2008 426. Structural Proteomics: HighThroughput Methods, edited by Bostjan Kobe, Mitchell Guss, and Huber Thomas, 2008 425. 2D PAGE: Sample Preparation and Fractionation, Volume 2, edited by Anton Posch, 2008
Osteoporosis Methods and Protocols
Jennifer J. Westendorf Editor Mayo Clinic, Department of Orthopedic Surgery Rochester, MN, USA
Editor Jennifer J. Westendorf Mayo Clinic Department of Orthopedic Surgery Rochester, MN USA
[email protected]
Series Editor John M. Walker University of Hertfordshire Hatfield, Hertz UK
ISBN: 978-1-58829-828-7 e-ISBN: 978-1-59745-104-8 DOI: 10.1007/978-1-59745-104-8 Library of Congress Control Number: 2008920309 © 2008 Humana Press, a part of Springer Science+Business Media, LLC All rights reserved. This work may not be translated or copied in whole or in part without the written permission of the publisher (Humana Press, 999 Riverview Drive, Suite 208, Totowa, NJ 07512 USA), except for brief excerpts in connection with reviews or scholarly analysis. Use in connection with any form of information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed is forbidden. The use in this publication of trade names, trademarks, service marks, and similar terms, even if they are not identified as such, is not to be taken as an expression of opinion as to whether or not they are subject to proprietary rights. While the advice and information in this book are believed to be true and accurate at the date of going to press, neither the authors nor the editors nor the publisher can accept any legal responsibility for any errors or omissions that may be made. The publisher makes no warranty, express or implied, with respect to the material contained herein. Cover Illustration: Background: Provided by Dr. Stefan Judex (Chapter 22, Fig. 2). Foreground: Provided by Dr. Ralph Muller (Chapter 19, Fig. 1). Printed on acid-free paper 9 8 7 6 5 4 3 2 1 springer.com
Preface
Osteoporosis is a disease characterized by reduced bone mass, quality, and strength, changes in skeletal micro-architecture, and increased fracture risk. An astonishing number of individuals have osteoporosis throughout the world, including 75 million people in the United States, Europe, and Japan alone (1). An estimated 30–50% of women and 15–30% of men will suffer on osteoporosis-related fracture in their lifetimes (1). Recovery time from osteoporotic fractures is lengthy, often includes terminal nursing home care, and places enormous financial burdens on society, health-care systems, businesses, families, and individuals. Thus, osteoporosis has a broad impact on human health. Osteoporosis, which literally means “porous bone,” is often referred to as the silent disease because symptoms are not noticed until a fracture occurs. A tremendous comprehensive effort has been made in the last two decades to improve diagnostic methods, increase awareness, and identify treatments for osteoporosis. Known causes of osteoporosis include aging, genetics, hormonal imbalances (e.g., estrogen deficiency, hyperparathyroidism), environmental factors (e.g., diet, nutrition, smoking and exercise), and medical drugs (e.g., corticosteroids and chemotherapy). Many advances have occurred in recent years because of novel research methodologies that allow scientists to study bone cells and tissues at the cellular and molecular levels and in a variety of models. The goal of this textbook is to provide a platform for leading scientists from around the world to share their protocols for studying bone biology. We begin with in vitro models to study bone cells and tissues. The emphasis is placed on osteoblasts (bone-forming cells) and osteoclasts (bone-resorbing cells), but it is recognized that other bone marrow–derived cells, notably adipocytes and hematopoietic cells, have major influences on the activities of osteoblasts and osteoclasts and thus contribute to the development and retention of bone mass. The second section details several in vivo rodent models that can be utilized for drug testing, tissue engineering, and studying osteoporosis in either gender. The third section contains three viral gene delivery protocols for studying the roles of specific gene products in bone cells. The fourth section includes relatively new state-of-the art molecular techniques that can be utilized to assess single genes or for global genomic analysis. Section five conv
vi
Preface
sists of protocols for powerful imaging techniques that allow for in situ observations of bone cells and tissues. The sixth and final section describes methods to analyze bone responses to mechanical stimuli. Many beautiful color reproductions of figures present in the textbook can be found on the compact disk (CD) that accompanies it. I hope that this Methods in Molecular Biology: Osteoporosis book will help both new and experienced experimentalists studying bone biology and osteoporosis. Of course, all potential topics could not be covered in one textbook. Thus, I refer the readers to another book in the Molecular Methods in Medicine series, Bone Research Protocols, edited by Helfrich and Ralston (2), wherein additional techniques and alternative descriptions to some of the topics covered in this volume are described. A second excellent book, Handbook of Histology Methods for Bone and Cartilage, edited by An and Martin (3), focuses on histology and histomorphometry techniques. Thank you to the authors who contributed their techniques and pointers to this manual. It was a pleasure to work with them. I learned a lot! Thank you Julia Hutchcroft at the University of Minnesota and Teresa Hoff at the Mayo Clinic for your administrative assistance. Finally, thank you John Walker and Humana Press for your patience and guidance and for giving me the opportunity to edit this textbook. Jennifer J. Westendorf, PhD
1. International Osteoporosis Foundation. www.iofbonehealth.org/facts-andstatistics.html 2. Helfrich, M. H., Ralston S. T. (2003) Molecular Methods in Medicine: Bone Research Protocols. Humana Press, Totowa, NJ. 3. An, Y. A., Martin, K. L. (2003) Handbook of Histology Methods for Bone and Cartilage. Humana Press, Totowa, NJ.
Contents
List of Color Plates ..........................................................................................
xi
Part I In Vitro Models 1
Isolation and Culture of Rodent Osteoprogenitor Cells .............................................................................. Thomas A. Owen and Lydia C. Pan
2
Osteoclast Culture and Resorption Assays ........................................... Elizabeth W. Bradley and Merry Jo Oursler
3
Assessing New Bone Formation in Neonatal Calvarial Organ Cultures ....................................................................... Khalid S. Mohammad, John M. Chirgwin, and Theresa A. Guise
4
Detection of Apoptosis of Bone Cells In Vitro ....................................... Teresita Bellido and Lilian I. Plotkin
3
19
37
51
Part II In Vivo Rodent Models 5
6
7
In Vivo Parathyroid Hormone Treatments and RNA Isolation and Analysis ............................................................ Xin Li, Ling Qin, and Nicola C. Partridge Assessment of Bone Formation Capacity Using In vivo Transplantation Assays: Procedure and Tissue Analysis ............................................................... Basem M. Abdallah, Nicholas Ditzel, and Moustapha Kassem
79
89
Ossicle and Vossicle Implant Model Systems ........................................ 101 Glenda J. Pettway and Laurie K. McCauley vii
viii
Contents
8
Utility of the Ovariectomized Rat as a Model for Human Osteoporosis in Drug Discovery ............................................................. 111 Yogendra P. Kharode, Michael C. Sharp, and Peter V.N. Bodine
9
Orchidectomy Models of Osteoporosis .................................................. 125 Stéphane Blouin, Hélène Libouban, Marie Françoise Moreau, and Daniel Chappard
Part III
Methods of Gene Delivery
10
Gene Delivery by Adenoviruses .............................................................. 137 Renny T. Franceschi and Chunxi Ge
11
Lentivirus Delivery of shRNA Constructs into Osteoblasts ................ 149 Marc N. Wein, Dallas C. Jones, and Laurie H. Glimcher
12
Gene Delivery by Retroviruses ............................................................... 157 Valerie Deregowski and Ernesto Canalis
Part IV Molecular and Genetic Techniques 13
Chromatin Immunoprecipitation Assays: Application of ChIP-on-Chip for Defining Dynamic Transcriptional Mechanisms in Bone Cells ........................................... 165 Margaretha van der Deen, Mohammad Q. Hassan, Jitesh Pratap, Nadiya M. Teplyuk, Daniel W. Young, Amjad Javed, Sayyed K. Zaidi, Jane B. Lian, Martin Montecino, Janet L. Stein, Gary S. Stein, and Andre J. van Wijnen
14
Identification of Transcription Factor Target Genes by ChIP Display ....................................................................................... 177 Artem Barski, Steven Pregizer, and Baruch Frenkel
15
Application of the Laser Capture Microdissection Technique for Molecular Definition of Skeletal Cell Differentiation In Vivo ............................................................................. 191 Dafna Benayahu, Rina Socher, and Irena Shur
16
Quantitative Trait Loci Mapping ........................................................... 203 Dong-Hai Xiong, Jian-Feng Liu, Yan-Fang Guo, Yan Guo, Tie-Lin Yang, Hui Jiang, Yuan Chen, Fang Yang, Robert R Recker, and Hong-Wen Deng
Contents
ix
Part V Imaging Techniques 17
In Situ Nuclear Organization of Regulatory Machinery ..................... 239 Shirwin M. Pockwinse, Sayyed K. Zaidi, Ricardo F. Medina, Rachit Bakshi, Krishna P. Kota, Syed A. Ali, Daniel W. Young, Jeffery A. Nickerson, Amjad Javed, Martin Montecino, Andre J. van Wijnen, Jane B. Lian, Janet L. Stein, and Gary S. Stein
18
Bioluminescent Imaging in Bone ............................................................ 261 Yoram Zilberman, Yossi Gafni, Gadi Pelled, Zulma Gazit, and Dan Gazit
19
Micro-Computed Tomography: A Method for the Non-Destructive Evaluation of the Three-Dimensional Structure of Biological Specimens ........................................................................... 273 Martin Stauber and Ralph Müller
20
Fourier Transform-Infrared Microspectroscopy and Microscopic Imaging ........................................................................ 293 Samuel Gourion-Arsiquaud, Paul A. West, and Adele L. Boskey
Part VI Mechanical Testing 21
Assessment of the In Vivo Adaptive Response to Mechanical Loading ............................................................................ 307 Leanne Kaye Saxon and Lance Edward Lanyon
22
Determination of Bone’s Mechanical Matrix Properties by Nanoindentation ............................................................... 323 Engin Ozcivici, Suzanne Ferreri, Yi-Xian Qin, and Stefan Judex
23
Fluid Flow Assays .................................................................................... 335 Ryan C. Riddle, Amanda F. Taylor, and Henry J. Donahue
Index .................................................................................................................. 347
List of Color Plates
The following color illustrations are printed in the insert Fig. 1.1
Fig. 1.1
Fig. 1.3
Fig. 1.4
Osteoblastic differentiation of mouse calvarial cultures. a. Calvarial cells were isolated from Col3.6-Topaz transgenic mice, which express green fluorescent protein (GFP) driven by a 3.6-kb rat type I procollagen promoter (15), and cultured under osteogenic differentiation conditions. On the indicated days after plating, cells were fixed with formalin. Using a 10× objective, the same fields were photographed using epifluorescence to visualize GFP and phase contrast optics. Nascent nodules first appeared as refractile areas where cells have begun to deposit collagen and form multilayers around day 8. Cells possessing GFP fluorescence presaged visible nodule formation and were preferentially localized within differentiating nodules. As seen in the day 14 phase contrast image, nodules grew in area and acquired a granular appearance as mineralization proceeded b. A transverse section through a mineralized nodule from a rat calvarial osteoblast culture (study day 19) that was fixed with formalin, embedded in Spurr’s resin, sectioned and stained with toluidine blue. Bright field microscopy revealed mineralized matrix (purple) on the side of the nodule next to the culture dish and a layer of cuboidal osteoblasts atop of a layer of osteoid adjacent to the culture medium Alkaline phosphatase enzyme histochemical staining during osteoblastic differentiation. Rat calvarial cells were isolated and cultured under osteogenic differentiation conditions. On the indicated days after plating, a dish was fixed and air dried. At the end of the timecourse, all dishes were stained for alkaline phosphatase enzyme activity. While alkaline phosphatase activity was already detectable at day 6 (red color), it increased in intensity with osteoblastic differentiation and the strongest staining was associated with the multilayered nodules of differentiated osteoblasts Extracellular matrix mineralization during osteoblastic differentiation. Rat calvarial or bone marrow cells were isolated and cultured under osteogenic differentiation conditions. a. On the indicated days after plating, rat calvarial cultures were von Kossa stained and viewed by phase xi
xii
Fig. 20.3
List of Color Plates
contrast optics using a 4× objective. Small mineralized nodules as well as unmineralized nodules were detectable by day 8. With time, the mineralized nodules increased in size and number. b. Unmagnified views of von Kossa (VK) stained cultures showing a differentiated rat calvarial culture (ROB) in a 12-well dish 12 days after plating (ROB) and a differentiated rat bone marrow culture in a 6-well dish 18 days after plating (BMC). c. Unmagnified view of mineralized nodules visualized by calcein incorporation. ROB and BMC cultures were stained with either calcein (left panel) or calcein blue (center and right panels) and photographed on top of an ultraviolet transilluminator Hyperspectral images obtained from normal baboon cortical bone tissue. a. A typical FTIR spectrum from a single image pixel showing frequencies of interest. b. Integrated matrix areas (1,592–1,712 cm−1) are calculated and displayed as 2D color-coded images to give an IR image of the matrix distribution. c. Integrated mineral areas (916–1,180 cm−1) and the tissue mineral distribution and (d) the carbonate (840–892 cm−1) distribution. Images of the mineral:matrix ratio (e), the carbonate:mineral ratio (f), the crystallinity ratio (1,030/1,020 cm−1) (g) and the crosslink ratio (1,660/1,690 cm−1) (h) are shown. Note: all spectra are baselined and PMMA (embedding media) corrected using ISYS software
Contributors
Basem M. Abdallah, MSc, PhD Department of Endocrinology, KMEB Laboratory, University Hospital of Odense, Denmark Syed A. Ali, MD Department of Cell Biology, University of Massachusetts Medical School, Worcester, MA Rachit Bakshi, PhD Department of Cell Biology, University of Massachusetts Medical School, Worcester, MA Artem Barski, PhD Department of Biochemistry & Molecular Biology, University of Southern California Keck School of Medicine, Los Angeles, CA Dafna Benayahu, PhD Department of Cell and Developmental Biology, Sackler School of Medicine, Tel-Aviv University, Tel-Aviv, Israel Teresita Bellido, PhD Division of Endocrinology & Metabolism and Center for Osteoporosis and Metabolic Bone Diseases, University of Arkansas for Medical Sciences, Little Rock, AR Peter V.N. Bodine, PhD Osteoporosis Research, Women’s Health and Musculoskeletal Biology, Wyeth Research, Collegeville, PA Adele L. Boskey, PhD Mineralized Tissue Research, Hospital for Special Surgery, Weill Medical College and Cornell University, New York, NY Stéphane Blouin, PhD INSERM, U 922, Université d’Angers, Angers, France xiii
xiv
Contributors
Elizabeth Bradley Mayo Clinic Graduate School, Rochester, MN Ernesto Canalis, MD Department of Research, Saint Francis Hospital and Medical Center, Hartford, CT, and The University of Connecticut School of Medicine, Farmington, CT Daniel Chappard, MD, PhD INSERM, U 922, Université d’Angers, Angers, France Shu-Gui Chen The Key Laboratory of Biomedical Information Engineering of Ministry of Education and Institute of Molecular Genetic, School of Life Science and Technology, Xi’an Jiaotong University, Xi’an, P.R. China Yuan Chen Laboratory of Molecular and Statistical Genetics, College of Life Sciences, Hunan Normal University, Changsha, Hunan, P.R. China John M. Chirgwin, PhD Department of Internal Medicine, Division of Endocrinology, University of Virginia, Charlottesville, VA Hong-Wen Deng, PhD Departments of Orthopedic Surgery and Basic Medical Sciences, University of Missouri-Kansas City, Kansas City, MO Valerie Deregowski, PhD Department of Research, Saint Francis Hospital and Medical Center, Hartford, CT Nicholas Ditzel, MSc Department of Endocrinology, KMEB Laboratory, University Hospital of Odense, Denmark Henry J. Donahue, PhD Division of Musculoskeletal Sciences and Department of Orthopaedics and Rehabiliation, Pennsylvania State University College of Medicine, Milton S. Hershey Medical Center, Hershey, PA Suzanne Ferreri Department of Biomedical Engineering, State University of New York at Stony Brook, Stony Brook, NY Renny T. Franceschi, PhD Department of Periodontics & Oral Medicine, School of Dentistry, and Biological Chemistry, School of Medicine, University of Michigan, Ann Arbor, MI Baruch Frenkel, DMD, PhD Departments of Orthopaedic Surgery and Biochemistry & Molecular Biology, University of Southern California Keck School of Medicine, Los Angeles, CA
Contributors
Yossi Gafni, MSc, DMD Skeletal Biotechnology Laboratory, Hebrew University- Hadassah Medical Center, Jerusalem, Israel Dan Gazit, PhD, DMD Skeletal Biotechnology Laboratory, Hebrew University- Hadassah Medical Center, Jerusalem, Israel Zulma Gazit, PhD Skeletal Biotechnology Laboratory, Hebrew University- Hadassah Medical Center, Jerusalem, Israel Chunxi Ge, MD, PhD Departments of Periodontics & Oral Medicine, School of Dentistry, and Biological Chemistry, School of Medicine, University of Michigan, Ann Arbor, MI Laurie H. Glimcher, MD Harvard Medical School and Harvard University of Public Health, Boston, MA Samuel Gourion-Arsiquaud, PhD Mineralized Tissue Research, Hospital for Special Surgery, New York, NY Theresa A. Guise, MD Department of Internal Medicine, Division of Endocrinology, University of Virginia, Charlottesville, VA Yan Guo The Key Laboratory of Biomedical Information Engineering of Ministry of Education and Institute of Molecular Genetic, School of Life Science and Technology, Xi’an Jiaotong University, Xi’an, P.R. China Yan-Feng Guo The Key Laboratory of Biomedical Information Engineering of Ministry of Education and Institute of Molecular Genetic, School of Life Science and Technology, Xi’an Jiaotong University, Xi’an, P.R. China Mohammad Q. Hassan, PhD Department of Cell Biology, University of Massachusetts Medical School, Worcester, MA Amjad Javed, PhD School of Dentistry, University of Alabama at Birmingham Institute of Oral Health Research, Birmingham, AL Hui Jiang Laboratory of Molecular and Statistical Genetics, College of Life Sciences, Hunan Normal University, Changsha, Hunan, P.R. China Dallas C. Jones, PhD Harvard Medical School and Harvard University of Public Health, Boston, MA
xv
xvi
Contributors
Stefan Judex, PhD Department of Biomedical Engineering, State University of New York at Stony Brook, Stony Brook, NY Moustapha Kassem, MD, PhD, DSc Department of Endocrinology, KMEB Laboratory, University Hospital of Odense, Denmark Yogendra P. Kharode, MS Osteoporosis Research, Women’s Health and Musculoskeletal Biology, Wyeth Research, Collegeville, PA Krishna P. Kota, PhD Department of Cell Biology, University of Massachusetts Medical School, Worcester, MA Lance E. Lanyon, CBE, BVSc, PhD, DSc, MRCVS, FMedSci Royal Veterinary College, London, UK Xin Li, PhD Department of Periodontics & Oral Medicine, School of Dentistry, University of Michigan, Ann Arbor, MI Jane B. Lian, PhD Department of Cell Biology, University of Massachusetts Medical School, Worcester, MA Hélène Libouban, PhD INSERM, U 922, Université d’Angers, Angers, France Jian-Feng Liu Departments of Orthopedic Surgery and Basic Medical Sciences, University of Missouri-Kansas City, Kansas City, MO Laurie K. McCauley, DDS, PhD Departments of Periodontics & Oral Medicine and Pathology, University of Michigan, Ann Arbor, MI Ricardo F. Medina, PhD Department of Cell Biology, University of Massachusetts Medical School, Worcester, MA Khalid S. Mohammad, MD, PhD Department of Internal Medicine, Division of Endocrinology, University of Virginia, Charlottesville, VA Martin Montecino, PhD Departamento de Biologia Molecular, Universidad de Concepcion, Facultad de Ciencias Biologicas, Concepcion, Chile
Contributors
xvii
Marie Françoise Moreau, PhD INSERM, U 922, Université d’Angers, Angers, France Ralph Müller, PhD Institute for Biomechanics, ETH Zürich, Zürich, Switzerland Jeffrey A. Nickerson, PhD Department of Cell Biology, University of Massachusetts Medical School, Worcester, MA Merry Jo Oursler, PhD Endocrine Research Unit, Mayo Clinic, Rochester, MN Thomas A. Owen, PhD Department of Cardiovascular, Metabolic and Endocrine Diseases, Pfizer Global Research and Development, Groton, CT Engin Ozcivici Department of Biomedical Engineering, State University of New York at Stony Brook, Stony Brook, NY Lydia C. Pan, PhD Department of Cardiovascular, Metabolic and Endocrine Diseases, Pfizer Global Research and Development, Groton, CT Nicola C. Partridge, PhD Department of Physiology and Biophysics, University of Medicine and Dentistry of New Jersey, Robert Wood Johnson Medical School, Piscataway, NJ Gadi Pelled, DMD Skeletal Biotechnology Laboratory, Hebrew University- Hadassah Medical Center, Jerusalem, Israel Glenda J. Pettway, MS Departments of Periodontics & Oral Medicine and Biomedical Engineering, University of Michigan, Ann Arbor, MI Lilian I. Plotkin, PhD Division of Endocrinology & Metabolism and Center for Osteoporosis and Metabolic Bone Diseases, University of Arkansas for Medical Sciences, Little Rock, AR Shirwin M. Pockwinse, BA Department of Cell Biology, University of Massachusetts Medical School, Worcester, MA Jitesh Pratap, PhD Department of Cell Biology, University of Massachusetts Medical School, Worcester, MA
xviii
Contributors
Steven Pregizer Department of Biochemistry & Molecular Biology, University of Southern California Keck School of Medicine, Los Angeles, CA Ling Qin, PhD Department of Physiology and Biophysics, University of Medicine and Dentistry of New Jersey, Robert Wood Johnson Medical School, Piscataway, NJ Yi-Xian Qin, PhD Department of Biomedical Engineering, State University of New York at Stony Brook, Stony Brook, NY Robert R. Recker, MD, MACP, FACE Osteoporosis Research Center, Creighton University, Omaha, NE Ryan C. Riddle, PhD Division of Musculoskeletal Sciences and Department of Orthopaedics and Rehabilitation, Pennsylvania State University College of Medicine, Milton S. Hershey Medical Center, Hershey, PA Leanne K. Saxon, Ba Appl Sci (Hon), PhD Royal Veterinary College, London, UK Michael C. Sharp Osteoporosis Research, Women’s Health and Musculoskeletal Biology, Wyeth Research, Collegeville, PA Irena Shur, MD, PhD Department of Cell and Developmental Biology, Sackler School of Medicine, Tel-Aviv University, Tel-Aviv, Israel Rina Socher. MSc Department of Cell and Developmental Biology, Sackler School of Medicine, Tel-Aviv University, Tel-Aviv, Israel Martin Stauber, PhD Institute for Biomechanics, ETH Zürich, Zürich, Switzerland Gary S. Stein, PhD Department of Cell Biology, University of Massachusetts Medical School, Worcester, MA Janet L. Stein, PhD Department of Cell Biology, University of Massachusetts Medical School, Worcester, MA Amanda F. Taylor, PhD Division of Musculoskeletal Sciences and Department of Orthopaedics and Rehabiliation, Pennsylvania State University College of Medicine, Milton S. Hershey Medical Center, Hershey, PA
Contributors
xix
Nadiya M. Teplyuk, PhD Department of Cell Biology, University of Massachusetts Medical School, Worcester, MA Margaretha van der Deen, Dr Department of Cell Biology, University of Massachusetts Medical School, Worcester, MA André J. Van Wijnen, PhD Department of Cell Biology, University of Massachusetts Medical School, Worcester, MA Marc N. Wein, PhD Harvard Medical School and Harvard University of Public Health, Boston, MA Paul A. West, PhD Mineralized Tissue Research, Hospital for Special Surgery, New York, NY Dong-Hai Xiong Osteoporosis Research Center, Creighton University, Omaha, NE Fang Yang Laboratory of Molecular and Statistical Genetics, College of Life Sciences, Hunan Normal University, Changsha, Hunan, P.R. China Tie-Lin Yang The Key Laboratory of Biomedical Information Engineering of Ministry of Education and Institute of Molecular Genetic, School of Life Science and Technology, Xi’an Jiaotong University, Xi’an, P.R. China Daniel W. Young, PhD Novartis Institute for Biomedical Research, Cambridge, MA Sayyed K. Zaidi, PhD Department of Cell Biology, University of Massachusetts Medical School, Worcester, MA Yoram Zilberman, PhD, DMD Skeletal Biotechnology Laboratory, Hebrew University-Hadassah Medical Center, Jerusalem, Israel
Chapter 1
Isolation and Culture of Rodent Osteoprogenitor Cells Thomas A. Owen and Lydia C. Pan
Abstract Osteoblasts are the cells responsible for formation of new bone throughout life. Rats are one of the most widely studied mammalian species in skeletal biology and serve as useful models for many aspects of human skeletal physiology. The availability of genetically modified mice as research tools has greatly enabled our understanding of how specific genes contribute to the process of skeletogenesis. In order to explore the impact of biochemical, genetic, or pharmacological manipulation on bone formation, various osteogenic cell culture systems have been developed. Two of the most widely accepted rodent osteogenic culture models, using osteoprogenitor cells isolated from calvaria or bone marrow, are described in this chapter. Keywords Osteoblast, bone marrow, mineralization, alkaline phosphatase, calcein.
1
Introduction
Cells of the osteoblastic lineage are integral to the process of bone formation, which maintains integrity of the skeleton. The molecular events that regulate osteoblastic differentiation are frequently studied using in vitro cellular models that recapitulate events occurring in vivo but are more accessible to biochemical, pharmacological, and genetic manipulation than whole organisms. Many widely used in vitro osteoblastic cell model systems employ either karyotically normal but immortal cell lines (e.g., MC3T3) or genetically atypical cell lines derived from tumors (e.g., UMR-106, ROS 17/2.8). However, tumor origins and genetic alterations associated with the prolonged culture of cell lines can raise questions about the extent to which they truly replicate in vivo phenomena. Two laboratory species that are widely employed in biomedical research on the mammalian skeleton are rats and mice. Rats have been widely used to model many aspects of human skeletal physiology, including responses to hormone deficiencies and mechanical loading. Recent advances in genetic technologies have made mice important models for investigating the roles of individual genes in the skeleton. From: Methods in Molecular Biology, Vol. 455: Osteoporosis: Methods and Protocols Edited by Jennifer J. Westendorf © Humana Press, Totowa, NJ
3
4
T.A. Owen and L.C. Pan
In vitro model systems can be effectively developed using genetically and phenotypically normal primary osteoblast precursor cells derived from either the developing calvaria of embryonic or neonatal rodents or from the bone marrow of rodents of any age. Osteogenic cultures derived from rodent calvaria undergo a series of proliferative and maturational events characterized by highly regulated changes in gene expression. Eventually, the cells form multi-layered nodules that resemble islands of bone tissue when examined histologically (1–3). The cells in these nodules are surrounded by an organized collagenous extracellular matrix onto which hydroxyapatite crystals are deposited (2–4). Similarly, osteoblastic precursor cells contained within the bone marrow undergo sequential stages of attachment, clonal growth, extracellular matrix deposition, and mineralization (5, 6). Osteogenic precursor cells derived from bone marrow are believed to represent cells at an early stage in the mesenchymal lineage, since they readily give rise not only to osteoblastic colonies, but also to adipocytic and fibroblastic colonies. In contrast, osteogenic cells isolated from the calvarium represent a more committed cell at a later stage in the lineage. The choice of experimental species and cellular model depends upon the question under investigation. Methods for preparing osteogenic precursor cells from rodents were first published more than 40 years ago (6–8) and have been refined by many investigators over time. Isolation of osteogenic cells from rodent calvariae involves sequential enzymatic digestion, typically using trypsin and collagenase, first to remove contaminating cell types such as blood cells and fibroblasts, and then to liberate the osteogenic cells from their surrounding extracellular matrix. Late-term fetal or neonatal calvaria are most often used because they provide a relatively abundant source of osteoprogenitors without the need for extensive tissue dissection. Although they originate from a different embryonic tissue than most of the rest of the skeleton, a large body of research indicates that they share many of their physiological and pharmacological responses with osteoblasts derived from other bone sites. Bone marrow yields cultures with a lower percentage of osteogenic cells than calvaria (9); however, these cells are responsible for the formation of cancellous and endocortical bone that continuously turns over and can be studied throughout the entire lifespan of the organism. With either cell source, the isolated cells proliferate and produce an organized collagenous extracellular matrix that eventually mineralizes in a manner highly reminiscent of intact bone.
2 2.1
Materials Cell Culture Media and Reagents for Rodent Calvarial Osteoblast Isolation
1. Cell digestion PBS: Cold, sterile phosphate-buffered saline, calcium and magnesium-free (PBS) with 50 µg/mL gentamicin sulfate (or 100 units/mL penicillin plus
1 Culture of Rodent Cells
2.
3.
4.
5.
6.
7.
5
100 µg/mL streptomycin) and antimycotic solution (e.g., Fungizone, 2.5 µg/mL amphotericin B in 20 µg/mL sodium deoxycholate, Invitrogen, Carlsbad, CA). Trypsin-collagenase digestion cocktail: Dissolve an amount of Collagenase P equivalent to 15 Wünsch units in 45 mL serum-free Minimal Essential Medium (MEM). When dissolved, add 5 mL of 2.5% trypsin solution in PBS to give 0.25% final trypsin concentration (see Note 1). Final collagenase activity should be 0.3 Wünsch units/mL. Sterilize by filtration through a 0.22-µm filter. Store on ice until needed, but warm to 37 °C just prior to use. Prepare fresh each day it will be used. L-Ascorbic acid: Dissolve 1 g in 20 mL of water to give 50 mg/mL final concentration (see Note 2). Sterilize by filtration through a 0.22-µm filter and freeze in 1.2-mL aliquots at −20 °C. Each aliquot is sufficient for two 500-mL bottles of culture medium and may be frozen and thawed up to three times. 0.5 M Sodium phosphate, pH 7.4: Mix 77.4 mL of 0.5 M Na2HPO4 with 22.6 mL of 0.5 M NaH2PO4. Confirm the pH (±0.1 pH unit) and sterilize by filtration through a 0.22-µm filter. Store at room temperature (see Note 3). Dexamethasone: Dissolve cell culture grade dexamethasone in 100% ethanol to a final concentration of 10−2 M. Store in small aliquots at −20 °C. Dilute 10−2 M dexamethasone 1:100 in ethanol to make a 10−4 M stock. Diluted (10−4 M) dexamethasone is stable at −20 °C for up to 6 months even after repeated freezethaw cycles. Growth medium: Supplement 500 mL of MEM with 50 mL (10%) of fetal bovine serum (FBS) (see Note 4), 0.55 mL of 50 mg/mL gentamicin sulfate (50 µg/mL final concentration). For the preparation of rat calvarial osteoblasts, add 55 µL of 10−4 M dexamethasone (10−8 M final concentration). Dexamethasone is not necessary for the growth of murine calvarial osteoblasts. Freezing medium: MEM + 30% FBS + 10% DMSO.
2.2
Rodent Calvarial Osteoblast Growth and Differentiation
1. Growth medium as in Section 2.1 Item 6 (above). 2. Differentiation medium: Supplement 500 mL of minimal essential medium, alpha modification (MEMα) with 50 mL (10%) of FBS (see Note 4), 0.55 mL of 50 mg/mL gentamicin sulfate (50 µg/mL final concentration), 0.55 mL of 50 mg/mL L-ascorbic acid (100 µg/mL final concentration including that present in the basal medium), 2.2 mL of 0.5 M sodium phosphate, pH 7.4. This yields a 3 mM total concentration of phosphate ions including what is already present in the basal medium (see Note 5). For the preparation of rat calvarial osteoblasts, add 55 µL of 10−4 M dexamethasone (10−8 M final concentration). Dexamethasone is not necessary for the differentiation of murine calvarial osteoblasts.
6
2.3
T.A. Owen and L.C. Pan
Rodent Bone Marrow Osteoblast Isolation, Growth, and Differentiation
1. Bone marrow PBS: Cold, sterile PBS with 50 µg/mL gentamicin sulfate (or 100 U/mL penicillin plus 100 µg/mL streptomycin) and antimycotic solution (e.g., Fungizone, 2.5 µg/mL amphotericin B in 20 µg/mL sodium deoxycholate, Invitrogen, Carlsbad, CA). 2. Isomet low speed saw with diamond wafering blade (Buehler, Lake Bluff, IL). 3. 10-cc syringe with flexible tubing (~5 cm) attached. The tubing diameter should allow a snug fit around the bone from which the bone marrow cells will be isolated (i.e., adult rat vs juvenile mice) and should be sterilized by immersion in 70% ethanol for at least 10 minutes. 4. L-ascorbic acid as in Section 2.1 Item 3. 5. 0.5 M sodium phosphate, pH 7.4 as in Section 2.1 Item 4. 6. Dexamethasone as in Section 2.1 Item 5. 7. Growth medium: Supplement 500 mL of MEMα with 75 mL (15%) of FBS (see Note 4) and 0.58 mL of 50 mg/mL gentamicin sulfate (50 µg/mL final concentration). 8. Differentiation medium: Supplement 500 mL of MEMα medium with 50 mL (10%) of FBS (see Note 4), 0.55 mL of 50 mg/mL gentamicin sulfate (50 µg/mL final concentration), 0.55 mL of 50 mg/mL L-ascorbic acid (100 µg/mL final concentration including that present in the basal medium), 2.2 mL of 0.5 M sodium phosphate pH 7.4 (3 mM final concentration including the phosphate present in the basal medium). For the preparation of rat calvarial osteoblasts, add 55 µL of 10−4 M dexamethasone (10−8 M final concentration). Dexamethasone is not necessary for the differentiation of murine bone marrow-derived osteoblasts (see Note 5).
2.4
Alkaline Phosphatase Enzyme Activity
1. Alkaline phosphatase lysis buffer: 150 mM Tris-HCl, pH 9.0, 0.1 mM ZnCl2, 0.1 mM MgCl2, 1% Triton X-100. Prepare by dissolving Tris base, ZnCl2, and MgCl2 in water, adjusting the pH to 9.0 with HCl, and then adding the Triton X-100 and water to the required volume. Sterilize by filtration through a 0.22 µm filter and store at room temperature. 2. Measurement of cell protein for normalization of enzyme activity: Detergentcompatible DC protein assay reagent (BioRad, Hercules, CA) or any other protein measurement method compatible with 1% Triton X-100. 3. Enzyme substrate: 1-Step PNPP (p-nitrophenyl phosphate) (Pierce, Rockford, IL) or any other detection reagent based on PNPP for colorimetric detection of absorbance at 405 nm (see Note 6). Prewarm to 37 °C prior to use.
1 Culture of Rodent Cells
2.5
7
Alkaline Phosphatase Histochemical Staining
1. Fixative: 100% methanol chilled to −20 °C (see Note 7). 2. 0.2 M Tris-maleate, pH 8.4: Titrate 0.5 M Tris base solution with maleic acid to the desired pH, and then adjust volume to obtain a final concentration of 0.2 M. 3. Alkaline phosphatase histochemical stain: Dissolve 10 mg of Napthol AS-MX phosphate, sodium salt, in 0.56 mL of dimethylformamide. Add 9.44 mL of water, 10 mL of 0.2 M Tris-maleate, pH 8.4, and 20 mg of Fast Red TR salt and stir until dissolved. If the solution is not completely clear, filter it (vacuum or gravity) through Whatman 3 M paper. Filtration is not usually necessary unless high resolution microscopy is being performed. Do not use if the solution has turned brownish yellow and cloudy. Prepare only as much stain as is needed for immediate use (see Note 8).
2.6
von Kossa Staining for Assessment of Mineralization
1. Fixative: 100% methanol (−20 °C) (see Note 7). 2. 5% AgNO3 in water. Store at room temperature protected from light. Solutions containing silver nitrate should be disposed of as hazardous waste.
2.7
Calcein Staining for Assessment of Mineralization
1. Fix: 100% methanol (−20 °C) (see Note 7). 2. Calcein (CAS Number: 1461-15-0) or calcein blue (CAS Number: 54375-47-2): 2.5 mg/mL in PBS. Sterilize by filtration through a 0.22-µm filter and store in the dark at 4 °C.
3 3.1
Methods Rodent Calvarial Osteoblast Isolation
1. This isolation protocol works best with rat pups ranging in age from day 21 of gestation through day 1–2 postpartum or with neonatal mice up to 7 days old. In accordance with institutional policies and procedures, euthanize pregnant dams, remove fetuses by Caesarean section and euthanize (e.g., by decapitation). For neonates, euthanize in accordance with institutional policies and procedures
8
2.
3.
4.
5.
T.A. Owen and L.C. Pan
(e.g., CO2 asphyxiation followed by decapitation). Dispose of the bodies and retain heads for dissection of the calvariae. Using sterile instruments, dissect each calvarium consisting of parietal bones plus adjoining sections of frontal and occipital bones, but exclude lateral sutures in order to minimize soft tissue contamination. This procedure may be done in a laminar flow hood to minimize the chance of bacterial or fungal contamination, but has also been performed on the benchtop without negative consequences. Cut each calvarium in half along the midline to prevent bones from stacking and sticking together during enzymatic digestion. Collect bone pieces in cell digestion PBS. Rinse bones twice with Cell Digestion PBS to eliminate as much blood as possible. Bones may be digested immediately or stored in this solution at 4 °C for up to 3 days without significant loss of cell viability (see Note 9). Transfer bones to a sterile 50-mL conical tube using sterile forceps for enzymatic digestion. a) Add pre-warmed trypsin-collagenase digestion cocktail, using 10 mL for up to 50 calvariae. Scale up volume for larger numbers of bones, but use no less than 10 mL for small numbers of bones. Fasten the tube horizontally in an orbital shaker and agitate at 120 rpm at 37 °C for 15 minutes. The cells released by this digestion (fraction I) are primarily red blood cells with a small percentage of fibroblasts. This first digest also releases loosely attached fragments of overlying tissue. Remove the tube from the orbital shaker and allow the bone fragments to settle to the bottom of the tube. Pipet off the supernatant and discard it. b) To the bone fragments, add the same volume of fresh, pre-warmed trypsincollagenase digestion cocktail as used in step 5.1. Fasten the tube horizontally in an orbital shaker and agitate at 120 rpm at 37 °C for 30 minutes. This digestion (fraction II) generally releases a large percentage of the osteogenic cell population from the calvarial fragments. Remove tube from the orbital shaker and allow the bone fragments to settle to the bottom. Using a pipet, transfer the supernatant containing cells into a clean sterile tube and add FBS to 10% (v/v) to neutralize protease activity. Keep cell suspension on ice until the next step is completed. c) To the bone fragments, add the same volume of fresh, pre-warmed trypsincollagenase digestion cocktail as used in step 5.1. Fasten the tube horizontally in an orbital shaker and agitate at 120 rpm at 37 °C for an additional 30 minutes. This digestion (fraction III) releases most of the remaining osteogenic cells from the calvarial fragments. Remove the tube from the orbital shaker and allow the bone fragments to settle to the bottom of the tube. Pipet the supernatant containing cells into a clean sterile tube and add FBS to 10% (v/v) to neutralize the activity of the digestion enzymes. Rinse the bone fragments with 10 mL of growth medium to recover any additional cells and add this rinse to the tube containing the fraction III digest.
1 Culture of Rodent Cells
9
6. Combine the FBS-neutralized fraction II and III supernatants and filter through a sterile 70- or 100-µm nylon mesh such as a cell strainer (BD Falcon) into a sterile 50-mL collection tube. 7. Pellet the cells by centrifugation at 250 g for 5 minutes. 8. Wash the cell pellet once with growth medium and centrifuge again as in step 7. 9. Resuspend the cell pellet in growth medium using 0.5 mL medium per calvarium. 10. Dilute a sample of cells 1:1 with 0.4% Trypan Blue in PBS and count viable cells using hemacytometer. Typical cell yields are in the range of 1.0 × 106 cells/calvarium for rats and 1.0–1.5 × 106 cells/calvarium for mice. 11. Plate cells in growth medium at a density of 20,000 cells/cm2 (rat) or 60,000 cells/cm2 (mouse). Incubate for 24 hours at 37 °C in a 5% CO2 incubator (see Note 10). 12. Detach cells using 0.25% trypsin with 1 mM EDTA, pellet by centrifugation for 5 minutes at 250 g, then resuspend cells in growth medium and count a sample of the cells. Pellet the cells again by centrifugation at 250 g for 5 minutes. Based on the cell count, resuspend the resulting pellet in freezing medium at 2 × 106 cells/mL. Freeze in cryovials at 1 mL/vial. Once frozen, transfer to liquid nitrogen for long-term storage.
3.2
Rodent Calvarial Osteoblast Growth and Differentiation
1. Thaw a vial of frozen osteoblasts (2 × 106 cells) into a T75 flask containing 20 mL of growth medium and incubate overnight. 2. The following day (day 1), cells may be trypsinized and replated for experiments as primary cultures. Alternatively, cultures may be re-fed with growth medium to remove residual DMSO and allowed to proliferate prior to subculturing (see Note 11). 3. Trypsinize the cells, determine viable cell concentration by Trypan Blue exclusion and plate these cells for the desired experiment at 20,000 cells/cm2 in growth medium. The use of culture vessels with a surface area of at least 2 cm2 per well (24-well dishes) is recommended (see Note 12). The day of plating is study day 0. Feed all wells again with growth medium on study day 2. 4. The cells should be completely confluent by study day 4. On study day 4 and every other day thereafter, feed wells with differentiation medium. After this point, care must be taken not to dislodge the cell layer when feeding. Multilayered nodules of differentiated osteoblasts first become visible as irregular, opaque foci in regions of highest cell density, typically beginning around study day 8. These nodules should be extensively mineralized by study day 14. The rate and extent of differentiation can be affected by several factors including genetics and variability in fetal bovine serum lots. Fig. 1.1a shows the appearance of differentiating calvarial cultures bearing a fluorescent transgene by phase contrast and fluorescence microscopy. Histologically, a cross-section of one of these multilayered
10
T.A. Owen and L.C. Pan
Fig. 1.1 Osteoblastic differentiation of mouse calvarial cultures. a. Calvarial cells were isolated from Col3.6-Topaz transgenic mice, which express green fluorescent protein (GFP) driven by a 3.6kb rat type I procollagen promoter (15), and cultured under osteogenic differentiation conditions. On the indicated days after plating, cells were fixed with formalin. Using a 10× objective, the same fields were photographed using epifluorescence to visualize GFP and phase contrast optics. Nascent nodules first appeared as refractile areas where cells have begun to deposit collagen and form multilayers around day 8. Cells possessing GFP fluorescence presaged visible nodule formation and were preferentially localized within differentiating nodules. As seen in the day 14 phase contrast image, nodules grew in area and acquired a granular appearance as mineralization proceeded (See Color Plates).
nodules of cells resembles that of newly formed woven bone. A layer of cuboidal osteoblasts can be seen on top of the nodule overlaying a region of dense collagenous extracellular matrix, the bottom of which is the site of active mineral deposition (Fig. 1.1b).
3.3
Rodent Bone Marrow Osteoblast Isolation, Growth and Differentiation
1. This procedure has been used to isolate total bone marrow cells capable of osteoblastic differentiation from rodents varying in age from 6 weeks to 28 months.
1 Culture of Rodent Cells
11
Fig. 1.1 (continued) b. A transverse section through a mineralized nodule from a rat calvarial osteoblast culture (study day 19) that was fixed with formalin, embedded in Spurr’s resin, sectioned and stained with toluidine blue. Bright field microscopy revealed mineralized matrix (purple) on the side of the nodule next to the culture dish and a layer of cuboidal osteoblasts atop of a layer of osteoid adjacent to the culture medium (See Color Plates).
In accordance with institutional policies and procedures, euthanize the rats or mice and dissect out the tibiae (rats) or femorae and tibiae (mice). Using a lowspeed Isomet saw with a diamond edged wafering blade and bone marrow PBS in the reservoir, cut off the distal end of the tibia immediately proximal to the tibiofibular junction. Remove the proximal end of the tibia including the metaphysis. Both ends of mouse femorae should be removed, leaving a cylinder of bone (see Note 13). 2. Rinse the syringe with attached tubing with PBS or medium to remove residual ethanol and fill with growth medium. Fit tubing over the distal end of the tibia and depress the plunger. Typically, the marrow cells will be extruded as an intact “plug.” This step may be performed in a laminar flow hood if desired, but can also be done aseptically on an open lab bench. 3. After all bone marrows have been extruded, carry out the remainder of the procedure in a laminar flow hood. Triturate the bone marrow using a 5- or 10-mL tissue culture pipet until it is a well dispersed cellular suspension. Adjust the volume with growth medium to approximately 5 mL per rat tibia or 1 mL per mouse bone. Remove any tissue fragments by filtration through a sterile 70- or 100-µm nylon filter such as a Cell strainer (BD Falcon) into a sterile 50-mL collection tube. 4. Remove 100 µL of the cell suspension and add to 900 µL of PBS, then further dilute with an equal volume of Trypan Blue solution. This results in a 1:20 dilution of the cells and will typically yield a convenient concentration of cells for counting. The dilution factor can be adjusted as needed. Count all cells, regardless of size. Calculate the total cell concentration, remembering to take the appropriate dilution factor into account. A typical yield of cells is in the range of 1.0 × 108 cells/tibia for adult rats and 4–5 × 107 cells from two tibiae plus two femorae of one adult mouse.
12
T.A. Owen and L.C. Pan
Fig. 1.2 Osteoblastic differentiation of rat bone marrow cultures. Rat bone marrow cells were isolated and cultured under osteogenic differentiation conditions. a. Unmagnified view of differentiated osteoblastic colonies in a 12-well dish after 18 days in culture. These mineralized colonies are visible as opaque patches in the culture dish. b. Enlarged view of a mineralized osteoblastic colony photographed under a 4× objective.
5. Plate the isolated bone marrow cells into 6-well (or larger) dishes at 1.8 × 105 cells/cm2 in growth medium (see Note 12). The day of plating is called study day 0. Feed all wells again with growth medium on study day 4. There are usually not many attached cells visible prior to study day 4. 6. On study day 7 and every third day thereafter, feed all wells with differentiation medium. Small colonies of cells can be observed by study day 7 and by study day 10, some colonies will begin to form multilayered nodules of differentiated osteoblasts, becoming opaque as a result of mineral deposition. Fig. 1.2 shows a rat marrow culture with mineralized nodules on study day 18 (see Note 14).
3.4
Alkaline Phosphatase Enzyme Activity
1. Aspirate the medium from cultures to be analyzed and rinse once with PBS. Add alkaline phosphatase lysis buffer to the cells (50 µL/cm2 of cell growth area) and incubate on a rocking platform for 30 minutes at room temperature. Assay for alkaline phosphatase enzyme activity either immediately (below) or store the samples in lysis buffer at −80 °C. 2. To assay for alkaline phosphatase enzyme activity, thaw the samples at room temperature or use freshly lysed samples. Do not subject lysates to more than one freeze-thaw cycle prior to analysis. Unless known from previous experiments, pipet various volumes of lysate in triplicate, ranging from 1 to 50 µL, into a 96-well assay plate. Add 200 µL of pre-warmed (37 °C) alkaline phosphatase substrate (PNPP) to each well and read the change in absorbance as a function of time at 405 nm (see Notes 6 and 15). 3. Determine the protein concentration of each sample using a reagent such as the DC protein assay reagent (BioRad) that is compatible with the presence of 1% Triton X-100. Use this value to normalize the amount of enzyme activity determined in step 3.4.2 and express the data as substrate converted per minute per µg protein (10).
1 Culture of Rodent Cells
13
Fig. 1.3 Alkaline phosphatase enzyme histochemical staining during osteoblastic differentiation. Rat calvarial cells were isolated and cultured under osteogenic differentiation conditions. On the indicated days after plating, a dish was fixed and air dried. At the end of the timecourse, all dishes were stained for alkaline phosphatase enzyme activity. While alkaline phosphatase activity was already detectable at day 6 (red color), it increased in intensity with osteoblastic differentiation and the strongest staining was associated with the multilayered nodules of differentiated osteoblasts (See Color Plates).
3.5
Alkaline Phosphatase Histochemical Staining
1. Aspirate medium from the cultures and rinse once with PBS. Fix the cells for 10 minutes at room temperature using 100% methanol (−20 °C). Following incubation, remove the fixative and rinse once in PBS (see Notes 7 and 16). 2. Remove the PBS and add a sufficient volume of freshly prepared alkaline phosphatase staining solution to cover the dish. Incubate at 37 °C for up to 20 minutes, protected from light, until the desired intensity of staining is reached. Aspirate and discard the stain, rinse with water, and air dry (9) (Fig. 1.3).
3.6
von Kossa Staining for Assessment of Mineralization
1. Aspirate medium from the cultures and rinse once with PBS. Fix the cells for 10 minutes at room temperature using cold 100% methanol (see Note 7). Remove the fixative and rinse twice with water. It is critical to rinse with water prior to staining or else unacceptable background may occur due to precipitation of the stain. Fixed dishes may be stained immediately or stored dry at room temperature for staining at a later time. 2. Stain the cells by adding 5% AgNO3 (250 µL/cm2 of cell growth area). Develop stain by exposure to ultraviolet light until the mineralized nodules become dark brown to black in color (Fig. 1.4a). Traditionally, this has been done by placing the dishes containing AgNO3 solution in bright sunlight until the desired level of staining is obtained (11); however, use of a calibrated ultraviolet source gives more controlled and reproducible results. We routinely use a high-intensity ultraviolet light box designed for cross-linking nucleic acids to membranes (UV Stratalinker 1800, Stratagene) set to deliver 2.4 × 105 µJoules/cm2. 3. Following exposure to ultraviolet light, remove the AgNO3 stain and dispose of it as hazardous waste in accordance with institutional policies. Rinse the dishes twice with water and invert to air dry (see Note 17).
14
T.A. Owen and L.C. Pan
Fig. 1.4 Extracellular matrix mineralization during osteoblastic differentiation. Rat calvarial or bone marrow cells were isolated and cultured under osteogenic differentiation conditions. a. On the indicated days after plating, rat calvarial cultures were von Kossa stained and viewed by phase contrast optics using a 4× objective. Small mineralized nodules as well as unmineralized nodules were detectable by day 8. With time, the mineralized nodules increased in size and number. b. Unmagnified views of von Kossa (VK) stained cultures showing a differentiated rat calvarial culture (ROB) in a 12-well dish 12 days after plating (ROB) and a differentiated rat bone marrow culture in a 6-well dish 18 days after plating (BMC). c. Unmagnified view of mineralized nodules visualized by calcein incorporation. ROB and BMC cultures were stained with either calcein (left panel) or calcein blue (center and right panels) and photographed on top of an ultraviolet transilluminator (See Color Plates).
4. Images of von Kossa stained nodules can be captured for quantitation by photography on a white light box using the same manual exposure settings to maintain image uniformity throughout an experiment (Fig. 1.4b). Entire plates can be easily documented using a flatbed scanner. The parameter most frequently quantified is total nodule area using image analysis software such as ImagePro (Media Cybernetics, Bethesda, MD).
3.7
Calcein Staining for Assessment of Mineralization
1. To assess mineralization using calcein, add calcein or calcein blue from the 2.5 mg/mL stock solution to cell culture medium to achieve a final concentration of 25 µg/mL (1:100 dilution) and return cultures to the incubator for 16–24 hours.
1 Culture of Rodent Cells
15
If quantitation of mineralized nodule area is desired, the time of exposure should be kept constant for all samples being compared. 2. To terminate staining, aspirate the medium and rinse once with PBS. Fix the cells for 10 minutes at room temperature using 100% methanol (−20 °C) or other suitable fixative (see Note 7). Following fixation with methanol, invert and air dry the plates. If an aldehyde fixative is used, rinse the cultures once in PBS, and then invert to air dry. Plates may then be stored indefinitely at room temperature. Protection from strong light is advisable, even though calcein and calcein blue are relatively resistant to photobleaching. 3. Images of calcein or calcein blue fluorescence can be captured by photographing the stained dishes on an ultraviolet transilluminator using the same manual exposure settings to maintain image uniformity throughout an experiment (Fig. 1.4C). These images are also suitable for determination of total nodule area using image analysis software such as ImagePro (Media Cybernetics).
4
Notes
1. An alternative method to prepare trypsin/collagenase digestion cocktail is to dissolve 15 Wünsch units of collagenase P in 50 mL 0.25% trypsin-1 mM EDTA since this is the trypsin-EDTA solution commonly used in cell culture. The low concentration of EDTA in this solution does not practically affect the activity of the collagenase in this application. In many publications, collagenase P concentration has been specified in mass units, for example, 0.2 mg/mL. However, we have specified the collagenase concentration by activity units to facilitate use of commercial preparations with markedly different specific activities. Recommended sources for Collagenase P are Roche and Worthington. 2. All reagents should be prepared in tissue culture quality deionized (18 MΩ) water. All chemicals should be ACS (American Chemical Society) grade or higher or qualified for tissue culture use. 3. β-Glycerol phosphate (10 mM final concentration) may be substituted for the additional sodium phosphate. A stock solution of 1 M β-glycerol phosphate can be prepared in PBS or MEMα medium, sterilized by filtration through a 0.22µm filter, divided into 6-mL aliquots (5.5 mL is added to each 500-mL bottle of differentiation medium), and stored at −20 °C. 4. Fetal bovine serum is not a defined product and varies significantly with manufacturer and individual lots. Serum lots should be tested for their ability to support osteoblast differentiation based on mineralized bone nodule formation under these culture conditions. 5. Ascorbic acid is absolutely required for collagen synthesis and is essential for osteoblasts to lay down an organized collagenous extracellular matrix and form multilayered nodules. The basal medium, MEMα contains 50 µg/mL ascorbic acid and the addition of an extra 50 µg/mL ensures that ascorbate will not be limiting for the necessary collagen production. MEMα contains 1 mM sodium phosphate so that supplementation with an additional 2 mM sodium phosphate is sufficient to allow efficient mineralization of the collagenous extracellular matrix.
16
T.A. Owen and L.C. Pan
6. Alternative alkaline phosphate substrates may be utilized, depending on the level of sensitivity required and on the equipment available to read the assay plates. For example, a fluorescent substrate (AttoPhos®) is available from Promega (Madison, WI) and a chemiluminescent substrate (LumiPhos®) can be obtained from Pierce (Rockford, IL). Mature calvarial or bone marrow-derived osteoblast cultures produce high levels of alkaline phosphatase activity, so sensitivity is not generally an issue. As a rule, differentiated calvarial cultures will express significantly higher levels of alkaline phosphatase than will bone marrow-derived cultures. 7. Many alternative fixatives can be used prior to alkaline phosphatase histochemical staining or analysis of mineralization. Limited fixation with either 4% paraformaldehyde, made by diluting 10 mL of 32% paraformaldehyde stock (Electron Microscopy Sciences, Ft. Washington, PA) with 8 mL of 10× PBS and 62 mL of water and storing at room temperature or commercially prepared 10% formalin for 10 minutes at room temperature works well. These aldehyde fixatives are preferable in situations where organic solvents must be avoided, for example, to preserve lipid deposits for adipocyte staining. 8. Commercial alkaline phosphatase histochemical staining kits, for example, leukocyte alkaline phosphatase (kit 86-C) from Sigma Diagnostics are satisfactory and provide other colors of stained product. An analogous kit that produces a fluorescent product, ELF 97 Endogenous Phosphatase Detection Kit (Invitrogen kit E6601) can be substituted for the Napthol AS-MX phosphate and fast red in protocol 3.5.2 (12). 9. The viability of calvarial osteoblasts for several days when stored in this manner allows for shipping between laboratories or for additional analyses that may be needed to genotype or phenotype genetically modified mice. Although we have had considerable success with this option, it is possible that this step could introduce additional experimental variables related to differential cell survival. 10. All cell culture incubations are done at 37 °C in a 5% CO2 incubator. Trypan Blue exclusion does not accurately predict cell viability after the enzymatic digestion protocol, so it is advisable to culture the cells as described in step 11 to allow full recovery prior to initiating experiments. Relatively little proliferation has been observed during this first 24 hours of culture. Freshly isolated calvarial cells can be plated for experiments at this point without freezing. If this is desired, proceed directly to protocol step 3.2.3. 11. Mouse calvarial cultures show a marked loss of differentiation capacity with subculturing; therefore, it is advisable to use primary cultures if osteoblastic differentiation is the desired endpoint. In contrast, rat calvarial cells can be expanded for three to four population doublings without significant loss of the potential for mineralized nodule formation (13). 12. For experiments with osteoblastic differentiation as an endpoint, plating calvarial osteoblasts in wells <2 cm2 (24-well plate) or rodent bone marrow cells in wells <9 cm2 (6-well) will result in an uneven distribution of the osteoblastic precursor cells among the wells. As a result, there will be a high degree of variability in measurements of osteoblast phenotype endpoints such as alkaline
1 Culture of Rodent Cells
13.
14.
15.
16.
17.
17
phosphatase enzyme activity and mineral deposition. For immunohistochemical analysis, these cells may also be plated on glass microscope slides (which are designed for cell culture, for example, Chamber Slides, BD Falcon). The amount of the proximal end of the tibia or femur that must be removed depends upon the region occupied by trabecular bone, which is affected by age, hormone status, and other factors. If not removed, the network of trabecular bone will prevent extrusion of the marrow and reduce the yield of cells. If extensive trabecularization is visible at the cut end, remove additional bone until it is gone. Alternatively, a sharp scalpel can be used to cut mouse bones or a Dremel tool with a circular blade may be used to cut rat bones. A subpopulation of the bone marrow cells attached to cell culture plastic by study day 4 in this procedure are multipotent, with the potential to differentiate into either osteoblasts or adipocytes in response to experimental manipulation. Adipogenic agents may be added to the osteogenic differentiation media, allowing development of both cell types in the same culture dish (14). Adipocytes can be detected by staining of lipids by Oil Red O. If the same dishes are to be stained for both adipocytes and mineralized nodules, Oil Red O staining should be performed before von Kossa staining. If a plate reader with a kinetic reading mode is not available, use an endpoint assay. Make certain to visually monitor product formation and read the plate during the linear phase of the enzyme reaction (before substrate becomes limiting). Ideally, a baseline absorbance reading should be subtracted from the endpoint absorbance reading. If it is desirable to delay staining for alkaline phosphatase until a later time, dishes fixed in methanol can be air dried without rinsing and stored for >1 year at room temperature. Rehydrate with water for 10 minutes prior to staining as in step 3.5.2. Dishes fixed in paraformaldehyde should be rinsed once in PBS and then stored under PBS at 4 °C prior to staining as in step 3.5.2. Both von Kossa staining and alkaline phosphatase histochemical staining can be performed on the same culture dishes. Perform the alkaline phosphatase staining first, rinse the dishes thoroughly with water, and then perform von Kossa staining.
Acknowledgments The authors thank David Rowe, of the University of Connecticut Health Center, for the pOBCol3.6GFPtpz transgenic mice.
References 1. Bhargava, U., Bar-Lev, M., Bellows, C. G., et al. (1988). Ultrastructural analysis of bone nodules formed in vitro by isolated fetal rat calvaria cells. Bone 9, 155–163. 2. Ecarot-Charrier, B., Shepard, N., Charette, G., et al. (1988). Mineralization in osteoblast cultures: a light and electron microscopic study. Bone 9, 147–154. 3. Owen, T. A., Aronow, M., Shalhoub, V., et al. (1990). Progressive development of the rat osteoblast phenotype in vitro: reciprocal relationships in expression of genes associated
18
4.
5.
6. 7. 8. 9.
10. 11. 12. 13.
14. 15.
T.A. Owen and L.C. Pan with osteoblast proliferation and differentiation during formation of the bone extracellular matrix. J Cell Physiol 143, 420–430. Bellows, C. G., Aubin, J. E., Heersche, J. N. M., et al. (1986). Mineralized bone nodules formed in vitro from enzymically released rat calvaria cell populations. Calcif Tissue Int 38, 143–154. Scutt, A., Bertram, P. (1995). Bone marrow cells are targets for the anabolic actions of prostaglandin E2 on bone: induction of a transition from nonadherent to adherent osteoblast precursors. J Bone Min Res 10, 474–487. Friedenstein, A. J., Piatetzky-Shapiro, I. I., Petrakova, K. V. (1966). Osteogenesis in transplants of bone marrow cells. J Embryol Exp Morphol 16, 381–390. Peck, W. A., Birge, S. J. J., Fedak, S. A. (1964). Bone cells: biochemical and biological studies after enzymatic isolation. Science 146, 1476–1477. Wong, G., Cohn, D. V. (1974). Separation of parathyroid hormone and calcitonin-sensitive cells from non-responsive bone cells. Nature 252, 713–715. Aubin, J. E. (1999). Osteoprogenitor cell frequency in rat bone marrow stromal populations: role for heterotypic cell-cell interactions in osteoblast differentiation. J Cell Biochem 72, 396–410. Lowry, O. H., Roberts, N. R., Wu, M.-L., et al. (1954). The quantitative histochemistry of brain. II. enzyme measurements. J Biol Chem 207, 19–38. Clark, G. (ed.) (1981). Staining Procedures, 4th ed. Williams & Wilkins, Baltimore. Cox, W. G., Singer, V. L. (1999). A high-resolution, fluorescence-based method for localization of endogenous alkaline phosphatase activity. J Histochem Cytochem 47, 1443–1456. Bellows, C. G., Heersche, J. N. M., Aubin, J. E. (1990). Determination of the capacity for proliferation and differentiation of osteoprogenitor cells in the presence and absence of dexamethasone. Dev Biol 140, 132–138. Li, M., Pan, L. C., Simmons, H. A., et al. (2006). Surface-specific effects of a PPARγ agonist, darglitazone, on bone in mice. Bone 39, 796–806. Kalajzic, I., Kalajzic, Z., Kaliterna, M., et al. (2002). Use of type I collagen green fluorescent protein transgenes to identify subpopulations of cells at different stages of the osteoblast lineage. J Bone Min Res 17, 15–25.
Chapter 2
Osteoclast Culture and Resorption Assays Elizabeth W. Bradley and Merry Jo Oursler
Abstract Bone homeostasis depends on balanced bone deposition and bone resorption, which are mediated by osteoblasts and osteoclasts, respectively. The process of bone turnover requires the coordination of these cells. Changes in the ability of either cell type to perform its function results in pathological conditions such as osteoporosis and tumor-induced bone loss (osteolysis). The number of osteoclasts present at the site of bone remodeling as well as the activity of those osteoclasts the control amount of bone resorbed (1). Therefore, factors affecting overall numbers of osteoclasts and osteoclast activation are key to regulating bone loss. Osteoclast numbers are in part controlled by osteoclast differentiation from bone marrow precursors of the monocyte/macrophage lineage (2). Differentiation of these hematopoietic precursors into osteoclasts is supported by bone marrow stromal cell production of two cytokines, receptor activator of NF-κB ligand (RANKL) and macrophage colony stimulating factor (M-CSF), which are both necessary and sufficient to mediate osteoclast differentiation (3, 4). Although RANKL production by the stroma supports osteoclast differentiation, this process is antagonized by osteoprotogerin (OPG) production, which acts as a soluble decoy receptor for RANKL (5, 6). Mechanistic studies to elucidate the factors influencing bone metabolism necessitate in vitro studies of osteoclast differentiation, activation and survival. There are a number of in vitro methods used to culture and study osteoclasts, some of which are described in this chapter. Keywords Osteoclast, co-culture, RANKL, resorption assay, CD11b, CD14.
1
Introduction
Osteoclasts are the multinucleated cells of hematopoietic origin responsible for bone resorption associated with bone modeling and remodeling and, in pathological conditions, excess bone loss. Given our knowledge of the RANKL/OPG system, a number of in vitro culture systems can be utilized to obtain mature osteoclasts. From: Methods in Molecular Biology, Vol. 455: Osteoporosis: Methods and Protocols Edited by Jennifer J. Westendorf © Humana Press, Totowa, NJ
19
20
E.W. Bradley and M.J. Oursler
The first method for generation of mature osteoclasts from marrow or spleen precursors utilizes soluble forms of RANKL and M-CSF (6, 7). The second method involves a co-culture system wherein osteoclasts are differentiated from marrow or spleen precursors, in tandem culture with marrow stromal cells, frequently the ST2 cell line (8, 9). Soluble and/or membrane-bound RANKL and M-CSF are produced by the stromal cells and act as paracrine factors to promote precursor differentiation into mature osteoclasts. In addition to these two methods, osteoclasts can be differentiated from CD11b/CD14-positive precursors that are isolated from the monocyte/ macrophage pool. Finally, the immortalized RAW264.7 cell line can be used to obtain osteoclast-like cells. RAW264.7 cells were derived from mouse macrophages transformed with the Abelson murine leukemia virus (10). This system allows for generation of stromal cell-free osteoclast precursors using soluble RANKL supplementation only. All four culture systems can be employed to generate mature osteoclasts capable of resorbing bone; however, each has associated strengths and weaknesses. Both the co-culture-generated and RANKL/M-CSF-generated systems can be utilized to study osteoclast phenotypes associated with knockout and transgenic mice ex vivo. Thus, providing powerful tools to determine molecular mechanisms. However, as with many primary cell culture systems, gene delivery must be accomplished through a viral system in these two culture models.
2 2.1
Materials Isolation of Osteoclast Precursors from Bone Marrow and Spleen
1. Phenol red-free alpha modified Minimal Essential Medium (αMEM) with L-glutamine and sodium pyruvate supplemented with 10% fetal bovine serum (FBS) (Hyclone, Logan, UT) and 1% antibiotic/antimycotic. 2. M-CSF (R&D Systems, Minneapolis, MN) dissolved to 2.5 µg/mL in PBS with 0.1% bovine serum albumin (BSA) stored in aliquots at −20 °C until diluted (100×) in culture medium. 3. Red blood cell (RBC) lysis buffer (eBioscience, San Diego, CA). 4. 100-mm tissue culture dishes. 5. 27-gauge needles. 6. 3-cc syringe. 7. Phosphate buffered saline (PBS). 8. 70% ethanol. 9. Dissecting scissors and forceps in 70% ethanol. 10. Cell scrapers. 11. Freezing medium: 88% FBS, 12% DMSO.
2 Osteoclast Culture and Resorption Assays
2.2
21
Generation of Osteoclasts Differentiated with Soluble RANKL and M-CSF
1. Phenol red-free αMEM with L-glutamine and sodium pyruvate supplemented with 10% FBS and 1% antimycotic/antibiotic solution. 2. RANKL (Chemicon International, Tamecula, CA) dissolved to 3.0–10.0 µg/mL in water, stored in aliquots at −80 °C until diluted (100x) in culture medium. 3. M-CSF dissolved to 2.5 µg/mL in PBS with 0.1% BSA stored in aliquots at −20 °C until diluted (100×) in culture medium.
2.3
Co-Culture System
2.3.1
Generation of Osteoclasts Co-Cultured with the ST2 Stromal Cell Line
1. ST2 stromal cell line expanded and stored under liquid nitrogen in FBS supplemented with 12% DMSO until use. 2. αMEM with L-glutamine and sodium pyruvate supplemented with 10% FBS and 1% antimycotic/antibiotic solution. 3. 10× Trypsin/EDTA diluted to 1X in sterile PBS, aliquoted and stored at −20 °C. 4. Vitamin D3 dissolved in ethanol to 10−3 M, stored in aliquots in amber tubes at −20 °C and diluted 1:1000 to obtain a working concentration of 10−7 M in culture medium. 5. Concentrated stock solution of 10−4 M dexamethasone prepared in ethanol, stored in aliquots at −20 °C and diluted 1:1,000 in culture medium to obtain a working concentration of 10−7 M in culture medium. 6. Fresh media: Base media supplemented with (12) concentrated stock solution of biotin (dissolved in 2 M NH4OH to a concentration of 5 mg/mL, then diluted 1:25 in water and aliquoted for daily use, stored at −80 °C and diluted to a final concentration of 100 ng/mL in base medium), 200 mM L-glutamine diluted to a final concentration of 200 µM in base medium, 1 mM sodium pyruvate diluted to a final concentration of 1 µM in base medium, concentrated stock solution of lipoic acid dissolved in ethanol to a concentration of 50 mg/mL, then diluted 1:25 in ethanol and aliquoted for daily use, stored at −80 °C and diluted to a final concentration of 200 ng/mL in base medium, vitamin B12 prepared in water to a stock concentration of 25 mg/mL, aliquoted and stored at −80 °C and diluted to a final concentration of 1.36 µg/mL in base medium, ascorbic acid dissolved in water to a stock concentration of 5 mg/mL, aliquoted and stored at −80 °C, then diluted in base medium for a final concentration of 50 µg/mL.
22
E.W. Bradley and M.J. Oursler
2.3.2 1. 2. 3. 4.
Enzymatic Purification of Co-Cultured Osteoclasts
Ham’s F12 media with sodium pyruvate and L-glutamine. Type III collagenase (Worthington Biochemicals, Lakewood, NJ). Dispase (Boehringer Mannheim, Framingham, MA). Transfer pipettes.
2.4
Generation of Osteoclasts from CD11b/CD14-Positive Cells Sorted Using FACS, MACS or Magnetic Beads
2.4.1
Antibody Sorting Using FACS
1. 2. 3. 4. 5.
Sterile PBS. 100-mm tissue culture dishes. Staining buffer: Sterile PBS supplemented with 1% BSA. PE (or another suitable fluorophore)-conjugated CD11b or CD14 antibody. Phenol red-free αMEM with L-glutamine and sodium pyruvate supplemented with 10% FBS and 1% antimycotic/antibiotic solution. 6. RANKL dissolved to 3.0–10.0 µg/mL in water, stored in aliquots at −80 °C until diluted (100×) in culture medium. 7. M-CSF dissolved to 2.5 µg/mL in PBS with 0.1% BSA stored in aliquots at −20 °C until diluted (100×) in culture medium. 2.4.2 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12.
Antibody Sorting Using MACS
Sterile PBS. 0.5 M EDTA. Sodium azide. Rinsing solution: PBS supplemented with 4 mM EDTA and 0.9% (w/v) sodium azide. Running buffer: Rinsing solution supplemented with 0.5% BSA. MACS colloidal microbeads conjugated with an anti-CD11b or CD14 antibody (Miltenyi Biotech, Auburn, CA). Magnetic cell separator and column adaptor (Miltenyi Biotec). Nylon filters (Milenyi Biotec). Positive selection columns (Miltenyi Biotech). Phenol red-free αMEM with L-glutamine and sodium pyruvate supplemented with 10% FBS and 1% antimycotic/antibiotic solution. RANKL dissolved to 3.0–10.0 µg/mL in water, stored in aliquots at −80 °C until diluted (100×) in culture medium. M-CSF dissolved to 2.5 µg/mL in PBS with 0.1% BSA stored in aliquots at −20 °C until diluted (100×) in culture medium.
2 Osteoclast Culture and Resorption Assays
2.4.3
23
Magnetic Bead Isolation of CD14-Positive Human Osteoclast Precursors
1. Sterile PBS. 2. Anti-CD14 primary antibody coated Dyanbeads M-450 and magnetic separator (Invitrogen). 3. Serum (Biochrom, Cambridge, UK). 4. Ficoll-Paque. 5. Phenol red-free αMEM with L-glutamine and sodium pyruvate supplemented with 10% FBS and 1% antimycotic/antibiotic solution. 6. RANKL (Chemicon International, Tamecula, CA) dissolved to 3.0–10.0 µg/mL in water, stored in aliquots at −80 °C until diluted (100×) in culture medium 7. M-CSF dissolved to 2.5 µg/mL in PBS with 0.1% BSA stored in aliquots at −20 °C until diluted (100×) in culture medium.
2.5
Generation of Multinucleated Cells (MNCs) from the RAW264.7 Cell Line
1. RAW264.7 cell line (ATCC, Manassas, VA). 2. Dulbeco’s Modified MEM supplemented with 10% FBS and 1% antimycotic/ antibiotic solution. 3. RANKL dissolved to 3.0–10.0 µg/mL in water, stored in aliquots at −80 °C until diluted (100×) in culture medium. 4. Cell lifters.
2.6 1. 2. 3. 4.
Isolation of Murine Osteoclasts Differentiated In Vivo
70% ethanol. Dissecting forceps and scissors in 70% ethanol. αMEM supplemented with 10% FBS and 1% antibiotic/antimycotic. 100-mm tissue culture dish.
2.7
Resorption Assays
2.7.1
Dentin or Bone Slice Resorption Assay
1. Dentin or bone. 2. Beuhler isomet low speed saw. 3. Diamond wafering blade.
24
E.W. Bradley and M.J. Oursler
4. Paper hole punch. 5. 70% ethanol. 6. 1% paraformaldehyde. 7. Phenol red-free αMEM with L-glutamine and sodium pyruvate supplemented with 10% FBS and 1% antimycotic/antibiotic solution. 8. RANKL dissolved to 3.0–10.0 µg/mL in water, stored in aliquots at −80 °C until diluted (100×) in culture medium. 9. M-CSF dissolved to 2.5 µg/mL in PBS with 0.1% BSA stored in aliquots at −20 °C until diluted (100×) in culture medium. 10. Concentrated ammonium hydroxide. 11. Toludine blue. 12. Sodium borate (O-certified, Sigma-Aldrich). 13. Whatman #1 filter paper.
2.7.2
Artificial Matrix Coating
1. BD BioCoat Osteologic Bone Cell Culture System (BD Biosciences, San Jose, CA). 2. Osteometric’s OsteoMeasure or Bioquant’s OSTEO II system. 3. Femur wash medium: αMEM with L-glutamine and sodium pyruvate supplemented with 15% FBS, 1 mg/mL penicillin G, 0.5 mg/mL gentamicin, 3 mg/mL Fungizone. 4. Osteoclast culture medium: αMEM with L-glutamine and sodium pyruvate supplemented with 15% FBS, 0.1 mg/mL penicillin G, 0.05 mg/mL gentamicin, 300 ng/mL fungizone, 0.28 mM L-ascorbic acid 2-phosphate, 10 nM dexamethasone. 5. Bleach solution: 6% NaOCl, 5.2% NaCl.
2.7.3 1. 2. 3. 4. 5. 6. 7. 6.
Calvarial Organ Culture
DMEM. Horse serum (Hyclone). L-glutamine. Sodium heparin (Elkin-Sinn, Cherry Hill, NJ). Penicillin/streptomycin. 17 × 100 mm polypropylene round bottom tubes. Roller drum (New Brunswick Scientific Co., Inc., Edison, NJ). Ciba-Corning 634 ISE Ca2+/pH Analyzer (Chiron Diagnostics Corp., East Walpole, MA).
2.7.4
Bone Particle Assay
1. 3H labeled proline. 2. 70% ethanol.
2 Osteoclast Culture and Resorption Assays
3 3.1
25
Methods Isolation of Osteoclast Precursors from Bone Marrow and Spleen
1. Sacrifice mice (see Note 1) by carbon dioxide asphyxiation. 2. Sterilize mouse epidermis with 70% ethanol. 3. Remove the spleen by snipping a small whole in the dorsal skin of the mouse between hind limbs and tearing the skin anteriorly to expose the underlying organs. The spleen should be visible on the left hand side of the animal and appear as a tongue-shaped red organ. Snip a small hole in the fascia and remove the spleen using sterile forceps. Place removed spleen in sterile PBS on ice. 4. Remove the femur and tibia by skinning both hind legs using sterilized dissecting scissors and forceps by cutting through the skin surrounding the hip joint. Tear the skin posteriorly toward the feet to remove. De-articulate the hip joint and remove the muscle surrounding the femur and tibia. Place clean bones in sterile PBS on ice. 5. Remove precursors form the spleen by placing the organ in a 100-mm tissue culture dish with sterile PBS. Snip the small end of the organ with sterile scissors and forceps. Milk the cells from the spleen into PBS using a sterile cell scraper. 6. Remove the marrow precursors by snipping the ends of each bone to expose the marrow cavity using sterile scissors and forceps. Using a 27-gauge needle and syringe, flush the marrow cavity with sterile PBS to remove precursors. 7. Pellet the cells by centrifugation at 1,000 g for 10 minutes. Aspirate the supernatant and suspend the cells in 5 mL RBC lysis buffer for 5 minutes (see Note 2). Pellet cells by centrifugation at 1,000 g for 10 minutes. 8. Plate the obtained cells in 100-mm dishes at a density of 2.9 × 107 cells in 8 mL in αMEM supplemented with 25 ng/mL M-CSF (80-µL) per dish. Culture cells overnight to remove contaminating stroma (see Note 3). 9. Remove the non-adherent cells from each dish for use in subsequent experiments or store precursors for future use under liquid nitrogen in freezing medium.
3.2
Generation of Osteoclasts Differentiated with Soluble RANKL and M-CSF
The first method described here for generation of mature osteoclasts from marrow or spleen precursors utilizes soluble forms of RANKL and M-CSF (11). Because precursors are supplemented with RANKL and M-CSF, the use of a stromal cell line to support differentiation is not required. Therefore, this method generates a pure population of mature osteoclasts derived from primary tissue origin. 1. Collect non-adherent precursors as described. 2. Culture precursors at a density of 4 × 105 cells/cm2 in 1 mL phenol red-free αMEM/cm2 supplemented with 25 ng/mL M-CSF (10 µL/mL of stock) and 30–100 ng/ml RANKL (10 µL/mL of stock) (see Note 4).
26
E.W. Bradley and M.J. Oursler
3. After 3 days in culture, feed differentiating osteoclasts. Aspirate medium and replace with 1 mL phenol red-free αMEM/cm2 supplemented with 25 ng/mL M-CSF and 30–100 ng/mL RANKL. 4. Treat mature osteoclasts (see Note 5) as per experimental design.
3.3
Generation of Osteoclasts Co-Cultured with the ST2 Stromal Cell Line
The co-culture system differentiates osteoclasts through bone marrow stromal cellmediated production of RANKL and M-CSF by utilizing the ST2 cell line. This system mimics the physiological conditions for osteoclast differentiation and maintains cell–cell contact between the osteoclast precursors and the stroma. The inherent drawback to this system is the co-culture with a stromal cell line. To generate a pure population of osteoclasts, these cultures must be enzymatically treated to remove contaminating stromal cells.
3.3.1
Generation of Osteoclasts Co-Cultured with the ST2 Stromal Cell Line
1. Culture ST2s in αMEM. To passage cells, rinse each flask with sterile PBS and add 2 mL 1× trypsin/EDTA solution. Incubate 10 minutes at 37 °C and dislodge cells. Reseed cells in αMEM until use (see Note 6). 2. Plate ST2s at a density of 1.6 × 105 cells/cm2 the day prior to obtaining nonadherent precursors (day –1) in αMEM (see Note 3). 3. Add non-adherent precursors (day 0) to the ST2s in culture at a density of 1.18 × 105 cells/cm2 in 1 mL fresh medium/cm2 prepared the day of use supplemented with 10−7 M dexamethasone and 10−7 M vitamin D3 (1:1,000 dilutions of stock solutions). 4. Feed cells with 1 mL/cm2 fresh medium supplemented with dexamethasone and vitamin D3 as above prepared the day of use on day 3, 6, and 9 of culture (see Note 7). 5. Treat mature osteoclasts as per experimental design or proceed to enzymatic purification protocol.
3.3.2
Enzymatic Purification of Co-Cultured Osteoclasts
1. Prepare 0.5 mL/cm2 of a 0.2% collagenase (w/v) solution in Ham’s F12 (see Note 8). 2. Aspirate culture medium from co-cultured osteoclasts and rinse twice with PBS. Add 0.5 mL 0.2% collagenase/cm2 to each well. Incubate for 30 minutes at 37 °C. 3. Prepare 0.5 mL/cm2 of a 0.2% Dispase (w/v) solution in Ham’s F12 (see Note 8). 4. Agitate medium with a transfer pipette and aspirate. Rinse each well twice with PBS (see Note 9).
2 Osteoclast Culture and Resorption Assays
27
5. Add 0.5 mL/cm2 0.2% Dispase to each well. Incubate 20 minutes at 37 °C (see Note 10). 6. Remove stromal cells through agitation of the media with a transfer pipette. Ensure all mononuclear cells have lifted and treat purified osteoclasts as per experimental design.
3.4
Generation of Osteoclasts from CD11b/CD14-Positive Cells Sorted Using FACS, MACS, or Magnetic Beads
Osteoclasts are derived from marrow/spleen precursors of the monocyte/macrophage lineage. These precursors express two cell-surface markers, CD11b and CD14, which can be used to enrich a cell mixture for osteoclast precursors. Sorting of these CD11b/CD14-positive precursors from mouse or human origin can be accomplished using a number of methods including fluorescence-activated cell sorting (FACS), magnetic-assisted cell sorting (MACS) or magnetic bead-based sorting described here. The advantage of FACS over MACS or magnetic beads is that it is possible to separate cell populations with high versus low levels of surface antigen expression.
3.4.1
Antibody Sorting Using FACS
1. Collect marrow or spleen precursors as described above and omit overnight culture in M-CSF. 2. Pellet obtained osteoclast precursors by centrifugation at 400 g for 5 minutes. 3. Resuspend cells in 2 mL staining buffer. Aliquot 1 × 106 cells for use as a negative control for cell sorting. Pellet the remaining cells by centrifugation at 400 g for 5 minutes. 4. Dilute the primary PE conjugated CD11b or CD14 antibody to a concentration of 0.2 µg/uL in staining buffer. 5. Resuspend pelleted cells in 200 µL diluted primary antibody and incubate cells for 30 minutes in the presence of antibody at 4 °C. 6. Wash antibody-labeled cells three times. Resuspend in 2 mL staining buffer and pellet the cells by centrifugation at 400 g for 5 minutes. 7. Resuspend washed cells to a concentration of 4 × 106 cells/mL staining buffer and sort CD11b- or CD14-positive cells using FACS. 8. Culture obtained CD11b- or CD14-positive cells at a density of 2.9 × 107 cells in a 100-mm tissue culture dish overnight in αMEM supplemented with 25 ng/ mL M-CSF (see Note 11). 9. Collect the non-adherent cells for use in the generation of osteoclasts using the co-culture method or supplementation with RANKL and M-CSF.
28
3.4.2
E.W. Bradley and M.J. Oursler
Antibody Sorting Using MACS
1. Isolate marrow/spleen precursors as described above and omit overnight culture in M-CSF (see Note 12). 2. Pellet cells by centrifugation at 100 g for 10 minutes and resuspend cells in 80 µL running buffer per 1 × 107 cells. 3. Add 20 µL MACS microbeads per 1 × 107 cells (see Note 12) and incubate cells at 4 °C for 20 minutes—do not incubate cells on ice. 4. Pellet cells by centrifugation at 1,000 g for 10 minutes and remove supernatant. 5. Resuspend pelleted cells in 1 mL room temperature running buffer. 6. Assemble the positive selection column according to manufacturer’s instructions and place the column in the magnetic field of the MACS separator. 7. Wash the column with running buffer and apply the flow resistor to the threeway-stopcock of the column. 8. Apply the cell suspension to the column and allow negative cells to pass through column. Rinse column three times with 0.5 mL running buffer. 9. Remove the column from the MACS separator and place column in a collection tube. Add 1 mL running buffer to the column to isolate the CD11b- or CD-14positive cells. 10. Plate the obtained cells in 100-mm dishes at a density of 2.9 × 107 cells in 8 mL in αMEM supplemented with 25 ng/mL M-CSF (80 µL) per dish. Culture cells overnight (see Note 11). 11. Remove the non-adherent cells from each dish for use in subsequent experiments or store precursors for future use under liquid nitrogen in freezing medium.
3.4.3
Magnetic Bead Isolation of CD14-Positive Human Osteoclast Precursors
1. Obtain human peripheral blood and dilute 1:1 with sterile PBS. 2. Pipette 25 mL of blood/PBS mixture on top of 25 mL Ficoll-Paque in a 50-mL conical centrifuge tube. 3. Centrifuge at 707 g for 12 minutes and collect lymphocytes present at the bloodFicoll interface. Wash collected lymphocytes two times with 5 °C PBS and pellet cells by centrifugation at 707 g for 12 minutes. 4. Resuspend pelleted cells in PBS supplemented with 2% serum. 5. Wash 250 µL Dynabeads per 50 mL thrombocyte concentrate three times with 5 °C PBS by placing beads in magnetic separator for 2 minutes. Discard supernatant and remove beads from magnetic separator. 6. Add isolated lymphocytes to washed beads and mix end-over-end for 20 minutes at 4 °C. 7. Place sample in magnetic separator for 2 minutes and discard supernatant.
2 Osteoclast Culture and Resorption Assays
29
8. Remove sample from magnetic separator and wash cells three times. Gently resuspend cell/bead mixture in 5 mL PBS with 2% serum. Place sample in magnetic separator for 2 minutes. 9. Plate the obtained cells in 100-mm dishes at a density of 2.9 × 107 cells in 8 mL in αMEM supplemented with 25 ng/mL M-CSF (80 µL) per dish. Culture cells overnight (see Note 11). 10. Culture isolated CD14-positive cells in αMEM supplemented with 25 ng/mL M-CSF and 30 ng/mL RANKL at a density of 175 cells/cm2.
3.5
Generation of Multinucleated Cells (MNCs) from the RAW264.7 Cell Line
An additional method for generating multinucleated cells utilizes the RAW264.7 cell line. As an osteoclast precursor cell line, the RAW264.7 cells can be transfected using standard molecular techniques prior to differentiation. This system allows for generation of stromal cell-free MNC precursors. In addition, RAW264.7 cells need only be supplemented with RANKL to support MNC differentiation. Two primary weaknesses to this system, however, need consideration. One weakness is the use of a cell line in place of marrow precursors derived from primary tissue origin. The second drawback concerns the ability of these cells to differentiate in the absence of M-CSF. RAW264.7 cells either produce and secrete M-CSF, or contain a defect in the cellular signaling associated with this pathway. We have unpublished observations showing that inhibition of MEK1/2 increases RAWgenerated MNC survival, which is contrary to the known role for MEK in the promotion of marrow and spleen in vitro–derived osteoclasts. Thus, osteoclasts generated from RAW264.7 cells are not a suitable system for all studies. 1. Culture RAW264.7 cells in DMEM. To passage cells, scrape each flask with a cell lifter (see Note 13). 2. To differentiate osteoclasts from RAW264.7 cells, plate 2,000–5,000 cells/cm2 in 1 mL/cm2 DMEM supplemented with 100 ng/mL RANKL (10 µL/mL of stock). 3. Feed everyday in culture with 1 mL/cm2 DMEM supplemented with recombinant RANKL. Mature MNCs are present on day 5 of culture.
3.6
Isolation of Murine Osteoclasts Differentiated In Vivo
1. Aseptically remove the femur and tibia, skin both hind legs using sterilized dissecting scissors and forceps by cutting through the skin surrounding the hip joint. Tear the skin posteriorly toward the feet to remove. De-articulate the hip joint and remove the muscle surrounding the femur and tibia.
30
E.W. Bradley and M.J. Oursler
2. Place the bones in αMEM and mince with sterile scissors. Gently pipette resulting bone pieces to release the mature osteoclasts from the bone surface (13). 3. Incubate the resulting cell suspension overnight in αMEM. 4. After overnight incubation, remove non-adherent cells by aspiration. 5. Enzymatically purify the mature osteoclasts as described above to remove stromal cells and treat isolated mature osteoclasts as per experimental design.
3.7
Resorption Assays
Most reports have found that it is difficult to recover in vitro generated osteoclasts from plastic surfaces for re-plating on resorbable surfaces. For this reason, precursors are usually plated on resorbable surfaces and the total amount of resorption monitored. Therefore, data regarding cumulative responses must be viewed as the results of potential effects on differentiation, activity, and survival and not exclusively the result of activity. To some extent, this can be circumvented in the case of measurement of soluble resorption products if normalized to the number of osteoclasts present on the bone surface and if the cultures are well rinsed prior to addition of test substances. In the case of in vivo differentiated osteoclasts, these can be directly seeded onto the bone. This will allow for direct measurement of influences on mature cell functions and/or survival. The caveat with this approach is that these cell preparations will not be as pure, with multiple contaminating cell types that will likely influence responses.
3.7.1
Preparation on Dentin or Bone Slices
1. Slice cleaned bone or dentin with a Buehler Isomet low speed saw and diamond wafering blade to a thickness of approximately 0.2 mm. Obtain circles with a paper hole punch. Soak in 70% ethanol overnight to sterilize. 2. Rinse slices repeatedly with sterile water and air-dry in a laminar flow hood. 3. Place slices into 48-well plates and add non-adherent precursor cells as in the above RANKL/M-CSF culture protocol. 4. On day 3, move the slices to a six-well plate for final differentiation (see Note 14). 5. Depending on the experimental design, treatment will commence when precursors are plated, on day 3, or when the osteoclasts are mature on day 4. For in vivo generated osteoclasts directly seeded onto bone, after an overnight incubation, slices with adherent osteoclasts are transferred to six-well plates and treatment commenced (see Note 14). 6. Fix slices with 1% paraformaldehyde in PBS and TRAP stain. Count osteoclasts using a light microscope. 7. Remove cells and debris by sonicating twice with 15-second bursts in concentrated ammonium hydroxide. 8. Prepare fresh stain each time by dissolving 0.1 g sodium borate in 10 mL water and then adding 0.1 g Toluidine Blue. Stir until completely in solution.
2 Osteoclast Culture and Resorption Assays
9.
10. 11. 12. 13.
31
Filter through Whatman #1 filter paper. Stain may be stored for 3 weeks with filtering prior to use. Prepare bones for staining by adding 200 mL water to bones. Incubate at room temperate for 1 hour. Wash a second time for 1 hour with fresh water. Store overnight at 4 °C if desired. Rinse the slices twice with water with 15-second bursts in the sonicator. Immerse the bone slices in stain. Stain the bone slices for 30 seconds and then rinse three times in water. Blot bone slices on Whatman #1 filter paper. Air dry bone slices. Image the samples. There are three visual approaches that are used to evaluate both dentin and bone slices. These visual methods are using: scanning EM, confocal microscopy, or light microscopy to determine the size, depth, and number of pits (14). These assays are time consuming and labor intensive. In addition, the existing bone structure can make recognizing and scoring pits on bone slices difficult. A further drawback with this approach is that it requires terminating the experiment and does not allow for analysis of the same cells over time. Assays that assess bone resorption by measuring the release of the collagen degradation products into the media also allow for measurement of mature cell activation separate from influences on differentiation. The CrossLaps for Culture ELISA from Nordic Biosciences can be used to detect collagen type I fragments generated by resorption of either bone or dentin slices. When paired with assessment of osteoclast numbers, this will provide information on osteoclast activity independent of osteoclast differentiation effects (see Note 15).
3.7.2
Artificial Matrix Coating
Ceramic biomaterial substrates are now being used by some laboratories in place of biologically derived substrates such as dentin or ivory slices. The BD BioCoat Osteologic Bone Cell Culture System utilizes sub-micron synthetic proprietary calcium phosphate 0.6-µM thick films coated onto various culture vessels. Given the difficulties with lifting and replating mature osteoclasts, this system also suffers from the disadvantage of requiring that osteoclast precursors be plated on the matrix prior to differentiation. Thus, data will therefore be the sum total of differentiation and activation. In addition, there is some concern as this is an artificial matrix and may not accurately reflect how osteoclasts will perform on bone surfaces. Moreover, the matrix is so thin that it can be dissolved in a random pattern, creating false indications of resorption. The analysis is a visual assessment of removal of a surface film. Thus, quantitation requires use of an image quantification system such as Osteometric’s OsteoMeasure or Bioquant’s OSTEO II (see Note 16). 1. Remove femurs from a young mouse following protocol outlined above. 2. Wash femurs 4× with 30–40 mL femur wash.
32
E.W. Bradley and M.J. Oursler
3. Wash femurs 1× with 30–40 mL osteoclast medium. 4. Remove the marrow precursors by snipping the ends of each bone to expose the marrow cavity using sterile scissors and forceps. Using a 27-gauge needle and syringe, flush the marrow cavity with 10 mL osteoclast medium to remove precursors. Mix cell suspension. 5. Add 1 ml of cell suspension to each well. Incubate overnight. 6. Replace medium with 1 mL osteoclast medium after 24 hours. 7. Feed every 3 days with osteoclast medium until day 8–10. 8. Aspirate medium and rinse cells with water. 9. Add 1 mL bleach solution to each well. Pipette solution to remove cells. Incubate at room temperature for 5 minutes. 10. Wash 3× with water and air dry. Examine for resorption using microscopy. 3.7.3
Calvarial Organ Culture
1. Dissect calvariae from mouse pups (aged 3–5 days), remove soft tissue, and cut in half along the sagittal suture (15, 16). 2. Place half calvariae in 17 × 100 mm polypropylene round bottom tubes containing 1 mL of DMEM supplemented with 15% heat-inactivated horse serum, 2.8 mM L-glutamine, 10 units/mL sodium heparin, and penicillin/streptomycin. 3. Gas tubes with 50% oxygen, 5% carbon dioxide, and 45% nitrogen, seal and incubate for 24 hours at 37 °C on a Roller Drum. 4. After 72 hours in culture, collect medium, and analyze ionized calcium is using a Ciba-Corning 634 ISE Ca2+/pH Analyzer. 3.7.4
Bone Particle Assay
This is an approach that has been in use for many years but is not routinely used today. 1. Label bone in vivo by injecting animals, such as pregnant mice, with 3H proline. 2. Harvest, clean, and pulverize the long bones. Alternatively, aseptically remove and culture bones with 3H proline for 24 hours followed by 24-hour incubation without radioactivity to permit the 3H proline to either wash out or incorporate into the matrix. 3. Obtain large bone particles, which cannot be readily phagocytosed, by sieving the bones. 4. Sterilize by suspension in 70% ethanol. Rinse with sterile water. 5. Resuspend bone particles in sterile buffer and add to the cells (see Notes 17 and 18). 6. Monitor the release of 3H with a scintillation counter (see Note 18). 3.8.5
Long Bone Organ Culture
Radiolabeled long bones are obtained as outlined for bone particle generation and the intact bone is cultured. Release of 3H into the culture media is measured as above (see Note 18).
2 Osteoclast Culture and Resorption Assays
4
33
Notes
1. Approximately 5.0 × 109 total cells per mouse will be obtained with a corresponding 75% recovery of non-adherent cells. 2. When obtaining precursors for use in co-culture with the ST2 cell line, lysis of the red blood cells is not required. 3. When co-culturing precursors along with the ST2 stromal cell line, overnight removal of the bone marrow stromal cells is not required. However, if cells of mixed genotype are utilized, removal of the bone marrow stromal cells by overnight incubation of the osteoclasts precursors may be required. 4. The concentration of recombinant RANKL may be lowered to slow osteoclast differentiation until day 7 if desired. 5. Mature osteoclasts differentiated with soluble RANKL and M-CSF are present on day 4 of culture and persist through day 6, at which point viability greatly diminishes. 6. Do not allow ST2s in culture to reach confluence, which promotes adipogenesis in this cell line. The ST2 cell line can be passaged approximately 20 times before support of osteoclast differentiation severely declines. 7. Osteoclasts differentiated using this co-culture method will reach maturity at day 6 of culture and will persist until day 13, after which cell viability will be significantly lost. 8. During the enzymatic purification of osteoclasts, pre-loading of cell-permeable treatments can be accomplished. 9. Efficacy of stromal cell removal significantly decreases when collagenase and Dispase are combined. Ensure each well is rinsed between treatments. 10. The ST2 monolayer will appear as a white sheet when cells are lifting. Ensure that this sheet of cells lifts through agitation of the medium. If cells are not lifting after 20 minutes of incubation, add additional incubation in the 0.2% Dispase solution at 37 °C. 11. Although it would be assumed that this step would not be necessary for sorted cells, we have found that this step improves the purity of the final osteoclast cultures. 12. If cell clumps are visible, use the nylon filter to prevent clogging. 13. RAW264.7 cells are trypsin-resistant. Do not allow these cells to reach confluence, as this causes a loss of osteoclast differentiation capacity. 14. The rationale for this is that, once differentiation is complete, the calcium released by resorption will be detrimental to the cells. 15. A minimum of 50 pits per slice is needed in the control groups. This assay is best when the pits are small round circles and tunneling is at a minimum. Thus, osteoclasts are only allowed to resorb for a short period of time. Tunneling of osteoclasts makes it difficult to distinguish between the activities of one or two osteoclasts and increases the variability of the assay. 16. Osteoclasts plated on this surface cannot be TRAP stained as the acidic conditions remove the matrix (R. Galvin, personal communication). 17. The advantages of this approach are that the particles can be added once the cells are mature while they are attached to the plastic surface, results can be normalized
34
E.W. Bradley and M.J. Oursler
to cell numbers, and the same cells can be analyzed over time. The disadvantage of this system, apart from the tediousness of obtained radio-labeled bone, is that one cannot be certain that released radio-label is present in the culture supernatant because of phagocytosis and intracellular degradation. 18. Alternatively, the radio-labeling step can be eliminated and medium can be collected and ionized calcium analyzed as in the calvarial organ culture. These approaches allow the evaluation of bone resorption within the full context of the dynamic population of the cells within bone. Thus, although this is an advantage in relating effects to in vivo responses, it is recognized that the cells responding to a particular stimulus remain unclear in these studies. Acknowledgments The authors thank R. Galvin and E. Greenfield for their helpful insight. Funding sources are NIH grants DE14689 and AR52004 and the Mayo Clinic Foundation.
References 1. Suda, T., et al. (1997) Regulation of osteoclast function. J Bone Miner Res 12, 869–879. 2. Udagawa, N., et al. (1990) Origin of osteoclasts: mature monocytes and macrophages are capable of differentiating into osteoclasts under a suitable microenvironment prepared by bone marrow-derived stromal cells. Proc Natl Acad Sci U S A 87, 7260–7264. 3. Arai, F., et al. (1999) Commitment and differentiation of osteoclast precursor cells by the sequential expression of c-Fms and receptor activator of nuclear factor kappaB (RANK) receptors. J Exp Med 190, 1741–1754. 4. Wong, B. R., et al. (1997) TRANCE is a novel ligand of the tumor necrosis factor receptor family that activates c-Jun N-terminal kinase in T cells. J Biol Chem 272, 25190–25194. 5. Simonet, W. S., et al. (1997) Osteoprotegerin: a novel secreted protein involved in the regulation of bone density. Cell 89, 309–319. 6. Yasuda, H., et al. (1998) Osteoclast differentiation factor is a ligand for osteoprotegerin/osteoclastogenesis-inhibitory factor and is identical to TRANCE/RANKL. Proc Natl Acad Sci U S A 95, 3597–3602. 7. Lacey, D. L., et al. (1998) Osteoprotegerin ligand is a cytokine that regulates osteoclast differentiation and activation. Cell 93, 165–176. 8. Udagawa, N., et al. (1989) The bone marrow-derived stromal cell lines MC3T3-G2/PA6 and ST2 support osteoclast-like cell differentiation in cocultures with mouse spleen cells. Endocrinology 125, 1805–1813. 9. Takahashi, N., et al. (1988) Osteoblastic cells are involved in osteoclast formation. Endocrinology 123, 2600–2602. 10. Raschke, W. C., et al. (1978) Functional macrophage cell lines transformed by Abelson leukemia virus. Cell 15, 261–267. 11. Sells Galvin, R. J., et al. (1999) TGF-beta enhances osteoclast differentiation in hematopoietic cell cultures stimulated with RANKL and M-CSF. Biochem Biophys Res Commun 265, 233–239. 12. Ragab, A. A., et al. (1998) Osteoclast differentiation requires ascorbic acid. J Bone Miner Res 13, 970–977. 13. Chambers, T. J., et al. (1984) Resorption of bone by isolated rabbit osteoclasts. J Cell Sci 66, 383–399.
2 Osteoclast Culture and Resorption Assays
35
14. Boyde, A., et al. (1986) Scanning electron microscopy in bone pathology: review of methods, potential and applications. Scan Electron Microsc Pt 4, 1537–1554. 15. Onyia, J. E., et al. (2004) Novel and selective small molecule stimulators of osteoprotegerin expression inhibit bone resorption. J Pharmacol Exp Ther 309, 369–379. 16. Fukayama, S., et al. (1988) Human parathyroid hormone (PTH)-related protein and human PTH: comparative biological activities on human bone cells and bone resorption. Endocrinology 123, 2841–2848.
Chapter 3
Assessing New Bone Formation in Neonatal Calvarial Organ Cultures Khalid S. Mohammad, John M. Chirgwin, and Theresa A. Guise
Abstract Bone formation is a complex process, and testing anabolic effects on the skeleton of agents is slow and expensive in animals. Neonatal mouse calvariae cultured ex vivo show strong anabolic or catabolic bone responses to 1-week treatments and can be analyzed by quantitative histomorphometry. Changes in new bone area and osteoblast number caused by added proteins, drugs, or transfected genes can be quantified and analyzed for statistical significance. The organ cultures preserve much of the skeletal architecture and cellular diversity present in vivo and offer greater physiological relevance than cell lines studied in vitro. Keywords Osteoblast, bone formation, neonatal calvaria, organ culture.
1
Introduction
Organ cultures of neonatal mouse calvariae can be used to assess bone resorption and bone formation. This ex vivo culture system has advantages over osteoblast cell lines and tissue culture. The calvarial sections retain both the three-dimensional architecture of developing bone as well as a physiologically relevant array of cell types, including osteoblasts, osteoclasts, osteocytes and stromal cells, as well as mineralized matrix. In addition, the calvariae contain cells at all stages within the spatiotemporally complex lineages leading to mature osteoblasts and osteoclasts. External factors are applied to the calvariae, and bone formation is assessed by quantitative bone histomorphometric measurement of new bone area and osteoblast number (1).
1.1
Scope of the Assay
Although the assay has been used to study bone resorption in response to interleukins and other factors (2–4), it is most useful for the quantitative assessment of factors that stimulate new bone formation. Factors tested in the assay include From: Methods in Molecular Biology, Vol. 455: Osteoporosis: Methods and Protocols Edited by Jennifer J. Westendorf © Humana Press, Totowa, NJ
37
38
K.S. Mohammad et al.
polypeptides such as parathyroid hormone (5), bone morphogenetic proteins (6), endothelin-1 (7), and dickkopf-1 (8), and small molecule drugs such as statins (6), proteasome inhibitors (9), and histone deacetylase inhibitors (10). RNA can be isolated from treated calvariae and analyzed by gene microarray to identify pathways activated by exogenous factors. Activation of the pathways can be validated using primary osteoblasts isolated from additional neonatal calvariae (8). Proteins that are not available in purified form can also be tested by transduction of the cultures with lentiviral vectors that express osteogenic factors, such as the secreted protein Cyr61 (unpublished). It is also possible to express proteins from plasmid DNAs introduced into the cultures (11).
1.2
Strengths of the Assay
The output of the assay, new bone formation, is of greater physiological relevance than mineralized nodule formation or the expression of markers, such as alkaline phosphatase, in cell cultures. New bone area can be quantified and the results directly analyzed for statistically significant responses to treatment. The assay is considerably simpler, faster, and less expensive than whole animal experiments. The assay is conducted in serum-free medium and can be carried out with calvariae from transgenic and gene knock-out animals (12). It is possible to apply mechanical stimulation to the cultures (13) to mimic mechanical loading in vivo. Osteocytes can also be studied in the assay (14), and modified versions of the culture system can be used to study closure of the sutures (15).
1.3
Limitations of the Assay
A classical assay for osteoclastic resorption determines release into the medium of previously incorporated 45Ca from calvariae (16, 17), and the technique remains in standard use (18). The assay described here has not been optimized for the determination of osteolytic responses by histology, but continuous treatment with parathyroid hormone-related protein 1-34 stimulates massive osteolysis to the point that all bone is destroyed by the end of a 1-week experiment (unpublished). However, it might be possible to count osteoclasts and quantify bone destruction at earlier times. Because the calvariae are from neonatal animals, they lack mature immune cells, whose contributions are thus excluded. The calvaria is a specialized site within the skeleton and may not accurately reflect bone metabolism in the appendicular skeleton at sites such as the growth plate and trabecular and cortical surfaces. Although gene products can be delivered to the calvarial cells by viral vectors in vitro, the three-dimensional complexity of the bone pieces blocks transduction of the majority of the cells when tested with a high-titer β-galactosidase
3 Ex Vivo Bone Formation Assay
39
lentivirus followed by β-gal staining of the calvariae (G.A. Clines, personal communication). The assay uses animals and takes up to 1 month to complete, which is not what a pharmacologist would consider high throughput. Since osteogenic responses vary among litters of mouse pups, it is essential to analyze self-contained experimental groups of hemi-calvariae derived from a single litter with n = 4 per group. A positive and a negative control group must be included. A typical litter allows the analysis of three experimental groups, such as three concentrations of a test factor. The histological preparation and staining of the sections must be modified for these samples and cannot be carried out in a routinely configured tissue processor such as that available in a clinical pathology laboratory.
2
Materials
1. Animals: Timed-pregnant Swiss white mice are utilized. We purchase mice from Harlan (http://www.harlan.com/). Calvariae are harvested from 4day-old mouse pups. Orders should be scheduled so that pups are born at a time convenient for laboratory harvest. We order 15-day gestation timedpregnant mice to be delivered on Monday. Mice usually give birth on day 19, which is Thursday, and the pups are 4 days old on the following Monday. A Swiss white female mouse typically delivers 10 to 14 pups. There is substantial heterogeneity of bone responses between litters, so it is essential to use separate litters (with appropriate positive and negative controls) to assess and compare specific treatments. It is not recommended to mix bones from different litters. Dams and pups should be handled according to the regulations of the investigator’s institutional animal care and use committee, with an approved protocol. Decapitation with scissors is presently a permitted method of euthanasia in the United States for neonatal mouse pups. 2. Culture Media: BGJb medium is supplemented with 0.1% bovine serum albumin and 100 U/mL each of penicillin and streptomycin. 3. Stainless Steel Grids: The calvariae are cultured on #60 mesh, stainless steel type 304, 0.0075” wire diameter grids, which are inserted into tissue culture wells. Grids are cut to fit in the individual culture wells from mesh sheet (CX-0600, http://www.smallparts.com/products/descriptions/cx,%20cxx.cfm): 1 × 1.5 cm rectangles are cut from the stainless steel sheet. From the long side of the mesh, bend the edges about 2 mm from both ends using a metal ruler. After every assay, soak the grids in 10% nitric acid for 15 minutes and rinse thoroughly for another 15–20
40
K.S. Mohammad et al.
minutes. Let the grids air dry and autoclave. One can also use disposable Netwell polystyrene inserts (see Note 1) with 500-µm mesh fitted with a polyester mesh bottom for use in 12-well plates. (http://www.corning.com/Lifesciences/technical_ information/techDocs/Transwell_InstructionManual.pdf). These are more expensive than reusable steel grid for frequent use. 4. Tools 1. 2. 3. 4. 5. 6. 7. 8.
Curved, fine tip forceps. Straight, fine tip forceps. Microscissors, 12-mm cutting edge. Standard straight scissor and large forceps. Autoclaved beaker with 95% ethanol. Hot surface such as hot plate. Autoclaved beaker with double-distilled, autoclaved water. Petri dishes.
5. Fixation and Decalcification a) 10% neutral buffered formalin is used for tissue fixation. b) Decalcify calvariae in 10% w/v EDTA (ethylenediamine tetraacetic acid, adjusted to pH 7.2 with and made fresh weekly) with continuous stirring in a sufficient volume to cover the cassettes to a depth of at least 2.5 cm for 48 hours at room temperature. 6. Staining Solutions: Harris’s acidified hematoxylin with mercuric oxide is used for nuclear staining with eosin, phloxine B, and orange G as counterstains. a) Harris’s acidified hematoxylin: Add 25 g aluminum ammonium sulfate to 235 mL of distilled water in a 500-mL Erlenmeyer flask and heat at 37 degrees for 20 minutes with stirring until dissolved. Add 1.25 g of hematoxylin and 0.6 g of mercuric oxide powders to 15 mL of distilled water; stir to mix. Pour this mixture into the aluminum ammonium sulfate solution, and rinse with distilled water to make a complete transfer. Return flask to heat and bring the entire solution to a boil with high heat. Allow the solution to boil rapidly for 2–3 minutes and then cool on ice. Store covered in darkness for 8 hours. Filter and add 8 mL of glacial acetic acid. Filter before each use (see Note 2). b) Eosin Y 0.6%: Add 6 g of eosin to 900 mL of ethanol and 100 mL of water. Stir to dissolve. Add 50 mL of glacial acetic acid until pH is between 4.6 and 5.0. The color of the solution will change from opaque green to clear red. Dissolve 1 g of phloxine B in 100 mL water and filter to make a 1% solution. Dissolve 2 g of orange G in 100 mL water and filter to make a 2% solution. Make a working solution by adding 6 mL of 1% phloxine and 6 mL of 2% orange G to 238 mL of 0.6% eosin.
3 Ex Vivo Bone Formation Assay
3 3.1
41
Methods Calvarial Dissection
1. Sterilize all tools and conduct the procedure under a sterile hood. 2. Place a beaker with distilled water on the hot surface and bring it to boil, use the water to clean surgical tools between each dissection. 3. Have two Petri dishes filled with BGJ medium ready to receive the dissected calvariae. 4. Place the mouse pups under the hood. 5. Pick the first pup with the large forceps. 6. While holding with the forceps, dip the pup for 2 seconds in the 95% ethanol. 7. Decapitate the pup and place the head in a Petri dish with BGJ medium. 8. Rinse the head in the medium and transfer to a second dish with medium. 9. With the back of the curved forceps pick a good fold of the scalp and excise the scalp using the microscissors. Make sure to remove most of the scalp. 10. Identify the sutures of the skull, sagittal, coronals and lambdoids (Fig. 3.1A). 11. Starting about 2 mm behind the posterior fontanel, penetrate the calvaria with the tip of the scissors and make a straight cut all the way across the sagittal suture, passing through the coronal sutures and ending at the anterior fontanel. 12. Starting at the posterior fontanel make a straight cut along the lambdoid sutures to about the level of the eye. Stop here and make another straight cut to connect to the eye (Fig. 3.1B). From the end of the cut near the eye make another straight cut to connect to the anterior fontanel. Make this cut at a 45-degree angle. 13. Make similar cuts on the other half of the calvaria.
Fig. 3.1 Anatomy of calvaria. a. Identification of sutures. b. Diagram of cuts for preparation of hemi-calvarial bone pieces.
42
K.S. Mohammad et al.
Fig. 3.2 Handling of hemi-calvarial bone pieces. a. Orientation of bone piece.
14. Using the straight fine tip forceps pick up each half of the calvaria away from the sagittal suture surface and place it in medium (Fig. 3.2A). 15. Repeat the process on all pups. Do not pool calvariae from different litters.
3.2
Calvarial Culture
1. In 12-well tissue culture plate, place one grid in each well. 2. With fine-tip forceps pick up the calvariae away from the sagittal suture surface and place on top of the grids. No more than four hemi-calvariae at a time should be done to avoid drying out the calvariae. 3. Add 1 mL of medium to each well, slowly, to avoid washing the calvariae off the grids. The bones should be almost covered by the medium. 4. Use at least four bones (four hemi-calvariae) for each treatment group. For each litter, one group each is needed for respective positive and negative controls (see Note 3). The duration of the assay ranges from 4 to 14 days. The standard protocol is 7 days of treatment, but a time course may be necessary to optimize for new factors (see Note 4). 5. Change the medium 24 hours after setting up the assay and every 72 hours thereafter.
3.3
Harvesting and Fixation
At the end of the culture period calvariae can be harvested by removal from culture medium and directly placed in fixative. Calvariae from one treatment group are placed in 20 mL scintillation vial with 10 mL of 10% neutral buffered formalin for
3 Ex Vivo Bone Formation Assay
43
24 hours. Pick up the calvariae from the grids with a fine-tip forceps, away from the sagittal suture. The trapezoid shape of the calvariae will help in identifying the surfaces; the longer of which is the sagittal suture surface. Place the calvariae one at a time in the formalin.
3.4
Decalcification
Remove calvariae from the fixative, rinse in PBS and place in tissue-processing cassettes between two sponges (previously immersed in water to purge air bubbles). In a standard experiment using four calvariae/group, place all calvariae on the sponge in one cassette (one at each corner in similar orientation) (Fig. 3.2B). Place the second sponge on top and close the cassette. Label the cassette with pencil (not marker pen, which is dissolved by the subsequent processing steps) and place in 10% EDTA solution for 48 hours.
3.5
Tissue Processing
After decalcification, calvariae are processed in an automated tissue processor, such as a Shandon Excelsior (http://www.thermo.com/) specifically programmed for neonatal mouse calvariae: 1. 2. 3. 4. 5.
70% Ethanol, 30 minutes. 95% Ethanol, 30 minutes. 95% Ethanol, 30 minutes. 100% Ethanol, 20 minutes. 100% Ethanol, 20 minutes.
Fig. 3.2 (continued) b. Placement of hemi-calvariae in histology cassette.
44
6. 7. 8. 9. 10. 11. 12.
3.6
K.S. Mohammad et al.
100% Ethanol, 20 minutes. 100% Xylene, 20 minutes. 100% Xylene, 20 minutes. 100% Xylene, 20 minutes. Paraffin wax, 45 minutes. Paraffin wax, 45 minutes. Paraffin wax, 45 minutes.
Embedding
1. After processing, open the cassette and carefully remove the top sponge to avoid loss of any calvariae. 2. Pick up each calvaria carefully and stack all four calvariae in the cassette on top of each other in the same orientation. 3. After stacking them, grab all calvariae in one grip with the forceps and place them in the molds filled with some wax. Here, the sagittal surface should be facing down toward the base of the mold. 4. Move the mold to the cold surface and slowly release the forceps. Make sure the calvariae stay in place while the wax hardens. 5. Place the labeled cassette on top of the mold and fill it with paraffin wax (see Note 5).
3.7
Tissue Sectioning
1. Manually trim the blocks until tissue is visible through the wax. Trim to within 500–600 µm of the sagittal suture. 2. After trimming, cut 3.5-µm thick sections using an automated microtome, and collect four sections. 3. Trim the block another 300 µm and collect another four sections. 4. Trim the block 300 µm further and collect four more sections. 5. Collect the sections on glass slides: two sections per slide. 6. Allow slides to air dry in 37 °C oven overnight before staining.
3.8
Staining Procedure
1. Xylene 2. 100% Ethanol 3. 95% Ethanol
3× 3× 1×
2 minutes 1 minute 1 minute
3 Ex Vivo Bone Formation Assay
4. 80% Ethanol 5. Water 6. Hematoxylin 7. Water 8. 0.5% w/v aqueous ammonia 9. Water 10. 80% Ethanol 11. 95% Ethanol 12. Eosin 13. 95% Ethanol 14. 100% Ethanol 15. Xylene 16. Coverslip with permount
45
1× 1× 1× 1× 1× 1× 1× 1× 3× 3× 3×
3.9
Quantitative Assessment of Bone Formation by Histomorphometry
3.9.1
Systems Used
1 minute 3 minutes 30 seconds change until rinse water is clear 10 seconds 3 minutes 1 minute 1 minute 2 minutes 1 minute 1 minute 1 minute
Quantification is carried out on digital images. Calvarial sections are viewed on a Leica DMLB microscope outfitted with Q-imaging micropublisher camera (http://www.vashaw.com/) at 200× magnification with a pan apochromat 20× lens. The images can be analyzed with histomorphometric software such as MetaMorph (http://www.moleculardevices.com/pages/software/metamorph.html), OsteoMeasure (http://www.osteometrics.com/software_update.htm), or Bioquant (http://www. bioquant.com/). 3.9.2
Orientation and Defining Area for Analysis
1. Tissues are first examined under low power (1.6×) for orientation and to identify the sutures (Fig. 3.3A). 2. Define the coronal suture. Bone tissue should be visible on both sides of the coronal suture. There is a long bone surface on one side and a short surface on the other (Fig. 3.3B). Most consistent results are obtained by analyzing the bone formation on the long surface. 3. Under the 20× magnification, identify the suture and move two or three optical fields away from the suture along the long surface. Capture the image from this area for analysis. An equivalent area should be analyzed for each section unless the tissue is damaged, in which case the next intact field should be used. 4. Define the area of old bone and areas of new bone on each image. Eosin Y with orange G will stain the old bone dark orange, while the new bone formed during the assay will be light orange (Fig. 3.4A).
46
K.S. Mohammad et al.
Fig. 3.3 Orientation of fixed sections to choose areas for analysis. a. Long and short arms at 1.6×.
Fig. 3.3 (continued) b. Orientation of coronal suture at 5×.
3.9.3
Tracing
Using the tracing tool in the software, trace the borders of the old bone area. Trace the borders of the total bone area (which include the old and new bone) (Fig. 3.4B).
3.9.4
Quantification and Calculation of New Bone Area
1. Using the color threshold tool mark the bone area. The system will pick up any area in the defined field that matches the color threshold for the bone. 2. Subtract the two threshold areas: The total bone area minus the old bone equals the new bone area. 3. Repeat the same steps on all the images. 4. New bone area can be calculated in µm2 (Fig. 3.5A) (see Note 6).
3 Ex Vivo Bone Formation Assay
47
Fig. 3.4 Histological appearance of stimulated calvarial section at 20×. a. Lighter and darker staining of new and old bone.
Fig. 3.4 (continued) b. Outlining areas for computerized analysis of areas.
3.9.5
Calculation of Osteoblast Number
Osteoblasts may be counted manually or determined with the counting function of the software and expressed as number of osteoblasts per 20× visual field (Fig. 3.5A) (see Note 6).
3.9.6
Data Analysis
Groups of sections should be coded for blinded analysis to prevent unintentional reading bias during the microscopic analysis. Data can be collected in a spreadsheet such as Excel and analyzed with a statistical package such as GraphPad Prism (http://www.graphpad.com/). Changes in new bone area or osteoblast number can be compared between groups for statistically significant differences by a measure such as one-way analysis of variance (ANOVA).
48
K.S. Mohammad et al.
Fig. 3.5 Stimulation of new bone formation with endothelin-1. a. Results of quantitative histomorphometric analysis on new bone area and osteoblast number. * = p < 0.05; ** = p < 0.01.
Fig. 3.5 (continued) b. Representative sections from the experiment analyzed in panel (A). New bone formation indicated between arrowheads. Similar data may be found in reference 7.
4
Notes
1. Stainless steel grids are preferable to Netwell mesh, being less expensive and reusable. However, the Netwell mesh is easier to handle, and mesh heights are consistent in all wells, so the calvariae are less likely to slide off the mesh.
3 Ex Vivo Bone Formation Assay
49
2. Best used within 1 week of preparation. The solution should be filtered if the color starts to change. Proper safety procedures should be used when working with and disposing of toxic mercuric oxide. 3. Insulin (50 µg/mL) is a stable, inexpensive and reproducible positive control; other anabolic factors such as BMP-2 and endothelin (ET)-1 can be used. The high concentration of insulin is believed to stimulate new bone through activation of the IGF-1 receptor. 4. The 7-day assay gives strong new bone formation and a less significant increase in osteoblast number. The 4-day assay shows greater osteoblast proliferation and less new bone formation. 5. Careful calvarial embedding is crucial in the assay. Proper orientation is essential for consistent histological sections and reproducible results. 6. Fig. 3.5 shows an example of quantitative analysis of new bone area and osteoblast number from calvariae treated for 7 days with ET-1, a 21 amino acid peptide. The response was blocked by an ETA receptor antagonist, atrasentan.
References 1. Garrett, I. R. (2003) Assessing bone formation using mouse calvarial organ cultures. Methods Mol Med 80, 183–198. 2. Yoneda, T., Williams, P., Boyce, B. F., et al. (1995) Suramin suppresses hypercalcemia and osteoclastic bone resorption in nude mice bearing a human squamous cancer. Cancer Res 55, 1989–1993. 3. Cornish, J., Gillespie, M. T., Callon, K. E., et al. (2003) Interleukin-18 is a novel mitogen of osteogenic and chondrogenic cells. Endocrinology 144, 1194–1201. 4. Kusano, K., Miyaura, C., Inada, M., et al. (1998) Regulation of matrix metalloproteinases (MMP-2, -3, -9, and -13) by interleukin-1 and interleukin-6 in mouse calvaria: association of MMP induction with bone resorption. Endocrinology 139, 1338–1345. 5. Bringhurst, F. R., Potts, J. T., Jr. (1981) Bone collagen synthesis in vitro: structure/activity relations among parathyroid hormone fragments and analogs. Endocrinology 108, 103–108. 6. Mundy, G., Garrett, R., Harris, S., et al. (1999) Stimulation of bone formation in vitro and in rodents by statins. Science 286, 1946–1949. 7. Yin, J. J., Mohammad, K. S., Kakonen, S. M., et al. (2003) A causal role of endothelin-1 in the pathogenesis of osteoblastic bone formation. Proc Natl Acad Sci U S A 100, 10954–10959. 8. Clines, G. A., Mohammad, K. S., Bao, Y., et al. (2007) Dickkopf homolog 1 mediates endothelin-1-stimulated new bone formation. Mol Endocrinol 21, 486–498. 9. Garrett, I. R., Chen, D., Gutierrez, G., et al. (2003) Selective inhibitors of the osteoblast proteasome stimulate bone formation in vivo and in vitro. J Clin Invest 111, 1771–1782. 10. Schroeder, T. M., Westendorf, J. J. (2005) Histone deacetylase inhibitors promote osteoblast maturation. J Bone Miner Res 20, 2254–2263. 11. Premaraj, S., Mundy, B., Parker-Barnes, J., et al. (2005) Collagen gel delivery of Tgf-beta3 nonviral plasmid DNA in rat osteoblast and calvarial culture. Orthop Craniofac Res 8, 320–322. 12. Woitge, H. W., Kream B. E. (2000) Calvariae from fetal mice with a disrupted Igf1 gene have reduced rates of collagen synthesis but maintain responsiveness to glucocorticoids. J Bone Miner Res 15, 1956–1964. 13. Rawlinson, S. C., Mosley, J. R., Suswillo, R. F., et al. (1995) Calvarial and limb bone cells in organ and monolayer culture do not show the same early responses to dynamic mechanical strain. J Bone Miner Res 10, 1225–1232.
50
K.S. Mohammad et al.
14. Gu, G., Mulari, M., Peng, Z., et al. (2005) Death of osteocytes turns off the inhibition of osteoclasts and triggers local bone resorption. Biochem Biophys Res Commun 335, 1095–1101. 15. Opperman, L. A., Fernandez, C. R., So, S., et al. (2006) Erk1/2 signaling is required for Tgfbeta 2-induced suture closure. Dev Dyn 235, 1292–1299. 16. Nyberg, L. M., Marks, S. C., Jr. (1975) Organ culture of osteopetrotic (ia) rat bone: evidence that the defect is cellular. Am J Anat 144, 373–378. 17. Lerner, U. (1980) Indomethacin inhibits bone resorption and lysosomal enzyme release from bone in organ culture. Scand J Rheumatol 9, 149–156. 18. Palmqvist, P., Lundberg, P., Persson, E., et al. (2006) Inhibition of hormone and cytokinestimulated osteoclastogenesis and bone resorption by interleukin-4 and interleukin-13 is associated with increased osteoprotegerin and decreased RANKL and RANK in a STAT6-dependent pathway. J Biol Chem 281, 2414–2429.
Chapter 4
Detection of Apoptosis of Bone Cells In Vitro Teresita Bellido and Lilian I. Plotkin
Abstract Studies during the last decade demonstrated that apoptosis is as important as mitosis for the growth and maintenance of the skeleton and provided information on the significance and molecular regulation of apoptosis of bone cells. It is now known that: (1) all osteoclasts die by apoptosis after completing a bone resorption cycle; (2) the majority of osteoblasts also die, whereas the remainder become lining cells or osteocytes; and (3) osteocytes, although long-living cells, also can die prematurely. Furthermore, mounting evidence indicates that systemic hormones, local growth factors, cytokines, and pharmacological agents, as well as mechanical forces regulate the rate of bone cell apoptosis. This chapter summarizes the methods developed in the last few years to examine apoptosis of cultured bone cells and identify the signaling pathways and molecules involved in apoptosis regulation by diverse skeletal stimuli. Keywords Apoptosis, osteoblasts, osteocytes, osteoclasts, bone.
1
Introduction
Apoptosis or programmed cell death is a form of cell suicide originally defined by morphological alterations in the cell nucleus and cytoplasm. Cells undergoing apoptosis exhibit dramatic nuclear changes, resulting in aggregation of the chromatin around the nuclear membrane followed by disintegration of the nuclear envelope and accumulation of bodies of condensed chromatin in the cytoplasm. Along with these nuclear alterations, the cytoplasm of apoptotic cells contracts and cell volume decreases due to loss of fluid through the damaged plasma membrane. During apoptosis, cells also lose contact with the extracellular matrix and detach from neighboring cells. The apoptotic process culminates with cellular disintegration into apoptotic fragments that are quickly removed by surrounding phagocytic cells. These morphological changes are accompanied by changes in the activity of intracellular enzymes and profound alterations in the composition of the cell membrane and the structure of the DNA. The family of intracellular cysteinyl aspartate-specific proteases, From: Methods in Molecular Biology, Vol. 455: Osteoporosis: Methods and Protocols Edited by Jennifer J. Westendorf © Humana Press, Totowa, NJ
51
52
T. Bellido and L.I. Plotkin
named caspases, plays a central role in the regulation and execution of apoptosis. “Initiator” caspases, such as caspase-8 and caspase-9, are activated by dimerization. On the other hand, “effector” or “executioner” caspases, such as caspase-3, are activated by cleavage at specific sites by the initiator caspases. These morphological and biochemical features of apoptosis have been used to detect apoptotic cells in bone tissues as well as in cultured bone cells. Thus, detection of increases in the activity of initiator or effector caspases is widely used to identify apoptotic cells. In addition, the presence of cleaved genomic DNA induced by caspase-mediated activation of DNases is another sine qua non feature of apoptosis. Thus, DNA fragmentation is detected by gel electrophoresis, or quantified using TUNEL (TdT-mediated dUTP-biotin nick end-labeling), ISNT (in situ nick translation) and ISEL (in situ nick end-labeling) techniques. Nuclear fragmentation and chromatin condensation can be visualized using fluorescent dyes that bind to DNA. Furthermore, examination of nuclear morphology of cells transfected with fluorescent proteins containing a nuclear localization sequence is a particularly useful tool for studying apoptosis in cells co-transfected with genes of interest. Cell detachment from the substrate and changes in the composition of the plasma membrane are also features that have been used to detect and quantify apoptotic cells. This chapter summarizes the methods utilized to determine the effect of systemic hormones, local growth factors, cytokines, pharmacological agents, and mechanical forces on apoptosis of cultured bone cells. The methodological approaches described here have been valuable for identifying signaling pathways and molecules involved in apoptosis regulation by these various stimuli.
2
Materials
2.1 1. 2. 3. 4. 5. 6. 7. 8.
Cells and Reagents for Tissue Culture
OB-6 osteoblastic cell line, established from murine bone marrow (1, 2). MLO-Y4 osteocytic cell line, established from murine long bones (3). Raw 264.7 monocytic cells (American Type Culture Collection; ATCC) (4, 5). α-MEM. Fetal bovine serum (FBS). Bovine calf serum (BCS). Penicillin, streptomycin, glutamine (PSG). Collagen type I from calf skin.
2.2
Regulators of Apoptosis, Kinase Inhibitors, and Receptor Antagonists
1. Parathyroid hormone (PTH) (2, 6). 2. Calcitonin (7, 8).
4 Detection of Apoptosis of Bone Cells In Vitro
53
3. Sex steroids: 17-β-estradiol, dihydrotestosterone (9–13). 4. 1α,25(OH)2 vitamin D3 and vitamin D3 metabolites, 24R,25(OH)2D3 and 25(OH)D3 (14). 5. Interleukin-6, oncostatin M, and interleukin-6 soluble receptor (15). 6. TGFβ (16–18). 7. IGF-I and II, PDGF, and bFGF (19). 8. Bisphosphonates (provided by Gador, S.A., Buenos Aires, Argentina and E. Oldfield, University of Illinois, Urbana, IL) (7, 13, 20–23). 9. CD40L (24). 10. Kinase inhibitors: Src inhibitor PP1 (12, 14, 20), ERK inhibitors PD98059 (7, 12, 14, 25, 26) and U0126 (7, 12, 14), PI3K inhibitor wortmannin (11, 25, 26), JNK inhibitor SP600125 (14), p38 inhibitor SB203580 (14, 26). 11. Receptor antagonists: estrogen receptor antagonist ICI182780 (12), androgen receptor antagonist flutamide (12), glucocorticoid receptor antagonist RU486 (22). 12. Dexamethasone (2, 6, 7, 22, 24, 26). 13. Etoposide (2, 6, 10, 12, 24–26). 14. Murine TNFα (6, 7, 24). 15. H2O2 (27, 28).
2.3
Measurement of the Catalytic Activity of Caspase-3 by Cleavage of a Fluorogenic Substrate
1. Lysis Buffer: 50 mM HEPES, pH 7.4, 0.1% CHAPS, 1 mM DTT, 0.1 mM EDTA; store at −20°C. Add protease inhibitors to the lysis buffer immediately before preparing the lysates. Use 2 µg/mL aprotinin (Stock: 0.4 mg/mL in 0.01 M HEPES, pH 7.4); 2 µg/mL leupeptin (Stock: 5 mg/mL in water); 1 mM phenylmethylsulfonyl fluoride, PMSF (Stock: 100 mM or 17.4 mg/mL). Store at −20°C. 2. Assay Buffer: 50 mM HEPES, pH 7.4, 100 mM NaCl, 0.1% CHAPS, 10 mM DTT, 1 mM EDTA, 10% glycerol. Store at −20°C. 3. Fluorogenic Substrate: 0.2 mM Ac-DEVD-AFC (Biomol) in assay buffer (Stock: 2 mM in DMSO). Store at −20°C protected from light.
2.4
Sodium Dodecylsulfate (SDS)-Polyacrylamide Gels and Western Blot Analysis
1. Resolving Gel Buffer: 1.5 M Tris-HCl, 1% SDS, pH 8.8. Store at room temperature. 2. Resolving acrylamide solution (RAS): 30% acrylamide, 1.5% bis-acrylamide in water (see Note 1). Filter through Whatman paper. Store at 4°C protected from light.
54
T. Bellido and L.I. Plotkin
3. Stacking Gel Buffer: 0.5 M Tris-HCl, 1% SDS, pH 6.8. Store at room temperature. 4. Stacking acrylamide solution (SAS): 30% acrylamide, 0.8% bis-acrylamide in water. Filter through Whatman paper. Store at 4°C protected from light. 5. 10× running buffer: 0.4 M Tris base, 2.5 M glycine, 1% SDS, pH 8.3. Store at room temperature. To prepare a 1× solution, mix 100 mL of 10× running buffer and 900 mL of water. 6. 10× transfer buffer: 2 M Tris base, 0.4 M glycine, 0.37% SDS, pH 8.3. Store at room temperature. To prepare a 1× solution, mix 100 mL of 10× transfer buffer, 700 mL of water, and 200 mL of methanol. 7. 10× Sample Buffer: 250 mM Tris-HCl, pH 6.8, 20% SDS, 10% glycerol, 0.05% bromophenol blue. Store at room temperature. To prepare a 5× solution (reducing loading buffer), mix an equal volume of 10× sample buffer and 1 M dithiothreitol (DTT). 8. Molecular Weight Markers: 8 µL of pre-stained molecular weight marker (Amersham Biosciences, Pittsburgh, PA), 12 µL of lysis buffer, 5 µL of 5× reducing loading buffer (amounts given for two gels, total volume: 25 µL). Pre-stained markers from Amersham require the addition of loading buffer; molecular weight markers from other sources may not.
3
Methods
3.1
Cell Models
The methods described in this chapter were established to examine apoptosis of OB-6 osteoblastic cells (1, 2), MLO-Y4 osteocytic cells (3) and osteoclast-like cells developed from primary cultures or from Raw 264.7 cells (21). The techniques described are the same for all these cells, unless otherwise indicated. The procedures should be adapted when working with other cell lines or primary cultures. The techniques described in this chapter can be used to identify the role of a molecule in a particular pro- or anti-apoptotic pathway. Cells can be isolated from mice lacking or overexpressing a wild type or a mutated form of the protein of interest (2, 20). In addition, cells can be transiently transfected with a plasmid encoding the protein of interest, either wild-type or mutated (dominant negative, truncated, unphosphorylatable, constitutively active, etc.) together with a molecule that allows the identification of apoptotic cells (e.g., nuclear targeted fluorescent proteins or caspase-3 sensor, see the following) (2, 9, 12, 14, 20, 21, 25, 29, 30). 3.1.1
Osteoblastic OB-6 Cells
1. Growth Medium: α-MEM containing 10% FBS and 1% PSG. 2. Medium to Induce Apoptosis: α-MEM containing 2% FBS and 1% PSG (see Note 2) (2). 3. Cells are cultured on tissue culture-treated plates.
4 Detection of Apoptosis of Bone Cells In Vitro
3.1.2
55
Osteocytic MLO-Y4 Cells
Growing Medium: α-MEM containing 2.5% FBS, 2.5% BCS, and 1% PSG. This medium is also suitable to induce apoptosis in these cells (3, 7). Cells are cultured in collagen coated tissue culture-treated plates, prepared as follows: 1. Prepare a 0.1% solution of calf-skin collagen type I in phosphate buffered saline (PBS) containing 1% glacial acetic acid. This stock can be stored at 4°C for 1 month. Alternatively, a 0.1% collagen type I solution can be purchased. 2. Dilute the stock 0.1% solution 1 to 10 in PBS containing 1% glacial acetic acid to obtain a 0.01% solution and use it immediately. 3. Cover the surface of the well with the 0.01% collagen type I solution by adding 300 µL/cm2. 4. Incubate for at least 3 hours to overnight at 37°C. Remove the solution, which can be reused twice on new plates. If previously used collagen solution is employed, extend the incubation time to 48 hours. Discard the solution after using it three times. 5. Let the plates dry under the tissue culture hood with the UV light on for not more than 30 minutes. 6. For storage, wrap in Saran wrap and store up to 3 months at 4°C. Rinse with PBS prior to use.
3.1.3
Osteoclast Precursors from Non-Adherent Bone Marrow Cells
1. Mice are sacrificed by placing them in a jar containing absorbent paper napkins soaked with metaphane for approximately 3 minutes. 2. Spray the mouse with 70% ethanol and place it under a tissue culture hood. Do all procedures under sterile conditions. 3. Extract femurs and tibiae and clean off muscle with sterile gauze. 4. Snip off the distal and proximal ends of the bones and place them into a centrifuge tube, with the proximal side down, containing 0.5 mL of medium (α-MEM containing 10% FBS and 1% PSG). 5. Cover the tubes and place them on ice until all femurs have been taken. 6. Spin in tabletop centrifuge at high speed for 2 minutes. Spray the external part of the tubes with ethanol before introducing back into the tissue culture hood. Use sterile forceps to shake the pellet loose from the bottom of the femur. Transfer the cell suspension to a 15-mL centrifuge tube and spin down. 7. Count cells with a Coulter Counter or a hemocytometer. 8. Plate cells in tissue culture flasks at 40–50 × 10 6/T175 flask. Culture for 48 hours in a tissue culture incubator (37°C, 5% CO2). 9. Harvest the non-adherent cells. Rinse the cell monolayer once with PBS to collect remaining non-adherent cells and add to the previous suspension of non-adherent cells. The amount of non-adherent cells recovered is approximately half of the total number of cells plated. Count the cells and plate 1,000 cells/well in a 96-well plate to develop osteoclasts (see Section 3.1.6).
56
3.1.4
T. Bellido and L.I. Plotkin
Osteoclast Precursors from Non-Adherent Spleen Cells
1. Mice are sacrificed by placing them in a jar containing absorbent paper napkins soaked with metaphane for approximately 3 minutes. 2. Spray the mouse with 70% ethanol and place it under a tissue culture hood. Perform all procedures under sterility conditions. 3. Remove the spleen and cut it in small pieces into a 3-cm Petri dish containing 2 mL growing medium for osteoclasts. 4. Place on ice until all spleens have been taken. 5. Disperse cells by pipetting up and down with a 5-mL pipette. Filter the cell suspension through a nylon mesh cell strainer to eliminate tissue and cell aggregates. 6. Count cells with a Coulter counter or hemocytometer. 7. Plate cells in tissue culture flasks at 40–50 × 106/T175 flask. Culture for 48 hours in a tissue culture incubator (37°C, 5% CO2). 8. Harvest the non-adherent cells. Rinse the cell monolayer once with PBS to collect remaining non-adherent cells and add to the previous set of non-adherent cells. The amount of non-adherent cells recovered is approximately half of the total number of cells plated. Count the cells and plate 1,000 cells/well in a 96-well plate to develop osteoclasts (see Section 3.1.6).
3.1.5
RAW 264.7 Cells
1. Thaw RAW 264.7 cells and plate them in a T75 flask. Culture in α-MEM containing 10% FBS and 1% PSG for 3–4 days. 2. Split cells by adding trypsin. Plate 2,000 cells/well in a 96-well plate to develop osteoclasts (see Section 3.1.6).
3.1.6
Osteoclast Development
1. Plate osteoclasts precursors in a 96 well plate in α-MEM containing 10% FBS (non-heat inactivated) (see Note 3) and 1% PSG containing human recombinant M-CSF and sRANKL (final concentration of 30 ng/mL for each cytokine). Do not remove from incubator for 3 days. 2. Remove half of the medium and add fresh medium containing 30 ng/mL M-CSF and sRANKL at day 3 or 4. Culture until good osteoclast development is observed (6–8 days for non-adherent bone marrow– or spleen-derived cells, 4–6 days for Raw cells) (see Note 4). 3. Add the agents to prevent and/or to induce apoptosis and culture for the appropriate time.
4 Detection of Apoptosis of Bone Cells In Vitro
3.2
Prevention of Apoptosis
3.2.1
Hormones, Cytokines, Growth Factors, or Pharmacological Stimuli
57
Cells are treated with the stimulus of interest for 1 hour, followed by the addition of the pro-apoptotic agent. Some of the hormones, cytokines, growth factors, and pharmacological agents that have been used to prevent osteoblast and osteocyte apoptosis are listed in the Materials section. Table 4.1 indicates some of these agents used to prevent osteoblast and osteocyte apoptosis, as well as the concentrations of the stock and working solutions. The following protocol is an example of treatment of osteoblastic cells with parathyroid hormone (PTH) (2, 6): 1. Plate 104 OB-6 cells/well in growing medium in a 48-well plate. Culture overnight. 2. Change the medium to medium containing 2% FBS. Culture for 3 hours to overnight. 3. Add bovine PTH (1–34) to a final concentration of 5 × 10−8 M or the equivalent amount of vehicle (PBS). Culture for 1 hour. 4. Add the pro-apoptotic agent (see below) or the equivalent volume of the corresponding vehicle. Culture for 6 hours. 5. Harvest floating cells by centrifugation of culture medium and adherent cells by trypsinization. Combine the pellets. 6. Determine the percentage apoptotic cells. (See the following for detection methods.)
3.2.2
Mechanical Stimuli
Osteoblasts and osteocytes can be mechanically stimulated by stretching or by applying fluid flow shear stress. The following protocol has been used to determine the effect of mechanical stimulation by biaxial stretching using a FX-4000 Flexercell Strain Unit (Flexcell International, Hillsborough, NC) on osteocyte apoptosis (26). Some other examples of prevention of apoptosis of osteoblasts and osteocytes by mechanical stimulation are described in references (31–33).
Table 4.1 Prevention of osteoblast and osteocyte apoptosis Inducer Stock solution Parathyroid hormone Bisphosphonates Sex steroids 1α,25(OH)2 vitamin D3 and vitamin D3 metabolites
Final concentration
10−4 M 10−2 M 10−2 M
5 × 10−8 M 10−7 M 10−8– 10−7 M
10−2 M
10−8 M
58
T. Bellido and L.I. Plotkin
1. Plate 2 × 105 MLO-Y4 cells on collagen coated flexible bottom 6-well plates in growing medium. Culture overnight. 2. Remove the medium and add fresh growing medium containing vehicle or pharmacologic inhibitors (see the following). Culture for 30 minutes. 3. Stretch for 10 minutes at 2% or 5% elongation using a regimen consisting on 20 seconds of stretching followed by 0.1 seconds of release. 4. Add vehicle or appropriate concentration of pro-apoptotic stimuli and determine apoptosis (see the following).
3.2.3
Pharmacological Inhibitors
The role of a particular molecule in pro- or anti-apoptotic events can be determined by using specific pharmacological inhibitors. Some commonly used inhibitors are listed in the Materials section. As examples, the protocols for inhibiting extra signal regulated kinase (ERK) activation (7) or caspase-3 activation (7) in MLO-Y4 cells are detailed. Inhibitors should be added to the cell culture 30 minutes before adding the pro-apoptotic stimulus when the objective is to determine whether the molecule that is inhibited is required for induction of apoptosis. Inhibitors should be added to the cell culture 30 minutes before adding the anti-apoptotic stimulus when the objective is to determine whether the molecule that is inhibited is required for prevention of apoptosis.
3.2.3.1
Inhibition of ERK Activation
1. Plate MLO-Y4 cells in collagen-coated tissue culture plates. Culture until they reach 75–80% confluence (18–36 hours). 2. Remove the medium and add fresh medium containing 50 µM PD98059, a specific inhibitor of MEK, the upstream kinase that activates ERKs (34), or the equivalent amount of vehicle (dimethyl sulfoxide). Culture the cells for 30 minutes. 3. Add the anti- and/or pro-apoptotic stimulus. 4. Continue with the chosen assay for apoptosis detection.
3.2.3.2
Inhibition of Caspase-3 Activation
1. Plate MLO-Y4 cells in collagen-coated tissue culture plates. Culture until they reach 75–80% confluence (18–36 hours). 2. Remove the medium and add fresh medium containing 50 nM DEVD-CHO, a specific inhibitor of capase-3 activation (35), or the equivalent amount of vehicle (dimethylsulfoxide). Culture cells for 30 minutes. 3. Add the pro-apoptotic stimulus. 4. Continue with the chosen assay for apoptosis detection.
4 Detection of Apoptosis of Bone Cells In Vitro
3.3
59
Induction of Apoptosis
The inducers of apoptosis vary depending on the cell type. Thus, agents that induce osteoblast and osteocyte apoptosis may not affect osteoclasts or may even prevent osteoclast apoptosis (6, 7, 9, 10, 12, 22, 29, 36). When adding a pro-apoptotic stimulus, always include a vehicle control in which only the vehicle in which the pro-apoptotic agent was diluted is added. The optimum concentrations and time of treatment should be determined for each particular cell line or primary culture.
3.3.1
Osteoblast and Osteocyte Apoptosis
Some of the agents commonly used to induce osteoblast and osteocyte apoptosis, as well as concentrations of the stock and working solutions, are indicated in Table 4.2 (2, 6, 7, 12, 25, 27–29, 36). To induce apoptosis, cells are treated with the pro-apoptotic agents for 3 to 48 hours. To induce apoptosis by serum starvation, growing medium is removed, and cells are washed thoroughly to remove the remaining medium containing serum. This is followed by addition of medium alone or medium containing 0.5% bovine serum albumin (25, 36). To induce apoptosis by cell detachment (anoikis) cells grown up to 70–80% confluence are detached from the extracellular matrix by treatment with EDTA/collagenase, washed with PBS and re-suspended in growing medium. Subsequently, 105 cells/mL are plated in Ultra Low Cluster polystyrene plates (Costar, Corning Incorporated Life Sciences, Lowell, MA) (37). These plates do not allow cell attachment, thereby inducing anoikis.
3.3.2
Osteoclast Apoptosis
Some of the agents commonly used to induce osteoclast apoptosis, as well as concentrations of the stock and working solutions, are indicated in Table 4.3 (9, 21, 38, 39). To induce apoptosis cells are treated with the pro-apoptotic agents for 6 to 24 hours. Table 4.2 Induction of osteoblast and osteocyte apoptosis Inducer Stock solution Dexamethasone Etoposide Murine TNFα H2O2 Serum starvation Cell detachment (anoikis)
10−2 M 5 mM 2.78 × 10−7 M 8.8 M (30 vol) — —
Final concentration 10−6 M 5 × 10−5 M 10−9 M 50 µM — —
60
T. Bellido and L.I. Plotkin
Table 4.3 Induction of osteoclast apoptosis Inducer Stock solution Bisphosphonates Sex steroids (17-β-estradiol, dihydrotestosterone) TGFβ
Final concentration
10−2 M
10−5 – 10−4 M
10−2 M 80 nM
10−8 M 0.4 nM
3.4
Detection of Apoptosis
3.4.1
Visualization of DNA Fragments by Gel Electrophoresis
The appearance of DNA fragments of 180 bp and multimers that results from the induction of apoptosis can be detected by electrophoresis. 1. Treat the cells with the agent of study. Make sure to include a recognized proapoptotic agent as positive control and vehicle as negative control. 2. Remove medium, add PBS, and scrape the cells off. 3. Transfer the cells to a centrifuge tube using more PBS to wash them, and pellet them by centrifugation. 4. Incubate the cell pellets for 10 seconds with Tris-EDTA-NP40 lysis buffer: 20 mM EDTA, 1% NP40, 50 mM Tris-HCl, pH 7.5 to lyse the cells. Add 10 µL per 1 million cells, minimum 50 µL. 5. Spin down for 5 minutes at 1,600g. Save supernatant and repeat extraction with same amount of lysis buffer. Combine supernatants. 6. Bring supernatants and nuclear pellet (as a control for total recovery of DNA fragments) to 1% SDS. 7. Treat for 2 hours with RNAse A (final concentration 5 µg/µL) at 56°C. 8. Treat with proteinase K (final concentration 2.5 µg/µL) for at least 2 hours at 37°C. 9. Precipitate DNA by addition of 1/2 volume 10 M ammonium acetate and 2.5 volume ethanol. 10. Spin down and re-dissolve in DNA loading buffer. 11. Run 5–10 µg of DNA preparation on a 1% agarose gel containing 0.5 µg/mL of ethidium bromide. 12. Visualize the 180 bp DNA fragments, and multimers thereof, under an UV lamp. DNA can be isolated using commercially available kits, for example, Apoptotic DNA ladder kit (Roche Applied Science), followed by agarose gel electrophoresis to separate the DNA fragments, as indicated in the preceding (steps 11–12).
3.4.2
ISEL
The appearance of double and single strand breaks (“nicks”) in the genomic DNA that results from the process of apoptosis can be identified by labeling free 3’-OH
4 Detection of Apoptosis of Bone Cells In Vitro
61
terminal ends with modified nucleotides in an enzymatic reaction. The product of the reaction is detected by the appearance of a brown precipitate resulting from the action of peroxidase on the diaminobenzidine (2). Alternatively, the peroxidasestreptavidin conjugate can be replaced by fluorescent tagged-streptavidin (e.g., FITC-streptavidin) and the apoptotic cells can be identified under a fluorescence microscope. A positive and negative control should be included every time the assay is performed. As a positive control, a sample in which apoptosis is known to occur or cells treated with DNAse I (3,000 U/mL in 50 mM Tris-Cl H, pH 7.5 1 mg/ mL l BSA) for 10 minutes at 15–25:C to induce DNA strand breaks, can be used. As negative control, a sample incubated as indicated for the experimental samples, but excluding the Klenow enzyme should be used. The protocol described in the following has been developed using a kit from Oncogene Research Products with modification to the manufacturer’s instructions. All the steps are performed at room temperature. Briefly: 1. Treat the cells with the agent of study. Make sure to include a recognized proapoptotic agent as positive control and vehicle as negative control. 2. Fix the cells with 4% paraformaldehyde for 15 minutes. 3. Permeabilize with 2 µg/mL proteinase K for 5 minutes. 4. Quench endogenous peroxidase with 3% H2O2 for 15 minutes. 5. Incubate cells with Klenow enzyme for 1 hour and 50 minutes in a humidified chamber. 6. Rinse with PBS five times. 7. Incubate with peroxidase-streptavidin conjugate for 20 minutes. 8. Develop using diaminobenzidine. 9. Counterstain the cells with 1% methyl green for 1 minute. 10. Cover with a coverslip using hydrophobic mounting medium. The percentage of apoptotic (brown) cells is determined by examining at least 250 cells from fields selected by systematic random sampling for each experimental condition (2).
3.4.3
Caspase Activation
The protocols described in the following were designed to measure caspase-3 activation and can be adapted to investigate the role of other caspases on apoptosis using specific substrates or antibodies for each caspase, which are commercially available.
3.4.3.1
Detection of Caspase Activity by Fluorometry or Colorimetry
The method detailed in the following uses a fluorogenic substrate and can be adapted to be used with a chromogenic substrate. The use of the latter will yield a less sensitive assay.
62
T. Bellido and L.I. Plotkin
1. Plate the cells in order to obtain approximately 75–80% confluence after overnight culture (see Note 5). 2. Culture for the desired time, adding the pro- and anti-apoptotic agents as indicated in the preceding. 3. Collect floating cells as follows: transfer the culture medium to an Eppendorf tube; wash the cell monolayer once with PBS and add the washing PBS to the same Eppendorf tube that contains the culture supernatant. Spin for 5 minutes at 2,000g and 4°C to pellet floating cells (see Note 6). 4. While spinning the tubes containing the floating cells, add 20 µL of lysis buffer/ well to the cells still attached. Scrape off the cells and pipette up and down several times to break up the cells. 5. Remove as much supernatant as you can from the spun tubes containing the floating cells. Combine lysates obtained from the monolayer with the pellets from the floating cells and vortex vigorously. Incubate on ice 15 minutes. 6. Continue with the caspase assay or store the lysates at −70°C until use.
3.4.3.2
Detection of Caspase Activity in Osteoclasts Derived from Non-Adherent Bone Marrow, Spleen, or RAW 264.7 Cells
1. Obtain osteoclasts and treat as desired. 2. Collect floating cells as follows: Transfer the culture media to an Eppendorf tube, wash the cell monolayer once with PBS and add the washing PBS to the same Eppendorf tube that contains the culture supernatant. Spin for 5 minutes at 2,000g and 4°C to pellet floating cells (see Note 6). 3. While spinning the tubes containing the floating cells, add 20 µL of lysis buffer/ well to the cells still attached. Scrape off the cells and pipette up and down several times to break up the cells. 4. Remove as much supernatant as you can from the spun tubes containing the floating cells. Combine lysates obtained from the monolayer with the pellets from the floating cells and vortex vigorously. Incubate on ice 15 minutes. 5. Continue with the caspase assay or store the lysates at −70°C until use.
3.4.3.3
Measurement of Caspase-3 Activity
1. Thaw the samples on ice. 2. Spin the cell lysates for 15 minutes at 11,000g and 4°C. 3. Measure protein concentration in a lysate aliquot using a detergent-compatible kit. Calculate the volume containing 15 µg of protein. 4. Load the appropriate lysate volume per well in a 96-well black frame, clear flat bottomed plates. Complete to 20 µL with lysis buffer (see Note 7). 5. Load 20 µL of lysis buffer in two separate wells, to be used as blanks. 6. Prepare a standard curve by diluting the recombinant caspase-3 standard (Biomol, Plymouth Meeting, PA) 1:500 (corresponding to 2 Unit/well) and
4 Detection of Apoptosis of Bone Cells In Vitro
6. 7.
8. 9. 10.
63
make 1:2 serial dilutions (final volume, 10 µL/well). Add 10 µL of lysis buffer to each well of the standard curve to a final volume of 20 µL. Add 70 µL of assay buffer per well to blanks, samples and standard curve. Turn on the fluoroplate reader and the lamp 10 minutes before initiating the readings. Create a file in the fluoroplate reader software. Instructions are given for the Packard fluoroplate reader. Refer to the manual of your apparatus for setting up the reading. Dilute the caspase substrate DEVD-AFC (Biomol) 1:10 in assay buffer. Protect from light. Add 10 µL of diluted caspase substrate per well using a repeater or multichannel pipette. Put the plate inside the reader. Set the reading settings to the following: a. Pair of filters: excitation 400 nm, emission 510 nm. b. In the well of max RFU; write AUTO. At this step leave the default gain or PMT voltage.
11. Start to read and let the plate reader do several reading at different sensitivity settings until it determines which well has the highest relative fluorescence units (RFU). Go back to the read settings. In the slot test enter the name of the well with the highest RFU. If the reading is between 10,000 or 20,000, use this sensitivity to read your plate. If the reading is higher, decrease the sensitivity; if it is lower, increase it. 12. Read the plate automatically 25 times every 5 minutes. 13. Export the data to an Excel file. 14. Run a linear regression curve for each point of the standard curve. Select the fluorescence units as dependent (y) and the time as independent (x) variables. From each equation get the slope: y = a + bx, where b is the slope (relative fluorescent units/min). 15. Run a new linear regression curve with using the slope of the standard curve as dependent and the units of caspase-3 added/well as independent variables. Obtain a new equation: y´ = a´ + b´x´. 16. Run a linear regression curve to calculate the slope for each sample. 17. With the slope obtained in Step 17, calculate the units of caspase-3 for each sample using the following equation: Units caspase-3 = (slope for each sample-a´)/b´. 18. Divide the Units/well by the amount of protein in each well to obtain Units/µg protein.
3.4.3.4
Detection of the Active Caspase Form by Western Blotting
Activation of caspase-3 (as well as other caspases) can be detected by the appearance of the active fragment of the caspases. Thus, pro-caspase-3 has a molecular weight of 32 kDa and when it is activated, it is cleaved to render a fragment of 17 kDa. Antibodies that recognize both pro- and active caspase-3 have been generated
64
T. Bellido and L.I. Plotkin
and can be used to detect pro- and active caspase-3 by western blot analysis (31, 40). In addition, an antibody that recognizes specifically active caspase-3, but not pro-caspase (Cell Signaling Technology, Danvers, MA), can be used to determine caspase-3 activation by western blot analysis and by immunocytochemistry. This latter antibody has been used in our laboratory to determine the effect of PTH on the etoposide-induced apoptosis (see supplementary Fig. 1 in reference (2) ). 3.4.3.4.1 Sample Preparation The instructions are given for a BioRad Mini Protean III apparatus and should be adapted when other systems are used. 1. 2. 3. 4.
Thaw the cell lysates on ice. Spin the samples for 15 minutes at 11,000g and 4°C. Measure protein concentration using a detergent compatible kit. Determine the amount of lysate required to obtain 50 µg of protein. (This amount is for OB-6 cells. It can vary for other cell types.) 5. Pipette this amount to a labeled Eppendorf tube. Complete the volume to 20 µL with lysis buffer. Add 5 µL of 5× loading buffer/tube. 6. Boil samples and molecular weight markers for 5 minutes; spin them for 5 seconds. 7. Store on ice until ready to load onto SDS-PAGE gel. 3.4.3.4.2 SDS-Polyacrylamide Gel Electrophoresis 1. Set up the glass plates using 1.5-mm spacers. 2. Prepare 10% resolving gel by mixing 5 mL of 4× resolving gel buffer, with 6.7 mL of acrylamide/bis solution (RAS), 5 mL of water, 200 µL of 10% ammonium persulfate, and 20 µL of TEMED. (The amounts given are sufficient to prepare two gels.) Mix carefully, without making foam (see Note 8). Pour 7.5 mL of the gel and overlay with water. The gel should polymerize in about 30 minutes. (A clear separation between the gel and water can be seen when the gel is ready.) 3. Remove the extra water using vacuum. 4. Prepare the stacking gel by mixing 2.5 mL of 4× stacking gel buffer with 1.33 mL of acrylamide/bis-acrylamide solution (SAS), 6.15 mL of water, 50 µL of 10% ammonium persulfate, and 10 µL of TEMED. (The amounts given are sufficient to prepare two gels.) Mix carefully without making foam. Pour the stacking gel solution and insert the comb. The stacking gel should polymerize within 30 minutes. 5. Remove the comb, place the gels in the electrophoresis chamber, and fill the inner and outer reservoirs with running buffer. 6. Load the samples and 12.5 µL of the molecular weight markers/well (see Materials section for the preparation of the molecular weight markers). If you have empty wells, add 10 µL of 1× loading buffer to avoid distortion of the bands. 7. Run the gel at 100 V until a good separation between the markers corresponding to 14 and 30 kDa (approximately 1 cm) is obtained (approximately 2 hours).
4 Detection of Apoptosis of Bone Cells In Vitro
65
3.4.3.4.3 Electrophoretic Transfer 1. Chill 1L of transfer buffer at −20°C. 2. Cut a piece of 5.5 × 9 cm of PVDF membrane and two pieces of Whatman filter paper of the same size as the sponges. 3. Prepare the membrane by soaking in methanol for 1 to 2 minutes, followed by washing with water for 3 minutes. Place membrane in transfer buffer for at least 15 minutes to equilibrate. 4. After gel electrophoresis, set up the transfer unit: Submerge the cassette in transfer buffer. With the white face of the cassette away from you, place in ascending order: a. b. c. d. e. f.
Sponge. Whatman filter paper. Gel. PVDF membrane. Whatman paper. Sponge.
Close the cassette carefully to avoid introduction of bubbles into the arrangement. The proteins will not transfer to the PVDF membranes through the bubbles. Place the cassette in the chamber with the black side of the cassette facing the black side of the holder. Fill the chamber with the chilled transfer buffer and place the cooling block in the chamber. Transfer for 1 hour 15 minutes at 100 V. After transfer, remove the membrane from the cassette and perform western blotting. If desired, stain membrane with the non-permanent dye Ponceau red to visualize proteins before the western blot. For this, place the membrane in a container containing 6 mL 0.1% Ponceau red in 5% acetic acid for 1–2 minutes, and remove the excess dye by rinsing with distilled water. The remaining dye will be removed when incubating the membranes with blocking solution. 3.4.3.4.4 Western Blotting 1. Place membrane in blocking solution: 5% skim milk in TBS containing 0.05% Tween-20 (TBS-T). 2. Incubate at room temperature with shaking for at least 30 minutes. 3. Incubate the membrane with a 1:1,000 dilution of anti-active caspase-3 antibody (Cell Signaling Technology, Danvers, MA) in blocking solution overnight at 4°C. Make sure that the antibody solution is covering completely the membrane. Cover the container to avoid evaporation of the solution. 4. Wash the membranes three times, 10 minutes each with 10 mL of TBS-T. 5. Incubate the membranes with 1:2,000 dilutions of anti-rabbit-horseradish peroxidase (HRP) in blocking solution for 2 hours at room temperature. 6. Wash the membranes three times, 10 minutes each with 10 mL of TBS-T. 7. Wash the membranes once for 20 minutes with TBS (without Tween-20). 8. Mix equal volume of stable peroxide solution and luminol/enhancer solution. 9. Place the membrane in the mixture for ~5 minutes.
66
T. Bellido and L.I. Plotkin
10. Remove the membrane, drip the excess of reagents and put the membranes inside a plastic bag inside a cassette for autoradiography (tape the membrane down so that it does not shift during exposure) (see Note 9). 11. In a dark room, place film in the cassette and began timing the exposure. 12. A good initial time of exposure is about 20 seconds. From there, it can be increased or decreased as needed.
3.4.3.5
Detection of the Active Caspase Form by Immunocytochemistry
1. Cells can be grown of tissue culture plates, chambers, or coverslips. 2. Fix the cells for 15 minutes in 3.7% formaldehyde. 3. Incubate the cells for 1 hour at room temperature with 20% normal goat serum (see Note 10) to block non-specific binding. 4. Incubate the cells with anti-active caspase-3 antibody (Cell Signaling Technology) (1:75 in 2% normal goat serum) for 2 hours. 5. Incubate for 1 hour with biotin-conjugated anti-rabbit antibody (1:200), followed by a 30-minute incubation with avidin-TRITC or avidin-HRP (1:100) (see Note 11). 6. Score at least 250 cells expressing (fluorescent) and not expressing active capase-3 under a fluorescence microscope from fields selected by systematic random sampling. 7. Determine the percentage of fluorescent cells.
3.4.3.6 Detection of Caspase Activation by Determining the Subcellular Localization or Size of Fusion Proteins Containing Capase-3 Cleavage Sites Plate the cells at the appropriate density for transient transfection and culture overnight. The conditions for DNA transfection may vary among different cell types and need to be optimized in advance. The method described is for evaluating apoptotic cells under a fluorescence inverted microscope. If such microscope is not available, cells can be plated in chambers or coverslips inside a 24-well plate. 1. Plate the 3 × 104 MLO-Y4 cells/well in Growth Media in collagen-coated 48well plates in the late afternoon. Culture overnight. 2. Mix DNA-Plus reagent-Lipofectamine as indicated (equal amount of the fluorescent probe and the plasmid of interest should be mixed). a. Mix 2 µL of Plus reagent (Invitrogen), 0.05 µg of each plasmid (total amount/ well: 0.1 µg) and 50 µL of medium without serum and without PSG for each well. Incubate at room temperature for 15 minutes. b. Dilute 0.6 µL of Lipofectamine (Invitrogen) in 100 µL of medium without serum and without PSG for each well and transfer to the tube containing plasmid DNA and Plus reagent. Incubate at room temperature for 15 minutes.
4 Detection of Apoptosis of Bone Cells In Vitro
67
3. Remove the medium from the wells containing the cells, wash once with 300 µL of medium without serum and without PSG and transfer the DNA-Plus reagent-Lipofectamine mixture (150 µL/well). Incubate for approximately 3 hours. If the cells start to look distressed, stop the transfection immediately. 4. To stop the transfection, add 150 µL/well of 2× of growing medium (α-MEM containing 5% FBS, 5% BCS, and 1% PSG). Culture overnight. 5. Check the cells under fluorescent microscope to determine the efficiency of the transfection. It should be about 40% for osteoblastic or osteocytic cells. Change the medium to normal growing medium and culture overnight. 6. Remove the medium and add fresh growing medium. 7. Treat as desired. 8. Remove the medium and wash once with 250 µL/well of PBS. 9. Fix the cells for 8 min (see Note 12) with 250 µL/well of 3.7% formaldehyde in PBS. 10. Remove the fixative and wash once with 250 µL/well of PBS. Add 500 µL/well of PBS containing 0.01% thimerosal to avoid bacterial growth. Store at 4°C.
3.4.3.7
Caspase-3 Sensor YFP-DEVD
Caspase-3 activation can be determined in transfected cells by analyzing the subcellular localization of a caspase-3 sensor (YFP-caspase-3; Clontech). The sensor protein contains a dominant N-terminal nuclear export signal, a caspase-3 cleavage site (DEVD), a yellow fluorescent protein (YFP), and a C-terminal nuclear localization signal. When caspase-3 is inactive, the nuclear export sequence prevails and the fluorescent protein is located in the cytoplasm. When caspase-3 is activated, it cleaves the protein at the DEVD sequence, removing the nuclear export sequence, and as a consequence, the fluorescent protein is located in the nucleus. Cells are transfected with the caspase-3 sensor, and the plasmid(s) encoding the protein(s) of interest together with nuclear targeted fluorescent protein (red or cyan) to allow the visualization of the cell nuclei. The percentage of cells expressing nuclear localization of YFP-DEVD is determined under a fluorescence microscope. At least 250 cells from fields selected by systematic random sampling should be scored (21, 25, 30).
3.4.3.8
c-myc-CFP-DEVD-YFP
This method consists of transfecting the cells with a plasmid encoding a fusion protein, comprising cyan fluorescent protein (CFP) linked to YFP by an 18 amino acid peptide containing the caspase-3 cleavage sequence, DEVD (41). The fusion protein also contains a c-myc tag in the N-terminus, to facilitate its detection using anti-c-myc antibodies. In living cells, the c-myc-CFP-DEVD-YFP protein is intact (60 kDa) and when caspase-3 is activated, it cleaves the c-myc-CFP-DEVD-YFP, rendering two fragments of approximately 30 kDa. Both the intact protein and
68
T. Bellido and L.I. Plotkin
fragments can be detected using anti-c-myc antibody or an anti-green fluorescent protein (GFP) antibody (which recognizes both CFP and YFP). Cells are transfected as indicated in Section 3.4.3.6. Usually, cells are plated in 10 cm plates to obtain enough material to perform the western blot analysis. After transfecting the cells, perform treatment and prepare cell lysates as indicated for detection of active caspase-3 by western blot analysis (Section 3.4.3.4.), with the following modifications: 1. Separate 15 µg proteins (see Note 13) in a 10% gel resolving gel. Prepare the gel by mixing 5 mL of 4× resolving gel buffer, with 6.7 mL of acrylamide/bis solution (RAS), 8 mL of water, 200 µL of 10% ammonium persulfate solution, and 20 µL of TEMED. 2. After blocking, incubate the membrane with anti-GFP antibody 1:2,000 or antic-myc antibody 1:1,000 overnight at 4°C. 3. Continue with the corresponding secondary antibody. This construct can also be used to detect apoptosis by fluorescence resonance energy transfer (FRET). Thus, in intact fusion proteins YFP and CFP are in close proximity and the emission spectrum from CFP overlaps with the excitation spectrum of YPF, allowing FRET to occur. When the protein is cleaved, CFP and YFP are no longer associated, resulting in an increase in CFP/YFP emission rate and a decrease in FRET (42).
3.4.4
Detection of Proteolytic Products of Caspases
PARP (poly (ADP-ribose) polymerase) is a 120-kDa nuclear enzyme involved in DNA repair. This protein is a target of the caspases and it is cleaved during the apoptosis process, resulting in two fragments of 89 and 24 kDa that can be detected by western blot analysis (43). Perform treatment and prepare cell lysates as indicated for detection of active caspase-3 by western blot (see Section 3.4.3.4), with the following modifications: 1. Separate the proteins in an 8% gel resolving gel. Prepare the gel by mixing 5 mL of 4× resolving gel buffer, with 5.4 mL of acrylamide/bis solution, 9.3 mL of water, 200 µL of 10% ammonium persulfate solution, and 20 µL of TEMED. 2. Block for 2 hours at room temperature in 5% skim milk in PBS containing 0.1% Tween-20. 3. Incubate with 1:4,000 dilution of rabbit polyclonal antibody against PARP overnight at 4°C. 4. Wash the membrane five times with PBS containing 0.1% Tween-20. 5. Incubate the membranes with the corresponding HRP-conjugated anti-immunoglobulin antibody in blocking buffer for 1 hour.
4 Detection of Apoptosis of Bone Cells In Vitro
3.4.5
69
Changes in Plasma Membrane Permeability
Increased plasma membrane permeability that occurs during the apoptosis process can be visualized using trypan blue, a dye that is excluded from living cells and is incorporated into the cells when they are undergoing apoptosis (2, 7, 9, 11–13, 20, 21, 44). 1. Plate 1 × 104 cell/well MLO-Y4 cells in growing medium, in collagen-coated 48-well plate. Culture overnight. 2. Remove the medium and add fresh growing medium and treat the cells as desired. 3. Harvest cells by trypsinization: a. Remove the medium containing the floating cells and put it in Eppendorf tubes previously labeled. (Put the medium from each well in a separate tube.) b. Add 0.25 mL of trypsin/well and incubate in the incubator until the cells detach (typically 5 minutes) (see Note 14). 4. Once the cells have detached, add 0.25 mL of growing medium/well. Collect the cells by pipetting up and down the medium. Mix with the culture supernatant corresponding to the well. 5. Spin 2 minutes at 6,300g and remove the supernatant carefully. 6. Add 25 µL of 0.04% trypan blue in PBS/tube. (Dilute the 0.4% stock solution 1:10 in PBS.) Resuspend and transfer 10 µL of the cell suspension to a hemocytometer. Count living (clear) and dead (blue) cells and determine the percentage of death cells for each condition. 3.4.6 3.4.6.1
Changes in Nuclear Morphology Staining with Nuclear Fluorescent Dyes (Hoechst 33258 Staining)
1. Fix the cells for 15 minutes with 3.7% formaldehyde in ethanol. 2. Wash the cells with PBS, 5 minutes at room temperature. 3. Prepare a stock solution of 1.9 mM Hoechst 33258 fluorescent dye (Polysciences, Inc., Warrington, PA) in distilled water. 4. Dilute the dye to a final concentration of 1.9 µΜ in 0.5 M NaCl, 10 mM TrisHCl, 1 mM EDTA, pH 7.4. 5. Overlay the cells with dye solution. Incubate for 2 minutes at room temperature in the dark. 6. Wash with PBS, 5 minutes at room temperature and: a. Add 0.01% thimerosal to the cells to avoid bacterial contamination if they were cultured in culture plates. b. Remove the PBS, add mounting media and coverslip if the cells were grown on chambers. 7. Evaluate the pyknotic fragmented nuclei typical of apoptotic cells under a fluorescence microscope. Score the percentage of apoptotic cells.
70
3.4.6.2
T. Bellido and L.I. Plotkin
Staining with TRAPase-Hematoxylin (for Osteoclasts)
After preparing osteoclasts and treating them as desired, spin down the plates for 10 minutes at 900g. If an adaptor to centrifuge the whole plate is not available, collect supernatants with floating cells and wash adherent cells twice with 1 mL PBS. Pool supernatant and washings for each well. Spin to collect floating cells and resuspend in 150 µL of PBS. Prepare cytospins on poly-L-lysine coated glass microscope slides by spinning at 300g for 10 minutes. 3.4.6.2.1 Fixation 1. Remove the medium and dry completely the plate (and the cytospin slides). 2. Add 0.5 ml of 3.7% formaldehyde in PBS/well for a 24-well plate. Incubate for 10–15 minutes at room temperature. 3. Rinse three times with tap water and set right side down to dry completely. The plates can be stored at room temperature. 3.4.6.2.2 Staining The amounts indicated are enough for four 24-well plates. Modify the volumes accordingly if you need to stain more or fewer plates. 1. Prepare 1 mL of a solution of 42 mg/mL NaNO2 in distilled water. 2. Mix well 0.75 ml of NaNO2 solution with 0.75 mL of pararosaniline. Incubate for 5 minutes at room temperature. 3. Add the mixture NaNO2 – pararosaniline to 21 mL of NaK tartrate and sodium acetate pH 5.0. 4. Add 2.5 mL of Naphtol AS-BI phosphoric acid. 5. Add 0.25 mL/well of the solution. Incubate for 20–30 minutes at 37°C, until red color develops. 6. Wash three times with distilled water. 7. Air dry right side up. 8. Prepare a 1:1 dilution of hematoxylin with water. Add 0.250 mL/well. Incubate for 1 minute at room temperature. Alternatively, cells can be stained with 0.2 mM Hoechst 33258 dye (39, 45). 9. Wash five times with tap water. Dry and store at room temperature. 10. Count TRAP + cells, distinguish between mono and multinuclear, and also between apoptotic and non-apoptotic. Apoptotic cells present condensed, dark nuclei. Express as percentage of apoptotic osteoclasts (46, 47).
3.4.6.3
Transfection with Nuclear Fluorescent Proteins
Changes in nuclear morphology exhibited by apoptotic cells can be better determined in cells expressing fluorescent proteins containing a nuclear localization sequence.
4 Detection of Apoptosis of Bone Cells In Vitro
71
Several fluorescent proteins have been developed. We have added a nuclear localization sequence to GFP (7, 29), and to red fluorescent protein (RFP) (13), and both proteins have been used to identified apoptotic cells (see Note 15). Cells can be transiently transfected with the plasmid encoding for the fluorescent protein targeted to the nucleus together with the plasmid(s) encoding for the protein(s) of interest, as indicated in Section 3.4.3.6. (steps 1–9). This method allows the quantification of apoptosis only in transfected cells that will be fluorescent. Alternatively, cells can be stably transfected with the fluorescent protein targeted to the nucleus and treated with pharmacologic inhibitors, anti- and pro-apoptotic stimuli (7). The percentage of cells exhibiting chromatin condensation and nuclear fragmentation is determined under a fluorescence microscope. At least 250 cells from fields selected by systematic random sampling should be examined for each experimental condition.
3.4.7
Other Methods to Measure Apoptosis of Bone Cells
Table 4.4 summarizes other methods reported in the literature that have been used to measure apoptosis of osteoblasts, osteocytes, and/or osteoclasts.
4
Notes
1. When preparing RAS and SAS, add the water slowly in a graduated cylinder to make sure that the final volume is not higher than desired. Acrylamide-bis-acrylamide mixture expands in contact with water. 2. OB-6 cells are resistant to apoptosis in the presence of 10% serum. Therefore, to induce apoptosis the amount of serum has to be reduced to 2% at least 3 hours before adding the pro-apoptotic stimuli. 3. Serum is very important for osteoclast development. Use the newest batch available, aliquot it as soon as arrived and store it at −70°C. Thaw one aliquot Table 4.4 Other methods used to measure apoptosis in bone cells Phenomenon Method Changes in mitochondrial permeability Presence of small, aqueous soluble DNA fragments Presence of cytoplasmic histoneassociated DNA fragments Translocation of phosphatidylserine from the inner side to the outer side of the plasma membrane
References
Mitochondria membrane potential
(48–50)
Cytochrome C release FACS analysis and detection of hypodiploid DNA content ELISA (Roche Applied Science, Indianapolis, IN)
(40, 51–53) (54–56)
Annexin V staining
(16, 24, 59, 60)
(57–59)
72
T. Bellido and L.I. Plotkin
at a time, and prepare just the amount of medium required for the experiment. Keep medium at 4°C and use it within 1 week. 4. It is important to start the treatments as soon as osteoclasts develop. Osteoclasts will start to die spontaneously in culture approximately 24 hours after developing, even in the presence of M-CSF and sRANKL. This increases basal levels of apoptosis and makes it difficult to detect changes in viability induced by the stimuli of interest. 5. The number of cells required to detect caspase-3 activity varies depending on the cell type. We found that at least a well of a 6-well plate is required to detect caspase-3 activity in MLO-Y4 cells. In the case of OB-6 cells, caspase-3 activity can be detected in lysates obtained from a well in a 24-well plate. 6. The whole plate can be spun down if a microplate adaptor for the centrifuge is available. If this is the case, following treatment, plates are spun down for 5 minutes. (Please refer to the centrifuge and adaptor manuals to determine the speed for spinning.) The supernatants are discarded and cell lysates are prepared as indicated. 7. If the volume required to have 30 µg is >20 µL, reduce the volume of assay buffer to get a final volume of 90 µL. Do not use >30 µL of sample and maintain the final volume of the standard to the same value as the samples (i.e., if you add 25 µL of sample, add 15 µL lysis buffer to the standards). 8. Foam will incorporate oxygen in the solution and inhibit the polymerization of the gel. 9. An image analysis system able to detect chemiluminescence can be used instead of film. This method has the advantage that the images are directly stored as electronic files, avoiding the need to scan the film. However, the sensitivity is lower and some weak bands may not be detected using image analysis. 10. The normal serum use to block has to be from the species in which the secondary antibody was raised. 11. The biotin-avidin system allows the amplification of the system, obtaining a better signal. However, FITC-labeled antibodies can be used if biotinylated antibodies are not available or the signal is strong. Alternatively, an HRPlabeled secondary antibody can be used, followed by addition of a peroxidase substrate such as diaminobenzidine. 12. Do not leave with fixative for more than 10 minutes, as fluorescence will be reduced. Four percent p-formaldehyde (15 minutes at room temperature) can also be used to fix the cells. Do not use methanol or acetone, as these fixatives destroy the fluorescent proteins. 13. The amount of protein to load depends on the efficiency of transfection: The higher the efficiency, the lower the amount of total lysate that needs to be loaded in the gel. 14. Make sure you do not leave trypsin in too long, as cells will start dying. 15. GFP targeted to the nucleus is also expressed in the cytoplasm, although with lower intensity than in the nucleus. This property can be used to evaluate cell morphology by quantifying the number of cytoplasmic projections per cell. The expression of nuclear-targeted RFP (and cyan fluorescent protein, commercially available) is restricted to the nucleus and cannot be detected in the cytoplasm; therefore, they cannot be used to evaluate cell morphology.
4 Detection of Apoptosis of Bone Cells In Vitro
73
References 1. Lecka-Czernik, B., Gubrij, I., Moerman, E. A., et al. (1999) Inhibition of Osf2/Cbfa1 expression and terminal osteoblast differentiation by PPAR-gamma 2. J Cell Biochem 74, 357–371. 2. Bellido, T., Ali, A. A., Plotkin, L. I., et al. (2003) Proteasomal degradation of Runx2 shortens parathyroid hormone-induced anti-apoptotic signaling in osteoblasts. A putative explanation for why intermittent administration is needed for bone anabolism. J Biol Chem 278, 50259–50272. 3. Kato, Y., Windle, J. J., Koop, B. A., et al. (1997) Establishment of an osteocyte-like cell line, MLO-Y4. J Bone Min Res 12, 2014–2023. 4. Hsu, H., Lacey, D. L., Dunstan, C. R., et al. (1999) Tumor necrosis factor receptor family member RANK mediates osteoclast differentiation and activation induced by osteoprotegerin ligand. Proc Natl Acad Sci U S A 96, 3540–3545. 5. Huang, L., Xu, J., Wood, D. J., et al. (2000) Gene expression of osteoprotegerin ligand, osteoprotegerin, and receptor activator of NF-kappaB in giant cell tumor of bone: possible involvement in tumor cell-induced osteoclast-like cell formation. Am J Pathol 156, 761–767. 6. Jilka, R. L., Weinstein, R. S., Bellido, T., et al. (1999) Increased bone formation by prevention of osteoblast apoptosis with parathyroid hormone. J Clin Invest 104, 439–446. 7. Plotkin, L. I., Weinstein, R. S., Parfitt, A. M., et al. (1999) Prevention of osteocyte and osteoblast apoptosis by bisphosphonates and calcitonin. J Clin Invest 104, 1363–1374. 8. Selander, K. S., Harkonen, P. L., Valve, E., et al. (1996) Calcitonin promotes osteoclast survival in vitro. Mol Cell Endocrinol 122, 119–129. 9. Chen, J. R., Plotkin, L. I., Aguirre, J. I., et al. (2005) Transient versus sustained phosphorylation and nuclear accumulation of ERKs underlie anti- versus pro-apoptotic effects of estrogens. J Biol Chem 280, 4632–4638. 10. Kousteni, S., Chen, J. R., Bellido, T., et al. (2002) Reversal of bone loss in mice by nongenotropic signaling of sex steroids. Science 298, 843–846. 11. Kousteni, S., Han, L., Chen, J. R., et al. (2003) Kinase-mediated regulation of common transcription factors accounts for the bone-protective effects of sex steroids. J Clin Invest 111, 1651–1664. 12. Kousteni, S., Bellido, T., Plotkin, L. I., et al. (2001) Nongenotropic, sex-nonspecific signaling through the estrogen or androgen receptors: dissociation from transcriptional activity. Cell 104, 719–730. 13. Plotkin, L. I., Aguirre, J. I., Kousteni, S., et al. (2005) Bisphosphonates and estrogens inhibit osteocyte apoptosis via distinct molecular mechanisms downstream of ERK activation. J Biol Chem 280, 7317–7325. 14. Vertino, A. M., Bula, C. M., Chen, J. R., et al. (2005) Nongenotropic, anti-apoptotic signaling of 1alpha,25(OH)2-vitamin D3 and analogs through the ligand binding domain of the vitamin D receptor in osteoblasts and osteocytes. Mediation by Src, phosphatidylinositol 3-, and JNK kinases. J Biol Chem 280, 14130–14137. 15. Bellido, T., O’Brien, C. A., Roberson, P. K., et al. (1998) Transcriptional activation of the p21(WAF1,CIP1,SDI1) gene by interleukin-6 type cytokines. A prerequisite for their pro-differentiating and anti-apoptotic effects on human osteoblastic cells. J Biol Chem 273, 21137–21144. 16. Bodine, P. V., Billiard, J., Moran, R. A., et al. (2005) The Wnt antagonist secreted frizzledrelated protein-1 controls osteoblast and osteocyte apoptosis. J Cell Biochem 96, 1212–1230. 17. Karsdal, M. A., Larsen, L., Engsig, M. T., et al. (2002) Matrix metalloproteinase-dependent activation of latent transforming growth factor-beta controls the conversion of osteoblasts into osteocytes by blocking osteoblast apoptosis. J Biol Chem 277, 44061–44067. 18. Fuller, K., Lean, J. M., Bayley, K. E., et al. (2000) A role for TGFbeta(1) in osteoclast differentiation and survival. J Cell Sci 113, 2445–2453.
74
T. Bellido and L.I. Plotkin
19. Hill, P. A., Tumber, A., Meikle, M. C. (1997) Multiple extracellular signals promote osteoblast survival and apoptosis. Endocrinology 138, 3849–3858. 20. Plotkin, L. I., Manolagas, S. C., Bellido, T. (2002) Transduction of cell survival signals by connexin-43 hemichannels. J Biol Chem 277, 8648–8657. 21. Plotkin, L. I., Manolagas, S. C., Bellido, T. (2006) Dissociation of the pro-apoptotic effects of bisphosphonates on osteoclasts from their anti-apoptotic effects on osteoblasts/osteocytes with novel analogs. Bone 39, 443–452. 22. Weinstein, R. S., Chen, J. R., Powers, C. C., et al. (2002) Promotion of osteoclast survival and antagonism of bisphosphonate-induced osteoclast apoptosis by glucocorticoids. J Clin Invest 109, 1041–1048. 23. Kogianni, G., Mann, V., Ebetino, F., et al. (2004) Fas/CD95 is associated with glucocorticoidinduced osteocyte apoptosis. Life Sci 75, 2879–2895. 24. Ahuja, S. S., Zhao, S., Bellido, T., et al. (2003) CD40 ligand blocks apoptosis induced by tumor necrosis factor alpha, glucocorticoids, and etoposide in osteoblasts and the osteocytelike cell line murine long bone osteocyte-Y4. Endocrinology 144, 1761–1769. 25. Almeida, M., Han, L., Bellido, T., et al. (2005) Wnt proteins prevent apoptosis of both uncommitted osteoblast progenitors and differentiated osteoblasts by beta-catenin-dependent and independent signaling cascades involving Src/ERK and phosphatidylinositol 3-kinase/AKT. J Biol Chem 280, 41342–41351. 26. Plotkin, L. I., Mathov, I., Aguirre, J. I., et al. (2005) Mechanical stimulation prevents osteocyte apoptosis: requirement of integrins, Src kinases and ERKs. Am J Physiol Cell Physiol 289, C633–C643. 27. Fatokun, A. A., Stone, T. W., Smith, R. A. (2006) Hydrogen peroxide-induced oxidative stress in MC3T3-E1 cells: the effects of glutamate and protection by purines. Bone 39, 542–551. 28. Park, B. G., Yoo, C. I., Kim, H. T., et al. (2005) Role of mitogen-activated protein kinases in hydrogen peroxide-induced cell death in osteoblastic cells. Toxicology 215, 115–125. 29. Bellido, T., Huening, M., Raval-Pandya, M., et al. (2000) Calbindin-D28k is expressed in osteoblastic cells and suppresses their apoptosis by inhibiting caspase-3 activity. J Biol Chem 275, 26328–26332. 30. Liu, Y., Porta, A., Peng, X., et al. (2004) Prevention of glucocorticoid-induced apoptosis in osteocytes and osteoblasts by calbindin-D28k. J Bone Miner Res 19, 479–490. 31. Pavalko, F. M., Gerard, R. L., Ponik, S. M., et al. (2003) Fluid shear stress inhibits TNFalpha-induced apoptosis in osteoblasts: a role for fluid shear stress-induced activation of PI3kinase and inhibition of caspase-3. J Cell Physiol 194, 194–205. 32. Bakker, A., Klein-Nulend, J., Burger, E. (2004) Shear stress inhibits while disuse promotes osteocyte apoptosis. Biochem Biophys Res Commun 320, 1163–1168. 33. Kitase, Y., Jiang, J. X., Johnson, M. L., et al. (2006) The anti-apoptotic effects of mechanical strain on osteocytes are mediated by PGE2 and monocyte chemotactic protein-3 (MCP-3): selective protection by MCP-3 against glucocorticoid (GC), but not TNF-a induced apoptosis. J Bone Miner Res 21, S48. 34. Alessi, D. R., Cuenda, A., Cohen, P., et al. (1995) PD 098059 is a specific inhibitor of the activation of mitogen-activated protein kinase kinase in vitro and in vivo. J Biol Chem 270, 27489–27494. 35. Nicholson, D. W., Ali, A., Thornberry, N. A., et al. (1995) Identification and inhibition of the ICE/CED-3 protease necessary for mammalian apoptosis. Nature 376, 37–43. 36. Jilka, R. L., Weinstein, R. S., Bellido, T., et al. (1998) Osteoblast programmed cell death (apoptosis): modulation by growth factors and cytokines. J Bone Min Res 13, 793–802. 37. Frisch, S. M., Vuori, K., Ruoslahti, E., et al. (1996) Control of adhesion-dependent cell survival by focal adhesion kinase. J Cell Biol 134, 793–799. 38. Murakami, T., Yamamoto, M., Ono, K., et al. (1998) Transforming growth factor-beta1 increases mRNA levels of osteoclastogenesis inhibitory factor in osteoblastic/stromal cells and inhibits the survival of murine osteoclast-like cells. Biochem Biophys Res Commun 252, 747–752. 39. Kameda, T., Mano, H., Yuasa, T., et al. (1997) Estrogen inhibits bone resorption by directly inducing apoptosis of the bone-resorbing osteoclasts. J Exp Med 186, 489–495.
4 Detection of Apoptosis of Bone Cells In Vitro
75
40. Chua, C. C., Chua, B. H., Chen, Z., et al. (2003) Dexamethasone induces caspase activation in murine osteoblastic MC3T3-E1 cells. Biochim Biophys Acta 1642, 79–85. 41. Mahajan, N. P., Harrison-Shostak, D. C., Michaux, J., et al. (1999) Novel mutant green fluorescent protein protease substrates reveal the activation of specific caspases during apoptosis. Chem Biol 6, 401–409. 42. Tyas, L., Brophy, V. A., Pope, A., et al. (2000) Rapid caspase-3 activation during apoptosis revealed using fluorescence-resonance energy transfer. EMBO Rep 1, 266–270. 43. Mohr, S., McCormick, T. S., Lapetina, E. G. (1998) Macrophages resistant to endogenously generated nitric oxide-mediated apoptosis are hypersensitive to exogenously added nitric oxide donors: dichotomous apoptotic response independent of caspase 3 and reversal by the mitogen-activated protein kinase kinase (MEK) inhibitor PD 098059. Proc Natl Acad Sci U S A 95, 5045–5050. 44. O’Brien, C. A., Jia, D., Plotkin, L. I., et al. (2004) Glucocorticoids act directly on osteoblasts and osteocytes to induce their apoptosis and reduce bone formation and strength. Endocrinology 145, 1835–1841. 45. Hiroi-Furuya, E., Kameda, T., Hiura, K., et al. (1999) Etidronate (EHDP) inhibits osteoclasticbone resorption, promotes apoptosis and disrupts actin rings in isolate-mature osteoclasts. Calcif Tissue Int 64, 219–223. 46. Hughes, D. E., Wright, K. R., Uy, H. L., et al. (1995) Bisphosphonates promote apoptosis in murine osteoclasts in vitro and in vivo. J Bone Min Res 10, 1478–1487. 47. Hughes, D. E., Dai, A., Tiffee, J. C., et al. (1996) Estrogen promotes apoptosis of murine osteoclasts mediated by TGF-beta. Nat Med 2, 1132–1136. 48. Chang, C. C., Liao, Y. S., Lin, Y. L., et al. (2006) Nitric oxide protects osteoblasts from oxidative stress-induced apoptotic insults via a mitochondria-dependent mechanism. J Orthop Res 24, 1917–1925. 49. Wiren, K. M., Toombs, A. R., Semirale, A. A., et al. (2006) Osteoblast and osteocyte apoptosis associated with androgen action in bone: requirement of increased Bax/Bcl-2 ratio. Bone 38, 637–651. 50. Ho, W. P., Chen, T. L., Chiu, W. T., et al. (2005) Nitric oxide induces osteoblast apoptosis through a mitochondria-dependent pathway. Ann NY Acad Sci 1042, 460–470. 51. Kim, G. S., Hong, J. S., Kim, S. W., et al. (2003) Leptin induces apoptosis via ERK/cPLA2/ cytochrome c pathway in human bone marrow stromal cells. J Biol Chem 278, 21920–21929. 52. Oursler, M. J., Bradley, E. W., Elfering, S. L., et al. (2005) Native, not nitrated, cytochrome c and mitochondrial-derived hydrogen peroxide drive osteoclast apoptosis. Am J Physiol Cell Physiol 288, C156–C168. 53. Wu, X., McKenna, M. A., Feng, X., et al. (2003) Osteoclast apoptosis: the role of Fas in vivo and in vitro. Endocrinology 144, 5545–5555. 54. Qin, Y. J., Zhang, Z. L., Yu, L. Y., et al. (2006) A20 overexpression under control of mouse osteocalcin promoter in MC3T3-E1 cells inhibited tumor necrosis factor-alpha-induced apoptosis. Acta Pharmacol Sin 27, 1231–1237. 55. Contardi, E., Palmisano, G. L., Tazzari, P. L., et al. (2005) CTLA-4 is constitutively expressed on tumor cells and can trigger apoptosis upon ligand interaction. Int J Cancer 117, 538–550. 56. Postiglione, L., Di, D. G., Ramaglia, L., et al. (2003) Behavior of SaOS-2 cells cultured on different titanium surfaces. J Dent Res 82, 692–696. 57. Alikhani, M., Alikhani, Z., Boyd, C., et al. (2007) Advanced glycation end products stimulate osteoblast apoptosis via the MAP kinase and cytosolic apoptotic pathways. Bone 40, 345–353. 58. Kim, H. J., Zhao, H., Kitaura, H., et al. (2006) Glucocorticoids suppress bone formation via the osteoclast. J Clin Invest 116, 2152–2160. 59. Mogi, M., Togari, A. (2003) Activation of caspases is required for osteoblastic differentiation. J Biol Chem 278, 47477–47482. 60. Wang, Y.H., Liu, Y., Rowe, D.W. (2007) Effects of transient PTH on early proliferation, apoptosis, and subsequent differentiation of osteoblast in calvarial osteoblast cultures. Am J Physiol Endocrinol Metab 292, E594–E603.
Chapter 5
In Vivo Parathyroid Hormone Treatments and RNA Isolation and Analysis Xin Li, Ling Qin, and Nicola C. Partridge
Abstract Intermittent parathyroid hormone (PTH) administration increases bone formation and bone mass and is being used as a therapy for osteoporosis. By contrast, chronic hyperparathyroidism results in the metabolic bone disease osteitis fibrosa characterized by local bone resorption and peritrabecular bone marrow fibrosis in humans. The animal models that can mimic the paradoxical effects of PTH provide the basis for further study of the functions of this hormone in skeletal tissues. In both rats and mice, the anabolic effects of PTH on bone can be achieved by daily injections subcutaneously and the catabolic effects can be achieved by continuous infusion with osmotic pumps. This chapter offers detailed information, such as the dosage and preparation of PTH, using the example of treatment of rats with PTH intermittently or continuously. High-quality, RT-PCR ready RNA is required for the analysis of gene expression. For the analysis of gene expression in bone, usually long bones are used for RNA extraction. Here we describe how to extract RNAs from the metaphyseal trabecular primary spongiosa of rat femur by a method based on two commercially available kits. This protocol can be used in other tissues with slight modification of the amount of reagent used according to the tissue size. Keywords PTH, injection, infusion, RNA, bone, in vivo.
1
Introduction
Excess PTH as seen in primary hyperparathyroidism (PHPT) has been long known for its catabolic effect to promote osteoclast activity and bone resorption and considered as a cause of secondary osteoporosis (1). Later studies have shown that PTH also has an anabolic effect on the mammalian skeleton showing stimulation of new bone formation in humans and experimental animals (2–5). Intermittent PTH application is an established pharmacological principle for its capability to stimulate bone formation. Catabolic effects of PTH can be achieved by infusion with implanted ALZET osmotic pumps continuously delivering PTH. The animal models mimic PTH’s dual effects and are a basis for further investigations
From: Methods in Molecular Biology, Vol. 455: Osteoporosis: Methods and Protocols Edited by Jennifer J. Westendorf © Humana Press, Totowa, NJ
79
80
X. Li et al.
of the molecular mechanisms underlying the effects of this hormone. Performing microarray analysis with RNA isolated from these animal models provided entire gene expression profiles associated with the anabolic or catabolic effects of the two different PTH treatments. Bone is a hard tissue and rich in degradative enzymes, which makes RNA extraction difficult. We obtained high-quality RNA from rat (and mouse) long bones using a method based on two commercially available kits. Rat femora were resected with all connective tissue removed completely. A region subjacent to the growth plate about 3 mm wide of the metaphyseal trabecular primary spongiosa was stored in RNAlater reagent (Qiagen, Valencia, CA) for RNA isolation using Tri Reagent (Sigma, St. Louis, MO). A RNA cleanup kit (Qiagen) was used to further purify the RNA after isolation to increase the quality of the RNA.
2 2.1
Materials Equipment
1. Sterile syringes, needles, scissors, hemostats, clips, sterile scalpels, forceps, and bone cutter. 2. A scale to weigh animals and a balance to weigh osmotic pumps. 3. ALZET osmotic pumps (DURECT Corp., Cupertino, CA). 4. Isoflurane anesthetic system to anesthetize rats and perform the implantation of pumps. 5. PT 10-35 Polytron homogenizer (Kinematica GA, Littau, Switzerland) to pulverize bone samples. 6. UV spectrophotometer and quartz cuvettes.
2.2 1. 2. 3. 4. 5.
Reagents
Synthetic PTH (1–34) (human) (Bachem, King of Prussia, PA). Saline. Sprague-Dawley rats. Heat-inactivated sera from rats of the same gender and age without treatments. Diethylpyrocarbonate (DEPC)–treated water: Add DEPC to distilled water to a final concentration of 0.1% and incubate at room temperature overnight. Autoclave the next day to inactivate the DEPC. 6. Sodium hydroxide (1 N). 7. Chloroform. 8. Ethanol, 75% and 96–100%.
5 PTH In Vivo and RNA Isolation
9. 10. 11. 12. 13. 14.
15. 16. 17. 18. 19.
3 3.1
81
Isopropanol. β-Mercaptoethanol (β-ME), 100% (see Note 1). RNAlater RNA Stabilization Reagent (Qiagen, Valencia, CA). Tri Reagent (Sigma). RNeasy kit (Qiagen). 5 × MOPS buffer: 0.4 M MOPS (morpholinopropanesulfonic acid), 100 mM sodium acetate, 10 mM EDTA. The buffer is adjusted to pH 7.0 with 1 M NaOH and sterilized by autoclaving. Agarose. Formaldehyde. Ethidium bromide, 10 mg/mL. RNA denaturation buffer: 10 mL 100% deionized formamide, 3.5 mL 40% formaldehyde, 1.5 mL 5 × MOPS buffer. 10 × RNA loading buffer: 50% glycerol, 0.4% bromophenol blue, and 0.4% xylene cyanol.
Methods PTH Preparation
1. The synthetic PTH peptide is a lyophilized white powder when purchased and must be dissolved in sterile saline to make an aliquot stock (50 ng/µL), which should be stored at or below −20 °C. 2. Make vehicle solution, saline with 1% heat-inactivated sera: Collect blood from untreated animals of the same gender and age as the animals used in the experiments. Leave the blood at room temperature for at least 30 minutes before centrifuging at 250 g for 10 minutes to separate sera. Heat sera at 56 °C for 30 minutes. Dilute with saline to the final concentration of 1%. Use a 0.20 micron sterile syringe filter (Corning) to make the solution sterile. 3. Weigh and record body weight of each rat to calculate the amount of PTH used in each rat. For injections, the final dose of PTH was 8 µg/100 g (rat body weight) at a final volume of 200 µL for each injection. For infusion, the dose of PTH was 4 µg/100 g (rat body weight) each day. Add vehicle to the PTH stock solution before loading for injections and infusions.
3.2
Injections
Use a 1-mL syringe with 26G 3/8 needle to inject the rats subcutaneously into the hind region.
82
X. Li et al.
3.3
Infusions
3.3.1
Loading Pump (see Notes 2–5)
1. Weigh the empty pump together with its flow moderator. 2. With the flow moderator removed, hold the pump in an upright position, and insert the filling tube through the opening at the top of the pump until it can go no further (see Note 6). 3. Use the smallest volume syringe to fill through a 0.20-µ syringe-end filter; holding the pump in an upright position, slowly push the plunger of the syringe (see Note 7). Stop filling when the solution appears at the outlet and carefully remove the tube. 4. Wipe off the excess solution and insert the flow moderator until the cap is flush with the top of the pump (see Note 8). 5. Weigh the filled pump with its flow moderator in place. The filled volume should be over 90% of the reservoir volume (see Note 9). 6. Place each loaded pump in a tube filled with sterile saline and incubate at 37 °C overnight (or at least for 6 hours) to normalize the flow speed before implantation into rats.
3.3.2
Implantation Surgery
1. Anesthetize rats by isoflurane at the induction concentration of 4.5%, then at maintenance concentration of 2% isoflurane during the surgery. 2. Make a 1-cm-long mid-scapular incision on the back of each rat. Insert a hemostat into the incision and spread the subcutaneous tissue to create a pocket for the ALZET osmotic pump. 3. Insert a filled pump into the pocket and the close the wound skin with two clips.
3.4
Long Bone Collection
Euthanize animals with CO2 immediately after the infusion has ceased or at the indicated time after injection. One or both femora can be collected for RNA extraction (6) (see Note 10). 1. Cut the skin with a scissor to expose one femur. Resect one femur with a bone cutter. Completely remove all connective tissue, including periosteum, with a scalpel. 2. Place the femur on a sterile 100-mm culture dish face up (anterior view). Remove the distal epiphysis, including the growth plate, with a sterile scalpel. Cut from the border between cartilage and cortical bone and flip the condyles off (Fig. 5.1, see Note 11).
5 PTH In Vivo and RNA Isolation
83
Fig. 5.1 Schematic diagram of longitudinal sections of rat femur showing the trabecular bone region resected for total RNA preparation and analyses. (Reprinted from Onyia, J. E., Bidwell, J., Herring, J., et al. (1995) In vivo, human parathyroid hormone fragment (hPTH 1-34) transiently stimulates immediate early response gene expression, but not proliferation, in trabecular bone cells of young rats. Bone 17, 479–484, with permission from Elsevier).
3. With a bone cutter (dip in NaOH and DEPC-treated water each for 5 seconds before use), resect a subjacent 3-mm-wide band of the metaphyseal trabecular primary spongiosa and store it in 300 µL RNAlater RNA Stabilization Reagent (Qiagen). The tissue can be stored in RNAlater Stabilization Reagent for up to 4 weeks at 2–8 °C, up to 7 days at 18 to 25 °C, or up to 1 day at 37 °C (see Note 12).
3.5
Isolation of RNA from Bone
1. Using sterile forceps (dip into DEPC-treated water before use), remove the sample from the tube into a sterile culture dish containing Kimwipes on the bottom to quickly drain the extra solution. Transfer the sample into a round bottom Falcon tube with 3 mL Tri Reagent (Sigma) inside. 2. Immerse the bottom of the tube into a beaker containing a mixture of ice/water, pulverize the sample by using a PT 10-35 Polytron homogenizer. Pulse several times for 1 minute (see Note 13). 3. Check the bottom of the tube, if it is not a fine suspension, repeat the previous step. 4. Aliquot the 3-mL suspension from the previous step into three labeled, 1.7-mL conical bottom screw-top Eppendorf tubes on ice (see Note 14). 5. Add 200 µL of chloroform per 1 mL Tri Reagent in the tube. Tightly close the tube cap. Shake vigorously for at least 15 seconds. Leave at room temperature for 2–15 minutes.
84
X. Li et al.
6. Centrifuge the resulting mixture at 12,000 g for 15 minutes at 4 °C. Centrifugation separates the mixture into three phases: a red organic phase (containing protein), an interphase (containing DNA), and a colorless upper aqueous phase (containing RNA). 7. Transfer the aqueous phase to a fresh tube and add 500 µL of isopropanol per 1 mL Tri Reagent and mix. Allow the sample to stand for 5–10 minutes at room temperature. 8. Centrifuge at 12,000 g for 10 minutes at 4 °C to precipitate the RNA pellet on the side and bottom of the tube. Remove the supernatant. 9. Wash the RNA pellet by adding 1 mL (minimum) of 75% ethanol per 1 mL Tri Reagent in the tube at step 4. Vortex the sample and then centrifuge at 7,500 g for 5 minutes at 4 °C (see Note 15). Remove the ethanol and air dry the pellet.
3.6
RNA Clean-Up
1. Add 100 µL of RNase-free water to dissolve the RNA pellet in each tube from Section 3.5. 2. Follow the RNA clean-up steps as instructed in the handbook from the RNeasy kit (Qiagen). 3. Pool the RNAs from three Eppendorf tubes of the same bone sample in a single labeled Eppendorf tube.
3.7
RNA Quality and Quantity Test
1. Use 4 or 10 µL of RNA and dilute with RNase-free water to 1 mL in a quartz cuvette. Measure the absorbance at 260 and 280 nm with a spectrophotometer. One A260 unit/mL = 40 µg/mL RNA (see Note 16). 2. The 260/280 nm ratio should be calculated to ensure the absence of protein (see Note 17). 3. Gel electrophoresis is recommended to check the integrity of the RNA samples. Agarose is prepared by melting the required amount of agarose (1%) in DEPC-treated water, cooling to approximately 60 °C (hand hot) and adding 40% formaldehyde and 5 × MOPS to give 2.2 M formaldehyde and 1 × MOPS, respectively (see Note 18). 4. RNA samples are prepared by adding up to 25 µg of RNA in a maximum of 5 µL DEPC-treated water to 15 µL RNA denaturation buffer. Add 1 µL of 10 mg/mL ethidium bromide to aid in the visualization of RNA after electrophoresis. 5. Immediately prior to loading, RNA samples are heated to 55 °C for approximately 15 minutes to denature any secondary structure and cooled on ice. Add 2 µL of sterile 10 × RNA loading buffer.
5 PTH In Vivo and RNA Isolation
85
Fig. 5.2 Gel electrophoresis of RNAs extracted from rat femora. Each lane was loaded with a different amount of RNA.
6. Samples are loaded onto the gel and electrophoresis conducted at no more than 5 V/cm until the bromophenol blue dye front has migrated approximately threefourths the length of the gel. 7. Visualization of RNA is achieved by irradiation with short wave (254 nm) UV light. Two bands corresponding to the 28S and 18S ribosomal RNAs should be clearly visible (Fig. 5.2).
4
Notes
1. The chloroform used for phase separation should not contain isoamyl alcohol or other additives. 2. ALZET pumps operate because of an osmotic pressure difference between a compartment within the pump, called the salt sleeve, and the tissue environment in which the pump is implanted. The high osmolality of the salt sleeve causes water to flux into the pump through a semipermeable membrane that forms the outer surface of the pump. As the water enters the salt sleeve, it compresses the flexible reservoir, displacing the test solution from the pump at a controlled, predetermined rate. Because the compressed reservoir cannot be refilled, the pumps are designed for single use only. 3. The rate of delivery by an ALZET pump is controlled by the water permeability of the pump’s outer membrane. Thus, the delivery profile of the pump is independent of the drug formulation dispensed. The molecular weight of a compound, or its physical and chemical properties, has no bearing on its rate of delivery by ALZET pumps. 4. The volume delivery rate of ALZET pumps is fixed by the manufacturer. ALZET osmotic pumps are available with a variety of delivery rates between 0.25 and 10 µL/hour and delivery durations between 1 day and 4 weeks. Because the volume delivery rate of the pump is fixed, different dosing rates are achieved by varying the concentration of agent in the solution used to fill the pump reservoir. 5. If a pump becomes contaminated, its surface may be wiped with an aqueous solution of 70% isopropanol immediately before use.
86
X. Li et al.
6. The syringe should be free of air bubbles and allow extra syringe volume for spillage. 7. Make sure the tip of the filling tube meets the bottom of the pump reservoir to avoid air bubbles. 8. A small amount of back pressure is normal due to the tight seal at the filling portal. Avoid rapid filling, which will introduce air bubbles into the reservoir. The insertion of the moderator will displace some of the solution from the filled pump. This overflow should be wiped off. 9. If the solution volume filled in the pump is < 90% of the reservoir volume, the pump should be evacuated and refilled. 10. RNA extraction should be carried out in an RNase-free environment whenever possible. Wear gloves throughout the process and frequently change them while handling samples. It is best to use RNase-free disposable plastic ware and filter-tips. 11. Although tibiae can also be used for RNA isolation, femora are preferred for greater RNA yields due to the tissue size and higher proportion of trabecular to cortical bone. Typical yields of total RNA from a rat femur sample is 40 µg or more. If RNA is extracted from the same region in mice, 1 mL TRI reagent is enough for each sample which yields about 10 µg of RNA. 12. For archived storage at −20 °C, first incubate the sample overnight in reagent (RNAlater) at 2–8 °C and then transfer to −20 °C for storage. For archived storage at −80 °C, incubate the sample overnight in reagent at 2–8 °C, and then remove the tissue from the RNAlater RNA stabilization Reagent and transfer it to −80 °C for storage. 13. The probe of the homogenizer needs to be cleaned before use. First wash with NaOH (1 N) twice. Then wash twice with DEPC-treated water. Finally, wash with 3 mL of Tri Reagent. 14. At this step, sample suspensions can be stored at −20 °C for weeks. Before proceeding to RNA isolation, allow samples to stand for 5 minutes at room temperature to ensure complete dissociation of nucleoprotein complexes. 15. This step can be skipped if immediately performing the first step in Section 3.6. Samples can be stored in ethanol at 4 °C at least 1 week and up to 1 year at −20 °C. 16. Alternatively, the absorbance can be measured with as little as 1 µL of RNA with a NanoDrop spectrophotometer (ND-1000). 17. Final preparation of RNA should have a 260/280 ratio within the range of 1.8– 2.0 to be used in realtime RT-PCR and microarray analysis. 18. Formaldehyde is extremely toxic. Gloves should be worn when preparing and handling solutions containing formaldehyde. Electrophoresis utilizing formaldehyde-containing buffers should be performed in a fume hood. Agarose/ formaldehyde gels are inherently weaker than the equivalent percentage agarose gels and rupture of the well bottom occurs easily. Care should be taken in removing the combs prior to loading of samples.
5 PTH In Vivo and RNA Isolation
87
References 1. Bauer, W., Aub, J., Albright, F. (1929) Studies in calcium phosphorus metabolism: study of bone trabeculae as ready available reserve supply of calcium. J Exp Med 49, 145–162. 2. Burr, D. B., Hirano, T., Turner, C. H., et al. (2000) Intermittently administered human parathyroid hormone (1-34) treatment increases intracortical bone turnover and porosity without reducing bone strength in the humerus of ovariectomized cynomolgus monkeys. J Bone Miner Res 16, 157–165. 3. Neer, R. M., Arnaud, C. D., Zanchetta, J. R., et al. (2001) Effect of parathyroid hormone (1-34) on fractures and bone mineral density in postmenopausal women with osteoporosis. NEJM 344, 1434–1441. 4. Dempster, D. W., Cosman, F., Kurland, E. S., et al. (2001) Effects of daily treatment with parathyroid hormone on bone microarchitecture and turnover in patients with osteoporosis: a paired biopsy study. J Bone Miner Res 16, 1846–1853. 5. Brommage, R., Hotchkiss, C. E., Lees, C. J., et al. (1999) Daily treatment with human recombinant parathyroid hormone-(1-34), LY333334, for 1 year increases bone mass in ovariectomized monkeys. J Clin Endocrinol Metab 84, 3757–3763. 6. Onyia, J. E., Bidwell, J., Herring, J., et al. (1995) In vivo, human parathyroid hormone fragment (hPTH 1-34) transiently stimulates immediate early response gene expression, but not proliferation, in trabecular bone cells of young rats. Bone 17, 479–484. 7. Alzet Guide to Pump Use (Alzet). 8. Protocol for Sigma’s TRI Reagent (Sigma). 9. RNase Mini Handbook (Qiagen). 10. Li, X. (2006) Parathyroid hormone’s dual effects (anabolic versus catabolic) on osteoblast gene expression. Doctoral dissertation. University of Medicine and Dentistry of New Jersey, Newark, NJ.
Chapter 6
Assessment of Bone Formation Capacity Using In vivo Transplantation Assays: Procedure and Tissue Analysis Basem M. Abdallah, Nicholas Ditzel, and Moustapha Kassem
Abstract In vivo assessment of bone formation (osteogenesis) potential by isolated cells is an important method for analysis of cells and factors controlling bone formation. Currently, cell implantation mixed with hydroxyapatite/tricalcium phosphate in an open system (subcutaneous implantation) in immunodeficient mice is the standard method for in vivo assessment of bone formation capacity of a particular cell type. The method is easy to perform and provides reproducible results. Assessment of the donor origin of tissue formation is possible, especially in the case of human-to-mouse transplantation, by employing human specific antibodies or in situ hybridization using human specific Alu-repeat probes. Recently, several methods have been developed to quantitate the newly formed bone using histomorphometric methods or using non-invasive imaging methods. This chapter describes the use of in vivo transplantation methods in testing bone formation potential of human mesenchymal stem cells. Keywords Mesenchymal stem cell, in vivo bone, heterotropic bone, osteoblast, hydroxyapatite.
1
Introduction
Assessment of osteogenic potential of cells has traditionally been examined in vitro using a combination of molecular biology and cell biology based assays including gene expression of osteoblast specific genes and proteins as well as in vitro mineralization assays (1). However, the expression of these markers does not directly correlate with in vivo bone formation capacity (2). Also, the in vitro environment does not usually reflect the complex in vivo conditions, and thus may be a limiting factor when studying the cellular and molecular control of in vivo bone formation. For these reasons, assessment of bone formation capacity of cells in vivo is an important assay to be employed by researchers interested in understanding cellular and molecular control of bone formation. From: Methods in Molecular Biology, Vol. 455: Osteoporosis: Methods and Protocols Edited by Jennifer J. Westendorf © Humana Press, Totowa, NJ
89
90
B.M. Abdallah et al.
Several methods have been employed to assess the in vivo bone formation capacity of cells. Traditionally, implanting cells in diffusion chambers has been employed by several investigators (3). Diffusion chambers are small plastic tubes bound with semi-permeable membranes that can be loaded with cells and then implanted intraperitoneally in immunodeficient mice. Bone and cartilage are usually formed inside the tubes after 20 days. The advantage of diffusion chambers is the ease of identification of donor cells because they are physically separated from recipient cells. However, diffusion chambers are small and difficult to be loaded with cells and the assay has a low success rate. They also cause discomfort for the animals. Alternatively, transplantation of cells mixed with a carrier, hydroxyapatite/tricalcium phosphate (HA/TCP), has been very successful in supporting osteogenesis (4). The advantage of this method is the ease of the procedure, its reproducibility and the possibility of quantitating the bone formed. This assay has been used to test for the stem cell potential of mesenchymal stem cells and quantitate the effects of factors on bone formation (5).
2 2.1
Materials Human Mesenchymal Stem Cell Culture
1. Heparin (100 U/mL). 2. Lymphoprep. 3. Standard growth medium: Modified essential medium (MEM), 10% (v/v) fetal calf serum (FCS), 1% penicillin/streptomycin. Store at 4 °C, for 1 month.
2.2
Loading Cells with HA/TCP Vehicles
1. Hydroxyapatite/tricalcium phosphate (HA/TCP) ceramic powder (Zimmer Scandinavia, Hørsholm, Denmark).
2.3
Transplantation Procedures
1. Non-obese diabetic/severe combined immunodeficient NOD/SCID (NOD/LtSzPrkdcscid) mouse recipients (8-week-old females, ca. 20 g) (6) (see Note 1). Mouse breeding pairs are originally obtained from Jackson Laboratories (Bar Harbor, ME). 2. Ketaminol (Ketamine) Store at 4 °C. 3. Rompun (Xylazine). Store at 4 °C.
6 In Vivo Bone Formation Assay
2.4
Fixation and Embedding of Implants
2.4.1
Plastic and Paraffin Embedding
91
1. Methyl methacrylate and glycol methacrylate Technovit 9100 kit (Heraeus, Kulzer GmbH, Wehrheim, Germany). 2. Heavy duty microtome Leica SM2500. 3. Neutral-buffered formaldehyde. 4. Formic acid solution: 0.4 M formic acid, 0.5 M sodium formate.
2.5
Histological Analysis of the Tissues Within the Implants
2.5.1
Hematoxylin/Eosin (H&E) Staining
1. Xylene. 2. Ethanol: 95%, 93%, 77%. 3. Hematoxylin/eosin stain: To prepare 1 L of hematoxylin staining solution, dissolve 1 g of hematoxylin (Merck 15938) in 1 L of distilled water. Add 0.2 g of sodium iodate (Merck 6525) and 50 g of potassium aluminium sulfate (Merck 1047). When this is dissolved, add 50 g of chloral hydrate (Ph Eur) and 1 g of citric acid (Merck 244). To prepare 1 L of eosin staining solution, make an eosin stock solution of 2.1% (w/v) eosin in distilled water. Mix 250 mL of the 2.1% eosin stock solution with 750 mL distilled water and add 150 µL acetic acid (100%). 2.5.2 1. 2. 3. 4.
Goldner’s Trichrome Stain
2-Methoxyethyl-acetate. Ethanol: 99%, 96%. 1% Acetic acid. Weigert Hematoxylin: Prepared fresh by mixing two solutions, A + B 1:3 (v/v). Solution A: Dissolve hematoxylin 1% (w/v) in 99% ethanol. Solution B: Dissolve ferrichloride • 6H2O 2.5% (w/v) and ferrosulfate • 7H2O 4.5% (w/v) in 98 mL distilled water and then mix with 2 mL concentrated hydrochloric acid. 5. Ponceau/Fuchsin Solution: Dissolve 3 g Ponceau 2 R (Merck) and 1.5 g Fuchsin acid (Merck) in 300 mL distilled water and mix with 0.6 mL 100% acetic acid, pH 3.1. Store in dark place at room temperature for 3 months. 6. Orange G Staining Solution: Dissolve 6 g Orange G (Merck 15925) and 12 g Phosphorwolframic acid (Merck 583) in 300 mL distilled water, pH 2.1. Store in dark place at room temperature for 3 months. 7. Fast green FCF solution: Mix 0.6 g Fast Green FCF (Sigma F-7252) and 300 mL distilled water. Then mix with 0.6 mL 100% acetic acid. Store in a dark place at room temperature for 3 months.
92
B.M. Abdallah et al.
2.5.3
1. 2. 3. 4.
5. 6. 7. 8. 9.
Identifying the Origin of Heterotropic Bone Formation (Immuno-Histochemical Staining)
H2O2 (hydrogen peroxide) 1.5% in TBS buffer. TBS: 100 mL of 0.05 M Tris-HCl, pH 7.4, and 900 mL of 0.15 M NaCl. Normal goat serum. Primary Antibodies: Polyclonal rabbit anti-human Collagen type I antibody Coll I (LF-67), polyclonal rabbit anti-human osteonectin (ON) (LF-37) antibody (7, 8), monoclonal mouse anti-human osteopontin (OPN) (NCL-O-Pontin) (Novocastra, UK, Newcastle upon Tyne) and mouse IgG (negative control). Secondary antibodies: biotinylated goat anti-mouse (Dako, Glostrup, Denmark) and biotinylated goat anti-rabbit (Dako). Peroxidase-conjugated streptavidin. 3-Amino-9-ethyl carbazole. Mayers hematoxylin. Glycergel.
3 3.1
Methods Human Bone Marrow Mesenchymal Stem Cell Cultures
1. Bone marrow is obtained by aspiration from the iliac crest of healthy volunteers (five females 24–30 years old) after infiltration of the area with local anesthetic (Lidocaine, 10 mg/mL) as described previously (9). 2. Aspirate 10 mL of bone marrow and mix with 10 mL of heparin (100 U/mL). Isolate the low-density mononuclear cells by centrifugation over a Lymphoprep density gradient (density = 1.077 ± 0.001 g/cm3) at 1,050 g for 25 minutes at room temperature. Wash the cells once in standard growth medium. 3. Count the viable cells after methylene blue staining in a Burker-Turk counting chamber and seed round 1 × 107 hMSC per T-75 flask. 4. Incubate the cultures in a humidified atmosphere of 5% CO2 at 37 °C and feed the cells by completely replacing the medium once a week until 80–90% confluence.
3.2
Loading Cells with HA/TCP Vehicles
1. Wet the hydroxyapatite/tricalcium phosphate (HA/TCP) powder (40 mg) with 100 µL of standard growth medium to avoid static electricity and transfer it into 1 mL syringe (Codan Medical, Denmark) as shown in Fig. 6.1A. 2. Trypsinize the cultured hMSC cells and add around 5 × 105 cells (in 200 µL medium) carefully on the top of HA/TCP granules in 1 ml syringe. Incubate
6 In Vivo Bone Formation Assay
93
overnight at 37 °C in 5% CO2. Use the HA/TCP granules without any cells as a control (see Note 2). 3. Prepare four samples of hMSC loaded onto HA/TCP granules (as mentioned in step 2) per each donor to produce four implants. 4. For dose-response experiments, the number of hMSC loaded onto HA/TCP granules are as follows: 30 × 105, 15 ×105, 5 × 105, 0.75 × 105, and 0.1 × 105 cells. 5. To examine the effect of different growth factors on the BFC of hMSC, cells are induced with different growth factors in culture for 4 days, then trypsinized and loaded onto the HA/TCP granules for transplantation (5).
3.3
Transplantation Procedures
Transplant the HA/TCP loaded without (vehicle alone as control) or with hMSC into immuno-deficient NOD/SCID (NOD/LtSz-Prkdcscid) mouse recipients (8-week-old females, ca. 20 g) as described previously (11) and as shown in Fig. 6.1B–F). 1. Perform the operations under anesthesia, which is achieved by intra-peritoneal injection of ketaminol (100 mg/kg ketamine) and Rompun (10 mg/kg xylazine). 2. Shave the mouse and disinfect the skin with 70% ethanol prior to operation.
Fig. 6.1 Step-by-step transplantation procedure of hMSC cells with HA/TCP in NOD/LtSzPrkdcscid mice. a. Prepare the cells for implantation by loading them onto HA/TCP granules in 1-mL syringes as shown and described in the Methods section. b. Shave the mouse and clean the area with ethanol. Make an incision 10- to 12-mm long through the skin only. c. Insert a sterile 1-mL syringe into the incision to form a subcutaneous pocket approximately 3 cm deep. Prepare the first implant by removing the medium above the HA/TCP (leave about 1 mm of medium). Insert the syringe with the implant into the pre-formed pocket. Inject the implant by pushing the piston. d. Withdraw the syringe slowly. e. Suture the incision and perform the procedure on the other side of the mouse. f. Arrows indicate the positions of the implants under the skin at the end of the transplantation procedure.
94
B.M. Abdallah et al.
3. Make two mid-longitudinal skin incisions of about 1 cm in length on the dorsal surface of each mouse, and make subcutaneous pouches by blunt dissection. 4. Place one HA/TCP implant with or without cells into each pouch (two implants per mouse). Close the incisions with surgical sewing and then keep the operated mouse in separate cage until the end of the experiment. 5. Repeat steps 2–5 on another mouse. 6. Keep the implants in mice for 8 weeks.
3.4
Fixation and Embedding of Implants
3.4.1
Plastic Embedding (for Non-Decalcified Sections)
1. Sacrifice the mice 8 weeks post-transplantation by cervical dislocation. Remove the implants and fix them in 70% ethanol. 2. Use the Technovit 9100 Methyl Methacrylate kit according to the manufacturer’s instructions to obtain plastic blocks with mineralized implants ready for sectioning (see Note 3). 3. Use a heavy duty microtome (e.g., Leica SM2500, Leica Microsystems) to cut three serial sections (7.5-µm thick) close to the block surface and repeat it every 100 µm depth through the implant to obtain a total of nine sections per implant at different levels.
3.4.2
Paraffin Embedding (for Decalcified Sections)
1. Sacrifice the mice at 8 weeks post-transplantation by cervical dislocation. Remove the implants and fix them in 4% neutral buffered formaldehyde for 24 hours. 2. Decalcify the implants by immersing them in formic acid solution for 3 days. 3. Use the standard paraffin embedding procedure previously described by other investigators (12–14) to prepare decalcified sections for hematoxylin/eosin (H&E) or immunohistochemical staining.
3.5
Histological Analysis of the Tissues Within the Implants
3.5.1
Hematoxylin/Eosin (H&E) Staining
1. Deparaffinize the sections by submerging in the following series of solutions: xylene for 10 minutes, 99% ethanol for 5 minutes, 93% ethanol for 2 minutes, and 77% ethanol for 2 minutes. 2. Rinse in tap water. 3. Stain with the standard protocol for H&E (Fig. 6.2A).
6 In Vivo Bone Formation Assay
95
Fig. 6.2 Histological analysis of heterotropic bone formation. a. Histological analysis of heterotropic bone in either decalcified paraffin-embedded implants (a, b, and c) or non-decalcified plastic-embedded implants (d, e, and f) harvested after 8 weeks of subcutaneous transplantation of hMSC with HA/TCP in immunodeficient mice. The paraffin sections were stained with H&E, and plastic sections were stained with Goldener’s Trichrome. hMSC cells formed Osteoid (O), trabecular bone (B), within the HA/TCP granules (H). Note that only the Goldener’s Trichrome staining could distinguish between osteoid and mineralized bone. Sections show the advantage of using the in vivo bone assay to compare the capacity of different hMSC clones to differentiate into bone in vivo. Sections were scored as, non-bone forming cells (a, d), cells with moderate bone formation (b, e) and cells with high bone formation (c, f) (magnification a–f ×10). b. Identifying the origin of heterotropic bone formation in the implants by immuno-staining using specific antihuman Collagen type I, osteonectin and osteopontin. Arrows show positive stained bone matrix and osteocytes (magnification a–c × 20).
3.5.2
Goldner’s Trichrome Stain
1. Stain the tissue sections for 30 minutes in 2-methoxyethyl-acetate followed by hydration in 99% ethanol, 96% ethanol, and water for 5 minutes each (see Note 4). 2. Perform the counter-staining with Weigert-Hematoxylin consisting of solution A and solution B 1:3 (v/v) for 5 minutes. 3. Wash the tissue sections for 10 minutes in water and stain with Ponceau/fuchsin solution for 15 minutes. 4. Rinse the sections in 1% acetic acid (2 × 1 minute) and stain with orange G solution for 7 minutes. Sections are rinsed again in 1% acetic acid (2 × 1 minute) followed by staining in Fast green for 20 minutes. 5. Perform a final wash in 1% acetic acid (2 × 1 minute). Dehydrate the tissue sections in 96% ethanol followed by 99% ethanol and mount with glass coverslips (Fig. 6.1A, d–e).
96
3.5.3
B.M. Abdallah et al.
Identifying the Origin of Heterotropic Bone Formation (Immuno-Histochemical Staining)
To identify the donor origin of the osteogenic tissue, immuno-histochemical staining with human-specific antibodies against osteonectin (ON), osteopontin (OPN) and collagen type I (Col I) was employed as described previously (11). Another method to demonstrate the human origin of the formed bone is in situ hybridization using human specific alu-repeats probes (15). 1. Immerse the tissue sections (from paraffin embedded implants) in 1% acetic acid for 10 minutes and treat with H2O2 in TBS (20 minutes) to abolish endogenous peroxidase. 2. After washing in TBS for 5 minutes, incubate the sections in 10% normal goat serum (NGS). 3. Incubate the tissue sections with primary antibodies overnight at 4 °C (see Note 5), wash with TNT buffer and incubate with secondary antibodies for 1 hour and wash again with TNT buffer. 4. Immuno-reactivity is detected by incubation with peroxidase conjugated streptavidin for 1 hour and developed for 30 minutes with 3-amino-9-ethyl carbazole (AEC). 5. Counter-stain the sections with Mayers Hematoxylin for 30 seconds, and mount in glycergel (Fig. 6.2B).
3.6
Quantifying the Amount of Bone Tissue Formed in the Implants
The following three quantitative methods can be independently performed on an identical set of sections or the whole implant (only for micro-CT method).
3.6.1
Visual Scoring Method (Based on Pixels Counting Method)
This simple and fast analysis method can be used to semi-quantitatively estimate the amount of bone formation using either the plastic or the paraffin-embedded sections as described previously (16, 17). 1. Scan serial sections of paraffin or plastic embedded implant (every 10th section) using Leica microscope linked through a camera Leica DFC 480 to a Leica Dellcompatible computer using the Image-Pro Plus 5.1 image analysis program. 2. Determine the total bone area for each section by selecting a color range that highlighted bone (as bone pixels) based on eosin-stained tissue (in case of paraffin-embedded sections) or Goldner’s Trichrome-stained tissue (in case of plastic-embedded sections) and import into Adobe Photoshop (Adobe Systems, Mountain View, CA).
6 In Vivo Bone Formation Assay
97
3. Estimate the bone area by dividing the total number of bone pixels by the total number of ceramic pixels based on color. 4. Values of serial sections through the implant are summed and then divided by the number of sections to obtain an average section score.
3.6.2
Bone Histomorphometric Analysis
This is a more accurate method and is based on the Cavalieri principle (see Note 6). 1. Serial sections (every 10th section) from each implant are cut and stained with Goldner’s trichrome stain as described. The amount of bone formed is estimated by point counting according to Cavalieri’s principle using CAST software (Olympus, Denmark, Ballerup) as described by our group (11). 2. To quantify the amount of bone tissue formed, two sets, each with the first section placed randomly in the implant, of five successive histological sections stained with Goldner’s Trichrome stain with a fixed intra-distance (75 µm) are prepared and analyzed from each implant according to Cavalierie’s Principle, which allows an unbiased volume estimate of an irregular three-dimensional (3D) structure from uniform random sections (18). 3. The amount of bone formed is determined in each section as a percentage of bone volume per total volume (BV/TV) by point counting using a grid as described previously (19). All counts are performed without knowledge of how many cells are implanted or the nature of donor cells. Intra-observer and interobserver variation for bone quantification are 9% and 17%, respectively.
3.6.3
Micro Computer Tomography
This is a newly developed procedure by our group and it may provide an easy and quick method for quantitation of in vivo formed bone. However, the major limitation is that micro-CT scanning technology is unable to reliably distinguish between the implanted hydroxyapatite/tricalcium phosphate granulates and newly formed mineralized bone. Nevertheless, the micro-CT analysis can provide data on parameters relating bone volume to total implant volume (BV/TV). It is essential that all the implants compared in a given data set were made with the same type of hydroxyapatite granulate because the interpretation of the data assumes equal mean hydroxyapatite particle size and equal distance between particles. If these criteria are met, the contribution of bone-forming cells to increases in the volume of bone in the implant can be measured as an increase in the BV/TV, when compared with control implants made without cells. 1. Carefully lift whole implants out of the fixation solution and place them in a sample-tube supplied with the micro-CT scanner (choose the tube with the smallest possible diameter) (see Note 7). Place the implant in an upright position to minimize the number of image slices required to enclose the entire implant.
98
B.M. Abdallah et al.
Fig. 6.3 Evaluation of heterotropic bone formation by 3D micro-CT analysis. The micro-CT analysis was performed on the whole implants obtained from the transplantation of HA/TCP granules alone (control) or with hMSC clones (with different capacity for in vivo bone formation) using the viva CT 40-system, which gives a cubic voxel size of 12*12*12 µm3 in the 3D reconstructions.
2. Perform the scanning procedure at high resolution using for example, the viva CT 40-system (Scanco Medical AG, Zürich, Switzerland) that gives a cubic voxel size of 12*12*12 µm3 in the 3D reconstructions. 3. Evaluate the scans by drawing a “region of interest” around the external surface of each implant. Perform 3D evaluation using the µCT Evaluation Program V5.0 (Scanco Medical AG, Zürich, Switzerland). As an example, the 3D micro-CT reconstruction images and parameters obtained from the scanning of two different implants with low and high bone formation are compared with an implant without cells and presented in Fig. 6.3.
4
Notes
1. Handle all the mice under sterile conditions and maintain in a separate NODSCID facility. All animal experiments presented in this study were carried out with permission from the Danish Experimental Animal Inspectorate. 2. As a model of hMSC, our immortalized hMSC named hMSC-TERT cell line (1, 10) were used through the whole in vivo procedure. hMSC-TERT cells are grown in standard growth medium.
6 In Vivo Bone Formation Assay
99
3. The Technovit 9100 kit is an embedding medium system based on methyl methacrylate (MMA). It is designed for the embedding of mineralized tissues with extensive possibilities of staining for light microscopy. 4. Negative controls for histochemical staining protocols are tissue sections from a mouse femur and a human bone biopsy processed in a similar way. 5. As a negative control, mouse IgG (Dako) is used as a primary antibody. 6. Cavalierie’s Principle sums up the areas of object cross-sections and multiplies the section thickness according to the formula: Vreference = T*a*ΣP(treference), where T is the section thickness, a is the area associated with the grid point, and ΣP(treference) is the number of points hitting the profile area. 7. Micro-CT analysis should be done on the whole fixed implant before decalcification or any histological evaluation.
References 1. Abdallah, B. M., Haack-Sorensen, M., Burns, J. S., et al. (2005) Maintenance of differentiation potential of human bone marrow mesenchymal stem cells immortalized by human telomerase reverse transcriptase gene in despite of extensive proliferation. Biochem Biophys Res Commun 326, 527–538. 2. Kuznetsov, S. A., Krebsbach, P. H., Satomura, K., et al. (1997) Single-colony derived strains of human marrow stromal fibroblasts form bone after transplantation in vivo. J Bone Miner Res 12, 1335–1347. 3. Bab, I., Ashton, B. A., Gazit, D., et al. (1986) Kinetics and differentiation of marrow stromal cells in diffusion chambers in vivo. J Cell Sci 84, 139–151. 4. Krebsbach, P. H., Kuznetsov, S. A., Satomura, K., et al. (1997) Bone formation in vivo: comparison of osteogenesis by transplanted mouse and human marrow stromal fibroblasts. Transplantation 63, 1059–1069. 5. Kratchmarova, I., Blagoev, B., Haack-Sorensen, M., et al. (2005) Mechanism of divergent growth factor effects in mesenchymal stem cell differentiation. Science 308, 1472–1477. 6. Shultz, L. D., Schweitzer, P. A., Christianson, S. W., et al. (1995) Multiple defects in innate and adaptive immunologic function in NOD/LtSz-scid mice. J Immunol 154, 180–191. 7. Fisher, L. W., Robey, P. G., Tuross, N., et al. (1987) The Mr 24,000 phosphoprotein from developing bone is the NH2-terminal propeptide of the alpha 1 chain of type I collagen. J Biol Chem 262, 13457–13463. 8. Fisher, L. W., Hawkins, G. R., Tuross, N., et al. (1987) Purification and partial characterization of small proteoglycans I and II, bone sialoproteins I and II, and osteonectin from the mineral compartment of developing human bone. J Biol Chem 262, 9702–9708. 9. Kassem, M., Mosekilde, L., Eriksen, E. F. (1993) 1,25-dihydroxyvitamin D3 potentiates fluoride-stimulated collagen type I production in cultures of human bone marrow stromal osteoblast-like cells. J Bone Miner Res 8, 1453–1458. 10. Simonsen, J. L., Rosada, C., Serakinci, N., et al. (2002) Telomerase expression extends the proliferative life-span and maintains the osteogenic potential of human bone marrow stromal cells. Nat Biotechnol 20, 592–596. 11. Stenderup, K., Rosada, C., Justesen, J., et al. (2004) Aged human bone marrow stromal cells maintaining bone forming capacity in vivo evaluated using an improved method of visualization. Biogerontology 5, 107–118. 12. Bruder, S. P., Kurth, A. A., Shea, M., et al. (1998) Bone regeneration by implantation of purified, culture-expanded human mesenchymal stem cells. J Orthop Res 16, 155–162.
100
B.M. Abdallah et al.
13. Kuznetsov, S. A., Mankani, M. H., Robey, P. G. (2000) Effect of serum on human bone marrow stromal cells: ex vivo expansion and in vivo bone formation. Transplantation 70, 1780–1787. 14. Mankani, M. H., Krebsbach, P. H., Satomura, K., et al. (2001) Pedicled bone flap formation using transplanted bone marrow stromal cells. Arch Surg 136, 263–270. 15. Hatano, H., Tokunaga, K., Ogose, A., et al. (1998) Origin of bone-forming cells in human osteosarcomas transplanted into nude mice—which cells produce bone, human or mouse? J Pathol 185, 204–211. 16. Dennis, J. E., Konstantakos, E. K., Arm, D., et al. (1998) In vivo osteogenesis assay: a rapid method for quantitative analysis. Biomaterials 19, 1323–1328. 17. Kaigler, D., Wang, Z., Horger, K., et al. (2006) VEGF scaffolds enhance angiogenesis and bone regeneration in irradiated osseous defects. J Bone Miner Res 21, 735–744. 18. Gundersen, H. J., Bendtsen, T. F., Korbo, L., et al. (1988) Some new, simple and efficient stereological methods and their use in pathological research and diagnosis. APMIS 96, 379–394. 19. Kerndrup, G., Pallesen, G., Melsen, F., et al. (1980) Histomorphometrical determination of bone marrow cellularity in iliac crest biopsies. Scand J Haematol 24, 110–114.
Chapter 7
Ossicle and Vossicle Implant Model Systems Glenda J. Pettway and Laurie K. McCauley
Abstract Bone regeneration and repair is a goal of many skeletal therapies and numerous agents positively or negatively impact these processes. New therapeutic agents and effective model systems are continually sought to identify agents and characterize their mechanisms of action are in constant demand. In addition, investigations of tumor cell–bone interaction in the skeletal metastatic microenvironment require well-defined and readily orchestrated models. This chapter describes a novel ectopic ossicle model and a vossicle modification that can be used to provide focused and rapid feedback of bone growth and bone–cellular interactions. The ossicle model is a bone marrow stromal cell (BMSC)–based model and the vossicle model is a neonatal vertebral bone transplant model. These models offer opportunities to mix and compare mesenchymal (donor derived) and hematopoietic elements (host derived). Multiple implants can be placed in one mouse to facilitate various outcome analyses, such as histomorphometry, micro-CT, gene expression studies, and cell tracking using markers such as luciferase, in response to pharmacological or genetic manipulation. Implants can also be combined with other cell types, such as cancer cells to evaluate the bone–tumor microenvironment. Keywords Bone marrow stromal cell, mesenchymal stem cell, parathyroid hormone, bone implant, bone growth, bone regeneration, ectopic ossicle.
1
Introduction
There are a wide variety of agents that positively or negatively impact bone growth and regeneration and there is a great deal of interest to better characterize their actions in bone and identify new therapeutic agents for tissue repair and engineering. In vitro model systems using osteoblastic cells have been beneficial to identify transcriptional mediators and agents with direct actions on cells of the osteoblast lineage, but many agents have disparate actions in vitro vs. in vivo. Often the From: Methods in Molecular Biology, Vol. 455: Osteoporosis: Methods and Protocols Edited by Jennifer J. Westendorf © Humana Press, Totowa, NJ
101
102
G.J. Pettway and L.K. McCauley
in vivo model systems, such as the mouse ovariectomy model, take a long time to orchestrate and evaluate. Model systems in which agents could be tested in a relatively short time period and with discrete and readily measurable endpoints would be beneficial. Our laboratory has focused on investigations of the anabolic actions of parathyroid hormone (PTH) in bone and the impact of tumor-derived PTH-related protein (PTHrP) in skeletal metastasis. PTH is currently the only FDA-approved anabolic agent for osteoporosis treatment in the United States. PTH is also being investigated for potential use in enhancing fracture healing (1, 2), the promotion of implant integration (3), and in tissue engineering applications. The use of PTH in healing osseous defects may be addressed through the use of bone regeneration models, such as the cell-based ectopic ossicle model system presented here and well characterized by other groups (4, 5). This bone-engineered model involves the implantation of bone marrow stromal cells (BMSCs) in immunocompromised mice (Fig. 7.1) to generate ossicles containing cortical and trabecular bone and a hematopoietic marrow. The mesenchymal components of this model are from the donor and the hematopoietic elements are from the host. This model system offers the advantage that multiple distinct bone ossicles can be generated in a single host and used for various endpoint analyses, such as histomorphometry, immunohistochemistry, micro-computed tomography (micro-CT), bioluminescent imaging (BLI), and gene expression studies. Another benefit of this model is that implants can be placed and/or removed from the same mouse in a temporal fashion. This model facilitates the ability to compare the effects of pharmaceutical agents on developing
Fig. 7.1 Schematic of ectopic ossicle model. Tissue engineered bone is generated from transplanted bone marrow stromal cells (BMSCs). This model system can be used to characterize the effects of pharmacologic agents, such as parathyroid hormone (PTH), on bone. (Reproduced from Schneider, A., Taboas, J. M., McCauley, L. K., et al. (2003) Skeletal homeostasis in tissue-engineered bone. J Orthop Res 21:859–864 with permission of John Wiley & Sons, Inc.)
7 Osseous Implants
103
bone versus endogenous host bone. Furthermore, this model can be used to compare mesenchymal elements from mice with certain genetic mutations with hematopoietic elements from normal mice or visa versa. The ectopic ossicle model system has been effectively used to study skeletal responses to PTH (6, 7) and may also be used to evaluate the effects of other biological agents on both modeling and remodeling bone. A modification of this model is the vossicle model (8), in which instead of implanting bone marrow stromal cells, intact vertebral bodies from neonatal mice are utilized. This modification has the advantage of being more rapid because the step of expanding cells ex vivo is eliminated. A limitation is that there is carryover of hematopoietic elements from the donor with the vertebral implants. So although the vossicles are composed mostly of hematopoietic elements from the host, there is contamination from the donor. An additional feature of this model is that cells such as tumor cells can be co-implanted with the vertebral bodies to simulate a tumor–bone microenvironment.
2 2.1
Materials Bone Marrow Stromal Cell Isolation and Culture
1. Four- to eight-week-old donor mice for bone marrow stromal cell (BMSC) isolation. 2. Surgical Instruments: Forceps, scissors, scalpel, #15 blades, 22- and 25-gauge needles, 1-mL syringes. 3. Growth Medium: Alpha-modified minimum essential medium (α-MEM) supplemented with 20% fetal bovine serum (FBS, Hyclone, Logan, UT), 1% penicillin/streptomycin sulfate/glutamine (PSG), and 10 nM dexamethasone. 4. Hanks’ balanced salt solution (HBSS). 5. Solution of trypsin (0.25%) and ethylenediamine tetraacetic acid (EDTA). 6. 70% Ethanol.
2.2
Surgical Implantation of Bone Marrow Stromal Cells
1. Gelatin sponges (Gelfoam®, Sullivan-Schein, a Henry Schein Company, Melville, NY). 2. No. 3 Whatman filter paper. 3. Surgical Instruments: forceps, scissors, surgical wound clips (Autoclip, 9 mm, Becton Dickinson, Sparks, MD). 4. Anesthesia: 1.5% isoflurane/air anesthesia or 90 mg/kg ketamine and 5 mg/kg xylazine.
104
G.J. Pettway and L.K. McCauley
5. Four- to six-week-old male nude mice (NIH III Nude, Charles River Laboratories, Wilmington, MA) (see Note 1).
2.3 1. 2. 3. 4. 5.
Vertebral Disk (Vossicle) Isolation
Neonatal mice (d4 postpartum). 70% EtOH. Surgical instruments: forceps, scissors, scalpel, #15 blades. 60-mm Petri dishes. Phosphate buffered saline (PBS).
2.4
Surgical Implantation of Vossicles
1. Surgical Instruments: forceps, scissors, surgical wound clips (Autoclip, 9 mm, Becton Dickinson, Sparks, MD). 2. Hamilton syringe (if co-implantation with other cells is desired). 3. 70% EtOH. 4. Anesthesia: 1.5% isofluorane/air anesthesia or 90 mg/kg ketamine and 5 mg/kg xylazine. 5. Four- to six-week-old male athymic nude mice (Athymic Nu/Nu, Harlan Laboratories, Madison, WI).
3 3.1
Methods Bone Marrow Stromal Cell Isolation
1. Sacrifice mice that will be used for BMSC isolation (see Note 2). 2. Wash each mouse in 70% ethanol to remove loose fur and disinfect. 3. Remove hind limbs and forelimbs from each mouse and place in HBSS containing 2% PSG. The PSG is used to reduce contamination. 4. In a sterile environment, dissect muscle from femurs, tibiae, and humeri and cut epiphyses (see Note 3). 5. Insert a 1-mL syringe with 25-gauge needle filled with growth medium into marrow cavity of bones and flush out bone marrow. 6. After collecting bone marrow from all bones, aspirate up and down in cell suspension with a 22-gauge needle to break apart clusters of marrow. 7. Add 15 mL of growth medium to a T75 flask followed by cell suspension.
7 Osseous Implants
105
8. Incubate cells at 37 °C in an atmosphere of 100% humidity and 5% carbon dioxide until a confluent adherent cell layer is observed (see Note 4). 9. When approaching confluency, harvest BMSCs using two washes with HBSS followed by trypsinization with trypsin/EDTA at 37 °C for 10 minutes (see Note 5). 10. Perform a 1:2 split of cells and maintain culture until a confluent adherent layer is observed (see Note 6).
3.2
Surgical Implantation of Bone Marrow Stromal Cells
1. In a sterile environment, place Gelfoam in a 100-mm dish and cut desired number of Gelfoam scaffolds (approximately 5 × 5 × 5 mm cubes) using a scalpel (see Note 7). 2. Saturate Gelfoam cubes with growth medium and place in 37° incubator. 3. Harvest BMSCs for implantation using two washes with HBSS followed by trypsinization with trypsin/EDTA (see Note 8). 4. Pool together BMSCs from all donor mice and count cells. 5. Add volume of growth medium required to obtain 2–3 × 106 cells per 1 mL of cell suspension. 6. Add 1-mL of cell suspension to Eppendorf tubes and centrifuge at 210 g for 5 minutes. 7. While cells are being centrifuged, place presoaked Gelfoam in between two pieces of sterile Whatman filter paper to remove excess growth medium. 8. Aspirate medium from cell pellet. 9. Incorporate BMSCs into Gelfoam using capillary action by placing one Gelfoam scaffold/Eppendorf tube (containing 2–3 × 106 BMSCs) and gently rubbing Gelfoam to bottom of tube where it will absorb the cells. 10. Anesthetize host mice with isoflurane/air anesthesia or with intraperitoneal injection of ketamine/xylazine. 11. Clean the dorsal surface of mice using 70% ethanol and make a small (approximately 1 cm long) vertical incision at hind limbs and forelimbs level. 12. Create bilateral subcutaneous pouches by dissecting skin with blunt scissors (see Note 9). 13. Place one BMSC implant in each “quadrant” of the animal’s back (see Note 10). 14. Close wounds using surgical wound clips. 15. Monitor the recovery of mice and remove clips 2 weeks post-surgery. 16. Treat BMSC recipient mice with desired pharmacologic agent and harvest implants at the end of treatment regime (see Note 11). 17. Perform appropriate endpoint analyses, such as radiographic analysis (see Fig. 7.2A), microCT analysis (see Fig. 7.2B), and histomorphometry using paraffin (see Fig. 7.2C) embedded and/or undecalcified plastic (see Fig. 7.2D) sections. The effects of treatment regime on proliferation of implanted BMSCs
106
G.J. Pettway and L.K. McCauley
Fig. 7.2 PTH stimulation of bone formation in ectopic tissue-engineered bone. Representative (a) microradiographic images, (b) micro-computed tomography (microCT), (c) paraffin embedded H&E stained histological sections, and (d) undecalcified plastic embedded tetrachrome stained histological sections of ossicles containing trabecular (T) and cortical (C) bone. Ossicles were treated with PTH or vehicle for 3 weeks, with treatment initiated 1 week after implanting BMSCs. Note the qualitative increase in radiopacity and trabecular bone in implants treated with PTH versus implants treated with vehicle (0.9 % saline). Reconstructed microCT images were generated using a microCT 40 system (Scanco USA, Inc., Wayne, PA) at a resolution of 6 µm.
may also be evaluated using bioluminescence imaging (BLI) (see Fig. 7.3). For this application, BMSCs from mice expressing luciferase are used to generate ossicles and then tracked in vivo with BLI (see Note 12).
3.3
Vossicle Isolation
1. Sacrifice mice that will be used for vertebral isolation (see Note 13). 2. Wash back of mouse with 70% ethanol. 3. In a sterile environment, dissect skin and muscle from vertebrae. Isolate lumbar vertebrae by cutting just under where the ribs attach to the spine and cut just above the pelvis. 4. Transfer vertebrae to a 60-mm dish containing PBS. 5. Carefully dissect the vertebrae with a #15 surgical blade and cut to separate single vertebral disks (vossicles) (see Note 14). 6. Transfer isolated vossicles into fresh PBS in a new 60-mm dish until implantation (see Note 15).
7 Osseous Implants
a
Baseline (D0)
107
D28
b
Baseline (D0)
D28 High
Low Vehicle
PTH
Fig. 7.3 Bioluminescent images of ossicles derived from BMSCs isolated from mice that constitutively express luciferase. Prior to imaging mice were injected by intraperitoneal (ip) injections with 100 µL of 40 mg/mL luciferin dissolved in sterile PBS. Bioluminescent images were acquired 12 minutes after injection under isofluorane/air anesthesia on a cooled CCD IVIS system equipped with a 50-mm lens (Xenogen Corp., Alameda, CA) and coupled to a data-acquisition PC running LivingImage Software (Xenogen Corp.). Luciferase activity was similar at (a) baseline (i.e., day of BMSC implantation), and markedly increased at (b) D28 in implants exposed to PTH for 21 days, with treatment initiated 7 days after implanting BMSCs.
3.4
Surgical Implantation of Vossicles
1. In a sterile environment, anesthetize host mice with isoflurane/air anesthesia or an intraperitoneal injection of ketamine/xylazine. 2. Disinfect the dorsal surface of mice using 70% ethanol and make a small (approximately 1 cm long) vertical incision at hind limbs and forelimbs level (see Note 16). 3. Create bilateral subcutaneous pouches by dissecting skin with blunt scissors. Place one vertebral disc (vossicle) implant in each “quadrant” of the animal’s back. If co-implantation studies are desired, inject cells into vertebral bodies just prior to implantation (see Notes 10 and 17). 4. Close wounds using surgical wound clips. 5. Monitor the recovery of mice and remove clips 2 weeks post-surgery. 6. Treat vossicle recipient mice with the desired pharmacological agent and harvest implants at the end of treatment regime (see Note 18). 7. Perform appropriate endpoint analyses, such as radiographic and histomorphometric analysis (see Fig. 7.4) or other analyses described for ossicles (see Note 19). The effects of treatment regimes on proliferation of implanted vossicles may also be evaluated using BLI where either the vossicle is tracked (using vertebral bodies from luciferase expressing mice) or the tumor is tracked (using luciferase tagged tumor cells).
108
G.J. Pettway and L.K. McCauley
e
a
b
c
d
f
Woven bone Tumor
Fig. 7.4 Vossicle implants. Microradiographic images of vossicles 3 weeks after implantation in a vehicle treated mouse (a), or mouse treated with PTH for 3 weeks (b). Photomicrograph of vossicle from a vehicle treated mouse (c) or mouse treated with PTH for 3 weeks (d). Vossicles can also be co-implanted with tumor cells. (e) Control sham injected without cells, or (f) ACE-1 prostate cancer cells (10,000) cells injected to vertebral bodies just prior to implantation and implants removed 4 weeks later. ACE-1 cells formed tumors and stimulated an “osteoblastic” woven bone phenotype.
4
Notes
1. Implants of BMSCs can also be placed back into the same strain of mice if it is an inbred strain of mice. 2. Surgical instruments used to isolate and implant BMSCs should be autoclaved prior to use to prevent contamination. Bone marrow stromal cells isolated from the long bones of two to three mice can be plated to one T75 flask. Therefore, two to three mice can be sacrificed at a time for BMSC harvest. 3. Ends of bones should be cut so as to preserve as much of the epiphyses as possible since the BMSCs are enriched in the trabecular bone area. Hence, a minimal amount of bone should be removed to enable access with a needle.
7 Osseous Implants
109
4. BMSCs isolated from the long bones of two to three mice plated in one T75 flask usually reach confluency after approximately 7 days in culture. 5. The adherent BMSCs may be difficult to detach from flask. It may be necessary to gently tap the flask to aid in detaching cells. A higher concentration (i.e., 0.5%) of trypsin/EDTA and/or a cell scraper may also be used. 6. BMSCs are implanted at passage 2. Cells should be passaged when they are approximately 90% confluent. If cells reach 100% confluency they may begin to differentiate into osteoblasts, which could compromise bone formation when implanted in vivo. At passage 2, approximately 6–8 × 106 cells may be obtained from bone marrow harvested from the long bones of one 4- to 6-week-old mouse. 7. Sterile forceps can be used to hold Gelfoam in place while cutting with scalpel. 8. When preparing BMSCs for implantation, it is important to work as quickly as possible. The capacity of cells to form bone is reduced if they remain at room temperature for a long period of time. 9. It is important that mouse remains in a lateral position during surgery to place implants securely in pouches created on the backs of mice. If implants are not placed far enough laterally, they may migrate back to midline and coalesce with other implants and/or reside too close to the surgical wound. 10. A total of four implants may be placed in each mouse. 11. The ideal time to initiate injections of the anabolic agent PTH (40–80 µg/kg per day) is 1 week after implantation. Administer PTH daily for 3 weeks. 12. If BMSCs are taken from mice expressing luciferase (e.g., B6; C3-Tg TettTALuc mice from Jackson Laboratory, Bar Harbor, ME) bioluminescent imaging can be used to monitor ossicle growth as a reflection of increased numbers of BMSCs in the implants (see Fig. 7.3). 13. Surgical instruments used to isolate and implant BMSCs should be autoclaved prior to use to prevent contamination. 14. Carefully but firmly hold vertebrae with forceps when cutting single disks. One mouse donor results in approximately six vossicles. 15. Donor and host mice should be coordinated so that vossicle implantation is done just after vossicle isolation. 16. It is important that mouse remains in a lateral position during surgery to place implants securely in pouches created on the backs on mice. 17. For applications of investigating tumor/bone interface, tumor cells can be co-implanted with vossicles (see Fig. 7.4F). Just prior to implantation (see step 3) cells can be injected into the vossicle implant using a Hamilton syringe in a volume of 3–5 µL. Typically 5,000–10,000 tumor cells are used, but this varies widely depending on the growth characteristics of the cell line and the optimal numbers should be determined for each cell line. Depending on the volume of cells injected, two vertebral bodies may be used for one vossicle implant. 18. Mice bearing vossicle implants can have administration of pharmacological agent the next day after implantation. 19. Often cartilage elements are noted in the vossicles upon histological evaluation. They are typically restricted in size and inactive in appearance and can easily be omitted from calculation of bone area or other osseous parameters.
110
G.J. Pettway and L.K. McCauley
Acknowledgments The authors acknowledge the following individuals who have contributed to the development and/or characterization of these models: Amy Koh, Paul Krebsbach, Abraham Schneider, Ana Mattos, and Jinhui Liao. Rashesh Kapadia (Scanco) is acknowledged for the micro-CT images of the ossicle model, as is Thomas Rosol for providing the ACE-1 prostate cancer cells. This work was supported by the National Institutes of Health DK53904 and CA93900.
References 1. Andreassen, T. T., Ejersted, C., Oxlund, H. (1999) Intermittent parathyroid hormone (1-34) treatment enhances callus formation and mechanical strength of healing rat fractures. J Bone Miner Res 14, 960–968. 2. Holzer, G., Majeska, R. J., Lundy, M. W., et al. (1999) Parathyroid hormone enhances fracture healing. A preliminary report. Clin Orthop 366, 258–263. 3. Skriptiz, R., Aspenberg, P. (2001) Implant fixation enhanced by intermittent treatment with parathyroid hormone. J Bone Jt Surg Br 83, 437–440. 4. Kuznetsov, S. A., Friedenstein, A. J., Robey, P. G. (1997) Factors required for bone marrow stromal fibroblast colony formation in vitro. Br J Haematol 97, 437–440. 5. Krebsbach, P. H., Kuznetsov, S. A., Satomura, K., et al. (1997) Bone formation in vivo: comparison of osteogenesis by transplanted mouse and human marrow stromal fibroblasts. Transplantation 63, 1059–1069. 6. Schneider, A., Taboas, J. M., McCauley, L. K., et al. (2003) Skeletal homeostasis in tissueengineered bone. J Orthop Res 21:859–864. 7. Pettway, G. J., Schneider, A., Koh, A. J., et al. (2005) Anabolic actions of PTH (1-34): use of a novel tissue engineering model to investigate temporal effects on bone. Bone 36, 959–970. 8. Koh, A. J., Demiralp, B., Neiva, K., et al. (2005) Cells of the osteoclast lineage as mediators of the anabolic actions of parathyroid hormone in bone. Endocrinology 146, 4584–4596.
Chapter 8
Utility of the Ovariectomized Rat as a Model for Human Osteoporosis in Drug Discovery Yogendra P. Kharode, Michael C. Sharp, and Peter V.N. Bodine
Abstract Ovariectomy-induced osteopenia in the rat produces skeletal responses similar to that in a post-menopausal woman. In the ovariectomized (ovx) rat, high bone turnover, and subsequent bone loss, like in the human post-menopausal condition, can be prevented by estrogen replacement. Because of the striking resemblance of skeletal responses in humans and rats in the state of estrogen deficiency, the ovx rat is considered to be a gold standard model for evaluating drugs for prevention and reversal of osteoporosis. This chapter describes the procedure for performing ovariectomy on the rat and the utility of the ovx rat model we have utilized over the last two decades in our laboratory. Keywords Ovariectomy, bone mineral density, peripheral dual x-ray absorptiometry, peripheral quantitative computed tomography, micro-CT.
1
Introduction
Osteoporosis is a disease characterized by reduction in bone mass, micro-architectural deterioration, and a decrease in biomechanical competence of appendicular and axial skeleton resulting in increased incidence of fractures (1). A variety of physiological and pharmacological conditions are known to cause osteoporosis—of which, the most prevalent is post-menopausal osteoporosis. Estrogen deficiency, as a consequence of menopause, dramatically increases bone turnover (2, 3)—with bone resorption far exceeding bone formation. Estrogen receptors are present in both osteoblasts (4, 5) and osteoclasts (6). Estrogen also promotes development of osteoblasts, increases their proliferation and increases a number of important osteoblast proteins like insulin-like growth factor-1, transforming growth factor-β, and bone morphogenic protein-6 (7–9). At high concentrations, estrogen increases histomorphometric indices of bone formation possibly via increasing synthesis of osteoblastic growth factors. Estrogen restricts osteoclast recruitment and suppresses bone resorptive activities mediated by stromal or monocytic cellular cytokines (IL-1, IL-6, TNF) and RANKL (10, 11). Estrogen also inhibits osteoclast differentiation and recruitment From: Methods in Molecular Biology, Vol. 455: Osteoporosis: Methods and Protocols Edited by Jennifer J. Westendorf © Humana Press, Totowa, NJ
111
112
Y.P. Kharode et al.
by increasing osteoblastic TGFβ (8). Moreover, estrogen increases the cellular level of nitric oxide, which is a potent inhibitor of bone resorption (12). In addition to decreased osteoblastic activity and increased osteoclastic activities, decreased intestinal calcium absorption (13) and increased urinary calcium excretion (14) contribute to the excessive bone loss in the estrogen-deficient state. Estrogen therapy (ET)—although associated with endometrial and breast cancer risk in a limited population—has been utilized and proven to be effective in prevention of postmenopausal bone loss and the incidence of fractures (15–18). Moreover, because estrogen deficiency results in an increase in bone turnover, mainly favoring an increase in bone resorption, a number of non-hormonal therapies also have been used to counteract the deleterious effects in both humans and rats. These non-hormonal agents act by attenuation of bone resorption (19, 20), enhancement of bone formation (21, 22) or dual-bone anti-resorptive and osteogenic-actions (23, 24). The rat skeleton has many similarities to that of humans with the exception of continuous modeling and longitudinal bone growth in the rat (25). The skeletal responses to depletion and repletion of estrogen (in the estrogen deficient state), as well as the pharmacological effects of known agents on the rat skeleton, are comparable with that in humans; thus, making the rat the species of choice for pre-clinical evaluation of new drugs as potential future therapies for human metabolic bone disorders such as osteoporosis.
1.1
Ovariectomy-Induced Osteopenia Model
The striking resemblance of the ovx rat to humans with respect to estrogen deficiency related pathophysiology—i.e., increase in bone turnover, bone loss—osteopenia (26–29) and prevention of the same by estrogen replacement (30–32) make the ovx rat a gold standard model of human osteoporosis. Since the ovx model effectively demonstrated the skeletal response of known agents such as conjugated equine estrogen (31) and parathyroid hormone (33, 34) in a fashion similar to that in postmenopausal women, the ovx rat model has been an animal model of choice for evaluation of new therapies for both prevention (8, 35–41) and treatment (21, 22, 42–45) of osteoporosis. In fact, the Food and Drug Administration also has recommended the utilization of the ovx model for pre-clinical evaluation of anti-osteoporotic drugs (46).
2 2.1
Materials Surgical Procedure
1. Surgical Instruments and Supplies: Clippers, scissors, disposable cauterizers, blunt forceps, tissue forceps, hemostats, scalpels #10 and #15 blades, 3-0 suture, gauze packs, syringe/needle (25 G), surgical apparel.
8 Rat Models for Human Osteoporosis
2. 3. 4. 5. 6.
113
Betadine. Ethyl alcohol. Ketamine (Bristol Laboratories, Syracuse, NY). Xylazine (Bristol Laboratories). Acepromazine (Aveco, Ft. Dodge, IA).
2.2
Evaluation of Agents in OVX Models
1. Ketamine, xylazine, and acepromazine as listed in Section 2.1. 2. Syringe/needle (25 G) and surgical apparel. 3. Ear-tags.
2.3
BMD Measurements Using Peripheral Dual X-Ray Energy Absorptiometer
1. Peripheral dual x-ray energy absorptiometer (pDXA), Sabre (Norland Corp., Ft. Atkins, WI).
2.4
BMD Measurements Using Peripheral Quantitative Computed Tomography
1. XCT Research SA+ (pQCT; Stratec Medizintechnik, Pforzheim, Germany).
2.5
Histology
1. Automated tissue processor (Shandon Hypercenter XP, Pittsburgh, PA). 2. Microtome (Leica RM2065, Heidelberg, Germany). 3. Bioquant Osteo image analysis system (Bioquant Image Analysis Corp., Nashville, TN). 4. Nikon E800 microscope (Nikon Corp., Melville, NY). 5. Methyl methacrylate (MMA) (Aldrich, St. Louis, MO). 6. Silane-coated positively charged slides (Surgipath Snowcoat X-tra Micro Slides, Richmond, IL).
2.6
Bio-Mechanical Testing
1. Instron 5543 with Instron 1 kN. load cell (Instron Corp., Canton, MA). 2. Isomet diamond blade wafer saw (Buehler Ltd., Lake Bluff, IL).
114
2.7
Y.P. Kharode et al.
Biochemical Analysis of Serum/Urine Parameter
1. Hitachi 911 analyzer. 2. Biomarker (osteocalcin, CTX, PYD) assay kits from suppliers.
3
Methods
Rats from a wide range of age groups (8 weeks to 15 months) and a variety of treatment periods are used by different investigators for examining the effects of agents in the ovx model. While use of older animals for the ovx model is attractive because of lower bone modeling and steady bone turnover rate, with careful study design and proper controls, the ovx model using young adult rats also can provide consistent, reproducible, and interpretable results (see Notes 1 and 2).
3.1
Ovariectomy Procedure
Bilateral ovariectomy (or sham operation) may be performed as follows: 1. Anesthetize rats with an intraperitoneal injection of 45 mg/kg ketamine, 8.5 mg/ kg, xylazine, and 1.5 mg/kg acepromazine, shave the fur over the dorsal lumbar area, disinfect the skin with Betadine followed by an alcohol rinse. 2. Make a 2-cm skin incision along the dorsal midline (just caudal to the last rib) and through the abdominal musculature (Fig. 8.1). 3. Gently grasp the ovarian fat pad using forceps, expose and remove the ovary. Use cautery to control any bleeding (Fig. 8.2).
Fig. 8.1 Incision site for ovariectomy in rats.
8 Rat Models for Human Osteoporosis
115
Fig. 8.2 Anatomy of the urogenital system of the female rat.
4. Stitch the muscle with 3-0 absorbable sutures and use stainless steel wound clips (two to three each side) to close the skin incision. The wound clips may be removed in 7 to 10 days post-surgery. 5. Sham operation may be performed using above steps but without removing the ovaries. Surgery takes about 10 minutes per animal. Instruments are sterilized between animals using a glass bead sterilizer.
3.2
Prevention Protocol
Bone anti-resorptive agents like estrogens, bisphosphonates, selective estrogen receptor modulators (SERMs), and others can be evaluated for their bone sparing activities using the protocol listed in the following. We have utilized the prevention protocol for evaluation of agents such as CEE (31) and SERMs (40) following 5 or 6 weeks of treatment in ovx rats. 1. Weigh and ear-tag ovx and sham operated rats 2 days after surgery and divide them randomly in groups of 8 to 10 rats/group so that the average body weights for each ovx group is similar. Provide casein diet and water ad libitum. 2. Initiate daily treatment and continue the regimen for 6 weeks. Administer 10 mg/kg calcein, sc 2 and 9 days before necropsy (see Note 3). 3. During the study, at the desired time point, collect urine overnight using metabolism cages. 4. Prior to necropsy, evaluate bone mineral density using pDXA and/or pQCT as described in the following sections.
116
Y.P. Kharode et al.
5. Two hours after the last day’s treatment, euthanize animals by over exposure to CO2. Collect tibiae, femurs, spine, blood, and uteri. Other tissues may also be collected as required by the study design. 6. Place tibia in 40% ethyl alcohol (EtOH), store femurs and spine at −20 °C. Separate serum/plasma as needed and store as required by assay conditions. Excised bones can be analyzed for skeletal parameters by techniques listed in the following sections.
3.3
Evaluation of Agents in OVX Rats with Established Osteopenia
Ovx produces significant trabecular bone loss in long bones of rats in 3 to 4 weeks after the surgery. Osteogenic agents like PTH reverse the ovx-induced bone loss. The length of treatment period usually depends on the severity of osteopenia, e.g., if the osteopenia development period (post-ovx) is longer prior to initiation of the osteogenic regimen, the reversal of osteopenia typically requires longer treatment period. We have observed that the 3 weeks post-ovx osteopenia development period is adequate for establishing a significant trabecular bone loss in mature Sprague-Dawley rats and the lost bone may be completely restored by daily sc administration of 10 µg/kg per day, hPTH for 3 weeks (43). Maintenance of therapeutic effects of hPTH by anti-catabolic agents like estrogen and SERMs may also be evaluated using this protocol (47). However, longer durations of osteopenia are also utilized to establish a lower remodeling turnover model. The variation of this protocol from the prevention protocol are listed below: 1. Weigh and ear-tag ovx and sham operated rats 2 days after surgery and divide them in groups of 8 to 10 rats/group. Provide casein diet and water ad libitum. 2. Four weeks after ovx, evaluate base line bone mineral density using pDXA and/ or pQCT as described in the following. 3. Initiate daily treatment and continue for 4 weeks. Administer 10 mg/kg calcein, sc 2 and 9 days before necropsy (see Note 3). 4. Prior to necropsy, evaluate bone mineral density using pDXA, and/or pQCT as described in the following. 5. Two hour after the last day’s treatment, euthanize animals by overexposure to CO2. Collect tibiae, femurs, spine, blood, and uteri. 6. Place tibia in 40% EtOH, store femurs, and spine at −20 °C. Separate serum/ plasma as needed and store as required by assay conditions. For evaluation of skeletal parameters (in vivo and in excised tissues) and other parameters, we have utilized the following techniques in our laboratory.
8 Rat Models for Human Osteoporosis
3.4
117
Evaluation of Areal BMD of 4th Lumbar Vertebra (L4) Using pDXA
BMD of the fourth lumbar vertebra (L4) may be evaluated in anesthetized rats using a pDXA. The dual energy x-ray absorptiometry (pDXA) measurements in the ovx (or sham operated) rat is performed as follows: 1. Fifteen minutes prior to DXA measurements, the anesthetized rat is placed on the Plexiglas window under the scanner perpendicular to the path of scanner aligning the spine straight in the center of Y axis. 2. A preliminary scout scan (6 × 6 cm) is performed at a scan speed of 40 mm/ second with a scan resolution of 1.0 × 1.0 mm to determine the region of interest in L4. 3. A measurement-scan (4 × 4 cm) at a scan speed of 5 mm/second and a resolution of 0.2 × 0.2 mm is taken for precise BMD measurement of L4. 4. The software allows the operator to define the total region of L4 from the images generated by the computer following the scan. The BMD is computed by the software as a function of attenuation of the dual beam (28 and 48 KeV) x-ray generated by the source underneath the subject and the detector traveling along the defined area above the subject. 5. The data for BMD values (expressed in g/cm2) and individual scans may be stored and later subjected to the statistical analysis (see Note 4).
3.5
Evaluation of Volumetric BMD of Proximal Tibia Using pQCT
Total and trabecular density of the proximal tibia can be evaluated in anesthetized rats using an XCT Research SA+ (see Note 5). 1. The right hind limb is passed through a polycarbonate tube with a diameter of 25 mm and taped to an acrylic frame with the ankle joint at a 90° angle and the knee joint at 180°. The polycarbonate tube is affixed to a sliding platform that maintained it perpendicular to the aperture of the pQCT. The platform is adjusted so that the distal end of the femur and the proximal end of the tibia would be in the scanning field. 2. A two-dimensional scout view is run for a length of 10 mm and a line resolution of 0.2 mm. After the scout view is displayed on the monitor, the proximal end of the tibia is located. The pQCT scan is initiated 3.4 mm distal from this point. The pQCT scan of 0.5 mm thickness with a voxel (three-dimensional pixel) size of 0.07 mm is obtained by 360 projections through the slice. 3. After the pQCT scan is completed, the image is displayed on the monitor. A region of interest including the tibia, but excluding the fibula is outlined. The soft tissue is automatically removed using an iterative algorithm.
118
Y.P. Kharode et al.
4. The density of the entire scanned bone (total density) is reported in mg/cm3. The outer 55% of the bone is peeled away in a concentric spiral. The density of the remaining bone (trabecular density) is reported in mg/cm3. 5. The results are saved in a database file for statistical analysis and the original scan are saved in the hard disk.
3.6
Micro-Computed Tomography
Three-dimensional volumetric analysis of the excised proximal tibia can be performed using Scanco µCT 40 (48, 49). 1. A scout view of the tibia is performed to include the proximal end extending approximately 7 mm distally and parallel to the growth plate. 2. A lead reference line for the CT measurement is placed at the proximal end of the tibia. Measurement begins at this point and continued distally for 350 scan slices (20 µm thickness). 3. A region of interest (ROI) distal to the growth plate is selected for analysis. 4. Images are reconstructed into three-dimensional arrays with an isotropic voxel size of 20 µm. The resulting gray-scale images are segmented using a constrained 3D Gaussian filter (σ = 1.0, support = 0.8), and bone structure is extracted using a fixed threshold (normally around 20% of maximal grayscale value; see Note 6). The trabecular bone within the proximal tibia is identified using manually drawn contouring algorithms on the region of interest (primary and secondary spongiosa). 5. Morphometric variables are computed within the volume of interest from the binary images. Morphometric analysis is performed by a direct, three-dimensional technique using Scanco software. Morphometric parameters included in the analysis are bone volume fraction (BV/TV, %), connectivity density (Conn.D.), trabecular number (Tb.N., mm−1), trabecular thickness (Tb.Th. µm) and trabecular spacing (Tb.Sp., µm) (see Notes 6, 7, and 8).
3.7
Histology
3.7.1
Evaluation of Histomorphometric Parameters in Proximal Tibia Samples
1. The proximal tibia is prepared for tissue processing, following necropsy, by cutting longitudinally through the anterior eminence with a scalpel to provide both a stable base for orientation and to allow rapid penetration of fixative into the cancellous bone. The diaphysis is then cross sectioned 1.5 cm from end of the proximal metaphysis and the resulting sample is placed into a labeled cassette and fixed in cold, approximately 5–8 °C, 40% EtOH for 48 hours for preservation
8 Rat Models for Human Osteoporosis
119
of cellular adhesion to bone surfaces before being placed into cold 70% EtOH for subsequent processing and storage. 2. The metaphysis is placed in an automated tissue processor (Shandon Hypercenter XP) and infiltrated with MMA. The MMA infiltrated specimen is embedded with the cut surface of the anterior eminence placed face down along the bottom of a glass vial containing MMA polymerization mixture. The open vials are then put under vacuum for 3 hours and nitrogen is run over samples before capping. Complete polymerization by placing samples in a water bath at 37 °C overnight. 3. Samples are rough ground to expose the longitudinal midline of the metaphysis from which thin and thick sections are taken on a microtome (Leica RM2065) containing a tungsten carbide knife. Sections are collected on positively charged glass slides with thick sections cut to 10 µm for observation of dynamic indices and thin sections cut to 5 µm for subsequent staining and evaluation of static parameters (see Note 9).
3.7.2
Acquisition of Dynamic and Static Histomorphometric Measures in Rat Proximal Tibia
Dynamic bone indices are evaluated from fluorescein (calcein) labeled proximal tibia sectioned at 10 µm on a Bioquant Osteo image analysis system. 1. Cancellous bone in the region of secondary spongiosa, which begins approximately 450 µm below the endochondral plate, is examined with a Nikon E800 microscope using plan fluor objectives for mineralized trabecular surface for a total length of 1,500 µm into the diaphysis taking care not to include cortical bone. 2. Fluorescence emission of the calcein labels is achieved by exciting in the blue spectrum at a wavelength of 465–495 nm to reduce background signal. Mineral apposition rate (MAR) is determined by measuring the inter-label distance at 20 µm intervals between the mineralized calcein double labels and calculating the mean inter-label distance divided by the inter-label period. The resulting primary histomorphometric index is reported as µm/day. To account for error, in measuring the inter-label distance, due to oblique orientation of mineralizing surfaces the mean inter-label distance is modified by a divisor of 1.2 to improve accuracy of the measure (50, 51). 3. Total mineralized surface in the region of the secondary spongiosa is measured to include individual single label length and double label length to determine the mineralizing surface (MS = sum double label surface + ½ sum single label surface) as well as the total bone surface to derive the bone formation rate (BFR). BFR = MAR × (MS/BS) expressed in mm3/mm2 per day 4. Static parameter measures (bone volume, osteoid volume, osteoblast, and osteoclast number and surface) are taken from 5-µm sections, which are stained to reveal cellular detail and distinguish structural components using a modified Goldner’s trichrome or 0.1% toluidine at pH 3.5.
120
3.8
Y.P. Kharode et al.
Bio-Mechanical Testing: Compressive Strength of Vertebrae
The compressive strength of fifth lumbar vertebra (L5) is determined using Instron 5543 equipped with a 1,000 Newton (N) load cell as follows (43). 1. Spine is thawed at room temperature and L5is dissected. 2. Using a low speed Isomet diamond blade wafer saw (Buehler Ltd., Lake Bluff, IL), remove all processes. Two coplanar cuts perpendicular to the cephalocaudal axis 4.9 mm apart are made. This produces a uniform sample for compression. 3. The sample is placed in a fixture with a piston moving parallel to the cephalocaudal axis compressing the sample. 4. Data are transmitted to Instron Merlin software, which produces a load-deformation curve. Maximum load is calculated from the curve and expressed as Newtons (N).
3.9
Biochemical Parameters
Serum bone formation parameter (osteocalcin) and bone resorption parameter (CTX, cross-lap) as well as urinary pyridinoline cross-links (PYD, a bone resorption parameter) may be measured using commercially available kits and the procedures described by the manufacturers of the assay kits. Serum cholesterol may be analyzed on a Hitachi 911 instrument using Boehringer Mannheim reagents (for cholesterol). The cholesterol is measured via o-quinone imine dye (which is formed following enzymatic reactions with cholesterol) and is analyzed photometrically at 505 nm at 37 °C and expressed as mg/dL (40).
3.10
Closing Remarks
Many similarities in skeletal effects of estrogen deficiency between rat and human allow the utility of ovx rat as a gold standard model of human osteoporosis. The limitations of the rat model, however, are: (1) continuous modeling through out life, and (2) minimal cortical effects of ovx (49). Nevertheless, the ovx rat model can be used for evaluation of agents for both prevention and treatment of osteoporosis. The ovx rat model with proper study design and use of densitometry, µCT, histomorphometry, bio-mechanical testing, and bio-chemical markers may allow investigators to collect interpretable data on test compounds that can possibly help in predicting clinical effects of the compounds in humans with osteoporosis.
8 Rat Models for Human Osteoporosis
4
121
Notes
1. In our ovx studies, we have routinely utilized 10 to 12 weeks old Sprague Dawley rats weighing 225 to 250 g. 2. Alternatively, ovx and sham operated rats may be purchased from suppliers such as Taconic Farms (Germantown, NY) or Charles River (Wilmington, MA). 3. Other chromophores such as tetracycline, demeclocycline, or xylenol orange may also be used to label the bone. The interval between the two labels may be changed as well. 4. We use n = 8 to 10 rats/group for evaluation of skeletal effects in rats. 5. Areal BMD of tibia may also be analyzed using the preceding procedure with proper positioning of the rat and defining the region of interest from the image. 6. This method may be utilized using other similar pQCT equipment such as XCT Research M or XCT 960M (43). 7. Threshold value must be set prior to analyzing the first sample of the study and should be kept the same for entire study. 8. Bone histomorphometry parameters are accepted from standardized nomenclature by Parfitt et al. (52). 9. Micro-CT evaluation of vertebrae samples may be performed using the same technique (described for tibia) by scanning the whole vertebrae and analyzing the region between the two end plates. Acknowledgment The authors wish to dedicate this article to our late director, Shunichi Harada, who devoted his professional life to osteoporosis research. We also thank Fred Bex and Barry Komm for their guidance, direction, and support of the Wyeth osteoporosis research group. Finally, they are grateful to many talented scientists whose excellent technical support over the last two decades have been instrumental in developing and implementing animal models and analytical methods in their laboratory: special thanks to Joseph Tamasi, James Morzolf, Michelle French, Stephani Smoluk, Andrea Scarcia, Vanessa Dell, Mellisa Wasco, Paula Green, Sally Selim, Colleen Millgan, and Jennifer Pirrello as well as the BioResources staff at Wyeth Research for their valuable assistance.
References 1. Consensus Development Conference. (1993) Diagnosis, prophylaxis and treatment of osteoporosis. Am J Med 94, 646–650. 2. Heaney, R. P., Recker, R. R., Saville, P. D. (1978) Menopausal changes in bone remodeling. J Lab Clin Med 92, 964–970. 3. Garnero, P., Sornayrendu, E., Chapau, M. C., et al. (1996) Increased bone turnover in late postmenopausal woman is a major determinant of osteoporosis. J Bone Miner Res 11, 337–349. 4. Komm, B. S., Terpening, C. M., Benz, D. J., et al. (1988) Estrogen binding receptor mRNA, and biologic response in osteoblast-like osteosarcoma cells. Science 241, 81–84.
122
Y.P. Kharode et al.
5. Ericksen, E. F., Colvard, D. S., Bergn N. J., et al. (1988) Evidence of estrogen receptors in normal human osteoblast-like cells. Science 241, 84–86. 6. Oursler, M., Osdoby, P., Pyfferon, J., et al. (1991) Avian osteoclasts as estrogen target cells. Proc Natl Acad Sci USA 88, 6613–6617. 7. Ernst, M., Heath, J. K., Rodan, G. A. (1989) Estradiol effects on proliferation, messanger ribonucleic acid for collagen and insulin-like growth factor-1, and parathyroid hormone-stimulated adenylate cyclase activity in osteoblastic cells from calvariae and long bones. Endocrinology 125, 825–833. 8. Yang, N. N., Bryant, H. U., Hardikar, S., et al. (1996) Estrogen and raloxifene stimulate transforming growth factor-beta-3 gene expression in rat bone: a potential mechanism for estrogenor raloxifene-mediated bone maintenance. Endocrinology 137, 2075–2084. 9. Rickard, D. J., Hofbaur, L. C., Bonde, S. K., et al. (1998) Bone morphogenic protein-6-production in human osteoblast-like cell lines: selective regulation by estrogen. J Clin Invest 101, 413–422. 10. Girasolle, G., Jilka, R. L., Passeri, G., et al. (1992) Marrow-derived stromal cells and osteoblasts in vitro: a potential mechanism for the anti-osteoporotic effects of estrogens. J Clin Invest 89, 883–891. 11. Rogers, A., Estell, R. (1998) Effects of estrogen therapy of postmenopausal women on cytokines measured in peripheral blood. J Bone Miner Res 13, 1577–1586. 12. Armour, C. E., Ralston, S. H. (1998) Estrogen upregulates endothelial constitutive nitric oxide synthase expression in human osteoblast-like cells. Endocrinology 139, 799–802. 13. Arjandi, B. H., Sahil, M. A., Herbert, D. C., et al. (1993) Evidence for estrogen receptorlinked calcium transport in the intestine. Bone Miner 21, 63–74. 14. Heaney, R. P., Recker, R. R., Saville, P. D. (1978) Menopausal changes in calcium balance performance. J Lab Clin Med 92, 953–963. 15. Recker, R. R., Saville, P. D., Heaney, R. P. (1977) Effects of estrogen and calcium carbonate on bone loss in postmenopausal women. Ann Int Med 87, 649–655. 15. Gallagher, J. C., Kable, W. T., Goldgar, D. (1991) Effects of progestin therapy on cortical and trabecular bone: comparison with estrogen. Am J Med 90, 171–178. 15. Vedi, S., Compston, J. E. (1996) The effect of long-term hormone replacement therapy on bone remodeling in postmenopausal women. Bone 19, 535–539. 16. Lindsay, R., Cosman, F. (2006) Effects of estrogen intervention on the skeleton, in (Favus, M. J., ed.), Primer on the Metabolic Bone Disease and Disorders of Mineral Metabolism, 5th ed. The American Society for Bone and Mineral Research, Washington, DC. 17. Kalu, D. N., Liu, C. C., Salerno, E., et al. (1991) Skeletal response of ovariectomized rats to low and high doses of 17-beta estradiol. Bone Miner 14, 175–187. 18. Plosker, G. L., Goa, K. L. (1994) Clodronate: a review of its pharmacological properties and therapeutic efficacy in resorptive bone disease. Drugs 47, 945–982. 19. Plum, L. A., Fitzpatrick, L. A., Ma, X., et al. (2006) 2MD, a new anabolic agent for osteoporosis treatment. Osteoporosis Int 17, 704–715. 20. Dempster, D. W., Cosman, F., Parisien, M., et al. (2006) Anabolic actions of parathyroid hormone on bone. Endocr Rev 14, 690–709. 21. Shen, V., Dempster, D. W., Mellish, R. W. E., et al. (1992) Effects of combined and separate intermittent administration of low-dose human parathyroid hormone fragment (1-34) and 17β-estradiol on bone histomorphometry in ovariectomized rats with established osteopenia. Calcif Tiss Int 50, 214–220. 22. Wronski, T. J., Yen, C. F., Scott, K. S. (1991) Estrogen and diphosphonate treatment provide long-term protection against osteopenia in ovariectomized rats. J Bone Miner Res 6, 387–394. 23. Jee, W. S. S., Yao, W. (2001) Overview: animal models of osteopenia and osteoporosis. J Musculoskel Neuron Interact 1, 193–207. 24. Wronski, T. J., Lowery, P. L., Walsh, C. C., et al. (1985) Skeletal alterations in ovariectomized rats. Calcif Tissue Int 37, 324–328. 25. Wronski, T. J., Walsh, C. C., Ignaszewski, L. A. (1986) Histologic evidence for osteopenia and increased bone turnover in ovariectomized rats. Bone 7, 119–123.
8 Rat Models for Human Osteoporosis
123
26. Wronski, T. J., Cintron. M., Dann, L. M. (1988) Temporal bone loss and increased bone turnover in ovariectomized rats. Calcif Tissue Int 43, 179–183. 27. Wronski, T. J., Dann, L. M., Scott, K. S., et al. (1989) Long-term effects of ovariectomy and aging on the rat skeleton. Calcif Tissue Int 45, 360–366. 28. Turner, R. T., Vandersteenhoven, J. J., Bell, N. H. (1987) The effects of ovariectomy and 17βestradiol on cortical bone histomorphometry in growing rats. J Bone Miner Res 2, 115–122. 29. Hayward, M. A., Kharode, Y. P., Becci, M. M., et al. (1990) The effect of conjugated equine estrogen on ovariectomy-induced osteopenia in the rat. Agents Actions 31, 152–156. 30. Kalu, D. N. (1991) The ovariectomized rat model of postmenopausal bone loss. Bone Miner 15, 175–192. 31. Gunness-Hey, M., Hock, J. M. (1984) Increased trabecular bone mass in rats treated with human synthetic parathyroid hormone. Metab Bone Dis Rel Res 5, 177–181. 32. Hock, J. M., Fonesca, J., Gunness-Hey, M., et al. (1989) Comparison of the anabolic effects of synthetic parathyroid hormone-related protein (PTHrP) 1-34 and PTH 1-34 on bone in rats. Endocrinology 125, 2022–2027. 33. Seedor, J. G., Quartuccio, H. A., Thompson, D. D. (1991) The bisphosphonate alendronate (MK-217) inhibits bone loss due to ovariectomy in rats. J Bone Miner Res 6, 339–346. 34. Muller, K., Wisenberg, I., Jaeggi, K., et al. (1998) Effects of the bisphosphonate Zoledronate on bone loss in the ovariectomized and in the adjuvant arthritic rat. Arzneim-Forsch Drug Res 48, 81–86. 35. Marie, P. J., Hott, M., Modrowski, D., et al. (1993) An uncoupling agent containing strontium prevents bone loss by depressing bone resorption and maintaining bone formation in estrogendeficient rats. J Bone Miner Res 8, 607–615. 36. Black, L. J., Sato, M., Rowley, E. R., et al. (1994) Raloxifene (LY139481 HCl) prevents bone loss and reduces serum cholesterol without causing uterine hypertrophy in ovariectomized rats. J Clin Invest 93, 63–69. 37. Ke, H. Z., Chen, H. K., Simmon, H. A., et al. (1997) Comparative effects of droloxifene, tamoxifen, and estrogen on bone, serum cholesterol, and uterine histology in the ovariectomized rat model. Bone 20, 31–39. 38. Komm, B. S., Kharode, Y. P., Bodine, P. V. N., et al. (2005) Bazedoxifene acetate: a selective estrogen receptor modulator with improved selectivity. Endocrinology 146, 3999–4008. 39. Uchiyama, Y., Higuchi, Y., Takeda, S., et al. (2002) ED-71, a Vitamin D analog, is a more potent inhibitor of bone resorption than alfacalcidol in an estrogen-deficient rat model of osteoporosis. Bone 30, 582–588. 40. Kimmel, D. B., Bozzato, R. P., Kronis, K. A., et al. (1993) The effect of recombinant human (1-84) or synthetic human (1-34) parathyroid hormone on the skeleton of adult osteopenic ovariectomized rats. Endocrinology 132, 1577–1584. 41. Murrills, R. J., Matteo, J. J., Samuel, R. L., et al. (2004) In vitro and in vivo activities of C-terminally truncated PTH peptides reveal a disconnect between cAMP signaling functional activity. Bone 35, 1263–1272. 42. Meng, X. W., Liang, X. G., Birchman, R., et al. (1996) Temporal expression of the anabolic action of PTH in cancellous bone of ovariectomized rats. J Bone Miner Res 11, 421–429. 43. Ke, H. Z., Jee, W. S. S., Zeng, Q. Q., et al. (1993) Prostaglandin E2 increased rat cortical mass when administered immediately following ovariectomy. Bone Miner 21, 189–201. 44. Guidelines for Preclinical and clinical evaluation of agents used in the prevention or treatment of postmenopausal osteoporosis. (1994) Division of Metabolism and Endocrine Drug Products: Food and Drug Administration. 45. Kharode, Y. P., Marzolf, J. T., Bodine, P. V. N., et al. (2002) Maintenance of therapeutic effects of hPTH in ovariectomized rats with established osteopenia: evaluation of bazedoxifene, raloxifene and ethinyl estradiol. J Bone Miner Res 17, S209. 46. Kharode, Y. P., Green, P. D., Marzolf, J. T., et al. (2003) Comparison of the effects of bazedoxifene, raloxifene, lasofoxifene, and risedronate co-treatment on hPTH-induced reversal of established osteopenia in ovariectomized rats. J Bone Miner Res 18, S273.
124
Y.P. Kharode et al.
47. Kimmel, D. B., Wronski, T. J. (1990) Nondestructive measurement of bone mineral in femurs from ovariectomized rats. Calcif Tiss Int 46, 101–110. 48. Frost, H. M. (1983) Bone histomorphometry: analysis of trabecular bone dynamics, in (Recker, R. R., ed.), Bone Histomorphometry: Techniques and Interpretations. CRC Press, Boca Raton, FL. 49. Reinhold, E. (2003) Bone-Labeling Techniques in (An, Y. H., Martin, K. L. eds.) Handbook of Histology Methods for Bone and Cartilage. Humana Press, Totowa, NJ. 50. Parfitt, A. M., Drezner, M. K., Glorieux, F. H., et al. (1987) Bone histomorphometry: standardization of nomenclature, symbols, and units. J Bone Miner Res 2, 595–610.
Chapter 9
Orchidectomy Models of Osteoporosis Stéphane Blouin, Hélène Libouban, Marie Françoise Moreau, and Daniel Chappard
Abstract Long considered a disease of post-menopausal women, osteoporosis is increasingly being recognized among the growing population of elderly men. Androgen deficiency may be associated with an increase of bone resorption in elderly men and so, with remodeling imbalance and fracture risk. It is firmly established that androgen withdrawal induced by orchidectomy (ORX) results in decreased bone mass in animal models especially in rodents. The mature rat is the model of choice. Skeletal effects of ORX in rats have been studied at the tissular and cellular level. It induces a decrease of BMD and BV/TV with microarchitecture alterations due to an increased bone remodeling. The present chapter focuses on the ORX surgery in rats and mice. Keywords Orchidectomy, osteoporosis, rat, mice, DEXA, histomorphometry.
1
Introduction
Long considered a disease of post-menopausal women, osteoporosis is increasingly being recognized among the growing population of elderly men. There is increasing evidence for a relationship between age-related endocrine changes and osteoporosis in men. Androgen deficiency may be associated with an increase of bone resorption in elderly men and so, with remodeling imbalance and fracture risk (1). The Food and Drug Administration (FDA) and World Health Organization (WHO) have recommended the ovariectomized (OVX) rat for postmenopausal osteoporosis research (2); the orchidectomized rat (ORX) has been proposed to simulate male osteoporosis due to hypogonadism (3–15) and to evaluate injectable biomaterials (16). ORX mice were also used in few studies (17–19). This model is interesting to evidence regulation pathway in transgenic mice (20, 21). In this introduction, we will discuss the major results obtained in the rat model. The most important drawback of the rat skeleton is its lifelong growth; however, the modeling influence can be minimized by using mature animals with preponderant bone remodeling (from 4-month-old rats) (5, 9, 22). Moreover, analysis should From: Methods in Molecular Biology, Vol. 455: Osteoporosis: Methods and Protocols Edited by Jennifer J. Westendorf © Humana Press, Totowa, NJ
125
126
S. Blouin et al.
be performed on the secondary spongiosa to avoid the remaining modeling activity in the primary spongiosa. Skeletal effects of ORX have been studied at the tissular and cellular level by several techniques: dual energy X-ray absorptiometry (DEXA), histomorphometry, serum dosage, and X-ray microtomograph. By DEXA, a decrease of the femoral bone mineral density (BMD) in the ORX rat was evident between 1 to 4 months postsurgery (5) but histomorphometric evaluation revealed the trabecular bone loss more rapidly and with greater sensitivity. Decreases in BV/TV in tibia were found between 2 and 3 weeks post-ORX (5, 9, 10, 23), whereas DEXA found differences after 8 weeks (24, 25). Histomorphometry showed decreased trabecular bone volume (BV/TV) of 40% in OVX mice, whereas DEXA indicated a decrease of BMD of 5% (26). ORX also induced a decrease of vertebral BV/TV between 2 and 3 weeks (5, 23) and cortical bone loss associated with increased endosteal porosity (15, 22). Histomorphometry allows measurement of microarchitecture and bone remodeling parameters. ORX induces a decrease of trabecular number and an increase of trabecular separation without alteration in trabecular thickness (9, 23, 25). Fractal dimension of the trabecular network most efficiently describes trabecular alterations because it is the earliest altered parameter (2 weeks) (25). Trabecular bone loss due to ORX is associated with increased bone remodeling. In adult rat, the osteoclast number and the osteoblastic surfaces increased early after the androgen deprivation (27). Increased bone remodeling was found 1 month post ORX; this increase was found to be transient and occurred before the trabecular bone loss observed 4 month after ORX (28). The bone remodeling can be measured by the dosage of markers evaluating resorption (TRAcP, deoxypyridinoline, and pyridinoline) or formation (osteocalcin) (23, 29–31). The mineralization rate can be affected transiently by ORX (5, 9) or not (28). In cortical bone, ORX induces a decrease of periosteal bone formation (5). The combination of different osteopenic factors has also been proposed to obtain severe bone loss in the rat. Iwamoto et al. studied the conjoined effects of ORX and sciatic neurectomy in adult rats. The association of both factors was associated with a more pronounced decrease of trabecular bone mass and increase of cortical porosity than with neurectomy alone (6, 7). Massive bone loss was obtained in a study combining the effects of ORX (acting on the whole skeleton) and paralysis induced by injection of botulinum toxin, acting on a single hind limb (31). The association of both factors led to considerable bone loss with marked alterations in bone microarchitecture (marked disconnection of trabeculae). The cumulative effect of disuse and ORX induced a peak of serum TRAcP higher than ORX alone (31). This chapter focuses on the ORX surgery in rats and mice.
2
Materials
Use of trade names is for identification purposes only and does not imply endorsement.
9 Orchidectomy Models of Osteoporosis
2.1
127
Anesthesia
Anesthesia of rats or mice is performed with one of the following agents (see Note 1).
2.1.1
Isoflurane
Isoflurane is the method of choice for rodent anesthesia. It is safe and very easy to use. Induction and awakening are rapid. Gas waste must be scavenged properly. The following material is needed: 1. An induction chamber in order to progressively put to sleep a conscious animal. 2. An oxygen flowmeter to regulate flow of the oxygen bottle so that the outlet pressure is compatible with its use. 3. A conical mask adaptable to the animal’s snout to maintain animals under anesthesia. 4. A gas vaporizer filled with isoflurane. It ensures a stable concentration of the gas mixture and provides a continuous flow of anesthetic. 5. HALOSORB cartridges (Minerve, Esternay, France), which are activated carbon cartridges used to trap residual anesthesia particles protect the operator against residual anesthetic.
2.1.2
Xylazine–Ketamine
1. Rompun containing the active ingredient xylazine, solution 2% (Bayer Pharma, Puteaux, France) (20 mg xylazine/mL). It is currently used in veterinary medicine for sedation, analgesia and muscle relaxation. 2. Ketalar containing the active ingredient ketamine chlorohydrate (Parke-Davis, Courbervoie, France) (50 mg ketamine/mL). Ketamine is a powerful anesthetic used primarily in veterinary medicine for anesthesia of farm animals during routine surgery. 3. Syringes of 1 or 2 mL according to animal weights. 4. Hypodermic needles 26 G.
2.2
Animal Preparation
1. An infrared lamp to prevent heat loss (see Note 2). 2. An electric clipper or a depilatory cream to remove hair/fur from the area of surgery. 3. A weight scale to adapt the volume of anesthetic to the weight of each animal.
128
2.3 1. 2. 3. 4.
S. Blouin et al.
Surgery (Fig. 9.1A) A beaker to immerse surgery instrument into disinfectant solution. Ethanol 70%. Michel clips for the wound closure (see Note 3). Betadine, which contains a 10% povidone-iodine solution to spray the surgical site.
Fig. 9.1 Methodology of orchidectomy. (a) Surgical equipment, (b) induction equipment of isoflurane anesthesia, (c) anesthesia maintenance of the rat using the conical mask, (d) site of incision (after hair/fur removal), (e) skin and muscle incision; testis is visible, (f) cauda epididymis, testis, caput epididymis, vas deferens and testicular blood vessels are extracted from the scrotum sac, (g) testicular blood vessels are visible after careful disconnection of epididymis and fatty tissue from testis, (h) ligature around blood vessels, (i) removal of testis 2 mm under the node.
9 Orchidectomy Models of Osteoporosis
129
5. 6. 7. 8.
Surgical blunt forceps. Michel clip applying forceps. Fine straight scissors. Vicryl (polyglactin 910) suture, which is a synthetic absorbable braided sterile surgical suture (Ethicon, Johnson and Johnson, Somerville, NJ). The suture gauge should be 2-0 for rats and 4-0 for mice. 9. Sterile gauze. 10. Lab coat. 11. Sterile gloves.
3
Methods
3.1
Anesthesia
3.1.1
Isoflurane
1. 2. 3. 4.
Place the animal in the induction chamber (see Note 4) (see Fig. 9.1B). Adjust the flowmeter on the oxygen bottle to 3 L/minute. Adjust the isoflurane vaporizer to 5%. Within approximately one minute for mice and 2 minutes for rats, the animal will become anesthetized. Initially, respiratory rate will increase and then decrease. First the animal is agitated then losses its stability. After that, it falls asleep and the animal is unreacting to noise. A simple test is to knock on the induction chamber. 5. When the animal is anesthetized, it is removed from the induction chamber and placed on the surgical area with the snout in the conical mask. Prevent heat loss until the animal recovers by using the infrared lamp (see Fig. 9.1C). 6. For maintenance, adjust the isoflurane vaporizer to 3%.
3.1.2
Xylazine–Ketamine
1. If the animal is anesthetized with a combination of xylazine and ketamine, dosage must be adapted to the weight of the animal. 2. Prepare a syringe with the proper mixture of ketamine (100 mg/kg) and xylazine (10 mg/kg). For mice, a dilution of ketamine (1:10) and xylazine (1:20) should be prepared due to small weight of these animals (20 g on average). 3. The anesthetic mixture is injected intraperitoneally with a 26G needle (see Note 5). 4. Repeat when necessary, half a dose at a time (approx. every 30 minutes). 5. Prevent heat loss until the animal recovers by using the infrared lamp.
130
3.1.3
S. Blouin et al.
Monitoring of Anesthesia Depth
The anesthesia depth must be checked to ensure that the animal is not too lightly anesthetized (i.e., that it does not experience pain or regain consciousness), or too deep (since vital functions can be compromised, leading to death). One can pinch the foot pad, tail, or ear of the animal to assess its anesthetic state. Any reaction from the animal indicates that it is too lightly anesthetized. Respiratory pattern and frequency give an indication of anesthesia depth as well. A decreasing respiratory rate is an indicator of increasing depth of anesthesia.
3.2
Animal Preparation
1. When the animal is properly anesthetized (reduced respiratory rate, lack of spontaneous movement to foot pad pinch), position it in dorsal recumbence. 2. Shave the hair/fur surrounding the scrotal sacs. 3. On the shaved area, apply a Betadine scrub using a sterile compress, followed by a final soaking with Betadine solution.
3.3
Surgery
3.3.1
Preparation of the Surgical Area, the Instruments, and the Operator
1. Aseptic surgical techniques must be used for this procedure. The operator must wear a lab coat (or surgical gown) and sterile gloves. 2. The surgical area should be located in a laboratory zone that is not heavily traveled. 3. Hard surface (table tops or equipment) must be disinfected with ethanol 70%. The contact time should be several minutes. 4. All instruments must be pre-sterilized by acceptable methods, such as steam sterilization. When performing surgery on more than one animal, effective sterilization can best be achieved by pre-sterilization of multiple sets of instruments. Dipping instruments in 70% alcohol between surgeries does not achieve sterility (>30 hours of contact time required) and is not an acceptable method. 3.3.2
The Surgical Procedure
Two incisions are made to separately reach each testis. Each incision is done with a scissor. Both sides are performed separately one after another from the incision to the suturing. 1. Check that the testes are localized within the scrotal sacs (see Note 6) (see Fig. 9.1D).
9 Orchidectomy Models of Osteoporosis
131
2. Make an incision through the skin on the ventral side of the scrotum (about 1 cm for rats and 2 mm for mice). 3. Cut the cremaster muscles with a small incision (see Fig. 9.1E). 4. Localize the testicular fat pad and pull it through the incision using a blunt forceps. 5. Pull out the cauda epididymis together with the testis, followed by the caput epididymis, the vas deferens, and the testicular blood vessels (see Fig. 9.1F). 6. Disconnect the cauda epididymis from the testis. 7. Disconnect carefully the caput epididymis embedded in fat from the testis and avoid cutting blood vessels (see Fig. 9.1G). 8. Perform a single ligature (two or three nodes and cut with 1 mm of free end suture) around the blood vessels (see Fig. 9.1H). 9. Remove the testis by cutting with scissors and make sure no bleeding occurs (see Fig. 9.1I). 10. Replace the remaining pieces of the vas deferens, the fat, and the blood vessels back in the scrotal sac with the blunt forceps. 11. Join the lips of the wound by using the blunt forceps and close the skin with Michel clips (about three to four clips per incision) by using applying forceps. 12. Repeat steps 1 to 8 for the other testis. 13. After both testes have been removed (see Note 7), return the animal to its cage (see Note 8) and monitor anesthesia recovery as described in the following.
3.4
Surgical Recovery
1. One must prevent heat loss until the animal is ambulatory. 2. With isoflurane anesthesia, the animal is ambulatory within few minutes after the end of surgery. With xylazine/ketamine anesthesia, the following clinical parameters must be monitored at a minimum of 15-minute intervals until the animal is ambulatory: — Respiratory rate — Movement — Ability to maintain sternal recumbency. It is estimated that animals will recover within 1–3 hours post-surgery (see Note 9). 3. During the first week after surgery, infection signs and clip elimination should be monitored by checking on each animal. Clips are eliminated by animals 2 or 3 weeks after surgery.
4
Notes
1. Isoflurane and xylazine/ketamine are the most commonly used anesthetic compounds. However, pentobarbital or Avertin can also be used according to each laboratory’s practice.
132
S. Blouin et al.
2. Hypothermia is the most commonly overlooked complication in rodent surgery and can result in a prolonged recovery period or death. To protect the animal from hypothermia, avoid metal surfaces during surgery. Animal could be wrapped in a towel to preserve body temperature. A supplemental heat source should be provided in the pre-, intra-, and post-operative periods (water re-circulating heating blanket or thermal packs can be used). 3. Clips are more suitable than suture material because animals can more easily damage the suture. 4. Animals waiting for surgery should be kept at a visual and olfactory distance from other animals undergoing surgery to minimize pre-operative stress. 5. Animals injected with a combination of xylazine and ketamine should be kept in a quiet environment to avoid anesthesia failure. 6. Sometimes testes are not localized within scrotal sacs and a massage of the peritoneal cavity should be performed from the top to the bottom of the abdomen until testes reach their normal position in the scrotal sacs. 7. For an experienced operator, the expected time for ORX is about 15 to 20 minutes. 8. Each animal under recovery should be placed in a separate cage to avoid injury by cage mates. 9. Animals may crawl into the corners of the recovery cage and bedding can sometimes block the airway. Acknowledgments The authors are greatly indebted to P. Legras and J. Roux (SCAHU) for their help with animal care.
References 1. Boonen, S., Vanderschueren, D., Geusens, P., et al. (1997) Age associated endocrine deficiencies as potential determinants of femoral neck (type II) osteoporotic fracture occurrence in elderly men. Int J Androl 20, 134–143. 2. Thompson, D. D., Simmons, H. A., Pirie, C. M., et al. (1995) FDA guidelines and animal models for osteoporosis. Bone 17, 125S–133S. 3. Audran, M., Chappard, D., Legrand, E., et al. (2001) Bone microarchitecture and bone fragility in men: DXA and histomorphometry in humans and in the orchidectomized rat model. Calcif Tissue Int 69, 214–217. 4. Fournier, P., Boissier, S., Filleur, S., et al. (2002) Bisphosphonates inhibit angiogenesis in vitro and testosterone-stimulated vascular regrowth in the ventral prostate in castrated rats. Cancer Res 62, 6538–6544. 5. Gunness, M., Orwoll, E. (1995) Early induction of alterations in cancellous and cortical bone histology after orchiectomy in mature rats. J Bone Miner Res 10, 1735–1744. 6. Iwamoto, J., Takeda, T., Katsumata, T., et al. (2002) Effect of etidronate on bone in orchidectomized and sciatic neurectomized adult rats. Bone 30, 360–367. 7. Iwamoto, J., Yeh, J. K., Takeda, T. (2003) Effect of vitamin K2 on cortical and cancellous bones in orchidectomized and/or sciatic neurectomized rats. J Bone Miner Res 18, 776–783. 8. Lerouxel, E., Libouban, H., Moreau, M. F., et al. (2004) Mandibular bone loss in an animal model of male osteoporosis (orchidectomized rat): a radiographic and densitometric study. Osteoporos Int 15, 814–819.
9 Orchidectomy Models of Osteoporosis
133
9. Li, M., Jee, W. S., Ke, H. Z., et al. (1995) Prostaglandin E2 administration prevents bone loss induced by orchidectomy in rats. J Bone Miner Res 10, 66–73. 10. Libouban, H., Moreau, M. F., Legrand, E., et al. (2002) Comparison of histomorphometric descriptors of bone architecture with dual-energy X-ray absorptiometry for assessing bone loss in the orchidectomized rat. Osteoporos Int 13, 422–428. 11. Moreau, M. F., Libouban, H., Legrand, E., et al. (2001) Lean, fat and bone masses are influenced by orchidectomy in the rat. A densitometric X-ray absorptiometric study. J Musculoskel Neuron Interact 1, 209–213. 12. Peng, Z., Tuukkanen, J., Zhang, H., et al. (1994) The mechanical strength of bone in different rat models of experimental osteoporosis. Bone 15, 523–532. 13. Vanderschueren, D., Vandenput, L., Boonen, S., et al. (2000) An aged rat model of partial androgen deficiency: prevention of both loss of bone and lean body mass by low-dose androgen replacement. Endocrinology 141, 1642–1647. 14. Verhas, M., Schoutens, A., L’hermite-Baleriaux, M., et al. (1986) The effect of orchidectomy on bone metabolism in aging rats. Calcif Tissue Int 39, 74–77. 15. Wink, C. S., Felts, W. J. (1980) Effects of castration on the bone structure of male rats: a model of osteoporosis. Calcif Tissue Int 32, 77–82. 16. Blouin, S., Moreau, M. F., Weiss, P., et al. (2006) Evaluation of an injectable bone substitute (betaTCP/hydroxyapatite/hydroxy-propyl-methyl-cellulose) in severely osteopenic and aged rats. J Biomed Mater Res A 78, 570–580. 17. Erben, R. G. (2001) Skeletal effects of androgen withdrawal. J Musculoskelet Neuronal Interact 1, 225–233. 18. Most, W., Van Der Wee-Pals, L., Ederveen, A., et al. (1997) Ovariectomy and orchidectomy induce a transient increase in the osteoclastogenic potential of bone marrow cells in the mouse. Bone 20, 27–30. 19. Weinstein, R. S., Jia, D., Powers, C. C., et al. (2004) The skeletal effects of glucocorticoid excess override those of orchidectomy in mice. Endocrinology 145, 1980–1987. 20. Bellido, T., Jilka, R. L., Boyce, B. F., et al. (1995) Regulation of interleukin-6, osteoclastogenesis, and bone mass by androgens. The role of the androgen receptor. J Clin Invest 95, 2886–2895. 21. Vandenput, L., Ederveen, A. G., Erben, R. G., et al. (2001) Testosterone prevents orchidectomy-induced bone loss in estrogen receptor-alpha knockout mice. Biochem Biophys Res Commun 285, 70–76. 22. Danielsen, C. C., Mosekilde, L., Andreasen, T. T. (1992) Long-term effect of orchidectomy on cortical bone from rat femur: bone mass and mechanical properties. Calcif Tissue Int 50, 169–174. 23. Erben, R. G., Eberle, J., Stahr, K., et al. (2000) Androgen deficiency induces high turnover osteopenia in aged male rats: a sequential histomorphometric study. J Bone Miner Res 15, 1085–1098. 24. Libouban, H., Moreau, M. F., Baslé, M. F., et al. (2001) Comparison insight dual X-ray absorptiometry (DXA), histomorphometry, ash weight and morphometric indices for bone evaluation in an animal model of male osteoporosis (the orchidectomized rat). Calcif Tissue Int 68, 31–37. 25. Libouban, H., Moreau, M. F., Legrand, E., et al. (2002) Bone Architecture Measured by Fractal Dimension and Connectivity Indices Is More Precociously Altered Than Mineral Content in the Orchidectomized Rat. Birkhauser Press, Cambridge, MA. 26. Rosen, H. N., Tollin, S., Balena, R., et al. (1995) Differentiating between orchidectomized rats and controls using measurements of trabecular bone density: a comparison among DXA, histomorphometry, and peripheral quantitative computerized tomography. Calcif Tissue Int 57, 35–39. 27. Wakley, G. K., Schutte, H. D., Hannon, K. S., et al. (1991) Androgen treatment prevents loss of cancellous bone in the orchidectomized rat. J Bone Miner Res 6, 325–330. 28. Vanderschueren, D., Van Herck, E., Suiker, A. M. H., et al. (1992) Bone and mineral metabolism in the adult guinea pig: long-term effects of estrogen and androgen deficiency. J Bone Miner Res 7, 1407–1415.
134
S. Blouin et al.
29. Rubinacci, A., Villa, I., Sibilia, V., et al. (1998) Responsiveness of urinary markers of bone resorption to orchiectomy and clodronate treatment in mature rats: a comparative study. Eur J Endocrinol 138, 120–127. 30. Vanderschueren, D., Jans, I., Van-Herck, E., et al. (1994) Time-related increase of biochemical markers of bone turnover in androgen-deficient male rats. Bone Miner 26, 123–131. 31. Blouin, S., Gallois, Y., Moreau, M. F., et al. (2006) Disuse and orchidectomy have additional effects on bone loss in the aged male rat. Osteoporos Int 18, 85–92.
Chapter 10
Gene Delivery by Adenoviruses Renny T. Franceschi and Chunxi Ge
Abstract Adenoviruses have a number advantages as gene delivery vectors, including ability to transduce a wide variety of non-dividing and dividing cells with high efficiency, relative ease of construction, and ability to be purified as high-titer viral stocks. These characteristics make adenoviruses particularly attractive for overexpressing specific genes in vitro and for evaluating in vivo biological activity in animal models. In this chapter, procedures will be described for adenovirus construction and virus delivery to in vivo sites for measurement of osteogenic activity. Keywords Adenovirus, gene therapy, bone, regeneration, bone morphogenetic protein, osteogenesis.
1
Introduction
Adenoviruses are among the most widely used gene therapy vectors. They have highly evolved mechanisms for efficient delivery of DNA to cells, are relatively easy to construct and amplify and, unlike retroviruses, are not dependent on cell replication for infection. Adenoviruses infect cells by binding of the viral fiber capsid protein by coxsackievirus and adenovirus receptor (CAR) and binding of the viral penton base by av integrins on the cell surface. The broad distribution of these receptors explains why adenoviruses can be used to infect such a wide range of cell types (1, 2). After infection, adenoviruses do not normally integrate into the host genome and, instead, remain in the nucleus as an episome that is gradually diluted and degraded as cells divide (3). Most gene therapy studies conducted to date used first-generation adenoviruses. These vectors have been genetically modified via deletion of the E1 gene to be replication incompetent, but still contain most of the viral genome, including genes encoding the major coat proteins. Because of this, cells infected with first-generation adenovirus vectors secrete viral proteins and elicit an immune response in the host that eventually result in clearance of transduced cells from the body (4). The in vivo From: Methods in Molecular Biology, Vol. 455: Osteoporosis: Methods and Protocols Edited by Jennifer J. Westendorf © Humana Press, Totowa, NJ
137
138
R.T. Franceschi and C. Ge
half-life of transduced cells can vary from a few days up to several weeks, depending on the degree to which the site of implantation is accessible by the immune system. For example, we measured in vivo luminescence after subcutaneous implantation of fibroblasts previously transduced with an adenovirus expressing firefly luciferase and observed decay of luciferase to baseline activity within 1 week (5). Safety issues related to the immunogenicity of virally transduced cells have severely restricted the use of first-generation adenoviruses for human applications (6). Because of these problems, second-generation adenovirus-based vectors have been developed that lack genes for most or all viral proteins (7, 8). These vectors have the advantage of being able to package up to 30 kb of foreign DNA. However, they can only be propagated in the presence of helper viruses that contain the missing viral genes necessary to form a viable capsid. In spite of their limitations, first-generation adenovirus vectors have been extremely useful for expressing specific genes in cell culture with greater efficiency and duration than traditional plasmid-based systems as well as for evaluating the effectiveness of regenerative factors in animal models. In the mineralized tissues field, adenoviruses have been extensively used to evaluate the in vivo biological activity of a wide range of molecules, including members of the BMP/TGF-b family; growth factors including FGF2, IGF-1, and PDGF; and osteogenic transcription factors such as RUNX2 (9). In addition, adenovirus expression was used to explore cooperative interactions between groups of regenerative factors such as combinations of BMPs (5, 10) or combinations of angiogenic and differentiation-promoting signals (11). Lastly, adenovirus expression was used to suppress the activity of inflammatory cytokines in rheumatoid arthritis by overexpressing decoy receptors to the interleukin 1 receptor (12). There is no doubt that adenovirus expression will continue to be the method of choice for the initial in vivo evaluation of new regenerative molecules. For this reason, this chapter focuses on the construction and use of first-generation adenovirus vectors. The approaches described in this chapter were specifically developed for bone regeneration studies in the author’s laboratory using virally expressed BMPs and osteogenic transcription factors. However, these approaches can be used to express a variety of molecules in vitro or in vivo.
2 2.1
Materials Adenovirus Production
1. HEK 293T Cells. This cell line is used for packaging and amplification of adenovirus stocks and is available from several commercial sources, such as Stratagene(La Jolla, CA) (AD-293 cells). 2. HEK 293T Cell Growth Medium: Dulbecco’s modified Eagle’s medium, 4.5 g/L glucose, 110 mg/L Na pyruvate, 4.0 mM L-glutamine, 10% heat-inactivated fetal bovine serum.
10 Gene Delivery by Adenoviruses
139
3. Phosphate Buffered Saline (PBS): 137 mM NaCl, 2.6 mM KCl, 10 mM Na2HPO4, 1.8 mM KH2PO4, adjust pH to 7.4 with HCl. 4. Virus Storage Buffer: 2.5% glycerol, 25 mM NaCl, 20 mM Tris-HCl, pH 8.0. 5. 5% Agarose in PBS: Dissolve 5 g SeaPlaque GTG agarose (BioWhittaker Molecular Applications, Cambrex Corp., East Rutherford, NJ) in 100 mL of sterile PBS and autoclave. Store in 10-mL aliquots at 4 °C in sterile 50-mL conical tubes. 6. Adeno-X Virus Purification Kits from BD Biosciences (San Jose, CA).
2.2
Ex Vivo Transduction and In Vivo Transplantation
1. Implant Lysis Buffer: 10 mM Tris-HCl, pH 7.4, 0.2 % NP40, 2.0 mM PMSF (add from 100× ethanol stock immediately before use). 2. Gelatin sponges (Gelfoam, Upohn, Kalamazoo, MI) or type I collagen hydrogels (BD Biosciences, Bedford, MA). 3. Sterile filter paper. 4. Immunodeficient mice (N: NIH-bg-nu-xid mice, Charles River Labs, Boston, MA). 5. Ketamine. 6. Xylazine. 7. 7-mm diameter trephine drill (Fine Science Tools, Foster City, CA). 8. 4-0 absorbable chromic gut (Ethicon, Inc., Somerville, NJ). 9. BLK cells. 10. Polytron Tissue Disruptor. 11. TCA: 15% and 30% in water. 12. Sigma Diagnostics kit for total calcium and phosphate.
3 3.1
Methods Vector Construction (see Note 1)
A number of kits and custom services are available for adenovirus construction (e.g., Stratagene; MP Biomedicals, Solon, OH). For this reason, specific protocols and reagents are not included here. The general strategy for making adenovirus vectors involves initial subcloning of the cDNA of interest into a shuttle vector that can either be recombined with the virus genome in E. coli as is used in the AdEasy method (13) or directly in a HEK 293T helper cell line containing CRE recombinase (14). In both methods, the use of homologous recombination greatly improves the efficiency of vector production. Commercially available shuttle vectors contain the strong cytomegalovirus promoter to drive expression of the desired gene and are also available with cDNA cloning sites 5' to an internal ribosome entry site followed by a green fluorescent protein marker gene (Stratagene). This is very useful for
140
R.T. Franceschi and C. Ge
marking virally transduced cells. The Ad5 serotype vectors most commonly used have deletions in the E1 gene necessary for DNA replication. For this reason, these viruses cannot replicate except in helper cells such as HEK 293T cells that have been stably transfected with E1 (commercially available from Stratagene as Ad293T cells).
3.2
Plaque Purification
The presence of the E1 gene in HEK293T cells introduces the possibility that replication-competent viruses can be generated via recombination during virus amplification. Although the frequency of recombination is very low, the percentage of replication-competent particles increases with each amplification. For this reason, the primary virus stock for each new adenovirus made should be used for all subsequent amplifications. To assure the homogeneity of this stock, it is recommended that it be plaque purified as follows: 1. Plate HEK293T cells at a density of 5 × 105/well of a 6-well tissue culture plate. 2. After an overnight incubation, prepare a series of 1:10 dilutions of the primary viral lysate to give final dilutions from 10−4 to 10−7. 3. Add 1 mL of each dilution to a separate well of the 6-well plate of HEK293T cells. 4. Incubate for 2 hours with gentle rocking to uniformly distribute virus. 5. Melt a 10-mL aliquot of agarose by placing in boiling water and cool to 45 °C. 6. Add 30 mL growth medium, previously equilibrated at 37 °C, and mix (final agarose concentration is 1.25%). 7. Remove medium from virally transduced cells and gently pipette 3 mL agarose solution into each well. Incubate for 1 to 2 weeks. Plaques appear as white spots. 8. Use a sterile Pasteur pipet to core out well-isolated plaques and transfer each to 250 µL growth medium in a sterile centrifuge tube. After a 24-hour incubation, virus can be used to reinfect HEK293T cells for amplification and purification.
3.3
Adenovirus Amplification and Large-Scale Purification
For large-scale adenovirus purification, we routinely use Adeno-X Virus Purification Kits from BD Biosciences that use proprietary filter technology to obtain pure virus. Alternatively, virus can be purified by CsCl ultracentrifugation. 1. Infect ten 150-cm dishes of HEK293T cells (50–90% confluent) with adenovirus stock at 5–10 pfu/cell (approx. 2 × 107 pfu virus/dish). After a 2- to 5-day incubation period, cytopathic effects of viral infection are complete (2–7 days). Cells generally are not lysed, but detach from plates and appear as floating clusters in the medium. 2. Transfer cells and medium to 50-mL conical tubes. Cells that remain attached to the culture dishes can be dislodged by gentle pipetting.
10 Gene Delivery by Adenoviruses
141
3. Centrifuge at 1,500 g for 5 minutes to obtain a cell pellet and save the supernatant. 4. Resuspend cells in 25 mL sterile of 100 mM Tris-HCl and disrupt cells with three consecutive freeze-thaw cycles. 5. Centrifuge at 1,500 g for 5 minutes to remove cell debris. Mix the cell extract with supernatant from step 3 and filter using a sterile 0.45 µ bottle top filter. 6. Virus is then purified from the filtered solution using reagents and protocols included in the Adeno-X kit. Purified virus is transferred to virus storage buffer and kept at −70 °C until use (see Note 2).
3.4
Determining Virus Titer
This method can be used to determine actual plaque forming units of a virus stock (15). It is critical that virus titer be very carefully controlled in experiments. Use of approximate methods for determining titer such as light scattering should be avoided. 1. Prepare a 96-well plate of HEK293T cells (105 cells/well) for each adenovirus preparation to be titered and let cells incubate overnight. 2. Prepare eight serial 1:10 dilutions of virus stock in growth medium containing 2% FBS. 3. For each dilution, add 100 µL to each of 10 wells. 4. After culturing for 10 days, count the number of wells showing a cytopathic effect of the virus (CPE). Make sure the lowest dilution shows 100% infection and the highest dilution has no CPE. 5. For each dilution, calculate the ratio of CPE-positive wells to total wells. Titer(T) is calculated using the following equation: T = 101+d(S – 0.5) where d = Log10 of dilution (i.e., 1 for a 1:10 dilution), S = sum of the ratios (CPE wells/total wells) for all eight dilutions. Sample calculation:
Dilution
Ratio (CPE wells/total wells)
10−1 10−2 10−3 10−4 10−5 10−6 10−7 10−8
10/10 = 1 10/10 = 1 10/10 = 1 10/10 = 1 6/10 = 0.6 2/10 = 0.2 0/10 = 0 0/10 = 0
142
R.T. Franceschi and C. Ge
In this example, 100% of wells at the 10−4 dilution have a positive CPE and 0% of wells at the 10−8 dilution have a positive CPE. Therefore, d = 1, S = 1 + 1+ 1+ 1 + 0.6 + 0.2 + 0 + 0 = 4.8 Titer = 101 + 1 (4.8 – 0.5) = 105.3 pfu/100 µl or 106.3/mL = 2 × 106 pfu/mL.
3.5
Ex Vivo and In Vivo Transduction Methods
Adenovirus vectors can either be used to transduce cells in tissue culture for subsequent in vitro analysis or in vivo implantation (ex vivo transduction) or can be directly applied to in vivo sites (in vivo transduction). The same basic procedure can be used to transduce a variety of cell types. We used this procedure with primary human and mouse dermal and gingival fibroblasts, human and mouse marrow stromal cells, and mouse and rat primary calvarial osteoblast cultures. The following mouse cell lines have also been successfully used: ST2 marrow stromal cells, MC3T3-E1 preosteoblasts, BLK fibroblasts, C3H10T1/2 mesenchymal cells, and C2C12 myoblasts.
3.5.1
Ex Vivo Transduction
1. Plate cells at the desired density in their preferred growth medium. 2. After a 24-hour attachment period, add adenovirus at the desired multiplicity of infection (MOI, which reflects the number of pfu virus added per cell) in serumfree medium (see Note 1). 3. After 4 hours, add FBS to a final concentration of 2% and grow cells for an additional 24 hours. 4. Transfer cells to the desired growth or differentiation medium and feed every 2 days for the duration of the experiment or trypsinize and implant into animals on a suitable carrier (see Note 3).
3.5.2
In Vivo Cell Implantation
We routinely use either gelatin sponges or type I collagen hydrogels as substrates for in vivo cell implantation (16) (see Note 4). 1. Twenty-four hours after virus transduction, cells are trypsinized and suspended in complete medium at a concentration of 1 × 107 cells/mL. 2. For loading cells onto gelatin sponges, the sponge (precut into 0.4 cm2 pieces) is pre-wetted with compete medium and air bubbles are removed by placing the sponge between two pieces of sterile filter paper and applying gentle pressure. Sponges are then seeded with 0.5 to 1 × 106 cells by capillary action. 3. For type I collagen hydrogels, the desired number of cells is mixed with type I collagen solution and medium to give a final collagen concentration of 2 mg/mL.
10 Gene Delivery by Adenoviruses
143
A 200-µL aliquot of the cell/collagen suspension is pipetted into a microtiter well to form a gel cylinder (see Notes 5 and 6).
3.5.2.1
In Vivo Implantation of Virally Transduced Cells
Two in vivo sites are routinely used to measure the osteogenic activity of virally transduced cells. Subcutaneous implants allow detection of osteogenic activity in the absence of endogenous bone cells and their precursors, whereas cranial implants allow detection of activity in an intramembranous bone microenvironment.
3.5.2.2
Subcutaneous Implants
Cells seeded on either Gelfoam sponges or collagen hydrogels are implanted on the backs of immunodeficient mice as follows. 1. Mice are anesthetized with ketamine (75 mg/kg) and xylazine (10 mg/kg) ip. 2. Two midline longitudinal skin incisions of approximately 1 cm in length are made on the dorsal surface of each mouse and a subcutaneous pocket is formed on each flank by blunt dissection. 3. A single implant is placed in each pocket with up to four implants per animal. 4. Incisions are closed with surgical staples.
3.5.2.3
Implanting Cells into Critical-Sized Calvarial Defects
1. Mice are anesthetized and an 0.8- to 1.0-cm incision is made in the scalp. 2. The cranial vault is exposed and a defect is created using a 7-mm diameter trephine drill, taking care not to injure the underlying dura. 3. Cells are seeded onto a precut Gelfoam sponge that is placed over the defect and secured in place with 4-0 absorbable chromic gut. In both in vivo models, mice are sacrificed after 3 to 6 weeks and bone formation is assessed using both histological and biochemical analysis as described (see Section. 3.4.3.).
3.5.3
In Vivo Transduction
Adenovirus vectors can also be directly administered to in vivo sites. In our experience, this does not generally give as reproducible results largely because of variations in efficiency of viral transduction of host cells in vivo. Both the subcutaneous and cranial models described in the preceding have been examined. Although virus can be directly injected into either site, we obtain better results when the virus is
144
R.T. Franceschi and C. Ge
immobilized on a suitable carrier such as a collagen hydrogel as described in Section 3.4.1. In this case, 1 to 2 × 109 pfu adenovirus is administered in a 10- to 20-µL volume of collagen gel.
3.5.4
Standardized System for Quantifying Osteogenic Activity of Virally Expressed Osteogenic Factors
We developed a standardized system to objectively compare the in vivo osteogenic activity of different regenerative factors using adenovirus-based expression in subcutaneous implants (17). This system uses a clonal mouse fibroblast cell line (BLK cells) derived from C57BL6 mice that is transduced ex vivo with adenoviruses expressing the desired regenerative factors. Because cells are syngeneic with this mouse line, they can be implanted into mice with a minimal immune response. To further avoid immune responses elicited by virus coat proteins, this method can also be used with immunodeficient lines as in Section 3.5.1. However, in our experience the use of C57BL6 hosts gives highly reproducible data on the in vivo osteogenic activity of regenerative factors while sparing the expense of conducting experiments in immunodeficient mice. Adenovirus expression of osteogenic factors such as BMPs in this system will produce a complete ectopic bone organ including cortical and trabecular bone and marrow. Figure 10.1 shows an example of how this assay is used to compare the osteogenic activity of adenovirus expressing BMP2, BMP7, or a BMP2/7 combination. Of interest, our studies showed that combined adenoviral expression of BMP2 and 7 had approximately four times the in vivo osteogenic activity of BMP2 or BMP7 alone (5). These results were explained by the formation of BMP2/7 heterodimers having enhanced biological activity. 1. Titered adenovirus stocks encoding each regenerative molecule to be compared are used to transduce BLK cells. Normally, BLK cells are transduced with increasing titers of virus over a range of MOI from 50 to 300. 2. To ensure that each factor is expressed at similar levels, cell layers and media (for secreted molecules) are harvested and regenerative factor levels are measured by western blotting, ELISA, or another quantitative method. 3. BLK cells expressing comparable levels of each factor are implanted on the backs of mice as described in Section 3.5.1. 4. After 4 weeks, mice are sacrificed by CO2 inhalation and implants are removed for biochemical analysis and/or histology.
3.5.5
Biochemical Analysis
When comparing a large number of samples, it is easiest to use the simplest index of new bone formation, accumulation of ectopic mineral, as a means of assessing osteogenic activity.
10 Gene Delivery by Adenoviruses
145
Fig. 10.1 In vivo osteogenic activity of AdBMP combinations. BLK fibroblasts were transduced with the indicated adenoviruses at an MOI of 100 each and implanted into C57BL/6 mice, as described in Methods. After 4 weeks, implants were harvested for determination of (a) ALP activity, (b) calcium, or (c) phosphate. Significant difference: *p < 0.05, **p < 0.01, ***p < 0.001. Histology of the implants is shown in panels D-G; (d) AdLacZ, (e) AdBMP-2, (f) AdBMP-7, or (g) AdBMP-2/7 (original magnification 4× for all sections). (Reprinted from Zhao, M., Zhao, Z., Koh, J.-T., et al. (2005). Combinatorial gene therapy for bone regeneration: cooperative interactions between adenovirus vectors expressing bone morphogenetic proteins 2, 4 and 7. J Cell Biochem 95, 1–16, with permission.)
1. Implants are homogenized in 1-mL lysis buffer with a Polytron Tissue Disruptor. Insoluble material is removed by centrifugation for 5 minutes in a microfuge. A 100-µL aliquot is removed for measurement of alkaline phosphatase (18) and the remainder of the sample is mixed with an equal volume of 30% TCA. 2. After overnight incubation at 4 °C, samples are centrifuged in a microfuge to remove precipitated protein and nucleic acids. Calcium and phosphate are solubilized by the 15% TCA. Supernatants are used to measure total calcium (Sigma Diagnostics, La Jolla, CA) and phosphate (19). Pellets can be used to measure total DNA or protein for normalization as necessary.
146
R.T. Franceschi and C. Ge
3.5.6
Histological Analysis
Immediately after removal, implants are fixed at 4 °C in 10% neutral-buffered formalin for 24 hours and then demineralized in 10% formic acid for 2 days. Samples are then rinsed and stored in 70% ethanol until paraffin embedding. After sectioning, samples are de-paraffinized, hydrated, and stained with hematoxylin and eosin.
4
Notes
1. Safety Issues. Adenovirus vectors are derived from infectious human respiratory viruses and are theoretically capable of recombining with these viruses to form infectious particles. All procedures associated with adenovirus production and use must be conducted under NIH Biosafety Level 2 conditions (see http://www. cdc.gov/od/ohs/biosfty/bmbl4/bmbl4s3.htm). Furthermore, construction of vectors expressing potentially harmful genes such as oncogenes should be done with extreme caution. Researchers are advised to consult with the Biosafety Office of their institution before commencing adenovirus work. 2. Adenoviruses are denatured by freeze-thawing. It is recommended that stocks not be thawed more than two times. 3. Large variations are observed between cell types in terms of the efficiency of adenovirus transduction and expression of recombinant proteins. When beginning studies with a new virus preparation, it is important to determine the optimal titer of virus stock for expression of the protein of interest in the specific cell line being transduced. This can be accomplished using western blotting or ELISAs. 4. In our experience with a variety of cell lines and primary cultures, adenovirus expression normally persists for 1 to 2 weeks in cell culture and for approximately 1 week after subcutaneous in vivo implantation of cells (5). However, these values vary between different cell types and are dependent on factors such as proliferation rate and stability of the expressed protein and must be determined on a case-by-case basis. 5. Virally transduced cells can also be implanted in a wide variety of synthetic tissue engineering scaffolds based on a variety of chemistries, including poly-lactate/ polyglycolate polymers, alginates and ceramics (20). 6. We routinely cast gels in microtiter wells mounted on glass slides (Lab-Tek Chamber Slides, Nalge Nunc International, Rochester, NY). After the hydrogel is formed, the gel cylinder can be easily removed from the well by removing the well from the slide and gently extruding the gel with a small spatula. Acknowledgments Work cited from the authors’ laboratory was supported by NIH grants DE13386 and DE11723.
10 Gene Delivery by Adenoviruses
147
References 1. Bergelson, J. M., Cunningham, J. A., Droguett, G., et al. (1997). Isolation of a common receptor for Coxsackie B viruses and adenoviruses 2 and 5. Science 275, 1320–1323. 2. Neumann, R., Chroboczek, J., Jacrot, B. (1988). Determination of the nucleotide sequence for the penton-base gene of human adenovirus type 5. Gene 69, 153–157. 3. Oligino, T. J., Yao, Q., Ghivizzani, S. C., et al. (2000). Vector systems for gene transfer to joints. Clin Orthop 379, S17–30. 4. Mahr, J. A., Gooding, L. R. (1999). Immune evasion by adenoviruses. Immunol Rev 168, 121–130. 5. Zhao, M., Zhao, Z., Koh, J.-T., et al. (2005). Combinatorial gene therapy for bone regeneration: cooperative interactions between adenovirus vectors expressing bone morphogenetic proteins 2, 4 and 7. J Cell Biochem 95, 1–16. 6. Yang, Y., Li, Q., Ertl, H. C., et al. (1995). Cellular and humoral immune responses to viral antigens create barriers to lung-directed gene therapy with recombinant adenoviruses. J Virol 69, 2004–2015. 7. Armentano, D., Zabner, J., Sacks, C., et al. (1997). Effect of the E4 region on the persistence of transgene expression from adenovirus vectors. J Virol 71, 2408–2416. 8. Hartigan-O’Connor, D., Amalfitano, A., Chamberlain, J. S. (1999). Improved production of gutted adenovirus in cells expressing adenovirus preterminal protein and DNA polymerase. J Virol 73, 7835–7841. 9. Franceschi, R. T. (2005). Biological approaches to bone regeneration by gene therapy. J Dent Res 84, 1093–1103. 10. Zhu, W., Rawlins, B. A., Boachie-Adjei, O., et al. (2004). Combined bone morphogenetic protein-2 and -7 gene transfer enhances osteoblastic differentiation and spine fusion in a rodent model. J Bone Miner Res 19, 2021–2032. 11. Peng, H., Wright, V., Usas, A., et al. (2002). Synergistic enhancement of bone formation and healing by stem cell-expressed VEGF and bone morphogenetic protein-4. J Clin Invest 110, 751–759. 12. Roessler, B. J., Allen, E. D., Wilson, J. M., et al. (1993). Adenoviral-mediated gene transfer to rabbit synovium in vivo. J Clin Invest 92, 1085–1092. 13. He, T. C., Zhou, S., da Costa, L. T., et al. (1998). A simplified system for generating recombinant adenoviruses. Proc Natl Acad Sci USA 95, 2509–2514. 14. Hardy, S., Kitamura, M., Harris-Stansil, T., et al. (1997). Construction of adenovirus vectors through Cre-lox recombination. J Virol 71, 1842–1849. 15. Doherty, R. (1964) Animal virus titration techniques, in (Harris, R., ed.), Techniques in Experimental Virology. Academic Press, New York. 16. Krebsbach, P. H., Gu, K., Franceschi, R. T., et al. (2000). Gene therapy-directed osteogenesis: BMP-7-transduced human fibroblasts form bone in vivo [in process citation]. Hum Gene Ther 11, 1201 –1210. 17. Franceschi, R. T., Yang, S., Rutherford, R. B., et al. (2004). Gene therapy approaches for bone regeneration. Cells Tissues Organs 176, 95–108. 18. Manolagas, S. C., Burton, D. W., Deftos, L. J. (1981). 1,25-Dihydroxyvitamin D stimulates the alkaline phosphatase activity of osteoblast-like cells. J Biol Chem 256, 7115–7117. 19. Heinonen, J. K., Lahti, R. J. (1981). A new and convenient colorimetric determination of inorganic orthophosphate and its application to the assay of inorganic pyrophosphatase. Anal Biochem 113, 313–317. 20. Alsberg, E., Hill, E. E., Mooney, D. J. (2001). Craniofacial tissue engineering. Crit Rev Oral Biol Med 12, 64–75.
Chapter 11
Lentivirus Delivery of shRNA Constructs into Osteoblasts Marc N. Wein, Dallas C. Jones, and Laurie H. Glimcher
Abstract Osteoblasts are the sole cell responsible for bone formation in vivo (1). Although genetic techniques have been extremely valuable to study the functions of certain genes in these cells in vivo, this approach is time consuming and expensive. An alternative loss-of-function approach that has been validated in many mammalian systems is shRNA-mediated gene silencing. This chapter describes methodology designed to introduce shRNA constructs into primary murine osteoblasts ex vivo in order to quickly assess the function of genes in osteoblast differentiation and extracellular matrix mineralization. Both the production of shRNA-expressing lentiviruses and the infection of calvarial osteoblasts with these lentiviruses are detailed. Keywords Osteoblasts, shRNA, lentivirus, extracellular matrix mineralization, 293T cells, NIH 3T3 cells.
1
Introduction
RNA interference (RNAi) is a rapid and powerful tool to reduce levels of target gene expression in a sequence-specific fashion. RNAi was first demonstrated to function in model systems such as A. thaliana, D. melanogaster, and C. elegans (2). RNAi also suppresses gene expression in mammalian cells, although it is technically difficult to introduce synthetic small interfering (si) RNA oligonucleotides into many cell types grown in vitro. An alternative approach is to transduce mammalian cells with viruses expressing short hairpin (sh) RNAs that will subsequently generate siRNAs within cells (3, 4). A variety of viral vectors expressing shRNA sequences have been developed. Lentiviruses are particularly powerful as they can be used to infect dividing and non-dividing cells. Additionally, lentiviruses are able to infect the vast majority of cell lines and primary cells, making lentiviral-mediated shRNA expression a powerful resource for loss-of-function studies in mammalian systems. As lentiviruses can be further engineered to express drug-resistance genes (e.g., a puromycin resistance marker, as used here), it is straightforward and convenient to infect cells From: Methods in Molecular Biology, Vol. 455: Osteoporosis: Methods and Protocols Edited by Jennifer J. Westendorf © Humana Press, Totowa, NJ
149
150
M.N. Wein et al.
with lentiviruses, add puromycin to kill all non-infected cells, and then analyze the shRNA-expressing infected cells. Previously we used lentivirus (LV)-mediated shRNA delivery to study the function of a family of E3 ubiquitin ligases in the regulation of extracellular matrix mineralization by murine calvarial-derived osteoblasts in vitro (5). This chapter provides detailed methods for the generation of lentiviruses, the propagation of murine calvarial osteoblasts in vitro, and the infection of calvarial osteoblasts with lentiviruses.
2 2.1
Materials Production of Lentiviruses in 293T Cells
1. HEK 293T cell line can be acquired from the American Type Culture Collection (ATCC, Manassas, VA). 2. DMEM-10%: 900 mL Dulbecco’s Modified Essential Medium (DMEM), 10 mL penicillin/streptomycin, 10 mL L-glutamine, 10 mL 1 M Hepes pH 8.0, 100 mL fetal bovine serum (FBS, Hyclone, Ogden, UT). 3. Effectene Transfection Reagent (Qiagen, 301427, Valencia, CA). 4. 0.45 µM filters (Costar, 8112, Corning, NY).
2.2
Determination of Viral Titer in NIH3T3 Cells
1. The murine fibroblastic NIH3T3 cell line can be acquired from the ATCC. 2. Polybrene (hexadimethrine bromide) dissolved in sterile water at a stock concentration of 8 mg/mL and kept at 4 °C. 3. Puromycin dihydrochloride dissolved in sterile water at 2 mg/mL and kept at −20 °C.
2.3
Isolation of Calvarial Osteoblasts
1. At least six neonatal mice 2–5 days old (see Note 1 for more details). 2. 70% Ethanol to sterilize mice and surgical instruments. 3. Ice cold Dulbecco phosphate-buffered saline (DPBS) dispensed into sterile sixwell tissue culture dish (Falcon, 35-3046, Franklin Lakes, NJ). 4. Digestion solution: 50 mL Alpha Minimum Essential Medium (AMEM), 50 mg collagenase type II, 100 mg Dispase II (Roche, Germany), 500 µL penicillin/ streptomycin. Filter sterilize prior to use. 5. Osteoblast differentiation medium (OBDM-10%): 500 mL AMEM, 5 mL penicillin/streptomycin, 5 mL non-essential amino acids, 5 mL L-glutamine, 5 mL
11 Lentivirus Delivery of shRNA Constructs
151
1 M Hepes pH 8.0, 50 mL fetal bovine serum, 25 mg L-ascorbic acid, 540 mg Glycerol 2 phosphate disodium salt hydrate. Filter sterilize prior to use. 6. 70-µm cell strainer (BD Falcon, 352350, Bedford, MA).
3
Methods
Production of shRNA-expressing lentiviruses is first accomplished by transfecting viral components into packaging 293T cells. LV-containing cell supernatant is collected and filtered. Rough estimates of viral titer are then determined by infecting murine NIH3T3 cells with viruses. Finally, we describe how to isolate calvarial-derived osteoblasts from neonatal mice, and how to infect these cells with shRNA-expressing LV in vitro.
3.1
Production of Lentiviruses in 293T Cells
1. HEK 293T cells are passaged in DMEM-10% at 37 °C until ready to use. The day before transfection, cells are harvested by trypsinization, counted, and plated at a density of 2 × 106 cells in 10 mL DMEM-10% in a 10-cm dish. Plate out one dish for each virus to be produced. 2. The next day, the cells should appear 25–35% confluent and are ready for transfection with Effectene. At this time, allow DMEM-10% medium to equilibrate to room temperature. In a tissue culture hood, prepare one sterile 1.5 mL Eppendorf tube for each transfection (see Note 1). In each tube, add 300 µL buffer EC, then add 2 µg LV cDNA, 2 µg psPAX2 (gag/rev), and 200 ng pMD2. G (VSV-G) plasmid DNA to buffer EC (see Note 2). Add 34 µL enhancer solution to each tube, vortex gently for 1–3 seconds, and incubate tubes at room temperature for 5 minutes. 3. Add 80 µL Effectene solution to each tube, vortex gently for 1–3 seconds to mix, and incubate tubes at room temperature for 5 minutes. 4. Aspirate medium from adherent HEK 293T cells and gently replace with 10 mL fresh DMEM-10%. Take care to slowly pipet this fresh medium along the edge of the plate so as to not disturb the cells. 5. To each transfection tube, add 500 µL DMEM-10% and mix by gently vortexing as above. Add this mix drop wise to each plate of HEK 293T cells and gently swirl plate to mix. Place cells back in 37 °C incubator overnight. At this point, cells should be treated with standard BSL2 safety precautions. 6. The next day, aspirate off medium into bleach-containing solution. Add 10 mL fresh DMEM-10% (pre-equilibrated to room temperature) to each plate and place plates back at 37 °C. 7. The next day, harvest HEK 293T cell LV-containing supernatant. Pipette off 10 mL medium from each plate and filter through 0.45 µm filter into 10 sterile 1.5 mL Eppendorf tubes. Label LV-containing Eppendorf tubes and freeze at −80 °C until further use.
152
3.2
M.N. Wein et al.
Determination of Viral Titer in NIH3T3 Cells
1. NIH3T3 cells are passaged in DMEM-10% at 37 °C until ready for use. The day before titer determination, collect cells by trypsinization, count, and plate at a density of 3 × 104 cells in 1 mL DMEM-10% in each well of a 24-well plate. Plate enough cells for four wells per LV to be tested. 2. The next day, cells should appear 25–35% confluent and are ready for LV infection. Thaw LV stocks on ice. For each LV to be tested, prepare four sterile 1.5-mL Eppendorf tubes with 100 µl DMEM-10% plus 2 µg polybrene. To these tubes, add 0-, 1-, 10-, or 100-µL LV stock. Add this LV/medium/polybrene solution to each well in a drop wise fashion and place the cells back at 37 °C. 3. The next day, aspirate medium and replace with 1 mL DMEM-10% plus 2 µg/ mL puromycin. As described, puromycin at this concentration will effectively kill all uninfected cells. 4. Forty-eight hours later, there should be obvious signs of cell death in uninfected, puromycin-treated wells (see Note 3). At this point, collect cells by trypsinization and determine cell viability by trypan blue staining. The ratio of trypan blue-negative/trypan blue positive cells should indicate the relative cell survival in each well. Alternatively, cell viability can be easily determined with flow cytometry using the forward scatter/side scatter profiles (see Note 4). 5. Wells infected with increasing amounts of LV should show decreased numbers of dead (trypan blue–positive) cells. If, for example, cells infected with 10 µL LV show 50% survival under these conditions, the LV titer of this stock is calculated as follows: 10 µL LV infected approximately 60,000 NIH3T3 cells (assuming doubling time of 24 hours); therefore, 1 µL LV can infect 6 × 103 cells, and 1 mL LV can infect 6 × 106 cells. Determine viral titer for each desired LV, as it is important to subsequently infect osteoblasts with comparable numbers of infectious particles for each gene to be silenced (see Note 5).
3.3
Isolation of Murine Calvarial Osteoblasts
1. At least six P2–P5 neonatal mice are sacrificed by decapitation (see Note 6). Dissection tools are kept sterile by submersion in 70% ethanol, and mice are sprayed with 70% ethanol prior to decapitation. 2. Remove skin over the skull, expose calvarial bones, cut coronally just anterior to suture overlying cerebellum and just posterior to suture at rostral end of cortex. Cut sagittally along the lateral edge of skull (near the ear) to free the overlying calvarial bones from underlying brain tissue. Place calvarial bones in sterile DPBS kept cold on ice in a six-well tissue culture dish. 3. After removing calvarial bones from each mouse, carefully dissect away soft tissue and sutures from bones to avoid contaminating fibroblasts.
11 Lentivirus Delivery of shRNA Constructs
153
4. Place dissected calvariae in 4-mL digestion solution at the bottom of a 50 mL Falcon tube. Incubate at 37 °C with shaking 275 rpm for 10 minutes. Remove digestion solution by pipetting, wash bones with 5 mL serum-free AMEM, add 4 mL fresh digestion solution and incubate again for 10 minutes at 37 °C on a rotating shaker set at 275 rpm. Wash again with 5 mL serum-free AMEM. Discard solution from these first two digestions and washes (see Note 7). 5. Add fresh 4 mL digestion solution and incubate for 10 minutes at 37 °C with shaking at 275 rpm. Aspirate off digestion solution and place in 14 mL OBDM10% (see Note 8) in a 50-mL Falcon tube. Repeat three times such that four digestion fractions are collected and pooled in 14 mL OBDM for a total of 30 mL solution. 6. Filter 30 mL solution collected in step 5 through a 70-µm cell strainer. Spin down cells at 365 g for 5 minutes at 4 °C (cell pellet will be barely visible). 7. Wash cells with 10 mL OBDM-10% and spin down again. For every two mice initially sacrificed, resuspend pellet in 3 mL OBDM-10% and plate in six-well tissue culture dish. For example, if eight mice were initially sacrificed, resuspend pellet in 12 mL OBDM-10% and plate in four wells of a six-well plate. Place cells at 37 °C. 8. Three days later, cells should appear in adherent colonies. Aspirate medium and replace with 3 mL fresh OBDM-10% pre-equilibrated to room temperature. 9. Two to three days later (5–6 days after dissection) cells should be confluent in dishes. Collect by trypsinization and count. Replate cells at a density of 3 × 104/ mL in OBDM-10% for LV infection. For example, if cells are to be infected in six-well plates (recommended for subsequent analysis of gene expression by RNA isolation), 9 × 104 cells are plated in 3 mL OBDM-10% per well.
3.4
Infection of Calvarial Osteoblasts with shRNA-Expressing Lentiviruses
1. The day after osteoblasts are plated at 3 × 104 cells/mL (step 9), cells should be 50–60% confluent. Aspirate off medium and replace with OBDM-10% containing 2 µg/mL polybrene. 2. Dilute LVs into 100 µl OBDM-10% in sterile 1.5 mL Eppendorf tubes such that the MOI is approximately 1–1.5 (see Note 9). 3. Add LV/medium mix drop wise to cells and mix by gently swirling. Place cells back at 37 °C. It is crucial to leave at least one well uninfected with LV (“mock infected”) to assess the efficiency of subsequent puromycin-induced selection. 4. The next day, add puromycin to cells to reach final concentration of 2 µg/mL. Dilute puromycin into OBDM-10% prior to adding to cells. Mix by gently swirling and place back at 37 °C. 5. Two to four days later, the mock-infected cells should be completely killed by puromycin. At this point, the medium may be changed to OBDM-10% without
154
M.N. Wein et al.
puromycin. The medium should be subsequently changed every 2–3 days to promote robust extracellular matrix production and mineralization. 6. To confirm shRNA-mediated silencing (knock-down), we typically harvest cells 5–7 days after puromycin-induced killing of mock-infected cells. Knock-down is interrogated by protein immunoblotting when possible or real-time quantitative PCR on cDNA prepared from control or gene-specific LV-shRNA cultures using gene-specific primers. 7. To investigate the effects of shRNA-mediated gene silencing on osteoblast function, we typically perform standard alkaline phosphatase stains on day 10–14 cultures to assess osteoblast differentiation and von Kossa stains on day 14–21 cultures to assess matrix mineralization.
4
Notes
1. It is essential to prepare control LV for subsequent experiments. For this purpose, we typically use LV expressing shRNA against either GFP or luciferase, genes normally not expressed in osteoblasts. 2. The empty pLKO.1 lentiviral vector as well as pLKO.1 vectors expressing specific human or mouse shRNA can be purchased from Open Biosystems (Huntsville, AL) or Sigma-Aldrich. The psPAX2 (gag/rev) and pMD2.G (VSV-G) plasmid necessary for generation of viral particles in HEK 293T cells can be acquired from Addgene.org. 3. In the uninfected, puromycin-treated wells, dying cells appear rounded and become non-adherent. In contrast, cells infected with lentivirus expressing a puromycin resistance gene adhere to the tissue culture plate and maintain their fibroblastic morphology 4. For analysis of cell death by flow cytometry, it is necessary to have a control group of cells that have not been treated with puromycin. The amount of viable cells in this population should be >90%, as determined by trypan blue exclusion on the hemocytometer. This group of untreated cells can then be utilized to generate the “live-cell” gate on the forward scatter/side scatterplot. Run each sample on the flow cytometer and collect the same number of total events with each sample. The percent of live cells in each sample can then be determined by using the “live-cell” gate that was generated with the control cells. Utilizing this method should result in no viable cells being present in the uninfected, puromycintreated group. In addition, cells can be stained with propidium-iodide prior to analysis. Viable cells can then be measured utilizing the FL2 channel on the flow cytometer, which detects dying cells that are propidium-iodide positive. 5. Given that the differentiation and function of osteoblasts can be profoundly influenced by cell density, it is important to infect cells with lentiviruses that are of a similar titer. This is crucial to ensure that each experimental condition will have a comparable cell density following puromycin selection.
11 Lentivirus Delivery of shRNA Constructs
155
6. A relatively inexpensive and dependable source of neonatal mice must be determined. We have successfully ordered late gestational pregnant female mice from Taconic of Balb/c and C57BL/6 genetic backgrounds. Pregnant females are shipped late in gestation, and typically give birth to litters of more than six pups within 2–4 days of arrival. Other investigators (6) have found that calvarial osteoblasts are best cultured from P2-6 mice. It is important to consistently use mice of the same age to reduce inter-experimental variability. 7. The first two fractions contain mainly fibroblasts and should be discarded such that the final pooled digestions contain a relatively homogenous population of osteoblasts. 8. All collagenase/Dispase digestions must be performed in serum-free medium because serum contains anti-proteases that will inactivate these enzymes. Digestions are then stopped by collecting digestion solutions and placing them in serum-containing medium kept on ice. 9. Determine titer of virus as described in Section 3.2. For example, if virus is determined to contain 6 × 103 infectious units per 1 µL and it is desired to infect 9 × 104 cells, use 15 µL virus to achieve an MOI of approximately 1. We have found that infecting at an MOI >10 leads to significant cytotoxicity and an MOI <0.1 leads to inefficient infection. Acknowledgments The authors thank Nir Hacohen (Broad Institute) for help with the generation of lentiviruses and Tony Aliprantis (Harvard University) for helpful discussions.
References 1. Karsenty, G., Wagner, E. F. (2002) Reaching a genetic and molecular understanding of skeletal development. Dev Cell 2, 389–406. 2. Fire, A., Xu, S., Montgomery, MK., et al. (1998) Potent and specific genetic interference by double-stranded RNA in Caenorhabditis elegans. Nature 391, 806–811. 3. Brummelkamp, T.R., Bernards, R., Agami, R. (2002) A system for stable expression of short interfering RNAs in mammalian cells. Science 296, 550–553. 4. Moffat, J., Grueneberg, DA., Yang, X., et al. (2006) A lentiviral RNAi library for human and mouse genes applied to an arrayed viral high-content screen. Cell 124, 1283–1298. 5. Jones, DC., Wein, MN., Oukka, M., et al. (2006) Regulation of adult bone mass by the zinc finger adapter protein Schnurri-3. Science 312, 1223–1227. 6. Ducy, P., Zhang, R., Georffroy, V., et al. (1997) Osf2/Cbfa1: a transcriptional activator of osteoblast differentiation. Cell 89, 747–754.
Chapter 12
Gene Delivery by Retroviruses Valerie Deregowski and Ernesto Canalis
Abstract Gene delivery by retroviruses is an easy and safe tool to stably over express a gene of interest and determine its role in a cell model. The gene of interest is cloned into the multiple cloning site of a retroviral vector that also contains a packaging signal and an antibiotic resistance marker for selection. Packaging cell lines, transfected with a retroviral vector containing the gene of interest or with a control vector, produce retroviral RNA packaged into infectious, replication incompetent virus. This is a consequence of the stable integration of the viral genes necessary for particle formation and replication into the packaging cell genome. Retroviruscontaining conditioned medium from packaging cells is used to transduce target cells. After antibiotic selection, the target cells are tested for overexpression of the gene of interest and its biological effects. Keywords Osteoblast, retrovirus, gene expression, transduction.
1
Introduction
Retroviral gene transfer is a technique to introduce, efficiently, stable genetic material into the genome of any dividing cell type. The retroviral gene transfer technology is based on the coordinated design of packaging cell lines and retroviral expression vectors. Packaging cell lines package recombinant retroviral RNAs into infectious, replication incompetent particles (1, 2). In a packaging cell line, the viral gag, pol, and env genes, which are necessary for particle formation and replication, are stably integrated into the cell genome. Retroviral expression vectors provide the packaging signal Ψ+, transcription and processing elements, and the gene of interest. Transfection of the retroviral vector into a packaging cell line produces hightiter, replication incompetent virus. The separate introduction and integration of the structural genes minimizes the likelihood of producing replication competent virus due to recombination events occurring during cell proliferation (3, 4). Packaging cell lines are transfected with a retroviral expression vector containing a packaging signal, the gene of interest, and a selectable marker. The retroviral RNA is packaged From: Methods in Molecular Biology, Vol. 455: Osteoporosis: Methods and Protocols Edited by Jennifer J. Westendorf © Humana Press, Totowa, NJ
157
158
V. Deregowski and E. Canalis
into infectious virus within 24–72 hours of transfection. Viral particles cannot replicate within target cells and spread because the viral structural genes are absent. The viral env gene, expressed by the packaging cell line, encodes the envelope protein, which determines the range of infectivity of the packaged virus. Viral envelopes are classified according to the receptor used to enter host cells. We have used different packaging cell lines, including RetroPack PT67, AmphoPack-293 (both from Clontech, Palo Alto, CA), and Phoenix (American Type Culture Collection, ATCC, Manassas, VA) cell lines. Virus packaged by these cell lines can infect a broad range of mammalian cell types, including those from mouse, rat, and humans. Several retroviral expression vectors are commercially available (Clontech, Palo Alto, CA) and contain the extended retroviral packaging signal ψ+ and gene elements derived from Moloney murine leukemia virus (MoMuLV) and Moloney murine sarcoma virus (MoMuSV). They also contain antibiotic resistance markers for selection in eukaryotic cells. Often we have used pLPCX retroviral vectors (Clontech), which express the puromycin resistance gene controlled by promoter/enhancer sequences of the 5' viral long terminal repeat (LTR). The gene of interest is cloned into the multiple cloning site immediately downstream of the cytomegalovirus (CMV) immediate early promoter. Retroviral vectors can be used to express a gene of interest, and also to express hairpin small interfering (si)RNA (Imgenex, San Diego, CA) and used for RNA interference and gene knock-down experiments. Gene delivery by retroviruses is an efficient and relatively safe tool to overexpress or knock-down a gene of interest and study its function in a cell model (5–8). Guidelines from the National Institutes of Health require that retroviral production and transduction be performed in a Biosafety Level 2 facility (http://www.ors.od. nih.gov). A contained room should be used for this purpose, with dedicated laminar flow hood, CO2 incubator, centrifuge, sink, and plasticware. Protective laboratory coats, facial protection, and double gloves should be worn.
2
Materials
1. RetroPack PT67, AmphoPack-293 (Clontech), and Phoenix packaging cells (ATCC) grown in a humidified 5% CO2 incubator at 37 °C in Dulbecco’s Modified Eagle’s Medium (DMEM) supplemented with 10% fetal bovine serum (FBS, Atlanta Biologicals, Norcross, GA). 2. Target cells, such as ST-2 cells, cloned stromal cells isolated from bone marrow of BC8 mice, MC3T3-E1 osteoblastic cells derived from mouse calvariae or other cells grown in a humidified 5% CO2 incubator at 37 °C in Minimum Essential Medium (α-MEM), supplemented with 10% FBS (9). 3. Gene of interest cloned into a retroviral vector, such as pLPCX, pLNCX2, or pLHCX (Clontech). Before the ATG initiation codon of the coding sequence of the gene of interest, a Kozak consensus ribosome binding site (GCC GCC ACC) is inserted to ensure translation in mammalian cells (10). It is useful to clone a tag
12 Gene Delivery by Retroviruses
4. 5. 6. 7. 8. 9.
159
(HA, Flag, or c-myc) at the amino or carboxy terminal end of the coding sequence of the gene of interest. The tag can be used to detect the protein of interest by western blot analysis or to study protein–protein interactions by immunoprecipitation. The sequence of the tag must be cloned in frame with the sequence of the gene of interest. The design of the construct in the multiple cloning site will be as follows for an N-terminal Tag: Kozak sequence - ATG - tag - coding sequence of the gene of interest - STOP codon; and for a C-terminal Tag: Kozak sequence - ATG - coding sequence of the gene of interest - tag - STOP codon. The sequence of the construct should be checked to ensure it contains no mutations. TransFast Transfection Reagent (Promega, Madison, WI). Puromycin (or other selection antibiotics) is dissolved in culture medium with no serum at 1 mg/mL for puromycin, and stored in aliquots at −20 °C. Polybrene is dissolved in phosphate buffered saline (PBS) at 800 µg/mL (100× stock), and aliquots are stored at 4 °C. 10-mL syringes. Acrodisc Syringe Filter 0.45 µm HT Tuffryn membrane (VWR International, Westchester, PA). Freezing medium, 90% serum, and 10% dimethylsulfoxide (DMSO).
3
Methods
The following procedure must be performed in a Biosafety Level 2 facility. As a positive control for transfection, a control retroviral vector containing a marker gene coding for green fluorescent protein (GFP) can be used. Before starting the transfection into the packaging cells, three 100-mm plates of target cells are prepared to ensure 70% confluence 24 hours following plating (see Note 1). Two target cell 100-mm plates will be transduced with retrovirus-containing conditioned medium and one will be used as a control for antibiotic selection.
3.1
Transfection of Packaging Cells
Control vector and vector containing the gene of interest are transfected into two different 100-mm plates of 70% confluent packaging cells. At this confluence, transfection efficiency is high and packaging cells can survive the stress of transfection, giving the highest titer of virus possible (see Note 2). 1. In two different sterile tubes, add 15 µg of each retroviral vector DNA to 6 mL of pre-warmed DMEM without serum and mix. 2. Add 45 µL of TransFast reagent to each tube and mix. 3. Incubate TransFast reagent/DNA mixture for 15 minutes at room temperature. 4. Remove medium from Phoenix packaging cells and add TransFast reagent/DNA mixture to each 100-mm plate, and return the cells to a 5% CO2 incubator at
160
V. Deregowski and E. Canalis
37 °C for 1 hour. During the incubation, warm 6 mL of DMEM supplemented with 10% FBS to 37 °C. 5. Add 6 mL of pre-warmed DMEM with 10% FBS to packaging cells and incubate for 24 hours.
3.2
Transductions of Host Cells
Retrovirus-containing conditioned medium is harvested from packaging cells to transduce target cells. 1. Pipette 5 mL of retrovirus-containing conditioned medium from packaging cells and filter through a 0.45-µm low-protein-binding Acrodisc to remove cell debris, and add 100 µL of polybrene to enhance infection (see Note 3). Polybrene is a polycation that reduces charge repulsion between the virus and the cellular membrane. 2. Add 5 mL of pre-warmed DMEM with 10% FBS to packaging cells and return cells to the CO2 incubator (see Note 4). 3. Remove medium from each target cell 100 mm plate and add 5 mL of prewarmed α-MEM with 10% FBS to the plates to be exposed to the retroviralcontaining conditioned medium and add 10 mL of α-MEM with 10% FBS to a plate to be used as a control for antibiotic selection. 4. Add 5 mL of filtered retroviral-containing conditioned medium to transduce the target cells. 5. Return cells to the CO2 incubator and culture for 72 hours (see Note 5). 6. Remove conditioned medium from target cells and from the control plate 72 hours after the transduction and add 10 mL of pre-warmed α-MEM containing 10% FBS and 2 µg/mL puromycin (or alternate selection antibiotic) to carry out the antibiotic selection for 48 hours. 7. Determine the confluence of surviving cells under light microscopy. If 80% of the host cells survive the antibiotic selection, split them at a 1:2 ratio; otherwise, re-feed cells with pre-warmed α-MEM containing 10% FBS and 2 mg/ml puromycin. Maintain transduced cells in the presence of antibiotic selection medium until they start to grow at their usual rate, and the un-transduced control cells are dead, typically 1 week. When cellular death stops, remove puromycin from medium, and continue the culture in a Biosafety Level 1 facility. Freeze multiple aliquots of early passage cells in freezing medium under liquid nitrogen, and grow cells to determine the levels of gene expression.
3.3
Gene Expression
Multiple techniques are available to determine the expression of mRNA levels of the gene of interest. Northern blot analysis allows the comparison of levels of RNA
12 Gene Delivery by Retroviruses
161
expression in control vector cells and retroviral vector cells overexpressing the gene of interest. Real time reverse transcription polymerase chain reaction (RT-PCR) is an alternative that is highly sensitive and does not require the use of radioactivity. Because of these reasons we favor its use in the laboratory. To determine protein expression level in control vector and retroviral expressing cells, western blot analysis or enzyme-linked immunosorbent assay (ELISA) can be applied using an antibody detecting the protein of interest or its tag.
4
Notes
1. It is important to use 70% confluent target cells to allow the viral integration complex to enter into the nuclei of actively dividing cells. 2. Packaging cells should not be allowed to reach confluence, since this will reduce the efficiency of the transfection. 3. Supernatant can be frozen at −80 °C for later infections; however, viral titer declines after each freeze-thaw cycle. Phoenix cells might lose the expression of gag-pol or env genes over time. Re-select Phoenix cells in 1 µg/mL diphtheria toxin to increase env expression or 500 µg/mL hygromycin to increase gag-pol gene expression. Cells can be treated with the two drugs simultaneously for up to 2 weeks. 4. The transfected packaging cells can be cultured in DMEM containing puromycin to select cells that stably express the integrated vector. Virus produced by stably transfected packaging cells can infect target cells and transmit the gene of interest. Virus production is higher in newly transfected cells. 5. A second round of infection 24 hours after the first can be done. This allows cells to rest between each infection and ensures that cellular receptors are not occupied by the viral envelope. Acknowledgments This work was supported by grants from the National Institutes of Health (NIH) AR 21707, DK 45227 and DK 42424 and by a fellowship award from the Arthritis Foundation.
References 1. Mann, R., Mulligan, R. C., Baltimore, D. (1983) Construction of a retrovirus packaging mutant and its use to produce helper-free defective retrovirus. Cell 33, 153–159. 2. Miller, A. D., Buttimore, C. (1986) Redesign of retrovirus packaging cell lines to avoid recombination leading to helper virus production. Mol Cell Biol 6, 2895–2902. 3. Miller, A. D., Chen, F. (1996) Retrovirus packaging cells based on 10A1 murine leukemia virus for production of vectors that use multiple receptors for cell entry. J Virol 70, 5564–5571. 4. Morgenstern, J. P., Land, H. (1990) Advanced mammalian gene transfer: high titre retroviral vectors with multiple drug selection markers and a complementary helper-free packaging cell line. Nucleic Acids Res 18, 3587–3596.
162
V. Deregowski and E. Canalis
5. Deregowski, V., Gazzerro, E., Priest, L., et al. (2006) Notch 1 overexpression inhibits osteoblastogenesis by suppressing Wnt/beta-catenin but not bone morphogenetic protein signaling. J Biol Chem 281, 6203–6210. 6. Deregowski, V., Gazzerro, E., Priest, L., et al. (2006) Role of the RAM domain and ankyrin repeats on notch signaling and activity in cells of osteoblastic lineage. J Bone Miner Res 21, 1317–1326. 7. Gazzerro, E., Deregowski, V., Vaira, S., et al. (2005) Overexpression of twisted gastrulation inhibits bone morphogenetic protein action and prevents osteoblast cell differentiation in vitro. Endocrinology 146, 3875–3882. 8. Ross, D. A., Rao, P. K., Kadesch, T. (2004) Dual roles for the Notch target gene Hes-1 in the differentiation of 3T3-L1 preadipocytes. Mol Cell Biol 24, 3505–3513. 9. Rawadi, G., Vayssiere, B., Dunn, F., et al. (2003) BMP-2 controls alkaline phosphatase expression and osteoblast mineralization by a Wnt autocrine loop. J Bone Miner Res 18, 1842–1853. 10. Kozak, M. (1987) At least six nucleotides preceding the AUG initiator codon enhance translation in mammalian cells. J Mol Biol 196, 947–950.
Chapter 13
Chromatin Immunoprecipitation Assays: Application of ChIP-on-Chip for Defining Dynamic Transcriptional Mechanisms in Bone Cells Margaretha van der Deen, Mohammad Q. Hassan, Jitesh Pratap, Nadiya M. Teplyuk, Daniel W. Young, Amjad Javed, Sayyed K. Zaidi, Jane B. Lian, Martin Montecino, Janet L. Stein, Gary S. Stein, and Andre J. van Wijnen Abstract Normal cell growth and differentiation of bone cells requires the sequential expression of cell type specific genes to permit lineage specification and development of cellular phenotypes. Transcriptional activation and repression of distinct sets of genes support the anabolic functions of osteoblasts and the catabolic properties of osteoclasts. Furthermore, metastasis of tumors to the bone environment is controlled by transcriptional mechanisms. Insights into the transcriptional regulation of genes in bone cells may provide a conceptual basis for improved therapeutic approaches to treat bone fractures, genetic osteopathologies, and/or cancer metastases to bone. Chromatin immunoprecipitation (ChIP) is a powerful technique to establish in vivo binding of transcription factors to the promoters of genes that are either activated or repressed in bone cells. Combining ChIP with genomic microarray analysis, colloquially referred to as “ChIP-onchip,” has become a valuable method for analysis of endogenous protein/DNA interactions. This technique permits assessment of chromosomal binding sites for transcription factors or the location of histone modifications at a genomic scale. This chapter discusses protocols for performing chromatin immunoprecipitation experiments, with a focus on ChIP-on-chip analysis. The information presented is based on the authors’ experience with defining interactions of Runt-related (RUNX) transcription factors with bone-related genes within the context of the native nucleosomal organization of intact osteoblastic cells. Keywords ChIP-on-chip, chromatin, cross-linking, histone, immunoprecipitation, microarray, osteosarcoma, Runx, shearing, transcriptional regulation.
1
Introduction
Transcriptional control of gene expression is fundamental to the normal phenotypic development of both osteoblasts and osteoclasts that together control bone growth and mineral homeostasis. In addition, deregulation of transcriptional control in tumor cells may support bone metastasis. Chromatin immunoprecipitation
From: Methods in Molecular Biology, Vol. 455: Osteoporosis: Methods and Protocols Edited by Jennifer J. Westendorf © Humana Press, Totowa, NJ
165
166
M. van der Deen et al.
(ChIP) analysis has emerged in recent years as a powerful technique that reveals the fundamental transcriptional mechanisms regulating biological processes in the bone environment. ChIP assays reveal the location of histone modifications or the recruitment of transcription factors and their co-factors to the promoters of actively transcribed or suppressed genes in their in vivo genomic context. DNA retrieved by chromatin immunoprecipitations can be analyzed using PCR or combined with genomic microarrays (i.e., ChIP-on-chip analyses) (Fig. 13.1) (1). Our studies focus on Runt-related (RUNX) proteins, which are essential transcriptional regulators of genes that control cell growth and differentiation. We and others have shown that Runx2 plays a pivotal role during osteoblast growth and differentiation, as well as in hypertrophic chondrocyte maturation and in tumorigenesis (2–7). We have successfully applied chromatin immunoprecipitations in osteoblasts in previous studies (5–10). Here we present ChIP protocols based on our studies with osteoblastic cells expressing Runx2 to define target genes that are directly controlled by this osteogenic master regulator.
Crosslinking DNA
Shearing DNA/protein complexes
Immunoprecipitation with antibodies
Analysis with PCR
Amplify DNA
Cy3/Cy5 labeling
Hybridization to array Fig. 13.1 Schematic diagram of the ChIP protocol. Cells are fixed with formaldehyde and chromatin is sheared by sonication. Precipitation of protein–DNA complexes occurs with specific antibodies. Enriched DNA is purified for quantitative analyses. In conventional ChIP, DNA is analyzed with PCR (either real-time or gel analysis). For high throughput detection, ChIP DNA is amplified and labeled with Cy3/Cy5 and hybridized to a microarray.
13 Chromatin Immunoprecipitation Assays
2 2.1
167
Materials Cells and Cross-Linking of Protein–DNA Interactions
1. Saos-2 cells are grown in McCoy’s 5A medium supplemented with 15% fetal bovine serum (FBS; Atlanta Biologicals, Lawrenceville, GA). 2. Formaldehyde (37% v/v solution, methanol stabilized; high purity, cat. #501602) (Mallinckrodt Baker, Inc., Phillipsburg, NJ) for fixation (cross-linking) of chromatin–DNA. 3. 1.25 M Glycine: (10× stock; USB Corp., Cleveland, OH) for quenching excess formaldehyde. 4. Phosphate buffered saline (PBS). 5. Complete protease inhibitor cocktail (Hoffmann-La Roche, Inc., Nutley, NJ).
2.2
Sonication to Fragment DNA
1. SDS lysis buffer: 1% SDS, 10 mM EDTA, 50 mM Tris-HCl, pH 8.1. 2. Complete protease inhibitor (Roche). Protease inhibitor cocktail in DMSO (200×) (Upstate Biotechnology, Inc. Millipore, Billerica, MA). 3. Sonicator, Sonic Dismembrator Model 550 (Fisher Scientific International, Pittsburgh, PA), 1.6-mm tip. Set to 10% max power. Sonicate in pulses of 10 seconds each on wet ice.
2.3
Immunoprecipitation of Cross-Linked Protein–DNA Interactions
1. ChIP dilution buffer: 0.01% SDS, 1.1% Triton X-100, 1.2 mM EDTA, 16.7 mM Tris-HCl, pH 8.1, 167 mM NaCl. 2. Complete protease inhibitor cocktail (Hoffmann-La Roche, Inc.). 3. Primary antibody of choice, suitable for ChIP analysis (if commercially available, select ChIP grade). Runx2 antibody: rabbit polyclonal M70 (PEBP2αA, Santa Cruz Biotechnology Inc., Santa Cruz, CA). Negative control: IgG of same species and isotype as specific antibody. Positive control: e.g., anti-RNApolymerase II (clone 8WG16, Covance, San Diego, CA). 4. Protein A/G agarose beads (Santa Cruz Biotechnology, Inc.). For micro-array purposes, magnetic Dynabeads are recommended (Invitrogen Dynal AS, Oslo, Norway). The beads are supplied as a slurry of beads and buffer. Beads must be carefully handled and thoroughly mixed before each use to deliver equivalent amounts of beads in each sample. Use a tip with a wide opening. Some investigators use siliconized tips to minimize sticking of beads.
168
M. van der Deen et al.
5. 200 mM phenylmethylsulfonyl fluoride (PMSF; Sigma/Aldrich) in isopropanol (or ethanol). Stock-solution is stable at room temperature, but diluted PMSF is rapidly destabilized in aqueous solution (half-life ~30–60 minutes). 6. 25 mM proteasome inhibitor MG132 in dimethylsulfoxide (DMSO) (CalBiochem/EMD Biosciences, San Diego, CA). 7. Low salt immune-complex wash buffer: 0.1% SDS, 1% Triton X-100, 2 mM EDTA, 20 mM Tris-HCl, pH 8.1, 150 mM NaCl. 8. High salt immune complex wash buffer: 0.1% SDS, 1% Triton X-100, 2 mM EDTA, 20 mM Tris-HCl, pH 8.1, 500 mM NaCl. 9. LiCl immune complex wash buffer: 0.25 M LiCl, 1% IGEPAL-CA630, 1% deoxycholic acid (sodium salt), 1 mM EDTA, 10 mM Tris-HCl, pH 8.1. 10. TE buffer: 10 mM Tris-HCl, 1 mM EDTA, pH 8.0.
2.4
Elution of Protein–DNA Complexes
1. Elution buffer (freshly prepared): 10 µL SDS (20%), 20 µL NaHCO3 (1 M), and 170 µL sterile, distilled water.
2.5 1. 2. 3. 4. 5. 6. 7. 8. 9. 10.
Reversal of Cross-Linking, DNA Purification, and PCR 5 M NaCl. 10 µg/mL RNAse A in sterile water. 0.5 M EDTA, pH 8.0. 1 M Tris-HCl, pH 6.5. Proteinase K: 10 mg/mL in 50 mM Tris-HCl, pH 8.0, 10 mM CaCl2. DNA Purification: Phenol/chloroform/isoamyl alcohol (25:24:1). 20 mg/mL Glycogen (molecular biology grade, Roche). Power SYBR Green PCR Master mix (Applied Biosystems, Foster City, CA). ABI PRISM thermocycler (Applied Biosystems, CA). ChIP primers for genes of interest and negative controls (i.e., non-binding DNA regions).
2.6 Amplification of Immunoprecipitated DNA for ChIP-on-Chip 1. Determine DNA concentration with UV absorption (260 nm) by spectroscopy. The NanoDrop spectrophotometer (ND-100) (NanoDrop Technologies, Wilmington, DE) is useful for small sample volumes. 2. Whole Genome Amplification (WGA) kit (WGA-2, Sigma-Aldrich, St. Louis, MI). 3. QIAquick PCR purification kit (Qiagen).
13 Chromatin Immunoprecipitation Assays
169
4. Fluorophore labeling kit (e.g., Cy-3/Cy-5 labeling kit; NimbleGen, Inc., Madison, WI). 5. Microarray equipment (see Note 7).
3
Methods
Although ChIPs can be performed without fixatives to detect histone modifications near specific genes, detection of transcription factor binding to DNA in vivo is optimized by treatment with appropriate cross-linking agents. The most commonly used fixative to cross-link protein–DNA interactions within cells is formaldehyde. Upon cell harvest, protein–DNA complexes are fragmented by sonication to obtain DNA fragments in a size range from 200 to 1,000 bp. Transcription factor DNA binding elements are isolated by immunoprecipitation (IP) with specific antibodies against the cognate proteins. Appropriate negative and positive control antibodies are included to establish specificity of the precipitation reactions. Complexes are recovered using agarose beads coupled to protein A and/or G depending on antibody isotype. DNA is released by reversing the cross-links and is subsequently purified. DNA recovered from chromatin immunoprecipitations can be analyzed by different methods. Detailed protocols for quantitation of ChIP DNA by ethidium bromide staining (8, 9), radioactive detection (10), or quantitative PCR (5, 6) have been presented in the indicated references. This chapter emphasizes ChIP-on-chip analysis, because it represents the most state-of-the-art methodology and permits comprehensive genome-wide analysis of gene regulatory events. Amplification of DNA is necessary to generate sufficient DNA for hybridization to either whole genome or promoter-specific DNA arrays. Whole genome amplification (WGA) or ligation-mediated PCR (LM-PCR) are frequently used to increase the amount of DNA (11). The two key parameters for ChIP assays are the quality of the precipitating antibody and the efficiency of DNA shearing. These parameters are even more critical for experiments in which amplified ChIP DNA is hybridized to DNA microarrays. ChIP-on-chip experiments also demand highly efficient DNA amplification and relatively pure immunoprecipitated DNA that is not contaminated with non-specific genomic DNA.
3.1
Cross-Linking of Protein–DNA Interactions in Live Cells
1. Grow cells until 80–90% confluency in 100-mm culture dishes containing 10 mL of growth medium. Use one dish for cell counting. For 10 immunoprecipitations, we use approximately 2 × 107 Saos-2 cells (see Note 1). 2. To cross-link protein to DNA, add 270 µL of 37% formaldehyde to the medium and swirl the dish gently. Maintain cells at room temperature (RT) for 10 minutes (see Note 2).
170
M. van der Deen et al.
3. Add 1 mL of 10× glycine (1.25 M) to quench unreacted formaldehyde. Swirl the dish and incubate for 5 minutes at RT. 4. Aspirate the medium and place dish on ice. Wash two times with 10 mL of ice-cold PBS. 5. Remove PBS (as much as possible) and add 1 mL of PBS with freshly added protease inhibitors. Collect cells by scraping with a rubber policeman and transfer cells into a centrifuge tube. 6. Spin at 700 g for 5 minutes at 4 °C. The cell pellet can be stored at −80 °C at this step upon snap freezing the pellet in liquid nitrogen. 7. Resuspend cell pellet in SDS lysis buffer containing freshly added protease inhibitors (complete protease inhibitor cocktail; optionally supplemented with 1 mM PMSF and 25 µM MG132). Add 1 mL of SDS lysis buffer for 2 × 107 cells (see Note 3). 8. Aliquot suspension in 350 µL (or between 300–400 µL) in 1.5 mL microfuge tubes.
3.2
Sonication of Chromatin Containing Bound Transcription Factors
1. Sonicate the cell lysate on wet ice. The extent of DNA shearing should be carefully optimized (see Note 4). 2. Spin at 10,000 g for ~10–30 minutes at 4 °C and collect the supernatant. Optional: remove 5 µL and keep at −20 °C to check the shearing of DNA on an agarose gel (see Note 4). At this step chromatin can be frozen at −80 °C. 3. Aliquot 100 µL per microcentrifuge tube. Keep samples on ice for addition of antibodies for immunoprecipitations (see Section 3.3.).
3.3
Immunoprecipitation of Cross-Linked Protein–DNA Interactions
1. Per 100 µL of sheared DNA, add 900 µL of dilution buffer with freshly added protease inhibitors (Complete protease inhibitor cocktail; optionally supplemented with 1 mM PMSF and 25 µM MG132). 2. Add 60 µL of protein A/G agarose beads for each ChIP sample for pre-clearing. Gently mix protein A/G beads by inversion each time before removing an aliquot. 3. Incubate for 1 hour at 4 °C with rotation. 4. Pellet protein A/G agarose beads by centrifugation for 1 minute at 3,000–5,000 g. 5. Remove an Input sample: 10 µL (1%) of the supernatant and keep at 4 °C until the first step of Section 3.4. The remainder of the aliquot is stored at −20 °C. 6. Transfer supernatant into a fresh 1.5 mL microcentrifuge tube.
13 Chromatin Immunoprecipitation Assays
171
7. Add antibodies to each 1 mL solution of sheared DNA. The ChIP quality of the antibody is very important and the concentration should be empirically determined by titration (antibody range from 1–10 µg). For Runx2, we use the rabbit polyclonal antibody M70. As negative control, use IgG control of the same species and isotype as the antibody of interest. As positive controls, acetylated histones H3 and H4, or anti-polymerase II are often used as markers for open chromatin in actively transcribed genes (see Note 5). 8. Incubate 2–16 hours at 4 °C with continuous rotation. The optimal duration will depend on the quality of the antibody. 9. Add 60 µL of protein A/G agarose beads for each ChIP to collect antibody– chromatin–DNA complexes. 10. Incubate for 1 hour at 4 °C with continuous rotation. 11. Pellet protein A/G agarose beads by centrifugation for 1 minute at 3,000– 5,000 g and remove the supernatant. 12. Wash the pellet of protein A/G agarose beads for 3–5 minutes consecutively in the following solutions. Each of the following steps is performed at 4 °C, but non-specific binding (and thus background signals) may be reduced by carrying out washes at 20 °C. After each step, beads are recovered by centrifugation at 3,000–5,000 g for 1 minute, and the supernatant is carefully removed and discarded after every wash. — Low salt wash buffer (1×) — High salt wash buffer (1×) — LiCl wash buffer (1×) — TE buffer (2×)
3.4
Elution of Protein–DNA Complexes
1. For each tube (including input), make 200 µL of elution buffer. 2. For input tubes, add 200 µL of elution buffer and set aside at RT until step 6. 3. Add 100 µL of elution buffer to the protein A/G agarose pellet and mix gently by tapping. 4. Incubate at RT for 15 minutes. 5. Pellet protein A/G agarose beads by centrifugation for 1 minute at 3,000–5,000 g and collect the supernatant into a new centrifuge tube. 6. Repeat steps 3, 4, and 5 and combine supernatants (200 µL total).
3.5
Reversal of Cross-Linking, DNA Purification, and PCR
1. Add 8 µL of 5 M NaCl to all tubes (IPs and inputs) and incubate at 65 °C for 4–5 hours or overnight to reverse the DNA–protein cross-links.
172
M. van der Deen et al. 0.50 0.45 0.40
ng/ul DNA
0.35 0.30 0.25 0.20 0.15 0.10 0.05 0.00
Runx2
IgG
Fig. 13.2 Representative result of a ChIP assay. Saos-2 cells were subjected to a ChIP assay with Runx2 antibody (6 µg) or IgG as a negative control. Binding of Runx2 to the Runx2 promoter is eightfold enriched compared to the IgG control. Analysis was performed using real-time qPCR by using a standard curve of genomic DNA.
2. Add 1 µL of RNAse A (i.e., 10 µg in 200-µL sample) to all tubes and incubate for 30 minutes at 37 °C. 3. Add 4 µL of 0.5 M EDTA, 8 µL of 1 M Tris-HCl, and 1 µL of proteinase K and incubate at 45 °C for 1–2 hours. 4. Purify DNA using standard phenol/chloroform extraction and purify DNA. Resuspend in 25–50 µL of elution buffer. 5. Samples can be amplified by PCR and quantified using ethidium bromide staining, radioactivity, or real-time quantitative PCR using with well-optimized primer sets (see Note 6). To analyze genome-wide occupancy, ChIP DNA samples are examined using microarrays. However, because DNA yields are typically low, samples subjected to microarrays will need to be amplified using whole genome amplification (see Section 3.6.). See Fig. 13.2 for a representative result for Runx2 binding to its cognate motifs in the promoter of the Runx2 gene, which is controlled by its own gene product.
3.6
Amplify ChIP DNA for ChIP-on-Chip (see Note 7)
1. Determine DNA concentration with NanoDrop spectrophotometer. 2. Use 10 ng of DNA and adjust volume with sterile nuclease free water to 10 µL.
13 Chromatin Immunoprecipitation Assays
173
3. Use WGA kit to amplify DNA (see Note 8). Omit steps 2, 3, and 4 in the provided protocol when using sheared DNA (fragmentation step is not necessary). 4. Purify DNA with PCR purification kit. 5. Measure DNA concentrations with a NanoDrop spectrophotometer. DNA purity is assessed using a DNA spectrum from 230 to 320 nm, which permits determination A260/280 and A260/230 ratios. DNA concentrations should be at least 100 ng/µL, but ideally above 250 ng/µL. The optimal absorbance ratios are A260/280 > 1.7 and A260/230 > 1.5. 6. Check the quality of DNA after WGA amplification on a 1.5% agarose gel (~5–10% of the final reaction). The DNA size should range from 200 to 1,000 bp with an average of size of ~400 bp. 7. Quantify samples with qPCR to detect signals for positive and negative control genes (see Note 9) as in step 5 of Section 3.5. 8. DNA can now be labeled with Cy-5 dye (ChIP DNA) and Cy-3 dye (control input DNA). Check the efficiency of DNA labeling (see Note 10). 9. Hybridize DNA to promoter array (see Note 11). 10. Analyze array with software to identify “peaks” above a certain threshold (see Note 12). A typical peak search for a Runx2 target gene is depicted in Fig. 13.3.
Binding to probes
Peak search
Gene position Start site
Tiled promoter region
Fig. 13.3 ChIP-on-Chip analysis. Saos-2 cells were subjected to a ChIP assay and enrichment for Runx2 on specific genes was analyzed with a promoter-microarray. Analysis of the data identified a peak at the Smad 5 gene promoter. The binding of Runx2 was also established independently using real-time qPCR (data not shown).
174
4
M. van der Deen et al.
Notes
1. A typical cell lysate contains the equivalent of 2 × 107 nuclei per mL SDS of lysis buffer. Grow one extra 100-mm dish for cell counting. Different cell numbers (from 5 × 106 to 5 × 107) should be used to optimize ChIP DNA yields for each cell type. 2. Formaldehyde cross-linking stabilizes protein/DNA interactions and ensures co-precipitations of the DNA and proteins of interest. The formaldehyde should be of the highest possible grade as break-down products (e.g., formic acid) may interfere with the cross-linking efficiency and/or immunoprecipitation. For ChIP with antibodies against abundant proteins bound to DNA (e.g., histones) it is not always necessary to crosslink protein/DNA interactions within cells. 3. After lysis, the percentage of released nuclei can be checked under microscope by taking a 10-µL sample of the lysate. Typically >80% of the cells should exhibit release of chromatin, although this is difficult to determine with confidence. If necessary, trypan blue staining can be used to visualize residual intact nuclei. 4. DNA fragmentation by sonication should be carefully optimized. The sheared DNA fragments should ideally be in the range of 200 to 1,000 base pairs and should not contain large genomic fragments. Some cell lines are difficult to lyse by mechanical force, and it is necessary to test different conditions. Typically, we apply a range of two to 12 pulses of 10 seconds each. Always keep the lysates on wet ice because shearing introduces energy, and sample heating can potentially result in protein denaturation. Some investigators rapidly cool samples using a dry ice-ethanol bath for a few second to minimize protein denaturation. Optional: Remove a sample that is equivalent to 1 × 105 cells prior to shearing to permit monitoring of the efficiency of DNA sonication using electrophoresis in a 1.5% agarose gel. Remove 5 µL of the sonicated chromatin from each condition to a fresh tube. To each 5 µL of sample (unsheared and sheared), add 90 µL of ddH2O and 4 µL of 5 M NaCl (final 0.2 M NaCl). Incubate at least 4 hours (up to 16 hours) at 65 °C to reverse the DNA-protein cross-links. Add 1 µL of RNase A and incubate for 30 minutes at 37 °C. Add 2 µL of 0.5 M EDTA (pH8.0), 4 µL of 1 M Tris-HCl (pH 7), and 1 µL of Proteinase K (final 0.1 mg/mL) and incubate at 45 °C for 1–2 hours. Examine a sample of 10 or 20 µL (the equivalent of, respectively, 1 or 2 × 104 cells) in parallel with a 100-bp DNA size marker by 1.5% agarose gel electrophoresis. 5. Typical controls include, for example, ChIP samples incubated with normal immunoglobulin (IgG) of the same species of the ChIP antibody or omitting the ChIP antibody. Note that IgG controls from different sources can be very different in purity and quality. If available, a blocking peptide would be useful to show specificity of antibody binding. 6. Because of the expense of ChIP-on-chip experiments, it is important to pre-evaluate the ChIP DNAs using appropriate positive and negative control primers. Positive control primers typically amplify fragments spanning a known binding site, although the size of sheared DNA (~400 bp) allows more flexibility in the design.
13 Chromatin Immunoprecipitation Assays
7.
8.
9.
10.
11.
175
Negative control primers include 3' UTR sequences that are far enough apart from known binding sites to avoid inadvertent signals. Alternatively, promoter regions from entirely unrelated genes that are devoid of sites for specific transcription factors of interest can be selected as negative controls. Optimize ChIP primers by performing a standard curve with genomic DNA or input DNA. Efficiency should ideally be >95%. The dissociation peak should also be checked. There are several microarray platforms available for ChIP-on-chip analysis, including microarrays from NimbleGen Systems, Inc. (Madison, WI), Agilent (Santa Clara, CA), and Affymetrix (Santa Clara, CA). Our protocol focuses on the platform developed by NimbleGen because of its ease of use. NimbleGen provides comprehensive services for DNA labeling, DNA hybridization, and data analysis of ChIP-on-chip experiments. Other platforms require array equipment that is usually available in institutional genomics facilities. The concentration of ChIP-enriched DNA is usually low and needs to be amplified to yield enough DNA for hybridization to a gene microarray. One method for amplification is ligation-mediated PCR (LM-PCR) and another method is whole genome amplification (11). Both methods are based on ligation of linkers and linear amplification. The shearing of DNA should be very well optimized and samples should not contain residual large genomic DNA. A properly sheared DNA sample will permit linear amplification of ChIP-enriched DNA, reduces background and avoids skewing of data due to disproportional amplification of DNA. Ideally, ratios of ChIP-enriched and non-enriched samples (i.e., “input” samples) are the same before and after amplification and samples need to be validated with standard quantitative PCR for positive and negative control genes. Note that DNA concentrations should be measured before amplification. DNA concentrations may need to be diluted for qPCR to obtain concentrations optimal for amplification. The amplified ChIP DNA and amplified total genomic DNA are labeled with the fluorophores Cy-5 and Cy-3, respectively, and are co-hybridized to arrays. Labeling efficiency can be measured using a NanoDrop spectrophotometer. The type of fluorophore may differ between array platforms and manufacturers. Companies provide tiling arrays that are specially designed for ChIP-on-chip. ChIP arrays are easier to design compared to conventional microarrays used for determining mRNA levels. One advantage of tiled arrays over amplicon-based probe arrays is that a positive signal can span over several probes, called the “neighbor effect.” DNA fragments of 15–70 nucleotides are designed to span or “tile” across genomic target regions. The tile density determines the resolution of the array and can be restricted to known promoter regions of genes but also whole genome arrays are available. With respect to DNA shearing, shorter fragments require more closely tiled probes than longer fragments so that real positive signals are not missed (12). Some companies also provide customized arrays.
176
M. van der Deen et al.
12. All statistical methods are based on “peak searches,” but computational approaches differ and several publications propose newer and better methods to identify positive “true” signals. A simple method of analysis is to identify peaks above a certain user-defined threshold. One method called joint binding deconvolution (JBD) uses DNA sequence information (protein-DNA binding sites) to identify peaks (13). Another approach to distinguish specific signal from noise is called algorithm for capturing microarray enrichment (ACME) (12). ACME uses a sliding window approach and it is based on the assumption that the enriched signal must be positive in one direction (one-tailed) and that the signal must be represented by multiple probes (the neighbor effect). Although there are many different ways to compute ChIP-on-chip data, independent multiple replicates are necessary to identify true signals. A typical peak search for a Runx2 target gene is depicted in Fig. 13.3.
References 1. Ren, B., Robert, F., Wyrick, J. J., et al. (2000). Genome-wide location and function of DNA binding proteins. Science 290, 2306–2309. 2. Pratap, J., Javed, A., Languino, L. R., et al. (2005). The Runx2 osteogenic transcription factor regulates matrix metalloproteinase 9 in bone metastatic cancer cells and controls cell invasion. Mol Cell Biol 25, 8581–8591. 3. Javed, A., Barnes, G. L., Pratap, J., et al. (2005). Impaired intranuclear trafficking of Runx2 (AML3/CBFA1) transcription factors in breast cancer cells inhibits osteolysis in vivo. Proc Natl Acad Sci USA 102, 1454–1459. 4. Blyth, K., Cameron, E. R., Neil, J. C. (2005). The runx genes: gain or loss of function in cancer. Nat Rev Cancer 5, 376–387. 5. Young, D. W., Hassan, M. Q., Yang, X.-Q., et al. (2007). Mitotic retention of gene expression patterns by the cell fate determining transcription factor Runx2. Proc Natl Acad Sci USA 104, 3189–3194. 6. Young, D. W., Hassan, M. Q., Pratap, J., et al. (2007). Mitotic occupancy and lineage-specific transcriptional control of rRNA genes by Runx2. Nature 445, 442–446. 7. Galindo, M., Pratap, J., Young, D. W., et al. (2005). The bone-specific expression of RUNX2 oscillates during the cell cycle to support a G1 related anti-proliferative function in osteoblasts. J Biol Chem 280, 20274–20285. 8. Shen, J., Hovhannisyan, H., Lian, J. B., et al. (2003). Transcriptional induction of the osteocalcin gene during osteoblast differentiation involves acetylation of histones H3 and H4. Mol Endocrinol 17, 743–756. 9. Shen, J., Montecino, M. A., Lian, J. B., et al. (2002). Histone acetylation in vivo at the osteocalcin locus is functionally linked to vitamin D dependent, bone tissue-specific transcription. J Biol Chem 277, 20284–20292. 10. Hassan, M. Q., Javed, A., Morasso, M. I., et al. (2004). Dlx3 transcriptional regulation of osteoblast differentiation: temporal recruitment of Msx2, Dlx3, and Dlx5 homeodomain proteins to chromatin of the osteocalcin gene. Mol Cell Biol 24, 9248–9261. 11. O’Geen, H., Nicolet, C. M., Blahnik, K., et al. (2006). Comparison of sample preparation methods for ChIP-chip assays. Biotechniques 41, 577–580. 12. Scacheri, P. C., Crawford, G. E., Davis, S. (2006). Statistics for ChIP-chip and DNase hypersensitivity experiments on NimbleGen arrays. Methods Enzymol 411, 270–282. 13. Qi, Y., Rolfe, A., MacIsaac, K. D., et al. (2006). High-resolution computational models of genome binding events. Nat Biotechnol 24, 963–970.
Chapter 14
Identification of Transcription Factor Target Genes by ChIP Display Artem Barski, Steven Pregizer, and Baruch Frenkel
Abstract Transcription factors play pivotal roles in the control of cell growth and differentiation in health and disease. In the post-genomic era, it has become possible to locate the regions occupied by transcription factors throughout the genome, leading to better understanding of their mechanism of action and the genes that they regulate. All methods for transcription factor location analysis utilize chromatin immunoprecipitation (ChIP). Although ChIP was initially used to test whether a protein binds to a candidate promoter in living cells, newly developed methods allow the unbiased identification of novel targets of transcription factors. This chapter describes ChIP Display, an affordable method for transcription factor location analysis. Despite being relatively low throughput compared with alternative methods such as ChIP-chip and ChIP-SAGE, ChIP Display provides even small molecular biology laboratories with the opportunity to discover novel targets of any transcription factor, for which high-quality antibodies are available. Keywords Transcription factor location analysis, transcription factor target genes, protein-DNA interaction in vivo, chromatin immunoprecipitation, genomics.
1
Introduction
Transcription factors play pivotal roles in cell biology, animal physiology, and disease processes. The ongoing quest to understand their function includes the search for genes whose expression they regulate. Target genes for transcription factors can be identified by expression studies, although these are fraught with problems (see Note 1). A complementary group of approaches is based on the physical interaction of transcription factors with cis-acting regulatory elements, and the assumption that such elements are frequently located close to the genes that they regulate. Most of these approaches begin with chromatin immunoprecipitation (ChIP), in which DNA fragments bound by a transcription factor of interest in living cells are immunoprecipitated with antibodies against that protein. Once identified, these fragments can be mapped to the genome, and nearby genes can be tested for their regulation by the transcription factor of interest. From: Methods in Molecular Biology, Vol. 455: Osteoporosis: Methods and Protocols Edited by Jennifer J. Westendorf © Humana Press, Totowa, NJ
177
178
A. Barski et al.
The ChIP-based methods for transcription factor target discovery are challenged by the overwhelming excess of non-specifically precipitated fragments (Fig. 14.1A). Each of these methods takes a different path to identifying those fragments that were specifically immunoprecipitated. Of these methods, ChIP Display (CD) (1) is one of the least high throughput but offers relative simplicity and can be performed in a typical molecular biology laboratory without sophisticated equipment or complicated statistical analyses (see Note 2). To overcome the high background problem, CD effectively concentrates fragments representing each target while scattering the remaining DNA. Targets are concentrated via restriction digestion (see Fig. 14.1C): All fragments representing a certain target now have the same size, allowing one to resolve them as a single band on a gel. Scattering of the non-specifically precipitated fragments is achieved by dividing the total pool of restriction fragments into families based on the identity of nucleotides at the ends of such fragments (see Fig. 14.1E–G). Because all restriction fragments representing each given target have the same nucleotides at the ends, they remain in the same family and the signal is not eroded. In contrast, the other fragments, mostly background, are scattered into many families. The CD protocol is rather simple (see Fig. 14.1). Following ChIP, immunoprecipitated DNA is dephosphorylated by shrimp alkaline phosphatase (SAP) to prevent ligation of linkers to DNA ends generated by sonication (see Fig. 14.1B). SAP is then heat-inactivated and the DNA is digested with a restriction enzyme. We use AvaII, whose recognition sequence, GGWCC (W = A or T), can be expected approximately every 500 bp in a random sequence. This is followed by ligationmediated PCR (LM-PCR) using various primer combinations (see Fig. 14.1E). Up to 36 combinations of eight nested primers are employed to amplify fragments belonging to one family at a time (see Fig. 1F,G). Each such primer contains either A or T at the +3 position of the AvaII site and one nested nucleotide, A, T, G, or C, at the 3′ end (Fig. 14.1F). Amplified fragments from two to three independent immunoprecipitates and two to three control precipitates are resolved by PAGE. Bands enriched in the ChIP lanes are considered candidate targets and are excised from the gel for further characterization. Using CD, our lab has identified dozens of targets for both Runx2 and the androgen receptor (1–4).
2
Materials
2.1 1. 2. 3. 4. 5. 6.
Digestion and Linker Ligation
Short linker oligo: 5′-TTCGCGGCCGCAC-3′. Long A linker oligo: 5′-GACGTGCGGCCGCGAA-3′. Long T linker oligo: 5′-GTCGTGCGGCCGCGAA-3′. 10× React1 and 10× React2 buffers (Invitrogen, Carlsbad, CA). Shrimp alkaline phosphatase (SAP), 1 U/µL. AvaII, 10 U/µL.
a
#1 #1
#2
1Mb
b
#2
SAP 5kb
5kb
c
AvaII 5kb
5kb
d
Ligate linkers 5kb
5kb
PCR
e
1Mb
1Mb
1Mb G
f
B g +3 T
A AvaII G G T C C N +1+2 +4+5+6 +6 +3 C
GGWCG G C CCWGC
CGWCC G GCWGG C +6
T +3
A A A T A G A C T A T T T G V T C X GCA T GC +6 A T G AAA A T T T T +3
Fig. 14.1 Principles of ChIP Display (CD). See color figure on the disc accompanying this book. a. Precipitated DNA fragments (green, specific; black, non-specific) are aligned with the genome. Graph (red) describes the representation of each nucleotide in the immunoprecipitate. The 1-Mb area shown contains two hypothetical targets, #1 and #2. b. Magnification of the two regions of interest from panel (a). Precipitated DNA fragments are treated with shrimp alkaline phosphatase (SAP) to prevent linker ligation to DNA ends generated during sonication. Crossed circles depict dephosphorylated end. (c) DNA is digested with AvaII. d. Linkers are ligated to the ends of AvaII fragments. Also shown are the nested primers. Note that different primers will amplify targets #1 and #2 (see color plate). e. PCR products amplified in three reactions: Left: Target #1 is amplified in a reaction with a single primer (family X in panel G). Middle: Target #2 is amplified with two different primers (family V in panel G); note that target #1 is amplified here again; Right: Most of the PCR reactions will amplify neither target #1 nor target #2; however, they will amplify targets from other loci. f. Linkers and nested primers. Positions +1 to +6 are defined at the top. Nucleotides at positions +3 and +6 (highlighted) are used to segregate AvaII fragments into families. Linkers (blue) contain a 1:1 A:T mixture (W) at position +3 and a C at position +1 to destroy the AvaII sites. In addition, this panel shows one of the eight PCR nested primers, the one with T at position +3 and C at position +6, which would amplify the family indicated by X in panel G. g. Thirty-six families of fragments, each amplified using one or two of eight nested primers. White squares correspond to a single primer and shaded squares correspond to two different primers. (Reproduced from Barski, A., Frenkel, B. (2004) Nucleic Acids Res 32, e104, with permission from Oxford University Press.)
180
A. Barski et al.
7. T4 DNA ligase, 400 U/µL. 8. 10× T4 ligase buffer. 9. Enzymatic reaction cleanup kit (Qiagen, Valencia, CA).
2.2 1. 2. 3. 4. 5.
PCR Amplification
Taq DNA polymerase, 5 U/µL. 10× PCR Reaction Buffer, –MgCl2. MgCl2, 50 mM. dNTP mix (Invitrogen), 10 mM each. Up to eight PCR primers: 5′-CGGCCGCACGWCCN-3′, where W is A or T and N is A, G, C, or T (see Note 2).
2.3
PAGE
1. 2. 3. 4.
40% polyacrylamide solution [29:1 (w/w) acrylamide/bisacrylamide]. 10% (w/v) ammonium persulfate (APS) in water (store ≤1 month at 4 °C). TEMED solution (N,N,N′,N′-tetramethylethylenediamine; store at 4 °C). 10× TBE Electrophoresis Buffer: pH 8.0: 890 mM Tris, 890 mM boric acid, and 20 mM EDTA. 5. 5× Sample Loading Buffer: 10% Ficoll 400, 50 mM disodium EDTA, pH 8.0, 0.5% sodium dodecyl sulfate, 0.25% bromphenol blue, 0.25% xylene cyanol. 6. Gel loading pipet tips. 7. 1-kb DNA ladder.
2.4 1. 2. 3. 4. 5. 6. 7. 8.
Target Identification
Spin-X Centrifuge Tube Filters (Corning, Corning, NY). HaeIII, HinfI, MspI, RsaI, 10 U/µL each. Standard/High Melt Agarose. Gel extraction kit (Qiagen). BigDye Terminator 3.1 Cycle Sequencing Kit (Applied Biosystems, Foster City, CA). AutoSeq G-50 Microcentrifuge Columns (Amersham). Hi-Di™ Formamide (Applied Biosystems). For in-house sequencing: ABI PRISM 3100 Genetic Analyzer (Applied Biosystems) or an alternative system.
14 Target Genes by ChIP Display
3
181
Methods
CD begins after ChIP has been completed with the cells and antibodies of interest. To identify Runx2 targets in osteoblasts, we usually ChIP half the chromatin collected from two confluent 100-mm plates of MC3T3-E1 cells with 5 µg of Runx2-specific antibodies (Santa Cruz, cat. no. 10758X). The other half is immunoprecipitated with 5 µg of non-specific IgG to control for non-specific IP. Each of the specific and non-specific ChIPs is done in duplicate or triplicate to control for experimental variation. Thus, we typically initiate a CD experiment for a given experimental condition with two or three ChIP samples and two or three mock-ChIP samples. Each of the samples is cleaned with an Enzymatic Reaction Cleanup Kit (Qiagen) and eluted in 50 µL H2O, which is sufficient for about 100 separate CD reactions. Each CD reaction is then displayed in one lane of an acrylamide gel, and will typically yield up to two targets. Although this chapter does not describe the ChIP procedure, two critical points must be emphasized: (1) No salmon sperm DNA should be used for ChIP intended for CD. Instead, substitute with bacterial tRNA (see Note 3); and (2) high-quality ChIPs are essential for the success of CD. Before a CD protocol is initiated, one must ensure that the starting ChIP material is enriched for at least one, and preferably more, known targets. We recommend the use of real-time PCR to determine the enrichment levels, with both internal (non-targets) and external (IgG) controls (see Note 4). If the enrichment levels are satisfactory (∼10-fold or better), then this material may be used for CD.
3.1
Digestion and Linker Ligation
1. Resuspend each of the two long linker oligos (“A” and “T”) and the short linker oligo in dH2O at a 100 µM concentration. 2. Mix 15 µL of the short linker oligo with 15 µL of the long “A” linker oligo. Add 10 µL of 10× React2 buffer and 60 µL of dH2O. Likewise, mix 15 µL of the short linker oligo with 15 µL of the long “T” linker oligo and add 10 µL of 10× React2 buffer and 60 µL of dH2O. Mix each well. 3. Anneal oligos by placing samples in a beaker of boiling water for 1 minute. Turn off heat and allow water to return to room temperature without removing samples. After cooling, annealed linkers may be stored at −20 °C. 4. Mix 10 µL of each ChIP sample with 2 µL of 10× React1 buffer, 1 µL of SAP, and 7 µL of H2O. Incubate at 37 °C for 30 minutes. Heat denature for 15 minutes at 65 °C (see Note 5). 5. Add 1 µL (10 units) of AvaII to each sample. Incubate at 37 °C for 30 minutes (see Note 6). 6. Add 2 µL each of both annealed linkers, 3 µL of 10× T4 ligase buffer and 1 µL T4 DNA ligase (400 U). Incubate overnight at 16 °C.
182
A. Barski et al.
7. Add 0.5 µL of AvaII and 0.5 µL of T4 Ligase to each sample. Incubate 1 hour at room temperature (see Note 7). Purify the samples on QuickSpin columns using Enzymatic Reaction Clean-Up Buffer. Elute with 50 µL dH2O.
3.2
PCR Amplification
1. Make a PCR master mix containing 1× PCR Buffer, 1.5 mM MgCl2, 0.2 mM each dNTP, 0.025 U/µL Taq DNA Polymerase (see Note 8), and 0.5 µM of one or two of the eight possible primers (see Note 9). 2. Add 18 µL of the PCR master mix to 2 µL of each replicate of each columnpurified ChIP sample (see Notes 10 and 11). 3. Mix by pipetting up and down, and then place tubes in a thermocycler at room temperature (see Note 12). 4. Amplify the three samples representing each IP using two or three annealing temperatures between 68–70 °C. Specifically, amplify each set of three samples using the following program: (1) 72 °C for 30 seconds; (2) 95 °C for 5 minutes; (3) 95 °C for 1 minute; (4) 68 °C [or 69 or 70 °C] for 1 minute; (5) 72 °C for 1 minute; (6) go to 3, 45 times; (7) 72 °C for 5 minutes; (8) End (see Figs. 14.2A and 14.3, and Note 13).
a
ChIP 2 3 68° 69° 68° 69° 68° 69°
M bp 517 506
396 344
i.
mock 2 3 68° 69° 68° 69° 68° 69°
b M
1
1
M
M
i. U Ha Hi Ms M
ii. U Ha Hi Ms
bp 517, 506 396 344
ii.
298 220 201 154 134
298
Fig. 14.2 Target display and identification. a. DNA samples from three independent ChIPs and three mock ChIPs were. DNA samples from three independent ChIPs and three mock ChIPs were subjected to LM-PCR with one of the 36 possible primer combinations (Fig. 14.1G). Each PCR was performed with an annealing temperature of either 68 or 69 °C as indicated. The products were then resolved side-by-side on an 8% polyacrylamide gel. Bands indicated with white arrowheads are not putative targets because they are comparably present in both the ChIP and the mock lanes. Bands representing a putative target are indicated with black arrowheads. The bands labeled i and ii were excised from the gel, purified and re-amplified. b. The re-amplified products from panel A were treated with the restriction enzymes HaeIII (Ha), HinfI (Hi), and MspI (Ms). The digestion products were resolved on a 4% agarose gel. This example demonstrates that targets can be identified even in the presence of contaminating bands, seen in the lanes containing undigested products (U). The bands of interest, identified based on their size (compare with panel A), are indicated with arrows. A sub-fragment likely to have arisen from the band of interest is indicated in each of the HinfI lanes with an arrowhead. Indeed, after sequencing these AvaII-HinfI sub-fragments and after mapping of the “hit” to the mouse genome, it was shown to reside between two AvaII sites separated by 473 bp, consistent with the size of the band of interest that was excised from the acrylamide gel (panel A, and see Section 3.4.11.). M, Molecular weight marker.
14 Target Genes by ChIP Display a M
68°C + -
1:1 69°C + -
70°C + -
183
68°C + -
1:4 69°C + -
70°C + -
68°C + -
1:16 69°C + -
70°C + -
M
bp 517 506
396 344
298
b M
68°C + -
1:1 69°C + -
70°C + -
68°C + -
1:4 69°C + -
70°C + -
68°C + -
1:16 69°C + -
70°C + -
M
bp 517 506
396 344
298
Fig. 14.3 Effect of dilution and annealing temperature on CD. a. Three dilutions (1:1, 1:4, 1:16) of Runx2 ChIP (+) or mock ChIP (−) were amplified with one of the 36 possible primer combinations (Fig. 14.1G) at an annealing temperature of 68, 69, or 70 °C, and the products were resolved on an 8% non-denaturing polyacrylamide gel. Arrowheads indicate a putative target that was reproducible at all dilutions, and with both 70 and 69 °C, but not 68 °C, as the annealing temperature. b. Another example, using a different primer set. Arrowheads indicate a putative target that appeared only with annealing temperatures of 69 and 70 °C, and only without dilution. Although this candidate is inferior to the one in panel A, it can be considered a putative target, but might very well prove to be a false-positive during the validation steps (Sections 3.4.11. and 3.4.12.). M, Molecular weight marker.
3.3
PAGE and Target Isolation
We recommend that the CD procedure be completed with one or two of the 36 primer combinations (represented by the 36 boxes in Fig. 14.1G) before moving on to additional families. Only very experienced investigators should consider performing all of the LM-PCR reactions with all possible primer combinations (see Fig. 14.1G) prior to PAGE and target identification. 1. Cast a 1× TBE, 8% native polyacrylamide gel for resolving nucleic acids. We use gels 14 cm long and 17 cm wide. 2. After polymerization, load the wells with 1× sample loading buffer and pre-run in 1× TBE for 30 minutes at 150 V.
184
A. Barski et al.
3. Add 5 µL of 5× sample loading buffer to each sample after amplification and mix well. 4. Flush the wells of the gel thoroughly with a syringe. Using gel loading pipette tips, carefully apply 10 µL of each sample to the bottom of each well. We typically load samples immunoprecipitated with specific antibodies on one-half of the gel, whereas samples immunoprecipitated with non-specific antibody are loaded on the other half (see Fig. 14.2A). We load replicates next to each other and place molecular weight markers in the middle and on each side of the gel. 5. Allow the gel to run at 150–200 V, until the xylene cyanol dye is 3 cm from the bottom of the gel. 6. Remove the gel and incubate in 1× TBE supplemented with 100 ng/mL ethidium bromide. Visualize bands using a UV transilluminator. Capture and print the image. 7. Note any co-migrating bands that are reproducible in the ChIPs with the specific antibody. If these do not co-migrate with bands in the control lanes, then they are considered candidate targets (see Figs. 14.2A and 14.3, and Note 14). 8. Cut the bands that represent good candidate targets out of the gel using a fresh blade for each band. Be careful to avoid neighboring bands and lanes. 9. Place each excised gel piece in a Spin-X column, and freeze at −80 °C. Spin the columns at maximum speed in a microcentrifuge for 15 minutes at room temperature.
3.4
Target Identification
In principle, targets can be identified by cloning into a vector followed by sequencing. However, we prefer to identify targets without cloning, by directly sequencing sub-fragments obtained through secondary digestion as described in the following section (see Note 15). 1. Re-amplify the eluted DNA using the same primers and conditions as in Section 3.2., in a total reaction volume of 50 µL. 2. Split each amplified DNA sample into five 10 µL aliquots. Digest 2–4 of these aliquots with different restriction enzymes, each in a final volume of 20 µL (see Note 15). 3. After digestion, add 5 µL of 5× sample loading buffer to each sample, and mix thoroughly. Load 10 µL of each sample on a 4% agarose gel containing 100 ng/mL ethidium bromide. Load the undigested sample in lane #1, followed by the digested samples (see example in Fig. 14.2B). Repeat this for each group of samples, separating the groups with 1-kb DNA ladder. Allow the gel to run at 100 volts for 2–3 hours. 4. Place the agarose gel on a UV transilluminator, and note the pattern of bands (sub-fragments) resulting from the restriction digestion for each candidate target. Look for identical patterns of sub-fragments, which appear to have originated from a common target. Excise several such sub-fragments from the gel for sequencing (see Note 16).
14 Target Genes by ChIP Display
185
5. Purify the DNA on MinElute columns (Qiagen) according to the manufacturer’s protocol. Sequencing of the eluted DNA can be performed either in house or using a service lab. Steps 6–9 describe the sequencing procedure used in our lab. 6. Mix 10 µL of each eluted DNA sample with 2 µL 5× BigDye Sequencing Buffer, 4 µL Ready Reaction Premix, and 5 pmol of the primers used in Step 3.2.1. Add dH2O up to 20 µL. Place samples in a thermocycler and run the following program: (1) 95 °C for 4 minutes; (2) 93 °C for 30 seconds; (3) 50 °C for 15 seconds; (4) 60 °C for 4 minutes; (5) Go to 2, 30 times; and (6) End. 7. Remove samples from the thermocycler and clean DNA with AutoSeq G-50 Microcentrifuge Columns. 8. Place samples in SpeedVac and dry under low heat. Resuspend in 10 µL Hi-Di Formamide. 9. Transfer samples to a 96-well plate. Heat to 100 °C for 5 minutes and cool immediately on ice for 2 minutes. Load plate into ABI PRISM 3100 Genetic Analyzer and begin run. 10. Map the resulting sequences within the genome of interest, using BLAST or a similar search engine. 11. Search the regions surrounding the “hit” for AvaII recognition sequences. The distance (in bps) between the two sites on either side of the sequenced region should closely match the size of the fragment that was excised from the polyacrylamide gel. Additionally, the identities of the nucleotides at the ends of each fragment should match the primer pair used to generate the fragment in Section 3.2. Finally, the fragment should contain restriction enzyme recognition sequences at positions that are compatible with the digestion patterns that were seen on the agarose gel (step 3.4.4). 12. Validate each “hit” of interest using conventional ChIP assay with locus-specific primers (see Note 17). Annotated genes near confirmed targets are potentially regulated by the transcription factor of interest.
4
Notes
1. Target genes can be identified in comprehensive gene expression studies (e.g., using microarrays) based on their response to over-expression or activation (e. g., by ligand) of transcription factors. However, there are at least four problems with expression-based approaches. First, it is generally difficult to tell whether a responsive gene is a direct or indirect target of the transcription factor of interest. Second, in experiments involving over-expression of transcription factors, the response of some genes may be forced by exaggerated concentrations of the protein, resulting in physiologically insignificant results. Third, such studies do not provide information on the location of the cis-acting regulatory elements. Last, expression studies are unable to disclose genes, to which the transcription factor of interest binds without functional consequences under the experimental
186
2.
3.
4.
5.
A. Barski et al.
conditions employed. Such genes may become responsive to the transcription factor under different conditions, for example, in the presence of a specific extracellular stimulus and/or a particular co-activator (either in the same or a different cell type). CD is a “sampling” method and will only disclose a fraction of all the targets. High throughput methods for transcription factor target discovery include ChIPChip (5–7), SABE (8), STAGE (9), ChIP-PET (10), GMAT (11), SACO (12), and DamID (13). These methods are also used to map histone modifications, a task for which CD is not well suited. However, the high throughput methods share requirements, which render them prohibitively expensive for many labs. First, the amount of starting material needed is relatively large. Second, complex statistical analysis is needed to discern true targets from the large number of non-specifically immunoprecipitated fragments. Third, complicated equipment is required and the price is extremely high. This is exacerbated by the possibility of poor reproducibility, which usually necessitates many repetitions, especially in the beginning of a project with a new transcription factor. In addition, unlike some of the high throughput methods listed in the preceding, CD can lead to novel targets before a complete screen is finished. Of the 36 reaction types that constitute a full screen (see Fig. 14.1G), one can start with one or a subset of reactions and continue only if more targets are desirable. Furthermore, CD provides the opportunity to compare occupancy between different cells, and pursue only those with interesting phenotypes. For example, one can display side-byside targets occupied in two different cell types, or under different physiological conditions, and then pursue the identification of only selected targets. Any ChIP protocol can be adapted for CD. However, the typical high background of non-specifically immunoprecipitated fragments, which can be tolerated in conventional ChIP assays of candidate targets, is the Achilles’ heel of methods intended to identify unknown targets. Although CD is relatively forgiving in this sense, it will not distinguish between targets and contaminating DNA. Therefore, extra care should be taken to prevent contamination. Related to this, the beads used for immunoprecipitation should be pre-adsorbed with bacterial tRNA instead of salmon sperm DNA. tRNA will not be ligated or amplified during the following steps. Alternatively, the pre-adsorption step can be skipped altogether. Enrichment for known targets is a measure of the success of a ChIP experiment. Using real-time PCR with locus-specific primers, one compares the concentrations of known targets to those of non-targets after ChIP. To more rigorously measure the enrichment factor, the known-target-to-non-target ratio in the ChIP is compared to the respective ratio in samples immunoprecipitated with nonspecific antibodies, which should in theory be equal to 1 for single-copy genes. In our lab, we achieve a ~10-fold enrichment for Runx2 targets in MC3T3-E1 cells. Prior to AvaII treatment, we dephosphorylate the DNA with SAP to prevent ligation of linkers to the ends produced by sonication. We include this step even though we are not sure that sonication produces ends that are compatible with
14 Target Genes by ChIP Display
6.
7.
8.
9.
187
ligation. SAP was chosen because it can be efficiently heat denatured to prevent dephosphorylation of fragments generated later for LM-PCR by AvaII digestion. A key feature of CD is the restriction digestion, a step that standardizes the size of a diverse group of DNA fragments obtained from a given target after sonication and ChIP. The ability of CD to disclose a given target depends on the presence of two recognition sites for the restriction enzyme in the vicinity of that target. We use the 5-bp cutter AvaII because the genomic distribution of its cognate sites can be expected to result in many sonication fragments (which themselves are on the order of 300–900 bp) containing two AvaII sites. Sites for 6-bp cutters occur too infrequently, so it is unlikely that two sites would be present in one sonication fragment. Of course even for 5-cutters such as AvaII, many sonication fragments will not be long enough and targets will be missed. Digestion with 4-bp cutters, however, would result in a large number of fragments and would unnecessarily increase the complexity of the DNA fragment pool. Of several possible 5-bp cutters, we chose AvaII because it is active in salt-free React1 buffer, eliminating the need for desalting prior to ligation. Conveniently, AvaII is an inexpensive and stable enzyme. Noteworthy, digestion by AvaII can be blocked by CpG methylation. Although this enzyme property may limit target discovery, it can be considered advantageous because methylated regions may be associated with less accessible genes. Be that as it may, among the initial Runx2 targets identified by CD, a fair number were associated with CpG islands. The second dose of AvaII is added to digest concatemers of AvaII fragments that might have been generated during linker ligation. Re-digestion with AvaII recycles these concatemers, whereas the presence of ligase facilitates linker ligation. Because the linkers do not restore the AvaII site (see Fig. 14.1F), their ligation to digested DNA fragments is irreversible even in the presence of AvaII. We use Taq DNA polymerase because it lacks proofreading activity. This prevents “editing” of the nested primers, which may anneal to the template even when the nucleotides at positions +3 or +6 (see Fig. 14.1F) are not complementary. Such mismatches would be perfect substrates for a 3′–5′ proofreader, which would then use the edited primer to amplify the fragment even when it should not be amplified in that specific reaction. By amplifying targets in individual families (see Fig. 14.1G), we increase the signal-to-noise ratio for one group of targets at a time. Because all fragments originating from a given target have the same nucleotides at the ends, they remain in the same family and the signal is not eroded. In contrast, the background fragments are scattered into many families (see Fig. 14.1E). This feature of CD also provides investigators with the choice of performing either a full or a partial screen, i.e., using as many primer combinations (families) as they wish. In our search for Runx2 target genes in MC3T3-E1 cells, we limited the initial CD screen to eight families, those that are amplified by a single primer each (1), and later continued with other families (2).
188
A. Barski et al.
10. In addition to biological replicates, we strongly encourage the use of technical replicates. There is a huge number of non-specifically immunoprecipitated fragments in each ChIP sample that are all competing for a limited number of primers in the amplification reaction. Usually, a subset of these fragments wins out early on in the competition, often for random reasons. The so-called “founder effect” and the complexity of the template DNA result in a different pattern of amplification from run-to-run of the same sample (see Fig. 14.2A). Although fragments that represent true targets are present at a higher concentration and should be amplified reproducibly, technical replicates help one to distinguish between true targets and spuriously amplified fragments (see Fig. 14.2A). 11. As an alternative to using direct replicates, it may be worthwhile to perform the PCR on different dilutions of the ChIP material. Seeing a particular band with all dilutions enhances one’s confidence that it truly represents a target (see Fig. 14.3A). Other bands, which are visible with lower dilutions, but not with higher dilutions (see example in Fig. 14.3B) can be either dismissed or pursued contingent upon validation (step 3.4.12). 12. Do not pre-heat the thermocycler, because this might compromise amplification efficiency. The linkers used for CD are unphosphorylated oligonucleotide duplexes; thus, only the short oligo of the duplex is ligated to the AvaII fragment. The long oligo is kept in place through non-covalent interactions and will be lost upon heating. To allow the polymerase to fill-in the lost sequence before the strands are completely separated, we start at room temperature and hold the reaction for 30 seconds at 72 °C. Hot-start polymerases complexed with antibodies should not be used in this PCR as they become active only after the reaction is heated to 95 °C. 13. Due to the large amount of non-specifically precipitated fragments in a ChIP sample, non-specific amplification is a technical challenge that must be overcome by CD. To this end, we use short (14-bp) primers and high annealing temperatures. Unfortunately, it is possible for the annealing temperature selected to be too stringent for any amplification to occur. Because it is difficult to predict what annealing temperature will yield the right balance between too stringent and too permissive, we recommend that replicates be performed at slightly different annealing temperatures. This can be accomplished easily using a thermocycler with gradient capabilities. Two or three temperatures between 68–70 °C usually provide good results (see Figs. 14.2 and 14.3). 14. In theory, targets should appear as co-migrating bands in all of the ChIP lanes but none of the mock ChIP lanes. In practice, such perfect scenarios are infrequent (see Fig. 14.2A), and investigators must exercise their own good judgment in selecting bands for further analysis. In any case, each putative target must be validated using conventional ChIP assay with locus-specific primers (step 3.4.12). 15. There are at least three reasons for the secondary amplification and digest. First, it is useful as a screen to confirm that co-migrating bands from different lanes actually contain the same sequence. Second, it helps to eliminate contamination from overlapping and neighboring bands present on the polyacrylamide gel. Restriction digestion results in a unique pattern of sub-fragments arising from
14 Target Genes by ChIP Display
189
the target; sub-fragments that participate in such a pattern will likely lead to identification of true targets (see Fig. 14.2B). Third, the secondary digestion facilitates the sequencing step. Because the primers used for sequencing are the same as those used for amplification, they would anneal to both ends of an intact fragment, producing two simultaneous and thus uninterpretable sequences. Digestion of the fragments is intended to create a sequence-able template with only one end to which a primer can anneal. Because most of the fragments isolated after PAGE are up to 500-bp in length, it is best to start the secondary digestion with enzymes that have a short recognition sequence, usually 4-bp cutters. Restriction enzymes that work well in our hands for the secondary digestion are HaeIII, MspI, RsaI, and HinfI, all of which are active in the PCR buffer. If these enzymes do not yield sequence-able sub-fragments, digestion with a panel of additional 4-bp cutters, or even 5- or 6-bp cutters alone or in combination, is advisable. Again, to avoid a desalting step, it is helpful to select enzymes that are active in the PCR buffer (a.k.a. primer extension mix). 16. Bands excised from the agarose gel must have one intact end containing the PCR primer in order to be sequenced. Because some bands may have lost both original ends during the secondary digest, we usually excise more than one band from the agarose gel, with preference toward those that are only slightly smaller than the undigested fragment. It is also advisable to excise several replicate sub-fragments because even sequence-able sub-fragments can become corrupted during gel extraction or the sequencing step itself. Having replicates increases the odds of obtaining a sequence. 17. For validation by conventional ChIP, there are at least two important considerations. First, CD does not disclose the exact sequence to which transcription factors bind. The discrete binding site may lie anywhere within the AvaII fragment, or even just outside the fragment. Thus, when designing primers for validation, it may be useful to target more than one sequence within, or immediately surrounding, the AvaII fragment. Cognate binding motifs found in the region can provide guidance for primer design. Second, validation of a target disclosed by CD is most likely to occur when using the same ChIP material that was used as input for the CD experiment. This is due to the fact that transcription factor occupancy at some loci may be sensitive to minor alterations in culture conditions. For this reason, it may be wise not to use all of the original ChIP material for CD, so that the leftover can be employed for validation. In addition, we recommend the validation of each target in independent ChIPs, which would demonstrate that occupancy by the transcription factor of interest occurs regardless of minor alterations to the culture conditions. Acknowledgments ChIP Display was developed with NIH grant R21 AR49297 and further employed in projects supported by the NIH (RO1 DK071122 and RO1 CA109147) and the Department of Defense (PC040322). The work was conducted in a facility constructed with support from Research Facilities Improvement Program grant no. C06 (RR10600-01, CA62528-01, RR14514-01) from the NIH/NCRR. BF holds the J. Harold and Edna L. LaBriola Chair in Genetic Orthopaedic Research at the University of Southern California.
190
A. Barski et al.
References 1. Barski, A. and Frenkel, B. (2004) ChIP Display: novel method for identification of genomic targets of transcription factors. Nucleic Acids Res, 32, e104. 2. Pregizer, S., Barski, A., Gersbach, C.A., Garcia, A.J. and Frenkel, B. (2007) Identification of novel Runx2 targets in osteoblasts: Cell type-specific BMP-dependent regulation of Tram2. J Cell Biochem, 102, 1458–1471. 3. Jariwala, U., Prescott, J., Jia, L., Barski, A., Pregizer, S., Cogan, J.P., Arasheben, A., Tilley, W.D., Scher, H.I., Gerald, W.L. et al. (2007) Identification of novel androgen receptor target genes in prostate cancer. Mol Cancer, 6, 39. 4. Prescott, J., Jariwala, U., Jia, L., Cogan, J.P., Barski, A., Pregizer, S., Shen, H.C., Arasheben, A., Neilson, J.J., Frenkel, B. et al. (2007) Androgen receptor-mediated repression of novel target genes. Prostate, 67, 1371–1383. 5. Cawley, S., Bekiranov, S., Ng, H.H., Kapranov, P., Sekinger, E.A., Kampa, D., Piccolboni, A., Sementchenko, V., Cheng, J., Williams, A.J. et al. (2004) Unbiased mapping of transcription factor binding sites along human chromosomes 21 and 22 points to widespread regulation of noncoding RNAs. Cell, 116, 499–509. 6. Horak, C.E. and Snyder, M. (2002) ChIP-chip: a genomic approach for identifying transcription factor binding sites. Methods Enzymol, 350, 469–483. 7. Ren, B., Robert, F., Wyrick, J.J., Aparicio, O., Jennings, E.G., Simon, I., Zeitlinger, J., Schreiber, J., Hannett, N., Kanin, E. et al. (2000) Genome-wide location and function of DNA binding proteins. Science, 290, 2306–2309. 8. Chen, J. and Sadowski, I. (2005) Identification of the mismatch repair genes PMS2 and MLH1 as p53 target genes by using serial analysis of binding elements. Proc Natl Acad Sci U S A, 102, 4813–4818. 9. Kim, J., Bhinge, A.A., Morgan, X.C. and Iyer, V.R. (2005) Mapping DNA-protein interactions in large genomes by sequence tag analysis of genomic enrichment. Nat Methods, 2, 47–53. 10. Wei, C.L., Wu, Q., Vega, V.B., Chiu, K.P., Ng, P., Zhang, T., Shahab, A., Yong, H.C., Fu, Y., Weng, Z. et al. (2006) A global map of p53 transcription-factor binding sites in the human genome. Cell, 124, 207–219. 11. Roh, T.Y., Ngau, W.C., Cui, K., Landsman, D. and Zhao, K. (2004) High-resolution genomewide mapping of histone modifications. Nat Biotechnol, 22, 1013–1016. 12. Impey, S., McCorkle, S.R., Cha-Molstad, H., Dwyer, J.M., Yochum, G.S., Boss, J.M., McWeeney, S., Dunn, J.J., Mandel, G. and Goodman, R.H. (2004) Defining the CREB regulon: a genome-wide analysis of transcription factor regulatory regions. Cell, 119, 1041–1054. 13. van Steensel, B. and Henikoff, S. (2000) Identification of in vivo DNA targets of chromatin proteins using tethered dam methyltransferase. Nat Biotechnol, 18, 424–428.
Chapter 15
Application of the Laser Capture Microdissection Technique for Molecular Definition of Skeletal Cell Differentiation In Vivo Dafna Benayahu, Rina Socher, and Irena Shur
Abstract Laser capture microdissection (LCM) method allows selection of individual or clustered cells from intact tissues. This technology enables one to pick cells from tissues that are difficult to study individually, sort the anatomical complexity of these tissues, and make the cells available for molecular analyses. Following the cells’ extraction, the nucleic acids and proteins can be isolated and used for multiple applications that provide an opportunity to uncover the molecular control of cellular fate in the natural microenvironment. Utilization of LCM for the molecular analysis of cells from skeletal tissues will enable one to study differential patterns of gene expression in the native intact skeletal tissue with reliable interpretation of function for known genes as well as to discover novel genes. Variability between samples may be caused either by differences in the tissue samples (different areas isolated from the same section) or some variances in sample handling. LCM is a multi-task technology that combines histology, microscopy work, and dedicated molecular biology. The LCM application will provide results that will pave the way toward high throughput profiling of tissue-specific gene expression using Gene Chip arrays. Detailed description of in vivo molecular pathways will make it possible to elaborate on control systems to apply for the repair of genetic or metabolic diseases of skeletal tissues. Keywords Laser capture microdissection, skeletal tissues, gene expression, RNA isolation, chromatin immunoprecipitation.
1
Introduction
Skeletal tissues develop from mesenchymal stem cells that differentiate to chondrogenic, osteogenic, hematopoietic-supporting and skeletal muscle cells (1–5). Skeletal tissues are active throughout life, in skeletal growth and remodeling, and are regulated by multiple factors, including systemic hormones and local regulatory factors (5–8). A cell’s differentiation is coordinated by activation or repression of transcription factors implicated in regulation of proliferation, and of functional From: Methods in Molecular Biology, Vol. 455: Osteoporosis: Methods and Protocols Edited by Jennifer J. Westendorf © Humana Press, Totowa, NJ
191
192
D. Benayahu et al.
genes for certain cytokines, their receptors, and ECM proteins that serve a crucial role in skeletal differentiation. A decrease in osteoblast activity and an increase in bone resorption by osteoclasts enhance bone remodeling, which results in osteoporosis. Studying the alterations in cell differentiation will teach us more about the connection between stem cell differentiation and skeletal pathology which occurs in different metabolic disorders including osteoporosis. Disfunction of skeletal cells leads to osteopenia and osteoporosis, diseases that are responsible for millions of fractures annually, mostly involving the lumbar vertebrae, hip, and wrist. Studying the changes in bone physiology in animal models will shed light on the type of changes that may contribute to develop modalities to attenuated bone loss. The laser capture micro-dissection (LCM) method allows the selection of individual or clustered cells from intact tissues. Subsequent RNA/DNA and/or protein extraction enables one to interpret the molecular mechanisms that function in vivo in the particular cells and tissues. LCM provides the opportunity to uncover the molecular mechanisms controlling cellular fate in the natural microenvironment. This technique is particularly useful for selecting cells from skeletal tissues that are difficult to study individually because of their anatomical complexity (10, 11). The LCM technique is particularly attractive for in vivo characterization of mesenchymal cells because it allows the definition of molecular mechanisms that drive differentiation of distinct cells in skeletal tissues. Single cell analysis sheds light on alterations in regulatory processes, which occur in osteoporosis, or changes of physiological status, such as aging. The use of LCM overcomes the problem of tissue heterogeneity and allows for the selection of defined cells while minimizing contamination from neighboring unrelated tissues. It is now realistic to extract small numbers of cells, isolate RNA, DNA and protein, and perform various molecular analyses, including chromatin immunoprecipitation (ChIP), to compare different types of cells from various tissues. Such analysis provides reliable interpretation of gene function in specific tissues. An analysis of tissue-specific gene expression in vivo enables one to determine the specific tissue control systems required for their differentiation. In addition, these methods enable discovery of tissue-specific genes and provide detailed descriptions of in vivo molecular pathways. The presented method can be used to obtain data that could be applied to prevent or delay the onset of age-related skeletal disabilities and diseases. Acquired knowledge about the intrinsic control of osteoblast transcription factors and gene expression in vivo will serve to develop novel diagnostic and therapeutic tools for various pathologies of the bone.
2
Materials
For molecular analyses, use gloves throughout the entire procedure along with RNase-free instruments and reagents.
15 The Laser Capture Microdissection Technique
2.1
193
Preparation of Samples for Laser Capture Microdissection
1. Frozen samples are embedded in Tissue-Tek embedding medium (OCT, Torrance, CA, United States). 2. One-millimeter-thick sterile glass slides, or membrane covered slides (PALM, Germany). 3. Phosphate buffer solution (PBS) containing 40 units of ribonuclease inhibitor. 4. 70% ethanol for fixation.
2.2
Histological Staining
1. Mayer’s hematoxylin (H&E) solution. 2. 70%, 96%, 100% ethanol solutions.
2.3
Laser Pressure Catapulting of the Samples
1. Catapult Buffer: 0.5 µM EDTA pH 8.0, 20 µM Tris-HCl, pH 8.0. 2. Proteinase K: 20 mg/mL in 100 µL of double distilled water (DDW) that is DEPC-treated (see Note 1). 3. Sterile mineral oil.
2.4 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14.
RNA Analysis RNAeasy total kit (Qiagen, Germany). Trizol (Sigma) for extraction in a phenol precipitation-based method. PALM RNA-extraction kit (PALM). RNeasyMini (Qiagen). Absolutely RNA nanoprep kit (Stratagene, La Jolla, USA). High Pure RNA tissue kit (Roche, Switzerland). Purescript RNA isolation kit (Gentra, Minneapolis, USA). ArrayPure Nano-scale RNA purification kit (Epicentre Biotechnologies, Madison, USA). SuperScript III reverse transcriptase (Invitrogen, Carlsbad, USA). iScript Select cDNA synthesis kit (Bio-Rad). Reverse-i MAX Blend kit (ABgene, United Kingdom). PCR mix (Sigma, USA). Tris Borate EDTA (TBE) buffer. SeaKem LE Agarose (Cambrex, New Jersey, USA).
194
2.5
D. Benayahu et al.
Chromatin Immunoprecipitation (ChIP) for LCM Samples
1. SDS Lysis Buffer: 1% SDS, 10 mM EDTA, 50 mM Tris-HCl, pH 8.1. 2. ChIP Dilution Buffer: 0.01% SDS, 1.1% Triton X-100, 1.2 mM EDTA, 16.7 mM Tris-HCl, pH 8.1, 167 mM NaCl. 3. Low Salt Wash Buffer: 0.1% SDS, 1% Triton X-100, 2 mM EDTA, 20 mM TrisHCl, pH 8.1, 150 mM NaCl. 4. High Salt Wash Buffer: 0.1% SDS, 1% Triton X-100, 2 mM EDTA, 20 mM TrisHCl, pH 8.1, 500 mM NaCl. 5. LiCl Wash Buffer: 0.25 M LiCl, 1% IGEPAL-CA630, 1% deoxycholic acid (sodium salt), 1 mM EDTA, 10 mM Tris-HCl, pH 8.1. 6. TE Buffer: 10 mM Tris-HCl, 1 mM EDTA, pH 8.0. 7. Elution Buffer: 1% SDS, 0.1 M NaHCO3.
3 3.1
Methods Preparation of Slides
1. Freeze isolated tissues in liquid nitrogen. 2. Embed frozen tissue in optimal cutting temperature (OCT) Tissue-Tek embedding medium. 3. Cut 6-µm thickness sections at −20 °C using a cryostat (e.g., Jung Frigocut 2800N microtome, Leica). 4. Mount cryo-tissue sections on the regular 1-mm thick sterile glass slides, or membrane covered slides. 5. Sections may be fixed or deep frozen at −80 °C for the longer storage (see Note 2).
3.2
Fixation
1. Air-dry slides for 20 seconds. 2. For RNA isolation it is recommended to pipet the RNase inhibitor solution on the slide, then dip the mounted sections for 90 seconds into 70% ethanol. The dip in RNase-free water (see Note 3). 3. Allow the fixed sections to dry at the room temperature for 60 minutes.
3.3
Histological Staining
Sections may be used unstained if areas of interest are recognized based on morphology, or can be stained for better visualization (see Note 4). Mayer’s hematoxylin/ eosin (H&E) is a routine histological staining, which results with nuclei stained in blue and cytoplasm is pink/red (Fig. 15.1).
15 The Laser Capture Microdissection Technique
195
Fig. 15.1 Morphology of stained (a) and unstained (b,c) sections. Sections stained with H&E are used for the visualization of periosteum (P) and bone (B). The isolated areas are marked by arrows.
1. Immerse the sections for 3 minutes in Mayer’s hematoxylin solution followed by 10 minutes of rinsing in water. 2. Immerse the sections for 3 minutes in eosin, quickly dip in water. 3. Following staining, slides are dehydrated in increasing ethanol solutions of 70%, 96%, and 100%. 4. Air dry for 60 minutes.
3.4
Specimen Isolation and Collection
3.4.1
Laser Cutting (Microdissection) of the Samples
1. Search for an area of interest on the section. 2. Pipette 5 µL of ethanol onto the depiction area to improve the visualization of the cells. 3. Mark the cells or cell area with the software tool. The ethanol will evaporate rapidly. Next you will catapult the marked cells or cell areas.
3.4.2
Laser Pressure Catapulting of the Samples
Use the cap with liquid that will allow the catapulted samples to stick to the cap. You will add autoclaved mineral oil (PCR oil) or catapult buffer in the inner ring of the cap. Aqueous solutions dry out after a while. The catapulted cells or cell areas stick onto the wet inner surface of the cap and do not fall down after the catapulting procedure (see Note 5). Alternatively use silicon caps that are clear adhesive caps without liquid. There is no danger of RNase activity in the absence of water, nor evaporation and crystal formation during extended sample harvesting. On the other hand, low energy is needed for catapulting due to short target distance. 1. Pipet 3–6 µL of catapult buffer into the middle of the cap (an optional method is described in the following). 2. Put the cap into the cap holder. 3. Perform laser microdissection and laser pressure catapulting of selected cells or cell areas into the cap (see Note 6).
196
D. Benayahu et al.
4. Remove the cap from the cap holder and put it onto a 0.5-mL microfuge tube containing lysis buffer or pipette lysis buffer into the tube with the attached cap. 5. Mix and perform the lysis of the cells by inversion. 6. Centrifuge the sample at 16,000 g for 3 minutes. 7. If not continue with isolation procedure, store the samples in lysis buffer at −80 °C. 3.4.3
Looking Into the Cap to See the Catapulted Samples
The efficiency of catapulting is observed by looking into the collection device (e.g., cap) with the 5×, 10×, 40×, and 63× objectives. By using the software function “go to checkpoint” the lid is moved out of the light path and the cap can be lowered further toward the objectives. Normally most catapulted areas/cells can be found within the small inner ring of the caps. 3.4.4
Getting the Collected Cells from the Cap into the Tip of the Tube
After microdissection, the fluid from the cap is spun down in a bench centrifuge (3 minutes, 16,000 g) and samples can be stored for later use. For RNA extraction, add an appropriate lysis buffer to the tube, close the tube, and mix by inversion. Then spin down the lysate as described in the preceding. (For future RNA isolation/ analysis, the tube is placed on ice or stored at −80 °C.)
3.5
RNA Isolation
Working with RNA is more demanding than working with DNA, because of the chemical instability of the RNA and the ubiquitous presence of RNases. In a case that LCM-obtained cells are going to be used for the RNA extraction the following steps should be taken: 1. Designate a special area for RNA work only. 2. Clean benches with 100% ethanol or a special cleaning solution (e.g., RNase Zap). 3. Always wear gloves. After putting on gloves, do not touch surfaces or equipment to avoid reintroduction of RNase to decontaminated material. 4. Use sterile, disposable plastic ware. 5. Use filtered pipetter tips. 6. Glassware should be baked at 180 °C for 4 hours. (RNases can maintain activity even after prolonged boiling or autoclaving!) 7. Use reagents that are RNase-free. 8. All solutions should be made with DEPC-(diethylpyrocarbonate) treated H2O. 9. For best results use either fresh samples or samples that have been quickly frozen in liquid nitrogen or at −80 °C. (This procedure minimizes degradation of RNA by limiting the activity of endogenous RNases.) All required reagents should be kept on ice.
15 The Laser Capture Microdissection Technique
197
10. To ensure RNase-free slides perform sterilization by autoclaving. Sections that are mounted on slides maybe treated to reduce the chance of contamination with exogenous RNases with RNase-ZAP followed by two separate washings in DEPCtreated water and drying at 37 °C up to 55 °C for 30 minutes up to 2 hours. 11. Store RNA by aliquoting in ethanol or RNA buffer at −80 °C and maintain the RNA at this temperature for long-term storage. Store prepared slides at −80 °C too. 12. RNA is not stable at elevated temperatures; therefore, avoid high temperatures (> 65 °C) because they affect the integrity of RNA. 13. RNA extraction is performed by using kits for the small-scale RNA purification. Kits are available from various companies (see Materials). 14. Two options are available to quantify RNA at a small scale: (1) Use a nanodrop to measure absorbance at A260 at the nanometer range. The assay can measure as low as 1 ng/µL and it is minimally affected by contamination likely to be found in small nucleic acid preps. (2) LabChips for analyzing RNA by the pico or nano Lab Chip Kit (Agilent) for quantitation on Agilent 2100 bioanalyzer (Agilent Technologies, Palo Alto, CA). The resulting electropherograms determine both integrity and concentration, of the RNA sample.
3.6
Reverse Transcription and RT-PCR
RNA reverse transcription (RT) to make cDNA includes the first strand cDNA synthesis with subsequent amplification by PCR using gene specific primers. It could be performed using several reverse transcriptases (e.g., Sensiscript III Reverse Transcriptase; iScript RNase H+ MMLV reverse transcriptase for sensitive detection using 1 pg to 1 µg of input total RNA; Reverse-iT MAX Blend, which includes a unique combination of RTases. The AMV component allows reverse transcription to be carried out at up to 57 °C when secondary structure is problematic, and the MMuLV provides maximum yields. RNase inhibitor is included in the blend to prevent degradation from RNAse A.) Both random hexamers and anchored oligo-dT can be used to prime the RT reactions. Random primers generate the most diverse pool of cDNA, whereas oligo-dT primers anneal only to mRNA poly-A tails and thus will be biased toward the 3′ end.
3.7
PCR Analysis
PCR analyses are performed based on protocols available at manufacturer web sites for the PCR kits. Two major technologies are used: (1) standard semi-quantitative PCR with the end-point product detection visualized on agarose gels for detection of PCR amplification is based on size determination of the product, or (2) real-time detection kinetic PCR is based on the detection and quantitation of a fluorescent reporter, which increases in direct proportion to the amount of PCR product in a reaction. By recording the amount of fluorescence emission at each cycle, it is possible
198
D. Benayahu et al.
to monitor the PCR reaction during the exponential phase, in which the first significant increase in the amount of PCR product correlates to the initial amount of target template. Specificity of amplification is assessed with a melting study. The presence of specific amplification products is confirmed by a single peak on the melting curve, which indicates an absence of non-specific amplification and primer–dimer formation.
3.8
Chromatin Immunoprecipitation from Tissues Samples Retrieved by LCM
1. Cryosections are fixed in 1% formaldehyde for 10 minutes. 2. Wash in PBS and air-dry for 30 minutes at room temperature. 3. Laser cut (microdissection) and laser pressure catapult (LPC) selected cells or cell areas. 4. Catapulting 2,500–3,000 cells is necessary for each sample. 5. Cups are inserted into the autoclaved 0.5-mL tubes and centrifugation is performed at 16,000 g for 3 minutes. 6. After the centrifugation the samples are processed for ChIP. 7. Resuspend cell pellet in 100 µL of SDS lysis buffer with protease inhibitors and incubate for 10 minutes on ice. 8. Sonicate for 3 minutes at max frequency using cup horn with ultrasonic water bath (e.g., MIXL-2020 Sonicator, MiSonix, Inc.) (see Note 7). 9. Centrifuge sample for 10 minutes at 16,000 g at 4 °C, and transfer the supernatant to a new 1.5 mL-microcentrifuge tube. Discard the pellet. 10. Dilute the sonicated cell supernatant 10-fold in ChIP Dilution Buffer with protease inhibitors (see Notes 8 and 9). 11. Add the antibody of interest for immunoprecipitation (the amount will vary per antibody) to the 1 mL of supernatant fraction and incubate overnight at 4 °C with rotation. 12. Add 30 µL of protein-A agarose/salmon sperm DNA (50% slurry). Incubate for 1 hour at 4 °C with rotation to collect the antibody/protein complex. 13. Pellet agarose by gentle centrifugation at 100–170 g at 4 °C for 1 minute. Carefully remove the supernatant that contains unbound non-specific DNA (see Note 10). 14. Wash the protein A agarose/antibody/protein complex for 5 minutes on a rotating platform with 1 mL of each of the buffers: (1) low salt wash buffer, (2) high salt wash buffer, (3) LiCl wash buffer, and (4) TE buffer. 15. Elute the protein–DNA complex from the antibody by adding 250 µL freshly prepared elution buffer to the pelleted protein A agarose–antibody–protein complex. Mix by vortex and incubate at room temperature for 15 minutes with rotation.
15 The Laser Capture Microdissection Technique
16. 17. 18. 19. 20.
199
Spin down agarose, and carefully transfer the supernatant fraction to another tube and repeat elution. Combine eluates, the total volume is now 500 µL. Add 20 µL of 5 M NaCl to the combined eluates (500 µL) and reverse proteinDNA cross-links by heating at 65 °C for 4 hours (see Notes 11 and 12). Add 10 µL of 0.5 M EDTA, 20 µL of 1 M Tris-HCl, pH 6.5, and 2 µL of 10 mg/ mL proteinase K to the combined eluates and incubate for 1 hour at 45 °C. Recover DNA by phenol/chloroform extraction and ethanol precipitation (see Notes 13 and 14). Wash pellets with 70% ethanol and air dry. Resuspend pellets in TE buffer for PCR (Fig. 15.2 and Tables 15.1 and 15.2).
Fig. 15.2 ChIP analysis was performed with the antibody that specifically recognizes chromatin remodeling protein CReMM/CHD9 (12-13). The unbound fraction was used for the Re-ChIP with anti-polymerase-II (Pol-II) antibody. We analyzed the amplification of Cbfa1, Biglycan, and PPARgamma promoters using specific primers listed in Table 15.1. The PCR products were visualized in an agarose gel and the results are summarized in Table 15.2. In: Input fraction; B: Bone, P: Periost, 1-ChIP with anti-CReMM/CHD9 antibody; 2- reChIP with anti-Pol-II antibody.
Table 15.1 PCR primers and expected size for PCR product Gene Forward primer Reverse primer
Expected PCR product
Cbfa1
AAAGCCACAGTG GTAGGCAG
TGAAGCATTCAC ACAATCCAA
142 bp
Biglycan
CCATATCCTGC TTCTCCCA
CCAGGGACCAA GTGTAAGGA
223 bp
PPARγ
ATGGTTATTCCA TTTGGGGG
CTTAAGGCCTTT GCCCTTTT
203 bp
Table 15.2 Results of chip and re-chip of periost and endochondral bone Re-ChIP with Pol-II ChIP with DB P B P B −
+
+
+
Cbfa1
−
+
−
+
Biglcan
−
−
−
+
PPARγ
200
4
D. Benayahu et al.
Notes
1. Addition of proteinase K is an optional step. The solution should be freshly prepared. 2. The best results are obtained when freshly prepared cryo-sections are used. 3. If OCT or another tissue freezing medium was used, it is important to get rid of the medium on the slide before fixation; otherwise, it will interfere with laser efficiency. Removing of the medium is achieved by gentle washing the slide for about 1 minute in water. 4. H&E-staining of cryo-sections is also suitable for RNA preparation. Slides should be treated carefully with solutions prepared with RNase–free water. The whole procedure is shortened from standard 10 to 3 minutes to minimize the activity of RNases. Alternatively, if RNA extraction is performed from the unstained slides, every ninth slide in the serial section could be stained with H&E to provide better visual reference. 5. For single cells or very small areas, spinning down from oil may be difficult; therefore, aqueous solutions should be preferred. 6. When using membrane mounted samples the dissected membrane acts as a backbone. The selected area/cell can be catapulted with a single laser shot from a remaining “bridge” at the border. Morphological integrity is completely preserved with this procedure. Important: Laser settings have to be adjusted to cells or tissue in concern. 7. After this step samples can be frozen at −80 °C. 8. A portion of the diluted cell supernatant (1%) should be kept. This is the input sample of cross-linked histone-DNA complexes that will be reversed by adding 1 µL of 5 M NaCl and heating at 65 °C for 4 hours. 9. For a negative control, perform a no-antibody immunoprecipitation by incubating the supernatant fraction with 30 µL of protein A agarose/salmon sperm DNA (50% slurry). 10. Supernatant (unbound fraction) can be further used for the subsequent ChIP analysis of the protein–DNA complexes that can be precipitated from the unbound fraction, by adding an antibody of interest to the unbound fraction and repeating the incubation step overnight, i.e., Re-chip. Then add protein A agarose/ salmon sperm DNA for 1 hour and pellet agarose by gentle centrifugation at 100–170 g at 4 °C for 1 minute. Carefully remove the supernatant that contains unbound non-specific DNA. 11. At this step the sample can be stored at −20 °C and the procedure continued at a later date. 12. Include the input material and perform the protein–DNA cross-link reversal by adding 1 µL of 5 M NaCl per 20-µL sample and heating to 65 °C for 4 hours. 13. The addition of an inert carrier, such as 20 µg glycogen or yeast tRNA, provides for better visualization of the DNA pellet. 14. Alternatively, DNA purification can be performed by using commercially available kits for the genomic DNA purification.
15 The Laser Capture Microdissection Technique
201
Acknowledgments This study was supported by funding from Ramot at Tel Aviv University and by the Chief Scientist of Ministry of Commerce and consortium “Bereshit” and the CellProm program from the EEC 6th framework to D.B.
References 1. Friedenstein, A. J., Piatetzky, S., II, Petrakova, K. V. (1966) Osteogenesis in transplants of bone marrow cells. J Embryol Exp Morphol 16, 381–390. 2. Benayahu, D. (2000) The hematopoietic microenvironment: the osteogenic compartment of bone marrow: cell biology and clinical application. Hematology 4, 427–435. 3. Jiang, Y., Jahagirdar, B. N., Reinhardt, R. L., et al. (2002) Pluripotency of mesenchymal stem cells derived from adult marrow. Nature 418, 41–49. 4. Pittenger, M. F., Mackay, A. M., Beck, S. C., et al. (1999) Multilineage potential of adult human mesenchymal stem cells. Science 284, 143–147. 5. Benayahu D, Akavia UD, Shur I. (2007). Differentiation of bone marrow stroma-derived mesenchymal cells. Current Medical Chemistry. 14 (2):173–9. 6. Syed, F., Khosla, S. (2005) Mechanisms of sex steroid effects on bone. Biochem Biophys Res Commun 328, 688–696. 7. Benayahu, D., Shur, I., Ben-Eliyahu, S. (2000) Hormonal changes affect the bone and bone marrow cells in a rat model. J Cell Biochem 79, 407–415. 8. Compston, J. (2001) Sex steroids and bone. Physiol Rev 81, 419–447. 9. Locklin, R., Williamson, M., Beresford, J., et al. (1995) In vitro effects of growth factors and dexamethasone on rat marrow stromal cells. Clin Orthop 313, 27–35. 10. Benayahu, D., Akavia, U. D., Socher, R., Shur, I. (2005) Gene expression in skeletal tissues: application of laser capture microdissection. J Microsc 220(Pt 1), 1–8. 11. Benayahu D, Socher R, Shur I. 2007. Laser Capture Microdissection of Bone Cells. Methods in Molecular Biology/Molecular Medicine Osteoporosis Editor: Jennifer Westendorf published by Humana Press, USA. 12. Shur, I., Benayahu, D. (2005) Characterization and functional analysis of CReMM, a novel chromodomain helicase DNA-binding protein. J Mol Biol 352, 646–655. 13. Shur, I., Socher, R, Benayahu, D. (2006) In vivo association of CReMM/CHD9 with promoters in osteogenic cells. J Cell Physiol 207, 374–378. 14. Shur, I., Salomon, R., Benayahu, D. (2006) Dynamic interactions of chromatin-related mesenchymal modulator, a chromodomain helicase-DNA-binding protein, with promoters in osteoprogenitors. Stem Cells 24, 1288–1293. 15. Marom, R., Shur, I., Hager, G. L., Benayahu, D. (2006) Expression and regulation of CReMM, a chromodomain helicase-DNA-binding (CHD), in marrow stroma derived osteoprogenitors. J Cell Physiol 207(3), 628–35.
Chapter 16
Quantitative Trait Loci Mapping Dong-Hai Xiong, Jian-Feng Liu, Yan-Fang Guo, Yan Guo, Tie-Lin Yang, Hui Jiang, Yuan Chen, Fang Yang, Robert R Recker, and Hong-Wen Deng
Abstract This chapter presents current methods for mapping quantitative trait loci (QTLs) in natural populations especially in humans. We discussed the experimental designs for QTL mapping, traditional methods adopted such as linkage mapping approaches and methods for linkage disequilibrium (LD) mapping. Multiple traits and interaction analysis are also outlined. The application of modern genomic approaches, which mainly exploit the microarray technology, into QTL mapping was detailed. The latter are very recent protocols and are less developed than linkage and association methods at present. The main focus of this chapter is technical issues although statistical issues are also covered to certain extent. Finally, we summarize the limitations of the current QTL approaches and discuss the solutions to certain problems. Keywords QTL, linkage, association, linkage disequilibrium, microarrays, eQTL, whole genome association.
1 1.1
Introduction Experimental Design for QTL Mapping
The classical quantitative genetics theory presumes that quantitative traits are controlled by an infinite number of genes, each with an infinitesimal effect, and are influenced by environmental factors. However, in addition to the polygenes with infinitesimal effects, there are other loci that may have major effects on quantitative traits (1) Such genes are called quantitative trait loci (QTL). A quantitative trait locus is defined as a region of the genome (i.e., segment of chromosome) that harbors one or more genes affecting a quantitative trait (2). The QTL concept is also used for trait variables with discrete distribution, for which continuous underlying processes can be assumed. These traits are often referred to as threshold traits. QTLs are believed to play an important role in the genetic mechanism underlying quantitative traits or complex discrete traits. In the past several decades, QTL
From: Methods in Molecular Biology, Vol. 455: Osteoporosis: Methods and Protocols Edited by Jennifer J. Westendorf © Humana Press, Totowa, NJ
203
204
D.-H. Xiong et al.
mapping has been greatly developed and widely used in various fields, such as medical genetics, livestock production, plant breeding and population genetics of model organisms. Particularly, searching for QTLs by means of genome-wide linkage and/or association scans has become a major method to identify the genes underlying complex traits/diseases in humans. The aim of QTL mapping is to estimate positions and effects of the unobserved QTL through the observed quantitative traits and markers, and finally to estimate the variance components and the heritability of the quantitative traits. The designs of such mapping experiment involve choosing the appropriate crossing types, the parental strains, and phenotyping and genotyping strategies. Sound experimental designs help maximize the statistical power of QTL mapping. According to the distinct characteristics of natural populations and inbred lines, different experimental design strategies need to be taken for mapping QTLs underlying complex traits of interest. To detect the effect of QTLs in natural populations, it is necessary to generate linkage disequilibrium. In inbred lines, we need to employ the reasonable crossing designs to identify the effects of the chromosomal regions linked to the genetic markers.
1.2
Methods for QTL Mapping in Humans: Linkage Analysis
Related individuals are likely to share genetic loci controlling the variation of quantitative traits (i.e., blood pressure, body mass index, bone mass density, etc.). If a marker locus is linked to a QTL, the sharing pattern would be similar. Model-free linkage methods are commonly used in QTL mapping because there is no need to assume the underlying inheritance model of the traits of interest. A popular modelfree linkage method is to exploit the inverse relationship between the difference of trait values within sib-pairs and their marker identity-by-descent (IBD) scores (3).
1.3
Methods for QTL Fine Mapping in Humans: Association Mapping
1.3.1
What Is the Association?
Genetic linkage studies based on pedigree data have limited resolution because of the relatively small number of meiosis events within the general pedigrees. Association mapping (also called linkage disequilibrium mapping) can be used to further narrow down the linkage region. The classical definition of linkage disequilibrium (LD) refers to the non-random association of alleles at tightly linked loci. Consider two markers that were on the same chromosome. We designate the corresponding alleles and frequencies as A/a (pa.pa), B/b (pb, pb). Four haplotypes consisting of these two markers are possible: AB, Ab, aB, and ab. If the frequencies of alleles A, a, B, and b in the
16 QTL Mapping
205
population were all 0.5 and these two loci were in linkage equilibrium, then we would expect the frequencies of each of the four haplotypes in the population to be 0.25. However, the significant deviation of the haplotype frequencies from 0.25 suggested the LD between the two loci. 1.3.2
Measures of LD
One measure of LD is D, calculated as (4): D = p(AB) × p(ab) – p(Ab) × p(aB) where p(AB) is the frequency of the AB haplotype in the population, and likewise for the other haplotypes. In practice, it is the standardized D—namely D′—that is widely used: D' = D/Dmax where Dmax is the maximum value of |p(AB) – p(A) × p(B)|. D′ varies between 0 (no LD) and ±1 (complete association) and is less dependent on the allele frequencies than D. As a rule of thumb, D′>0.33 is often taken as the threshold level of LD above which associations are considered as apparent. Hill et al. (5) proposed another statistic, r2, which is: r2 =
D2 PA Pa PB Pb
where pA is the frequency of the A allele in the population, and likewise for the other alleles in the population. The r2 parameter can also be calculated for two loci with multiple (more than two) alleles as follows, which is usually necessary for microsatellite markers: n m
r2 = ∑ ∑ i
j
Dij2 Pi Pj
where locus 1 has n alleles, locus 2 has m alleles, pi is the frequency of allele i of locus one, pj is the frequency of allele j of locus two, and Dij = pij – pi × pj, with pij being the frequency of haplotype ij.
1.3.3
Tagging SNP and Haplotype
Originally, haplotype was defined as “a set of alleles of closely linked loci on a chromosome that tend to be inherited together.” In the single nucleotide polymorphism (SNP) era, the haplotype concept is based on SNPs and closely related to the LD concept. There is remarkable genomic variability in LD across the human genome
206
D.-H. Xiong et al.
with a series of high LD regions separated by short discrete segments of very low LD. Those high LD regions usually exhibit limited haplotype diversity, so that a small number of haplotypes account for most of the chromosomes in such genomic regions (haplotype blocks) in a population. Thus far, there are a number of ways to define haplotype blocks (6, 7). For example, Gabriel et al. (7) adopted the standardized gametic disequilibrium coefficient D′ to identify the high LD regions. Within these regions, allelic dependence made it unnecessary to study the redundant SNPs. Only the representative SNPs such as tag SNPs need to be genotyped and analyzed to make the association studies much more cost-effective while maintaining the statistical power at the same time. In association studies, both SNP markers and haplotypes constructed by individual SNPs can be used. Whether a specific method is more effective depends on whether the true causal allele is in higher LD and has more closely matched allele frequency with the marker set that method utilizes.
1.4 Multiple Traits and Interaction Analysis Genetic multivariate analyses in which the correlations between the phenotypes are explicitly modeled can provide greater statistical power to identify quantitative trait loci (QTLs) whose effects are too small to be detected by univariate analysis of the individual traits (8). The multivariate framework also serves to formally integrate results from separate univariate analysis. Joint genetic linkage analysis of multiple correlated traits has been shown to improve the power to detect, localize, and estimate the effect of possible QTLs or disease susceptibility loci (9, 10). Pleiotropic effects or close-linkage loci that each influence different traits can also be identified by joint genetic analysis (11). Epistasis, defined as the interaction between different genes, has become a hot topic in complex disease genetics in recent years. For complex traits, such as diabetes, asthma, and osteoporosis, the search for susceptibility loci has, to date, been less successful than for simple mendelian disorders. Part of this complexity of mapping complex disease genes can be attributed to epistasis or gene–gene interactions (12). It is hoped that the identification of the mode of gene interaction will facilitate understanding of the pathological mechanisms involved in complex diseases.
1.5
Microarrays and QTLs
1.5.1
Expression Quantitative Trait Loci Analyses
In the year of 2001, Jansen (13) firstly proposed the concept of “genetical genomics,” aimed at merging the genetic map and genomics to unravel the function and interrelation of gene products and gene action from expression profiling, such as gene expression, protein, enzymatic, and metabolite levels. The experiments in
16 QTL Mapping
207
yeast, plant, mouse, and human exemplified the merit of this kind of combination (14, 15). For example, through analyzing the co-segregation of eQTL with clinical QTL, Yaguchi (16) identified four candidate genes from 106 genes in a QTL spanning 10 Mb. Schadt (15) found two independent genomic regions for fat-pad-mass trait according to a hint from two separate gene expression clusters. 1.5.2
Whole Genome Association
With the near completion of the International HapMap Project and the rapid improvements in microarray technology, there comes the era of whole genome association (WGA), in which a dense set of SNPs across the genome is genotyped to survey the most common genetic variation for a role in complex diseases or traits. Because no assumptions are made about the genomic location of the causal variants, this approach could exploit the strengths of association studies without having to guess the identity of the causal genes. Therefore, the WGA approach represents an unbiased yet fairly comprehensive option that can be attempted even in the absence of convincing evidence regarding the function or location of the causal genes (17). To date it has been used to find causal genes and/or polymorphisms underlying complex diseases or traits such as rheumatoid arthritis (RA), lung cancer, alcoholism, and obesity (18–23). It can be imagined that the WGA approach will also be applied to dissect the causative genetic variants influencing other complex diseases in the near future.
2 2.1
Materials Linkage Study
The necessary material for mapping QTLs is a linkage map of polymorphic marker loci that adequately covers the whole genome. There are several criteria for the marker loci: (1) highly polymorphic, so that individuals are likely to carry different alleles at each locus; (2) abundant to cover the genome comprehensively; (3) neutral, both with respect to the quantitative trait of interest and to reproductive fitness; and (4) co-dominant, that all possible genotypes at a marker locus can be identified phenotypically. The final criterion is less stringent, as dominant/recessive markers can also be used in some designs. The most frequently used marker for QTL mapping is microsatellite marker.
2.2
Association Study
In association studies, single nucleotide polymorphisms (SNPs) are the most widely used markers. SNPs can be classified as noncoding or coding SNPs. Non-coding SNPs are generally located in introns, non-transcribed or untranslated regions, and
208
D.-H. Xiong et al.
the intergenic regions. Coding SNPs may cause non-synonymous or synonymous mutations, depending on whether there is any associated amino acid change. Various systems have been developed to meet the challenge of genotyping a large number of SNPs for disease association studies (24). The features of SNP genotyping methods are: (1) endonuclease cleavage (e.g., restriction site analysis and invader assay), (2) primer extension (e.g., SNPstream, multiplex mini-sequencing, pyrosequencing, MassArray, fluorescence polarization, SnaPShot, etc.), (3) allele-specific PCR (e.g., FRET primers, AlphaScreen, intercalating dye, etc.), (4) oligonucleotide ligation (e.g., OLA (oligonucleotide ligation assay), microarray ligation, ligase chain reaction, padlock probes, etc.), and (5) hybridization methods (e.g., Genechip microarrays, TaqMan, Molecular Beacons, etc.). For genome-wide association studies requiring hundreds of thousands of SNP markers, highly multiplexed and high-density microarray–based SNP genotyping systems (e.g., GeneChip 500 K, BeadArray, SNPstream, etc.) have been rapidly developed, offering the potential of genome-wide SNP mapping of genes involved in complex diseases in the near future (25).
2.3
eQTL Mapping
Generally, there are two types of microarray technologies: in situ synthesized oligonucleotide microarray (developed by companies such as Affymetrix, Agilent, and Febit AG) and spotted array based on the physical deposition of biomolecules (typically oligonucleotides or cDNAs) in glass or nylon membrane matrices. In Affymetrix, three categories of expression analysis arrays (for genes, transcripts, and genomes, respectively) are available at http://www.affymetrix.com/products/ arrays/index.affx. 2.3.1
Instruments
1. Agilent 2100 Bioanalyzer (Agilent, Palo Alto, CA). 2. Hybridization Oven 640 (Affymetrix, Santa Clara, CA). 3. High-solution Laser scanner: GeneChip Scanner 3000 (Affymetrix), Axon 4000a scanner (Axon Instruments). 4. GeneChip Fluidics Station 450 (Affymetrix). 5. GeneAmp PCR System 9700 and 7900HT Fast Real-Time PCR System (Applied Biosystems, Foster, CA). 2.3.2
Total RNA/Poly-A mRNA Isolation from Mammalian Cells
1. RNeasy Mini Kit (Qiagen, Valencia, CA) for total RNA isolation, or Oligotex Direct mRNA kit (Qiagen) for isolating mRNA directly from mammalian cells, or Oligotex mRNA kit (Qiagen) for separating mRNA from total RNA.
16 QTL Mapping
2. 3. 4. 5. 6.
209
QIAshredder for simple and rapid homogenization of cell lysates (Qiagen). RNase inhibitor, 20 U/µL (Ambion, Austin, TX). RNase-free DNase set (Qiagen). DEPC-treated water (Ambion). RNase zap (Ambion).
2.3.3
Quantification of RNA/cRNA
1. RNA 6000 ladder (Ambion). 2. RNA 6000 Nano kit (Agilent, Palo Alto, CA). 2.3.4
cDNA Synthesis
1. Poly-A RNA control kit (Affymetrix). 2. Sample cleanup module (Affymetrix). 3. One-cycle cDNA synthesis kit for 1–15 µg of total RNA or 0.2–2 µg of mRNA, or two-cycle cDNA synthesis kit for 10–100 ng of total RNA (Affymetrix, Santa Clara, CA). 4. Cy-labeled dNTPs for spotted array (GE Healthcare). 5. Biotin-labeled cRNA synthesis. 6. GeneChip expression 3′-amplification reagent for IVT labeling (Affymetrix). 7. Sample cleanup module (Affymetrix). 2.3.5
Hybridization, Wash, and Stain
1. Hybridization control kit (Affymetrix). 2. GeneChip hybridization, wash, and stain kit (Affymetrix or home made). 3. Wash buffer A and B (Affymetrix or home made). 2.3.6 1. 2. 3. 4.
RT-PCR
TaqMan gold RT-PCR Kit (Applied Biosystems, Foster, CA). TaqMan gene expression assays (Applied Biosystems). Housekeeping gene control (Applied Biosystems). AmpliTaq Gold PCR Master Mix (Applied Biosystems).
2.4
Whole Genome Association
In whole-genome genotyping, currently there are two most representative genotyping platforms: Illumina Sentrix HumanHap550 Genotyping Beadchip, and Affymetrix
210
D.-H. Xiong et al.
GeneChip500 k. Both allow genotyping of >500,000 SNPs across the human genome. According to the HapMap Project, the characterization of 300,000–500,000 tag SNPs is sufficient to provide good genomic coverage of all possible common polymorphisms in various ethnic groups. The Sentrix HumanHap550 Genotyping BeadChip (Illumine, San Diego, CA) contains over 555 K SNP loci with high-density tagSNP content that provides comprehensive genomic coverage across multiple populations. The GeneChip Human Mapping 500 K Array Set (Affymetrix) also enables highly powered whole-genome association studies. It is comprised of a set of two arrays that enable genotyping of >500,000 SNPs. One array uses the Nsp I restriction enzyme (~262,000 SNPs), whereas the second uses Sty I (~238,000 SNPs). Total genomic DNA with the concentration of 50 ng/µL as determined by PicoGreen is required for this 500 K array set. We will elaborate the materials and major steps of WGA analysis using GeneChip human mapping 500 K array set.
2.4.1
Instruments for Genomic DNA Purification and Quantification
1. Sorvall RT7 plus refrigerated centrifuge (Sorvall, Guelph, ON). 2. VWR Multi-Tube Vortexer (VWR, West Chester, PA). 3. SpectraMax Plus384 High-Throughput Microplate Spectrophotometer (Molecular Devices, Palo Alto, CA).
2.4.2
Instruments and Software for WGA Using Affymetrix GeneChip500 k Platform
1. Dual 96-Well GeneAmp PCR System 9700 (Applied Biosystems). 2. SpectraMax Plus384 High-Throughput Microplate Spectrophotometer (Molecular Devices). 3. Maxi-Mix* II Mixer (VWR). 4. QIAvac Multiwell Unit (Qiagen). 5. GeneChip Hybridization Oven 640 (Affymetrix). 6. GeneChip Operating Software (GCOS) (Affymetrix). 7. GeneChip Fluidics Station 450 (Affymetrix). 8. GeneChip Scanner 3000 7 G (Affymetrix). 9. GeneChip Genotyping Analysis Software (GTYPE) (Affymetrix).
2.4.3 DNA Purification and Quantification 1. Puregene DNA purification kit (Gentra) containing: a. RBC lysis solution. Store at room temperature. b. Cell lysis solution. Store at room temperature.
16 QTL Mapping
211
c. Protein precipitation solution. Store at room temperature. d. RNase A solution. Store at room temperature. e. DNA hydration solution. Store at room temperature. 2. Quant-iT PicoGreen dsDNA Assay Kit (Invitrogen) containing: a. Quant-iT PicoGreen dsDNA reagent is dissolved at dimethylsulfoxide (DMSO) and stored at −20 °C. b. 20× TE: 200 mM Tris-HCl, 20 mM EDTA, PH7.5. Store at room temperature. c. Lambda DNA standard with the concentration of 100 µg/mL. Store at 2–6 °C.
2.4.4
WGA Using Affymetrix GeneChip500 k Platform
1. Affymetrix GeneChip Mapping 250 K Nsp and/or Sty Assay Kit (Affymetrix, Santa Clara, CA), containing: a. b. c. d. e. f. g. h. i.
Adaptor Nsp or Sty, 50 µM. DNA labeling reagent, 30 mM. Fragmentation reagent. Oligo control reagent, 0100. PCR primer 002, 100 µM. Terminal deoxynucleotidyl transferase. 5× terminal deoxynucleotidyl transferase buffer. Reference genomic DNA 103 (use as a positive control). 10× fragmentation buffer.
2. Sty I (10,000 U/mL) and NE buffer 3 (New England Biolabs). 3. Nsp I (10,000 U/mL) and NE buffer 2 (New England Biolabs). 4. BSA (bovine serum albumin). 5. T4 DNA ligase. 6. T4 DNA ligase buffer. 7. G-C melt (5 M) (Clontech). 8. Molecular grade biology water. 9. dNTP (2.5 mM each). 10. Clontech TITANIUM Taq DNA polymerase (50×) (Clontech): a. 50× Clontech TITANIUM Taq DNA polymerase. b. 10× Clontech TITANIUM Taq PCR buffer. 11. 2% TBE Gel:BMA reliant precast (2% SeaKem Gold) (Cambrex, East Rutherford, NJ). 12. Gel loading solution (Sigma). 13. All purpose Hi-Lo DNA Marker (Bionexus). 14. 5 M TMACL (Tetramethyl Ammonium Chloride) (Sigma). 15. 10% Tween-20, diluted to 3% in molecular biology grade water. 16. MES hydrate SigmaUltra (Sigma). 17. MES sodium salt.
212
D.-H. Xiong et al.
18. 19. 20. 21. 22. 23. 24. 25. 26.
DMSO. 0.5 M EDTA. Denhardt’s solution (Sigma). HSDNA (herring sperm DNA). Human Cot-1 DNA. 20× SSPE. SAPE (streptavidin, R-phycoerythrin conjugate) (Invitrogen). Biotinylated anti-streptavidin antibody (Vector Laboratories). Distilled water.
3
Methods
3.1
Experimental Design for QTL Mapping
3.1.1
Experimental Design for Model Organism Population
QTL mapping results in model organisms such as inbred mouse lines can be applied to the mapping of QTL or disease susceptibility loci in humans through comparative genomics approaches (26). Two experimental designs are often used for QTL mapping in inbred lines, i.e., back cross design and F2 design. Both of these designs usually start with a cross between two inbred lines from the F1 population. An alternative to the cross between inbred lines is to study natural outbred populations such as the human populations.
3.1.1.1
Back Cross
Starting with two completely inbred parental lines, P1 and P2, the back cross (BC) design examines marker/trait associations in the progeny formed by back crossing the F1 to one of the parental lines. The BC design is depicted (Fig. 16.1). Two parental lines have different marker and QTL genotypes. They are mated to form an F1. There is a general assumption with BC design that the two parental lines are homozygous for alternate alleles of both loci, so all F1 individuals have the same heterozygous genotypes in both marker locus and QTL. The F1 is then mated to one of the parental lines. The BC progeny can be classified into two groups according to the recombination status (Fig. 16.1). There are only two marker genotype groups for the BC design, i.e., individuals with marker genotype M 2M2 and those with M1M 2 (Fig. 16.1). The difference between the means of these two progeny marker genotype groups can be calculated to represent genotypic effect. The significance of such genotypic effect can be tested by traditional methods such as analysis of variance (ANOVA),
16 QTL Mapping
213
Fig. 16.1 The back cross design, r denotes the recombination rate between mark locus and QTL.
or t-test. Under the null hypothesis that there is no QTL linked to the marker, the ratio of the marker mean-squares to the residual mean-squares follows a central F-distribution when we perform ANOVA. The significant deviation of such statistic from the central F-distribution is the evidence that a QTL is linked to the marker. The t statistic can be derived as the ratio of the difference between the means of two genotype groups to the standard error of this difference. Such t-statistic will have a central t-distribution under the null hypothesis. Both of the preceding two tests can be easily conducted by most statistical packages, such as SAS, SPSS, etc. The contrast between marker group means in a back cross gives an estimate of the additive QTL effect (a), which is unbiased only in absence of dominance (d = 0), and recessive or partly recessive QTL may not be detected. This problem can be overcome by backcrossing to both parental lines (1). Furthermore, the back cross design is usually less powerful than the F2 design because it only detects differences between heterozygotes and one of the homozygotes when performing one back cross between F1 and the parental line.
214
3.1.1.2
D.-H. Xiong et al.
F2 Design
F2 design is depicted in Fig. 16.2. Firstly as in BC design, two homozygous inbred parents are mated to form a heterozygous F1. Then the F1 individuals are mated with each other to generate F2 progeny that include three non-recombinants and seven recombinants. The F2 design detects marker/trait associations in the progeny from an intercross mating of the F1 individuals, as three genotypes are generated at each marker locus, the estimation of the degree of dominance associated with the detected QTL is possible in the F2 design.
Fig. 16.2 The F2 design.
16 QTL Mapping
215
Similarly, significance of a segregating QTL can be detected by ANOVA, including all three marker genotypes. Additionally, comparisons between different marker genotypes can be tested by t-test. The main difference of interest is that between the means of the two distinct homozygotes. Note that, although it is possible to test for significance of additive and dominance effects, the test statistic is always the function of recombination rate, additive effect and/or dominance effect, i.e., QTL effect and recombination rate are always confounded. This concern is also present in BC design. Although F2 and back cross are the most widely used designs, other line crosses can offer further advantages. For example, the F1 can be used to create recombinant inbred lines (RIL) and doubled-haploid lines (DHL), which allow marker–trait associations to be scored in a completely homozygous background and across multiple environments (27).
3.1.2
Experiment Design for Human Population
Designs using inbred lines are frequently used in laboratory animals (e.g., mice) and plants. However, in human populations, inbred lines are unavailable. Instead, experimental designs should be based on the analysis of families within existing populations. Two main types of experimental designs are commonly used for QTL mapping studies in human populations, including: (1) sib-pair design for analysis of many small nuclear full-sib families based on a single generation; and (2) general pedigree design for analysis of large families with complex pedigree structures based on multiple generations.
3.1.2.1
Sib-pair
Sib-pair design in human outbred population was first proposed by Haseman and Elston in 1972 (28). This method is most suitable for human populations, which include many small full-sib families. The main idea of this method is illustrated in Fig. 16.3. We consider the simplest case to introduce this design, i.e., assuming each parent has different alleles in both the marker locus and the linked QTL locus denoted by M1, M2, M3, M4, and Q1, Q2, Q3, Q4, respectively. For simplicity, only three kinds of nonrecombinant sib-pairs are shown in Fig. 16.3. In sib-pair 1, both individuals’ marker alleles are identical by descent (IBD). In sib-pair 2, only M 1 alleles carried by two individuals are paternal IBD. In sib-pair 3, none of the marker alleles are IBD. According to the assumption, sibpairs with more marker loci IBD should also have higher probability of IBD in QTL and hence are expected to be more similar than sib-pairs sharing no IBD marker alleles.
216
D.-H. Xiong et al.
Fig. 16.3 The full sib-pair design.
Because QTL genotypes cannot be directly observed, in theory, we can detect the linked QTL compare means of different marker genotypes as BC or F2 design. However, these comparisons are not available because the differences between different marker genotypes maybe show negative or positive in different families. In order to resolve this issue, Haseman and Elston (28) proposed a regression model based on squared differences. They defined: x1 j = m + g1 j + e1 j x2 j = m + g2 j + e2 j where x1j and x2j are phenotypes for sibs 1 and 2 of family j, m is the general mean, gij and eij are the QTL and residual effects respectively for sib i.
16 QTL Mapping
217
Then let Yj = (x1j – X2j)2 = (g1j + e1j – g2j – e2j)2. If the two sibs share IBD of both QTL alleles, then g1j = g2j and so Yj = (e1j – e2j)2, which will be less than Yj for other sibs sharing no allele IBD or sharing only one allele IBD of QTL. According to this relationship between Yj’s, presence of a linked QTL can be detected by the following regression: Y j = a + bp j + e j where pj is the proportion of alleles IBD at the marker of interest between sib-pair, a is the y-intercept and b is the regression coefficient which will have a negative value. Even with incomplete linkage between the marker and the QTL, b will still be negative. Under null hypothesis, b should not be significantly different from zero. Through the squared sib-pair trait difference regressed on the proportion of marker alleles, the sib-pairs are estimated to share identity by descent: A significantly negative regression coefficient suggests linkage between the marker and a QTL. We need to note that, in this regression model, when both parents have complete marker informative, pj can be inferred unequivocally. However, it is impossible for all families in which both parents are fully informative in all marker loci. In the case that parents cannot provide complete marker information, pj can be estimate as fj2 + 0.5fj1, where fj2 and fj1 are the probabilities that sib-pair j has one or two marker alleles IBD, respectively.
3.1.2.2
General Pedigree
In general pedigreed design, mixed model based on variance components estimation is the popular method. Under this model, QTL mapping by variance components in general pedigrees has been developed independently by both the human genetics (29) and animal breeding (30) communities. Consider the “polygenic + QTL” model used to estimate additive genetic variance and QTL variance in a population with general pedigree. Using matrix algebra form, the animal model can be written: y = Xβ + Za + Zq + e where y is a vector of phenotypes of all pedigreed individuals, β is a vector of fixed effects, X is a design matrix relating the appropriate fixed effects to each individual, a is a vector of random effects (polygenic additive effects), q is a vector of additive QTL effects, Z is an incidence matrix relating the appropriate random effects to each individual, and e is a vector of residual errors. We depict the general pedigree design in Fig. 16.4. For any pair of individuals in the pedigree, the genetic covariance between them is a function of 2Θij and Rij. where Qij is the coefficient of co-ancestry, the probability that an allele randomly drawn from individual i is IBD with an allele randomly drawn from individual j.
218
D.-H. Xiong et al.
Fig. 16.4 QTL mapping in general pedigrees. The pedigree contains 10 individuals denoted by number 1, 2, …, 10. All of individuals are typed at a marker locus with four alleles A, B, C, and D. 2Θij (twice the coefficient of coancestry) and Rij (the IBD coefficient at the marker) between each pair of individuals is shown.
Note that the coefficient of co-ancestry is obtained from the pedigree structure of the individuals concerned instead of other information, such as marker genotypes. Rij is the proportion of alleles that two individuals i and j actually share IBD at a chromosomal location. Two alleles are IBD if we can trace them back to a single copy in a common ancestor. To obtain an REML solution of this model, marker data are used to infer Rij. Using all markers on a chromosome to estimate Rij at each location is known as multipoint mapping. It should be pointed out that Rij is an estimate rather than a probability, which varies at each test location.
16 QTL Mapping
219
Through this model, we can obtain estimates of the additive genetic variance, the variance attributable to a QTL at the location considered and the likelihood of of the REML solution (l1). A likelihood ratio test statistic (LRT) can be used to perform chi-square test as follows: LRT = – 2ln(l0 –l1) Under the null hypothesis of no QTL at the test location the test statistic follows a χ2 distribution (29, 30). Under the different null hypothesis of no QTL anywhere on the chromosome, the test statistic appears to approximate to a χ2 distribution with 1 degree of freedom (30). Using general pedigree design, mapping QTL utilizes more pedigree information than sib-pair designs, and hence appears more powerful to detect QTL (30, 31).
3.2
QTL Mapping in Humans: Linkage Analysis
3.2.1
Regression Analysis
Regression tries to predict a variable (Y) in terms of another variable (X) via a functional relationship Y = f(X). The functional relationship f could be linear (Y = a + bX), quadratic (Y = a + bX + cX2), logarithmic (Y = a log X), or more complex relationships. The simplest one is linear regression. The statistical problem of linear regression is to determine the “best” values of a and b that minimize the deviations from the linear relationship. This is done by “the least squares errors” method. Since some of the errors are positive and some are negative, minimizing the sum of errors (Σei) is not optimal. It is better to minimize the sum of squared errors (Σei2), that is (Σ(Yi – a – bXi)2), to minimize with respect to a, b. The best values: b0 = Cov (X,Y ) / Var (X ), a0 = Mean(Y ) – b0 Mean(X ) Haseman and Elston (1972) developed a regression approach for detecting linkage based on the squared difference in quantitative trait values of sib-pairs (Y) and their ∧ estimated marker IBD scores (p m ). The basis of the regression is the equation: ∧
∧
E (Y⎥ p m ) = a + b p m ,
(see Note 1)
where there is no dominance in the trait and b = −2p(1 − p)a2 (1 − 2q)2; p being the allele frequency of A1, a the conditional expectation of the trait given genotype A1A1, and q the recombination fraction between the QTL and the marker locus. Function of q, a test for no linkage (i.e., q = 0.5) is equivalent to testing b = 0 in Equation (1). The test can be performed via the usual t statistic based on the least squares regression estimate of b.
220
D.-H. Xiong et al.
Based on the traditional Haseman-Elston method, many extensions have been made so that it can be used for linkage analysis in more scenarios. First, Amos and Elston extended the above regression procedure to other relative pairs (32). Then, Fulker and Cardon applied the Haseman-Elston regression equation to interval mapping (33). Tiwari and Elston extended the traditional Haseman-Elston procedure to the case of two unlinked QTLs which might interact epistatically (34). Recently, Elston et al. (3) suggested that the mean-corrected cross-product of the sib-pair trait values carry more linkage information than the squared sib-pair trait difference.
3.2.2
Variance Components Analysis
Another popular statistical approach for QTL mapping is to dissect the genetic variation within the quantitative trait. Although the methods assume specific probability distributions for trait values, the advantage of using these methods is that larger sibships or entire pedigrees can be simultaneously analyzed (35). The general variance components model is given by: Y=m+g+G+e where m is the overall mean of the quantitative trait, g is a random effect due to a major gene with additive variance sa2 and dominance variance sd2, G is a random polygenic effect with variance sG2 and e is the non-shared environmental effect (or random error) with variance se2. The trait values of individuals in a pedigree are usually assumed to be distributed as multivariate normal with dispersion matrix V, where the variance of the trait value of each individual is sa2 + sg2 + sG2 + se2 and the covariance between the trait values of two individuals is given by fσa2 + ∆sd2 + fsG2, where f is the coefficient of relationship between the two individuals and ∆ is the probability that the two individuals share both their alleles IBD at the major locus. Conditioning on IBD score (p) at a marker locus, the preceding covariance is given by: f (q , p )s a2 + g(q , ∆ )s d2 + fs G2 , where q is the recombination fraction between the QTL and the marker locus. The log-likelihood of the data is given by: c−
1 1 log | V | − ∑ (Y − m1)′V −1 (Y − m1), ∑ 2 P 2 P
where c is a constant, Y and m1 are, respectively, the vector of trait values and that of the means within a pedigree and the summation is over independent pedigrees. The variance components methods use the maximum likelihood method to estimate the parameters. The test for linkage is equivalent to testing sa2 = 0 versus sa2 > 0.
16 QTL Mapping
221
The usual likelihood ratio test statistic is asymptotically distributed as a 1– : 1– mixture 2 2 of a χ12 variable and a point mass at zero. Based on the basic framework, Almasy and Blangero expanded the variance components model to calculate multipoint IBD probability using pedigrees of arbitrary size. They also extended the model to incorporate multiple QTLs. In addition, the model can also incorporate other environmental co-variates, gene by environment interaction terms, epistatic effects and so on.
3.3
QTL Fine Mapping in Humans: Association Mapping
3.3.1
Population-Based Association Mapping
Genetic association studies aim to correlate differences in between-group disease frequencies (or trait values for continuous variables) with differences in allele frequencies at candidate loci. The association study adopting a random sample chosen from a population is usually called population-based association study. The populationbase design is often more powerful and feasible to be conducted than the family-based design. However, this approach is prone to the confounding effect of population stratification that refers to differences in allele frequencies between cases and controls due to systematic differences in ancestry rather than association of genes with disease (see Note 2). Some solutions, such as genomic control (GC) and structured association (SA) have been developed and widely used to correct for the stratification confounder (36–38).
3.3.2
Transmission-Disequilibrium Test
The most popular method for family-based design is the transmission disequilibrium test (TDT) (39). TDT is carried out with data on transmission of marker alleles from heterozygous parents for that marker to affected offspring (case). In TDT, cases’ genotypes are compared to those of their parents to explore whether a specific allele, or marker, at a locus of interest appears to be transmitted in excess of what is expected on the basis of mendelian inheritance. Such apparent excess transmission indicates that cases are being selected for that allele, thereby providing evidence that this allele is a risk factor for disease. The test is unaffected by population stratification. The general procedure to conduct such study is as follows: 1. Affected probands are ascertained. 2. The probands and their parents are typed for the marker. 3. Those parents who are heterozygous for marker M1 are selected. They may or may not be affected. Let a be the number of times a heterozygous parent transmits M1 to the affected offspring, and b be the number of times the other allele is transmitted.
222
D.-H. Xiong et al.
The TDT statistic is (a − b)2/(a + b). It follows a χ2 distribution with 1 degree of freedom, provided the numbers are reasonably large. Whatever association test is used, the problem of multiple testing must be addressed. A full Bonferroni correction (dividing the nominal p value by N, the total number of individual tests, to obtain an experiment-wide threshold) is overconservative for large values of N. Thus other methods, such as false discovery rate (FDR) (40) and permutation test (41), were proposed to solve this problem. FDR method controls the expected proportion of incorrectly rejected type I errors in a list of rejected hypotheses. It is a less conservative comparison procedure with optimal power at a cost of increasing the likelihood of obtaining type I errors. The permutation test uses a Monte Carlo procedure to provide a global assessment of significance. It is more powerful than the Bonferroni correction procedure when alleles at linked markers are associated.
3.4
Multiple Traits and Interaction Analysis
3.4.1
Principal Component Analysis
It has been suggested that evaluating multiple phenotypes jointly is more powerful than using of each interrelated phenotypes separately (42). One of the most commonly used methods is principal component analysis (PCA), which can be performed by classic statistical package SPSS or SAS. PCA uses the information from the entire correlation coefficient matrix of some phenotypes to produce a smaller number of hypothetical factors (components) that help explain correlations among the original variables. PCA includes two stages: (1) factor extraction using PCA, and (2) rotation of the principal components using varimax option. Factors are defined by extracting the eigenvalues (which represent the variance explained by each of the principal components) >1.0. The results of PCA generate a new uncorrelated combination of the original phenotypes with a reduction of the number of variables. The principal component scores (the correlation coefficients between the variables and the factors) are extracted as “new” phenotypes representing the original phenotypes for other statistical genetic analysis.
3.4.2
Multivariate Linkage Analysis
In univariate linkage analyses, a variance component approach is widely used (29, 35), which is based on specifying expected phenotypic covariances between relatives as a function of the IBD relationships at one or more genetic loci. Briefly, the covariance matrix for a pedigree is given by: Ω = Πσ 2m + 2Φσ 2g + Iσ 2e
16 QTL Mapping
223
where smij2 is the additive genetic variance caused by the major locus, Π is a matrix whose elements (πmij) provide the probability that individuals i and j share IBD at a QTL linked to a genetic marker locus; P is a function of the estimated IBD matrix for the genetic marker itself and a matrix of the correlations between the proportions of genes IBD at the marker and at the QTL. sg2 is the genetic variance due to residual additive genetic factors, F is the kinship matrix for the entire pedigree, Se2 is the variance due to individual-specific environmental effects, and I is an identity matrix. By assuming underlying multivariate normality of the trait within pedigrees, the likelihood of any pedigree can be easily written and numerical procedures can be used to estimate the variance component parameters. The basic variance component model can extend to a multivariate framework (see Note 3). In the multivariate linkage model, the phenotype covariance is further decomposed to include the genetic correlation between traits caused by additive genetic effects and the shared effects of the QTLs, such that the covariation between two individuals for two traits is given by: Ω=
Ω 11Ω 12 Ω 12 Ω22
where Ω is a covariance matrix of 2×2 covariance matrices, where the elements are defined by: Ωab = Πmsmasmb + 2Φrgs gasgb + I reseaseb where a and b are 1 or 2, and rg and re are the additive genetic and environmental correlations between the two traits, respectively. The genetic correlation estimates the proportion of genes shared in common between the two traits. If a = b, then rg is 1, and the covariance of a pair of relatives simplifies to Equation 1. In this model, the marker-specific genetic correlation (rm) is assumed to be 1. This approach can be conducted in program SOLAR (43).
3.4.3
Epistasis and Gene–Environment Interaction Analysis
Methods for the detection of epistasis vary according to whether one is performing linkage or association analysis, and dealing with a quantitative or a qualitative trait (see Note 4). Here, we focus on quantitative traits, as most surrogate phenotypes for osteoporosis are quantitative. For genetic linkage analysis, one popular method is the variance component model. Test for epistasis uses an extension of the variance component model (43), which can be written as: Ω = Πm1σ 2 m1 + Π m 2 σ 2 m 2 + Π m1 m 2 σ 2 m1 m 2 +2Φσ 2g + Ισ e2
224
D.-H. Xiong et al.
2 2 where Sm1 and Sm2 are the additive genetic variance due to the observed marker m1 and m2, Pm1 and Pm2 are the matrixes whose elements provide the probability that any two individuals i and j share IBD for the observed genetic marker m1 and m2, 2 is epistatic component, and Pm1m2 is the element by element matrix multiplicaSm1m2 tion of Pm1 and Pm2 serves as the coefficient for the epistatic component. This model can be implemented in program SOLAR (43) by comparing it with models that do not contain these components using maximum-likelihood methods. Another popular method of linkage analysis is the Haseman-Elston method (28) and its extensions (3), which is based on a regression model. The squared difference or trait similarity between the members of a pair of relatives in a regression equation is a function of the underlying IBD sharing probabilities. To estimate epistasis between loci, it is required to include IBD sharing probabilities at different loci as predictors in the regression framework. In the quantitative genetics literature, the general epistatic model is a standard multiple linear (as opposed to logistic) regression equation (44):
y = m + a1x1 + d1z1 + a2x2 + d2z2 + iaax1x2 + iad x1z2 + idaz1x2 + idd z1z2 where y is a quantitative phenotype of interest, xi and zi are dummy variables related to the underlying genotype at locus i. The coefficients µ, a1, d1, a2, d2 represent genetic parameters to be estimated corresponding to the mean effect and additive and dominance effects at loci 1 and 2, whereas iaa, iad, ida, and idd correspond to epistatic interaction effects. An advantage of this formulation is that, provided the parameter estimates are almost independent, we can test the epistasis between specific alleles at the two loci rather than merely testing the overall existence of any interaction between two loci.
3.5
Using Microarrays in QTL Mapping
3.5.1
eQTL Mapping: Measuring Gene Expression
1. Extract mRNA from the samples (see Note 5). For oligonucleotide microarrays, mRNA should be converted to cDNA first, which in turn will be used to generate biotin-labeled cRNA. But for spotted microarrays, the labeled “target” could be cRNA or cDNA. 2. RNA extraction and quantification: RNA should be extracted using various kits according to different sample types (Qiagen). Usually the quantification of total RNA amount is done on Agilent 2100 Bioanalyzer (see Note 6). 3. Convert mRNA to cDNA using reverse transcriptase and a poly-T primer. (For the spotted array, the cDNA is synthesized with Cy-labeled dNTPs.) 4. Amplify resulting cDNA using T7 RNA polymerase in the presence of biotinUTP and biotin-CTP. Each cDNA will yield 50–100 copies of biotin-labeled cRNA.
16 QTL Mapping
225
5. Quantify the cRNA and determine the cRNA yield. 6. Fragment an appropriate amount of cRNA at 94 °C using fragmentation buffer to produce cRNA fragments of length 35–200 nucleotides. 7. Hybridize to chip using hybridization oven. 8. Wash away non-hybridized material and stain hybridized biotin-labeled cRNA with streptavidin-phycoerythrin and wash again. 9. Scan the chip in a high-resolution laser scanner. For oligo-nucleotide microarrays, absolute gene expression level is acquired. For spotted arrays, to which tested sample and the control was co-hybridized, expression values are based on the competitive hybridization of the two samples. 10. Validate microarray gene expression level with RT-PCR by reverse transcribing cDNA from RNA with reverse transcriptase (Applied Biosystems) and performing real-time quantitative PCR in a 50-µL reaction volume with ideal housekeeping gene control (Applied Biosystems).
3.5.2
Whole Genome Association
Similar to traditional candidate gene association approaches, both population-based and family-based designs are commonly used in whole genome association studies to locate genes and/or polymorphisms associated with complex diseases (see Note 7). Taking account for the high cost of whole genome association study and that many polymorphisms within a small genomic region may be in strong linkage disequilibrium (LD) with each other, the two-stage approach has been adopted, in which a subset of the data is used to select a proportion of promising SNPs from a larger set of all known polymorphic sites within a candidate gene or a candidate region, and only these promising SNPs are typed in additional subjects (45). Therefore, it is quite important to choose the sub-sample for promising SNP selection. Too large a sample obviously defeats the purpose of reducing costs by limiting the number of SNPs to be genotyped in additional subjects, whereas too small a sample could lead to unstable estimates of the parameters of a model that can be applied to the analysis of the data of the additional subjects. Nevertheless, as the cost of high-volume genotyping plummets and haplotype tagging information from the International HapMap project rapidly accumulates in public databases, such two-stage designs may soon become unnecessary.
3.5.2.1
Isolation Total Genomic DNA from Whole Blood
1. Add 5 mL of whole blood (or bone marrow) to a 50-mL centrifuge tube containing 15 mL of RBC lysis solution. Invert to mix and incubate 5 minutes at room temperature; invert again at least once during the incubation. 2. Centrifuge for 5 minutes at 2,000 g. Remove supernatant leaving behind the visible white cell pellet and about 200–400 µL of the residual liquid. Vortex the tube vigorously to resuspend the cells in the residual liquid.
226
D.-H. Xiong et al.
3. Add 5 mL of cell lysis solution to the resuspended cells and pipette up and down to lyse the cells. 4. Add 1.67 mL of protein precipitation solution to the cell lysate. 5. Vortex vigorously at high speed for 20 seconds to mix the protein precipitation solution uniformly with the cell lysate. Centrifuge at 2,000 g for 5 minutes. 6. Pour the supernatant containing the DNA (leaving behind the precipitated protein pellet) into a 15- or 50-mL tube containing 5 mL of 100% isopropanol (2-propanol). Mix the sample by inverting gently 50 times. 7. Centrifuge at 2,000 g for 3 minutes; the DNA will be visible as a small white pellet. 8. Pour off supernatant and drain tube briefly on clean absorbent paper. Add 5 mL of 70% ethanol and invert tube several times to wash the DNA pellet. Centrifuge at 2,000 g for 1 minute. Carefully pour off the ethanol. 9. Invert and drain the tube on clean absorbent paper and allow sample to air dry for 10–15 minutes. 10. Add 500 µL of DNA hydration solution. 11. Rehydrate DNA by incubating at 65 °C for 1 hour and overnight at room temperature. If possible, tap tube periodically to aid in dispersing the DNA. 12. For storage, sample may be centrifuged briefly and then transferred to a 1.5-mL tube. Store DNA at 4 °C. For long-term storage, store at −20 °C or −80 °C.
3.5.2.2
Quantitation of Genome dsDNA
1. Prepare a 1× TE working solution by diluting the concentrated buffer 20-fold with sterile, distilled, DNase-free water. 2. Prepare an aqueous working solution of the Quant-iT PicoGreen reagent by making a 1:200 dilution of the concentrated anhydrous dimethylsulfoxide (DMSO) solution in 1× TE. 3. Prepare a 2 µg/mL stock solution of bacteriophage lambda dsDNA in 1× TE. Determine the DNA concentration on the basis of absorbance at 260 nm (A260) in a cuvette with a 1-cm pathlength; an A260 of 0.04 corresponds to 2 µg/mL dsDNA solution. 4. Dilute the 2 µg/mL DNA stock solution into plastic test tubes as shown in the table that follows. Add 1.0 mL of the aqueous working solution of the Quant-iT PicoGreen Reagent to each plastic test tube. Mix well and incubate for 2–5 minutes at room temperature, protected from light. Vol (µL) DNA stock
Vol (µL) 1× TE
Vol (µL) PicoGreen working solution
Final DNA conc. in PicoGreen assay
1,000 100 10 1 0
0 900 990 999 1,000
1,000 1,000 1,000 1,000 1,000
1 µg/mL 100 ng/mL 10 ng/mL 1 ng/mL Blank
16 QTL Mapping
227
5. Measure sample fluorescence using the fluorescence microplate reader (excitation ~480 nm, emission ~520 nm). 6. Subtract the fluorescence value of the reagent blank from that of each of the samples. Use corrected data to generate a standard curve of fluorescence versus DNA concentration. 7. Add 1.0 mL of the aqueous working solution of the Quant-iT PicoGreen Reagent to each sample. Incubate 2–5 minutes at room temperature, protected from light. 8. Measure fluorescence of sample using instrument parameters that correspond to those used when generating standard curve (see step 5). 9. Subtract fluorescence value of reagent blank from that of each sample. Determine DNA concentration of the sample from standard curve.
3.5.2.3
Whole Genome Association Analysis
1. Dilute the DNA sample to 50 ng/µL using reduced EDTA:TE buffer and aliquot 5 µL of each sample per well to a 96-well plate. 2. Aliquot 14.75 µL of the Digestion Master Mix (1 µL of Nsp I or Sty I restriction enzyme, 2 µL of corresponding NE buffer, 2 µL of BSA, and 9.75 µL of H2O) to digest the genomic DNA. Put the plate on a thermal cycler and run the digest program (37 °C for 120 minutes, 65 °C for 20 minutes, and 4 °C hold). 3. Aliquot 5.25 µL of the ligation master mix (0.75 µL of adaptor Nsp I or adaptor Sty I, 2.5 µL of T4 DNA ligase buffer, 2 µL of T4 DNA ligase) into each digested DNA sample to ligate the digested samples. Put the plate on a thermal cycler and run the digest program (16 °C for 180 minutes, 70 °C for 20 minutes, and 4 °C hold). 4. Dilute each DNA ligation reaction by adding 75 µL of molecular biology-grade H2O per 25 µL ligated DNA. 5. Transfer 10 µL of diluted ligated DNA to the corresponding well of three reaction plates. Aliquot 90 µL of PCR Master Mix (39.5 µL of H2O, 10 µL of titanium Taq PCR buffer, 20 µL of GC-Melt, 14 µL of dNTP, 4.5 µL of PCR primer 002, and 2 µL of TITANIUM Taq DNA polymerase) to three reaction plates and perform PCR program using GeneAmp PCR system 9700 (one cycle of 94 °C for 3 minutes, 30 cycles of 94 °C for 30 seconds, 60 °C for 45 seconds, and 68 °C for 15 seconds, one cycle of 68 °C for 7 minutes, and 4 °C hold). 6. Run 3 µL of each PCR product mixed with 3 µL of 2× gel loading dye on 2% TBE gel at 120 V for 1 hour. Check the average size of ligated DNA between 200 and 1,100 bp. 7. Add 8 µL of 0.1 M EDTA to each PCR reaction. Consolidate three PCR reactions for each sample into one well of the clean-up plate. Apply a vacuum and maintain at ~600 mbar until the wells are completely dry by QIAvac multiwell unit. Wash the PCR products by adding 50 µL of molecular biology grade water and dry the wells completely.
228
D.-H. Xiong et al.
8. Add 45 µL of RB buffer to each well. Cover the plate with PCR plate cover film and moderately shake the clean-up plate on a plate shaker for 10 minutes at room temperature. Transfer the purified PCR product to a fresh plate. 9. Use spectrophotometric analysis to determine the purified PCR product yield. Add 2 µL of the purified PCR product to 198 µL molecular biology grade water and mix well. Read the absorbance at 260 nm. And 1 absorbance unit at 260 nm equals 50 µg/mL for double-stranded PCR product. 10. Transfer 90 µg of each of the purified DNA samples to a new plate. Bring the total volume of each well up to 45 µL by adding the appropriate volume of RB buffer. 11. Add 5 µL of 10× fragmentation buffer to each sample. Dilute the stock of fragmentation reagent to 0.05 U/µL, and add 5 µL of diluted fragmentation reagent to the fragmentation plate. Run the fragment program (37 °C for 35 minutes, 94 °C for 15 minutes, and 4 °C hold). 12. Dilute 4 µL of fragmented PCR product with 4 µL of gel loading dye and run on 4% TBE gel at 120 V for 30 minutes to 1 hour. Check the average size of fragment DNA <180 bp. 13. Aliquot 19.5 µL of labeling master mix (14 µL of 5× TdT buffer, 2 µL of GeneChip DNA labeling reagent, and 3.5 of µL TdT) into the fragmentation plate containing 50.5 µL of fragmented DNA samples to total 70 µL. Run the labeling program (37 °C for 4 hours, 94 °C for 15 minutes, and 4 °C hold). 14. Transfer each of the 70 µL of labeled samples from the plate to a 1.5-mL Eppendorf tube. Aliquot 190 µL of the hybridization cocktail master mix (12 µL of 12× MES, 13 µL of DMSO, 13 µL of Denhardt’s solution, 3 µL of EDTA, 3 µL of HSDNA, 2 µL of OCR 0100, 3 µL of human Cot-1 DNA (1 mg/ mL), 1 µL of Tween-20, and 140 µL of TMACL) into the labeled DNA samples. 15. Heat the 260 µL of hybridization mix and labeled DNA at 99 °C in a heat block for exactly 10 minutes to denature. Cool on crushed ice for 10 seconds. Place the tubes at 49 °C for 1 minute. Inject 200 µL of denatured hybridization cocktail into the array. Hybridize at 49 °C for 16–18 hours at 60 rpm. 16. Wash away non-hybridized material and stain hybridized biotin-labeled DNA with three stage processes consisting of a streptavidin phycoerythrin (SAPE) stain, followed by an antibody amplification step and final stain with SAPE using the fluidics station 450. 17. Scan the chip in the GeneChip Scanner 3000 7G scanner. 18. Get the genotyping file based experimental data. 19. Perform the comprehensive statistical analyses for the association study. Various statistical tools for the whole genome association are now available. The large-scale datasets produced by the high throughout arrays require computationally efficient manner of the tools. As for family-based studies, most methods based on a generalization of the transmission disequilibrium test and assesses whether the over- or under-transmission of an allele is correlated with offspring phenotypes, such as PLINK and Helixtree-PBAT, although most statistical tools for populationbased studies can be performed on PLINK and HelixTree.
16 QTL Mapping
3.6
229
Software Review
QTL linkage mapping is commonly based on regression or variance-components methods. The sib-pair Haseman-Elston regression method is available in GENEHUNTER (46) with heuristic adjustments for general pedigrees. The extension to more general pedigrees is implemented in MERLIN (47). This approach now has comparable power to variance-components methods. MERLIN and GENENHUNTER also provide rank-based tests, which are appropriate for nonnormally distributed traits. Implementations of variance-components methods are also available in MERLIN and GENENHUNTER. SOLAR (29) is another very flexible package for variance components analysis. Allegro also provide implementations of various quantitative trait methods. Although GENEHUNTER, ALLEGRO (48), MERLIN, and SOLAR can all be used to calculate multipoint IBD distributions, MERLIN is currently the only program that can perform multipoint variance components analysis on the X chromosomes. A very recent work developed a Markov chain Monte Carlo sampling method that uses LOKI (49) to QTL linkage mapping. LOKI performs linkage analysis by using oligogenic quantitative trait locus models and produces estimates of the best trait model and the percentage of variance in the trait explained by each QTL. The great advantage of this software is reducing genetic complexity and variability of large pedigrees, without the problem of the pedigree fragmentation. There is also a wide choice of software available for association analysis. For random sample association analysis, the usual methods are based on ANOVA. However, random sample may suffer bias from population stratification or admixture. So association tests for complex disease or quantitative trait while controlling population stratification become popular. These methods include genomic control (GA) approach, structure association (SA) approach, and semi-parametric (SPT) approach, for controlling false-positive results in the presence of population stratification. On the other hand, family-based design is applied extensively as it can avoid the effect of population stratification. Two common used family-based methods to test associations between markers and quantitative traits are QTDT (50) and FBAT (51). QTDT has the advantages of conducting the powerful total association analyses using the entire sample while correcting for internal correlations among all the members, as well as performing within family association tests that are robust against obvious population stratification. However, QTDT needs to assume that the tested phenotypes are normally distributed. In contrast, FBAT is a non-parametric approach and require no assumptions regarding the trait distribution and can also handle with qualitative traits. We briefly reviewed the well-known packages for QTL mapping. Furthermore, two internet sites provide useful lists of available software. A comprehensive list of software in statistical genetics can be found at http://www.nslij-genetics.org/soft/, with links to their sources. This list continues to be mirrored at its previous site http://linkage.rockefeller.org/soft/. It contains almost all the available genetic software, but without recommendations. The other collection at http://www.hgmp.mrc. ac.uk/Registered/Menu/linkage.html contains only the most popular programs for
230
D.-H. Xiong et al.
linkage analysis, and provides executable files, detailed documentations and a Webbased graphical interface for the most common applications.
4
Notes
1. The linkage approach does not rely on the linkage disequilibrium among genes or markers in adjacent genomic regions; therefore, it can be used to search for any genomic region that contributes relatively large variation in complex traits, without any prior knowledge. In contrast to the population association approach, the linkage approach is robust with respect to population admixture/stratification. However, genomic regions detected by linkage studies are generally large (~30 cM), and are thus not suitable for physical mapping. Therefore, fine-mapping techniques are needed to narrow down a QTL to a small region in order eventually to identify a specific gene (52, 53). The linkage approach has been successfully used to locate a gene(s) underlying mendelian inherited genetic traits, but its application to complex traits can be complicated (54). In this case, it is necessary to screen or genotype (or both) very large samples (55–57), or the power to detect linkage would be limited. 2. Although association analysis is widely used, it still has some limitations: a. Allele frequency distribution. A SNP useful for mapping in one pedigree is not necessarily useful in another one, as most SNPs are rare. b. The criteria of LD. There are different criteria of measures to detect the LD, such as D′ and r2, which make it difficult to take an exact measure of LD pattern. c. Population stratification. Each population has a unique genetic and social history, and thus ancestral patterns of geographical migration, mating practices, reproductive expansions and bottlenecks, and stochastic variation all yield differences in allele frequencies between populations. d. The molecular mechanisms. Although haplotype blocks hold great promise, the molecular mechanisms and population genetic forces that shape their characteristic are still little known. e. The size of LD blocks does not have a specific definition. LD can extend only a few kilobases (kb) around common SNPs (39), but in some cases it can extend to larger than 100 kb (39, 58). f. The number of haplotypes is not defined. When haplotypes are used, a block is usually defined when a small number of haplotypes account for a high proportion of the observations (75–90%). The precise thresholds for the number of haplotypes and proportion of observations are subjective. g. The choice of htSNPs. The various algorithms in tagging make it difficult to select the definite htSNPs. h. The rare alleles are not be fully used. Most previous studies have used the SNPs currently in public databases and available for genotyping that were identified
16 QTL Mapping
231
on a small set of chromosomes and are thus not a representative sample of variants in the population. Instead, common alleles are over-represented and rare alleles are under-represented in these samples. If it is true, suggested that genetic susceptibility to common complex disorders is influenced by relatively common polymorphisms, the SNPs selected by the preceding method will limit the genetic analysis to diseases caused by rarer alleles. i. The SNP density. Genotyping budgets and technology place practical limitations on large-scale association analysis. Some studies have suggested that the sparser marker panel have identified the longer blocks (7). If it is true in general, blocks detected at low marker density might not reflect all the diversity in the data and might not be reliable for association studies. j. Other factors. Although information on the pattern of linkage disequilibrium will allow the design of association studies that survey a substantial fraction of common variants, it will not account for genetic heterogeneity, phenotypic complexity, inadequate sample sizes and genotyping error. In addition, variant frequencies and their relations, over-interpretation of marginal findings is also a factor to make the association irreproducible. 3. Multiple traits analysis is not a panacea. In general, joint analysis of multiple traits increases the number of model parameters that must be estimated, and the additional degree of freedom increase the critical value of the test statistic required to achieve a given level of statistical significance. These factors can offset the potential gains from joint consideration of the correlated characters, with the result that multiple traits analysis may actually be less powerful than single-trait analysis, even with traits that are highly correlated (59). If the correlations between traits are not appropriately handled, for example, by true multivariate analysis or by means of orthogonal canonical variables, then some correction for multiple non-independent tests must be applied to control the rate of type error (60). 4. Until recently, a large amount of researchers have been devoted to the detection and investigation of epistatic interactions. However, it is well known that power to detect epistasis in a QTL framework is low. Epistasis considerably reduces the total amount of information available, considering the classical models of duplicate and complementary gene action (61). Using a variance components framework, the statistical power is also low to detect epistasis in an application of two-locus linkage in extended pedigrees (62). On the other hand, the methodology for estimating epistasis is still in its infancy. Lack of powerful statistical methods limits the identification and characterization of interactions (63). Thus, new powerful approaches are required to improve the ability to identify epistatic interactions. 5. To reduce biological variance, the sample selection for eQTL analysis should be careful. The potential confounding co-variates need to be considered when conducting the analysis, with significant ones adjusted for accurate analysis. Technical details are also essential, such as the quality and quantity of target hybridized, reagents used, and handling methods. The validation of gene expression
232
D.-H. Xiong et al.
level with alternative methods such as RT-PCR and western blotting is important as well. 6. The quality of RNA (integrality and purity) is essential to the overall success of eQTL mapping. Thus, the mature and commercially available kit is recommended when possible. 7. Multiple testing problems could be serious in whole genome association studies. Bonferroni correction is usually used to control for the overall type I error rate. However, this method may be too conservative to allow for the detection of significant SNPs in a genome-wide association study. Using the false discovery rate (FDR) method may provide a better balance between statistical stringency and power (64). Moreover, false or non-replicable findings in population based designs may be attributable to the population stratification problem, for which the methods like genomic control (GC) and structured association (SA) were proposed to detect and adjust for this confounder (65). Acknowledgments Investigators of this work were partially supported by grants from NIH (R01 AR050496, K01 AR02170-01, R01 AR45349-01, and R01 GM60402-01A1) and an LB595 grant from the state of Nebraska. The study also benefited from grants from National Science Foundation of China, Huo Ying Dong Education Foundation, HuNan Province, Xi’an Jiaotong University, and the Ministry of Education of China.
References 1. Falconer, D. S., Mackay, T. F. C. (1996) Introduction to Quantitative Genetics. Longman, Essex, UK. 2. Geldermann, H. (1975) Investigations on inheritance of quantitative characters in animals by gene markers. I. Methods. Theor Appl Genet 46, 319–330. 3. Elston, R. C., Buxbaum, S., Jacobs, K. B., et al. (2000) Haseman and Elston revisited. Genet Epidemiol 19, 1–17. 4. Hill, W. G. (1981) Estimation of effective population size from data on linkage disequilibrium. Genet Res 38, 209–216. 5. Hill, W. J., Robertson, A. (1968) Linkage disequilibrium in finite populations. Theor Appl Genet 38, 226–231. 6. Patil, N., Berno, A. J., Hinds, D. A., et al. (2001) Blocks of limited haplotype diversity revealed by high-resolution scanning of human chromosome 21. Science 294, 1719–1723. 7. Gabriel, S. B., Schaffner, S. F., Nguyen, H., et al. (2002) The structure of haplotype blocks in the human genome. Science 296, 2225–2229. 8. Williams, J. T., Van Eerdewegh, P., Almasy, L., et al. (1999) Joint multipoint linkage analysis of multivariate qualitative and quantitative traits. I. Likelihood formulation and simulation results. Am J Hum Genet 65, 1134–1147. 9. Amos, C. I., Elston, R. C., Bonney, G. E., et al. (1990) A multivariate method for detecting genetic linkage, with application to a pedigree with an adverse lipoprotein phenotype. Am J Hum Genet 47, 247–254. 10. Schork, N. J. (1993) Extended multipoint identity-by-descent analysis of human quantitative traits: efficiency, power, and modeling considerations. Am J Hum Genet 53, 1306–1319. 11. Almasy, L., Dyer, T. D., Blangero, J. (1997) Bivariate quantitative trait linkage analysis: pleiotropy versus co-incident linkages. Genet Epidemiol 14, 953–958.
16 QTL Mapping
233
12. Moore, J. H. (2003) The ubiquitous nature of epistasis in determining susceptibility to common human diseases. Hum Hered 56, 73–82. 13. Jansen, R. C., Nap, J. P. (2001) Genetical genomics: the added value from segregation. Trends Genet 17, 388–391. 14. DeCook, R., Lall, S., Nettleton, D., et al. (2006) Genetic regulation of gene expression during shoot development in Arabinosis. Genetics 172, 1155–1164. 15. Schadt, E. E., Monks, S. A., Drake, T. A., et al. (2003) Genetics of gene expression surveyed in maize, mouse and man. Nature 422, 297–302. 16. Yaguchi, H., Togawa, K., Moritani, M., et al. (2005) Identification of candidate genes in the type 2 diabetes modifier locus using expression QTL. Genomics 85, 591–599. 17. Hirschhorn, J. N., Daly, M. J. (2005) Genome-wide association studies for common diseases and complex traits. Nat Rev Genet 6, 95–108. 18. Herbert, A., Gerry, N. P., McQueen, M. B., et al. (2006) A common genetic variant is associated with adult and childhood obesity. Science 312, 279–283. 19. John, S., Shephard, N., Liu, G., et al. (2004) Whole-genome scan, in a complex disease, using 11,245 single-nucleotide polymorphisms: comparison with microsatellites. Am J Hum Genet 75, 54–64. 20. Klein, R. J., Zeiss, C., Chew, E. Y., et al. (2005) Complement factor H polymorphism in agerelated macular degeneration. Science 308, 385–389. 21. Maraganore, D. M., de Andrade, M., Lesnick, T. G., et al. (2005) High-resolution wholegenome association study of Parkinson disease. Am J Hum Genet 77, 685–693. 22. Namkung, J., Kim, Y., Park, T. (2005) Whole-genome association studies of alcoholism with loci linked to schizophrenia susceptibility. BMC Genet 6, S9. 23. Spinola, M., Meyer, P., Kammerer, S., et al. (2006) Association of the PDCD5 locus with lung cancer risk and prognosis in smokers. J Clin Oncol 24, 1672–1678. 24. Weaver, T. A. (2000) High-throughout SNP discovery and typing for genome-wide genetic analysis. A Trends Guide 20, 136–142. 25. Syvanen, A. C. (2005) Toward genome-wide SNP genotyping. Nat Genet 37, S5–10. 26. Stoll, M., Kwitek-Black, A. E., Cowley, A. W., et al. (2000) New target regions for human hypertension via comparative genomics. Genome Res 10, 473–482. 27. Lynch, M., Walsh, B. (1998) Genetics and Analysis of Quantitative Traits. Sinauer Associates, Sunderland, MA. 28. Haseman, J. K., Elston, R. C. (1972) The investigation of linkage between a quantitative trait and a marker locus. Behav Genet 2, 3–19. 29. Almasy, L., Blangero, J. (1998) Multipoint quantitative-trait linkage analysis in general pedigrees. Amer J Hum Genet 62, 1198–1211. 30. George, A. W., Visscher, P. M., Haley, C. S. (2000) Mapping quantitative trait loci in complex pedigrees: a two-step variance component approach. Genetics 156, 2081–2092. 31. Slate, J., Pemberton, J. M., Visscher, P. M. (1999) Power to detect QTL in a free-living puligenous population. Heredity 83, 327–336. 32. Amos, C. I., Elston, R. C. (1989) Robust methods for the detection of genetic linkage for quantitative data from pedigrees. Genet Epidemiol 6, 349–360. 33. Fulker, D. W., Cardon, L. R. (1994) A sib-pair approach to interval mapping of quantitative trait loci. Am J Hum Genet 54, 1092–1103. 34. Tiwari, H. K., Elston, R. C. (1997) Linkage of multilocus components of variance to polymorphic markers. Ann Hum Genet 61, 253–261. 35. Amos, C. I. (1994) Robust variance-components approach for assessing genetic linkage in pedigrees. Am J Hum Genet 54, 535–543. 36. Devlin, B., Roeder, K., Wasserman, L. (2001) Genomic control, a new approach to geneticbased association studies. Theor Popul Biol 60, 155–166. 37. Devlin, B., Roeder, K., Bacanu, S. A. (2001) Unbiased methods for population-based association studies. Genet Epidemiol 21, 273–284. 38. Pritchard, J. K., Rosenberg, N. A. (1999) Use of unlinked genetic markers to detect population stratification in association studies. Am J Hum Genet 65, 220–228.
234
D.-H. Xiong et al.
39. Schaid, D. J. (1998) Transmission disequilibrium, family controls, and great expectations. Am J Hum Genet 63, 935–941. 40. Benjamini, Y. A. H. Y. (1995) Controlling the false discovery rate: a practical and powerful approach to multiple testing. J Roy Stat Soc Series B 57, 289–300. 41. McIntyre, L. M., Martin, E. R., Simonsen, K. L., et al. (2000) Circumventing multiple testing: a multilocus Monte Carlo approach to testing for association. Genet Epidemiol 19, 18–29. 42. Boomsma, D. I., Dolan, C. V. (1998) A comparison of power to detect a QTL in sib-pair data using multivariate phenotypes, mean phenotypes, and factor scores. Behav Genet 28, 329–340. 43. Blangero, J., Almasy, L. (1997) Multipoint oligogenic linkage analysis of quantitative traits. Genet Epidemiol 14, 959–964. 44. Cockerham, C. C., Zeng, Z. B. (1996) Design III with marker loci. Genetics 143, 1437–1456. 45. Thomas, D., Xie, R., Gebregziabher, M. (2004) Two-Stage sampling designs for gene association studies. Genet Epidemiol 27, 401–414. 46. Kruglyak, L., Daly, M. J., Reeve-Daly, M. P., et al. (1996) Parametric and nonparametric linkage analysis: a unified multipoint approach. Am J Hum Genet 58, 1347–1363. 47. Abecasis, G. R., Cherny, S. S., Cookson, W. O., et al. (2002) Merlin: rapid analysis of dense genetic maps using sparse gene flow trees. Nat Genet 30, 97–101. 48. Gudbjartsson, D. F., Jonasson, K., Frigge, M. L., et al. (2000) Allegro, a new computer program for multipoint linkage analysis. Nat Genet 25, 12–13. 49. Shmulewitz, D., Heath, S. C., Blundell, M. L., et al. (2006) Linkage analysis of quantitative traits for obesity, diabetes, hypertension, and dyslipidemia on the island of Kosrae, Federated States of Micronesia. Proc Natl Acad Sci U S A 103, 3502–3509. 50. Abecasis, G. R., Cardon, L. R., Cookson, W. O. (2000) A general test of association for quantitative traits in nuclear families. Am J Hum Genet 66, 279–292. 51. Horvath, S., Xu, X., Laird, N. M. (2001) The family based association test method: strategies for studying general genotype: phenotype associations. Eur J Hum Genet 9, 301–306. 52. Deng, H. W., Chen, W. M., Recker, R. R. (2000) QTL fine mapping by measuring and testing for Hardy-Weinberg and linkage disequilibrium at a series of linked marker loci in extreme samples of populations. Am J Hum Genet 66, 1027–1045. 53. Deng, H. W., Shen, H., Xu, F. H., et al. (2002) Tests of linkage and/or association of genes for vitamin D receptor, osteocalcin, and parathyroid hormone with bone mineral density. J Bone Miner Res 17, 678–686. 54. Lander, E. S., Schork, N. J. (1994) Genetic dissection of complex traits. Science 265, 2037–2048. 55. Allison, D. B. (1997) Transmission-disequilibrium tests for quantitative traits. Am J Hum Genet 60, 676–690. 56. Risch, N., Merikangas, K. (1996) The future of genetic studies of complex human diseases. Science 273, 1516–1517. 57. Xiong, M. M., Krushkal, J., Boerwinkle, E. (1998) TDT statistics for mapping quantitative trait loci. Ann Hum Genet 62, 431–452. 58. Abecasis, G. R., Noguchi, E., Heinzmann, A., et al. (2001) Extent and distribution of linkage disequilibrium in three genomic regions. Am J Hum Genet 68, 191–197. 59. Mangin, B., Thoquet, P., Olivier, J., et al. (1999) Temporal and multiple quantitative trait loci analyses of resistance to bacterial wilt in tomato permit the resolution of linked loci. Genetics 151, 1165–1172. 60. Weller, J. I., Song, J. Z., Heyen, D. W., et al. (1998) A new approach to the problem of multiple comparisons in the genetic dissection of complex traits. Genetics 150, 1699–1706. 61. Eaves, L. J. (1994) Effect of genetic architecture on the power of human linkage studies to resolve the contribution of quantitative trait loci. Heredity 72, 175–192. 62. Mitchell, B. D., Ghosh, S., Schneider, J. L., et al. (1997) Power of variance component linkage analysis to detect epistasis. Genet Epidemiol 14, 1017–1022.
16 QTL Mapping
235
63. Moore, J. H., Williams, S. M. (2002) New strategies for identifying gene-gene interactions in hypertension. Ann Med 34, 88–95. 64. Storey, J. D., Tibshirani, R. (2003) Statistical significance for genomewide studies. Proc Natl Acad Sci U S A 100, 9440–9445. 65. Hao, K., Li, C., Rosenow, C., et al. (2004) Detect and adjust for population stratification in population-based association study using genomic control markers: an application of Affymetrix Genechip Human Mapping 10 K array. Eur J Hum Genet 12, 1001–1006.
Chapter 17
In Situ Nuclear Organization of Regulatory Machinery Shirwin M. Pockwinse, Sayyed K. Zaidi, Ricardo F. Medina, Rachit Bakshi, Krishna P. Kota, Syed A. Ali, Daniel W. Young, Jeffery A. Nickerson, Amjad Javed, Martin Montecino, Andre J. van Wijnen, Jane B. Lian, Janet L. Stein, and Gary S. Stein
Abstract Regulatory machinery for gene expression, replication, and repair are architecturally organized in nuclear microenvironments. This compartmentalization provides threshold concentrations of macromolecules for the organization and assembly of regulatory complexes for combinatorial control. A mechanistic understanding of biological control requires the combined application of molecular, cellular, biochemical, and in vivo genetic approaches. This chapter provides methodologies to characterize nuclear organization of regulatory machinery by in situ immunofluorescence microscopy. Keywords Nuclear organization, cell synchrony, Runx, confocal microscopy, immunofluorescence microscopy, FRAP, live cell microscopy, nuclear matrix.
1
Introduction
There is growing evidence that the organization of nucleic acids and regulatory proteins in microenvironments within the nucleus may be functionally linked with biological control (1–5). The non-random, punctuate localization of regulatory machinery for gene expression, replication and repair in sub-nuclear compartments is dynamic rather than static and supports physiological responsiveness to regulatory cascades. Such nuclear microenvironments appear to provide an architectural basis for the organization and assembly of genes, transcription factors, and co-regulatory proteins that are principal components of combinatorial control. The focal localization of regulatory macromolecules can effectively support the integration of regulatory networks and establish threshold levels of factors for positive and negative control in a broad spectrum of biological contexts that include development and tissue remodeling. Equally important, changes in the composition and organization of regulatory machinery in nuclear microenvironments provide insight into perturbed mechanisms that relate to human disease, which is strikingly illustrated by, but not restricted to, skeletal disorders and tumorigenesis. Examples are modifications in the size, From: Methods in Molecular Biology, Vol. 455: Osteoporosis: Methods and Protocols Edited by Jennifer J. Westendorf © Humana Press, Totowa, NJ
239
240
S.M. Pockwinse et al.
number, and composition of intranuclear sites that support transcription, replication, repair, and altered regulatory domains that are causally associated with cleidocranial dysplasia and competency for metastatic breast cancer cells to form osteolytic lesions in bone. The repertoire of genes and regulatory factors, together with expanding insight into the location of regulatory machinery for gene expression, replication, and repair from a temporal/spatial perspective is increasingly evident. However, there is a necessity for stringent definitions of regulatory machinery within the three-dimensional context of nuclear architecture to validate components of control that are linked to cellular biology and pathology. Specificity and quantitation are essential for mechanistically interrogating biological control and are similarly essential for diagnosing disease or providing a platform for development of targeted therapies. Traditionally, compartmentalization of regulatory machinery, including components of macromolecular complexes that reflect combinatorial control, has been identified and characterized by sub-nuclear fractionation followed by biochemical and molecular analyses. While informative, there are constraints that prevent directly relating isolated complexes with regulatory machinery that resides in sub-nuclear compartments where location is a major element of function. During the past several years, advances in microscopy, together with highly specific antibodies and epitope tags, have provided the basis for investigating the assembly and activities of regulatory machinery within cellular microenvironments. Nucleic acids and proteins that mediate gene expression, replication, and repair are now being examined in relation to composition, organization, and localization. Cellular biochemistry and morphology are not mutually exclusive. The combined use of high-resolution cellular, biochemical, and molecular approaches maximizes the extent to which regulatory mechanisms can be defined. With live cell imaging, the in vivo relevance for compartmental regulation of regulatory machinery takes on new dimensions, permitting regulatory mechanisms to be validated in time and space. This chapter focuses on visualization of nuclear microenvironments using Runx transcription factors as a paradigm for compartmentalization of regulatory machinery within nuclei of osteoblastic cells. It presents approaches for imaging of focally localized regulatory complexes in interphase nuclei. Because of the requirement to sustain control during cell division for epigenetic cell fate determination and phenotype commitment, this chapter presents strategies for visualizing transcription factors throughout mitosis, at foci on metaphase chromosomes and in progeny cells. Specificity and quantitation of regulatory complexes that are visualized by microscopy are required to informatively relate cell morphology with regulatory mechanisms. We recently developed an approach designated “Intranuclear Informatics” that quantitatively assimilates multiple parameters of regulatory protein localization within the nucleus into contributions toward skeletal gene expression from a temporal/ spatial perspective (6).
17 Organization of Regulatory Machinery
2 2.1
241
Materials Cell Synchronization
1. Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 2 mM Lglutamine, and 100 U/mL penicillin G, 100 µg/mL streptomycin. 2. DMEM supplemented with 2 mM L-glutamine, 100 U/mL penicillin G, 100 µg/ mL streptomycin, and 10% fetal bovine serum (FBS; Atlanta Biologicals, Lawrenceville, GA). 3. DMEM supplemented with 2 mM L-glutamine, 100 U/mL penicillin G, 100 µg/ mL streptomycin, and 20% FBS. 4. McCoy’s 5A medium supplemented with 2 mM L-glutamine, 100 U/mL penicillin G, 100 µg/mL streptomycin, and 15% FBS. 5. Phosphate buffered saline (PBS). 6. Propidium iodide-stained cells subjected to fluorescence-activated cell sorting (FACS). 7. Nocodazole: 2 mg/mL in DMSO Hybri-Max (Sigma) and stored at −20 °C. 8. Thymidine: 100 mM in serum-free medium, filter-sterilized in a cell culture hood and stored in aliquots at −20 °C. This is a 50× stock. 9. 2'-Deoxycytidine: 24 mM in serum-free medium, filter-sterilized in a cell culture hood and stored in aliquots at −20 °C. This is a 1,000× stock. Stock solution will change the medium color when dissolved, indicative of a change in pH. However, further dilution to the working concentration will have no effect on the medium pH.
2.2
Preparation of Metaphase Chromosome Spreads from Suspension and Adherent Cell Cultures
1. Karyomax colcemid (10 µg/mL). 2. 0.075 M Potassium chloride (KCl) solution. Pre-warmed to 37 °C in a water bath. 3. Fixative: Methanol/glacial acetic acid at 3:1, made fresh each time. The fixative should be ice-cold prior to use. 4. Microscopy glass slides (pre-chilled at 4 °C).
2.3
Nuclear Matrix Intermediate Filament Preparation
1. Sterile glass coverslips, 22-mm round, coated with 0.5% gelatin. 2. Cytoskeleton (CSK) Buffer (10× stock solution): 1 M NaCl, 100 mM PIPES pH 6.8, 30 mM MgCl2, 10 mM EGTA. 1× working solution: Freshly prepare 100 mL
242
3.
4.
5.
6. 7. 8. 9. 10.
2.4 1. 2. 3. 4.
5. 6.
7. 8. 9. 10.
S.M. Pockwinse et al.
of 1× CSK buffer by dissolving 10.27 g sucrose in 77.6 mL of double distilled water. Add 10 mL of 10× stock CSK buffer, 0.5 mL of Triton X-100, 0.8 mL of ribonucleoside-vanadyl complex (RVC, New England Biolabs, Ipswich, MA) and 0.8 mL of 150 mM AEBSF [4-(2-aminoethyl) benzenesulfonyl fluoride] (Sigma). Digestion Buffer (DB) (10× stock solution): 0.5 M NaCl, 100 mM PIPES pH 6.8, 30 mM MgCl2, 10 mM EGTA. Freshly prepare 1× DB as described in the preceding for 1× CSK buffer except use 10× DB instead of 10× CSK buffer. Phosphate Buffered Saline (PBS): 9.1 mM dibasic sodium phosphate (Na2HPO4), 1.7 mM monobasic sodium phosphate (NaH2PO4), and 150 mM NaCl. Adjust pH to 7.4 with NaOH. Fixatives: 3.7% formaldehyde in PBS (WC fixative), or in 1× CSK buffer (CSK fixative), or in 1× DB (NMIF fixative). All fixatives should be freshly prepared. Stop Solution: 250 mM ammonium sulfate in 1× DB. (Add 1 volume of 2 M ammonium sulfate to 8 volumes of 1× DB). Permeabilizing Solution: 0.25% Triton X-100 in PBS. RNAse-free DNAse. PBSA: 0.5% bovine serum albumin (BSA) in PBS. Prolong Gold (Invitrogen, Carlsbad, CA).
Microscopy 40-mm glass coverslips (Bioptechs, Butler, PA). 60-mm Corning culture dishes. Microwave oven. McCoy’s 5A complete medium: McCoy’s 5A medium with L-glutamine, supplemented with 10% FBS, 1% penicillin-streptomycin, and 1% L-glutamine (200 mM). U2-OS (ATCC, Manassas, VA) (see Note 1) plated at 0.3 × 106 per 60-mm culture dish. Transfection reagent: FuGENE 6 (Roche, Indianapolis, IN), McCoy’s 5A complete medium, McCoy’s 5A incomplete medium, DNA concentration ranges from 200–600 ng per coverslip (see Note 2). Live cell stage-closed system chamber (Bioptechs). Aqueduct slide, gaskets (Bioptechs). Perfusion Fluid: McCoy’s 5A complete medium with 20 mM Hepes buffer solution (1 M), heated to 37 °C. Bioptechs objective heater and slide heater.
17 Organization of Regulatory Machinery
3
243
Methods
3.1
Cell Cycle Progression
We present three methods currently used in our laboratory to synchronize cells. They allow the enrichment of a cell population in a particular cell cycle stage. By releasing cells from a block, progression through cell cycle stages can be easily monitored. The first two methods (double thymidine block and nocodazole treatment) are particularly useful to study cell cycle progression–related events, whereas the last method (growth stimulation) is of particular interest in investigating growth stimulation–related events, since serum starvation will allow transition from G0 to G1 to occur. An important consideration is the fact that cells must be exponentially growing during the entire procedure to ensure active proliferation.
3.1.1
Double Thymidine Block Synchronization of Monolayer Cultures
1. Plate cells at a density that will permit active growth throughout the time course of the synchronization procedure. (We routinely plate 0.4 × 106 human HeLa S3 cervical adenocarcinoma cells per 100-mm plate 24 hours prior to the procedure.) Seed 21 plates (one plate each for RNA isolation, protein extract preparation and fluorescence activated cell sorting (FACS) analysis (see Note 3). 2. Block cells by removing the growth medium and replacing with fresh DMEM supplemented with 10% FBS containing 2 mM thymidine. Incubate the cells for 12–16 hours (see Note 4). 3. Release the cells from the first thymidine block by removing the medium and washing twice with phosphate buffered saline (PBS; at 37 °C). Add DMEM supplemented with 10% FBS containing 24 µM 2'-deoxycytidine and incubate for exactly 9 hours. 4. Wash cells twice with PBS (at 37 °C) and impose a second thymidine block as described in the preceding step. 5. Release the second thymidine block washing the cells twice with PBS (at 37 °C) and add DMEM supplemented with 10% FBS containing 24 µM 2'-deoxycytidine. 6. Remove three plates at each of the following time points: 0, 1.5, 3, 4.5, 6, and 9 hours after release. Use 1 plate of each time point for RNA isolation, protein extract preparation, and FACS analysis (to assess cell cycle distribution) (see Note 5). 7. Analyze cell cycle distribution by FACS analysis, RNA by quantitative realtime PCR or northern blot, and protein by western blot. In our laboratory we routinely assess synchrony by both RNA (histone H4 mRNA) and proteins (cyclins E, A, and B1) known to fluctuate during the cell cycle in a characteristic manner.
244
3.1.2
S.M. Pockwinse et al.
Synchronization of Monolayer Cultures by Nocodazole Treatment
1. Plate cells at a density that will permit active growth throughout the time course of the synchronization procedure. (We plate 1 × 106 human SaOS-2 osteosarcoma cells per 100-mm plate 24 hours prior to the procedure.) Seed 18 plates (one plate each for RNA isolation, protein extract preparation and FACS analysis) (see Note 3). 2. Cells are first blocked by removing the growth medium and replacing with fresh McCoy’s 5A medium supplemented with 15% FBS containing 50 ng/mL nocodazole. Block the cells for 24 hours (see Note 4). 3. Cells are gently washed twice with PBS (at 37 °C) and released from mitotic block with fresh media. At this stage, cells are blocked in mitosis; therefore, they are rounded and very loosely attached to the plate. Handle cells carefully to minimize cell loss. 4. Process three plates at each of the following time points: 0, 13, 16, 20, and 24 hours after release. Use one plate of each time point for RNA isolation, protein extract preparation, and FACS analysis (to assess cell cycle distribution) (see Note 5). 5. Analyze cell cycle distribution by FACS analysis, RNA by quantitative real-time PCR or northern blot, and protein by western blot. In our laboratory we routinely assess synchrony by both RNA (histone H4 mRNA) and proteins (cyclins E, A, and B1) known to fluctuate during the cell cycle in a characteristic manner.
3.1.3
Growth Stimulation: Serum Deprivation of Monolayer Cultures
1. Plate cells at a density that will permit active growth throughout the time course of the stimulation procedure. (We routinely plate 0.6 × 106 human T98G glioblastoma cells per 100-mm plate 24 hours prior to the procedure.) Seed 30 plates (one plate each for RNA isolation, protein extract preparation, and FACS analysis). Due to material limitations, we routinely collect two plates each for RNA isolation and protein extract preparation, and one plate for FACS analysis for the first two time points and for the asynchronous control (Fig. 17.1). 2. Force cells to enter quiescence by removing the growth medium, washing twice with PBS (at 37 °C), and replacing with fresh DMEM supplemented with 2 mM L-glutamine, and 100 U/mL penicillin G, 100 µg/mL streptomycin (no FBS added). Starve cells for 72 hours (see Note 3). 3. Cells are washed with PBS (at 37 °C) and stimulated to proliferate with fresh DMEM containing 20% FBS (see Note 4). 4. Process three plates at each of the following time points: 0, 4, 12, 15, 18, 21, and 24 hours after release. Use one plate of each time point for RNA isolation, protein extract preparation, and FACS analysis (to assess cell cycle distribution) (see Note 5). 5. Analyze cell cycle distribution by FACS analysis, RNA by quantitative real-time PCR or northern blot, and protein by western blot. In our laboratory we routinely assess synchrony by both RNA (histone H4 mRNA) and proteins (cyclins E, A, and B1) known to fluctuate during the cell cycle in a characteristic manner.
17 Organization of Regulatory Machinery
245
Fig. 17.1 Cell synchrony. Different phases of the cell cycle can be assessed by enrichment of cell population in each cell cycle stage. a. Cartoon showing a modified approach taken to synchronize cells to allow simultaneous processing of samples. This approach is particularly useful for simultaneous sample preparation for microscopy. Cells are plated, starved (synchronized), and released from blocking at different times. Thus, cells will be ready at the same time by the end of the synchrony. This allows simultaneous harvesting of cells at different stages of the cell cycle. Examples of FACS data for the different approaches taken to synchronize cells are given for synchronization of cells at the G1/S phase transition (b. S-phase synchrony), enrichment of mitotic cells (c. mitotic block and release), and obtaining cells at G0 (d. serum stimulation).
3.2
Preparation of Metaphase Chromosome Spreads from Suspension and Adherent Cell Cultures
Metaphase chromosome spreads are traditionally used for identification of chromosomal abnormalities (translocation, deletions, and insertions) in patients. It has been recently observed that some lineage-specific proteins, such as Runx transcription factors, retain association with chromosomes during mitosis (Fig. 17.2A). In such cases, metaphase chromosome spreads provide powerful means for the identification of chromosomes in which Runx transcription factors reside during mitosis. For example, Runx proteins associate with nucleolar organizing regions (NORs) on metaphase chromosomes.
246
S.M. Pockwinse et al.
Fig. 17.2 Metaphase chromosome spreads. a. Runx2 is stable throughout mitosis. Synchronously growing Saos-2 cells were fixed and stained for DNA by using DAPI and for Runx2 by using a rabbit polyclonal antibody. Mitotic cells were identified by chromosome morphology. High-resolution images obtained by three-dimensional deconvolution algorithms reveal that Runx2 (green) is localized in mitotic chromosomes. A subset of Runx2 co-localizes with the microtubules, labeled by α-tubulin staining (red). b. A metaphase chromosome spread of mouse pre-myoblast C2C12 cells, immunolabeled for upstream binding factor (UBF; red) to identify nucleolar organizing regions (NORs).
NORs can be visualized in situ by immunolabeling metaphase chromosome spread preparations for upstream binding factor (UBF), a known regulator of ribosomal RNA transcription (see Fig. 17.2B). Below is a protocol used in our laboratory for preparation of metaphase chromosome spreads (7–9).
3.2.1
Suspension Cells
1. Passage 1 × 106 cells in regular growth medium, 1 to 2 days prior to performing the chromosome spreads. 2. Feed cells with fresh medium 12–14 hours prior to harvesting and add Colcemid to a final concentration of 0.05 µg/mL; incubate at 37 °C for 3–4 hours. 3. Transfer the cells into a centrifuge tube and pellet at 750 g for 5 minutes (see Note 6). 4. Aspirate the supernatant completely. 5. Add 10 mL of 0.075 M KCl solution drop by drop, i.e., hypotonic treatment. Resuspend the pellet by pipetting up and down gently. Incubate at 37 °C for 30 minutes (see Note 7).
17 Organization of Regulatory Machinery
247
6. Add 1 mL of ice-cold fixative (methanol/glacial acetic acid 3:1) to the cell suspension and keep at room temperature for 15 minutes. 7. Spin the cells at 400 g for 5 minutes. 8. Aspirate the supernatant completely. Add 2 mL of fresh fixative (methanol/glacial acetic acid in a 3:1 ratio) to the cells and keep at 4 °C for slide preparation. (At this stage the cell suspension can be stored at 4 °C for few days or at −20 °C long term.)
3.2.2
Adherent Cells
1. Follow the initial steps 1 and 2 as described in Section 3.2.1. for suspension cells. 2. During the mitotic block, some adherent cells become rounded and detach from the plate. In this case, medium should not be discarded but should be transferred to a centrifuge tube. 3. Rinse the plate with PBS and detach the cells with 0.5 mL of trypsin. Mix the detached cells with the medium that contains the cells from mitotic shake off (step 2). 4. Centrifuge the cells at 750 g for 5 minutes. Discard the supernatant and resuspend pellet in ice-cold PBS and centrifuge again at 750 g for 5 minutes. Repeat the PBS wash. 5. Continue as in step 5 for suspension cells and follow each step exactly as described.
3.2.3
Slide Preparation
1. Cool slides to 4 °C before using them for metaphase spreads. Adjust hot plate for medium heat (ideally 50–60 °C) (see Note 8). 2. Take 100–200 µL of the cell suspension and add drop by drop to the slide from a height of about 20 cm. Drain the excess solution by tilting the slide. 3. Immediately put the slide on hot plate (heat shock) for 1 minute. 4. Air dry the slide, check for chromosome spreading in the phase contrast microscope. 5. Keep the slides in a box at 4 °C (up to 8 weeks) or at −80 °C for a longer period of time.
3.3
Nuclear Matrix Intermediate Filament Preparation
It is becoming increasingly evident that regulatory proteins are organized in highly specialized compartments within the mammalian nucleus. The biological activity of proteins often correlates with their presence or absence in these nuclear microenvironments (1). The sub-nuclear organization of regulatory proteins can be
248
S.M. Pockwinse et al.
Fig. 17.3 In situ assessment of nuclear microenvironments. Regulatory proteins can be visualized by indirect immunofluorescence in situ. Proteins involved in distinct nuclear processes are localized in specialized nuclear microenvironments. These microenvironments can be further visualized by removing soluble cytosolic and nuclear proteins as well as chromatin. The procedure is schematically outlined. The upper panel shows a cell after sequential extractions that remove soluble cytosolic as well as soluble nuclear proteins. Finally, the chromatin is digested with DNase I to reveal a network of ribonuclear proteins, designated the nuclear matrix. The middle panel shows corresponding in situ immunofluorescence of an osteoblast co-stained with tubulin (red) and Runx2 (green). DNA is depicted as a blue colored circle. Each fraction can be also resolved by SDS-PAGE (sodium dodecyl sulfate-polyacrylamide gel electrophoresis) to identify proteins of interest. The bottom panel shows a schematic of western blot analysis of biochemical fractionation for Runx2, which is an architecturally associated protein primarily present in the NM-IF fraction.
assessed by the sequential removal of soluble proteins and chromatin from the mammalian cell (Fig. 17.3) followed by either in situ immunofluorescence or western blot analysis. The following is an optimized protocol that we routinely use for in situ assessment of parameters of gene expression.
3.3.1
Whole Cell Preparation
1. Plate cells at a density of 0.5 × 106 cells per 6-well plate and incubate in humidified incubator with 5% CO2 at 37 °C for 24 hours.
17 Organization of Regulatory Machinery
249
2. After 24 hours, wash cells twice with ice-cold PBS, fix the WC preparation on ice for 10 minutes (in an experiment, typically two wells of a six-well plate are allocated to each of the WC, CSK, and NMIF preparations) by adding 2 mL of WC fixative per well. 3. Wash cells once with PBS. 4. To facilitate antibody staining of WC preparations, permeabilize WC preparations with 1 mL of permeabilizing solution on ice for 20 minutes. 5. Aspirate permeabilizing solution and wash twice with PBS. 6. Add 1 mL of PBSA to the wells. 3.3.2
Cytoskeleton Preparation
1. Wash cells twice with ice-cold PBS. 2. Add 1 mL of 1× CSK buffer and incubate plates on ice for 5 minutes while swirling plates once or twice. 3. Wash wells allotted for CSK preparation (see Section 3.3.1.) once with ice-cold PBS and fix cells by adding 2 mL of CSK fixative per well. 4. Aspirate CSK fixative after 10 minutes and wash twice with PBS. 5. Add 1 mL of PBSA to the wells. 3.3.3
Nuclear Matrix Intermediate Filament Preparation
1. Wash cells twice with ice-cold PBS. 2. Add 1 mL of 1× CSK buffer and incubate plates on ice for 5 minutes while swirling plates once or twice. 3. Prepare 1 mL of DB by adding 400 units of RNase free DNase I to 1× DB. 4. Flatten parafilm on the covers of plates, label, and dispense 20 µL drop of DB containing RNase free DNase I on the covers of respective plates. (This step is to conserve the amount of DNase I, otherwise add 1 mL of DB containing RNase free DNase I to each well.) 5. To digest the chromatin with DNase I, carefully invert the coverslip, so that cells will face DB containing DNase I. 6. Incubate cells for 50 minutes at 30 °C. Place coverslips back in their respective labeled wells. Add 1 mL Stop Solution to the wells and incubate plates on ice for 10 minutes to stop the activity of DNase I. 7. Wash once with ice-cold PBS and fix NMIF preparations in 2 mL of nuclear matrix intermediate filament (NMIF) fixative on ice for 10 minutes. 8. Aspirate fixative and wash twice with PBS. 9. Add 1 mL of PBSA. 3.3.4
Immuno-Staining of the Samples
1. Dilute antibody in PBSA to an appropriate dilution. As quality and specificity of antibodies vary among suppliers and lots, we recommend testing several
250
2.
3. 4. 5. 6. 7. 8.
S.M. Pockwinse et al.
dilutions to optimize antibody concentration. When immuno-labeling cells with two proteins, caution must be practiced to assure that the antibodies used are raised in different species (e.g., mouse versus rabbit). If raised in the same species, they must be of different isotypes (e.g., IgG vs. IgM). On parafilm already flattened on the lids of plates, dispense a 20-µL drop of diluted antibody per well. Carefully invert a coverslip on the drop so that the cells are in direct contact with the antibody. Avoid creating air bubbles by gently placing the coverslip on one edge on the antibody droplet and slowly lowering it. Incubate for 1 hour at 37 °C. Place coverslips back in respective wells on ice with cells facing upward and wash four times with ice-cold PBSA. Stain cells with appropriate secondary antibodies conjugated with fluorochromes (e.g., Texas Red or FITC) for 1 hour at 37 °C. Place coverslips back in their respective wells and wash four times with ice-cold PBSA. Stain cells with DAPI (0.5 µg DAPI in 0.1% Triton X-100-PBSA) for 5 minutes on ice. Wash once with 0.1% Triton X-100-PBSA followed by two washes with PBS. Leave cells in last wash on ice. Immediately mount coverslips in an antifade mounting medium (e.g., Prolong Gold) and aspirate excess of mounting medium. After 10–15 minutes, seal coverslips with nail polish and store at −20 °C in dark.
3.4
Microscopy
3.4.1
Fluorescence Microscopy
Fluorescence microscopy provides a powerful tool to assess sub-cellular and subnuclear localization of regulatory proteins as well as nucleic acids. A variety of microscopes are available; each microscope has its own set of unique features. In the following are the instructions specific for the Zeiss Axioplan 2 Microscope. 1. Turn on the mercury lamp, microscope, charged-coupled device (CCD) camera and computer. Clean all lenses with microscopic lens paper. 2. Wipe your slide with tissue paper and lens cleaner. Place a small drop of lens oil on your slide. It is important that the cover slips are sealed to avoid any mixing of the mounting medium with lens oil. 3. To find your cells, start at 10× magnification and then proceed to 40× magnification to start your analysis. Once you have identified a cell or field to document, you can then increase the magnification, depending upon your specific requirements. 4. Once you have selected a cell or field for documentation, you are ready to acquire an image using MetaMorph Software (Molecular Devices, Downingtown, PA). 5. Before acquiring an image, make sure that the arrow on the knob, which diverts light either toward a 35-mm camera or a charged coupled device (commonly called as CCD camera), is pointing toward the CCD camera.
17 Organization of Regulatory Machinery
251
6. In the main menu, go to ACQUIRE to access the Acquire Dialogue Box. 7. Enter an exposure time. Set the region of interest by selecting Entire Chip or Central Quadrant option on the Acquire dialogue box. 8. Acquire the image on all filters (DAPI, Fitc, Texas Red, and Phase) if analyzing a dual labeled slide. 9. The default image depth of the CCD camera is 12 bits. However, these images can not be opened by Adobe Photoshop or Microsoft PowerPoint Software. Copy images to 8 bits by selecting “Copy to 8 Bits” command on the main task bar. Always keep your original image (i.e., raw data) as it contains the most information. 10. Once acquired, images can be presented (PowerPoint) or published (Adobe Photoshop, Illustrator) directly or further analyzed quantitatively by using MetaMorph Imaging Software or the Intranuclear Informatics mathematical algorithm (see the following).
3.4.2
Viewing Live Cells Using the Confocal Microscope
Our lab has characterized the Runx family of transcription factors, describing spatial distribution, sub-nuclear architectural scaffolding, and relationships with co-regulatory factors. Much of this work was done with fixed cells on an epifluorescence microscope with verification using a confocal microscope. This naturally led to an interest in documenting the Runx protein dynamics using live cell imaging; looking at mobility, mitotic labeling, and protein–protein interactions. The laser scanning confocal microscope coupled with a Bioptechs microobservation system enables us to simulate conditions of living cells during live cell microscopy by providing a continuity of temperature, pH and food by heating the slide and objective lens and perfusing cell nutrients with pH maintenance over living cells via an aqueduct slide and dialysis pump. The laser scanning confocal microscope offers us higher-image resolution with the ability to capture images that are sharply defined optical sections produced by the elimination of out of focus light and background information from which threedimensional renderings can be recreated. This coupled with the live cell stage allows us to verify the location of Runx foci throughout dynamic processes, for example, mitosis and to assess the mobility of Runx foci in interphase and during mitosis (Fig. 17.4B) using fluorescence recovery after photobleaching (FRAP) techniques. A common problem occurs while live cell imaging GFP-transfected cells. Cells become extremely sensitive when exposed to blue filtered UV light (green fluorescence) and die while viewing over long periods of time. For example, in a dynamic process such as mitosis, cells permanently stall in metaphase. Using the scanning laser confocal microscope relieves this situation. Red fluorescence proteins (RFPs) are not as sensitive to UV light; thus, we transfected a RFP mitotic stage marker, found the stage of mitosis we were interested in, then used the lasers to image the spatial localization of GFP-Runx labeled protein through mitosis. Another clear advantage to laser scanning confocal microscopy is the elimination of possible bleed-through from double-labeling by turning off one of the lasers to confirm specific localization.
252
S.M. Pockwinse et al.
Fig. 17.4 Live cell microscopy by confocal laser scanning. a. The Bioptics Live Cell StageClosed System Chamber allows the viewing of living cells by maintaining 37 °C temperature; pH and nutrient supply by perfusing medium across a cell-laden coverslip. The slide heater (1) warms an aqueduct slide (4) to 37 °C. The perfusion ring chamber (2) allows medium to enter, cross and exit the chamber, and gaskets (3, 5) sandwich the aqueduct slide and cell-laden coverslip (6) to prevent leaks. This assembly sits in the self-locking base (7) that is placed on the microscope stage. b. Live image of U2-OS cell in anaphase. Runx2-EGFP foci (green) localize to mitotic chromosomes in anaphase. Histone 2B-RFP and differential interference contrast (DIC) images are used to identify mitotic stage. c. Initially, using high intensity laser power, a spot or region of interest inside the fluorescent cell is bleached. After bleaching, a series of images are taken at predetermined time intervals to measure the recovery of fluorescence in the bleached spot (left panel). Y-axis represents the relative fluorescence after photo bleaching in the bleached spot and x-axis represents time in seconds. Post-normalization, the relative fluorescence in the bleached spot is zero. This time point is represented as zero time. The relative fluorescence increases with time until it reaches asymptote (right panel).
1. Sterilize 40-mm coverslips set in 60-mm dishes in a microwave for 20 minutes. 2. Plate cells (see Note 7) in McCoy’s 5A complete medium at a density of ∼0.3 × 106. Allow cells to grow for 40–48 hours to 50–80% confluency. 3. Transfection. Ascertain and document cell growth. Warm complete and incomplete McCoy’s 5A media. Using Roche FuGENE 6 Transfection Reagent, follow the standard protocol for a 60-mm dish using 5 mL total volume of medium, 200– 600 ng of DNA and 200 µL total volume of complex per coverslip (see Note 8).
3.4.3
Preparation of Live Cell Stage, a Closed System Chamber
The Focht’s Closed System (FCS) chamber allows the microscopic observation of living cells by duplicating conditions of an incubator on the microscope. Temperature is controlled by a slide heater and an objective heater. The slide heater works in conjunction with a micro-aqueduct slide that has a thermally conductive coating, on
17 Organization of Regulatory Machinery
253
which the slide heater arms rest. The temperature is set by a controller unit. The objective heater’s temperature is also set by a controller unit and has an adjustable loop that surrounds the objective lens. These heaters are designed to eliminate heat-sink loss. Over time, cells under microscopic observation must be fed, pH level maintained, and waste eliminated. A micro-peristaltic pump working in conjunction with a micro-aqueduct slide allows media to perfuse across the coverslip at a precise, very slow rate, feeding the cells, maintaining pH, and eliminating waste. 1. Pre-warm the following items to 37 °C. One hour prior to using the confocal microscope, warm water bath to 37 °C, placing a flask with water and a thermometer in the bath to verify bath temperature. 2. After reaching temperature, place near confocal microscope for perfusion media. 3. Warm perfusion medium, which is McCoy’s 5A medium with 20 mM Hepes buffered solution (Hepes maintains the medium pH throughout the imaging session), to 37 °C. 4. Warm the FCS and parts to 37 °C as well as an insulated transporter box with 2 × 50 mL centrifuge tubes of 37 °C water (on each side of the chamber) that will keep the chamber warm en route from the warm room to the scope. 5. One half hour prior to viewing, confirm that the objective lens and slide heater controller systems are calibrated to 37 °C, then attach the objective heater to objective lens and turn on to warm.
3.4.4
Microscope Preparation
1. Confirm DIC is aligned, correct condenser (wide field) and rotating condenser prism are in place, and stage height is correct. 2. Assemble chamber with cells in warm room. All chamber pieces, gaskets and tubing have been cleaned with 70% ethanol and distilled water after the last use and again before current use. 3. The order of assembly of the FCS is as follows (see Fig. 17.4A): Place upper gasket (part 3) on perfusion ring chamber (part 2) matching two holes over two pegs, then align the aqueduct slide (part 4) on the pegs. The aqueduct slide allows the perfusion fluid to flow across the coverslip, keeping the cells fed and warm. Place the lower gasket (part 5) on top of the aqueduct slide. Pipet 0.25 mL of medium onto the aqueduct slide from the 60-mm dish containing the coverslip with cells, filling the channels and mid-space area. Finally, invert the coverslip (part 6) with cells onto the layered assembly. It is very important to touch one edge of the coverslip to one side of the aqueduct slide and slowly lower the coverslip onto the aqueduct slide in order to avoid air bubbles. Invert this sandwich assembly and secure in the self-locking base (part 7). Wipe the medium off the underside of the apparatus. 4. Place chamber in the transporter box with the 2 × 50 mL, 37 °C warming tubes to keep it warm while walking from the warm room to the microscope room. Place the chamber on the stage, attach the inlet tubing, and turn on the dialysis
254
S.M. Pockwinse et al.
pump to a very slow rate (e.g., 0 and 2) and confirm perfusion fluid is entering, crossing, and exiting the chamber. Then attach exit tubing and drain to receiver vessel. Insert slide heater (part 1, see Fig. 17.4A) in the slide warmer receptacle and oil the objective lens.
3.5
Fluorescence Recovery after Photobleaching
It has long been suggested that sub-nuclear organelles are highly dynamic structures (10). A powerful approach for measuring the dynamics of these structures and their molecules is to track fluorescently tagged molecules in living cells. Time-lapse studies can characterize the plasticity of the structures in which those molecules are incorporated. The binding parameters of individual molecules with those structures are better determined by laser photobleaching techniques (11, 12). All structures that are not covalently cross-linked must arise from dynamic interactions between constituent molecules. These can be equilibrium or steady-state interactions, but there are always at least two pools of molecules: free and bound. Molecules will move at diffusion rates when free and at the same rate as the structure when bound. Taking advantage of this difference, photobleaching-recovery techniques can help characterize the binding equilibriums for molecules docking on a simple, stable structure. The analysis becomes complicated if the protein has multiple and heterogeneous interactions. Several studies have examined the dynamics of cellular and nuclear GFP-fusion proteins by FRAP (13–16). In FRAP, fluorescence in a spot-of-interest (SOI) or region-of-interest (ROI) inside the cell is bleached with a high intensity focused laser, so that fluorescence in this region is lost irreversibly (see Fig. 17.4C, left panel). This results in non-fluorescent molecules in the bleached areas surrounded by fluorescent molecules outside the bleached region. Since the binding of these molecules is dynamic, bleached molecules will unbind and diffuse away. Molecules that are still fluorescent and are bound in the unbleached region will unbind and diffuse into the bleached zone where they can bind. Photobleach recovery rates are determined by unbinding, diffusion, and binding rates (13, 14). For most biological FRAP experiments, diffusion is very fast relative to rates of recovery. In this case, recovery rates are a measure only of the binding equilibrium. Recovery of fluorescence in the bleached area is measured by low-intensity laser power. If the fluorescent molecules are not dynamic or very tightly bound to cellular structures, there is no measured recovery of fluorescence in the bleached area. It is worth noting that during a FRAP experiment, there is no loss of protein in the bleached region nor is there any change in the interactions between protein–DNA, protein–RNA, or protein–subcellular structures, but rather a replacement of bleached molecules with fluorescent molecules in these interactions. After bleaching, a series of images of the bleached cell are taken to measure the recovery of fluorescence in the bleached spot (see Fig. 17.4C, right panel). Most papers in the biological literature report FRAP recovery rates in terms of half time of recovery or t1/2.
17 Organization of Regulatory Machinery
255
Others report “apparent diffusion coefficients,” even though FRAP recovery rates are dependent on binding but not on diffusion (14). Therefore, measurement of binding and unbinding constants is very important in understanding FRAP recovery rates, especially for nuclear proteins (13, 14). After FRAP, data from the confocal system are exported into a spreadsheet software package, corrected for the loss of fluorescence in the whole cell and normalized for pre- and immediate post-bleach fluorescence in the bleached zone. Post-normalization, full recovery of the fluorescence in the bleached zone might be expected. If there is no full recovery, then a fraction of the protein is immobile, which means the protein is tightly bound to the sub-cellular structure and exchanges too slowly to be measured in the post-bleach session. FRAP has been valuable in many applications including, but not limited to, measurement of the binding of histone proteins to DNA (14, 17, 18), the binding of the estrogen receptor to promoters (19), the binding of the glucocorticoid receptor to promoters (20), and the dynamic binding of exon junction complex proteins to RNA splicing speckled domains (15, 16). 1. Once the live cell chamber is set up, fluorescent cells are identified and images of them are collected at low laser power. Optimum gain and offset values for images are determined and the settings are saved under a user profile. 2. Initially, 5 to 10 images are recorded before bleaching an ROI or an SOI. 3. In spot photobleaching, one or more than one spot inside the fluorescent cell is selected. Bleaching is usually done with maximum laser power from 1 to 3 seconds until about 70% of the fluorescence in the spot is bleached. If more than one spot is selected for bleaching, the confocal system performs bleaching sequentially. Bleaching can be done in larger regions of the cell, for example, in half of the nucleus, by zooming up to a ROI within the cell and scanning it 30–50 times at full laser intensity, resulting in the photobleaching of this enlarged region of the cell. 4. After photobleaching, routinely 30–50 images are taken at intervals of 1.7–20 seconds (it can be minutes if desired) depending on the dynamics of the fluorescent protein. (If large area bleaching was performed, images are enlarged to the size equal to the pre-bleach image.) 5. With the aid of Leica confocal software version 2.0, the fluorescence within the bleached spot, the fluorescence in the whole cell or nucleus, and the fluorescence in a region outside the bleach zone are measured for the entire stack of images. 6. For data analysis, fluorescence intensity values from the Leica software are exported to a spreadsheet software package, for example Microsoft Excel. Normalization is done at two levels. At the first level, the initial post-bleach intensity is subtracted for the fluorescence in the ROI so that any fluorescence in the bleached area after bleaching is normalized to zero. At the second level, the pre-bleach fluorescence intensity, corrected for the fluorescence loss in the whole cell that is caused by the bleach, is normalized to 1. 7. The relative fluorescence intensity (I rel) in the bleached spot is measured as described by (12): Irel = T0*If / Tt*I0, with T0 being the total cellular intensity before bleach, Tt the total cellular intensity at time t, I0 the intensity in the bleached area before bleach, and If, the intensity in the previously bleached area at time t.
256
S.M. Pockwinse et al.
8. Recovery curves are obtained using Microsoft Excel or Kaleidograph.3.5 (Synergy Development). 9. The immobile protein fraction is calculated by subtracting the relative intensity at the asymptote of the recovery curve from a relative recovery 1. For example, an asymptote at 0.7 reflects an immobile fraction of 30%.
3.6
Intranuclear Informatics
Intranuclear informatics is a mathematical algorithm that is designed to identify and assign unique quantitative signatures that define regulatory protein localization within the nucleus (Fig. 17.5) (6). Quantitative parameters that can be assessed include nuclear size and variability in domain number, size, spatial randomness, and radial positioning (see Fig. 17.5, top panel). The significance and implication of Intranuclear Informatics can be shown by two distinct biological examples (see Fig. 17.5). Regulatory proteins with different activities can be subjected to Intranuclear Informatics analysis that assigns each protein a unique architectural signature. The overlap between the architectural signatures of different proteins is often correlated to their functional overlap. Alternatively, the sub-nuclear organization of a protein domain can be linked with sub-nuclear targeting, biological function, and disease. For example, biologically active Runx2 and its inactive sub-nuclear targeting defective mutant (mSTD) show distinct architectural signatures, indicating that the biological activity of a protein can be defined and quantified as sub-nuclear organization. These architectural signatures have the potential to discriminate between intranuclear localization of proteins in normal and cancer cells. Intranuclear informatics can be combined with proteomics (changes in protein–DNA and protein–protein interactions) and genomics (altered gene expression profiles) to develop a novel platform for identification and targeting of perturbed regulatory pathways in cancer cells.
4
Notes
1. In this method we are using U2-OS cells as a model; however, the procedure can be used for any cell type provided appropriate culture medium is substituted to support cell growth. 2. The DNA concentration depends on your protein of interest. Try several dilutions to find the concentration that yields optimum transfection and immunofluorescence. 3. All procedures described in the preceding are performed in a warm (at 37 °C) room. 4. Twenty-four hours after plating, remove one plate of actively proliferating cells and fix cells in 70% ethanol (stored at 4 °C until use). This will be the asynchronous control for FACS analysis. Also remove one plate each for RNA isolation and
17 Organization of Regulatory Machinery
257
Fig. 17.5 Intranuclear informatics. This figure shows how intranuclear informatics can be used to examine nuclear alterations in cancer cells compared with normal cells. The top panel shows the conceptual framework for the quantitation of sub-nuclear organization by intranuclear informatics. Four main groups of parameters, selected based on inherent biological variability, can be examined. Example 1. Regulatory proteins with different activities can be subjected to intranuclear informatics analysis, which assigns each protein a unique architectural signature. The overlap between the architectural signatures of different proteins is often correlated to their functional overlap. Shown here are Runx transcription factor, SC35 splicing protein, and RNA Pol II. Example 2. The sub-nuclear organization of Runx domains is linked with sub-nuclear targeting, biological function, and disease. Biologically active Runx2 and inactive sub-nuclear targeting defective mutant (mSTD) show distinct architectural signatures, indicating that the biological activity of a protein can be defined and quantified as sub-nuclear organization.
protein extract preparation. These are the asynchronous controls for RNA and protein analyses. 5. Some cell synchronies require harvesting samples at time points only a couple of hours apart from each other. Consequently, preparing samples for microscopy at every time point is challenging. In this case a different, time-consuming approach must be taken. We routinely plate, block, and release cells at different times in a way that all samples will be ready at the same time, i.e., by the end of the synchrony. This allows us to handle the samples for microscopy concomitantly (see Fig. 17.1A for a T98G cell synchrony). 6. If the metaphase chromosomes are highly condensed, use a lower concentration of colcemid and decreased time of colcemid treatment.
258
S.M. Pockwinse et al.
7. Appropriate hypotonic treatment is vital to the quality of metaphase spreads. The concentration of KCl can be changed according to the quality of chromosome spread. 8. Drop the cell suspension on cold, slightly moist slides. Chromosomes will spread poorly on dry slides. If necessary, breathe on the slide before dropping the suspension. Acknowledgments Studies reported in this article were in part supported by grants from NIH (5PO1CA82834-05, 5PO1AR048818-05, 2R01GM32010, 5R01AR049069). The authors thank Judy Rask for editorial assistance with preparation of the manuscript.
References 1. Stein, G. S., Zaidi, S. K., Braastad, C. D., et al. (2003). Functional architecture of the nucleus: organizing the regulatory machinery for gene expression, replication and repair. Trends Cell Biol 13, 584–592. 2. Zaidi, S. K., Young, D. W., Choi, J. Y., et al. (2004). Intranuclear trafficking: organization and assembly of regulatory machinery for combinatorial biological control. J Biol Chem 279, 43363–43366. 3. Zaidi, S. K., Young, D. W., Choi, J. Y., et al. (2005). The dynamic organization of gene-regulatory machinery in nuclear microenvironments. EMBO Rep 6, 128–133. 4. Zink, D., Fischer, A. H., Nickerson, J. A. (2004). Nuclear structure in cancer cells. Nat Rev Cancer 4, 677–687. 5. Taatjes, D. J., Marr, M. T., Tjian, R. (2004). Regulatory diversity among metazoan co-activator complexes. Nat Rev Mol Cell Biol 5, 403–410. 6. Young, D. W., Zaidi, S. K., Furcinitti, P. S., et al. (2004). Quantitative signature for architectural organization of regulatory factors using intranuclear informatics. J Cell Sci 117, 4889–4896. 7. Henegariu, O., Heerema, N. A., Lowe, W. L., et al. (2001). Improvements in cytogenetic slide preparation: controlled chromosome spreading, chemical aging and gradual denaturing. Cytometry 43, 101–109. 8. Claussen, U., Michel, S., Muhlig, P., et al. (2002). Demystifying chromosome preparation and the implications for the concept of chromosome condensation during mitosis. Cytogenet Genome Res 98, 136–146. 9. Deng, W., Tsao, S. W., Lucas, J. N., et al. (2003). A new method for improving metaphase chromosome spreading. Cytometry A 51, 46–51. 10. Berezney, R., Basler, J., Kaplan, S. C., et al. (1979). The nuclear matrix of slowly and rapidly proliferating liver cells. Eur J Cell Biol 20, 139–142. 11. Kruhlak, M. J., Lever, M. A., Fischle, W., et al. (2000). Reduced mobility of the alternate splicing factor (ASF) through the nucleoplasm and steady state speckle compartments. J Cell Biol 150, 41–51. 12. Phair, R. D., Misteli, T. (2000). High mobility of proteins in the mammalian cell nucleus. Nature 404, 604–609. 13. Lele, T., Oh, P., Nickerson, J. A., et al. (2004). An improved mathematical approach for determination of molecular kinetics in living cells with FRAP. Mech Chem Biosyst 1, 181–190. 14. Lele, T., Wagner, S. R., Nickerson, J. A., et al. (2006). Methods for measuring rates of protein binding to insoluble scaffolds in living cells: histone H1-chromatin interactions. J Cell Biochem 99, 1334–1342. 15. Wagner, S., Chiosea, S., Ivshina, M., et al. (2004). In vitro FRAP reveals the ATP-dependent nuclear mobilization of the exon junction complex protein SRm160. J Cell Biol 164, 843–850.
17 Organization of Regulatory Machinery
259
16. Wagner, S., Chiosea, S., Nickerson, J. A. (2003). The spatial targeting and nuclear matrix binding domains of SRm160. Proc Natl Acad Sci U S A 100, 3269–3274. 17. Lever, M. A., Th’ng, J. P., Sun, X., et al. (2000). Rapid exchange of histone H1.1 on chromatin in living human cells. Nature 408, 873–876. 18. Misteli, T., Gunjan, A., Hock, R., et al. (2000). Dynamic binding of histone H1 to chromatin in living cells. Nature 408, 877–881. 19. Stenoien, D. L., Patel, K., Mancini, M. G., et al. (2001). FRAP reveals that mobility of oestrogen receptor-α is ligand- and proteasome-dependent. Nat Cell Biol 3, 15–23. 20. Stavreva, D. A., Muller, W. G., Hager, G. L., et al. (2004). Rapid glucocorticoid receptor exchange at a promoter is coupled to transcription and regulated by chaperones and proteasomes. Mol Cell Biol 24, 2682–2697.
Chapter 18
Bioluminescent Imaging in Bone Yoram Zilberman, Yossi Gafni, Gadi Pelled, Zulma Gazit, and Dan Gazit
Abstract Monitoring gene expression in vitro and in vivo, is crucial when analyzing osteogenesis and developing effective bone gene therapy protocols. Until recently, molecular analytical tools were only able to detect protein expression either in vitro or in vivo. These systems include histology and immunohistochemistry, fluorescent imaging, PET (micro-PET), CT (micro-CT), and bioluminescent imaging. The last is the only system to date that can enable efficient quantitative monitoring of gene expression both in vitro and in vivo. Effective bioluminescent imaging in bone can be achieved by using transgenic mice harboring the luciferase reporter gene, downstream of an osteogenesis specific promoter. The aim of this chapter is to comprehensively describe the various protocols needed for the detection of bioluminescence in bone development and repair. Keywords Luciferase, luciferin, cooled charge coupled device imaging, bioluminescence, gene expression, bone development, bone regeneration, stem cells.
1
Introduction
Tracking gene expression is important in developmental studies and gene therapy systems, which aim to reveal the spatial, temporal, and intensity of expression patterns of the gene of interest. However, it requires the joint use of several techniques, which are mostly invasive. Non-invasive quantitative monitoring of gene expression can potentially measure all three parameters without the use of postmortem tissue assay techniques and therefore offers an important tool for tracking gene expression in vivo (1–3). Reporter genes are widely used for monitoring gene expression in prokaryotic, plant, and mammalian systems in cells, reconstituted systems, and whole organisms. The firefly luciferase gene, which codes for an enzyme that catalyzes the oxidation of luciferin in the presence of ATP to generate light, and the green fluorescent proteins (GFP) are commonly used reporters. In recent years, new technologies that facilitate detection of the amount of light emitted from mammalian internal organs were developed (4). From: Methods in Molecular Biology, Vol. 455: Osteoporosis: Methods and Protocols Edited by Jennifer J. Westendorf © Humana Press, Totowa, NJ
261
262
Y. Zilberman et al.
These technologies are based on the ability of photons to pass through mammalian tissue despite its light-absorbing and -scattering properties, thus enabling us to visualize gene expression non-invasively in intact animals. A cooled charged coupled device (CCCD) camera is a molecular imaging system that can be used to visualize low quantities of photons emitted by internal mammalian tissue both in vitro and in vivo (1–3). Osteocalcin (OC) is a bone tissue-specific protein expressed by osteoblasts, odontoblasts, and hypertrophic chondrocytes at the onset of tissue mineralization and it accumulates in the bone extracellular matrix (5, 6). OC that has affinity for the mineral phase in bone is expressed at maximal levels in the subsequent stage of differentiation at the onset of mineralization: post-proliferatively in osteoblasts, pre-osteocytes, and mature osteocytes (5). Some newly synthesized and secreted OC circulates in blood. A segment of the human OC promoter was used in previous studies for different purposes, including the analysis of OC promoter regulation and as an osteogenic tissue specific promoter (7–9). Clemens et al. (10) cloned preosteoblasts containing a stably integrated human OC promoter-luciferase construct (hOC-Luc) to map bone-specific promoter elements. This section presents the use of transgenic mice harboring a luciferase marker gene under the regulation of the human osteocalcin (hOC) promoter. The luciferase marker gene is expressed only when the osteocalcin promoter is induced, i.e., when the cells undergo osteogenic differentiation (Fig. 18.1). Transgenic hOC-Luc mice allow the investigation of OC regulation through bone remodeling and mesenchymal stem cell (MSC) osteogenic differentiation, in vivo and in vitro utilizing the bioluminescent imaging system (Figs. 18.2 and 18.3).
Fig. 18.1 Immunohistochemistry for luciferase. Right: Murine skeletal muscle tissue used as positive control. Left: Murine experimental tissue of newly formed bone. Arrows indicates positive cells. B, Bone; CT, connective tissue; M, skeletal muscle.
18 Bioluminescent Imaging in Bone
b
263
Luciferase expression of C3H-GFP/Luc clones 3.00E+06 2.50E+06 2.00E+06 1.50E+06 1.00E+06 5.00E+05 0.00E+00 C3H-wt
C3H-GFP
C3H-GFP-Luc
Clones Fig. 18.2 In vitro analysis of C3H-GFP/Luc clone. C3H-GFP/Luc clonal cells were cultured in 24 well plate and selected for Luc expression. Quantitative luciferase expression in culture was assessed by bioluminescent analysis. Luciferin solution was added to the cultures as described in Methods. Bioluminescence signal was detected in several clones (a) and quantified; highest expressing clone versus negative controls is presented in graph (b).
2
Materials
2.1 Tissue Fixation, Embedding, Sectioning, and Decalcification 1. Fixative solution: Use one recommended for a corresponding antibody. 2. Decal solution: EDTA 0.5 M, pH 7.4. Dissolve EDTA in saline 0.9% not PBS. Bring 186.1 g Na2EDTA • 2H2O to 700 mL of saline 0.9%, and put on the stirrer. Dissolve as much as possible, but it might not dissolve completely. Adjust pH to 7.4 with 10 M NaOH (~50 mL). Add saline up to 1 L. Check pH again, and adjust it if necessary.
264
Y. Zilberman et al.
Fig. 18.3 In vivo temporal monitoring of osteogenesis: Genetically engineered adult MSCs, expressing the osteogenic gene BMP-2 in a Tet-OFF system were transplanted into a calvarial defect inflicted in hOC-Luc transgenic mice.
3. EtOH solutions: 50%, 70%, 95%, 100%. 4. Xylene. 5. Paraffin.
2.2
1. 2. 3. 4. 5.
In Situ Detection of Luciferase Expression (Immunohistochemistry)
4% formaldehyde/PBS. EDTA 0.5 M (pH = 8). Paraffin. Glass slides (Superfrost). ZYMED HistoStain SP Kit, rabbit primary, AEC (cat. no. 95-6143) (see Note 1). a. Reagent 1A. One dropper bottle (105 mL) of ready-to-use serum blocking solution, 10% goat non-immune serum. b. Reagent 1B. One dropper bottle (105 mL) of ready-to-use biotinylated second antibody. c. Reagent 2. One dropper bottle (105 mL) of ready-to-use streptavidinperoxidase (HRP).
6. Biotin-conjugated secondary antibody. The authors use a rabbit polyclonal antibody (Zymed 626140), diluted 1:100 in PBS. 7. Humidity chamber (see Note 2).
18 Bioluminescent Imaging in Bone
265
8. Primary antibody against luciferase (Cortex cat. no. CR2029RAP). Make aliquots from the antibody 1:100 dilutions in PBS×1. 9. Peroxidase quenching solution: 3% H2O2 in methanol. Add 1 part of 30% H2O2 to nine parts of absolute methanol and mix well. 10. Dako-Pen (SIGMA Z37,782-l). 11. Xylene. 12. Ethanol: 100%, 95%. 13. Double distilled water (DDW) and PBS×1. 14. Citrate buffer × 20 (Zymed cat. no. 00-5000). 15. Pepsin solution Digest-All 3 (Zymed cat. no. 00-3009). 16. CAS-block (Zymed cat. no. 00-8120).
2.3 1. 2. 3. 4.
Protein and Luciferase Assays
Phosphate-buffered saline ×1 (PBS×1) solution. Cell culture lysis 5× reagent (Promega cat. no. E1531). Luciferase assay system substrate (Promega cat. no. E151A). BCA Protein Assay Kit (Pierce, Rockford, IL). a. BCA reagent A, containing sodium carbonate, sodium bicarbonate, bicinchoninic acid, and sodium tartrate in 0.1 M sodium hydroxide. b. BCA reagent B, containing 4% cupric sulfate. c. Albumin standard, 2 mg/mL, 10 × 1 mL ampules, containing bovine serum albumin (BSA) at 2.0 mg/mL in 0.9% saline and 0.05% sodium azide.
5. BCA working reagent: Add 50 parts of BCA reagent A to one part of BC A reagent B mix well (clear green color is observed). You will need 200 µL for each replication of the sample (it is recommended to triple each sample).When stored in a closed container at RT, working reagent is stable for at least 1 day. 6. Turner Designs Luminometer (Turner Designs, Sunnyvale, CA). Preparation of diluted bovine serum albumin (BSA) standards: for working range of 20 to 2,000 µg/mL in concentration: Volume of the BSA
Volume of the diluent
Final BSA concentration
300 µL stock 375 µL of stock 325 µL of stock 175 µL of A 325 µL of B 325 µL of D 325 µL of E 100 µL of F
0 µL 125 µL 325 µL 175 µL 325 µL 325 µL 325 µL 400 µL
2000 µg/mL 1500 µg/mL A 1000 µg/mL B 750 µg/mL C 500 µg/mL D 250 µg/mL E 125 µg/mL F 25 µg/mL G
Note: For working range of 5–250 µg/mL protein concentration there is a different way to prepare diluted BSA standards (see original protocol for the BCA Protein Assay Kit.)
266
2.4
Y. Zilberman et al.
Molecular Imaging Device
1. The imaging unit consists of an intensified Peltier cooled charge-coupled device (CCCD), model LN/CCD-1300EB, which is equipped with an ST-133 controller and a 50-mm Nikon lens (Roper Scientific, Princeton Instrument, Trenton, NJ). In this system, a pseudo color image represents light intensity (with blue representing least intense and red most intense). In all cases, the integrated light is the result of a 2-min exposure and acquisition. The exposure conditions (including time, f-stop, position of the stage, binding ratio, and time after injection with luciferin) are maintained at identical levels so that all measurements are comparable. 2. This CCCD camera converts photons to electrons with an efficiency of about 90% in the wavelength emission range of luciferase-luciferin and converts about 1.8 electrons per count. Thus, every count represents two photons. The measurement is a total integrated signal of a constant exposure interval. All measurements can be verified by using an ROI that is progressively larger and then subtracting backgrounds that are the same. An increase in ROI size has no appreciable effect on the total measured intensity. In summary, measurements are the total integrated signal of a standard time interval subtracted by a background area of equal size. This is equivalent to the peak volume measurement.
2.5 1. 2. 3. 4.
3
In vivo Imaging
Beetle luciferin (Promega Corp., Madison, WI). Ketaset (equivalent to 100 mg/mL ketamine) (see Note 21). Xylazine 2% (see Note 22). PBS×1, sterile.
Methods
3.1 Tissue Fixation, Embedding, Sectioning, and Decalcification The following is a brief description of immunohistochemistry previously described (11). Operated and control organs should be fixed in 4% formaldehyde/PBS for 3 days, and then decalcified with EDTA 0.5 M (pH = 8) for 14 days. Samples will be then, embedded in paraffin, and cut into 5-mm–thick sections. 1. Fix tissue in appropriate fixative solution recommended for a corresponding antibody. The time and conditions of the fixation are mentioned in the antibodies description.
18 Bioluminescent Imaging in Bone
267
2. In the case of hard tissue when decalcification is required: remove samples from fixative, wash in PBS×1 15 minutes × 2. In the case of soft tissue when decalcification is not required, after appropriate fixation go to step 5. 3. Transfer samples to Decal solution: EDTA 0.5 M, pH 7.4. 4. Wash in PBS×1 three times, 1 hour each step. 5. Process samples through 9 hours of dehydration, clearing and paraffin embedding as follows: a. b. c. d. e. f. g. h. i. j. k.
50% EtOH 45 minutes. 70% EtOH 45 minutes. 95% EtOH 45 minutes. 100% EtOH 45 minutes. 100% EtOH 45 minutes. 100% EtOH 45 minutes. Xylene 45 minutes. Xylene 45 minutes. Xylene 45 minutes. Paraffin 60 minutes in the oven (70 °C). Paraffin 60 minutes in the oven (70 °C).
6. Following infiltration of the tissue with paraffin, samples should be embedded into paraffin block for storage until microtome sectioning.
3.2
In Situ Detection of Luciferase Expression (Immunohistochemistry)
Paraffin sections will be transferred to glass slides. Slides are incubated at 37 °C overnight, followed by deparaffinization with xylene. Sections are subsequently hydrated in a descending series of ethanol baths and rinsed in PBS. Endogenous peroxidase activity is re-enabled by treatment with H2O2. Sections are incubated with blocking solution before primary luciferase antibody is applied. The slides are then rinsed in PBS, and a secondary goat anti-rabbit IgG antibody is applied to the slides. After washing with PBS, slides are incubated with horseradish peroxidase (HRP) conjugated to streptavidin, followed by 3-amino-9-ethyl-carbazole (AEC) dye for visualization with light microscopy. Finally, slides are stained with hematoxylin, washed, and mounted. 1. Heat sections overnight at 37 °C to promote adherence of the tissue to the slide (see Note 3). 2. Deparaffinize in xylene (2 × 3 minutes) (see Note 4). 3. Dehydrate in ethanol 100% (2 × 3 minutes) (see Note 5). 4. Dehydrate in ethanol 95% (2 × 3 minutes) (see Note 6). 5. Rinse in DDW (2 × 3 minutes) (see Note 7). 6. If pepsin will be used, put it in 37 °C to warm at this stage (see Note 8).
268
Y. Zilberman et al.
7. Do not allow the specimens to dry from this point on. Incubate specimens in the humidity chamber (see Note 2). 8. Transfer to peroxidase quenching solution, which blocks endogenous peroxidase activity, for 10 minutes (see Note 9). 9. Rinse well with 1 × PBS (3 × 2 minutes) (see Note 10). 10. Use either one of the following techniques: Transfer slides to a plastic container with citrate buffer × 1. Run in the microwave for 2–5 minutes to reveal the antigen (see Note 11). a. Microwave antigen retrieval high power, until the buffer boils. Run on medium power for 10 minutes in which cycles of boiling are seen. Leave for 20 minutes to cool. b. Use a DAKO-PEN to create a border around specimens. Apply a few drops of pepsin (37 °C) on each specimen, and incubate for 5 minutes at 37 °C. 11. Transfer to 1 × PBS (3 minutes). 12. Use a DAKO-PEN to create a border around the specimens (if the microwave technique was used) (see Note 12). 13. Carefully blot each solution from the specimens in the following steps (see Note 13). 14. Apply two drops of serum blocking solution (ready to use reagent 1a) on each specimen. Incubate 10 minutes at room temperature in the humidity chamber. 15. Apply two drops of CAS blocking solution on each specimen. Incubate for 30 minutes at RT in the humidity chamber. 16. Apply primary antibody on each sample. On negative control samples, apply PBS × 1 instead. Incubate for 60 minutes at room temperature in the humidity chamber. 17. Rinse well with PBS×1 (3 × 2 minutes) (see Notes 10 and 14). 18. Apply two drops of anti-rabbit biotinylated secondary antibody (ready to use reagent 1b) on each specimen. Incubate for 10 minutes at room temperature. 19. Rinse well with PBS×1 (3 × 2 minutes) (see Note 10). 20. Apply two drops of enzyme conjugate (ready to use reagent 2) on each specimen. Incubate for 10 minutes at room temperature in the humidity chamber. 21. Rinse well with 1 × PBS (3 × 2 minutes) (see Note 10). 22. Prepare substrate Chromagen mixture: Add one drop of reagent 3a, one drop of 3b, and one drop of 3c to 1 mL of DDW. Mix well, protect from light, and use within 1 hour (see Note 15). 23. Apply two drops of substrate Chromagen mixture to each specimen. Incubate 10 minutes at room temperature in the humidity chamber. 24. Rinse well with DDW (3 × 2 minutes) (see Notes 10 and 16). 25. Warm your mounting solution to 37 °C in a water bath (see Note 17). 26. Apply two drops of hematoxylin solution (ready-to-use reagent 4) to each specimen. Incubate for 1–3 minutes at room temperature in the humidity chamber. 27. Rinse in tap water for 1 minute. 28. Rinse in PBS × 1 for 30 seconds, until specimens turn blue (see Note 10).
18 Bioluminescent Imaging in Bone
269
29. Move slides to DDW (see Note 10). 30. Apply mounting solution (ready-to-use reagent 5) and cover the slides (see Note 18). 31. Positive specimens appear red-brown (see Note 1).
3.2.1
Luciferase Assay
1. Add four volumes of DDW to one volume of Cell Culture Lysis 5× reagent. Equilibrate the 1× lysis reagent at RT before use. Diluted cell culture lysis reagent is good for 48 hours. 2. Carefully remove the growth medium from the cells to be assayed. Rinse cells with PBS, being careful not to dislodge attached cells. 3. Add enough 1× cell culture lysis reagent to cover the cells (e.g., 250 uL/24-well, 300 µL/30-mm culture dish, 400 µL/60-mm culture dish, 900 uL/100-mm culture dish, or 20 µL per well of a 96-well plate). 4. Rock culture dishes several times to ensure complete coverage of the cells with the lysis buffer. Scrape attached cells from the dish using 200-µL tips (use the base side). Transfer cells and all liquid to Eppendorf tubes and place on ice. 5. Vortex 10–15 seconds, then centrifuge at 12,000 g for 15 seconds at RT or for up to 2 minutes at 4 °C. Transfer the supernatant to a new tube. (The pellet will not always be visible, so if total volume is 200 µL, transfer only 160 µL, etc.). 6. The cell lysates may be stored at −70 °C or assayed directly with a luminometer. 7. Equilibrate the substrate and cell lysates at room temperature (see Note 19). Transfer 100 µL substrate to a fresh transparent Eppendorf tube. Add 20 µL of the sample, mix by vortexing, and assess activity. 8. Multiply the reading by 1,000 to obtain the measurement in RLUs (relative light units). Divide RLUs by the protein concentration to determine activity per microgram.
3.2.2
BCA Protein Assay
1. Pipette 25 µL of each standard or unknown sample into the appropriate microwell plate wells. Use 25 µL of the diluent for the blank wells. 2. Add 200 µL of the WR to each well, and mix the plate well on a plate shaker for 30 seconds. 3. Protect from light and incubate plate at 37 °C for 30 minutes. 4. Measure the absorbance at 562 nm on a plate reader. (Wavelengths from 540 to 590 nm have been used successfully.) 5. Subtract the average A (562) reading for the blanks from the A (562) reading for each standard or unknown sample. 6. Prepare a standard curve by plotting the average blank-corrected A (562) reading for each BSA standard versus its concentration in µg/mL. Using the standard curve, determine the protein concentration for each unknown sample.
270
Y. Zilberman et al.
3.3
Bioluminescence
3.3.1
Bioluminescence Analysis In Vitro
1. Apply luciferin to cell culture medium prior to imaging (315 µg/mL). 2. Place plates in the dark chamber of the Roper Chemiluminescence Imaging system (Roper Scientific, Princeton, NJ). 3. Shooting: A grayscale surface reference image (digital photograph) should be taken under weak illumination, then switch off the light source, and another image should be taken of photons emitted from luciferase expressing cells. 4. Photon emission is then integrated over a period of 2 minutes and recorded as pseudocolor images (1–3). 5. Analyze image with the aid of computer software. 3.3.2
Bioluminescence Analysis In Vivo
1. Anesthetize mice with ketamine/xylazine mixture (1 µL/g body weight) (see Notes 21 and 22). 2. Inject luciferin aqueous solution (126 mg/kg body weight) intraperitoneally (ip) 5 minutes prior to imaging (see Note 23). 3. Place animals in the dark chamber (see Note 24). 4. Capture and analyze image as described in the preceding section.
4
Notes
1. Always make sure that you use the right immunohistochemistry kit for your antibody. We prefer a specific secondary antibody rather than a broad-spectrum reagent (non-specific staining is more common with the latter). Use of DAB staining system will generate stable staining. If using AEC staining, make sure to photograph your slides within 24 hours from the end of the procedure because the slides tend to lose their stain with time. 2. A humidity chamber is fabricated by placing some tissue paper inside a closed glass or plastic container big enough to hold several slides in horizontal state. Place Pasteur pipettes on the tissue paper in small intervals, allowing room for the slides. Soak the paper in DDW. During incubations make sure that you close the lid of the container to prevent the slides from dehydrating. When storing the chamber, always store with the lid open, otherwise fungus will grow inside. 3. Make sure that the sections are incubated for at least 12 hours at 37 °C, so that they adhere sufficiently to the slides. Do not incubate in higher temperature, since it will decrease the signal. Keep slides and blocks at 4 °C prior to staining.
18 Bioluminescent Imaging in Bone
271
4. Use fresh and clean xylene. Transfer the slides with care, since the sections can easily fall off the slides. 5. Use fresh and clean ethanol. 6. Dilute ethanol with filtered DDW. 7. Use filtered DDW in this stage. 8. Do not heat all your pepsin each time. Aliquot the amount needed and heat it only. 9. Make sure to use clean solutions. This stage is very important in order to avoid unspecific staining. 10. Put the slides carefully in the solution and incubate for the time noted. Be careful not to damage the sections in the process. Do not reuse solutions in the washing procedure, use only fresh and clean solutions. This applies also for steps 17, 19, 21, 24, 28, and 29. 11. The microwave technique is more aggressive than the use of pepsin. We suggest using both on a small number of slides to compare and determine which technique is better for your specimens. 12. Make sure no leakage occur between different specimens mounted on the same slide. Draw the DAKO-PEN line as close to the specimen as you can. 13. Use some folded tissue paper to blot solutions carefully. Do not touch specimens with the paper. 14. Make sure that no leakage occurs between specimens. Blot the solution before rinsing with PBS×1. 15. Be sure to wrap aluminum foil around your tube to protect your reagent from light. Prepare approximately 100 µL solution per specimen (depends on the specimen size). 16. Wash the slides by gently moving them up and down. Change water when they are to be stained with hematoxylin. 17. Warming the mounting medium will prevent the formation of bubbles under the mounting glass. 18. Make sure that no bubbles are accumulated in the specimen area beneath the covering slide. Apply gentle pressure with pincers on the glass in order to remove such bubbles. 19. Samples equilibrated with substrate must be protected from light until activity assessment. 20. Ensure that the time period from adding the luciferin solution until reading will be identical for all samples. 21. When imaging for the first time, take a control picture before luciferin injection to eliminate auto-luminescence of your subject. 22. Add negative controls to your assay. 23. Ensure that the time period from luciferin injection until shooting will be identical for all animals. 24. When BLI detection site is located on the animal’s ventral side, it is recommended to place the animal on its back.
272
Y. Zilberman et al.
References 1. Bar, I., Zilberman, Y., Zeira, E., et al. (2003) Molecular Imaging of the skeleton: quantitative real time bioluminescence in transgenic mice. J Bone Miner Res 18, 570–578. 2. Wu, J. C., Sundaresan, G., Iyer, M., et al. (2001) Noninvasive optical imaging of firefly luciferase reporter gene expression in skeletal muscles of living mice. Mol Ther 4, 297–306. 3. Honigman, A., Zeira, E., Ohana, P., et al. (2001) Imaging transgene expression in live animals. Mol Ther 4, 239–249. 4. Contag, P. R., Olomu, I. N., Stevenson, D. K., et al. (1998) Bioluminescent indicators in living mammals. Nat Med 4, 245–247. 5. Lian, J. B., Stein, G. S., Stein, J. L., et al. (1998) Osteocalcin gene promoter: unlocking the secrets for regulation of osteoblast growth an differentiation. J Cell Biochem 30/31, 62–72. 6. Hauschka, P. V., Lian, J. B., Cole, D. E. C., et al. (1989) Osteocalcin and matrix Gla protein: vitamin K-dependent proteins in bone. Physiol Rev 69, 990–1047. 7. Sims, N. A., White, C. P., Sunn, K.L., et al. (1997) Human and murine osteocalcin gene expression: conserved tissue restricted expression and divergent responses to 1,25-dihydroxyvitamin D3 in vivo. Mol Endocrinol 11, 1695–1708. 8. Thomas, G. P., Bourne, A., Eisman, J. A., et al. (2000) Species-divergent regulation of human and mouse osteocalcin genes by calciotropic hormones. Exp Cell Res 258, 395–402. 9. Zhao, G., Monier-Faugere, M. C., Langub, M. C., et al. (2000) Targeted overexpression of insulin-like growth factor I to osteoblasts of transgenic mice: Increased trabecular bone volume without increased osteoblast proliferation. Endocrinology 141, 2674–2682. 10. Clemens, T. L., Tang, H., Maeda, S., et al. (1997) Analysis of osteocalcin expression in transgenic mice reveals a species difference in vitamin D regulation of mouse and human osteocalcin genes. J Bone Miner Res 12, 1570–1576. 11. Lavon, I., Goldberg, I., Amit, S., et al. (2000) High susceptibility to bacterial infection, but no liver dysfunction, in mice compromised for hepatocyte NF-kappaB activation. Nat Med 6, 573–577.
Chapter 19
Micro-Computed Tomography: A Method for the Non-Destructive Evaluation of the Three-Dimensional Structure of Biological Specimens Martin Stauber and Ralph Müller
Abstract The large increase in interest in micro-computed tomography (microCT) over the last decade reflects the need for a method able to non-destructively visualize the internal three-dimensional structure of an object. Thereby, the real beauty of computed tomography lies in the fact that it is available for a large range of nominal resolutions, which allows hierarchical imaging from whole bodies down to the tissue level. Although micro-CT is currently mainly used for imaging of hard tissue (i.e., bone and tooth), future developments might also allow high soft tissue contrast either using appropriate contrast agents or x-ray contrast mechanisms. This chapter aims to review the steps necessary for a successful micro-CT measurement. Although the actual measurement is often machine dependent, the chapter does not describe a specific system but rather lists all steps that eventually have to be considered to set up a measurement, run the measurement, process the image data, and get morphometric indices as a result. The chapter provides an easy understandable manual that should allow newcomers to perform successful measurements and hence to best profit from this powerful technique. Keywords Computed tomography, micro-computed tomography, three-dimensional imaging, non-destructive imaging, 3D structural analysis, morphometry.
1 1.1
Introduction History
Computed tomography (CT) is a non-destructive technique that provides threedimensional images of the internal structure of an object. The basic idea of this imaging technique goes back to J. Radon, who proved in 1917 that an n-dimensional From: Methods in Molecular Biology, Vol. 455: Osteoporosis: Methods and Protocols Edited by Jennifer J. Westendorf © Humana Press, Totowa, NJ
273
274
M. Stauber and R. Müller
object can be reconstructed from its (n-1)-dimensional projections (1). However, the mathematical basis for the actual CT image reconstruction was presented in two papers by Cormack (2, 3) in 1964 and 1965, respectively. About 10 years later, Hounsfield submitted a patent, describing the first CT scanner (4, 5), which was then built in 1975 (6). The possibility of non-invasively imaging three-dimensional sections of a human body was of such importance that Cormack and Hounsfield were awarded with the Nobel Prize for Medicine in 1979. Because CT provides an excellent contrast of bone to soft tissue, new medical developments of this technique were often driven by the bone research field. Feldkamp et al. (7) were first to build a micro-computed tomography (micro-CT) scanner for the evaluation of the three-dimensional micro-structure of trabecular bone. At this stage, micro-CT was an experimental technique available to a few research groups only. However, with the presentation of the first commercially available bone micro-CT scanner in 1994 (8, 9), this technique became quickly a standard in bone research. Nowadays, micro-CT scanners are available from several manufacturers, who provide a whole palette of different scanner types targeting many applications that range from in-vivo measurements down to the analysis of bone tissue on a micrometer scale. With the advent of third generation synchrotron radiation (SR) facilities, microCT with resolutions of 1 µm and even better became feasible (10). Synchrotron radiation (SR) has several advantages over x-ray, including its high brilliance, which allows the achievement of higher resolutions (11) and the use of a monochromatic beam for accurate density representations (12, 13). The high-resolution of SR micro-CT and nano-CT systems (systems with resolutions <1 µm) opened a new level of magnification, which in bone research was necessary to investigate threedimensional properties of ultrastructural features, such as osteocyte lacunae, intrabone vasculature, or micro-cracks and micro-fractures. With the ongoing development of CT systems, this technique became available on many different levels of resolution, while always using the exact same physical working principle. Thus, CT is an excellent technique for investigation in a hierarchical fashion ranging from whole bodies down to the sub-cellular level (Fig. 19.1). With the development of micro-CT, complementary techniques as well as new image processing algorithms and analysis techniques have evolved. These new techniques opened the field of micro-CT to many new applications, such as the analysis of trabecular bone samples under linear compression (Fig. 19.2A) (14, 15), analysis of bone in growth in implants (Fig. 19.2B), evaluation of the performance of bone replacement material in fracture repair (Fig. 19.2C), investigation of optimal implant design and coating (Fig. 19.2D), and many other applications. Future developments in micro-CT will include improving the systems with respect to speed, increasing the spatial resolution, and developing new imaging modes. A very promising imaging mode is the so-called phase contrast micro-CT, in which not only the absorption of the photons but also their phase shift can be measured. This method allows a high contrast of soft tissues, which opens the micro-CT field to a large number of new applications.
19 Micro-Computed Tomography
275
Fig. 19.1 Hierarchical imaging using computed tomography (CT). The technique can be used on a large scale with different resolutions, while always using the same physical principles. In highresolution CT (HR-CT) domain, normal x-ray tubes can be used as a source, whereas for microCT special microfocus x-ray tubes are required. The lower range of micro-CT as well as the nano-CT domain is currently best assessed using synchrotron radiation (SR). The images show from left to right, human hand (courtesy of Thomas L. Mueller, ETH Zürich), trabecular bone structure, microcallus, murine cortical bone surface of a femur with internal vasculature (courtesy of Philipp Schneider, ETH Zürich), capillary in bone surrounded by osteocyte lacunae (courtesy of Philipp Schneider, ETH Zürich).
1.2
Theory
The basic physical principle of computed tomography is the interaction of ionizing radiation, such as x-ray or synchrotron radiation (see Note 1), with matter, where, in the energy range typically used for CT imaging, the so-called photo-effect builds the main interaction mechanism. The photo-effect attenuates the photons proportional to the third power of the order number of the elements and inverse proportional to the third power of the photon energy. Thus, the actual attenuation not only depends on the material but also on the energy spectrum of the x-ray source (see Notes 2 and 3). As an x-ray beam penetrates an object, it is exponentially attenuated according to the material along its path. The energy-dependent material constant appearing in
276
M. Stauber and R. Müller
Fig. 19.2 Typical applications for micro-CT. a. Trabecular bone structure in a two-plate compression testing experiment. b. Three-dimensional reconstruction of scaffold (white) and bone ingrowth (yellow); only part of the cylindrical scaffold is displayed, as to be able to look inside the scaffold (courtesy of Harry van Lenthe, ETH Zürich). c. Performance analysis of a bone replacement material in fracture repair; a gap was cut in a long-bone (gray) and filled with bone replacement material (yellow), the fracture was fixed using a conventional T-plate (courtesy Thomas L. Müller). d. Titanium screw (red) in bone (gray). e. Murine cranium. f. Murine femur; in bone research typically different compartments (gray = full bone, red = cortical ring, yellow = trabecular region) are selected to compute morphometric indices (courtesy of Thomas Kohler, ETH Zürich).
the exponent of this attenuation formula is called the linear attenuation coefficient. It expresses the amount of radiation that is attenuated on an infinitely small distance, in which the final attenuation reflects the sum of all these local linear attenuations along the x-ray beam. Therefore, an x-ray projection (or x-ray image) represents an image of the sum of all local attenuations along the x-ray beam.
19 Micro-Computed Tomography
277
Fig. 19.3 Main components and working principle of a micro-CT scanner. A micro-focus x-ray tube emits x-ray, which is collimated and filtered to narrow the energy spectrum. The x-ray passes then the object and is recorded by a two-dimensional CCD array. A full scan involves a set of projections under different rotations of the object.
To produce a three-dimensional CT image, a whole set of such two-dimensional projections need to be acquired. In micro-CT, these projections are usually taken in a setup in which the source and detector are at a fixed position and the object is rotated around its long axis (Fig. 19.3). The source is mostly either a microfocus x-ray tube or an insertion device of a synchrotron radiation facility and the detector is normally based on a CCD camera with a phospholayer to convert x-ray to visible light. Since CCD cameras have a limited number of pixels, the projections are recorded in discrete points with a so-called sampling distance (distance between neighboring pixels) and a maximal number of samples (which may correspond to the number of pixels on the CCD). It can be shown that the number of projections taken over 180 degrees should be about twice the number of samples per projection to avoid aliasing artifacts (see Note 4). The two-dimensional projections can then be used to reconstruct a three-dimensional image. In this sense, CT images can be seen as images that represent linear attenuation coefficients.
2
Materials
Micro-CT scanners are available for in vivo as well as in vitro imaging. Where the in vivo imaging systems, which are mostly designed for animal investigations, do not require much more than to fix the animal in the appropriate position, the in vitro imaging systems require more extensive sample preparation. This section thus concentrates on the sample preparation for in vitro micro-CT systems.
278
M. Stauber and R. Müller
As opposed to laboratory methods in which clear protocols can be given about how to perform a certain assay, the sample preparation for micro-CT measurements greatly depends on the sample to be scanned and also on the scanner that will be used. For this reason, this section is not a clear assay but rather highlights the points to be considered when preparing samples for micro-CT measurements. 1. Sample Size. Most micro-CT manufacturers provide sample holders in the form of cylindrical vials of different sizes, where the size of the sample holder reflects the maximal nominal resolution that can be acquired (see Note 5). Thus, in a first step the sample has to be prepared to fit into the sample holder matching the respective resolution. Some systems, such as experimental micro-CT systems or SR micro-CT systems, provide an open platform to mount the sample. In this case, the user has to make sure that the sample is not placed outside the reconstruction circle, which means that the sample has to be completely visible on all projections. If this condition is not met, the image has to be reconstructed locally (see Note 6), which may result in reduced image quality. 2. Sample Fixation. It is of great importance to properly fix the sample in the sample holder, such that no movement during the measurement, is possible. If a sample changes its position during the measurement, the final image will most likely display motion artifacts (see Note 7). Sample fixation should be done using a material that has low absorption for x-rays, such as styrofoam, which also has the advantage of providing the possibility to easily build sample specific casts. 3. Sample Hydration. Samples can be measured both, in liquid and air (see Note 2). If the sample needs to be measured in liquid (typically saline or 70% ethanol), then the liquid should be added using a syringe to avoid creating large air bubbles. It is also a good practice to tap the sample holder on a table to remove remaining air bubbles, which may be trapped under the sample or the sample fixation. To prevent the liquid from evaporating during the measurement, the sample holder should be closed using either the lid provided by the manufacturer or by simply using laboratory film. However, when using an additional material such as a laboratory film, make sure that not too much material is added on the outside of the sample holder, which could cause mechanical problems or even result in local tomography (see Note 6). 4. Sample Orientation. Small elongated samples (e.g., long bones of small animals) may fit with their long axis both parallel or perpendicular to the vial axis in the sample holder. The advantage of putting the sample perpendicular to the vial axis is a shorter measurement time. Furthermore, in systems that allow for batch scanning, several samples may be stacked on top of each other into one sample holder, which may then be scanned automatically overnight. Thus, horizontal placement and stacking should be used for high-throughput studies where hundreds or thousands samples of the same type have to be measured. However, horizontal placement is compromised with a slightly reduced image quality due to different beam-hardening along and perpendicular to the long bone (see Note 2). Thus, if best image quality is required, the long axis of the sample should be aligned with the rotation axis of the scanner.
19 Micro-Computed Tomography
279
5. Loading the Sample Revolver. Recent developments in micro-CT included a so-called sample revolver, an apparatus, which allows loading several sample holders that are all processed in batch mode. These systems were designed for high throughput analysis of hundreds or thousands of similar samples. The preparation of the individual sample holders is done as for single sample holder systems, but instead of placing the sample holder directly onto the rotation axis they are placed onto the sample revolver. The sample revolver then loads one sample holder after the other fully automatically into the scanner, which allows scanning for several days or even weeks without further operator assistance.
3
Methods
The actual micro-CT measurement is even more machine dependent than the sample preparation, and different manufacturers may provide different features and software to perform the actual measurement. Thus, it may be possible that this section describes features that are missing on certain scanners. However, the basic scanning procedure should be very similar for all scanner types and it can be assumed that if an option cannot be set, the manufacturer chose to use a standard setting to yield reasonable results. The points in this section can therefore be seen as a list of things to be considered to perform an optimal measurement, and they should increase understanding of the consequences of changing a certain parameter. Micro-CT measurement does not only involve the actual measurement of the specimen, but also computation of a number of indices as well as the two- and three-dimensional visualization. Therefore, this section is structured in three subsections; the first one describes the steps that have to be taken for the actual measurement, the second one describes the standard image processing procedures, and the third one describes standard procedures that can be applied for structure analysis.
3.1
Scan and Image Reconstruction Parameters
1. Energy. The photon source in desktop micro-CT systems is in general a microfocus x-ray tube, which allows setting the peak energy of the x-ray spectrum (Fig. 19.4). The optimal energy depends on different factors, such as material type, object thickness, and whether the sample is measured in air or liquid. Furthermore, it must be taken into account that x-ray tubes are designed to work optimally in a given range and that performance may be too low in the actual optimal range. High energies allow better penetration of high-density materials, low energies yield better contrast for different materials. Thus, the optimal energy selection is a trade-off between contrast and intensity. At SR facilities the energy can be selected quasi-monoenergetic and
280
M. Stauber and R. Müller
A
0
10
B
20
Peak energy (80 kVp)
Relative intensity
Characteristic radiation
Shift of Eeff
30 40 50 Photon Energy, keV
60
70
80
Fig. 19.4 Beam hardening (see Note 2). This sketch depicts the phenomenon of beam hardening. Line A and line B show the energy spectrum before and after the x-ray beam passed the object, respectively. The peak energy (80 kVp) and the place of the characteristic radiation remain at the same position. Lower energies are higher absorbed than high energies, which causes the effective energy to shift toward higher values.
its optimal value may be computed from the thickness and material of the sample according to Grodzins (16). 2. Filtering. The energy spectrum of the laboratory x-ray sources entails a problem, which is known as beam hardening (see Note 2). This means that low energy photons are more highly absorbed than high energy photons, causing the mean of the energy spectrum to shift to higher values (see Fig. 19.4), which in consequence makes it difficult to use the images for quantitative analysis of the attenuation coefficients. To minimize this effect, some manufacturers provide a set of different filters that can be used to absorb the very low-energy photons directly in the filter. Using an x-ray filter narrows the energy spectrum and makes the images more suitable for quantitative analysis of the gray values. However, filtering also reduces the overall intensity, therefore a higher integration time has to be chosen to get enough signal. Thus, by narrowing the energy spectrum the measurement will also take longer. 3. Integration Time. The longer the integration time, the better the signal-to-noise ratio and hence, the better the image quality. On the other hand, an increase in integration time is directly related to measurement time and radiation dose, where the latter is of minor importance in in vitro imaging (see Note 8). Furthermore, if the integration time is chosen to be too high, the CCD detector could be fully saturated, yielding in image artifacts that are especially pronounced for the high absorbing materials. Thus, the integration time is a tradeoff between image quality, measurement time and radiation dose. 4. Frame Averaging. Another technique to increase the signal-to-noise ratio is to image each projection several times and to use the average for image reconstruction; a technique referred to as frame averaging. Increasing number of
19 Micro-Computed Tomography
5.
6.
7.
8.
9.
10.
281
frames does not only increase image quality, but also time and radiation dose. Thus, the trade-off is the same as increasing the integration time. However, the advantage of frame averaging over increasing the integration time is that the CCD device can be used in an optimal range and that saturation can be prevented. Resolution. The optimal resolution depends of the features that should be detected in the images. However, the resolution can in general not be chosen arbitrarily because it is linearly coupled with the object dimensions (see Note 5). Nevertheless, a method called local tomography (see Note 6), which is occasionally used at SR facilities, allows imaging and reconstructing a local part of the object only. With this method, the object size is no longer limited by the nominal resolution but rather by mechanical limitations as well as by a maximal diameter that could cause total absorption. It also has to be noted that image quality generally decreases in local tomography, which may involve intensive post-processing of the scanned data. Number of Projections. From theory, the number of projections chosen should be about twice the number of sample points on a projection (given by detector array). However, in practice it was shown that the same number or even half the number of samples still yields reasonable image quality. The number of projections is a trade-off between image quality and measurement time. For fast overview scans, a relatively low number of projections can be chosen. However, if the number is chosen too low, so-called aliasing artifacts may distort the image (see Note 4). Number of Stacks. As opposed to most clinical CT systems which operate in a helical or spiral mode, micro-CT systems acquire only a certain number of slices per stack. If the object is longer than one stack, then several consecutive stacks have to be imaged. This is usually done automatically by the software from the micro-CT manufacturer. Cone Beam vs. Fan Beam. Most modern micro-CT systems operate in cone beam mode. However, because cone beam reconstruction may yield geometrical incorrectness at the stack boundary, some manufacturers also provide an option to select the more accurate fan beam mode. This usually results in images with better geometrical representation. The drawback of fan beam mode imaging is the shorter stack size and hence the longer scan time. Stack Overlapping. To make micro-CT scanners faster, most manufacturers use a cone-beam scanner. A problem that may arise with such systems is that the image quality at the stack boundary is reduced as compared with the center of the stack. To correct for this problem there are now scanners that provide stack overlapping, which increases overall image quality. Stack overlapping is similar to frame averaging in that projections are taken several times; however, the projections are shifted along the rotation axis by the given offset. The higher stack overlapping chosen, the better the image quality, but also the longer a scan will take and the higher the radiation becomes. Phantom for Calibration. If the density values in the object are to be used for a quantitative analysis, a phantom with well-known density has to be measured
282
M. Stauber and R. Müller
as well. The phantom can either be measured simultaneously in the same scan or previously if it can be assumed that the phantom and object densities are about of the same order (see Note 2). Some manufacturers provide phantoms that should be measured on a regular basis (daily or weekly) for quality control. If the scanner is then stable over a long time, the calibration can be made previously for all energies and imaging modes and then be used in the measurement for the actual correction and calibration. 11. Image Reconstruction. Most manufacturers use a standard convolution back projection image reconstruction method and do not allow the user to change the algorithm. However, in principle there exist different image reconstruction methods, which may enhance image quality for special purposes. Nevertheless, the field of image reconstruction should be left to experts because it requires a deep mathematical understanding of CT. 12. Binning. To reduce the noise in the images it is possible to average several pixels on the CCD chip, a method referred to as binning. Mostly, a two- or threefold binning is used, which means that 2 × 2 or 3 × 3 pixels are averaged, respectively. This method not only increases the signal-to-noise ratio due to the better photon statistics, but it also effectively reduces the actual data size (see Note 9). However, a drawback of this method is the reduced nominal resolution.
3.2
Standard Image Processing
Digital image processing is a large field and to cover all aspects would be beyond the scope of this chapter. Thus, there are many different ways that data can be maximally processed. The procedures that are applied may also depend on the object scanned. This section describes the most commonly applied procedures to extract the structural features as it is used in the bone field. 1. Filtering. Photon statistics, fluctuations of the CCD, or thermal fluctuations are all factors that may increase noise in the final images. To reduce this noise and to best extract the desired features, a large set of different filtering methods have been introduced. Examples of such filters are Gauss-, Hamming-, Hann-, Blackman-, Bartlett-, Welch-, or Wiener-filter (17). Most filters are applied in the Fourier space, where the filters are classified as either low- or high-pass filters, according to their function to either remove the high or the low frequencies from the images. High-pass filters are commonly used for feature extraction and edge enhancement, whereas low-pass filters are commonly used for noise reduction. Where high-pass filters may increase some noise, low-pass filters blur the images. This trade-off can be solved optimally using Gaussian filtration. Furthermore, the Fourier Transform of the Gauss function yields a Gauss function again, wherefore this filter can be applied in direct space, which makes this filter
19 Micro-Computed Tomography
283
easy implementable and fast in computation. For these reasons, Gaussian filtration is one of the most commonly used filters for noise reduction (see Note 10). However, because the Gaussian filter also blurs the edges, more advanced smoothing techniques have been developed to smooth the image while preserving the edges. An example of such an edge-preserving smoothing filter is the anisotropic diffusion filter. Other filters, which are frequently used, are the Median-, Minimum-, and Maximum-filters, which can also be applied in direct space. 2. Segmentation. Many applications require segmenting the images into a few discrete phases for further processing. An example of a typical three-phase object application is a bone with an implant, where the third phase is the background (see Note 11). Segmenting images is a difficult topic and has led to many publications trying to establish new methods for special purposes. The main problem is that because of beam-hardening (see Note 2), aliasing artifact (see Note 4), partial volume effect (see Note 12), or ring artifact (see Note 13), the values of different materials in a single image may not be clearly distinguished. Therefore, it becomes relatively difficult to select all voxels of a certain value-range and to assign it to an object. Nevertheless, the most common segmentation technique is still the so-called thresholding, where all voxels above a certain value are set to white (1) and all remaining voxels are set to black (0). This procedure can also be applied to two three or more levels, if image quality is good enough and absorption contrast is sufficient for the different materials to be resolved (see Note 14). 3. Component Labeling. For structural analyses, it might in certain cases be reasonable to remove small unconnected objects from the main object, which may arise from image noise or from improper sample preparation. This is done by an algorithm, which labels all components in the structure. Depending on the software it can then be decided, whether the largest, or smallest, or the components with a certain size should be kept for further evaluations.
3.3
Standard Structure Analysis
The beauty of CT is that it allows non-destructive access to the internal structure of a body. Thus, along with the development of micro-CT systems, new morphometric indices have also evolved for the characterization of the structural features. Because CT yields a good contrast of bone versus non bone, many of these indices were developed in the bone field (18). However, these measures may be used in any field in which three-dimensional structures have to be analyzed. This subsection gives an overview of the most common morphometric indices as derived from the threedimensional images. The indices listed here relate to the bone field and may be renamed if used in a different environment. It has to be noted that different microCT manufacturers provide different software packages for the computation of these
284
M. Stauber and R. Müller
indices; however, the algorithms should all be implemented along the references indicated and the name of the indices should always be the same (19). 1. Bone Volume Density (BV/TV). Among the basic indices is the measurement of the bone volume (BV) and the total volume of interest (TV). These indices can either be derived from a simple voxel counting method or by a more advanced volume rendering method, also referred to as volumetric marching cubes (VOMAC) (20). The latter method may be more accurate for small or very complex structures. The measurement of BV/TV is of major importance in bone research because it is the principal determinant of bone strength. 2. Specific Bone Surface (BS/BV). Another basic measure is the bone surface (BS), which is conventionally computed by triangularization of the object surface using a method called marching cubes (21). The specific bone surface can then easily be derived by dividing the measured bone surface by the bone volume. 3. Mean Trabecular Thickness (Tb.Th), Mean Trabecular Separation (Tb.Sp), and Mean Trabecular Number (Tb.N). These measures are all based on a sphere fitting method, where for thickness measurement the spheres are fitted to the object and for separation the spheres are fitted to the background (22). The principal idea is to label each voxel with the radius of the largest possible sphere that can be fitted through this voxel, which is completely contained within the object (or background), and then to average all these radii. This yields a reasonable average thickness of the whole structure or background, where the latter can be interpreted as the mean trabecular separation. The mean trabecular number is computed as the inverse of the mean distance between the mid-axes of the structure, which may be derived from the distance transformation method (23). 4. Connectivity density (Conn.D). To characterize the number of possible paths that connect one side of the object to the opposite side, an index called connectivity has been introduced. Conn.D is derived from the Euler number (24), which is a fundamental topological measure counting the number of objects, the number of marrow cavities fully surrounded by bone, and the number of connections that must be broken to split the structure in two parts. Because the connectivity depends a lot on the structure size, it is more convenient to present this index as a density by dividing it by the total volume (24). 5. Structure model index (SMI). To estimate the plate-rod characteristic of a trabecular bone structure an index called structure model index was invented (25). This index was designed to be 0 for perfect plates, 3 for perfect rods, and 4 for perfect spheres. Thus, any value in this range indicates how plate- or rod-like a structure is. However, although this index was designed to yield values in the range of 0 to 4, it may happen that values outside this range occur. The interpretation of these values is difficult, and the structure type may be expressed more precisely by measuring the actual amount of rods and plates within a bone structure (26).
19 Micro-Computed Tomography
285
6. Mean Curvature áH ñ, Trabecular Bone Pattern Factor (TBPf). The mean curvature of the object surface can be derived from the so-called parallel surface method (27), where the triangulated surface of an object is slightly dilated. The increase in the surface is then related to the dilation distance via the mean curvature and the Gaussian curvature. It can be shown that the frequently used trabecular bone pattern factor, which was originally devised for two-dimensional image analysis (28), equals 2áH ñ if expanded to three-dimensional space. Furthermore, the mean curvature is closely related to the structure model index (29) via the equation SMI = 12 · áH ñ · BV / BS · 7. Degree of Anisotropy (DA). It is thought that the degree of anisotropy in combination with bone volume density may explain a significant part of the mechanical properties of a three-dimensional structure. Therefore, a large number of different methods, such as mean intercept length (MIL) (30), volume orientation (VO) (31), star volume distribution (SVD) (32), or star length distribution (SLD), have been proposed to estimate the anisotropy of trabecular bone. Those and other measure of architectural anisotropy are reviewed in detail elsewhere (18). 8. Line Skeleton Graph Analysis (LSGA). To compute topological indices of trabecular bone structures, a method has been introduced, which computes from the original three-dimensional structure a line skeleton graph (33). This line skeleton graph is an abstraction of the structure, in which the structure is reduced to a one voxel thick line with the same topology. This abstraction allows for a simple computation of topological indices as well as the length and volume of single trabecular elements. 9. Topological Classification. Another approach for the computation of topological indices uses surface skeletons combined with a point-classification algorithm (34). With this approach, the anisotropy of trabecular bone could be determined based on single trabecular elements and it could be shown that these measures perform better than MIL (35). 10. Local Morphometry. Local morphometry denotes a method that measures morphometric indices of single trabecular elements. For this, the elements have to be extracted from the structure, which can be done using a method called volumetric spatial decomposition (26). With this method it is hence possible to perform more detailed analyses of three-dimensional structures and it becomes possible to analyze the individual contribution of rods and plates to the competence of trabecular bone (36). The decision on which morphometric indices should be computed depends a lot on the object under investigation. A typical example from bone research that demonstrates this is the mouse femur. In these femora, typically three different compartments are analyzed (37), the full bone, a cortical ring, as well as the trabecular region above the growth plate (see Fig. 19.2F). For the analysis of these compartments masks are generated either by hand, by semi-automatic or fully automatic algorithms, to define the region of interest.
286
4
M. Stauber and R. Müller
Notes
1. X-Ray, γ-Ray, and Synchrotron Radiation. Physically, there is no difference among x-ray, g -ray, and synchrotron radiation. They all are photons at the high end of the electromagnetic spectrum. The different names arise from the different origins of the radiation, where x-ray originates from the atom shell, g-ray originates from the atom nucleus, and synchrotron radiation originates as a relativistic effect from acceleration of fast flying charged particles (e.g., electrons). Thus, the interaction mechanism with matter for these three radiation types is exactly the same. The advantage of using synchrotron radiation is the extremely high flux (photons per seconds), which also allows one to narrow the energy band resulting in a quasi-mono-energetic beam. The benefit of using a monoenergetic beam is that beam-hardening artifacts (see Note 2) do not occur, which in consequence allows for an accurate quantitative analysis of the reconstructed images. 2. Beam-Hardening. An x-ray tube emits radiation with an energy spectrum peaking at the maximum voltage used to accelerate the electrons, the so-called peak energy (kVp). The exact shape of the spectrum depends on the anode material, the filtering (see Note 3), and on the energy applied between cathode and anode. Since the absorption in the photo-effect is proportional to the inverse of the third power of the energy, the low energy beams are more highly absorbed, which results in a shift of the mean energy to higher values as the x-rays pass through the object; an effect referred to as beam-hardening (see Fig. 19.4). It has to be noted that the peak energy remains at all times constant and that only the mean and the shape of the spectrum are shifted toward higher values. The fact that the energy spectrum is altered as the beam transverses the object further leads to different absorptions along the beam, even if the object material is exactly the same at all points. The beam-hardening effect depends on the initial energy spectrum, the object material and the object shape and dimensions. Thus, if the object is composed of different materials, an accurate quantitative measurement of the local densities is basically impossible. However, if the object is composed of a known material and if the thickness and shape of the object can be estimated, it is possible to correct for a certain amount of beam hardening by measuring a well-known wedge phantom. If no beam-hardening correction is done, the images are brighter at the boundary as compared to the center. To correct for beam-hardening, the manufacturers use different methods. Whereas some scanners automatically correct for an “average measurement,” other scanners provide the possibility to select an appropriate correction method, which is based on a wedge phantom measurement of well known density. The phantom is then either measured in air or water, which requires the sample of the actual measurement also to be measured in air or water, respectively, for accurate beam-hardening correction. The manufacturers provide this information, which is important to follow for best image quality.
19 Micro-Computed Tomography
287
3. X-Ray Filtering. By adding x-ray filters, the very low energies in the energy spectrum can be removed and as a consequence, the spectrum is narrowed, which reduces beam-hardening artifacts (see Note 2). X-ray filtering may improve contrast in cases, where different materials of similar linear attenuation coefficient should be distinguished. Some micro-CT manufacturers provide several filters to select the desired energy spectrum. X-ray filters are typically made from aluminum, copper, or titanium. 4. Aliasing Artifact. To acquire an image with reasonable quality, the number of projections should equal at least half the number of samples per projection. If this rule is violated, the image gets distorted by aliasing artifacts (Fig. 19.5).
Fig. 19.5 Aliasing artifact (see Note 4). This figure shows a computer simulation of a micro-CT measurement and reconstruction in a 512 × 512 pixel matrix with different settings for the number of projections and number of samples. If the number of projections chosen is too low relative to the number of samples, the images get distorted by aliasing artifacts. The figure also demonstrates that a higher number in both parameters yields a better resolution. Furthermore, it has to be noted that the eyes do not differ a lot in their gray values from the mouth or the nose, and that they seem to be brighter on the outside, an effect caused by beam-hardening.
288
M. Stauber and R. Müller
5. Sample Size and Resolution. There is a simple rule that connects the sample size to the maximal nominal resolution: The smaller the sample size the larger the resolution. To estimate the maximal nominal resolution that can be acquired the largest sample diameter can be divided by the number of samples taken per projection. Current micro-CT systems take about 1,000–2,000 samples per projection. Thus, an object of a maximal diameter of 10 mm could be measured with a maximal nominal resolution of 10 µm and 5 µm, respectively. However, this theoretical maximal nominal resolution may be lower due to mechanical limitations of the micro-CT. 6. Local Tomography. An object is in the strict mathematical sense not reconstructable if certain parts of the object are not included on one or several projections. However, if this is done, an image can be reconstructed nonetheless, and it is assumed that the missing information is smeared over the image. Because this of course is not quite true, image quality in local micro-CT is always lower than in normal micro-CT. Especially toward the boundary where the values are much higher because the missing values from the outside accumulate there. 7. Motion Artifact. If the object moves or rotates during the measurement, the projections do not fit together at the reconstruction, resulting in distortion of the object and tails, enclosing it (see Fig. 19.7C). For this reason, it is of major importance to properly fix the sample prior to measurement. 8. Radiation Dose. As x-ray transfers a body, a certain amount of the radiation is absorbed, which in biological samples may cause some damage. This is generally no problem for in vitro measurements of samples, but must be taken into account, when imaging live biological system, such as tissues in a bioreactor or animals. The radiation dose increases roughly with the fourth power of the inverse of the nominal resolution. Thus, in an image taken at a nominal resolution of 10 µm the deposed radiation dose is about 16 times larger than in an image taken at a nominal resolution of 20 µm. The exact dose deposition depends on many things, such as energy, total integration time, scatter, and leakage, and must be measured for each system and each scanning mode. Micro-CT scanners designed for in vivo measurements all provide these data, whereas these data may not be available for in vitro systems. 9. 32-/64-Bit Operating Systems. The data size of micro-CT images increased with the improvement of computers and detector systems and may nowadays easily reach several gigabytes per image. For example, an image of 10243 voxels acquired at a gray level depth of 16 bit reaches a size of 2 GB. In image processing, it is often required to keep two or three copies of an image at the same time in the memory. Thus, it is required that the operating system is able to address several gigabytes of memory. Unfortunately, common 32-bit systems are limited (the actual limit is on most operating systems even lower) by being able to address 4 GB of memory only and may thus not be suitable for measuring or processing micro-CT data. Fortunately, all modern operating systems will use 64-bit technology in the near future which sets the theoretical limit of addressable memory to 16 EB (16 EB = 1.718 * 1010 GB).
19 Micro-Computed Tomography
289
10. Image Processing, Gaussian Filtration. Due to the optimal noise-reduction, Gaussian filtration is the most often used filter to enhance the images. This filter uses digital image processing applied in a window and hence has two parameters, which are the filter width (σ) and the filter support (s), describing the number of neighboring voxels that are used for the computation (window). The optimal values depend on the initial image quality and the desired features to be extracted. Fig. 19.6 demonstrates the effect of different filter width on noise reduction. 11. Metal Artifact. Materials with a high linear attenuation coefficient may cause total absorption of the x-ray beam yielding in star-formed artifacts in the reconstructed images (Fig. 19.7). Because this effect typically occurs when imaging metals (e.g., steel), it is frequently called metal artifact, resulting in a bright shadow (halo) over the metal object and its immediate region. This effect can only be prevented by replacing the high absorbing materials. Metals such as
Fig. 19.6 Gaussian-filtration (see Note 10). As the filter width is increased, noise is clearly reduced and also the ring artifacts (see Note 13) can be removed. However, the images are also blurred and contrast is reduced.
Fig. 19.7 a. Ring artifact (courtesy of Philipp Schneider, ETH Zürich, see Note 13). b. Metal artifact (see Note 11). c. Motion artifact (see Note 7). The ring artifact in this image was caused by a defect of the scintillator that converts x-ray to visible light. The screw in (B) absorbs quasi all radiation, which results in star-shaped artifacts as well as to distortions of the image in its neighborhood. Image (C) shows the ulna of a patient measurement, where the patient moved its arm during the measurement. For this reason, the cortex is not a closed ring but rather composed of two half rings that go along with two tails.
290
M. Stauber and R. Müller
steel could be replaced by aluminum or titanium, which have a low enough linear attenuation coefficient that can be measured. 12. Partial Volume Effect. The voxel on the surface of an object always lies partially inside and partially outside the object. Therefore, the image value of such voxels is an average of the object and non-object contribution, an effect called partial volume effect. For this reason, the surface cannot be sharply defined and it has to be avoided to use these voxels for numerical computation. If for instance the bone mineral density should be computed, these outer voxels must be removed prior to computation to get a reasonable result. 13. Ring Artifact. A common artifact in CT imaging is rings or half-rings around the rotation center of the reconstructed images. The source of these rings is a defective pixel on the CCD, a defect in the scintillator that converts x-ray to visible light, or simply a piece of dust on the detector system. In latter case, the problem can easily be solved by cleaning the system. For defective pixels, some manufacturers of micro-CT systems provide the possibility to electronically replace it by an average of some given neighboring pixels. Although the data at this very spot are lost, averaging yields generally good results and removes the ring artifacts. 14. Image Processing and Segmentation. In many applications it is useful to separate the object from its background (e.g., bone from marrow) for further processing. The simplest method is to select a threshold in which all voxels above are set as object voxels (1) and all voxels below are set as background voxels (0). As simple as this may sound, it is difficult for several reasons. It has been shown that changing the threshold also changes the outcome of the morphometric indices. However, it has also been shown that if the threshold is chosen in a “reasonable range” the indices are scaled “reasonably” as well. Thus, using a global thresholding procedure can be justified in many applications. The best approach to find an appropriate threshold is by visually comparing the gray-level images to the binarized image both, in 2D and in 3D.
References 1. Radon, J. (1917) Über die Bestimmung von Funktionen durch ihre Integralwerte längs bestimmter Mannigfaltigkeiten. Ber Verb Sächs Akad Wiss Leipzig Math-Nat Kl 69, 262–277. 2. Cormack, A. M. (1963) Representation of a function by its line integrals with some radiological applications. J Appl Phys 34, 2722–2727. 3. Cormack, A. M. (1964) Representation of a function by its line integrals with some radiological applications. II. J Appl Phys 35, 2908–2913. 4. Hounsfield, G. N. (1972) A method of and apparatus for examination of a body by radiation such as x-ray or gamma radiation. Patent Specification GB. 5. Hounsfield, G. N. (1973) Computerized transverse axial scanning (tomography). 1. Description of system. Br J Radiol 46, 1016–1022. 6. Hounsfield, G. N. (1977) EMI scanner. Proc the Roy Soc London Series B Biol Sci 195, 281–289. 7. Feldkamp, L. A., Goldstein, S. A., Parfitt, A. M., et al. (1989) The direct examination of threedimensional bone architecture in vitro by computed tomography. J Bone Miner Res 4, 3–11.
19 Micro-Computed Tomography
291
8. Müller, R., Rüegsegger, P. (1994) Morphological validation of the 3D structure of non-invasive bone biopsies. Abstracts 10th Int. Workshop on Bone Densitometry. Bone Miner 25, 8. 9. Rüegsegger, P., Koller, B., Müller, R. (1996) A microtomographic system for the nondestructive evaluation of bone architecture. Calcif Tissue Int 58, 24–29. 10. Bonse, U., Busch, F. (1996) x-ray computed microtomography (micro-CT) using synchrotron radiation (SR). Prog Biophys Mol Biol 65, 133–169. 11. Stampanoni, M., Borchert, G., Abela, R., et al (2002) Bragg magnifier: A detector for submicrometer x-ray computer tomography. J Appl Phys 92, 7630–7635. 12. Nuzzo, S., Lafage-Proust, M. H., Martin-Badosa, E., et al. (2002) Synchrotron radiation microtomography allows the analysis of three-dimensional microarchitecture and degree of mineralization of human iliac crest biopsy specimens: effects of etidronate treatment. J Bone Miner Res 17, 1372–1382. 13. Nuzzo, S., Peyrin, F., Cloetens, P., et al. (2002) Quantification of the degree of mineralization of bone in three dimensions using synchrotron radiation microtomography. Med Phys 29, 2672–2681. 14. Müller, R., Gerber, S. C., Hayes, W. C. (1998) Micro-compression: a novel technique for the non-destructive assessment of local bone failure. Technol Health Care 6, 433–444. 15. Nazarian, A., Stauber, M., Muller, R. (2005) Design and implementation of a novel mechanical testing system for cellular solids. J Biomed Mater Res B Appl Biomater 73B, 400–411. 16. Grodzins, L. (1983) Optimum energies for x-ray transmission tomography of small samples. Nuclear Instr Methods 206, 541–545. 17. Gonzalez, R., Woods, R. (2002) Digital Image Processing. Prentice-Hall, Englewood Cliffs, NJ. 18. Odgaard, A. (1997) Three-dimensional methods for quantification of cancellous bone architecture. Bone 20, 315–328. 19. Parfitt, A. M., Drezner, M. K., Glorieux, F. H., et al. (1987) Bone histomorphometry: standardization of nomenclature, symbols, and units. Report of the ASBMR Histomorphometry Nomenclature Committee. J Bone Miner Res 2, 595–610. 20. Müller, R., Rüegsegger, P. (1995) Three-dimensional finite element modelling of non-invasively assessed trabecular bone structures. Med Eng Phys 17, 126–133. 21. Lorensen, W. E., Cline, H. E. (1987) Marching cubes: a high resolution 3D surface construction algorithm. Comput Graphics 21, 163–169. 22. Hildebrand, T., Rüegsegger, P. (1997) A new method for the model-independent assessment of thickness in three-dimensional images. J Microsc 185, 67–75. 23. Danielson, P-E. (1980) Euclidean distance mapping. Comp Vision Graph Image Processing 14, 227 –248. 24. Odgaard, A., Gundersen, H. J. (1993) Quantification of connectivity in cancellous bone, with special emphasis on 3-D reconstructions. Bone 14, 173–182. 25. Hildebrand, T., Ruegsegger, P. (1997) Quantification of bone microarchitecture with the structure model index. Comput Methods Biomech Biomed Engin 1, 15–23. 26. Stauber, M., Müller, R. (2006) Volumetric spatial decomposition of trabecular bone into rods and plates: a new method for local bone morphometry. Bone 38, 475–484. 27. Nishikawa, Y., Jinnai, H., Koga, T., et al. (1998) Measurements of interfacial curvatures of bicontinuous structure from three-dimensional digital images. 1. A parallel surface method. Langmuir 14, 1242–1249. 28. Hahn, M., Vogel, M., Pompesius-Kempa, M., et al (1992) Trabecular bone pattern factor-a new parameter for simple quantification of bone microarchitecture. Bone 13, 327–330. 29. Jinnai, H., Watashiba, H., Kajihara, T., et al. (2002) Surface curvatures of trabecular bone microarchitecture. Bone 30, 191–194. 30. Whitehouse, W. J. (1974) The quantitative morphology of anisotropic trabecular bone. J Microsc 101, 153–168. 31. Odgaard, A., Jensen, E. B., Gundersen, H. J. (1990) Estimation of structural anisotropy based on volume orientation. A new concept. J Microsc 157, 149–162. 32. Cruz-Orive, L. M., Karlsson, L. M., Larsen, S. E., et al. (1992) Characterizing anisotropy: a new concept. Micron Microscopica Acta 23, 75–76.
292
M. Stauber and R. Müller
33. Pothuaud, L., Porion, P., Lespessailles, E., et al. (2000) A new method for three-dimensional skeleton graph analysis of porous media: application to trabecular bone microarchitecture. J Microsc 199, 149–161. 34. Saha, P. K., Chaudhuri, B. B. (1996) 3D digital topology under binary transformation with applications. Comput Vis Image Understanding 63, 418–429. 35. Gomberg, B. R., Saha, P. K., Wehrli, F. W. (2003) Topology-based orientation analysis of trabecular bone networks. Med Phys 30, 158–168. 36. Stauber, M., Rapillard, L., van Lenthe, G. H., et al. (2006) Importance of individual rods and plates in the assessment of bone quality and their contribution to bone stiffness. J Bone Miner Res 21, 586–595. 37. Kohler, T., Beyeler, M., Webster, D., et al. (2005) Compartmental bone morphometry in the mouse femur: reproducibility and resolution dependence of microtomographic measurements. Calcif Tissue Int 77, 281–290.
Chapter 20
Fourier Transform-Infrared Microspectroscopy and Microscopic Imaging Samuel Gourion-Arsiquaud, Paul A. West, and Adele L. Boskey
Abstract For age- and sex-matched subjects, osteoporotic bone is more fragile than healthy bone. Vibrational infrared spectroscopy and in particular infrared microspectroscopic imaging is a useful tool for investigating and characterizing changes associated with metabolic bone diseases including osteoporosis in biopsied tissues. Strength-related measures such as bone mineral content/composition as well as spectroscopically determined bone quality–related measures such as mineral crystallinity, carbonate substitution, and collagen cross-linking consequently differ between osteoporotic patients and normal subjects. Validated IR parameters specific to the mineral and matrix components of bone have been defined and can now be used to quantify anatomical/spatial variations and the effect of new therapies on osteoporotic bone. Keywords FT-IR microspectroscopy, bone quality, mineralization, collagen crosslink, mineral crystallinity.
1
Introduction
Fourier transform infrared microscopy (FTIRM) and FT-IR Imaging Spectroscopy (FTIRI) are well-established vibrational spectroscopic techniques that can be successfully applied to investigate molecular changes in osteoporotic bone (1–3). The combination of an infrared focal-plane array (FPA) detector and a FT-IR microscope provides a powerful medium for the study of non-decalcified histological tissue sections at a spatial resolution of ∼6 µm. With the use of spectral parameters, images correlated to specific molecular vibrations can be generated (Fig. 20.1) to portray the spatial distribution of tissue components and identify changes both in the organic and inorganic phases of bone. In mineralized tissues the organic (~90% type I collagen) and inorganic (mineral) constituents produce intense, structure sensitive IR absorption bands (Fig. 20.2) that are characteristic of specific types of chemical bonds within the tissue. The primary absorption bands used in the IR analysis of bone are amide I (collagen matrix), From: Methods in Molecular Biology, Vol. 455: Osteoporosis: Methods and Protocols Edited by Jennifer J. Westendorf © Humana Press, Totowa, NJ
293
294
S. Gourion-Arsiquaud et al.
5
5
10
10 80 15
40
25
20
30
15
20
10
25 5 30 0
35
35
40
40 10
Single Channel Intensity
15
20 Pixels
30
80 60 40 20 0 10 20 30 Pixels
10
40
20
30
10
A 90 80 70 60 50 40 30 20 10 0
Single Channel Intensity
Pixels
20
Pixels
60
20 Pixels
30
15 10 5 0 −5 10 20
20 10 Pixels
30 Pixels
Pixels
40
30
B 18 16 14 12 10 8 6 4 2
Fig. 20.1 Hyperspectral images of the same section of trabecular bone obtained using ISYS imaging software. a. 2D (top) or 3D (bottom) images correspond to the peak of the phosphate (inorganic) absorbance band (1030 cm−1) showing the highest density of mineral is in the center of the trabeculae. b. 2D (top) or 3D (bottom) images correspond to the peak of the Amide I (organic) band (1660 cm−1).
Phosphate PO43-
1
(1180-916) v1, v3
Absorbance
0.8
Amide I 0.6
(1712-1592) Amide II
0.4
Carbonate CO32(892-840) v2
0.2
0 1800
1700
1600
1500
1400
1300
1200
1100
1000
900
800
Wavenumber(cm-1) Fig. 20.2 A typical FTIR spectrum from a single image pixel of healthy trabecular bone (baboon) showing frequencies of interest: Amide I and II, phosphate (ν1, ν3) and carbonate ν2 bands, Note: this spectrum was baselined and PMMA (embedding media) corrected.
20 FT-IR Microspectroscopy
295
1.4 1.2
Single Beam Intensity
6
1
5
5
10
4
0.8 Pixels
(a)
0.6 0.4
15
0.2
20
3 2
0
1 5
−0.2 2000
1800
1400
1600
1200
1000
800
Wavenumber cm
20 25 Pixels
30
35
1000
5
800
10
10
600
15
400
20
1700
1600
1500
1400
1300
10
15
(b)
20 25 Pixels
30
35
12 10 8
15
6 4
20
200
5
1200
Pixels
5
2 5
10
15
0
20 25 Pixels
30
35
(f) 1.35
1.2 6000
1
5
0.8 0.6 0.4
10
4000
Pixels
5 Pixels
Absorbance
0
14
0.55 0.5 0.45 0.4 0.35 0.3 0.25 0.2 0.15 0.1 0.05
Pixels
Absorbance
15
(e)
-1
1800
10
15 2000
0.2 950
900
850
5
800
10
15
(c)
20 25 Pixels
30
35
1.25 15 1.2
20
20
0 1300 1250 1200 1150 1100 1050 1000
1.3
10
5
0
10
15
(g)
20 25 Pixels
30
35
1.2
0.4 0.2 0 1200
40
15 20
20 1150
1100
(d)
1050
1000
950
900
850
800
5
10
15
20 25 Pixels
30
35
0
Pixels
0.6
10
4.2
5
60
5 Pixels
Absorbance
1 0.8
10
4
15 3.8
20 5
10
15
20 25 Pixels
30
35
3.6
(h)
Fig. 20.3 Hyperspectral images obtained from normal baboon cortical bone tissue. a. A typical FTIR spectrum from a single image pixel showing frequencies of interest. b. Integrated matrix areas (1592–1712 cm−1) are calculated and displayed as 2D color-coded images to give an IR image of the matrix distribution. c. Integrated mineral areas (916–1180 cm−1) and the tissue mineral distribution and (d) the carbonate (840–892 cm−1) distribution. Images of the mineral:matrix ratio (e), the carbonate:mineral ratio (f), the crystallinity ratio (1030/1020 cm−1) (g) and the crosslink ratio (1660/1690 cm−1) (h) are shown. Note: all spectra are baselined and PMMA (embedding media) corrected using ISYS software (See Color Plates).
phosphate (mineral), and carbonate. Quantitative data are obtained by calculating ratios, thereby eliminating errors associated with differences in specimen thickness. The mineral (phosphate)/matrix (collagen) area ratio calculated from the integrated areas of the ν1,ν3 phosphate absorbance band at 916–1180 cm−1 and the amide I (peptide bond C = O stretch) matrix band at ≈1592–1712 cm−1 is correlated to ash weight (4) and is a validated indicator of mineral content (Fig. 20.3). Similarly carbonate/ phosphate area ratios calculated from the integrated areas of ν2 carbonate peak at 840–892 cm−1 and the phosphate band indicate the extent of carbonate substitution (5) in the hydroxyapatite (HA) lattice. Analysis of the ν2 carbonate absorbance band by curve-fitting reveals whether the carbonate has replaced hydroxide (type A) or phosphate (type B) in the apatite lattice (6, 7). The deconvolution of the ν1, ν3 phosphate (mineral) absorbance band yields are 1030/1020 cm−1 intensity ratio (8) from two underlying bands that are sensitive to variations in crystal perfection. This ratio is linearly correlated with the hydroxyapatite crystal size and maturation as determined by x-ray diffraction (9). Mechanical properties of bone, such as tensile strength or
296
S. Gourion-Arsiquaud et al.
viscoelasticity, are dependent on the intermolecular cross-linking in collagen, the most abundant protein of the organic matrix (10). By applying difference spectroscopy it was shown that the ratio of two components of the organic (Amide I) band vary when osteoporotic and non-osteoporotic tissues were compared. This 1660/1690 cm−1 ratio has been used to estimate the amount of non-reducible (mature) to reducible (immature) cross-links (11). Changes in these IR absorption band ratios may be referenced to changes in tissue age, disease state (3, 12), and anatomical location. As such these specific IR parameters are being used to evaluate the effects of anti-resorptive and anabolic therapies in osteoporotic patients.
2 2.1
Materials Tissue Embedding
Polymethyl methacrylate (PMMA) embedding of mineralized tissues (specific to bone specimens for FT-IR analyses). 1. 2. 3. 4. 5. 6. 7. 8. 9. 10.
11. 12. 13. 14.
2.2
Ethanol (EtOH): 70%, 95%. 100% 2-propanol. 100% xylene. Methacrylate (MMA, Polysciences, Inc. Warrington, PA). 99% Butyl methacrylate. 99% Methyl benzoate. Benzoyl peroxide (stored at 4 °C). Polyethylene glycol 200 (PEG 200). 99% N,N-dimethyl-p-toluidine. Low Viscosity Embedding Media Spurr’s Kit (Electron Microscopy Sciences, Hatfield, PA) consisting of: vinyl cyclohexane dioxide (ERL 221), diglycidyl ether polypropylene glycol (DER 736), non-enyl succinic anhydride (NSA), and dimethylamino ethanol (DMAE). 15-mL plastic embedding molds (Nalge Nunc International Rochester, NY). Size may vary with sample. 50 mL polypropylene conical tubes (screw cap). Rocking platform. Securline marker (Fisher Chemicals, Springfield, NJ) or pencil.
Tissue Sectioning
1 Microtome (Leica SM 2500, Leica, Germany) or Microtome (Microm HM 360) depending of the bone sample size. 2. Knives with a tungsten carbide cutting edge (Delaware Diamond Knives, Wilmington, DE).
20 FT-IR Microspectroscopy
3. 4. 5. 6.
297
Variable speed Grinder-polisher (Buehler, Ecomet 3). Band saw. Ethanol 30% and 90%. 25 mm × 2 mm barium fluoride (BaF2) Infrared (IR) optical disks (SpectraTek, Hopewell Junction, NY).
2.3
FTIR Data Acquisition and Analysis
1. Imaging Spectrometers: The spectrometer used for the studies described is a Perkin-Elmer Spectrum Spotlight 300 Imaging System (Perkin Elmer Instruments, Shelton, CT), consisting of a step-scanning FT-IR spectrometer with an MCT (mercury-cadmium-telluride) focal plane array (FPA) detector placed at an image focal plane of an IR microscope. With this configuration we can analyze the frequency region from 7800–720 cm−1. Other related imaging systems may be used, e.g., Bruker, Nicolet, Bio-Rad. 2. Imaging Software: ISYS chemical imaging software (Spectral Dimensions Inc., Olney, MD) or other related imaging software, e.g., Matlab, CytoSpec, Biorad.
3
Methods
3.1
Preparation of Bone Samples
3.1.1
Tissue Fixation
1. Dissect fresh bone biopsies free of soft tissue and place in 50 mL polypropylene conical tubes filled with 70% ethanol (see Note 1) and place them on a rocking platform for 1–3 days depending on the sample size. 2. Conical tubes should be labeled using a Secureline marker or pencil as other markings can be removed by solvents during processing. The tissue must be completely immersed in the fixing solution (the ratio of ethanol volume to sample volume should be at least 15:1). The fixative may be changed daily depending on the fixation time.
3.1.2
Tissue Dehydration (13)
1. Solutions from the fixation step should be appropriately discarded before beginning tissue dehydration. All dehydration steps are carried out at room temperature on a rocking platform. 2. For the start of the dehydration phase, refill all labeled conical sample tubes with 95% ethanol. Change solutions accordingly while maintaining the same volume ratio:
298
S. Gourion-Arsiquaud et al.
— 95% ethanol (daily for 2 days) — 100% 2-propanol (daily for 2 days) — Xylene (daily for 2 days). 3. After dehydration, the samples need to be infiltrated and embedded with a plastic embedding mixture using a three-step methacrylate solution protocol (see Note 2) according to the Erben method (14) in the following. 3.1.3
Infiltration
Each methacrylate solution should be stirred for at least 10 minutes prior to use and used to infiltrate the tissue for 3 days. Change solutions (see Note 3) at the specified time points and return to rocking platform while maintaining a similar volume ratio (see Note 4). ●
●
●
MMA Solution I amounts per 100 mL: Stir 60 mL MMA (containing 100 ppm hydrochinon) + 25 mL butyl methacrylate (containing 10 ppm hydrochinon) + 5 mL methyl benzoate + 1.2 mL polyethylene glycol 200. MMA Solution II amounts per 100 mL: 100 mL MMA with 0.4 g dry benzoyl peroxide. Benzoyl peroxide must be fully dissolved before use (see Note 5). MMA Solution III amounts per 100 mL: 100 mL MMA with 0.8 g dry benzoyl peroxide. Benzoyl peroxide must be fully dissolved before use (see Note 6).
3.1.4
Embedding (Polymerization)
Polymerization is carried out in 15-mL plastic embedding molds. Depending on the sample size, smaller or bigger embedding molds may be used. 1. Before embedding the tissues, polymerize a solid layer of plastic (Spurr resin) (15) at the bottom of each embedding mold (about one-third the container volume) (see Note 7). This step creates a flat base on which the bone specimen can be oriented (see Note 8). The 100 ml of MMA solution III can be made cold by placing the aluminum foil-covered container inside a bucket of ice. Use a thermometer to verify the temperature of 4 °C before adding accelerant (N,N-dimethyl-p-toluidine) (see Note 9). 2. To embed samples, retrieve embedding molds (with prepared Spurr bases) and prepare polymerization mixture. Polymerization mixture: Add 400 µL of N,N-dimethyl-p-toluidine per 100 mL of cold (4 °C) MMA solution III and stir for a few minutes. Make sure polymerization mixture is kept cold (on ice) at all times. Completely fill embedding mold with polymerization mixture (see Note 10), place and orient the sample at the center of the plastic base, seal with an air-tight cap, and polymerize in deepfreezer at −20 °C. Polymerization takes approximately 1–3 days (see Note 11). To finish the polymerization, the samples can be placed for 24 hours in an oven at 60 °C. Polymerized blocks can be stored at room temperature.
20 FT-IR Microspectroscopy
3.1.5
299
Preparation of Bone Samples for the IR Analysis
1. Once the blocks have polymerized remove the plastic embedding molds and discard. 2. Cut and trim the PMMA blocks with a band saw and grinder/polisher to expose the embedded sample for sectioning (see Note 12) at room temperature. 3. Clamp the block in the microtome holder so that the bone is parallel to the knifeedge (see Note 13). 4. Start the cutting process from one end of the tungsten carbide blade; this will maximize the use of the whole blade edge. 5. Turn on the microtome and further trim the block until the desired sample area is reached. Lubricate the block with 30% alcohol with the aid of an artist’s brush while trimming and sectioning. 6. After trimming, cut desired sections at 1–3 µm (see Note 14). 7. Pick-up the cut section which the aid of a pair of fine forceps, place on a BaF2 window on which drops of 90% ethanol were added to stretch/flatten section (see Note 15). 8. Allow the sections to dry before storing at room temperature.
3.2
FTIRI Analysis
3.2.1
Collecting IR Data
1. First allow the spectrometer software to initialize the IR imaging system for data acquisition. Then place the sample on the motorized stage (see Note 16) of the Perkin Elmer Spectrum Spotlight 300 Imaging System or other imaging systems used, e.g., Bruker, Nicolet, Bio-Rad. Illuminate and focus the sample using the IR microscope. Check that the system energy level is adequate and the specimen is properly oriented. 2. A digital video camera attached to the microscope allows the acquisition of digital images and visualization of the sample areas selected for scanning. 3. Depending on the type of spectral analysis, scans may be acquired at a spectral resolution ranging from 2–16 cm−1 wavenumbers. For analysis of the phosphate or Amide I sub-bands a minimum spectral resolution of 4 cm−1 is advised. 4. Before data acquisition, a background reference spectrum should be collected (once for each sample window) using a clean sample free area on the BaF2 window (see Note 17). 5. IR data are collected in transmission as either line scans of individual spectra at discrete micron intervals or 2D images at spatial resolutions of ~6.25 µm or ~25 µm. Use of reflectance data is discouraged because of the well known dispersion artifact (16).
300
3.2.2
S. Gourion-Arsiquaud et al.
Processing the IR Data
1. The acquired IR data may be processed using a variety of different software platforms (BioRad, Grams, or Matlab). We have found ISYS (Spectral Dimensions Inc., Olney, MD) to be most straightforward. a. Bone sites and acquisition modes utilized for IR data collection depend on the research question. For example, when temporal changes in bone development are of interest, line scans going from the outer bone forming surface (periosteum) to the inner bone resorbing surface (endosteum) are most useful. For an osteoporosis study comparison between osteoporotic and healthy bone spatial images of different areas may prove more relevant (3). Parallel stained sections are useful for orientation purposes. Backscatter electron images can also be helpful in this regard. b. Because fully calcified tissues cannot be cut into thin sections without embedding in an equally hard material, embedding in plastic is generally performed (17). However, the process induces complications in the FTIR analysis of bone due to the IR active nature of PMMA. Spectral subtraction of the embedding medium must be used to remove these contributions. Depending on the IR data acquisition mode, image or single spectrum, the spectral subtraction of embedding medium is made either by using a single pure PMMA spectrum or specialized ISYS algorithms, respectively (3). 2. Before spectral processing, all data are truncated usually between 2,000 and 800 cm−1, linearly baselined and corrected (PMMA subtraction) using an ISYSspecific protocol (Fig. 20.4). A similar method may be developed using other software. 3. The primary IR peak areas of interest are then calculated: Amide I (matrix band at 1,592–1,712 cm−1), phosphate (mineral band at 916–1,180 cm−1) and carbonate (mineral band at 840–892 cm−1) peaks (see Note 18). These results are given by hyperspectral IR images showing their distributions (Fig. 20.3) in the selected sample area. 4. The IR parameters that are measured are: — Mineral/matrix area ratio. This validated parameter which measures bone mineral content (correlated to ash weight) is calculated by integrated area of phosphate (916–1,180 cm−1)/amide I (1,592–1,712 cm−1) (see Note 19). — Carbonate/phosphate area ratio, another validated parameter, indicates the total amount of carbonate substitution for hydroxide (type A), phosphate (type B), or surface (labile carbonate) in the mineral crystals (see Note 20). It is calculated by the integrated areas of carbonate (840–892 cm−1) / phosphate (916–1,180 cm−1). — Mineral crystallinity (1,030/1,020 cm−1 peak intensity ratio), a validated correlated mineral crystal size and perfection as determined by x-ray diffraction line broadening. — Collagen maturity (1,660/1,690 cm−1 peak intensity ratio) related to variations in collagen cross-linking network.
20 FT-IR Microspectroscopy bone spectrum (2,2)
0.7 0.6
0.5
0.5
0.4
0.4 0.3 0.2
0.3 0.2 0.1
0.1 0 2000
0
1800
1600
1400
1200
1000
Single Beam Intensity
4 3 2 1 0
1800
1600
1400
1200
1000
800
Wavenumber (cm−1)
1600
1400
1200
1000
800
1000
800
Wavenumber (cm−1) S-MMA.spf (1,1)
1.2
5
c
1800
b
bone spectrum (2,2)
6
Scaled Data
−0.1 2000
800
Wavenumber (cm−1)
a
−1 2000
bone spectrum (2,2)
0.6
Absorbance
Absorbance
301
1 0.8 0.6 0.4 0.2 0 −0.2 2000
d
1800
1600
1400
1200
Wavelength (microns)
bone spectrum (3,3)
14 12
Scaled Data
10 8 6 4 2 0 −2 2000
1800
1600
1400
1200
1000
800
Wavenumber (cm−1)
e
Fig. 20.4 Treatment of the IR spectra, (a) typical spectrum of ethanol-fixed bone in PMMA, (b) using ISYS software (Spectral Dimensions Inc., Olney, MD) the spectra were baseline corrected from 720 to 2,000 cm−1, (c) the spectra were “normalized” to the PMMA band at 1,728 cm−1 (see arrows), (d) spectrum from the embedding material (PMMA) alone after “normalization”; the normalized PMMA spectrum is then subtracted from each of the normalized spectra creating (e) spectra baselined without PMMA contribution from which the different IR parameters will be calculated.
4
Notes
1. Tissues processed for IR are fixed in 70% ethanol due to IR interference from other fixatives (e.g., formalin). 2. It is very important that tissues for FT-IR are dehydrated to prevent water interference in the IR spectra; therefore, water should not be in contact with any of the processing solutions! 3. All steps involved in the preparation of bone specimens must be handled under an operational fume hood and require the wearing of gloves.
302
S. Gourion-Arsiquaud et al.
4. Each MMA solution is per 100 mL (i.e., for a total volume of 400 mL, say 10 tubes with 40-mL solution in each), multiply each solution amount by 4). 5. The benzoyl peroxide must be stored at 4 °C. 6. Any waste solution containing methyl methacrylate must be discarded according to institutional waste disposal policies. 7. Add each component (by weight) in turn to a disposable plastic beaker; 10 g vinyl cyclohexane dioxide (ERL 4221), 4 g diglycidyl ether polypropylene glycol (DER 736), 26 g non-enyl succinic anhydride (NSA), 0.4 g dimethylamino ethanol (DMAE). Add the catalyst (DMAE) last after gently mixing the three other components. Mix the complete formula thoroughly, add to the embedding molds (~1/3 height) and polymerize overnight in an oven at 70 °C. 8. Preparation of a polymerized base at the bottom of the embedding mold may or may not be necessary depending on the container used. 9. Embedding molds should be prepared in advance for tissue embedding. 10. The polymerization process is sensitive to oxygen. The infiltrated bone samples will not be able to polymerize in the presence of oxygen. 11. If after 3 or 4 days some samples have not polymerized or are only partially polymerized, empty the embedding mold and refill with new polymerization mixture and place them in the freezer. 12. You need to conserve only between 1–2 mm of plastic around the embedded bone. The plastic remaining is superfluous. 13. Specimen orientation is critical for proper analysis. Bone samples should be oriented with cortical bone either perpendicular or parallel to the edge of the PMMA block, thus allowing transverse and longitudinal sections to be obtained. 14. It is important to maintain this thickness to prevent detector saturation. 15. To flatten and to avoid any rippling, the sections may be sandwiched between two 25 mm BaF2 windows. 16. To keep cut sections flat, a second BaF2 window may be placed on top of the sample window before FTIRI analysis. 17. Background and IR spectra must have the same spectral resolution. 18. Another phosphate contribution, the ν4 PO43− band around (500–650 cm−1) that falls in a frequency region below the cutoff of the mercury-cadmium-telluride detectors usually used in commercial IR microscopy, can become available with utilization of detectors with cutoff <500 cm−1 as DTGS-polyethylene detector. In this configuration, the phosphate ν4 mode would be observable; therefore, before spectral processing, all data should be truncated and baselined until 400 cm−1. 19. Following the previous note it is possible with this equipment to calculate the mineral/matrix parameter by the ratio ν4 PO43− (500–650 cm−1) to Amide I (1,592–1,712 cm−1). 20. As the carbonate band is weak compared to the other bands (Amide I, phosphate), be aware that overestimation of the background and the baseline correction may cutoff one part of the carbonate band reducing the carbonate:mineral ratio. Using a second thicker section for determination of the carbonate contribution can circumvent this error.
20 FT-IR Microspectroscopy
303
Acknowledgements This work was supported by NIH grant AR 041325 and Core Center grant AR046121.
References 1. Paschalis, E. P., Betts, F., DiCarlo, E., et al (1997) FTIR microspectroscopic analysis of human iliac crest biopsies from untreated osteoporotic bone. Calcif Tissue Int 61, 487–492. 2. Mendelsohn, R., Paschalis, E. P., Sherman, P. J., et al. (2000) IR microscopic imaging of pathological states and fracture healing of bone. Appln Spectrosc 54, 1183–1191. 3. Boskey, A. L., Mendelsohn, R. (2005) Infrared analysis of bone in health and disease. J Biomed Opt 10, 031102–031105. 4. Faibish, D., Gomes, A., Boivin, G., et al. (2005) Infrared imaging of calcified tissue in bone biopsies from adults with osteomalacia. Bone 36, 6–10. 5. Ou-Yang, H., Paschalis, E. P., Mayo, W. E., et al (2001) Infrared microscopic imaging of bone: spatial distribution of CO3(2–). J Bone Miner Res 16, 893–900. 6. Mayer, I., Schneider, S., Sydney-Zax, M., et al. (1990) Thermal decomposition of developing enamel. Calcif Tissue Int 46, 254–257. 7. Paschalis, E. P., DiCarlo, E., Betts, F., et al (1996) FTIR microspectroscopic analysis of human osteonal bone. Calcif Tissue Int 59, 480–487. 8. Gadaleta, S. J., Paschalis, E. P., Betts, F., et al. (1996) Fourier transform infrared spectroscopy of the solution-mediated conversion of amorphous calcium phosphate to hydroxyapatite: new correlations between X-ray diffraction and infrared data. Calcif Tissue Int 58, 9–16. 9. Pleshko, N., Boskey, A. L., Mendelsohn, R. (1991) Novel infrared spectroscopic method for the determination of crystallinity of hydroxyapatite minerals. Biophys J 60, 786–793. 10. Knott, L., Bailey, A. J. (1998) Collagen cross-links in mineralizing tissues: a review of their chemistry, function, and clinical relevance. Bone 22, 181–187. 11. Paschalis, E. P., Verdelis, K., Doty, S. B., et al. (2001) Spectroscopic characterization of collagen cross-links in bone. Bone Miner Res 16, 1821–1828. 12. Faibish, D., Ott, S. M., Boskey, A. L. (2006) Mineral changes in osteoporosis: a review. Clin Orthop Relat Res 443, 28–38. 13. Schenk, R. K, Olah, A. J., Herrmann, W. (1984) Preparation of calcified tissues for light microscopy, in (Dickson GR, ed.), Methods of Calcified Tissue Preparation, vol. 1. Elsevier Science Publishers Amsterdam. 14. Erben, R. G. (1997) Embedding of bone samples in methylmethacrylate: an improved method suitable for bone histomorphometry, histochemistry, and immunohistochemistry. J Histochem Cytochem 45, 307–314. 15. Spurr, A.R. (1969). A low-viscosity epoxy resin embedding medium for electron microscopy. J Ultrastruct Res 26, 31–43. 16. Romeo, A., Diem, M. (2005) Correction of dispersive line shape artifact observed in diffuse reflection infrared spectroscopy and absorption/reflection (transfection) infrared microspectroscopy. Vib Spectrosc 38, 129–132. 17. Aparicio, S., Doty, S. B., Camacho, N. P., et al. (2002) Optimal methods for processing mineralized tissues for Fourier transform infrared microspectroscopy. Calcif Tissue Int 5, 422–429
Chapter 21
Assessment of the In Vivo Adaptive Response to Mechanical Loading Leanne Kaye Saxon and Lance Edward Lanyon
Abstract The primary responsibility of the skeleton is to bear the loads involved in physical activity without sustaining damage. This capability involves a mechanism in which bone cells “assess” the suitability of the bones’ existing architecture in relation to their prevailing loading environment and adapt or maintain it accordingly. It is widely assumed that the loading-related variable to which the bone cells respond is the strain engendered within the bone tissue. Strain-related adaptation is essential for normal bone development, regulation of strength in relation to exercise, healing of fractures and the success of orthopedic interference. The most widespread failure of this adaptive response is osteoporosis. Although strain-related bone adaptation can be investigated in these situations in humans in vivo, both the inputs and outputs are difficult to assess and control, and the tissues are unavailable for study. As a consequence much of our understanding of the mechanisms involved in strain-related adaptation have come from studies in animals where the strains within the bone are measured, the loads imposed on the bones can be controlled, and the adaptive changes to cells and architecture determined. Although many animals have been used in the past (1–3), now the most commonly used animal is the mouse. Normal and transgenic mice are available and inbred strains of mice are well characterized physiologically (4). Most importantly, the techniques for measuring strains in mouse bones, loading these bones in a controlled manner, and assessing changes in architecture are all available. This chapter outlines the techniques involved in these three phases of investigation in mouse bone. Keywords In vivo, mechanical loading, biomechanics, strain gauging, mouse bone.
1 1.1
Introduction Measurement of Bone Strains In Vivo
The magnitude of load (in Newtons, N) used during the loading experiment depends on how much strain needs to be applied to bone. When the bone is loaded From: Methods in Molecular Biology, Vol. 455: Osteoporosis: Methods and Protocols Edited by Jennifer J. Westendorf © Humana Press, Totowa, NJ
307
308
L.K. Saxon and L.E. Lanyon
it deforms, and this deformation can be expressed as strain. Strain (ε) is the change in a dimension as a proportion of the original dimension (ε = δL/L). A high load may deform a bone by 3% of its original length, which is a strain of −0.03. Strain can be positive (tensile) or negative (compressive) and because it is a ratio, it is dimensionless. Strain is often expressed as a percentage, i.e., strain × 100 or as microstrain (strain × 106) (5). Because bone adapts to the strains it is subjected to, it is important in any loading experiment to know how much strain is engendered within a bone for a given amount of load. Because nearly all bones grow and develop under conditions of normal activity, it is also necessary to know the normal (usually locomotor) strains. During locomotion, mouse long bones experience compressive strains in the range 200 µε up to 600 µε from a 30-cm jump (6). If this were the whole story, strains >600 µε would be sufficient to initiate new bone formation. However, peak strain magnitude is not the sole strain-related stimulus, because strain rate and strain distribution are also important. In the mouse bone investigated, using the loading modalities employed, strains >1,200 µε are needed to stimulate significant osteogenesis (6–8). To date, the electrical resistance strain gauge, a device that measures the deformation of the structure to which it is attached, is the only method capable of measuring strains on the bone’s surface ex vivo or in vivo during normal activity or artificial loading under a closed incision. Strain gauges consist of a fine wire supported in a plastic or epoxy backing substance that is bonded to the bone’s surface. As the bone surface deforms under load, the backing material with its enclosed wire deforms with it. The change in length of the wire alters its resistance in proportion to the change in its dimension. Thus, by measuring the change in electrical resistance, the strain of the underlying bone can be determined. The gauge factor (GF) is a measure of the strain gauge’s sensitivity to strain. Gauge factor is defined as the ratio of fractional change in electrical resistance to the fractional change in length (strain). GF = (∆R/RG)/ε, where RG is the resistance of the undeformed gauge, ∆R is the change in resistance caused by strain, and ε is strain. Generally provided by the strain gauge manufacturer, the gauge factor specification is valid only for a specific excitation voltage and ambient temperature. For metallic foil gauges the gauge factor is 2; however, gauge factors for special strain gauges can be significantly larger or even negative. Strain gauges are available commercially with nominal resistance values from 30 to 3,000 Ω (Ohms), with 120 Ω being the most common for strain gauging mice and 350 Ω for load cell transducers. When the choice exists, the higher-resistance gauge is preferable because it decreases the signal-to-noise ratio by decreasing lead wire effects and unwanted signal variations caused by temperature fluctuations. However, higher-resistance gauges require more current. To measure the change in resistance from the strain gauge, the values need to be converted to a change in voltage; therefore, the strain gauge must be connected to an amplifier. Together the strain gauge and the amplifier form a Wheatstone bridge (full, half, or quarter bridge) that consists of four balanced resistors or arms. To measure the change in resistance of a single strain gauge using a Wheatstone bridge, the gauge is attached to one arm and the three remaining arms of the bridge
21 Mechanical Loading In Vivo
309
Three-wire, quarter-bridge strain gauge circuit R2
R1 V R3
Rwire1
Rgauge
Rwire3 Rwire2
Fig. 21.1 The bridge’s state of balance is solely dependent on the ratios of R1/R3 and R2/R4. To measure resistance with a Wheatstone bridge, an unknown resistance, i.e., strain gauge, is connected to one arm, i.e., R4, whereas the other three resistors are precision devices of known value. Thus, with no force applied to the strain gauge, the bridge is symmetrically balanced and the voltmeter indicates zero volts, representing zero force on the strain gauge. As the strain gauge is either compressed or tensed, its resistance decreases or increases, respectively, thus unbalancing the bridge and producing an indication at the voltmeter. Note that the wire resistances Rwire1 and Rwire2, being in series with Rgauge, also contribute to the resistance of the lower half of the arm of the bridge, and consequently contribute to the voltmeter’s indication. This, of course, will be falsely interpreted by the meter as physical strain on the gauge. Although this effect cannot be completely eliminated in this configuration, it can be minimized with the addition of a third wire, connecting the right side of the voltmeter directly to the upper wire of the strain gauge.
must have a known resistance. When a strain is applied to the strain gauge, it will unbalance the Wheatstone bridge and the output voltage (V0) is directly related to the strain applied to the strain gauge (Fig. 21.1). For a thorough bibliography of the use of strain gauges to measure bone deformation refer to the book chapter by Fritton and Rubin (2001) (8). To aid in the selection of strain gauges, go to www.vishay.com; and for a review of the advantages and disadvantages of strain gauges, see Note 1. Materials needed for strain gauging are listed in Section 2.1 and a description of how to measure strain in excised mouse bone follows in Section 3.1.
1.2
Application of Bone Loads In Vivo
Non-invasive axial loading has a number of advantages over other loading techniques because the loads are applied to the bone under anesthetic through the skin and the bones articular surfaces rather than through the periosteum as occurs in some other models (9). The non-invasive axial loading model was first developed in the rat ulna by Torrance et al. (7) and then extended to the mouse ulna (10) and tibia (9). This model enables loads to be applied to the bone without surgical intervention and with normal cage activity in between loading sessions (7). Another advantage of the axial loading model
310
L.K. Saxon and L.E. Lanyon
over other techniques, such as running on a treadmill or using a vibrating plate (11, 12), is that it allows side-to-side comparisons between the loaded versus non-loaded bone hence eliminating confounding factors, such as diet, background cage activity, hormone levels and weight, that may affect differences between experimental groups. Section 3.2. details the methods of applying an axial load to the mouse tibia or ulna using the Dartec machine (Zwick Roell Group, UK) (Fig. 21.2). The use of the tibia for axial loading is a recent development and, unlike the ulna, allows the measurement of loading related changes in trabecular as well as cortical bone. During loading, the bone is positioned in between custom-designed upper and lower loading cups that are screwed into the actuator and load cell, respectively, and need to: (1) reduce lateral movement of the bone during loading, and (2) minimize stress to the joints.
Fig. 21.2 Dartec HC10: used to apply an axial mechanical load to rodent bone.
21 Mechanical Loading In Vivo
311
Fig. 21.3 3D reconstruction of the cancellous bone in the proximal tibia, allowing quantification of BV/TV, Tb.Th. Tb.Sp to name a few (left). Microscopic slide of an un-decalcified section of ulna from a growing rat showing the fluorochrome labels in the bone deposited when these substances were in the circulation (right).
1.3
Assessment of Adaptive Modeling and Remodeling In Vivo
The adaptive response to mechanical loading can be assessed using a number of methods that are not discussed in detail. If the duration of the loading regimen is sufficient (i.e., 3 days/week for 2 weeks), changes in bone geometry and cancellous bone architecture can be measured by micro-CT (Fig. 21.3). If the duration is shorter (i.e., 1–3 days), the loading response is best measured by histology and dynamic histomorphometry. By giving two short courses of fluorochrome labels (i.e., calcein or alizarin), which binds to calcium at sites of new bone formation, bone formation can be visualized under the microscope in calcified sections of the bone (see Fig. 21.3). Keeping in mind that mechanical strain varies along the length of the bone, so will the response to loading. Finite element modeling can be used to estimate the strain distribution along the diaphysis and in the cancellous bone and predict sites of greatest bone formation. Static histomorphometry is also useful for assessing the response to loading in cancellous and cortical bone. That is, staining bone sections and quantifying the presence of osteoblasts, osteoclasts, and osteoid and eroded surfaces relative to how much bone is present. For cancellous bone, the percent bone volume fraction (BV/TV), surface of bone in the bone tissue volume (BS/TV), trabecular thickness (Tb.Th [µm]), and trabecular separation (Tb.Sp [µm]) can be quantified.
2
Materials
2.1 1. 2. 3. 4.
Measurement of Bone Strains In Vivo
Saline: 0.9% NaCl solution. Foam board. Multimeter. Strain gauge wire (38 TDQ wire from Phoenix Wire Co.).
312
L.K. Saxon and L.E. Lanyon
5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19.
Pins. 70% ethanol. Chloroform or degreaser spray (Vishay, PA). Strain gauges. Solder (361A-20R, Vishay). Light microscope. Iron razors. Cotton swabs. Scalpel blades. Tweezers. Glue catalyst-C (Vishay). Soldering M-flux (Vishay). M-Bond adhesive glue (Vishay) or superglue. Amplifier (e.g., 2100 Amplifier System, Vishay). Dartec machine.
2.2
Application of Bone Loads In Vivo
1. Anaesthetic unit. 2. Dartec HC10. This servo-hydraulic machine is designed to perform a range of general testing, from static to high frequency. When applying a mechanical load to animal bone it is used for dynamic testing. The frame consists of a bottom crosshead, which is bolted to two fabricated feet; the feet can be bolted to the bench in order to provide increased stability when required. Two 60-mm diameter columns are clamped to the bottom cross-head and form the support for the top cross-head and actuator assembly. The actuator applies the axial load to bone and houses a stroke transducer that measures the linear displacement of the actuator (movement in the up–down direction). The stroke transducer or LVDT is concentrically mounted within the hydraulic actuator and when the piston moves it precisely monitors the linear movement. The load cell is mounted on the end of the piston rod and measures the degree of force applied to the bone during loading. Finally, the Dartec consists of a servo valve that regulates the flow of oil that, along with the stroke transducer and load cell, is connected to a hydraulic power unit or pump (see Fig. 21.4).
3
Methods
3.1
Measurement of Bone Strains In Vivo
3.1.1
Strain Gauge Preparation
1. Strain gauges are assembled into an implantable unit before the day of strain gauging (Fig. 21.5). Using Vishay Micro Measurements strain gauges
21 Mechanical Loading In Vivo
L
R
313
P
L
Z G
A
F D
E C
B
D
Horizontal Testspace
410 mm
E F
Height of Bottom Tool Face Adjustable Vertical Testspace
175 mm 100-700 mm
G L P
Actuator Stroke Lifting Point Threads Hydraulic Pressure Input
100 mm M12 ¼” BSP
SPECIFICATION System Operating Pressure 207 Bar 0.03 mm / 100 mm Parallelism 0.05 mm 0-350 mm + 0.1mm / m Alignment Eccentricity above 350 mm Stiffness 280 kN / mm @ 725 mm vertical daylight. 160 kg Net Mass 20 l / min Standard (other ratings Servo Valve Rating available) Guards Optional Colour Polished Aluminium 1.6 x 0.8 x 0.7 m (Nominal) Export Shipping Dimensions
R
Hydraulic Return Output
¼” BSP
Maximum Speed
Z
Hydraulic Drain Connection
1/8” BSP
A B C
DRAWING KEY (All dimensions are in mm) Machine Maximum Height 1500 mm Machine Width 600 mm Machine Depth 500 mm
Frame Capacity
500 mm/sec with standard valve. 25 kN
Fig. 21.4 Schematic diagram and details of the Dartec HC10. All dimensions are in mm.
Fig. 21.5 Micro Measurement strain gauge (06-015-DJ-120) used to measure the strains applied to bone during mechanical loading.
314
2. 3.
4.
5.
6. 7.
L.K. Saxon and L.E. Lanyon
(06-015-DJ-120), use scotch tape to tape the gauge down flat onto a dissecting board and ensure it is visible under a microscope. Trim three sides of the backing using a single-edged razor blade. Lightly scratch each terminal using a pin (this will help the solder stay on the terminal) and apply a small drop of soldering M-Flux AR to each terminal. Then apply a small bead of solder to each terminal using a fine-tipped soldering iron. Prepare the strain gauge lead wires by skinning the tips of two short (~9-mm in length) 38 TDQ wires that have been wrapped together. Dip the ends in flux and apply a small bead of solder to each end. Re-melt the solder beads on the gauge terminals and insert the skinned strain gauge lead wires into the hot beads to establish electrical continuity. Allow the beads to cool and test the resistance of the gauge using a multimeter (should be 120.0 ± 0.3 Ω). Coat the gauge with a light spray of polyurethane (we use clear lacquer) to waterproof the gauge. Trim the last side of the strain gauge that is stuck down by the scotch tape.
3.1.2
Prepare the Bones
1. After euthanizing the mouse, extract the limb with the bone you wish to attach the strain gauge to, leaving all the muscle and skin attached. Store in 70% alcohol. 2. The day before strain gauging, transfer the limb to 0.9% NaCl solution and incubate overnight at room temperature.
3.1.3
Attach the Gauge
1. Securely pin the excised leg or arm to a Styrofoam board. If strain gauging the ulna, have the lateral side facing up. If strain gauging the tibia, expose the medial side. 2. Using a sharp scalpel blade, slice through the muscle to expose the underlying bone. Using a tightly rolled cotton swab, rub the bone to get a nice shine on the bone surface. Spray a clean swab with CSM-1 degreaser or dip in alcohol and quickly wipe down the center of the bone to degrease. 3. When strain gauging, it is important that the strain gauge is positioned at the same site for each bone. This site can be determined by measuring bone length from radiographs and positioning the gauge at a calculated length along the bone (i.e., 50% of bone length) or by using anatomical landmarks. The site chosen must allow the gauge to be attached evenly and correspond to a region where significant osteogenic changes occur in response to loading. 4. Use a pencil to mark the bone where the center of the gauge should be attached. Brush some Catalyst-C onto a soft surface (we use the foil wrapper of an opened scalpel blade) and immediately dip the back of the gauge into the film of catalyst. Squeeze a small dot of M-Bond Adhesive glue (we use superglue) onto the same surface and quickly dip in the back of the gauge.
21 Mechanical Loading In Vivo
315
5. Lay the gauge in the correct position on the bone (the gauge should run parallel to the length of the bone) and gently push down on the solder bumps on the gauge terminals one at a time. Hold for ∼20 seconds, and then prop up the wire with the tweezers so that the weight of the wire does not peel the gauge up. 6. Wait 1.5–2.0 minutes and apply another thin coating of polyurethane spray to the bone. The purpose of this coating is to seal the places on the gauge terminals where you pushed with the tweezers. 7. Until you are ready to measure strain, store the bones in a vial containing saline to keep them hydrated.
3.1.4
Measure Strains
1. For data collection, the stain gauge wires are connected to a shielded cable that leads to an amplifier that converts the change in resistance of each gauge to a change in voltage, which has a known relationship with microstrain. The amplifier is connected to the Dartec machine that will display the microstrain reading. (Note: Other amplifiers can be used; however, the following methods are specific to the 2100 Amplifier System in conjunction with the Dartec HC10 machine.) 2. Before turning on the amplifier, check EXCIT toggle switches are turned off and the CAL switches are in the center (OFF) position. Turn on the 2100 amplifier system. The red pilot lamp should light up. Attach the 3-OUPUT lead wires from the back of the amplifier to the power board of the Dartec machine. 3. Select a channel from which you will connect the strain gauge to. Ensure that the Dartec is programmed to read this channel (check software under workshop → configuration → define channels). 4. Connect the wires of a shielded cable to the input pin of the selected channel in accordance with Fig. 21.6. 5. Turn the channel selector to AC. The meter should read between 9 and 11 on the scale. 6. Turn the channel selector to DC. The meter should read close to 10 on the scale. 7. Turn the selector to the channel you will use. For strain gauging we use ∼1 V to minimize error due to self-heating. If the voltage needs to be adjusted, use a small screwdriver to read the desired BRIDGE VOLTS on the power supply meter.
2 leads that join together
1 lead
Note: Pins J and K not normally used
A
B C
J D
H K
E
G F
Fig. 21.6 Input plug pin arrangement.
Shield lead = ground lead Jumper required for half and quarter bridges
316
L.K. Saxon and L.E. Lanyon
8. Adjust the GAIN for the channel. The gain needs to be high enough to detect the lowest output voltage, if it is too low you will only detect background noise. 9. Adjust the amplifier balance for the channel with the EXCIT switch off. Using a small screwdriver, adjust the AMP BAL until both OUTPUT lamps are off. 10. Turn on the Dartec HC10 machine, follow steps 2–7 and 12–13. For step 5, have one Feedback window showing the microstrain (µε) output from the channel the strain gauge is attached. 11. Attach the two strain gauge wires to the shielded cable using a soldering iron. 12. Turn on the EXCIT switch for the selected channel and adjust the balance to extinguish the OUTPUT lamps. In doing so you are balancing the Wheatstone bridge. 13. To check the channel is calibrated correctly, turn the CAL switch to position A to show +500 µε and to position B to show −500 µε. Adjust these values if need be using the BALANCE. 14. Position the bone with the gauge attached in-between the loading cups. Use the cycle generator or the ramp program to apply a range of loads (N) and record the corresponding strain (µε) as shown in the Feedback window. 15. Ensure that you get at least three repeatable measures for each load and there is a linear increase in microstrain with load.
3.2
Application of Bone Loads In Vivo
The following methods are specific to the Dartec HC10. This machine is one of many materials testing units available on the market today, other machines can be purchased from EnduraTEC (BOSE Corp., US), Instron Ltd. (US), and Lloyd Instruments (AMTEK Inc., UK).
3.2.1
Setting Up the Dartec
1. To use the Dartec HC10 you must be trained as an authorized user; if you have not been trained contact Zwick/Roell (www.zwick.com). If working in the United Kingdom, you must be working under an approved institutional project license and have a personal license from the Home Office (http://scienceandresearch.homeoffice.gov.uk/animal-research/) that allows you to conduct mechanical loading and anesthetic procedures on rodents. 2. Turn on the Dartec machine by switching the button on the tower unit. 3. On the computer, open the software Workshop 96, select the window Toolkit 96. Ensure that the software shows a yellow light next to Isolator. The status display window will indicate the software is in SET UP mode. 4. Check the following are selected: under Options ® Toolbar and Monitor Line and under Settings® Auto Load Default Screen.
21 Mechanical Loading In Vivo
317
5. The feedback can be displayed by selecting Tools-> Status or by clicking the Feedback button from the toolbar. Under Settings-> Feedback Channel select Actuator 1 and then Clone to open another window. In one window, go to Monitor-> Upper Peak (the smallest recorded peak value) and in the other window select Lower Peak (the maximum recorded peak value). For both windows, select Rest Interval = 10 seconds. 6. On the top toolbar, click on Pump Start to start the hydraulic pump and select Main Pressure to apply hydraulic pressure to the actuator. 7. Open the Systems Configuration window. Open the Offset/Gains window and zero the load cell with no sample mounted. Check that the Offset Number for the load cell is not greater than the capacity of the load cell, if it is the load cell is damaged. 8. Under Cycle Generator-> Waveform enter in the desired loading protocol. (See Notes 2 and 3 for a typical loading program and factors that need to be considered.) Actuator needs to be Actuator 1. Control mode is the mode it will cycle in (i.e., Load Cell). Wave type is the pattern of loading and Number of Cycles is the number of impacts you want to apply to the bone. Enter in the low and high magnitudes of load you wish to apply in Level A and B, respectively. Note the loads are negative indicating a compressive load. Enter in Hold Time A and Hold Time B, for the time to be spent at Level A and B (seconds) and the Fall Time and Rise Time, or strain rate (N/second). Depending on which waveform selected you may also need to enter in Frequency (number of cycles per second). Other variables such as Amplitude, Mean, Fall Rate, and Rise Rate will be calculated automatically. Click on Send, Read to save. 9. Under Cycle Generator Panel-> Define Option, select Enable. Under Select Options Mode, select Peak Control and enter the maximum and minimum loads (N) to be applied to the bone (same as values entered for Level A and B). This allows the upper and lower levels to be controlled and maintained. Click Send, Read to save. 10. Under Cycle Generator-> Advance, ensure the Stop Load (A or B) is the resting load. 11. See Note 4 on Occupational Health and Safety for setting limits to the machine to prevent damage to the load cell and for checking the accuracy of the load cell. 12. Carefully screw the upper and lower loading cups into the end of the actuator and load cell, respectively. (Ideally screw in the lower cup when the machine is turned off.) Use the button ++ on the Manual Control panel to move the actuator up if needs be. The lower loading cup should protrude through a stage that the mouse will rest on during loading. Insure the stage does not touch the load cell or the lower cup. 13. In the Offset/Gains window, insure the Actuator is Actuator 1 and the Channel is Load Cell. The P (Proportional), I (Integral), and D (Derivative) gains need to be entered in the text boxes. The P value is less for stiffer samples and greater for more elastic samples. The I value is generally ~one-third of the p-value and the d-value is ∼1/10 of the p-value. When entered, click on Send Gains and
318
L.K. Saxon and L.E. Lanyon
Read Gains. The Dartec is a closed-loop system, which means the input signal is adjusted by the PID values to achieve the desired output. Therefore, these values are critical for ensuring the waveform is not over- or undershooting the desired load. For mouse bone, we use p = 12, i = 4, d = 1.2. 14. To determine if the PID gains are correct, place a bone sample in between the loading cups (see Section 3.2.2.) and from Tools open Oscilloscope; this allows high speed capture displays from the actuator. For Actuator select Actuator 1, the Capture Period is the duration you want to capture the waveform and the Feedback Channel is the channel you wish to monitor (i.e., Load Cell). To change any parameters turn the Scope off, conversely when all parameters are set, enable the Scope. It is best to run a square wave to check the gains are correct. The PID gains can be changed from the System window until the waveform is read back as a perfect square wave (Fig. 21.7). 15. When all parameters are set, induce anaesthesia to the animal using either an injectable or inhalant anaesthetic. It is preferable to use an inhalant anaesthetic (i.e., Halothane or Isoflurane) because it provides more control over the dosage and the animals recover faster.
3.2.2
Mechanical Loading of the Tibia/Ulna
1. If loading the tibia, position the mouse’s right ankle in the lower cup and bend the knee so the tibia is vertical. Alternatively if loading the ulna, place the mouse’s right elbow in the lower cup. 2. Use the Manual Control window to move the actuator down (− = slow, −− = fast) until a force of approximately −2 N is applied to the knee or −0.2 N to the dorsal surface of the volar flexed wrist (Fig. 21.8).
P gain too low
P gain too high
P gain correct
D gain applied to improve rise time
Fig. 21.7 Pattern from the oscilloscope indicating the effect P values have on a square waveform. The PID values can be adjusted to improve the feedback so that a perfect waveform is applied to bone.
21 Mechanical Loading In Vivo
319
Fig. 21.8 Compressive loading of the mouse tibia (left) and ulna (right) using a Dartec mechanical testing apparatus. The greatest bending moment occurs near the mid-shaft in the medial-lateral direction.
3. When the bone is positioned correctly, click Start to begin loading. The mode will change from SET UP to LOAD CELL. 4. The Cycle Generator window shows you the number of cycles completed and the Feedback window indicates what loads are being applied to the bone. When the loading cycles are complete, click the Global Setup button in the top right hand corner so that the actuator is in SETUP mode. (This will help prevent damage to the load cell.) The Manual Control window will now display mm, not N. 5. Using the Manual Control panel, move the actuator up (+ = slow, ++ = fast) and remove the mouse from the machine. 6. Weigh the mouse and observe their recovery. When the animal is no longer recombinant, they have fully recovered. 7. When all the mice have been loaded, click Pump Stop and remove the upper loading cup. Turn off the machine at the main tower unit and close the software.
4
Notes
1. Strain gauges have a number of advantages; they are universal, simple to use, have low mass, show high stability over time, demonstrate excellent linearity over a large strain range and are relatively cheap. However, they also have their disadvantages. The change in resistance is very low; therefore, an amplifier is needed to measure
320
L.K. Saxon and L.E. Lanyon
strain, they are for single use only, they need protection against temperature and moisture, and their correct positioning on bone is critical. 2. This is a typical protocol we use for mechanically loading mouse tibia. Actuator: Actuator 1 Control Mode: 20 N Load Cell Wave Type: Trapezoid No Cycles: 40 Level A: −2.00 N Amplitude: 5 N Time Period: 10.1 s Hold Time A: 10 s or 14.9 s Fall Time A: 0.025 s Fall Rate: 400 N/s
Level B: −13 N Mean: −7 N Frequency: (not relevant for trapezoid waveform) Hold Time B: 0.05 s Fall Time B: 0.025 s Fall Rise: 400 N/s
3. Several factors need to be considered when designing the loading program. It is important to consider how long each loading session will be, how many days per week you will apply loading, the frequency of the loading waveform, the type of waveform, the strain rate and whether you will introduce a rest period in between loading cycles. Although no data are available on how bone responds to three alternate days versus five continuous days of axial loading a week, rest days allow the animal to recover from anaesthetic and stress to the joints. Rest periods in between loading cycles (i.e., 10 or 14.9 seconds rest) also appear to enhance the osteogenic response to loading (13, 14). As a result, the number of cycles required to induce significant bone formation is diminished and the magnitude of the response to strain is enhanced (13, 14). This is based on the notion that rest periods enhance canalicular fluid flow and stimulation of osteocytes (15). Therefore, rest periods are important for recovery and for enhancing the response to loading. Fig. 21.9 examines the importance of wave form, number of cycles, strain rate, and magnitude of load when designing the loading regimen. 4. Remember these are Occupational Health and Safety tips: (1) In case of emergency, use the red emergency stop button on the left hand side of the Dartec machine. (2) Reduce the pressure of the pump to avoid unnecessarily large forces being applied by the machine. (3) Check the load cell is calibrated correctly by placing a known weight directly on the load cell (one that will not exceed the load cells capacity), and check that the load output in Newton’s is correct. Use the following equation to calculate the expected output of a known weight: Kilogram (kg) × 9.807 = × N Eg 2 kg × 9.807 = 19.614 N Therefore, if a 2-kg weight is placed on the load cell, the output should read −19.614 N. This is the maximum load you should place on a 20 N load cell. If a
21 Mechanical Loading In Vivo
321
LOADING REGIMEN Trapezoid Wave = constant strain rate = incorporates rest periods which ↑ osteogenice response = longer time mouse is under anaesthetic
Trapezoid Wave
Sine Wave = variable strain rate = no rest periods = shorter loading time, therefore less time mouse is anaesthetised
Sine Wave
Time (seconds)
0 −X N
−X Strain Rate Shorter Shorter time time to to peak peak == greater greater osteogenic response Magnitude of Load (N) = needs to be sufficient to apply a high enough strain (µε). Strain same as background cage activity = no response Strains above background = osteogenic response
Number of cycles 5-36 cycles = enough to show an osteogenic response >100 cycles = not necessarily more beneficial. May induce microdamage. Strain Frequency 1-10 Hz = osteogenic 10 -100 Hz = response depends on site chosen
Fig. 21.9 Factors to consider when designing a loading regimen for axial loading experiments in rodents.
heavier weight is used, it will permanently damage the load cell. (4) Under Tools → Limits set the error trips to stop the machine if it is overdriven. Error trips detect a difference between the required load and the actual load applied to the bone. To program error trips, select Mode → i.e., Load Cell and Lower Limit—enter in 1N over the load cell capacity (i.e., 21 N), select an Action to take place if the limit is exceeded (i.e., Stop Generator) and under Enable Limits select Active, Send, and Read. Should one or more limit detectors be activated, the Controller Status display will change to indicate either DETECTED or ACTIVATED. These limits are crucial to prevent damage to the load cell. (5) Do not place fingers in between the loading cups when the machine is running and ideally screw in the bottom-loading cup when the machine is turned off. Acknowledgments The authors thank Alan Thomas from Zwick/Roell, Alex Robling from Indiana University and Massimo Marenzana from the Royal Veterinary College for their contributions to this chapter. This work was supported by the Wellcome Trust.
References 1. Rubin, C. T., Lanyon, L. E. (1985) Regulation of bone mass by mechanical strain magnitude. Calcif Tissue Int 37, 411–417. 2. Lanyon, L. E., Rubin, C. T., Baust, G. (1986) Modulation of bone loss during calcium insufficiency by controlled dynamic loading. Calcif Tissue Int 38, 209–216. 3. Judex, S., Boyd, S., Qin, Y. X., et al. (2003) Adaptations of trabecular bone to low magnitude vibrations result in more uniform stress and strain under load. Ann Biomed Eng 31, 12–20. 4. Laboratory, T. J. (1999) http://www.jax.org
322
L.K. Saxon and L.E. Lanyon
5. Turner, C. H., Burr, D. B. (1993) Basic biomechanical measurements of bone: a tutorial. Bone 14, 595–608. 6. De Souza, R. L., Matsuura, M., Eckstein, F., et al. (2005) Non-invasive axial loading of mouse tibiae increases cortical bone formation and modifies trabecular organization: a new model to study cortical and cancellous compartments in a single loaded element. Bone 37, 810–818. 7. Torrance, A., Mosley, J., Suswillo, R., et al. (1994) Non invasive loading of the rat ulna in vivo induces a strain-related modeling response uncomplicated by trauma or periosteal pressure. Calcif Tissue Int 54, 241–247. 8. Fritton, S. P., Rubin, C. T. (2001) In vivo measurement of bone deformation using strain gauges, in (Cowin, S.C., ed.), Bone Mechanics Handbook. CRC Press, Washington, DC. 9. Robling, A. G., Burr, D. B., Turner, C. H. (2001) Recovery periods restore mechanosensitivity to dynamically loaded bone. J Exp Biol 204, 3389–3399. 10. Lee, K. C., Maxwell, A., Lanyon, L. E. (2002) Validation of a technique for studying functional adaptation of the mouse ulna in response to mechanical loading. Bone 31, 407–412. 11. Hagihara, Y., Fukuda, S., Goto, S., et al. (2005) How many days per week should rats undergo running exercise to increase BMD? J Bone Miner Metab 23, 289–294. 12. Judex, S., Lei, X., Han, D., et al. (2006), Low-magnitude mechanical signals that stimulate bone formation in the ovariectomized rat are dependent on the applied frequency but not on the strain magnitude. J Biomech 40, 1333–1339 13. Srinivasan, S., Weimer, D. A., Agans, S. C., et al. (2002) Low-magnitude mechanical loading becomes osteogenic when rest is inserted between each load cycle. J Bone Miner Res 17, 1613–1620. 14. Robling, A. G., Hinant, F. M., Burr, D. B., et al. (2002) Shorter, more frequent mechanical loading sessions enhance bone mass. Med Sci Sports Exerc 34, 196–202. 15. Gross, T. S., Poliachik, S. L., Ausk, B. J., et al. (2004) Why rest stimulates bone formation: a hypothesis based on complex adaptive phenomenon. Exerc Sport Sci Rev 32, 9–13.
Chapter 22
Determination of Bone’s Mechanical Matrix Properties by Nanoindentation Engin Ozcivici, Suzanne Ferreri, Yi-Xian Qin, and Stefan Judex
Abstract Osteoporosis is a devastating disease that is characterized not only by a reduction in bone quantity but also by deterioration in bone quality. The quality of bone tissue is greatly influenced by its mechanical properties and, therefore, investigations into the etiology and enhanced detection of osteoporosis, or the efficacy of interventions, may require the assessment of bone’s mechanical properties at the level of the tissue. Nanoindentation is a relatively new technique that is capable of evaluating bone’s quasi-static and dynamic mechanical properties on extremely small volumes of tissue. These data can be used directly to describe the pre-yield properties of the matrix, but can also be combined with imaging techniques and mechanical models to extrapolate the mechanical properties from the level of the tissue to that of the organ. Keywords Nanoindentation, mechanical properties, bone matrix, stiffness, elastic modulus, hardness, trabecular bone, cortical bone, loading rate.
1
Introduction
The mechanical integrity of the skeleton is directly related to its ability to withstand the loads that are placed on it. Unfortunately, a skeleton at risk of fracture cannot simply be determined by the amount of bone that exists, such as quantified by dual x-ray absorptiometry (1–3). To a large degree, the quality of the bone is just as important. Bone quality comprises all parameters other than bone quantity, but a complete understanding of how quality precisely determines bone’s mechanical behavior is currently absent (4, 5). It is clear, however, that bone quality is greatly influenced by its tissue mechanical properties (6). Material properties offer a direct measure of bone’s load-bearing capacity and may serve as an indicator of the likelihood for future fractures. Thus, the quantification of tissue mechanical properties is critical to a large range of studies in osteoporosis research: from the identification of chromosomal regions that define bone quality or fracture susceptibility, to improved detection of osteoporosis, to the assessment of efficacy of preventive or therapeutic interventions.
From: Methods in Molecular Biology, Vol. 455: Osteoporosis: Methods and Protocols Edited by Jennifer J. Westendorf © Humana Press, Totowa, NJ
323
324
E. Ozcivici et al.
Bone tissue is characterized by a hierarchal structure in its composition and morphology, and material properties can be obtained from measurements at different length scales. Traditionally, bone’s mechanical tissue properties have been derived from mechanical tests at the level of the whole bone or bone sample, including, but certainly not limited to, bending, compression, tension, or torsion tests. The direct results of such tests reflect both the material properties of the bone matrix as well as the morphology of the tissue and the geometric and technical assumptions required for extracting the role of bone geometry can be a source of error (7, 8). To avoid at least some of these assumptions, mechanical tests can be performed on micro-machined cortical and trabecular bone samples with well-defined geometry (9, 10). This approach isolates the tissue response at the millimeter scale; however, it does not capture the heterogeneity in tissue properties at the sub-millimeter scale and can be technically challenging to apply to very small bone samples (11). Nanoindentation is an emerging technology that can overcome these technical limitations by providing mechanical matrix data at very high resolutions of approximately 1 micron (12, 13). It was originally developed for testing of thin films and for applications in the microelectronics industry, but within the last decade, it has been adapted for the characterization of bone’s mechanical tissue properties (14). Nanoindentation uses a very fine diamond tip (typically <1 micron in diameter) to probe the surface of a specimen. The output parameters, including the stiffness and hardness of the matrix, are derived from the elastic response of the indented material to the applied load (Fig. 22.1). The necessity of mechanical measurements at the level of the material becomes further apparent when considering that measurements of the bulk elastic modulus of trabecular bone not only vary by orders of magnitude (0.02–5 GPa), but that bulk trabecular moduli are much smaller than those of cortical bone (~15 GPa) (15). Clearly, bulk mechanical properties are dependent on the amount and architecture of the bone present as well as on the material properties of the trabeculae themselves. Based on bulk measurements, it
2
e rv cu
urv e
3
Pmax = max force
hmax = displacement ing ad at max load Lo
gc
4
Hold period S = stiffness
Un loa din
Force (mN)
5
1 100
hmax 200 300
400
Displacement (nm)
Pmax S
500
Fig. 22.1 A loading–unloading curve recorded during nanoindention of cortical bone from the middiaphysis of a mouse tibia. During loading, the indenter tip penetrates the surface of the sample causing both elastic and plastic deformation. After a hold period, the indenter is withdrawn and the sample is unloaded. The elastic stiffness of the sample (S) is calculated during unloading by a curve-fit to the region between the two open circles. The impression of this indent is visualized in Fig. 22.2.
22 Nanoindentation of Bone
325
has been presumed previously that the material properties of trabecular bone are less than those of cortical bone; however, recent nanoindentation studies suggest that the elastic moduli of trabecular and cortical bone are indeed similar (16, 17). Because of its high resolution, nanoindentation permits the comparison of the stiffness of adjacent lamellae, or the properties of newly added bone relative to preexisting bone (18). Recent studies using nanoindentation have successfully determined the variation in material modulus across individuals and anatomical sites (14, 19), providing insight into the regulation of the spatial distribution of ultrastructural mechanical properties by genetic and environmental factors. Similar to the substantial influence of variations in trabecular thickness on bone strength (20), it is obvious that variations in ultrastructural tissue moduli will influence the overall mechanical properties of bone. Thus, nanoindentation can provide insight into the spatial distributions of material properties, including ultrastructural features, and thus will help to better define how mechanical loads are tolerated by the skeleton. The generation of rigorous spatial material maps that can be incorporated into micro-CT generated high-resolution micro-finite-element models to mimic the mechanical behavior of entire bones or anatomical regions holds great potential. Further, mechanical properties determined by nanoindentation can be used to validate tissue quality measurements from other imaging modalities, including quantitative ultrasound or high-resolution bone densitometry (21).
2 2.1
Materials Specimen Preparation
1. Variable speed diamond wheel saw (e.g., Model 650, South Bay Technology, Inc., San Clemente, CA). 2. High pressure pulsatile water irrigation system (e.g., Waterpik, Water Pik Technologies, Inc., Fort Collins, CO). 3. A series of increasing concentrations of ethanol (70%, 80%, 90%, and 100% EtOH). 4. Resin and hardener (e.g., Epo-thin, Buehler, Lake Bluff, IL). 5. Disposable embedding molds (e.g., Peel Away Molds, Electron Microscopy Sciences, Inc., Hatfield, PA). 6. Vacuum chamber. 7. Vacuum pump capable of achieving 22″ Hg (e.g., High-volume Aspirator Model 130, Schuco, Carle Place, NY). 8. Polishing wheel (e.g., Crystal Master 8, Crystalite Co., Lewis Center, OH). 9. Abrasive silicon carbide papers of decreasing grit size (220, 330, 600, and 1,000 µm). 10. Diamond suspension of decreasing particle size (3, 1, 0.25, and 0.05 µm). 11. Ultrasonic cleaner.
326
E. Ozcivici et al.
12. Metal disk or slide of uniform thickness. 13. Cyanoacrylate glue (e.g., Krazy Glue, Elmer’s Products, Inc., Columbus, OH). 14. Compressed air (e.g., Dust-Off, Falcon Safety Products, Inc., Branchburg, NJ).
2.2
Nanoindentation
1. Nanoindenter System (Triboindenter; Hysitron, Minneapolis, MN) (see Note 1). 2. Diamond Berkovich indenter tip with an included angle of 142.3 degrees and a half angle of 65.35 degrees (see Note 2). 3. Fused quartz calibration sample. 4. Single-crystal aluminum calibration sample.
2.3
Data Analysis
1. Nanoindenter user interface. 2. Statistical software.
3 3.1
Methods Preparation of Samples
1. Section the bone in the plane of interest (see Note 3) using a water lubricated diamond wheel saw. 2. At room temperature, remove as much bone marrow and soft tissues as possible with a pulsatile water irrigation system, taking care not to damage the sample with the high-speed water jet. 3. Dehydrate the bone sections over the course of 8 days in a series of ethanol solutions (70%, 80%, 90%, and 100% EtOH), increasing the ethanol concentration every 2 days (see Note 4). 4. Embed the samples in low viscosity epoxy resin (see Note 5). Be sure to mix the resin and hardener for 2 minutes using a folding motion to minimize the formation of air bubbles. Position the sample at the bottom of the embedding mold and pour the resin/hardener mixture over the sample. When pouring resin into the disposable embedding molds, allow the solution to stream over the end of a disposable wooden stick to further minimize the formation of air bubbles. Then, place the samples in a vacuum chamber for at least 1 hour to minimize the formation of air bubbles and promote bonding of resin to the bone surface. 5. Allow the resin to cure for 24 hours in a vacuum hood at room temperature. 6. Once the resin has cured, remove the disposable mold and polish the surface to be indented using the polishing wheel and abrasive silicon carbide papers of decreasing particle size (220, 330, 600, and 1,000).
22 Nanoindentation of Bone
327
7. Polish the samples further by using a diamond suspension with decreasing particle size (3 µm, 1 µm, and 0.25–0.05 µm) (see Note 6). 8. Clean samples ultrasonically to remove particles and debris that resulted from polishing. 9. Store the embedded bone samples in sealed containers at −20 °C prior to indentation to minimize changes in its material properties (see Note 7). 10. Glue the unpolished surface of the specimen to a metal disk or slide using a cyanoacrylate glue to allow the sample to be secured to the magnetized stage of the nanoindenter. 11. Use compressed air to remove dust and debris from the polished surface immediately before mounting the sample on the moving stage.
3.2
Calibration of the Tip Area Function
1. Perform a line of 100 indents on a calibration sample of fused quartz, increasing the peak load with each indent uniformly over the range of the transducer (e.g., from 2 µN to 9,800 µN.) Fused quartz is used to calibrate of the tip area function because of its low modulus to hardness ratio. Each indent should use a triangular load function with a constant loading/unloading rate. 2. Determine the computed contact area, A, for each indent using the equation: 2
A=
p⎛ S⎞ , 4 ⎜⎝ Er ⎟⎠
where S is the contact stiffness as determined from the initial portion of the unloading curve (usually between 95% and 20%, see Fig. 22.1) and Er is the reduced modulus of the fused quartz (Er = 69.6 GPa). A reduced elastic modulus reflects the combined elastic moduli of the material and the indenter tip. 3. Plot the computed contact area, A(hc), as a function of contact depth, hc, and fit a sixth order polynomial to the curve to obtain: 1
1
1
1
A(hc ) = C0 hc2 + C1hc + C2 hc2 + C3 hc4 + C4 hc8 + C5 hc16 , where Co is a constant equal to 24.5 for a Berkovich tip.
3.3
Quasistatic Nanoindentation
1. These instructions assume the use of hardware available as part of a nanoindenter system manufactured by Hysitron (Minneapolis, MN); however, they may be easily adapted to other systems. Mount the prepared samples on the magnetized stage in accordance with the manufacturer’s safety guidelines. Be sure that all safety limits account for potential inaccuracies in the focal point of the microscope optics as defined by the operator.
328
E. Ozcivici et al.
2. Select indent locations using the optical microscope mounted on the moving stage in parallel with the load sensing transducer. The nanoindenter has an in plane (X-Y) precision of 500 nm, a length scale at which bone is very much inhomogeneous (see Note 8). Thus, careful selection of indent locations is crucial for maintaining data consistency. For example, bone is much stiffer within its cement lines because of higher mineralization levels (22) and pooling data from different ultrastructural regions may lead to large variability in the collected data. 3. Static elastic material properties, such as bone’s elastic modulus and hardness, are determined by applying a trapezoidal load function to the indenter tip (see Note 9). The load function is characterized by constant loading and unloading rates, typically between 100 and 500 µN/s, and a peak load hold period of at least 10 seconds (see Note 10). Since bone is inherently viscoelastic, the peak load hold period serves to dissipate the tissue’s viscoelastic response and minimize the contribution of material plasticity, allowing the initial portion of the unloading curve to be used to calculate the elastic response (Fig. 22.1). Viscoelastic material properties, such as storage and loss modulus, may be calculated using dynamic indentation (see Note 11). 4. Immediately prior to indentation, perform time-sensitive calibrations such as an optics calibration and an electrostatic force constant calibration. Perform an optics calibration by indenting a regular pattern, such as seven indents forming the letter “H,” into a calibration sample of single-crystal aluminum. Then, locate the center indent using the system’s optics. Since the transducer/tip assembly and machine optics are mounted on the same stage, a coordinate map can be defined to relate the X-Y-Z position of the indenter tip to the focal point of the optical microscope. Calibrate the electrostatic force constant by estimating the distance between the drive plate and center electrode. To this end, indent air for a distance large enough to cause a change in the spacing of the two plates and then fit a quadratic curve to the plot of distance and spacing. 5. Perform the indents at the prescribed locations within the bone surface choosing an appropriate maximum indentation depth (see Note 12). Inherently, bone samples with large spatial heterogeneity (e.g., haversian bone) require more indents than more homogenous samples (e.g., lamellar bone). 6. Indents can be visualized with a light microscope or a scanning electron microscope (Fig. 22.2).
3.4
Post Processing
1. Calculate the static elastic material properties by analyzing the unloading portion of the trapezoidal load function (see Note 9). Fit a power law relation of the form: P = b(h − h )m , f
22 Nanoindentation of Bone
329
Fig. 22.2 Scanning electron micrograph of the specific indent characterized in Fig. 22.1. A Berkovich tip was used to generate a peak load of 4.4 mN with a corresponding vertical displacement of 469 nm.
to the unloading curve (typically starting at 5% of the recovered deformation and extending up to 80% of the recovered deformation) to determine the constants b, hf, and m. Then, calculate the contact stiffness, S, by taking the derivative of the power law relation, P, with respect to h and evaluating it at the maximum load, Pmax. 2. Calculate the contact depth, hc, using the following relation: hc = hmax − 0.75
Pmax , S
where hmax is the displacement at the maximum load. 3. Calculate the reduced elastic modulus, Er, using the formula: Er =
S p
, 2 Ac where Ac is the contact area evaluated at the contact depth, hc (see Fig. 22.1). 4. Calculate the sample elastic modulus, E, from the reduced elastic modulus, Er, with the formula: E=
( × (n
) − 1)
Ei × Er × 1 − n 2 Ei × Er
2 i
330
E. Ozcivici et al.
where i and b denote the indenter tip and bone, respectively. υi is Poisson’s Ratio of the diamond tip (0.07), Ei is the elastic modulus of the indenter tip (1140 GPa), and υb is Poisson’s Ratio for bone (0.3). 5. Hardness of the bone, H, is calculated using the formula (see Note 13): P H = max , A(hc ) 6. Perform appropriate descriptive statistics. If the data distribution does not approximate a Gaussian curve, median, rather than mean values, should be used to summarize the mechanical data. 7. Group data can be compared using either parametric or non-parametric statistical tests.
4
Notes
1. This protocol assumes the use of the Triboindenter nanoindentation system developed by Hysitron (Mineapolis, MN). Other producers of similar equipment include MTS Nano Instruments (Oak Ridge, TN) or Micro Materials Ltd. (Wrexham, UK). 2. The Berkovich tip is the most commonly used tip for determining bone’s mechanical matrix properties. It is characterized by an included angle of 142.3 degrees, a half angle of 65.35 degrees, and a radius of curvature between 100 and 200 nm. The relatively obtuse geometry of this tip allows probing of the sample at greater depths. A cube corner tip has an included angle of 90 degrees, a much smaller radius of curvature (40–100 nm) and can be used to probe at very shallow depths or to achieve a higher resolution for spatial modulus mapping. Cono-spherical tips have a much greater tip radius (~3 µm) and are often used to indent softer materials, including cartilage and for scratch testing that can generate maps of mechanical material properties over a larger area. 3. Nanoindentation measures the response of the material in the direction of the applied load. Material properties in other directions may be determined by sectioning the bone in such a manner that the surface of the sample is perpendicular to the desired measurement axis. Measurements in multiple direction facilitate the definition of anisotropic constitutive equations (23). 4. Bone samples are routinely dehydrated prior to embedding to enhance bonding and infiltration of the bone with the embedding material (14). Dehydration and embedding of the sample in resin are important steps particularly for the indentation of trabecular bone. However, dehydration can increase bone’s elastic modulus by as much as 55% (24, 25). Rehydration of the sample after polishing may result in the measurement of intrinsic material properties that are similar to those of fresh and fully hydrated tissue (24).
22 Nanoindentation of Bone
331
5. It is important to choose an embedding material appropriate for the specific bone under investigation. Factors such as the type of bone (e.g., lamellar vs haversian), location (e.g., cortical vs trabecular), or intrinsic stiffness (e.g., fracture callus vs. pre-existing bone) may need to be considered in particular when probing the hardness of the tissue. For instance, low viscosity resin may be more suitable for trabecular bone because air bubble formation can be minimized and bonding with the trabecular surface can be enhanced. Measurements of bone hardness, but not of the elastic modulus, are sensitive to the modulus and the viscosity of the embedding material (26). Given that the nature of the embedding material may affect the mechanical measurements, its properties should be specified in publications to facilitate comparisons to the literature. 6. A diamond suspension particle size of 0.05 µm is necessary to ensure that the sample’s mean surface roughness is smaller than the maximum indentation depth. 7. Material properties can be affected by the duration of tissue storage between embedding and nanoindentation. After 6 months of storage, the elastic modulus of the embedded tissue may increase significantly, whereas the hardness appears to be less affected by increased storage time (26). It is suggested to indent the bone samples upon embedding without significant delay. 8. Mechanical matrix properties obtained from nanoindentation can be associated with other physical or chemical matrix properties (18) to identify specific factors that define bone’s mechanical behavior at the level of the matrix. 9. During the loading phase of the trapezoidal load function, the indenter tip penetrates the matrix of the sample, thereby inducing both elastic (recoverable) and plastic (non-recoverable) deformation in the material. Thus, the slope of the loading curve cannot be used to describe bone’s elastic properties (see Fig. 22.1). Upon withdrawal of the force from the sample, however, only the elastic deformation is recovered and the slope of the unloading is directly related to the elastic stiffness of the material. 10. It is important to note that although some studies have suggested that both the magnitude of the applied maximum load as well as the applied loading rate may affect the determination of bone’s material properties (13, 24), others were not able to confirm this (26). Discrepancies between these studies may be related to the differences in species, preparation techniques, or loading regimens. 11. The Oliver-Pharr method is useful in describing bone’s elastic material properties, but is not suitable to capture bone’s dynamic response to loading. At the macroscopic level, bone is viscoelastic; therefore, bone’s elastic behavior exhibits strain rate dependency (27). At the microscopic scale, nanoindentation studies have indicated a similar relation (13, 28). To assess whether the elastic modulus of the sample is dependent upon the applied strain rate, a series of consecutive trapezoidal loading/unloading protocols with identical peak loads can be performed on the same spot. A power law relation between the elastic modulus and strain rate indicates strain rate dependency (13), and bone’s mechanical matrix properties can be further described by including its dynamic response. The complex modulus is equal to the sum of the storage modulus and the loss modulus and incorporates both bone’s static and dynamic response to
332
E. Ozcivici et al.
loading. The storage modulus is closely related to the mineral component of bone, whereas the loss modulus is indicative of bone’s damping characteristics primarily modulated by collagen fibers (29). Using the model proposed by Loubet et al. (30), the storage modulus, E', and loss modulus, E″, can be determined by solving the following system of equations: Xo =
Fo ( K s + K i − Mw 2 )2 + [(Ci + Cs )w ]2
tan(f ) =
(Ci + Cs )w K s + K i − Mw 2 )
,
,
E ″ ⎣w Cs (1+ n ) p ⎦/ A(hc ) = , E′ ⎣ K s (1+ n ) p ⎦/ A(hc ) where XO is the displacement amplitude, FO is the displacement force, Ki and KS are the stiffness’ of the indenter shaft springs and tip-sample contact, M is the mass of the indenter, ω is the excitation frequency of the tip, Ci and CS are the damping coefficients of the displacement-sensing capacitor and sample, ϕ is the phase shift, and A(hc) is the contact area evaluated at the contact depth hc as described. Most nanoindentation studies in bone have focused on bone’s static elastic properties but dynamic nanoindentation has recently been used to determine the complex modulus of bone by superimposing small amplitude (~3–5 nm) oscillations on a quasi-static indentation test at a frequency between 10 and 200 Hz (31). 12. Nanoindentation has been used to measure bone’s material properties in a wide range to species (e.g., rodents, human, or cows) with greatly different types of bone or the size of structural and ultrastructural features. Perhaps not surprisingly, the choice of maximum indentation depth varies greatly within the literature. There is currently no consensus on the “optimal” indentation depth. Generally, bone’s material properties demonstrate depth dependence at depths of <500 nm (24), whereas greater indentation depths may produce less data variability. The mechanisms underlying these observations have not been fully investigated. Indentation at smaller depths may offer increased spatial sensitivity, but it is also possible that the greater data variability is associated with greater error susceptibility. It is clear, however, that the chosen maximum indentation depth should be kept consistent to allow for inter-group and interstudy comparison and may ultimately reflect the nature of the samples and the study design. 13. Hardness in plastic materials, such as metals, is closely related to its yield properties, however in bone, tissue hardness is related both to its elastic properties and its yield properties (24, 32). Therefore, relative differences in bone’s elastic modulus between groups are in most studies mirrored by similar differences in bone’s hardness values.
22 Nanoindentation of Bone
333
Acknowledgments The authors thank Jim Quinn for assistance with acquiring the SEM image. Financial support by NASA NAG 9-1499 (S.J.), the Wallace Coulter Foundation (S.J.), the Whitaker Foundation RG-02-0564 (S.J.), the National Space Biomedical Research Institute (TD00207 & TD00405) through NASA Cooperative Agreement NCC 9-58 (Y.Q.), NIH (R01 AR49286 Y.Q, R01 AR49694) Y.Q, and R01 AR052778, S.J), and the US Army Medical Research and Material Command (DAMD-17-02-1-0218) (Y.Q.) is gratefully acknowledged.
References 1. Hans, D., Fuerst, T., Lang, T., et al. (1997) How can we measure bone quality? Baillieres Clin Rheumatol 11, 495–515. 2. Riggs, B. L., Hodgson, S. F., O’Fallon, W. M., et al. (1990) Effect of fluoride treatment on the fracture rate in postmenopausal women with osteoporosis. NEJM 322, 802–809. 3. Mashiba, T., Hirano, T., Turner, C. H., et al. (2000) Suppressed bone turnover by bisphosphonates increases microdamage accumulation and reduces some biomechanical properties in dog rib. J Bone Miner Res 15, 613–620. 4. Judex, S., Boyd, S. K., Qin, Y. X., et al. (2003) Combining high-resolution microct with material composition to define the quality of bone tissue. Curr Osteoporosis Repts 1, 11–19. 5. Mittra, E., Rubin, C., Qin, Y. X. (2005) Interrelationship of trabecular mechanical and microstructural properties in sheep trabecular bone. J Biomech 38, 1229–1237. 6. Judex, S., Boyd, S., Qin, Y. X., et al. (2003) Adaptations of trabecular bone to low magnitude vibrations result in more uniform stress and strain under load. Ann Biomed Eng 31, 12–20. 7. Keaveny, T. M., Borchers, R. E., Gibson, L. J., et al. (1993) Trabecular bone modulus and strength can depend on specimen geometry. J Biomech 26, 991–1000. 8. Silva, M. J., Brodt, M. D., Fan, Z., et al. (2004) Nanoindentation and whole-bone bending estimates of material properties in bones from the senescence accelerated mouse SAMP6. J Biomech 37, 1639–1646. 9. Wang, X., Shanbhag, A. S., Rubash, H. E., et al. (1999) Short-term effects of bisphosphonates on the biomechanical properties of canine bone. J Biomed Mater Res 44, 456–460. 10. Reilly, G. C., Currey, J. D. (2000) The effects of damage and microcracking on the impact strength of bone. J Biomech 33, 337–343. 11. Tanck, E., Van Donkelaar, C. C., Jepsen, K. J., et al. (2004) The mechanical consequences of mineralization in embryonic bone. Bone 35, 186–190. 12. Oliver, W. C., Pharr, G. M. (1992) An improved technique for determining hardness and elastic modulus using load and displacement sensing indentation experiments. J Mater Res 7, 1564–1583. 13. Fan, Z., Rho, J. Y. (2003) Effects of viscoelasticity and time-dependent plasticity on nanoindentation measurements of human cortical bone. J Biomed Mater Res A 67, 208–214. 14. Rho, J. Y., Tsui, T. Y., Pharr, G. M. (1997) Elastic properties of human cortical and trabecular lamellar bone measured by nanoindentation. Biomaterials 18, 1325–1330. 15. Keaveny, T. M., Hayes, W. C. (1993) A 20-year perspective on the mechanical properties of trabecular bone. J Biomech Eng 115, 534–542. 16. Roy, M. E., Rho, J. Y., Tsui, T. Y., et al. (1999) Mechanical and morphological variation of the human lumbar vertebral cortical and trabecular bone. J Biomed Mater Res 44, 191–197. 17. Turner, C. H., Rho, J., Takano, Y., et al. (1999) The elastic properties of trabecular and cortical bone tissues are similar: results from two microscopic measurement techniques. J Biomech 32, 437–441. 18. Busa, B., Miller, L. M., Rubin, C. T., et al. (2005) Rapid establishment of chemical and mechanical properties during lamellar bone formation. Calcif Tissue Int 77, 386–394. 19. Zysset, P. K., Guo, X. E., Hoffler, C. E., et al. (1999) Elastic modulus and hardness of cortical and trabecular bone lamellae measured by nanoindentation in the human femur. J Biomech 32, 1005–1012.
334
E. Ozcivici et al.
20. Iwamoto, J., Yeh, J. K., Aloia, J. F. (1999) Differential effect of treadmill exercise on three cancellous bone sites in the young growing rat. Bone 24, 163–169. 21. Xia, Y., Lin, W., Qin, Y. X. (2005) The influence of cortical end-plate on broadband ultrasound attenuation measurements at the human calcaneus using scanning confocal ultrasound. J Acoust Soc Am 118, 1801–1807. 22. Rho, J. Y., Zioupos, P., Currey, J. D., et al. (1999) Variations in the individual thick lamellar properties within osteons by nanoindentation. Bone 25, 295–300. 23. Fan, Z., Swadener, J. G., Rho, J. Y., et al. (2002) Anisotropic properties of human tibial cortical bone as measured by nanoindentation. J Orthop Res 20, 806–810. 24. Hoffler, C. E., Guo, X. E., Zysset, P. K., et al. (2005) An application of nanoindentation technique to measure bone tissue Lamellae properties. J Biomech Eng 127, 1046–1053. 25. Hengsberger, S., Kulik, A., Zysset, P. (2002) Nanoindentation discriminates the elastic properties of individual human bone lamellae under dry and physiological conditions. Bone 30, 178–184. 26. Mittra, E., Akella, S., Qin, Y. X. (2006) The effects of embedding material, loading rate and magnitude, and penetration depth in nanoindentation of trabecular bone. J Biomed Mater Res A 79, 86–93. 27. Carter, D. R., Hayes, W. C. (1976) Bone compressive strength: the influence of density and strain rate. Science 194, 1174–1176. 28. Sasaki, N., Nakayama, Y., Yoshikawa, M., et al. (1993) Stress relaxation function of bone and bone collagen. J Biomech 26, 1369–1376. 29. Yamashita, J., Li, X., Furman, B. R., et al. (2002) Collagen and bone viscoelasticity: a dynamic mechanical analysis. J Biomed Mater Res 63, 31–36. 30. Loubet, J. L. (1996) Some measurements of viscoelastic properties with the help of nanoindentation. NIST Spec Pub 31–34. 31. Donnelly, E., Baker, S. P., Boskey, A. L., et al. (2006) Effects of surface roughness and maximum load on the mechanical properties of cancellous bone measured by nanoindentation. J Biomed Mater Res A 77, 426–435. 32. Currey, J. D. (1990) Physical characteristics affecting the tensile failure properties of compact bone. J Biomech 23, 837–844.
Chapter 23
Fluid Flow Assays Ryan C. Riddle, Amanda F. Taylor, and Henry J. Donahue
Abstract Mechanical signals are major regulators of skeletal homeostasis as the addition of exogenous load is followed by enhanced bone formation and the removal of normal loads is followed by net bone loss. The mechanism by which bone cells perceive and respond to changes in their biophysical environment are still poorly understood, but it is widely accepted that the detection of interstitial fluid flow is an initiating cue. In this chapter, we describe two in vitro systems designed to examine the effects of fluid flow on bone cell behavior and to elucidate the signaling cascades activated by this stimulus. The first utilizes a parallel plate flow chamber designed to stimulate a single bone cell type grown on glass slides. The second employs a rotating disk fluid flow apparatus. Commerciallyavailable cell culture inserts allow one type of bone cell to be exposed to fluid flow and signals to be communicated to a second bone cell model not exposed to fluid flow. Keywords Interstitial fluid flow, parallel plate flow chamber, rotating disk flow apparatus, osteoblasts, osteocytes, mesenchymal stem cells.
1
Introduction
Adaptation of skeletal mass and architecture as a result of changes in osteoblastic bone formation and osteoclastic bone resorption is a well-established response of bone to mechanical loading (1, 2). Indeed, the sensitivity of bone cells to changes in their biophysical environment is such that only a single, short exposure to a dynamic load is necessary to activate bone-forming osteoblasts and increase bone formation (3). If the ability of bone cells to perceive changes in their biophysical environment is diminished, the resultant decrease in bone formation could be instrumental in the initiation of osteopenia. Thus, a comprehensive understanding of the cellular mechanisms by which biophysical signals regulate bone cell activity will lead to the development of novel treatments for osteopenia.
From: Methods in Molecular Biology, Vol. 455: Osteoporosis: Methods and Protocols Edited by Jennifer J. Westendorf © Humana Press, Totowa, NJ
335
336
R.C. Riddle et al.
Substrate strains, resulting from the deformation of skeletal tissue, are possible candidates for the biophysical signals perceived by bone cells and physical activity has been shown to induce strains on the order of 500–2,500 microstrain (µε) (4, 5). However, in vitro studies suggest that bone cells do not respond to mechanical strains within this range. For example, exposing MC3T3-E1 osteoblasts to substrate strains <5,000 µε failed to enhance the expression of osteopontin mRNA or induce the mobilization of intracellular calcium (6, 7). Rather, studies suggest that strain magnitudes 5- to 10-fold higher than those observed in vivo are necessary to induce the release of autocrine/paracrine factors and alter the transcription of genes associated with bone formation (8, 9). Emerging data suggest that bone cells may instead act as fluid sensors and respond to mechanical loading indirectly through the detection of interstitial fluid flow. The generation of substrate strains by mechanical loads induces the pressurization of interstitial fluid and stimulates its movement along pressure gradients, from the matrix into haversian systems (10). Due to the cyclical nature of physiological loading (ambulation), interstitial fluid flow is oscillatory in nature. In a normal gait cycle, loading, which drives fluid out of the matrix, is followed by a period of relaxation, thus reversing the pressure gradient and allowing interstitial fluid to flow back into the matrix. Within the confined geometry of bones’ lacunar-canalicular and haversian systems, fluid flow would impart shear stresses upon the cell membrane. Cellular shear stresses experienced were originally estimated to be up to 30 dynes/cm2 (11, 12) but are now predicted to range from 0 to 20 dynes/cm2 depending on the cellular location (13, 14). In vitro studies indicate that these predicted levels of fluid flow shear stress are potent regulators of bone cell behavior. Osteocytes, osteoblasts, and even mesenchymal stem cells respond to the levels of fluid flow shear stress, predicted to occur as a result of physiological loading, with increased production and release of a number of autocrine/paracrine factors, including prostaglandin E2 (15–17), ATP (18, 19), and nitric oxide (16, 20, 21). Importantly, these signaling molecules are crucial for the anabolic response of bone to mechanical load in vivo. For instance, pharmacological inhibition of prostaglandin production or genetic ablation of purinergic receptor expression abolishes load-induced bone formation (22–24). Downstream of these factors, fluid flow enhances the expression of several phenotypic markers of osteoblastic cells, including osteopontin (7, 19, 25), type I collagen (26), and osteocalcin (27, 28), and enhances cellular proliferation (27, 29–31). This chapter describes two in vitro systems designed to examine the effect of fluid flow on bone cell behavior. The first employs a parallel plate flow chamber designed to use a single cell type. The second utilizes a novel osteocyte-osteoblast co-culture model that mimics in vivo systems by permitting osteocytes to be exposed to physiological levels of fluid flow while osteoblasts are shielded from it. Both systems are amenable to examining changes in cellular proliferation, gene expression, protein kinase activation, and many other assays.
23 Fluid Flow Assays
2 2.1
337
Materials Cell Culture
1. Osteoblastic cell lines (MC3T3-E1/hFOB 1.19) maintained in Minimum Essential Medium alpha medium (α-MEM) (MC3T3-E1) or Dulbecco’s Modified Eagles Medium: F12 (hFOB 1.19) supplemented with 10% fetal bovine serum (FBS) (Hyclone) and 1% penicillin/streptomycin. 2. Osteocytic MLO-Y4 cells maintained on collagen type 1-coated tissue culture plastic (150 mg/mL, BD Biosciences, CA) in α-MEM supplemented with 2.5% FBS (BioWhittaker, Inc.), 2.5% calf serum (Hyclone), and 1% penicillin/streptomycin (32). 3. Primary human mesenchymal stem cells (Cambrex) maintained in Dulbecco’s Modified Eagles Medium supplemented with 10% FBS (Hyclone), 1% L-glutamine, and 1% penicillin/streptomycin (see Note 1) 4. Trypsin: 0.25% Trypsin, 1 mM ethylenediamine tetraacetic acid (EDTA). 5. Phosphate-buffered saline (PBS). 6. 70% ethanol. 7. Collagen type I. 8. 0.02 N acetic acid.
2.2
Parallel Plate Fluid Flow Apparatus
1. Parallel plate flow chambers consist of custom-machined polycarbonate plates based upon a design modified from the description of Frangos et al. (33). The base of the chamber contains a 34 × 74 × 0.28 mm channel which when covered with a 75 × 38 × 1.0 mm glass slide forms the flow channel. Inlet and outlet ports at the end the channel are fitted with female leurs and one-way stopcocks. A silastic gasket bridges the distance between the glass slide and the polycarbonate lid that is secured by Teflon thumbscrews (Fig. 23.1, see Note 2). 2. A small-scale, servopneumatic material testing device (EnduraTec, Eden Prairie, MN) is used to deliver an oscillatory flow pattern. Glass Hamilton syringes (Fisher Scientific) are connected to the actuator of the material testing device and held in place by a custom-designed syringe holder attached to the lower fixtures of the testing device. Computer-control of the materials testing device (QuikTest Software, EnduraTec) can be used to generate a number of waveforms, but a sine wave with a 1-Hz frequency is typically used (Fig. 23.2, see Note 3). 3. Rigid wall tubing (3.1 mm internal diameter) is used to connect the Hamilton syringes to the flow chamber. 4. An ultrasonic flow probe (Transonic systems, Ithaca, NY) is connected in series to monitor the flow rate in real-time.
338
R.C. Riddle et al.
Fig. 23.1 Parallel plate flow chamber. A 75 × 38 × 1.0 mm glass slide positioned in a polycarbonate chamber (1) and secured with Teflon thumb screws (2). One-way stopcocks (3) are connected to the inflow and outflow ports of the chamber. Rigid wall tubing (4), filled with media via a 20-mL syringe, connects the chamber to a glass Hamilton syringe (5).
Fig. 23.2 Oscillatory fluid flow rig. A computer-controlled servo-pneumatic materials testing device (a,b, EnduraTec) is used to drive flow media through the parallel plate chamber. Plungers of glass Hamilton syringes are connected to the upper fixture of the testing devices, while the syringes are connected to the lower fixture (c), tubing extends from the syringes into a 37 °C incubator that houses the flow chambers for the duration of the experiment.
5. 75 × 38 × 1.0 mm glass microscope slides on which cell monolayers are cultured. 6. 37 °C incubator to house the fluid flow chamber during experiments. 7. Neutral dextran.
2.3
Rotating Disk Fluid Flow Apparatus
1. 1 µm perforated polyethylene terephthalate (PET) membrane cell culture inserts and companion six-well plates (BD Biosciences, CA) seeded with osteoblastic and osteocytic cell models (34).
23 Fluid Flow Assays
339
2. The rotating disk apparatus was designed after Sill et al. (35). Custom-built, precision-machined Lexan cell culture chambers contain a precision-machined Teflon holder for the cell culture insert and a custom silicon gasket (Sylgard 184, Dow Corning, MI) (Fig. 23.3). 3. The rotational apparatus consists of a cylindrical stainless steel disk attached to a drive shaft. The cylindrical stainless steel disk is attached to a variable positioning device that enables the disk to be moved to a predetermined position within the cell chamber. 4. The control apparatus consists of a stationary cylindrical stainless steel disk attached to a variable horizontal positioning device that enables the disk to be moved to a predetermined position within the cell chamber. 5. Servo motor and step driver/controller (Exonic Systems, CA). 6. Custom motor mount enabling two identical rotational apparatuses to be connected in series via a drive belt to the motor. 7. Toothed belt and pulley connectors. 8. Retort stands and bosses to hold static control chambers. 9. Flow medium: Dulbecco’s Modified Eagle’s Medium: F12, 2% FBS, 1% penicillin/streptomycin.
3
Methods
3.1
Parallel Plate Fluid Flow
3.1.1
Cell Culture
1. Prepare rectangular glass slides (75 × 38 × 1.0 mm) for cell seeding by soaking in 70% ethanol and then air-drying under ultraviolet illumination in 100-mm cell
Fig. 23.3 Rotating disk fluid flow apparatus. Cell seeded inserts are loaded into Lexan cell chambers (1), and butted against the rotational apparatus (2) and attached via a compression ring (3). The rotational apparatuses are attached in series to the motor (4), which is positioned on a custom motor mount (5) and driven by a servo motor (6).
340
R.C. Riddle et al.
culture dishes in a laminar flow hood. For some cell types, such as MLO-Y4 osteocytes, it may be necessary to coat slides with collagen type I. This can be accomplished by coating slides with 300 µg/mL collagen type I in 0.02 N sterile acetic acid for 1 hour. Remove excess collagen and acetic acid by washing slides with PBS. 2. Remove spent growth medium from 75% to 85% confluent cultures of cells grown in 100-mm cell culture dishes and wash briefly with PBS. Collect cells by adding 1 mL of trypsin to the dish and incubating at 37 °C for 5–10 minutes. Once the cells have detached from the dish, add 9 mL of medium to the plate and transfer cells to a 15-mL falcon tube and centrifuge for 5 minutes. 3. Aspirate growth medium from pelleted cells and resuspend in 3–4 mL of growth medium. Count the cell number with a hemocytometer and further dilute cells to a concentration of 6.5 × 104 cells/mL. Seed each slide with 1.3 × 105 cells by overlaying 2 mL of the cell suspension on each slide and incubating at room temperature for 30–45 minutes. After the cells have attached to the slide, add an additional 10 mL of medium to the dish and culture 48–72 hours until cells are approximately 85% confluent. 4. Serum starve the cells 24 hours prior to the initiation of the fluid flow experiment by replacing the spent growth medium on the slides with flow medium containing 0.5–2% FBS.
3.1.2
Experimental Set-Up
1. If the experimental endpoint is to be assessed >8 hours after the completion of fluid flow exposure, then it is necessary to work aseptically and sterilize the parallel plate flow chambers, flow tubing, Hamilton syringes, silastic gaskets, and thumb screws prior to the set-up of flow experiments. Soak equipment in 70% ethanol (v/v) for 1–2 hours and then air-dry in a laminar flow hood. 2. Fill 20- or 60-mL syringes with flow medium warmed to 37 °C and connect to the flow tubing. Carefully fill the tubing and associated Hamilton syringes with medium. Be sure to push air bubbles out of the flow tubing, as they will disrupt the cell monolayer and flow profile. 3. Connect the flow tubing to the flow chamber via the one-way stopcocks at the inlet and outlet ports at each end of the chamber. After pushing approximately 5 mL of medium in to the flow channel, invert a glass slide seeded with a cell mono-layer and carefully lower into the channel. Placing one end near the outlet port and lowering the other end toward the inlet port helps to avoid air bubbles being captured in the flow chamber. 4. Position a silastic gasket on the back of the glass slide and cover with the lid of the flow chamber. Secure the lid of the chamber with Teflon thumbscrews, avoiding unnecessary perturbations to the flow chamber that might artificially activate experimental endpoints (see Fig. 23.1). 5. Transport the flow chamber to a 37 °C incubator and connect the Hamilton syringes to the materials testing device (see Fig. 23.2). For kinetic studies, it is
23 Fluid Flow Assays
341
often beneficial to incubate the cell 30–45 minutes in the chamber prior to the initiation of fluid flow to eliminate the effects of chamber set-up. 6. Static, control chambers should be set-up in the same manner. 7. In this experimental set-up, the flow rate and shear stress have a linear relationship governed by the equation τ = 6 µQ/bh2 (Fig. 23.4). τ is the wall shear stress (dyn/cm2), Q is the flow rate (cm3/s), µ is the viscosity of the flow medium (0.01 dyn·s·cm−2), and b and h are the width and height of the channel, respectively (33). The flow rate, monitored in real-time, can be varied by adjusting the amplitude of the waveform generated by the materials testing device. Alternatively, neutral dextran can be added to the flow medium to alter the viscosity (36, 37). 8. Cell monolayers can be exposed to fluid flow for varying lengths of time depending upon the experimental endpoint. However, due to the length of time necessary to disassemble the chamber and recover the slide, it is difficult to obtain samples for time points <1 minute (see Note 5).
3.2
Rotating Disk Fluid Flow
3.2.1
Cell Culture
1. Trypsinize, count, and adjust osteoblastic hFOB 1.19 cells to 8 × 104 cells/mL as described in Section 3.1.1. 2. Invert cell culture inserts onto a sterile flat-bottomed glass tray and seed the basal side of the membrane with 4 × 104 cells in 500 µL medium. Cover the tray to maintain sterility and incubate for 2 hours at 37 °C to permit cellular adhesion. 3. Trypsinize, count and adjust MLO-Y4 osteocytes to 1 × 104 cells/mL as described in Section 3.1.1. 4. Revert inserts into 6 well tissue culture plates with 2 mL hFOB 1.19 medium in the lower reservoir. Apply 2 × 104 MLO-Y4 osteocytes in 2 mL hFOB 1.19 medium to the apical side of the membrane. 5. Incubate co-cultures for 72 hours at 37 °C with 5% CO2 in air (v/v).
3.2.2
Experimental Set-Up
1. Sterilize cell chambers; insert holders, silicon gaskets, and rotational and control apparatus in 70% ethanol (v/v); and air-dry, as in Section 3.1.2.1. 2. Fix the rotation apparatus within the custom motor mount and connect in series to the motor via a toothed belt and pulleys. 3. Clamp control apparatus in retort stands. 4. Fill the lower media reservoir of the Lexan cell chamber with 40 mL of flow medium.
342
R.C. Riddle et al.
5. Position cell culture inserts into the Teflon insert holder, and apply a silicon gasket. Set this complete unit inside the Lexan cell chamber and fill the upper media reservoir with flow medium. 6. Abut the Lexan cell chamber against the stop plate on the rotation apparatus and tighten the compression ring, causing the expansion of the silicon gasket. This acts to separate the lower media reservoir from the upper media reservoir and orientates the insert membrane to a predetermined horizontal plane within the cell chamber (Fig. 23.3). 7. Using the variable positioning device, lower the disk to reside 500 µm above the insert membrane/MLO-Y4 osteocytes. 8. Set-up and attach control cell chambers to control system as described in Sections 3.2.2.4 to 3.2.2.7. 9. To initiate flow, the motor is set to a predetermined speed, which via the toothed belt and pulley system rotates the drive shaft of the attached disk at a rotational frequency of 26.7 rad/second. In this experimental set-up shear stresses imparted onto cells cultured on the insert increases radially, such that cells at the center are exposed to 0 dynes/cm2. Those at the edge experience maximum shear stress (tmax) where the average shear stress experience by cultured cells is 2/3 tmax. tmax is derived using the equation tmax = (m·ω·r)/h where the medium’s coefficient of friction (µ, 0.01 dyn·s·cm−2) and radius of disk (r, cm) are held constant, and height of disk from cells (h, cm) and rotational frequency (ω, rad/minutes) can be varied (35) (Fig. 23.4, see Note 4). 10. Expose co-cultures to fluid flow with a tmax of 4.4 dynes/cm2 for 1 hour at room temperature. 11. Following exposure to fluid shear or control conditions cells are either harvested by trypsinization, media samples collected or if required co-cultures are placed in 6 well plates with fresh medium and post-incubated at 37 °C with 5% (v/v) CO2 in air for the desired time (see Note 5).
Parallel Plate Q
Rotating Disc h
τ = 6µQ bh2 τ = Shear stress µ = Media coefficient of friction Q = Flow rate b = Flow channel width h = Flow channel height
w h
r
τ max = µ*ω*r h
τmax = Maximal rotational shear stress ω = Rotational frequency (radians / min) µ = Media coefficient of friction r = Radius of the disc h = Height of the disc above cells
Fig. 23.4 Comparison of the calculations utilized to determine fluid shear stress in the parallel plate flow chamber and the rotating disk flow apparatus. Values for each measure are given in the text.
23 Fluid Flow Assays
4
343
Notes
1. These bone cell types represent only a partial list of those that can be used in fluid flow experiments. Many other cell models, including ROS 17/2.8, MG-63, and primary mouse, rat, and chicken osteoblasts, have also been successfully employed in these studies. Indeed, virtually any adherent cell can be examined in these devices. 2. An advantage of this design over designs in which the flow channel is formed by a silastic gasket and the chamber is sealed via a vacuum channel, is that cells are not compressed between the gasket and the slide. Since the slide serves as the base of the flow channel, the entire cell population is exposed to laminar flow and can be used in biochemical assays, reducing contamination from damaged cells. 3. The description of this system is for the delivery of an oscillatory fluid flow regime, as we believe this to be more physiologically relevant. Other laboratories and a number of studies have utilized steady or pulsatile fluid flow regimes. Our system can be modified to deliver a pulsatile flow by adding a steady component as described by Jacobs et al. (38). Briefly, a Harvard syringe pump is connected via a Y-connector to the flow tubing prior to the inlet valve of the flow chamber. Alternative fluid flow systems for the delivery of steady and pulsatile flow have been described by Stevens and Frangos (39). 4. If a uniform value of shear stress is required across the surface of the cell culture insert, rather than the radially increasing one described here, the system can be modified by replacing the disk with a cone. 5. Each of these fluid flow models has its distinct benefits and limitations. The parallel plate flow chamber typically utilizes only a single cell type, whereas bone is composed of numerous cell types, including lining cells, osteoblasts, and osteocytes. Each of these cell types may individually be found to be responsive to fluid flow, but the interplay of multiple cell types is difficult to examine in this system. The rotating disk apparatus allows the inclusion of multiple bone cell types and more accurately models bone morphology. However, in order to prevent bowing of the membrane in response to fluid flow initiation, a small size must be maintained. As a result sample volumes are inherently small and can be used on a limited number of parallel assays, or in some cases must be pooled. Additionally, the closed design of this system prevents the measurement of response to fluid flow in real-time, such as intracellular calcium changes, that are possible in the parallel plate system.
References 1. Hillam, R. A., Skerry, T. M. (1995) Inhibition of bone resorption and stimulation of formation by mechanical loading of the modeling rat ulna in vivo. J Bone Miner Res 10, 683–689. 2. Rubin, C. T., Lanyon, L. E. (1985) Regulation of bone mass by mechanical strain magnitude. Calcif Tissue Int 37, 411–417.
344
R.C. Riddle et al.
3. Pead, M. J., Skerry, T. M., Lanyon, L. E. (1988) Direct transformation from quiescence to bone formation in the adult periosteum following a single brief period of bone loading. J Bone Miner Res 3, 647–656. 4. Burr, D. B., Milgrom, C., Fyhrie, D., et al. (1996) In vivo measurement of human tibial strains during vigorous activity. Bone 18, 405–410. 5. Rubin, C. T., Lanyon, L. E. (1982) Limb mechanics as a function of speed and gait: a study of functional strains in the radius and tibia of horse and dog. J Exp Biol 101, 187–211. 6. You, J., Yellowley, C. E., Donahue, H. J., et al. (2000) Substrate deformation levels associated with routine physical activity are less stimulatory to bone cells relative to loading-induced oscillatory fluid flow. J Biomech Eng 122, 387–393. 7. Owan, I., Burr, D. B., Turner, C. H., et al. (1997) Mechanotransduction in bone: osteoblasts are more responsive to fluid forces than mechanical strain. Am J Physiol 273, C810–815. 8. Harter, L. V., Hruska, K. A., Duncan, R. L. (1995) Human osteoblast-like cells respond to mechanical strain with increased bone matrix protein production independent of hormonal regulation. Endocrinology 136, 528–535. 9. Buckley, M. J., Banes, A. J., Levin, L. G., et al. (1988) Osteoblasts increase their rate of division and align in response to cyclic, mechanical tension in vitro. Bone Miner 4, 225–236. 10. Piekarski, K., Munro, M. (1977) Transport mechanism operating between blood supply and osteocytes in long bones. Nature 269, 80–82. 11. Weinbaum, S., Cowin, S. C., Zeng, Y. (1994) A model for the excitation of osteocytes by mechanical loading-induced bone fluid shear stresses. J Biomech 27, 339–360. 12. Kufahl, R. H., Saha, S. (1990) A theoretical model for stress-generated fluid flow in the canaliculi-lacunae network in bone tissue. J Biomech 23, 171–180. 13. Mi, L. Y., Basu, M., Fritton, S. P., et al. (2005) Analysis of avian bone response to mechanical loading. Part two: Development of a computational connected cellular network to study bone intercellular communication. Biomech Model Mechanobiol 4, 132–146. 14. Mi, L. Y., Fritton, S. P., Basu, M., et al. (2005) Analysis of avian bone response to mechanical loading, Part one: Distribution of bone fluid shear stress induced by bending and axial loading. Biomech Model Mechanobiol 4, 118–131. 15. Saunders, M. M., You, J., Zhou, Z., et al. (2003) Fluid flow-induced prostaglandin E2 response of osteoblastic ROS 17/2.8 cells is gap junction-mediated and independent of cytosolic calcium. Bone 32, 350–356. 16. Bakker, A. D., Soejima, K., Klein-Nulend, J., et al. (2001) The production of nitric oxide and prostaglandin E(2) by primary bone cells is shear stress dependent. J Biomech 34, 671–677. 17. Klein-Nulend, J., Burger, E. H., Semeins, C. M., et al. (1997) Pulsating fluid flow stimulates prostaglandin release and inducible prostaglandin G/H synthase mRNA expression in primary mouse bone cells. J Bone Miner Res 12, 45–51. 18. Genetos, D. C., Geist, D. J., Liu, D., et al. (2005) Fluid shear-induced ATP secretion mediates prostaglandin release in MC3T3-E1 osteoblasts. J Bone Miner Res 20, 41–49. 19. You, J., Reilly, G. C., Zhen, X., et al. (2001) Osteopontin gene regulation by oscillatory fluid flow via intracellular calcium mobilization and activation of mitogen-activated protein kinase in MC3T3-E1 osteoblasts. J Biol Chem 276, 13365–13371. 20. McAllister, T. N., Frangos, J. A. (1999) Steady and transient fluid shear stress stimulate NO release in osteoblasts through distinct biochemical pathways. J Bone Miner Res 14, 930–936. 21. Klein-Nulend, J., Semeins, C. M., Ajubi, N. E., et al. (1995) Pulsating fluid flow increases nitric oxide (NO) synthesis by osteocytes but not periosteal fibroblasts—correlation with prostaglandin upregulation. Biochem Biophys Res Commun 217, 640–648. 22. Forwood, M. R. (1996) Inducible cyclo-oxygenase (COX-2) mediates the induction of bone formation by mechanical loading in vivo. J Bone Miner Res 11, 1688–1693. 23. Pead, M. J., Lanyon, L. E. (1989) Indomethacin modulation of load-related stimulation of new bone formation in vivo. Calcif Tissue Int 45, 34–40. 24. Li, J., Liu, D., Ke, H. Z., et al. (2005) The P2X7 nucleotide receptor mediates skeletal mechanotransduction. J Biol Chem 280, 42952–42959.
23 Fluid Flow Assays
345
25. Kreke, M. R., Huckle, W. R., Goldstein, A. S. (2005) Fluid flow stimulates expression of osteopontin and bone sialoprotein by bone marrow stromal cells in a temporally dependent manner. Bone 36, 1047–1055. 26. Wu, C. C., Li, Y. S., Haga, J. H., et al. (2006) Roles of MAP kinases in the regulation of bone matrix gene expressions in human osteoblasts by oscillatory fluid flow. J Cell Biochem 98, 632–641. 27. Li, Y. J., Batra, N. N., You, L., et al. (2004) Oscillatory fluid flow affects human marrow stromal cell proliferation and differentiation. J Orthop Res 22, 1283–1289. 28. Kreke, M. R., Goldstein, A. S. (2004) Hydrodynamic shear stimulates osteocalcin expression but not proliferation of bone marrow stromal cells. Tissue Eng 10, 780–788. 29. Riddle, R. C., Taylor, A. F., Genetos, D. C., et al. (2006) MAP kinase and calcium signaling mediate fluid flow-induced human mesenchymal stem cell proliferation. Am J Physiol Cell Physiol 290, C776–784. 30. Jiang, G. L., White, C. R., Stevens, H. Y., et al. (2002) Temporal gradients in shear stimulate osteoblastic proliferation via ERK1/2 and retinoblastoma protein. Am J Physiol Endocrinol Metab 283, E383–389. 31. Kapur, S., Baylink, D. J., Lau, K. H. (2003) Fluid flow shear stress stimulates human osteoblast proliferation and differentiation through multiple interacting and competing signal transduction pathways. Bone 32, 241–251. 32. Kato, Y., Windle, J. J., Koop, B. A., et al. (1997) Establishment of an osteocyte-like cell line, MLO-Y4. J Bone Miner Res 12, 2014–2023. 33. Frangos, J. A., McIntire, L. V., Eskin, S. G. (1988) Shear stress induced stimulation of mammalian cell metabolism. Biotechnol Bioeng 32, 1053–1060. 34. Taylor, A. F., Saunders, M. M., Shingle, D. L., et al. (2006) Mechanically stimulated osteocytes regulate osteoblastic activity via gap junctions. Am J Physiol Cell Physiol, 292:C545–C552. 35. Sill, H. W., Chang, Y. S., Artman, J. R., et al. (1995) Shear stress increases hydraulic conductivity of cultured endothelial monolayers. Am J Physiol 268, H535–543. 36. Donahue, T. L., Haut, T. R., Yellowley, C. E., et al. (2003) Mechanosensitivity of bone cells to oscillating fluid flow induced shear stress may be modulated by chemotransport. J Biomech 36, 1363–1371. 37. Reich, K. M., Gay, C. V., Frangos, J. A. (1990) Fluid shear stress as a mediator of osteoblast cyclic adenosine monophosphate production. J Cell Physiol 143, 100–104. 38. Jacobs, C. R., Yellowley, C. E., Davis, B. R., et al. (1998) Differential effect of steady versus oscillating flow on bone cells. J Biomech 31, 969–976. 39. Stevens, H. Y., Frangos, J. A. (2003) Bone cell responses to fluid flow. Methods Mol Med 80, 381–398.
Index
µCT. See Microcomputed tomography 17-β-estradiol. See Sex steroids 3D structural analysis, 273
A Adaptive modeling and remodeling, 311 Adenoviruses, 137–147 amplification, 140–141 ex vivo transduction, 139, 142 in vivo implantation, 139, 142–143 in vivo transduction, 143–144 large-scale purification, 140–141 plaque purification, 140 production, 138–139 safety, 138, 146 titer, 141 vector construction, 139–140 Adipocytes, 4, 16, 17, 33 Agarose beads, 167, 169, 170, 171 Algorithm for capturing microarray enrichment, 175 Alkaline phosphatase, 3, 6–7, 12–13, 16, 38, 145, 154 enzyme assay, 6, 12 histochemical staining, 7, 13 substrates chemiluminescent, 16 colorometric, 6, 16 fluorescent, 16 ALZET osmotic pumps. See Osmotic pumps Amide I, 293, 296, 300, 302 Anesthesia acepromazine, 113, 114 Avertin, 131 isoflurane, 80, 82, 103–107, 127, 129, 131 ketamine, 90, 93, 103–107, 113, 114, 127, 129, 131, 132, 139, 143, 266 pentobarbital, 131
xylazine, 90, 93, 103–107, 113, 114, 127, 129, 131, 132, 139, 143 Anisotropy, 285 Annexin V, 71 Anoikis, 59 Apoptosis, 51–75 Artificial matrix coating, 24, 31 Association mapping, 204, 208–209 general pedigree design, 217–219 population-based, 221 transmission-disequilibrium, 221 Axial load, 310
B BCA protein assays, 265, 269 Berkovich tip, 330 BFR. See Bone formation rate β-galactosidase, 38 BioCoat osteologic bone cell culture system, 24 Bioluminescent imaging, 102, 107, 108, 261–272 in vitro, 270 in vivo, 270 Biomarkers CTX, 144, 120 osteocalcin, 114, 120, 126, 262, 336 serum, 114, 116 urine, 114, 115 Bio-mechanical testing, 113, 120 Bisphosphonates, 53, 57, 60, 115 Blades. See Saws BMD. See Bone mineral density BMP. See Bone morphogenic protein Bone formation, 37, 38, 79–80, 89, 101, 111, 112, 137, 261, 308, 335 Bone formation rate, 119 Bone lining cells, 51 Bone loading in vivo, 309–310, 312, 316–319
347
348 Bone marrow stromal cell, 3, 6, 10–12, 101–104, 107–108 Bone mineral density, 111, 113, 116, 126 areal BMD of 4th lumbar vertebra, 117, 121 volumetric BMD of proximal tibia, 117–118 Bone morphogenic protein, 38, 49, 111, 137, 138, 142 Bone particle assay, 24, 32 Bone quality, 293, 323 Bone quantity, 323 Bone regeneration, 101, 137, 138, 261 Bone remodeling, 125, 126, 200, 262 Bone repair, 101 Bone resorption, 19, 30, 37, 51, 79, 111, 112, 125, 126, 200, 335 Bone slice, 23, 30 Bone strength, 325 Bone surface, 311 Bone turnover, 111 Bone volume fraction, 97, 118, 126, 284, 311 BV/TV. See Bone volume fraction
C Calcein blue. See Fluorescent dyes Calcitonin, 52 Calcium 45 Ca incorporation, 38 ionized calcium, 33, 34 serum, 145 Calvaria critical-sized defects, 143 primary osteoblasts, 3, 4–5, 7–10, 142, 149–155 Calvarial organ culture. See Organ culture Cancellous bone. See Trabecular bone Carbonate, 295, 300 Caspase-3 activation colorimetry, 61–63 fluorometry, 53, 61–63 immunocytochemistry, 66 subcellular localization of fusion protein substrates, 66–68 western blotting, 63–66 inhibition, 58 Cavalieri principle, 97, 99 CCD camera, 277, 280 CD11b, 19, 20, 22, 27, 28 CD14, 19, 20, 22, 23, 27–29 cDNA synthesis, 209 Cell cycle progression, 243 Cell detachment. See Anoikis Cell synchrony, 239, 241, 243 double thymidine block, 243
Index nocodazole, 243, 244 serum deprivation, 244 Cellular proliferation, 336 CFP. See Fluorescent proteins Charged coupled device camera, 250–251 ChIP display, 177–190 ChIP. See Chromatin immunoprecipitation ChIP-on-chip, 165–175, 177, 186 Chromatin, 165, 167, 174, 181 Chromatin condensation apoptosis, 51, 52, 71 Chromatin immunoprecipitation, 165–175, 177, 181, 191, 192, 193, 198–199, 200 Co-culture assays, 19, 21 Collagen hydrogels, 139, 142–144, 146 type I, 52, 55, 92, 95, 96, 336, 337, 340 fragments, 31, 114, 120 Collagen cross-linking, 293, 296, 300 Collagenase, 4, 5, 15, 22, 26, 59, 150, 155 Computed tomography. See Microcomputed tomography Connectivity density, 118, 284 Cooled charge coupled device imaging, 261, 262, 266 Coronal suture, 41, 42, 45, 46 Cortical bone, 126, 310, 323–325 Counterstains. See Stains CpG methylation, 186 Cre recombinase, 139 Cross-linking, 165, 166, 167, 168, 169–170, 171 Cytochrome C release, 71
D Dartec HC10, 310, 315 safety, 320 setting up, 316–318 DC protein assay, 6, 12 Decalcification, 40, 43, 93, 263–264, 266–267 Dentin, 23, 30 Deparaffinization, 93 DEPC, 80, 83, 84, 86 DEXA. See Dual x-ray absorptiometry Dexamethasone, 5, 21, 26, 53, 60, 103 Diamond edged wafering blade. See Saws Diffusion chambers, 90 Dihydrotestosterone. See Sex steroids Dispase, 22, 26, 27, 33, 150, 155 Dissection calvarial, 41, 42 femur, 8, 25, 29 DNA digestion, 178–180, 181–182
Index purification, 168, 171–172 fragmentation, 51, 52, 60 Dnase, 52, 61 Dual x-ray absorptiometry, 323 DEXA, 125, 126 peripheral DXA, 111, 113, 115, 116, 117 Dynabeads, 23, 28
E Ectopic bone. See Heterotopic bone Elastic modulus, 323, 324, 327, 328, 331, 332 Electrophoretic transfer, 65 Electrophoresis agarose DNA, 60 RNA, 80–81, 84–85, 86 SDS-PAGE, 53–54, 64 Embedding paraffin, 44, 49, 91, 93, 263–264, 266–267 plastic, 91, 93, 99, 105, 113 polymerization, 298, 300 polymethyl methacrylate, 296 Endothelin-1, 38, 48, 49 Eosin. See Stains Epistasis, 206, 223–224, 231 eQTL, 203, 208, 224–225 ERK inhibitors, 53, 58 Eroded surfaces, 311 Estrogen receptors, 111 Estrogen. See Sex steroids Estrogen-deficiency, 111, 112 Etoposide, 53, 60 Euthanasia, 7–8, 39, 82 Ex vivo culture. See Organ culture Extracellular matrix, 4, 10, 262 Extracellular matrix mineralization, 149, 154
F FACS, 22, 27, 71, 243–244 Femur dissection for bone marrow isolation, 8, 25, 29 metaphyseal trabecular primary spongiosa, 79, 80 Fibroblasts cell lines, BLK, 142, 144 NIH3T3, 150, 151 ST2, 20, 21, 26, 33, 140, 158 stromal cells, 20, 26 Ficoll-Paque, 23, 28 Finite element modeling, 311
349 Fixation acetone, 72 formaldehyde, 70, 86 formalin, 40, 43 methanol, 72 Flow cytometry cell viability, 152, 154 Fluid flow assays, 335–345 interstitial fluid flow, 335, 336 oscillatory, 343 pulsatile, 343 steady, 343 Fluid flow shear stress. See Mechanical forces Fluorescence activated cell sorting. See FACS Fluorescence resonance energy transfer, 68 Fluorescent dyes calcein blue, 7, 14–15, 115, 116, 119, 311 Hoechst 33258, 70 Fluorescent proteins CFP, 68 GFP, 68, 71, 72, 154 RFP, 68, 71, 72 YFP, 67–68 Fluorophores Cy-3, 169, 173, 174 Cy-5, 169, 173, 174 Focal microscopy. See Imaging Formaldehyde, 91, 105, 167, 169, 174 Formalin. See Fixation Fourier transform-infrared microscopy, 293–303 data acquisition and analysis, 297, 299–300 preparation of bone samples, 299 Fractures, 102, 274, 307, 323 FRAP, 239, 254–256 FRET. See Fluorescence resonance energy transfer Frontal suture, 41, 42 Frozen tissue processing. See Tissue processing FT-IR imaging spectroscopy, 293 FT-IR microspectroscopy. See Fourier transform-infrared microscopy
G Gelatin sponges, 103, 105, 108, 139, 142–144 Gelfoam. See Gelatin sponges Gene silencing. See RNA interference, 149 Gene therapy, 137 GFP. See Fluorescent proteins Goldner’s trichrome stain. See Stains
350 H H&E staining. See Stains HA/TCP. See Hydroxyapatite/tricalcium phosphate Haplotype, 205–206 Hardness, 323, 324, 328, 331, 332 Harris’ acidified hematoxylin. See Stains, hematoxylin Haversian systems, 336 HEK 293T cells, 138–141, 149–151 Hematoxylin. See Stains Hematoxylin/eosin. See Stains Heparin, 90, 92 Heterotopic bone, 89, 92 Histological staining, 101, 194 Histology, 113, 118–119, 146, 261, 311 Histomorphometry, 37, 45, 97, 101, 102, 121, 125, 126, 311 Histone, 165, 166, 169, 171 Hoechst dye. See Fluorescent dyes Hydroxyapatite, 4, 89, 295, 302 Hydroxyapatite/tricalcium phosphate, 89, 90, 92–94, 97, 98 Hyperparathyroidism, 79 Hypothermia, 132
I Image analysis Bioquant OSTEO, 24, 31, 45, 113, 119 ImagePro, 9–10, 14, 96 MetaMorph, 45 OsteoMeasure, 24, 31, 45 Imaging confocal microscopy, 231, 39, 251–254 GFP fluorescence, 261 immunofluorescence, 239, 249–251 light microscopy, 31 live cell, 239, 249–251 non-destructive, 273 radiography, 105, 274–277, 286 scanning electron microscopy (EM), 31 three-dimensional, 273 Immunocompromised mice. See Immunodeficient mice Immunodeficient mice, 89, 90, 102 athymic nude, 104 NIH-bg-nu-xid, 139 NOD/SCID, 90, 93, 98 Immunohistochemistry, 92, 93, 96, 102, 261, 264, 266–269, 270 collagen type I, 92, 95, 96 osteonectin, 92, 95, 96 osteopontin, 92, 95, 96
Index Immunoprecipitation, 165, 167, 168, 169, 170–171, 174 Implants, 274 In situ hybridization, 89, 96 In situ nick end-labeling, 52 In situ nick translation, 52, 60 In vivo adaptive response, 307–322 Informatics, 256 INST. See In situ nick translation Interaction analysis, 223–224 ISEL. See In situ nick end-labeling Isofluorane. See Anesthesia
J Joint binding deconvolution, 175
K Ketalar. See Anesthesia, ketamine Ketamine. See Anesthesia Ketaminol. See Anesthesia, ketamine Kinase inhibitors regulators of apoptosis, 52, 53 Knockout mice, 20, 38 Kozak consensus sequence, 158, 159
L Lamboid suture, 41, 42 Laser capture microdissection, 191–201 catapulting samples, 193, 195–196 preparation of samples, 193 specimen isolation and collection, 195 Lentiviruses, 38, 149–155 infection, 153–154 production, 150, 151 safety, 151 titer, 150, 151, 152, 155 vectors, 154 Light microscopy. See Imaging Line skeleton graph analysis, 285 Linear attenuation coefficient, 276 Linear regression curve, 63 Linkage analysis, 204, 208, 219–221 Linkage disequilibrium, 203, 204–205, 230 Linkage mapping, 203 Linker ligation, 178–180, 181–182 Loading rate, 323 Luciferase, 101, 107, 108, 138, 154, 261, 262, 264, 265, 266, 269 antibody, 267–268 Luciferin, 261, 266, 271 Lymphoprep, 90, 92
Index M Macrophage colony stimulating factor (M-CSF), 19–33, 56, 72 Magnetic assisted cell sorting, 22, 27, 28 isolation of human CD14 postive osteoclast precursors, 22, 23, 27, 28 MAR. See Mineral apposition rate Matrix, 323–334 Mean curvature, 285 Measurement of bone strains in vivo, 307–309, 311–312 attach the gauge, 314–315 measure strain, 315–316 prepare the bones, 312–314 strain gauge preparation, 312–314 Mechanical forces, 51, 52, 335 fluid flow shear stress, 57 stretching, 57 Mechanical loading, 38, 307–322 Mechanical properties, 323–334 bending, 324 compression, 324 tension, 324 torsion, 324 viscoelastic complex modulus, 331 loss modulus, 328, 331 storage modulus, 328, 331 Mesenchymal cell lines C3H10T1/2, 144 Mesenchymal stem cells, 89, 90, 92–93, 101, 191, 261, 262, 335, 336, 337 Metaphase chromosome spreads, 241, 245–247 Methyl methacrylate. See Embedding, plastic Microarray, 38, 80, 165, 166, 169, 172, 174, 185, 203, 206–207, 224–225 Micro-computed tomography, 96–98, 99, 101, 102, 105, 118, 121, 261, 273–292, 311, 325 image processing artifacts, 283, 287–289 beam-hardening, 283, 286, 287 Fourier transformation, 282–283, 286–289 Gaussian filtration, 282–283, 286–289 in vitro, 277–279 in vivo, 277 phase contrast, 274 scan and image reconstruction, 279–282, 286–289 structure analysis, 283–285, 286–289
351 Micro-CT. See Microcomputed tomography Micro-finite-element models, 325 Microscopy. See Imaging Microstrain, 336 Microtome, 94, 119, 194, 297, 299 Mineral apposition rate, 119 Mineral crystallinity, 293, 300 Mineralized matrix, 37 MOI. See Multiplicity of infection MOPS, 81, 84 Morphometry, 273, 285 Multiplicity of infection, 142, 144, 153, 155 Multivariant analysis, 222–223 Myoblast cell lines C2C12, 142
N Nanoindentation, 323–334 calibration of tip area function, 327 post processing, 328–330 preparation of samples, 326–327 quasistatic, 327–328 Neonatal calvarial organ culture. See Organ culture NOD/SCID. See Immunodeficient mice Nodules, 10, 38 Non-decalcified sections. See Tissue processing, 93 Nuclear fragmentation apoptosis, 52, 71 Nuclear matrix, 239, 241–242, 247–249 Nuclear morphology, apoptosis, 69 Nuclear organization, 239 Nude mice. See immunodeficient mice
O ODX. See Surgery OPG. See Osteoprotegerin Optimal cutting temperature (OCT) Tissue-Tek embedding medium, 194, 200, Orange G. See Stains Orchidectomy. See Surgery Organ culture calvarial, 24, 32, 34, 37–50 long bone, 32 Osmotic pumps, 80, 82, 85–86 implantation surgery, 82 loading, 82 Ossicle, 101–110
352 OSTEO II. See Image analysis Osteoblasts, 3–18, 37, 89, 101, 108, 111, 126, 149, 157, 165, 166, 200, 262, 311, 335, 336, 337, 343 apoptosis, 51, 57, 59 bone marrow, differentiation, 6, 10–12 growth, 6, 10–12 isolation, 6, 10–12 calculation of number, 47, 48 calvarial organ cultures, 24, 32, 34, 37–50 calvarial cells differentiation, 5, 9–10 growth, 5, 9–10, 142 isolation, 4–5, 7–9 cell lines hFOB 1.19, 337, 341 MC3T3–E1, 3, 158, 181, 186, 336, 337 MG-63, 343 OB-6, 52, 54, 57, 64, 71, 72 ROS17/2.8, 3, 343 Saos-2, 167, UMR-106, 3, Osteocalcin, 114, 120, 126, 262, 336 Osteoclasts, 19–35, 37, 38, 111, 165, 200, 311 apoptosis, 51, 59, 62, 70 cell lines RAW264.7, 20, 23, 26, 29, 33, 52, 54, 56, 62 differentiation, 25, 56, 72 human osteoclast isolation, 23, 28 multinucleated cells, 23, 29 murine isolation from bone marrow precursors, 20, 25, 27, 28, 32, 55 isolation from spleen, 20, 23, 25, 27, 28, 56 Osteocytes, 37, 38, 335, 336, 337 apoptosis, 51, 57, 58, 59 MLO-Y4 cell line, 52, 54, 55, 66, 72, 337, 340, 341, 342 Osteogenesis. See Bone formation Osteolysis, 19, 38 OsteoMeasure. See Image analysis Osteonectin, 92, 95, 96 Osteopenia, 111, 112, 116, 335 Osteopontin, 92, 95, 96, 336 Osteoporosis, 19, 79, 102, 111, 112, 125, 200, 293, 300, 307, 323 Osteoprogenitor cells, isolation and culture of, 3–18 Osteoprotegerin, 19, 34 Osteosarcoma, 165
Index Ovariectomy. See Surgery OVX. See Surgery
P PAGE, 178, 180, 183, 185, 189 Paraffin wax. See Embedding Parallel plate fluid flow, 335, 336 apparatus, 337–338 benefits and limitations, 343 cell culture, 339–340 experimental setup, 340–341 Parathyroid hormone, 38, 52, 57, 79–81, 101, 102, 108, 112, 116 continuous infusion, 9, 81, 82 dosage, 79 intermittent injection,79, 81 preparation, 79, 81 Parathyroid hormone related protein, 38, 102 PARP. See Poly (ADP-ribose) polymerase PCR, 79, 154, 168, 169, 171–172, 197–198, 209 amplification, 180 ligation-mediated, 169, 174, 178, 183 real-time, 181 pDXA. See Dual x-ray absorptiometry Pepsin, 268, 271 Peripheral dual x-ray absorptiometry. See Dual x-ray absorptiometry Peripheral quantitative computed tomography, 111, 113, 115, 116, 117–118 Phloxine B. See Stains Phosphate, 145, 295, 300 Plasma membrane permeability apoptosis, 69 p-Nitrophenyl phosphate, 6, 12 PNPP. See p-Nitrophenyl phosphate Poly (ADP-ribose) polymerase, 68 Polybrene, 150, 152, 153, 159, 160 Polystyrene inserts, 40, 49 pQCT. See Peripheral quantitative computed tomography Primary spongiosa, 79, 80, 83, 126 Principle components analysis, 222 Programmed cell death. See Apoptosis Protease inhibitors, 167, 170 Proteinase K, 168 Protein-DNA interactions, 177 Proteosome inhibitors, 38, 170, PTH. See Parathyroid hormone PTHRP. See Parathyroid hormone related protein
Index Puromycin, 149, 150, 153, 158, 159, 160 Pyridinoline, 114, 120, 126
Q Quantitative trait loci mapping, 203–235 back cross design, 212–213 F2 design, 213–215 sib-pair design, 215–217 software, 229
R Radiography. See Imaging RANKL, 19–33, 56, 72, 111 Receptor activator of NF-κB ligand. See RANKL Receptor antagonists regulators of apoptosis, 52 Regression analysis, 219–221 Resorption assays. See Bone resorption Retroviruses, 137, 157–162 packaging cell lines, 158–161 transfection, 159–160 safety, 158, 159, 160 titer, 161 transduction of host cells, 160 vectors, 158 Reverse transcription, 197 RFP. See Fluorescent proteins RNA clean-up Rneasy, 81, 84 extraction from rat femur, 79, 82, 86 isolation, 79, 81, 83–84, 86, 160, 191, 192, 193, 196–197, 208–209 stabilization, RNAlater, 80, 81, 83, 86 RNA interference shRNA, 149, 151, 153–154 siRNA, 149, 158 Roller drum, 24, 32 Rompun. See Anesthesia, xylazine Rotating disk fluid flow, 335 Rotating disk fluid flow apparatus, 338–343 benefits and limitations, 343 cell culture, 341 experimental setup, 341–342
S Sagittal suture, 41–44 Saws band saw, 297, 299
353 isomet saw with diamond edged wafering blade, 11, 23, 30, 113, 120 low speed diamond wheel, 325, 326, 331 Scanning electron microscopy (EM). See Imaging SDS-PAGE. See Electrophoresis SERMs, 115, 116 Serum starvation apoptosis, 60 cell synchronization, 244 Sex steroids 17-β-estradiol, 53, 60 dihydrotestosterone, 53, 60 estrogen, 112, 115, 116 Shearing. See Sonication Sheer stress, 336, 342, 343 Silver nitrate. See Stains, von Kossa Single nucleotide polymorphisms (SNPs), 205–206, 208–209, 230–231 Sonication, 186 Specific bone surface (BS/BV), 284 Sprague-Dawley rats, 80, 116, 121 Stainless steel grids, 39, 40, 49 Stains eosin Y, 40, 41, 45, 46, 91 fast green FCF, 91 Goldner’s trichrome, 91, 94, 97, 119 hematoxylin, 40, 45, 70, 91, 267 Mayers, 92, 96, 193, 195, 200 Weigert, 91 hematoxylin/eosin, 91, 93 methyl green, 61 methylene blue, 91 oil red O, 17 orange G, 40, 41, 46, 91 phloxine B, 40, 41 Ponceau/fuchsin, 91 toludine blue, 24, 30 trypan blue, 152 von Kossa, 7, 13–14, 17, 154 Statistical analysis ANOVA, 49 Graphpad Prism, 48 Stiffness, 323, 324 Strain, 308, 336 compressive, 308 microstrain, 308, 315 tensile, 308 Strain gauging, 307–322 attach the gauge, 314–315 gauge factor, 308 measure strains, 315–316 preparation, 312–314 Strain rate, 331
354
Index
Stretching. See Mechanical forces Stromal cells. See fibroblast cell lines Structure model index, 284 Surgery implantation surgery, 82 orchidectomy, 125–134 ovariectomy, 102, 111–124, 125 subcutaneous implantation, 89, 103, 104, 106, 138, 142, 146 transplantation, 93, 102 Swiss white mice, 39 SYBR Green, 165, 167, 169, 170, 174, 186 Synchrotron radiation, 274–277, 286
location analysis, 177 Runx2, 138, 165, 166, 167, 171, 172, 173, 178, 181, 186, 239, 245, 251 Transfection, 70–72, 151 Transgenic mice, 20, 38, 125, 261 TRAP. See Tartrate resistance acid phosphatase Trypsin, 4, 5, 15, 69, 72, 92, 103, 104, 108, 151, 152, 340 Tumor, 101, 108 Tumor-induced bone loss. See Osteolysis TUNEL, 52 Tungsten carbide knife, 119, 297, 299
T Tartrate resistance acid phosphatase, 30, 33, 70, 126 Tb.N. See Trabecular number Tb.Sp. See Trabecular spacing Tb.Th. See Trabecular thickness Tensile strength, 296 Tetracycline, 121 Tiling arrays, 168 Tissue engineering, 102 Tissue processing, 43, 44, 174, 296–297 dehydration, 297 fixation, 263–264, 266–267, 297, 301 frozen tissues, 194 infiltration, 298 sectioning, 263–264, 266–267 Trabecular bone, 4, 274, 310, 323–325, 331 Trabecular bone pattern factor, 285 Trabecular number, 118, 126, 284 Trabecular separation, 284, 311 Trabecular spacing, 118, 126 Trabecular thickness, 118, 126, 284, 311, 325 Transcription factors androgen receptor, 178 fixation, 194 identification of target genes, 177–190
V Variance components analysis, 165 Vertebral bone transplant. See Vossicle Viscoelasticity, 296 Vitamin D3, 21, 26, 53, 57 von Kossa. See Stains Vossicle, 101–110
W Western blotting, 53–54, 63–66, 68, 144, 161 Wheatstone bridge, 308–309, 316 Whole genome amplification, 220–221 Whole genome association, 168–169, 172–174 Woven bone, 14, 15
X X-ray diffraction, 296 X-ray radiation. See Imaging, radiography Xylazine. See Anesthesia
Y YFP. See Fluorescent proteins
