Detection of Highly Dangerous Pathogens: Microarray Methods for BSL 3 and BSL 4 Agents

Detection of Highly Dangerous Pathogens: Microarray Methods for the Detection of BSL 3 and BSL 4 Agents Edited by Tanja...

Author: Tanja Kostic | Patrick Butaye | Jacques Schrenzel

18 downloads 586 Views 4MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!

Report copyright / DMCA form

DOWNLOAD PDF

Detection of Highly Dangerous Pathogens: Microarray Methods for the Detection of BSL 3 and BSL 4 Agents

Edited by Tanja Kostic Patrick Butaye Jacques Schrenzel WILEY-VCH

Detection of Highly Dangerous Pathogens

Edited by Tanja Kostic, Patrick Butaye, and Jacques Schrenzel

Related Titles X.-H. N. Xu (Ed.)

New Frontiers in Ultrasensitive Bioanalysis Advanced Analytical Chemistry Applications in Nanobiotechnology, Single Molecule Detection, and Single Cell Analysis 2007 ISBN: 978-0-471-74660-7

N. Khardori (Ed.)

Bioterrorism Preparedness Medicine — Public Health — Policy 2006 ISBN: 978-3-527-31235-1

R. F. Pilch, R. A. Zilinskas (Eds.)

Encyclopedia of Bioterrorism Defense 2005 ISBN: 978-0-471-46717-5

Detection of Highly Dangerous Pathogens Microarray Methods for the Detection of BSL 3 and BSL 4 Agents

Edited by Tanja Kostic, Patrick Butaye, and Jacques Schrenzel

The Editors Dr. Tanja Kostic Austrian Research Centers GmbH – ARC Department of Health and Environment Bacterial Ecology and Genomics 2444 Seibersdorf Austria

All books published by Wiley-VCH are carefully produced. Nevertheless, authors, editors, and publisher do not warrant the information contained in these books, including this book, to be free of errors. Readers are advised to keep in mind that statements, data, illustrations, procedural details or other items may inadvertently be inaccurate. Library of Congress Card No.: applied for

Prof. Patrick Butaye Vet. & Agrochem. Research Center Department Bacteriology/Immunology Groeselenberg 99 1180 Brussels Belgium Prof. Jacques Schrenzel University Hospitals of Geneva Service of Infectious Diseases Genomic Research Laboratory Rue Micheli-du-Crest 24 1211 Geneva 14 Switzerland

Cover illustration Left: Mapping of statistically relevant probes on a reduced 16S rDNA bacterial phylogenetic tree after hybridization of an unknown sample onto a 16S phylogenetic microarray. Red dots and lines depict nodes and branches where at least 1 probe yielded a statistically signiﬁcant signal. Right: Hybridization patterns obtained with 3 Bacillus anthracis strains using a universal microarray design, showing strain-speciﬁc resolution. Image and results with kind permission from Antoine Huyghe (Genomic Research Laboratory, Geneva, Switzerland).

British Library Cataloguing-in-Publication Data A catalogue record for this book is available from the British Library. Bibliographic information published by the Deutsche Nationalbibliothek The Deutsche Nationalbibliothek lists this publication in the Deutsche Nationalbibliograﬁe; detailed bibliographic data are available on the Internet at http://dnb.d-nb.de. # 2009 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim All rights reserved (including those of translation into other languages). No part of this book may be reproduced in any form – by photoprinting, microﬁlm, or any other means – nor transmitted or translated into a machine language without written permission from the publishers. Registered names, trademarks, etc. used in this book, even when not speciﬁcally marked as such, are not to be considered unprotected by law. Cover design Adam-Design, Weinheim Typesetting Thomson Digital, Noida, India Printing Strauss GmbH, Mörlenbach Binding Litges & Dopf GmbH, Heppenheim Printed in the Federal Republic of Germany Printed on acid-free paper ISBN: 978-3-527-32275-6

V

Contents This Publication is Supported by COST Preface

XIII

List of Contributors 1 1.1 1.2 1.2.1 1.2.1.1 1.2.1.2 1.2.1.3 1.2.2 1.2.2.1 1.2.2.2 1.2.2.3 1.2.2.4 1.2.2.5 1.2.2.6 1.2.3 1.2.3.1 1.2.3.2 1.2.3.3 1.2.4 1.2.4.1 1.2.4.2 1.2.5 1.2.5.1 1.2.5.2

XI

XV

Introduction to Microarray-Based Detection Methods 1 Jacques Schrenzel, Tanja Kostic, Levente Bodrossy, and Patrice Francois Introduction to Microarray Technology 1 Technical Aspects of Microarray Technology 2 Probes 2 Genome Fragments 3 PCR Products 3 Oligonucleotide Probes 3 Substrates for Printing 6 Slides with Poly-L-lysine Coating 9 Slides with Amino Silane Coating 9 Slides with Aldehyde Coating 9 Slides with Epoxy Coating 9 Proprietary Surface Chemistries 9 Probe Spacers 9 Targets for Microarray Analysis 10 Target Ampliﬁcations and Sensitivity Issues 10 Labeling of the Targets 11 Hybridization and Wash Conditions 11 Classical Commercially Available Microarray Formats 12 Spotting Approaches 12 In Situ Synthesis 12 Alternative Methods for Improving Microarray-Based Detection Sensitivity 14 Resonance-Light Scattering (RLS) 14 Planar-Waveguide Technology (PWT) 14

VI

Contents

1.2.5.3 1.2.5.4 1.2.6 1.3 1.4 1.4.1 1.4.2 1.4.3 1.4.4 1.4.5 1.5 1.6

Liquid Arrays 15 Three-Dimensional Microarray Formats 15 Marker Genes Used on MDMs 16 Analysis and Quality Control Aspects 17 Applications of Microarray Technology in Microbial Diagnostics Gene Expression Studies 18 Comparative Genomic Hybridization (CGH) 18 Generic or Universal Microarrays 19 Microarrays for Sequence Analysis 20 Microbial Diagnostic Microarrays (MDMs) 21 Further Developments and New Perspectives Regarding Array Sensitivity and Speciﬁcity 22 Conclusions 22 References 22 35

Part I

Methods

2

Long Oligonucleotide Microarray-Based Microbial Detection 37 Tanja Kostic and Levente Bodrossy Introduction 37 Method 38 DNA Extraction 38 F29 Ampliﬁcation 38 Klenow Ampliﬁcation/Labeling 39 Probe and Slide Preparation 40 Slide Processing Protocol (for Amino Surfaces) 41 Hybridization and Slide Washing 42 Comments 44 Our Test System and Results 44 Conclusions 44 References 46

2.1 2.2 2.2.1 2.2.2 2.2.3 2.2.4 2.2.5 2.2.6 2.2.7 2.3 2.4

3 3.1 3.2 3.3 3.4 3.5 3.6 3.7 3.8 3.9 3.10 3.11

Sequence-Specific End-Labeling of Oligonucleotides 47 Tanja Kostic and Levente Bodrossy Introduction 47 Probe Design 50 Slide Preparation (Spotting) 51 Slide Processing Protocol ( for Aldehyde Surfaces) 52 DNA Extraction and PCR Ampliﬁcation of the Targeted Gene 52 Shrimp Alkaline Phosphatase Treatment 53 Labeling 53 Hybridization and Slide Washing 54 Data Analysis 55 Costs 55 Microarray for Detection of Pathogenic Bacteria 55 References 56

18

Contents

4

4.1 4.2 4.2.1 4.2.2 4.2.3 4.2.4 4.2.5 4.3

5 5.1 5.2 5.3 5.4 5.4.1 5.4.2 5.5 5.5.1 5.5.2 5.5.3 5.6

6 6.1 6.2 6.2.1 6.2.2 6.2.3 6.2.4 6.2.5 6.3

Non-Cognate Approaches for Pathogen Detection on Microarrays 59 Antoine Huyghe, Patrice Francois, Yvan Charbonnier, David Hernandez, Jonathan Hibbs, and Jacques Schrenzel Introduction 59 Non-Cognate Hybridization System 60 Concept 60 Deﬁnition of the Optimal Probe Length 60 Virtual Assessment of Array Performances (in Silico Experiments) 61 Array Manufacturing and Hybridization (Wet-Lab Experiments) 63 Analysis 64 Perspectives 64 References 64 Patterning Techniques for Array Platforms 67 Erhan Pis¸kin, Bora Garipcan, Gökhan Demirel, and Og uzhan Çaglayan Introduction 67 Soft Lithography 68 Photolithography 70 Robotic Printing 71 Micro-Spotting 71 Ink-Jet Printing 72 Lithography with AFM 74 Dip-Pen Lithography with AFM 75 cAFM Lithography 76 Nanoshaving and Nanografting 77 Conclusions 80 References 81 Probe Immobilization Techniques in Array Technologies Erhan Pis¸kin, Bora Garipcan, and Memed Duman Introduction 85 Support Material 86 Glass 86 Silicon 90 Gold 90 Polymers 91 QDs 91 Immobilization 94 References 99

85

VII

VIII

Contents

105

Part II

Identification

7

Low-Cost and Low-Density Microarrays – A Novel Technique for Identification and Typing of Microorganisms 107 Dimitrios Frangoulidis, Volker Heiser, Olfert Landt, and Hermann Meyer Introduction 107 Chip Design/Array Description 107 Protocol for a LCD Array Experiment 108 Ampliﬁcation 108 Short Protocol for Hybridization and Labeling 109 Results and Discussion 111 Conclusions 111

7.1 7.2 7.3 7.3.1 7.3.2 7.4 7.5 8

8.1 8.2 8.3 8.4 8.5 8.6 8.7 8.8 8.9

9 9.1 9.2 9.3 9.4

DNA Microarray Technique for Detection and Identification of Viruses Causing Encephalitis and Hemorrhagic Fever 113 Henrik Nordström, Kerstin I. Falk, Peter Nilsson, and Åke Lundkvist Introduction 113 Principle of Applying Microarray Technology for Virus Detection and Identiﬁcation 113 Viruses and the Importance of Rapid Diagnostics 115 Advantages and Drawbacks of Using the Microarray Technique 115 Key Factors for Development of a Microarray-Based Test 116 Hantavirus Microarray 117 Flavivirus Microarray 117 Hemorrhagic Fever Viruses 121 Conclusions 123 References 123 Microarrays for Genomotyping of Pathogens 125 Jasper Kieboom, Ingrid Voskamp, and Martien P. Broekhuijsen Aim and Approach 125 Francisella Genomotyping 125 Brucella Genomotyping 128 Conclusions 130 References 131 133

Part III

Typing

10

Single Nucleotide Polymorphisms as Targets for DNA-Based Identification and Typing of Biosafety Level 3 Bacteria 135 Pierre Wattiau, Pieter Vos, and David Fretin Introduction 135 Real-Time Polymerase Chain Reaction: More Than a Simple Polymerase Chain Reaction 135

10.1 10.1.1

Contents

10.2 10.2.1 10.2.1.1 10.2.2 10.2.2.1 10.2.3 10.2.3.1 10.3

11

11.1 11.2 11.3 11.4 11.5

Results of Different SNP Typing Methods 136 Locked Nucleic Acid-containing TaqMan probes as SNP Typing Tools 136 Use of TaqMan LNA Probes for the Typing of Brucella suis Subspecies 136 Typing SNPs with Molecular Beacons 139 Use of Molecular Beacons for Typing Burkholderia pseudomallei Subspecies 139 Typing Large Number of SNPs by Ligation Detection Reaction 139 Use of LDR Probes and Low-Density Microarrays for the Multiplex Diagnostic of BSL3 Bacteria 142 Conclusions 144 References 145 Use of a Microchip to Detect Antibiotic Resistance Genes in Bacillus anthracis 147 Vincent Perreten and Joachim Frey Introduction 147 Conjugal Transfer of Antibiotic Resistance Genes Between Enterococcus Species and Avirulent Strains of B. anthracis 148 Determination of the MICs of Different Antibiotics for the B. anthracis Transconjugants 148 Detection of Antibiotic Resistance Genes in B. anthracis by Microchip-Based Hybridization System (ArrayTube) 149 Conclusions 150 References 151 153

Part IV

Quality Control

12

Progress Towards Development of Microarrays for Routine Diagnostic Use 155 Karen Kempsell, Sonal Shah, Susanna Sherwin, Richard Vipond, and Nigel Silman Introduction 155 Materials and Methods 157 Bacterial Strains, Culture and Nucleic Acid Puriﬁcation 157 Design and Printing of Oligonucleotide Probes 157 Random Ampliﬁcation and Cy3-labeling of Nucleic Acid 159 Hybridization of Labeled Targets and Data Processing 159 Results 160 Random Ampliﬁcation and Hybridization of Cy3-labeled Pathogen DNA Targets 160 Quality Assessment of Intra- and Inter-Operator Variation 161

12.1 12.2 12.2.1 12.2.2 12.2.3 12.2.4 12.3 12.3.1 12.3.2

IX

X

Contents

12.3.2.1 12.3.2.2 12.4

Sources of Intra-Operator Experimental Variation 162 Sources of Inter-Operator Experimental Variation 164 Discussion 165 References 165 Index

169

XI

This Publication is Supported by COST COST- the acronym for European Cooperation in Science and Technology- is the oldest and widest European intergovernmental network for cooperation in research. Established by the Ministerial Conference in November 1971, COST is presently used by the scientiﬁc communities of 35 European countries to cooperate in common research projects supported by national funds. The funds provided by COST – less than 1% of the total value of the projects – support the COST cooperation networks (COST Actions) through which, with EUR 30 million per year, more than 30 000 European scientists are involved in research having a total value which exceeds EUR 2 billion per year. This is the ﬁnancial worth of the European added value which COST achieves. A ‘‘bottom up approach’’ (the initiative of launching a COST Action comes from the European scientists themselves), ‘‘à la carte participation’’ (only countries interested in the Action participate), ‘‘equality of access’’ (participation is open also to the scientiﬁc communities of countries not belonging to the European Union) and ‘‘ﬂexible structure’’ (easy implementation and light management of the research initiatives) are the main characteristics of COST. As precursor of advanced multidisciplinary research COST has a very important role for the realisation of the European Research Area (ERA) anticipating and complementing the activities of the Framework Programmes, constituting a ‘‘bridge’’ towards the scientiﬁc communities of emerging countries, increasing the mobility of researchers across Europe and fostering the establishment of ‘‘Networks of Excellence’’ in many key scientiﬁc domains such as: Biomedicine and Molecular Biosciences; Food and Agriculture; Forests, their Products and Services; Materials, Physical and Nanosciences; Chemistry and Molecular Sciences and Technologies; Earth System Science and Environmental Management; Information and Communication Technologies; Transport and Urban Development; Individuals, Societies, Cultures and Health. It covers basic and more applied research and also addresses issues of pre-normative nature or of societal importance. Web: http://www.cost.esf.org

XII

This Publication is Supported by COST

Legal Notice by COST Office Neither the COST Ofﬁce nor any person acting on its behalf is responsible for the use which might be made of the information contained in this publication. The COST Ofﬁce is not responsible for the external websites referred to in this publication.

XIII

Preface The main objective of COST (European Cooperation in the ﬁeld of Scientiﬁc and Technical Research) Action B28 is to increase knowledge on Biosafety Level 3 and 4 agents by supporting the development of more accurate diagnostic assays, vaccines and therapeutics, and to better understand the epidemiology of these highly pathogenic microorganisms that can also potentially be used as biological weapons. The research initiatives and interests of COST Action B28 partners are organized in ﬁve working packages/groups:

WG1: Technology platform (including ﬂow cytometry and microarrays) WG2: Antigenicity WG3: Proteomics and glycomics WG4: Genomics WG5: Microbiology (bacteriology, virology and mycology)

Research groups united in WG1 possess extensive expertise in the development, optimization and/or application of microarray technology. Microarrays are genomic tools originally developed to monitor gene expression, but also applied for the detection of speciﬁc mutations in DNA sequences, and lately employed in the parallel detection, identiﬁcation and/or characterization of microorganisms. With the publication of the ﬁrst microarray studies in 1995, and the milestone of nearly 5000 published microarray papers in 2004, there is no doubt that this technology has rapidly spread into both basic and applied research. More information on WG1 and also of other working groups can be found on the COST Action B28 website (www.cost-b28.be). The aim of this book is to summarize actual general knowledge on DNA microarray technology, and to provide insights on the development and application of microarraybased experimental procedures. Contributions to the booklet illustrate the width and heterogeneity of microarray-based methods and possible applications, only with the framework of COST Action B28. The Methods section (Chapters 2 to 6) introduces ﬁve different techniques available for the setup of the microarray platforms. These contributions include concepts that were used to develop the selected approach. Some of them also provide straightforward step-by-step protocols and authors notes on the optimization steps required by method development. Chapters 7 to 11 describe

XIV

Preface

applications of the microarrays for identiﬁcation and/or characterization (typing) of microorganisms. These works depict once more a variety of the microarray-based methods. Furthermore, they demonstrate diverse experimental questions that can be tackled using microarray technology, from viruses to bacteria, from identiﬁcation, to characterization and detection of speciﬁc traits (i.e. antibiotic resistance). Finally, to put these new techniques into practice, e.g. in clinical diagnostics, appropriate quality control parameters, such as sensitivity speciﬁcity and reproducibility, should be guaranteed. These practical problems of quality control associated with array platforms are discussed in the last section (Chapter 12). This booklet is by no means a complete and ﬁnal microarray manual. As it is also such a fast evolving new technique, it would be nearly impossible to give a real up-todate overview. However, it gives the reader an interesting insight into this new and fast evolving ﬁeld. The information presented here may be helpful to researchers interested in microarray technology and its applications, and it may provide a good starting point for further research. For additional reading, we recommend the references provided by the authors. Tanja Kostic Patrick Butaye Jacques Schrenzel

XV

List of Contributors Levente Bodrossy Austrian Research Centers GmbH Department of Bioresources/ Microbiology 2444 Seibersdorf Austria Martien P. Broekhuijsen TNO Defence, Security and Safety Prins Mauritz Laboratory Business Unit Biological and Chemical Protection PO Box 45 2280 AA Rijswijk The Netherlands Oguzhan Ça g layan Hacettepe University Chemical Engineering Department and Bioengineering Division Beytepe 06800 Ankara Turkey Yvan Charbonnier University Hospitals of Geneva Service of Infectious Diseases Genomic Research Laboratory Rue Micheli-du-Crest 24 1211 Geneva 14 Switzerland

Gökhan Demirel Hacettepe University Chemical Engineering Department and Bioengineering Division Beytepe 06800 Ankara Turkey Memed Duman Hacettepe University Chemical Engineering Department and Bioengineering Division Beytepe 06800 Ankara Turkey Kerstin I. Falk Swedish Institute for Infectious Disease Control Centre for Microbiological Preparedness Nobels väg 18 17182 Solna Sweden and Karolinska Institutet Department of Microbiology, Tumor and Cell Biology Nobels väg 16 17177 Stockholm Sweden

XVI

List of Contributors

Patrice Francois University Hospitals of Geneva Service of Infectious Diseases Genomic Research Laboratory Rue Micheli-du-Crest 24 1211 Geneva 14 Switzerland

David Hernandez University Hospitals of Geneva Service of Infectious Diseases Genomic Research Laboratory Rue Micheli-du-Crest 24 1211 Geneva 14 Switzerland

Dimitrios Frangoulidis Bundeswehr Institute of Microbiology Neuherbergstrasse 11 80937 Munich Germany

Jonathan Hibbs University Hospitals of Geneva Service of Infectious Diseases Genomic Research Laboratory Rue Micheli-du-Crest 24 1211 Geneva 14 Switzerland

David Fretin Veterinary and Agrochemical Research Centre Section of Bacteriology Rue Groeselenberg 99 1180 Bruxelles Belgium Joachim Frey University of Bern Institute of Veterinary Bacteriology Laenggass Strasse 122 3012 Bern Switzerland Bora Garipcan Chemical Engineering Department and Bioengineering Division Hacettepe University Beytepe 06800 Ankara Turkey Volker Heiser Chipron GmbH Eresburgerstrasse 22–23 12103 Berlin Germany

Antoine Huyghe University Hospitals of Geneva Service of Infectious Diseases Genomic Research Laboratory Rue Micheli-du-Crest 24 1211 Geneva 14 Switzerland and Swiss Institute of Bioinformatics Rue Michel-Servet, 1 1211 Geneva 4 Switzerland Karen Kempsell Heath Protection Agency Centre for Emergency Preparedness and Response Novel and Dangerous Pathogens Diagnostic Technologies Group Porton Down Salisbury SP4 0JG United Kingdom

List of Contributors

Jasper Kieboom TNO Defence, Security and Safety Business Unit Biological and Chemical Protection PO Box 45 2280 AA Rijswijk The Netherlands

Peter Nilsson KTH – Royal Institute of Technology School of Biotechnology Albanova University Center Valhallavägen 79 10691 Stockholm Sweden

Tanja Kostic Austrian Research Centers GmbH Department of Health and Environment Bacterial Ecology and Genomics 2444 Seibersdorf Austria

Henrik Nordström Swedish Institute for Infectious Disease Control Centre for Microbiological Preparedness Nobels väg 18 17182 Solna Sweden and Karolinska Institutet Department of Microbiology, Tumor and Cell Biology Nobels väg 16 17177 Stockholm Sweden

Olfert Landt TIB MOLBIOL GmbH Eresburgerstrasse 22–23 12103 Berlin Germany Åke Lundkvist Swedish Institute for Infectious Disease Control Centre for Microbiological Preparedness Nobels väg 18 17182 Solna Sweden and Karolinska Institutet Department of Microbiology, Tumor and Cell Biology Nobels väg 16 17177 Stockholm Sweden Hermann Meyer Bundeswehr Institute of Microbiology Neuherbergstrasse 11 80937 Munich Germany

Vincent Perreten University of Bern Institute of Veterinary Bacteriology Laenggass Strasse 122 3012 Bern Switzerland Erhan Pis¸kin Hacettepe University Chemical Engineering Department and Bioengineering Division Beytepe 06800 Ankara Turkey

XVII

XVIII

List of Contributors

Jacques Schrenzel University Hospitals of Geneva Service of Infectious Diseases Genomic Research Laboratory and Clinical Microbiology Laboratory Rue Micheli-du-Crest 24 1211 Geneva 14 Switzerland

Richard Vipond Heath Protection Agency Centre for Emergency Preparedness and Response Diagnostic Technologies Group Novel and Dangerous Pathogens Porton Down Salisbury SP4 0JG United Kingdom

Sonal Shah Heath Protection Agency Centre for Emergency Preparedness and Response Diagnostic Technologies Group Novel and Dangerous Pathogens Porton Down Salisbury SP4 0JG United Kingdom

Pieter Vos Check-Points BV Agrobusiness Park 90 6708 PW Wageningen The Netherlands

Nigel Silman Heath Protection Agency Centre for Emergency Preparedness and Response Diagnostic Technologies Group Novel and Dangerous Pathogens Porton Down Salisbury SP4 0JG United Kingdom Susanna Sherwin Heath Protection Agency Centre for Emergency Preparedness and Response Diagnostic Technologies Group Novel and Dangerous Pathogens Porton Down Salisbury SP4 0JG United Kingdom

Ingrid Voskamp TNO Defence, Security and Safety Business Unit Biological and Chemical Protection PO Box 45 2280 AA Rijswijk The Netherlands Pierre Wattiau Veterinary and Agrochemical Research Centre Department of Bacteriology Groeselenberg 99 1180 Brussels Belgium

j1

1 Introduction to Microarray-Based Detection Methods Jacques Schrenzel, Tanja Kostic, Levente Bodrossy, and Patrice Francois

1.1 Introduction to Microarray Technology

Microarrays consist of an orderly arrangement of probes (oligonucleotides, DNA fragments, proteins, sugars or lectins) attached to a solid surface. The main advantages of microarray technology are high throughput, parallelism, miniaturization, speed and automation. Despite the fact that microarray analysis is a relatively novel technology, with the publication of the ﬁrst microarray studies in 1995 [101, 152], it is now broadly applied and the milestone of nearly 5000 published microarray papers was recorded in 2004 [80]. The scientiﬁc and technological background discussed here will be limited to DNA microarrays, excluding the new evolving ﬁelds of protein microarrays and glycomics [140]. Analogous to antigen–antibody interactions on immunoarrays, DNA microarrays rely on sequence complementarity of the two strands. They put in practice the fundamentals of complementary base-pairing (hybridization) that were ﬁrst described by Ed Southern [162]. In general, the strategy of microarray hybridization is reversed to that of a standard dot-blot, leading to recurring confusion in the nomenclature. Therefore, it has been suggested to describe tethered nucleic acid as the probe and free nucleic acid as the target [134]. Earlier studies on duplex melting and reformation, carried out on DNA solutions, have provided the basic knowledge about the reaction kinetics. Furthermore, a method for computational determination of the melting point as a function of nucleic acid composition and salt concentration was established [6, 148, 149]. Much of the pioneering work can be linked to the use of nitrocellulose membranes [69], dot-blots [84] and Southern blots [162]. Development of cDNA or oligonucleotide arrays was possible by combined innovations in micro-engineering, molecular biology [33, 120, 123, 156, 163, 164] and bioinformatics [59]. The real breakthrough in microarray technology was initiated by two key innovations: the use of non-porous solid supports (such as glass and silicon) and the development of methods for highdensity synthesis of oligonucleotides directly onto the microarray surface [59].

j 1 Introduction to Microarray-Based Detection Methods

2

Microarrays have demonstrated their applicability for a broad variety of applications such as cell differentiation [170], whole-genome expression analysis [67, 152], cancer research [5, 71, 131], comparative genome hybridization (CGH) [146], drug discovery [39], vaccine development [72] or single nucleotide polymorphism (SNP) analysis [56]. Technology transfer to diagnostic applications is therefore very appealing [3, 112, 196]. Reports have already shown that oligonucleotide microarrays can enable bacterial detection utilizing conserved bacterial genes [158], species identiﬁcation [189] as well as genotyping of bacterial pathogens, by using large sets of discriminative epidemiological markers [32, 36, 70, 183, 198]. Detection of genetically encoded virulence or antimicrobial resistance determinants [17, 36, 112, 176] may also afford a major beneﬁt for selection of an adequate chemotherapy. Current limitations for routine implementation of microarrays to detect DNA signatures are, initially, their high manufacturing costs and the requirement for large amounts of nucleic acid targets. Availability of large amounts of nucleic acid targets requires either a large volume of bacterial culture (biological ampliﬁcation) or target ampliﬁcation [122, 139]. Additional technical achievements in signal ampliﬁcation methods [60] and novel optical techniques [57, 128] can already improve target detection to the femtomole range. The development of microarray-based bacterial identiﬁcation starting directly from the biological sample, without any enzymatic target ampliﬁcation, is an important objective [168]. This procedure would signiﬁcantly reduce the turnaround time and overcome enzymatic-induced signal alterations or biases [16, 197].

1.2 Technical Aspects of Microarray Technology

The development of a new microarray platform requires consideration of many different features, most of them being co-dependent (see Figure 1.1). Different approaches have been reported, each of them exhibiting certain advantages and limitations. A summary of the most important technological features will be given to introduce the speciﬁc platforms in association with the intended experimental applications. 1.2.1 Probes

The nature of the probe used is related to the experimental question. In general it can be distinguished between genome fragments, polymerase chain reaction (PCR) products and oligonucleotide probes. Application potential as well as advantages and limitations of each probe type will be described brieﬂy. An additional summary of different probe types is presented in Table 1.1 (see p. 4).

1.2 Technical Aspects of Microarray Technology

Figure 1.1 Elements of the microarray experiment (adapted from [204]). Starting (clockwise) from a biological question, the process leads ultimately to data analysis and new knowledge, enabling us to address and refine the biological question.

1.2.1.1 Genome Fragments The use of entire bacterial or community genomes (suitably fragmented) as probes was ﬁrst employed for the reverse sample genome probing technique [184]. The same principle was later applied for community genome arrays [10, 124]. The major problem related to such microarray platforms is the huge complexity of the system. 1.2.1.2 PCR Products PCR products used as probes for microarray fabrication are mostly ampliﬁed inserts of the clone libraries. Different types of clone libraries can be used as a template for microarray fabrication (cDNA libraries, suppression subtractive hybridization (SSH) libraries, shotgun libraries, open reading frame (ORF) libraries, etc.). Microarrays utilizing PCR products are used for gene expression analysis [68, 97, 124, 152, 167, 199]. Furthermore, PCR products can be used as probes for the functional gene arrays [31, 191]. The inappropriate labeling of a substantial fraction of the PCR products (1–5%) can lead to poorly controlled microarrays, even when originating from prestigious research centers or commercial entities [89]. The IMAGE consortium (Integrated Molecular Analysis of Genomes and their Expression) revealed that, after resequencing, only 62% of 1189 cDNAs were pure and correct. 1.2.1.3 Oligonucleotide Probes Oligonucleotide microarrays provide a ﬂexible design, and are considered more reliable in terms of sensitivity and speciﬁcity [15, 18, 92, 94]. Differing from the previously described PCR probes, oligonucleotide probes are typically designed with a predeﬁned speciﬁcity. Generic or universal oligonucleotide probes form a distinct group of oligonucleotide probes that are predeﬁned only by length (and not by speciﬁcity). In this case a probe set is designed to contain all possible

j3

Length

not known (fragmented genome)

200–1500 bp

50–100 nt

15–30 nt

15–30 nt

Probe

Genome fragments

PCR products

Long oligonucleotides

Short oligonucleotides

Generic probes

yes

in silico probe design no

yes

in silico probe design

generic probes

no

no

Predefined specificity

clone library (shotgun, cDNA, SSH)

entire bacterial or community genome

Origin

Table 1.1 Summary of the probes used for DNA microarrays.

entire genomes

phylogenetic or functional

functional

functional

entire genome(s)

Targeted genes

many

one (few)

many

many

many

No. of targeted genes

species–strain

genus–species

genus–species

genus–species

Taxonomic resolution

(re)sequencing arrays, diagnostic arrays

diagnostic arrays

functional analysis, diagnostic arrays

gene expression, functional gene arrays

community genome arrays

Application

[43–46, 100, 115, 169]

[21, 78, 110, 155]

[185, 186]

[31, 68, 97, 124, 152, 167, 191, 199]

[10, 124]

References

4

j 1 Introduction to Microarray-Based Detection Methods

1.2 Technical Aspects of Microarray Technology

sequence permutations of a given length. Generic arrays have been used to study large-scale hybridization behavior [163] or for solid-state nucleic acid sequencing [43–46, 100, 115, 169]. Two main features inﬂuence probe speciﬁcity: probe length and the degree of conservation of the marker gene. In general, probe design is carried out in silico using different software tools (e.g. ARB [113] or OligoCheck [35]) and is based on a sequence database of the targeted marker gene. The extent and quality of the sequence database has a major effect on probe quality. The general criteria that need to be considered during probe design are the required probe speciﬁcity and uniformity of the probe set regarding hybridization behavior. In silico approaches allow for partial prediction of the hybridization behavior of designed probes. However, it has become clear that the simple notion that short oligonucleotides with a mismatch (MM) should hybridize less efﬁciently than perfect-match (PM) probes is not always applicable [137]. It has been demonstrated that the hybridization intensity of MM probes can depend on the nucleotide type (i.e. A, C, G or T) and position of the MM relative to the termini [178, 179], and that some MM probes yield even higher signal intensities with the target than those of corresponding PM probes [125]. Deﬁned mismatches are more difﬁcult to detect (e.g. GG or GT), as well as mismatches that reside in a context (GC-rich domain). Even well-designed probes can display differences in maximal hybridization capacity of 2 orders of magnitude under different hybridization conditions [22] and thus it is difﬁcult to ﬁnd one set of conditions that is optimal for all probes on an array [85, 178]. Factors affecting duplex formation on DNA microarrays include probe density, microarray surface composition, spacer length and the stabilities of oligonucleotide– target duplexes, intra- and intermolecular self-structures, and RNA secondary structures [75, 121, 133]. More generally, there is a lack of a simple relationship between hybridizations of probe–target duplexes as inferred from signal intensity values and in silico predictions based on Gibbs free energies [137]. This does not apply as strictly for highdensity microarrays where the high level of redundancy accounts for the speciﬁcity of the signals nor for long oligonucleotide microarrays where inter-allele distinction is not required. In any case, a thorough wet-lab validation with a set of reference strains or clones is warranted, as for the implementation of any other molecular tool [35]. Long Oligonucleotides The main advantages of long oligonucleotides (over 50 nucleotides in length) are high target binding capacity and irreversible hybridization kinetics. These features allow for enhanced detection sensitivity. However, the threshold for the differentiation is at 85–90% sequence similarity, resulting in reduced speciﬁcity. This can be compensated by the host speciﬁcity of the targeted genes. Due to their high sensitivity, long oligonucleotide microarrays are typically used in combination with universal ampliﬁcation techniques or without any ampliﬁcation, allowing the researcher to target an unlimited number of different genes [174]. Long oligonucleotide microarrays have been used for the detection of viruses [186] and pathogens [185].

j5

j 1 Introduction to Microarray-Based Detection Methods

6

Figure 1.2 Comparison of long versus short oligonucleotide probes.

Short Oligonucleotides Short oligonucleotide (15–30mer) microarrays are more precise at the detection of shorter nucleotide polymorphisms, including single nucleotide differences under optimized hybridization conditions. On the downside, short oligoarrays frequently require a larger number of probes for reliable diagnostics. Reversible hybridization kinetics and lower target binding capacity (in comparison to long oligonucleotides) are responsible for somewhat limited detection sensitivity (Figure 1.2). Short oligonucleotide are widely used for both environmental and diagnostic microbial diagnostic microarrays (MDMs) [21, 78, 110, 155]. 1.2.2 Substrates for Printing

The choice of the substrate for microarray printing depends primarily on the nature of the probes. An important factor to be taken into consideration is the effect of steric hindrance on the hybridization efﬁciency [21]. This may be a considerable problem in the case of short oligonucleotide probes and therefore such probes are generally appended to spacer molecules. Further important parameters to be considered during fabrication of the microarray are the probe concentration, spotting buffer and surface blocking strategies. Most of these features have been discussed thoroughly in the literature [79, 98, 172, 200]. The most widely used microarray format is a planar glass slide (1 3 in.). Slides for microarray printing are usually coated with different active surfaces that facilitate deposition of nucleic acids. An overview of the more commonly used substrates and their applications will be given below. A list of commercially available surface chemistries is given in Table 1.2.

Clontech Bioslide Technologies Asper Biotech Memorec Full Moon BioSystems Greiner Bio-One PerkinElmer Life Sciences

www.matrixtechcorp.com www.eriesci.com www.eriemicroarray.com www.clontech.com www.kaker.com www.asperbio.com www.api.com www.fullmoonbio.com/ www.greinerbioone.com www.perkinelmer.com

www.arcs.ac.at www.schott.com www.peqlab.de www.cel-1.com www.arrayit.com www.nuncbrand.com www.corning.com www5.gelifesciences.com

ARCS Schott

CEL Associates TeleChem Nunc Corning GE Healthcare (Amersham BioSciences) Matrix Technologies Erie Scientiﬁc

Web

Company

ü

ü? ü

ü

ü

ü

ü

ü

ü ü ü

ü

ü ü ü ü ü ü

CHO

Epoxy

ü

ü

Plastic

ü

ü

ü ü ü

ü

Plain glass

Table 1.2 Slides of different surface chemistries (reproduced with kind permission of Dr Preininger).

ü

ü ü

ü

ü

Other

ü ü ü

ü

ü ü

ü ü

Hydrogel

ü ü

ü

Photoactivatable

ü ü ü ü ü

ü

ü

NCS

ü ü ü ü

ü

NH2

1.2 Technical Aspects of Microarray Technology

j7

www.sciencestuff.com www.emsdiasum.com www.scienion.de http://www.zinsser-analytic.com www.noabdiagnostics.com www.chimera-biotec.com

www.sigmaaldrich.com www.whatman.com www.arraying.com www.XanTec.com www.xenopore.com www.biocat.de www.u-vision-biotech.com www.advalytix.de www.genetix.com www.cynmar.com www.genomicsolutions.com

Sigma-Aldrich Whatman (Schleicher & Schuell) Xantec Bioanalytics Xenopore U-Vision Biotech, Biocat

Advalytix Genetix Cymar Genomic Solutions (GeneMachines) Science Stuff Rinzl Scienion Geneworx NoAb BioDiscoveries Chimera Biotec

Web

Company

Table 1.2 (Continued)

ü

ü ü

ü ü

Plastic

ü

ü

Plain glass

ü ü ü

ü ü

ü

ü ü

ü ü ü

ü

CHO

ü ü ü

Epoxy

ü

ü

ü ü

ü ü

ü

NH2

NCS

Photoactivatable

ü

ü

Hydrogel

ü ü

ü

ü

ü

ü

Other

8

j 1 Introduction to Microarray-Based Detection Methods

1.2 Technical Aspects of Microarray Technology

1.2.2.1 Slides with Poly-L-lysine Coating The binding of DNA fragments to poly-L-lysine involves charge interactions that can be converted to covalent bonding by baking or ultraviolet (UV) cross-linking. Advantages of poly-L-lysine surfaces are low background and good signal intensities. The main disadvantage is low temperature resistance that can lead to the damage of the surface during denaturation or hybridization at higher temperature. PolyL-lysine-coated slides have been successfully used for the binding of PCR products [41, 152, 199] and short oligonucleotides [172]. 1.2.2.2 Slides with Amino Silane Coating Amino silane surface chemistry allows for electrostatic interactions between the amino groups of the silane (positively charged at neutral pH) and negatively charged phosphodiester backbone of the DNA. This interaction can be additionally stabilized by UV cross-linking. Amino silane coating demonstrates enhanced resistance towards high temperatures. However, somewhat higher background signals may occur when low-quality coatings are used. Substrates based on amino silane surface chemistry are widely used for deposition of PCR products [40, 68, 141, 145, 167, 175]. 1.2.2.3 Slides with Aldehyde Coating Covalent binding between aldehyde groups and DNA fragments can be facilitated, either through a 50 -amino linker on chemically modiﬁed DNA fragments or through aromatic amines of nucleotides. Probes with a 50 -amino group bind more efﬁciently than native DNA fragments. Furthermore, coupling via a 50 -amino group is directional, allowing for the deﬁned orientation of the probes on the microarray. In general, slides with aldehyde coating are characterized by high binding capacity and low background. Aldehyde-coated slides are mostly used for short oligonucleotide microarrays [21, 109]. 1.2.2.4 Slides with Epoxy Coating Similarly to aldehyde-coated slides, epoxy-coated substrates also allow for the covalent binding utilizing amino groups of the DNA fragments. Epoxy-coated slides have been used for the deposition of PCR products [31], and short [11] and long [187] oligonucleotides. 1.2.2.5 Proprietary Surface Chemistries Finally, it is worth mentioning that the array surface of various commercial microarrays can display very different properties (e.g. a highly hydrophobic surface on the Agilent SurePrint technology (www.agilent.com) due to proprietary surface chemistry). This leads to the use of speciﬁc hybridization and washing buffers adapted to the surface chemistry. 1.2.2.6 Probe Spacers To reduce the effect of steric interference (steric hindrance and surface electrostatic forces) [156, 180] on hybridization of targets to planar surfaces (e.g. glass and silicon), spacer molecules with a length of more than 50 A can be used to physically separate

j9

j 1 Introduction to Microarray-Based Detection Methods

10

the probes from the microchip surface [130, 156]. These are typically C6–C12 alkane spacers and/or 5–15 thymidine or adenine residues added to the tethered end of the oligo probe [7, 24, 61, 201]. 1.2.3 Targets for Microarray Analysis

Targets used for microarray analysis are typically ﬂuorescently labeled nucleic acid derivatives. Two basic types of targets can be distinguished: those derived from RNA and those derived from DNA templates. Selection of nucleic acids used as a template for target preparation is primarily dependent on the experimental question. Gene expression studies use mRNA-based targets, whereas microbial diagnostic arrays employ primarily DNA-based targets or rRNA targets that are more abundant. The parallel analysis of mRNA- and DNA-based targets provides a complex picture correlating presence and activity [23]. 1.2.3.1 Target Amplifications and Sensitivity Issues In general, targets used for short oligonucleotide diagnostic microarrays are previously PCR-ampliﬁed. PCR ampliﬁcation ensures enrichment of the targeted gene(s) and therefore increases the sensitivity of microarray detection, but also introduces an inherent PCR bias [185]. Long oligonucleotide probes exhibit higher target binding capacity and therefore allow hybridization with highly complex target mixes (i.e. unampliﬁed environmental DNA, a native mixture of mRNA from an organism or the products of universal, whole-genome ampliﬁcation methods). The main advantage of the latter targets is that they represent the entire gene pool to be studied, without reduction of its complexity. Thus, the potential of DNA microarray-based microbial screening and diagnostic technologies is currently limited by front-end target-speciﬁc nucleic acid detection. The presence of an ubiquitous poly-adenylated tail at the 30 -end of eukaryotic messenger RNAs offers the possibility to convert minute amounts of RNA to micrograms of labeled material, with minimal effects on the respective abundance of the mRNA mixture [8, 139]. Prokaryotic RNAs are not poly-adenylated and are thus more challenging to work with when starting material is scarce. In such cases, the use of generic primers able to amplify parts of the targeted gene is often required [7, 144], but the universality of such primers is questionable (false-negative signals are not rare [7]). Other options include intact or even degraded RNA ampliﬁcation using T3-coupled random primers [193] or a limited set of genome-derived cognate primers [205]. The interested reader is referred to two recent publications that have analyzed and validated different target ampliﬁcation strategies before array hybridization [62, 185]. Vora et al. investigated four front-end ampliﬁcation strategies: random primed, isothermal Klenow fragment-based, F29 DNA polymerase-based and multiplex PCR. Their results underscore the feasibility of using random ampliﬁcation approaches and begin to systematically address the versatility of these approaches for unbiased pathogen detection from environmental sources [165]. Francois et al. compared

1.2 Technical Aspects of Microarray Technology

commercially available ampliﬁcation methods such as MessageAmp or GenomiPhi. They showed that this type of enzyme represents an interesting alternative of moderate cost for transcriptomic studies. Such ampliﬁcations permitted the authors to obtain signiﬁcant amounts of nucleic acids, sufﬁcient to perform microarray studies even when starting with a few tens of nanograms of material. Importantly, these methods showed exquisite reproducibility, even considering data before normalization, which is the major requirement for their utilization in transcriptomic studies [188]. Finally, these nucleic ampliﬁcation methods can be coupled to signal ampliﬁcation (see, e.g. [27]) and/or array-based methods for improving detection sensitivity. 1.2.3.2 Labeling of the Targets Fluorescently labeled targets are in general prepared using one of the many commercially available kits [114] or following standardized labeling protocols [21]. Incorporation of the ﬂuorescently labeled nucleotides occurs during enzymatic ampliﬁcation of the nucleic acids (e.g. PCR ampliﬁcation, in vitro transcription, reverse transcription, random DNA ampliﬁcation). Alternatively, modiﬁed nucleotides (i.e. amino-allyl nucleotides) can be incorporated in the target followed by subsequent coupling with ﬂuorescent dye esters. 1.2.3.3 Hybridization and Wash Conditions Hybridization speciﬁcity is of paramount importance, especially when one has to differentiate targets from non-targets or to discriminate closely related DNA or RNA sequences that may possibly differ by only a single base pair. Probes on the microarray are subjected to the same hybridization and washing procedures (e.g. buffers, salt concentrations, and temperature). Strategies to overcome problems arising thereof include the acquisition of melting curves for every individual probe [104]; the careful design of probes with similar predicted hybridization properties (usually combined with the application of two to three probes per targeted group) [24, 147, 201]; the addition of tetramethylammonium chloride that equalizes the melting temperature of different probes by stabilizing the AT base pairs; composition of the hybridization solution [120]; or the use of highly redundant probe sets with multiple probes to target each speciﬁc group of microorganisms [189]. Secondary structure formation within the targets can reduce the binding constant of a speciﬁc probe by as much as 105–106 times [96], leading to an increase in falsenegative signals and a decrease in hybridization speciﬁcity [9, 164]. Several methods have been suggested to alleviate this problem, such as the use of helper oligonucleotides [130] and a two-probe proximal chaperon detection system [158] to mitigate the effects of target secondary structure hindrances, an appropriate labeling method [63], and a protocol to achieve optimized target lengths [126, 164, 197]. Since long targets can form secondary and tertiary structures that hinder efﬁcient probe–target duplex formation, the sizes of the target molecule and its amplicon are often reduced via chemical, enzymatic or thermal fragmentation methods [24, 87, 104, 126, 138, 158]. Liu et al. have recently elegantly reviewed these issues and experimentally demonstrated that microarray hybridizations with short rRNA fragments were more

j11

j 1 Introduction to Microarray-Based Detection Methods

12

dependent on target sequence than on the competition between probe–target interaction and RNA self-folding [103]. Hybridization with short gene fragments increases the potential for the accumulation of background signal from non-speciﬁc hybridization events. In order to circumvent this negative target effect an alternative protocol – sequence-speciﬁc end-labeling of oligonucleotides – was developed (see Chapter 3; Kostic et al., [206] and Ref. [145]). In this approach, targets are complementary to oligonucleotide probes on the array, and the labeling is performed by incorporating singlelabeled dideoxynucleotide in the presence of the targeted PCR product. This methods ensures both high speciﬁcity and sensitivity; however, it is still affected by PCR bias. 1.2.4 Classical Commercially Available Microarray Formats

All platforms share the common attribute that a sensor detects a signal from target sequences hybridized to immobilized nucleotidic probes. The intensity of this signal provides a measure of the amount of bound nucleic acid from a sample [137]. Schematically, we have divided this section into spotted and in situ synthesized arrays. Section 1.2.5 focuses on alternative platforms that provide improved detection sensitivities. 1.2.4.1 Spotting Approaches Currently, up to 50 000 gene fragments or oligonucleotides can be spotted onto a single microscope slide using robotic technology. The advantages of this technology are ﬂexibility in the design of the array, the relative ease of production and its relatively low cost. Multiple identical microarrays can be robotically printed in batches of over 100 in a single run. Most of the cost in printing such arrays is associated with the synthesis of oligonucleotide probes or primer pairs required for the ampliﬁcation of the probe gene fragments [42]. Here, we brieﬂy review various commercially available microarray formats. Operon The Qiagen Operon format (www.operon.com) uses optimized 70mer oligonucleotides to represent each gene in a given genome. Each 70mer probe is designed to have optimal speciﬁcity for its target gene and is melting-temperature normalized. This approach provides a reduction in cross-hybridization and an increase in the differentiation of overlapping genes or highly homologous regions. Theoretically, mutant alleles could be detected using such oligonucleotide microarrays owing to the shorter probe size compared to PCR product-based microarrays [42]. 1.2.4.2 In Situ Synthesis In situ synthesis allows higher yields, lower chip-to-chip variation, as well as higher probe densities. These methods also permit the manufacture of true random access arrays [164]. Manufacturing techniques include photolithographic masks to control

1.2 Technical Aspects of Microarray Technology

chemical activation by photodeprotection steps [59, 100], ink-jet deposition [81, 166], as well as physical barriers allowing for sequential ﬂooding of precursors [120]. Affymetrix Oligonucleotide probes are not deposited on Affymetrix microarrays (www.affymetrix.com), but are directly synthesized on the surface. The company has coupled photochemical deprotection to solid-phase DNA synthesis by adapting techniques from the semiconductor industry [100, 105, 129]. The main advantage of this approach is very high probe density (over 500 000 probes can be deposited on a surface of 1.6 cm2). Limitations are high price, low ﬂexibility and lack of properly validated probe sets. Therefore, in order to ensure the speciﬁcity of detection, applications of the Affymetrix platform require multiple probes to monitor single targets, relying on empirical algorithms [30]. NimbleGen Recent technical developments, such as Nimblegens micromirror device (www.nimblegen.com), facilitate maskless photoreactive synthesis of oligonucleotide probes, and currently permit the simultaneous deposition and analysis of as many as nearly 800 000 probes on one array platform [4]. Such probe density now permits detailed CGH for detecting small deletion changes in the studied genome. However, insertions of genes compared with the sequenced reference strain cannot be detected by CGH DNA microarray analyses. This problem can be alleviated by adding non-redundant ampliﬁed sequences from several closely related bacteria to the array, once new genetic information is available [26, 27, 34, 135]. Agilent A more versatile, but still essentially mechanical, method for producing DNA arrays is to use the print heads out of commercial piezoelectric ink-jet printers to deliver reagents to individual spots on the array [13, 19, 161]. A piezoelectric ink-jet head consists of a small reservoir with an inlet port and a nozzle at the other end. When a voltage is applied to the crystal, it contracts laterally thus deﬂecting the diaphragm and ejecting a small drop of ﬂuid from the nozzle. Such devices are inexpensive and can deliver drops with volumes of tens of picoliters at rates of thousands of drops per second. In conjunction with a computer-controlled XY stepping stage to position the array with respect to the ink-jet nozzles, it is possible to deliver different reagents to different spots on the array. Arrays of approximately 250 000 spots can be addressed in a few minutes with each spot receiving one drop of reagent. Agilent has developed a ﬂexible method for microarray production, centered around an in situ oligonucleotide synthesis method in which the ink-jet printing process is modiﬁed to accommodate delivery of phosphoramidites to directed locations on a glass surface [19]. Achieving high density with the ink-jet approach requires one more trick. Two drops of liquid applied too closely together on a surface will tend to spread into each other and mix. For 40-pl drops the minimal center-to-center spacing is about 600 mm. This limits the array density achievable with the ink-jet method. One way around this is to engineer patterns in the surface chemistry of the array to produce spots of a relatively hydrophilic character surrounded by hydrophobic barriers [19, 161]. Design ﬂexibility and high densities constitute the two major advantages of this technique that can generate arrays at moderate costs.

j13

j 1 Introduction to Microarray-Based Detection Methods

14

CombiMatrix CombiMatrixs technology (www.combimatrix.com) is a specially modiﬁed semiconductor adapted for biological applications. These integrated circuits contain arrays of microelectrodes that are individually addressable using embedded logic circuitry on the chip. Placed in a specially designed ﬂuidic chamber, the chip digitally directs the molecular assembly of biopolymers in response to a digital command. Under a controlled process, each microelectrode is addressed to selectively generate chemical reagents by means of an electrochemical reaction. These chemical reagents facilitate the in situ synthesis of complex molecules such as DNA oligonucleotides. The parallel process drastically reduces the cost and time of synthesizing hundreds or thousands of different molecules. Currently, this technology is able to produce arrays with approximately 45000 features. 1.2.5 Alternative Methods for Improving Microarray-Based Detection Sensitivity

Most microarray applications are limited by the starting amounts of nucleic acids to be studied. In other words, detection sensitivity is a major limitation of microarraybased approaches that has to be compensated by several enzymatic steps for target ampliﬁcation and/or labeling. The followings sections illustrate various array-based methods that can also improve detection sensitivity, independently from any target or signal ampliﬁcation. 1.2.5.1 Resonance-Light Scattering (RLS) New optical techniques are now available for microarray detection [57, 128] that provide sensitivities high enough to detect femtomolar amounts of targets. Francois et al. [60] nicely illustrated the improvements in detection sensitivity that can be achieved with different optical detection methods when using direct non-enzymatic labeling of bacterial nucleic acids. Microarrays detected by RLS (Genicon, Invitrogen, Carlsbad, CA, USA) offer short turnaround times and exquisite sensitivity. Interestingly, the labeling and detection schemes offer an affordable alternative at a reasonable cost to the expensive ﬂuorescence-based methods. The principle of RLS is the following: when a suspension of nano-sized gold or silver particles is illuminated with a ﬁne beam of white light, the scattered light has a clear (not cloudy) color that depends on composition and particle size. This scattered light can be used as the signal for ultrasensitive analyte detection [127]. 1.2.5.2 Planar-Waveguide Technology (PWT) Figure 1.3 (see p. 16) depicts PWT technology-based microarrays. A 150-nm to 300-nm thin metallic oxide ﬁlm (green) with a high refractive index (e.g. Ta2O5 or TiO2) is deposited on a transparent support (grey) with lower refractive index (e.g. glass or polymer). A parallel laser light beam (red) is coupled into the wave-guiding ﬁlm by a diffractive grating that is etched into the substrate. The light propagates within this ﬁlm and creates a strong evanescent ﬁeld perpendicular to the direction of laser propagation into the adjacent medium [48]. The ﬁeld strength decays exponentially

1.2 Technical Aspects of Microarray Technology

with the distance from the waveguide surface and its penetration depth is limited to about 400 nm (large orange arrow). This effect results in the selective excitation of ﬂuorophore molecules located at or near the surface of the waveguide (red circles). For microarray applications, speciﬁc capture probes or recognition elements are immobilized on the waveguide surface. Upon ﬂuorescence excitation by the evanescent ﬁeld, excitation and detection of ﬂuorophores by a CCD camera is restricted to the sensing surface, whilst signals from unbound molecules in the bulk solution (blue) are not detected. This yields a signiﬁcant increase in the signal/noise ratio compared to conventional optical detection methods [61]. PWT microarrays are also commercially available (e.g. Zeptosens; www.zeptosens.com). 1.2.5.3 Liquid Arrays The Luminex (www.luminexcorp.com) suspension array is simply a transfer of the microarray format from a glass slide to a high-throughput and efﬁcient bead format. With this type of assay, the DNA probes (e.g. oligonucleotides) are attached to 5.6-nm polystyrene microspheres (beads) containing an internal ﬂuorescent dye. Each probe is assigned to a particular bead set containing a unique mixture of ﬂuorescent dyes, or spectral address. Bead sets coupled to the probes of interest are then mixed together in the wells of a 96-well microtiter plate, allowing many different probes to be analyzed simultaneously. Target DNA molecules are labeled with a different and spectrally distinct ﬂuorescent dye and hybridized to the probes on the beads. Beads with the hybridized targets are then separated and quantiﬁed using a two-laser ﬂow cytometer. The unique internal color of the bead is read by one laser and serves to identify which probe is present on the bead. The second laser measures the ﬂuorescent signal of the reporter dye present on the labeled target DNA and allows one to assess the strength of the hybridization between the target DNA and the probe. As this technology allows up to 100 different probes to be analyzed in a single well of a 96-well plate, it promises to make microarray subtyping faster and less expensive. The established suspension array protocol requires that relatively short PCR products be used as targets [47]. Microsphere-based ﬁber-optic arrays (Illumina; www.illumina.com) provide many advantages over other array-based methods [2]: higher sensor packing density, smaller assay sample volumes, increased array reusability, ﬂexible array design, and reduced false-positives and false-negatives [50]. Previous work has demonstrated that the microsphere-based ﬁber-optic array can detect as few as 600 target DNA molecules and is sensitive enough to discriminate a single-base mismatch from a perfect match [49, 51]. 1.2.5.4 Three-Dimensional Microarray Formats Three-dimensional microarray formats offer the option to record hybridization and dissociation events in real time. This enables rapid establishment of the melting curves for all probes on the microarray, facilitating development of validated probe sets. Three-dimensional microarray systems include gel-pads [73, 104, 136], ﬂow-through systems such as PamGene (www.pamgene.com) [192] or MetriGenix (www.metrigenix.com) [88].

j15

j 1 Introduction to Microarray-Based Detection Methods

16

Figure 1.3 Schematic representation of PWT technology-based microarrays (from [61]).

1.2.6 Marker Genes Used on MDMs

Microarrays employing long oligonucleotide probes or gene fragments can target an unlimited number of different genes. Short oligoarrays, on the other hand, depend on PCR ampliﬁcation to reduce target complexity to a level compatible with the sensitivity of the probes. They are thus usually limited to a small number of marker genes (typically between one and 10). Marker genes used for phylogenetic analysis and the development of short oligonucleotide MDMs need to fulﬁll several criteria: (1) widespread distribution throughout the targeted organism group, (2) high degree of conservation allowing for universal PCR ampliﬁcation, (3) existence of variable regions allowing for the design of discriminating probes and (4) none or low rate of horizontal gene transfer. The most commonly employed phylogenetic marker for the detection of microorganisms is the 16S rRNA gene. rRNAs are particularly suitable for species identiﬁcation procedures as they occur universally, contain conserved as well as divergent regions and are highly abundant in bacterial cells. A further advantage of the 16S rRNA gene consists in the availability of large sequence and probe databases (www.arb-home.de, rdp.cme.msu.edu, greengenes.lbl.gov, http://www.microbialecology.net/probebase). The main limitation of the 16S rRNA gene lies in its extremely high degree of conservation. In many cases (a notable example being various genera of Enterobacteriaceae) it is not possible to design even species-speciﬁc probes based on it. Therefore, in order to augment discriminatory potential of short oligonucleotide microarrays, a scope of alternative, phylogenetic and functional marker genes has been suggested [108, 150]. These include the 23S rRNA gene, the rRNA intragenic

1.3 Analysis and Quality Control Aspects

spacer region, so-called house-keeping genes (e.g. gyrB, rpoB, recA, atpD, groEL), virulence genes, antibiotic resistance genes and functional genes (e.g. pmoA, amoA, nifH, nirK, nirS). Many of these have been successfully applied on various microarray platforms [21, 86, 171, 191]. The major limitation of these alternative marker genes is the limited organism coverage of published sequence databases.

1.3 Analysis and Quality Control Aspects

Each step of microarray experiments needs to be optimized and validated, from array design and manufacture to data collection and analysis. Among critical technical parameters that need to be controlled are microarray surface chemistry, probe sequence, probe deposition process and hybridization conditions. The MicroArray Quality Control (MAQC) Consortium, an unprecedented, community-wide effort, spearheaded by US Food and Drug Administration scientists, recently addressed experimentally the key issues surrounding the reliability of DNA microarray data. They assessed the performance of seven microarray platforms in proﬁling the expression of two commercially available RNA sample types. Results were compared not only at different locations and between different microarray formats, but also in relation to three more traditional quantitative gene expression assays. The MAQC Consortiums main conclusions conﬁrm that, with careful experimental design and appropriate data transformation and analysis, microarray data can indeed be reproducible and comparable among different formats and laboratories, irrespective of sample labeling format. The data also demonstrate that fold change results from microarray experiments correlate closely with results from assays like quantitative reverse transcription PCR [157]. However, different experimental setups require different validation approaches. Crucial aspect is array-to-array normalization [93, 159, 195]. Various methods for normalization have been suggested. For gene expression studies one approach is to determine a set of invariant genes for normalization [151, 177]. Another approach recommends replicating genes on the array and using this within-array replication for normalization [53–55]. Standard normalization protocols rely on the assumption that the majority of genes on the microarray are not differentially expressed between samples [194]. Jaeger et al. suggest including additional normalization genes on the small diagnostic microarrays and propose two strategies for selecting them from genome-wide microarray studies: data driven univariate selection of normalization genes or multivariate selection based on ﬁnding a balanced diagnostic signature [82]. In the case of diagnostic short oligonucleotide microarrays signals are usually normalized against positive controls. These positive controls are designed for conserved regions targeted gene, for the PCR primers used to amplify the targeted gene or against exogenously spiked DNA. Long oligonucleotide arrays can be normalized against general probes, targeting conserved regions of universal genes present in all bacteria [i.e. universal 16S probe(s)] or other housekeeping genes.

j17

j 1 Introduction to Microarray-Based Detection Methods

18

Normalized signals are compared to arbitrary threshold values and the targeted microbes are rated as present or absent. For low-density short oligonucleotide arrays, the threshold values are ideally individual values, reﬂecting the hybridization potential of the individual probes. For long oligoarrays and high-density arrays with a highly redundant set of short oligonucleotides, it is possible to devise universal rules for calling a signal/set of signals present or absent.

1.4 Applications of Microarray Technology in Microbial Diagnostics

Microarray technology offers a great potential for answering many different experimental questions. The nature of the experimental question is the main issue that has to be taken into consideration when developing a new microarray platform. Depending on this question, there is an initial decision on the nature of probes and on the method for target labeling. This subsequently inﬂuences the selection of substrates and hybridization strategies. Some of the most common experimental questions for which microarray technology is used will be reviewed here. 1.4.1 Gene Expression Studies

The most widely used application of DNA microarrays is the study of transcriptional responses. Consequently, targets are derived from mRNA. Initial studies were limited to an organism or tissue of interest and provided insights in particular aspects of the organisms physiology [68, 152, 167, 199]. For these studies, probes representing the genetic proﬁle of organism or tissue of interest were used (e.g. clone libraries containing either genomic DNA or cDNA fragments). Recent developments of microarray technology enabled environmental monitoring of gene expression. Even though these studies are still limited to a few genes of interest, they can provide valuable information regarding the functionality of the whole microbial community [202]. 1.4.2 Comparative Genomic Hybridization (CGH)

Traditional phylogenetic classiﬁcation of bacteria to study evolutionary relatedness is based on the characterization of a limited number of genes, rRNA or signature sequences. However, owing to the acquisition of DNA through lateral gene transfer, the differences between closely related bacterial strains can be vast [42]. By contrast, whole-genome sequencing comparisons allow a multitude of genes to be compared. Unfortunately, whole-genome sequencing is currently too expensive to allow comparison of a large number of isolates of a species in a high-throughput scenario, as the global surveillance of infectious diseases requires. Therefore, as microbial genotyping is increasingly being used to track infectious diseases as they spread in human populations, another usage of microarrays has emerged. CGH permits assessing

1.4 Applications of Microarray Technology in Microbial Diagnostics

genetic similarities and differences between closely related organisms. This approach is an adaptation of array methods used in gene expression studies, but applied to total genomic DNA [97, 124]. CGH enables a birds-eye view of all the genes absent or present in a given genome compared to the reference genome on the microarray. Whole-genome comparisons typically identify sets of core genes, which are shared by all strains in a species, and accessory genes, which are present in one or more strains in a species and often result from gene acquisition. It is these differences that can often be used to identify genes and/or genetic islands related to gain-of-function traits in pathogenic strains [42]. CGH approaches can be applied to further characterize strains to identify novel marker genes and chromosomal regions speciﬁc for given groups of isolates, thus providing better discrimination and additional information compared to classical genotyping methods [90]. However, ambiguities in the interpretation of the ratios of hybridization and cross-hybridizations to paralogous genes remain important limitations of the technique [181]. Solid statistical criteria for the absence/ presence of ORFs are still lacking as a result of the diversity of the microarray design approaches, affecting the meta-analysis of the data obtained by different investigators [83]. 1.4.3 Generic or Universal Microarrays

Generic arrays have been proposed as an inexpensive alternative to sequencing. Using all possible combinations of an n-mer allows walking at every position along a nucleotide sequence. This approach is currently limited by the complexity of the algorithm required to generate contigs (conversion of a listing of hybridized n-mers into a meaningful sequence). Also, if the sequence undergoing analysis contains a repeat region (the same sequence appearing more than once within the target molecule), the reconstruction diagram will have to display a corresponding number of branching points, leading to an ambiguous sequence. This type of array is the only one capable of detecting sequences that are lacking in large electronic libraries. In contrast, dedicated arrays are used for repetitive sequencing (resequencing) of the same target for detection of nucleotide polymorphisms or functional mutations. Universal arrays refer to strategies that can provide target identiﬁcation without any a priori sequence knowledge. This approach has been named the non-cognate hybridization system (see Chapter 4) [154]. By synthesizing all probes of a given length that can be generated by a combination of four nucleotides, microarrays could detect any single organism. Unfortunately, for generating realistic probe lengths (i.e. permitting unique sequence identiﬁcation as well as adequate hybridization behavior), total number of permutations of the four nucleotides would yield to very large numbers of possible probes (i.e. 413 ¼ 67108864 probes for 13mer nucleotides). Synthesis of such large microarrays is currently technically impossible. Furthermore, bioinformatics tools that would be necessary for microarray analysis of billons of probes are not yet available. Thus, reducing the probe set and the complexity of the

j19

j 1 Introduction to Microarray-Based Detection Methods

20

analytical approach is warranted. Random reduction of the probe set to a smaller number of probes might result to the potential missing of sequences characteristic for given pathogens. An alternative would be the pruning of low informative probes, i.e. probes that have no targets in a collection of selected microorganisms. However, this approach yields ultimately to another cognate design, with the inherent limitations described above. Another way to reduce the probe set is to consider permutations of only two nucleotides. This divides the probe set by 2(n–2), where n is the length of the probes. This microarray design is truly non-cognate and guarantees that no organism is a priori favored or missed. In analogy, Roth et al. have designed a universal microarray system combined with an enzymatic manipulation step that is capable of generating expression proﬁles from any organism without requiring a priori sequence-speciﬁc knowledge of transcript sequences [143]. Finally, we should mention here multi-purpose arrays that contain probes for detecting a series of molecular barcodes. These molecular barcodes can be used as tags in various experimental conditions, for probing different targets under various experimental formats. These multi-purpose arrays serve only to quantitatively detect the presence of barcoded probes, whose speciﬁcity is determined by the user. Molecular inversion probe (MIP) technology was initially developed for the detection of SNPs in human genes [76]. MIP technology has been shown to work well for multiplexing, i.e. massive parallel processing (12000 MIPs in the same reaction tube) [77]. The power and versatility of MIP technology makes it perfectly suited for the identiﬁcation and quantiﬁcation of microbes. MIPs high sensitivity and speciﬁcity in detecting large numbers of SNPs [76, 77] should allow us to harness this technology to detect a large number of pathogens and to identify multiple infections in an individual sample. A MIP is comprised of genomic recognition sequences, common ampliﬁcation sequences and a molecular barcode for each genotype assigned to a speciﬁc gene. This probe is a linear oligonucleotide with targetcomplementary sequences at the ends and a non-complementary linking segment in between [173]. 1.4.4 Microarrays for Sequence Analysis

Generic or universal microarrays are only one of the many, technologically very diverse, microarray technologies available for sequence analysis. Another possibility is an microarray containing a range of oligonucleotides covering a DNA sequence of interest, employing the so-called tiling strategy [74]. Sequence analysis is performed by comparison of the hybridization patterns of the reference versus test sample. Such microarrays have been used for sequencing [74, 197], detection of SNPs [74, 98] and analysis of secondary structures [160]. Alternative detection methods have been suggested such as using a labeled common oligonucleotide primer that is extended to the site of the match or mismatch [106]. Publications have already started addressing the issue of automated interpretation of resequencing on microarrays [116].

1.4 Applications of Microarray Technology in Microbial Diagnostics

Recent advances in high-density oligonucleotide arrays have enabled the development of high-throughput resequencing techniques. Resequencing arrays are designed to cover the entire genome by overlapping oligonucleotides. Multiple versions of each oligonucleotide are spotted on the array to represent the four possible base combinations (A, T, G and C) for each nucleotide position. To date, this technique has been applied to Bacillus anthracis with 56 strains being resequenced using a custom-designed resequencing array [203]. The same technique has also been used to track the evolution of the severe acute respiratory syndrome coronavirus [190]. Conversely, similar tiling-based resequencing has been employed for Staphylococcus speciation [37], genotyping [182] or detection of mutations conferring resistance to quinolone antimicrobials [38]. Finally, other bead-based arrays have been suggested for high-throughput sequencing-based approaches [29]. The latter approach resulted in a powerful highthroughput sequencing platform (Illumina-Solexa; www.illumina.com) that currently competes against the pyrosequencing method described by Margulies et al. [118]. 1.4.5 Microbial Diagnostic Microarrays (MDMs)

MDMs are used for the simultaneous identiﬁcation of microorganisms in clinical or environmental samples. Probes used for MDMs are usually oligonucleotides designed to be speciﬁc for a given strain, subspecies, species, genus or higher taxon. Classiﬁcation and nomenclature of MDMs throughout the literature is not unanimous. According to their intended use, environmental MDMs [21, 110] and detection/ identiﬁcation MDMs can be distinguished [78, 154]. The main difference between these two MDM types lies in their detection requirements. Environmental MDMs are generally used to assess the whole microbial community structure or a subset of the microbial community in a particular environment. Therefore, reliable parallel detection of many different microorganisms and the potential for some level of quantiﬁcation are required. Detection/identiﬁcation MDMs are primarily used in clinical (medical, veterinary), food and biodefense microbiology. For this purpose, highly sensitive and speciﬁc detection of a few microorganisms in a complex community are required. According to the nature of the marker gene, one can further distinguish phylogenetic [30, 110, 145] and functional MDMs, also referred to as functional gene arrays [21, 171, 191]. There are two concepts for sensitivity of MDMs, both of them potentially posing a bottleneck to the detection of the targeted microbes. Absolute sensitivity refers to the amount of target DNA or number of target microbes required for successful detection. Absolute sensitivity reﬂects the hybridization capacity and detection sensitivity of the microarray platform used. Relative sensitivity, on the other hand, refers to the ratio of the targeted microbe within the entire microbial community analyzed. It is primarily due to low level non-speciﬁc background signal accumulation and to the fact that the amount of target DNA applicable in microarray hybridizations is limited.

j21

j 1 Introduction to Microarray-Based Detection Methods

22

1.5 Further Developments and New Perspectives Regarding Array Sensitivity and Specificity

A promising approach to increase the sensitivity of a microarray assay is tyramide signal ampliﬁcation [40]. Upon hybridization, this method relies on enzymatic ampliﬁcation of the signal by employing horseradish peroxidase-mediated deposition of ﬂuorochrome-labeled tyramides at the location of the probe. The relative sensitivity can be improved by limiting the labeling to very short, speciﬁc regions (Kostic et al. [206], and Refs. [14, 145]). High-density microarrays employing multiple PM/MM probe sets for each targeted microbe also enable a signiﬁcant improvement in relative sensitivity [30]. A novel method to analyze microarray data holds promise for a signiﬁcant improvement in terms of relative sensitivity of MDMs [117]. The ultimate speciﬁcity of microarray technology depends on the discrimination between a fully complementary target and a non-target differing in only one single nucleotide. Various enzyme-assisted hybridization strategies (also used in SNP and resequencing assays [52, 98]) are being applied because of their promise in strongly discriminating single mismatches located near the 30 -end of microarray probes [34, 75, 101, 119, 135, 138, 145]. Isotope microarrays represent further development of the traditional phylogenetic MDMs, enabling linking phylogeny (community structure) to function. This approach employs double-labeled targets, where the ﬁrst radioactive labeling is substrate-mediated, and the second labeling is performed according to standard microarray protocols [1]. 1.6 Conclusions

Development of microarray-based bacterial identiﬁcation systems starting directly from the biological sample, without any enzymatic target ampliﬁcation, would be most welcome [168]. During a single hybridization, arrays can integrate probes that provide microbial identiﬁcation but also enclose large sets of discriminative epidemiological markers [31, 31, 70, 182, 183, 198], or contain probes to detect virulence or antimicrobial resistance determinants [17, 36, 38, 91, 176]. The high parallelism of microarray systems appears particularity adapted provided their design includes the following: (1) targeted universal gene(s), (2) simpliﬁed coupling and labeling protocols, (3) exquisite sensitivity, and (4) adapted analysis strategy.

References 1 Adamczyk, J., Hesselsoe, M., Iversen, N., Horn, M., Lehner, A., Nielsen, P.H., Schloter, M., Roslev, P. and Wagner, M. (2003) The isotope array, a new tool that

employs substrate-mediated labeling of rRNA for determination of microbial community structure and function. Appl. Environ. Microbiol., 69, 6875–6887.

References 2 Ahn, H.J., La, H.J. and Forney, L.J. (2006) System for determining the relative ﬁtness of multiple bacterial populations without using selective markers. Appl. Environ. Microbiol., 72, 7383–7385. 3 Aitman, T.J. (2001) DNA microarrays in medical practice. BMJ, 323, 611–615. 4 Albert, T.J., Norton, J., Ott, M., Richmond, T., Nuwaysir, K., Nuwaysir, E.F., Stengele, K.P. and Green, R.D. (2003) Light-directed 50 ! 30 synthesis of complex oligonucleotide microarrays. Nucleic Acids Res., 31, e35. 5 Alizadeh, A.A., Eisen, M.B., Davis, R.E., Ma, C., Lossos, I.S., Rosenwald, A., Boldrick, J.C., Sabet, H., Tran, T., Yu, X., et al. (2000) Distinct types of diffuse large B-cell lymphoma identiﬁed by gene expression proﬁling. Nature, 403, 503–511. 6 Allawi, H.T. and SantaLucia, J. Jr, (1998) Nearest-neighbor thermodynamics of internal AC mismatches in DNA: sequence dependence and pH effects. Biochemistry, 37, 9435–9444. 7 Anthony, R.M., Brown, T.J. and French, G.L. (2000) Rapid diagnosis of bacteremia by universal ampliﬁcation of 23S ribosomal DNA followed by hybridization to an oligonucleotide array. J. Clin. Microbiol., 38, 781–788. 8 Aoyagi, K., Tatsuta, T., Nishigaki, M., Akimoto, S., Tanabe, C., Omoto, Y., Hayashi, S., Sakamoto, H., Sakamoto, M., Yoshida, T., et al. (2003) A faithful method for PCR-mediated global mRNA ampliﬁcation and its integration into microarray analysis on laser-captured cells. Biochem. Biophys. Res. Commun., 300, 915–920. 9 Armitage, B.A. (2003) The impact of nucleic acid secondary structure on PNA hybridization. Drug Discov. Today, 8, 222–228. 10 Bae, J.W., Rhee, S.K., Nam, Y.D. and Park, Y.H. (2005) Generation of subspecies level-speciﬁc microbial diagnostic microarrays using genes ampliﬁed from subtractive suppression

11

12

13

14

15

16

17

18

19

20

hybridization as microarray probes. Nucleic Acids Res., 33, e113. Bailly, X., Bena, G., Lenief, V., de Lajudie, P. and Avarre, J. C. (2006) Development of a lab-made microarray for analyzing the genetic diversity of nitrogen ﬁxing symbionts Sinorhizobium meliloti and Sinorhizobium medicae. J. Microbiol. Methods, 67, 114–124. Bains, W. and Smith, G.C. (1988) A novel method for nucleic acid sequence determination. J. Theor. Biol., 135, 303–307. Baldeschwieler, J.D., Gamble, R.C. and Thierault, T.P. (1995) Method and apparatus for performing multiple sequential reactions on a matrix. WO9525116(19950314). Ballmer, K., Korczak, B.M., Kuhnert, P., Slickers, P., Ehricht, R. and Hachler, H., (2007) Fast DNA serotyping of Escherichia coli by use of an oligonucleotide microarray. J. Clin. Microbiol., 45, 370–379. Barczak, A., Rodriguez, M.W., Hanspers, K., Koth, L.L., Tai, Y.C., Bolstad, B.M., Speed, T.P. and Erle, D.J. (2003) Spotted long oligonucleotide arrays for human gene expression analysis. Genome Res., 13, 1775–1785. Bavykin, S.G., Akowski, J.P., Zakhariev, V.M., Barsky, V.E., Perov, A.N. and Mirzabekov, A.D. (2001) Portable system for microbial sample preparation and oligonucleotide microarray analysis. Appl. Environ. Microbiol., 67, 922–928. Bekal, S., Brousseau, R., Masson, L., Prefontaine, G., Fairbrother, J. and Harel, J. (2003) Rapid identiﬁcation of Escherichia coli pathotypes by virulence gene detection with DNA microarrays. J. Clin. Microbiol., 41, 2113–2125. Blanchard, A.P. and Friend, S.H. (1999) Cheap DNA arrays – its not all smoke and mirrors. Nat. Biotechnol., 17, 953. Blanchard, A.P., Kaiser, R.J. and Hood, L.E. (1996) High-density oligonucleotide arrays. Biosens. Bioelectron., 11, 687–690. Bo, T. and Jonassen, I. (2002) New feature subset selection procedures for

j23

j 1 Introduction to Microarray-Based Detection Methods

24

21

22

23

24

25

26

27

28

classiﬁcation of expression proﬁles. Genome Biol., 3, RESEARCH0017. Bodrossy, L. (2003) Diagnostic oligonucleotide microarrays for microbiology, in A Beginners Guide to Microarrays, (ed. E.M. Blalock,), Kluwer, New York, pp. 566–582. Bodrossy, L. and Sessitsch, A. (2004) Oligonucleotide microarrays in microbial diagnostics. Curr. Opin. Microbiol., 7, 245–254. Bodrossy, L., Stralis-Pavese, N., Konrad-Koszler, M., Weilharter, A., Reichenauer, T.G., Schofer, D. and Sessitsch, A. (2006) mRNA-based parallel detection of active methanotroph populations by use of a diagnostic microarray. Appl. Environ. Microbiol., 72, 1672–1676. Bodrossy, L., Stralis-Pavese, N., Murrell, J.C., Radajewski, S., Weilharter, A. and Sessitsch, A. (2003) Development and validation of a diagnostic microbial microarray for methanotrophs. Environ. Microbiol., 5, 566–582. Boldrick, J.C., Alizadeh, A.A., Diehn, M., Dudoit, S., Liu, C.L., Belcher, C.E., Botstein, D., Staudt, L.M., Brown, P.O. and Relman, D.A. (2002) Stereotyped and speciﬁc gene expression programs in human innate immune responses to bacteria. Proc. Natl Acad. Sci. USA, 99, 972–977. Borucki, M.K., Krug, M.J., Muraoka, W.T. and Call, D.R. (2003) Discrimination among Listeria monocytogenes isolates using a mixed genome DNA microarray. Vet. Microbiol., 92, 351–362. Borucki, M.K., Reynolds, J., Call, D.R., Ward, T.J., Page, B. and Kadushin, J. (2005) Suspension microarray with dendrimer signal ampliﬁcation allows direct and high-throughput subtyping of Listeria monocytogenes from genomic DNA. J. Clin. Microbiol., 43, 3255–3259. Brennan, T.M. and Heinecker, H. (1995) Methods and compositions for

29

30

31

32

33

34

35

36

determining the sequence of nucleic acids. 5,474,796 (US 5403708). Brenner, S., Johnson, M., Bridgham, J., Golda, G., Lloyd, D.H., Johnson, D., Luo, S., McCurdy, S., Foy, M., Ewan, M., et al. (2000) Gene expression analysis by massively parallel signature sequencing (MPSS) on microbead arrays. Nat. Biotechnol., 18, 630–634. Brodie, E.L., DeSantis, T.Z., Joyner, D.C., Baek, S.M., Larsen, J.T., Andersen, G.L., Hazen, T.C., Richardson, P.M., Herman, D.J., Tokunaga, T.K., et al. (2006) Application of a high-density oligonucleotide microarray approach to study bacterial population dynamics during uranium reduction and reoxidation. Appl. Environ. Microbiol., 72, 6288–6298. Call, D.R., Bakko, M.K., Krug, M.J. and Roberts, M.C. (2003) Identifying antimicrobial resistance genes with DNA microarrays. Antimicrob. Agents Chemother., 47, 3290–3295. Call, D.R., Borucki, M.K. and Loge, F.J. (2003) Detection of bacterial pathogens in environmental samples using DNA microarrays. J. Microbiol. Methods, 53, 235–243. Case-Green, S.C. and Southern, E.M. (1994) Studies on the base pairing properties of deoxyinosine by solid phase hybridization to oligonucleotides. Nucleic Acids Res., 22, 131–136. Cassat, J.E., Dunman, P.M., McAleese, F., Murphy, E., Projan, S.J. and Smeltzer, M.S. (2005) Comparative genomics of Staphylococcus aureus musculoskeletal isolates. J. Bacteriol., 187, 576–592. Charbonnier, Y., Gettler, B.M., Francois, P., Bento, M., Renzoni, A., Vaudaux, P., Schlegel, W. and Schrenzel, J. (2005) A generic approach for the design of whole-genome oligoarrays, validated for genomotyping, deletion mapping and gene expression analysis on Staphylococcus aureus. BMC Genomics, 6, 95. Chizhikov, V., Rasooly, A., Chumakov, K. and Levy, D.D. (2001) Microarray

References

37

38

39

40

41

42

43

44

45

analysis of microbial virulence factors. Appl. Environ. Microbiol., 67, 3258–3263. Couzinet, S., Jay, C., Barras, C., Vachon, R., Vernet, G., Ninet, B., Jan, I., Minazio, M.A., Francois, P., Lew, D., et al. (2005) High-density DNA probe arrays for identiﬁcation of staphylococci to the species level. J. Microbiol. Methods, 61, 201–208. Couzinet, S., Yugueros, J., Barras, C., Visomblin, N., Francois, P., Lacroix, B., Vernet, G., Lew, D., Troesch, A., Schrenzel, J., et al. (2005) Evaluation of a high-density oligonucleotide array for characterization of grlA, grlB, gyrA and gyrB mutations in ﬂuoroquinolone resistant Staphylococcus aureus isolates. J. Microbiol. Methods, 60, 275–279. Debouck, C. and Goodfellow, P.N. (1999) DNA microarrays in drug discovery and development. Nat. Genet., 21, 48–50. Denef, V.J., Park, J., Rodrigues, J.L., Tsoi, T.V., Hashsham, S.A. and Tiedje, J.M. (2003) Validation of a more sensitive method for using spotted oligonucleotide DNA microarrays for functional genomics studies on bacterial communities. Environ. Microbiol., 5, 933–943. Diehl, F., Grahlmann, S., Beier, M. and Hoheisel, J.D. (2001) Manufacturing DNA microarrays of high spot homogeneity and reduced background signal. Nucleic Acids Res., 29, e38. Dorrell, N., Hinchliffe, S.J. and Wren, B.W. (2005) Comparative phylogenomics of pathogenic bacteria by microarray analysis. Curr. Opin. Microbiol., 8, 620–626. Drmanac, R. and Crkvenjakov, R. (1999) Method of sequencing of genomes by hybridization of oligonucleotide probes US1991000723712 (US5202231). Drmanac, R. and Drmanac, S. (1999) cDNA screening by array hybridization. Methods Enzymol., 303, 165–78, 165–178. Drmanac, R., Drmanac, S., Strezoska, Z., Paunesku, T., Labat, I., Zeremski, M., Snoddy, J., Funkhouser, W.K., Koop, B.

46

47

48

49

50

51

52

53

and Hood, L. (1993) DNA sequence determination by hybridization: a strategy for efﬁcient large-scale sequencing. Science, 260, 1649–1652. Drmanac, S., Kita, D., Labat, I., Hauser, B., Schmidt, C., Burczak, J.D. and Drmanac, R. (1998) Accurate sequencing by hybridization for DNA diagnostics and individual genomics. Nat. Biotechnol., 16, 54–58. Dunbar, S.A., Vander Zee, C.A., Oliver, K.G., Karem, K.L. and Jacobson, J.W. (2003) Quantitative, multiplexed detection of bacterial pathogens: DNA and protein applications of the Luminex LabMAP system. J. Microbiol. Methods, 53, 245–252. Duveneck, G.L., Bopp, M.A., Ehrat, M., Balet, L.P., Haiml, M., Keller, U., Marowsky, G. and Soria, S. (2003) Two-photon ﬂuorescence excitation of macroscopic areas on planar waveguides. Biosens. Bioelectron., 18, 503–510. Epstein, J.R., Ferguson, J.A., Lee, K.H. and Walt, D.R. (2003) Combinatorial decoding: an approach for universal DNA array fabrication. J. Am. Chem. Soc., 125, 13753–13759. Epstein, J.R., Leung, A.P., Lee, K.H. and Walt, D.R. (2003) High-density, microsphere-based ﬁber optic DNA microarrays. Biosens. Bioelectron., 18, 541–546. Epstein, J.R. and Walt, D.R. (2003) Fluorescence-based ﬁbre optic arrays: a universal platform for sensing. Chem. Soc. Rev., 32, 203–214. Ericsson, O., Sivertsson, A., Lundeberg, J. and Ahmadian, A. (2003) Microarraybased resequencing by apyrase-mediated allele-speciﬁc extension. Electrophoresis, 24, 3330–3338. Fan, J., Chen, Y., Chan, H.M., Tam, P.K. and Ren, Y. (2005) Removing intensity effects and identifying signiﬁcant genes for Affymetrix arrays in macrophage migration inhibitory factor-suppressed neuroblastoma cells. Proc. Natl Acad. Sci. USA, 102, 17751–17756.

j25

j 1 Introduction to Microarray-Based Detection Methods

26

54 Fan, J., Peng, H. and Huang, T. (2005) Semilinear high-dimensional model for normalization of microarray data: a theoretical analysis and partial consistency (with discussion). J. Am. Statist. Ass., 100, 781–813. 55 Fan, J., Tam, P., Woude, G.V. and Ren, Y. (2004) Normalization and analysis of cDNA microarrays using within-array replications applied to neuroblastoma cell response to a cytokine. Proc. Natl Acad. Sci. USA, 101, 1135–1140. 56 Fan, J.B., Chen, X., Halushka, M.K., Berno, A., Huang, X., Ryder, T., Lipshutz, R.J., Lockhart, D.J. and Chakravarti, A. (2000) Parallel genotyping of human SNPs using generic highdensity oligonucleotide tag arrays. Genome Res., 10, 853–860. 57 Ferguson, J.A., Steemers, F.J. and Walt, D.R. (2000) High-density ﬁber-optic DNA random microsphere array. Anal. Chem., 72, 5618–5624. 58 Fodor, S.P., Rava, R.P., Huang, X.C., Pease, A.C., Holmes, C.P. and Adams, C.L. (1993) Multiplexed biochemical assays with biological chips. Nature, 364, 555–556. 59 Fodor, S.P., Read, J.L., Pirrung, M.C., Stryer, L., Lu, A.T. and Solas, D. (1991) Light-directed, spatially addressable parallel chemical synthesis. Science, 251, 767–773. 60 Francois, P., Bento, M., Vaudaux, P. and Schrenzel, J. (2003) Comparison of ﬂuorescence and resonance light scattering for highly sensitive microarray detection of bacterial pathogens. J. Microbiol. Methods, 55, 755–762. 61 Francois, P., Charbonnier, Y., Jaquet, J., Utinger, D., Bento, M., Lew, D.P., Kresbach, G., Schlegel, W. and Schrenzel, J. (2005) Rapid Bacterial identiﬁcation using evanescent waveguide oligonucleotide microarray classiﬁcation. J. Microbiol. Methods, 65, 390–403. 62 Francois, P., Garzoni, C., Bento, M. and Schrenzel, J. (2007) Comparison of

63

64

65

66

67

68

69

70

ampliﬁcation methods for transcriptomic analyses of low abundance prokaryotic RNA sources. J. Microbiol. Methods, 68, 385–391. Franke-Whittle, I.H., Klammer, S.H., Mayrhofer, S. and Insam, H. (2006) Comparison of different labeling methods for the production of labeled target DNA for microarray hybridization. J. Microbiol. Methods, 65, 117–126. Fredericks, D.N. and Relman, D.A. (1996) Sequence-based identiﬁcation of microbial pathogens: a reconsideration of Kochs postulates. Clin. Microbiol. Rev., 9, 18–33. Garaizar, J., Rementeria, A. and Porwollik, S. (2006) DNA microarray technology: a new tool for the epidemiological typing of bacterial pathogens? FEMS Immunol. Med. Microbiol., 47, 178–189. Garzoni, C., Francois, P., Couzinet, S., Tapparel, C., Charbonnier, Y., Huyghe, A., Renzoni, A., Lucchini, S., Lew, D.P., Vaudaux, P., Kelley, W.L. and Schrenzel, J. (2007) A global view of Staphylococcus aureus whole genome expression upon internalization in human epithelial cells. BMC Genomics, 8, 171. Gasch, A.P., Spellman, P.T., Kao, C.M., Carmel-Harel, O., Eisen, M.B., Storz, G., Botstein, D. and Brown, P.O. (2000) Genomic expression programs in the response of yeast cells to environmental changes. Mol. Biol. Cell, 11, 4241–4257. Gill, R.T., Katsoulakis, E., Schmitt, W., Taroncher-Oldenburg, G., Misra, J. and Stephanopoulos, G. (2002) Genome-wide dynamic transcriptional proﬁling of the light-to-dark transition in Synechocystis sp. strain PCC 6803. J. Bacteriol., 184, 3671–3681. Gillepsie, D. and Spiegelman, S.A. (1965) A quantitative assay for DNA–RNA hybrids with DNA immobilized on a membrane. J. Mol. Biol., 12, 829–842. Gingeras, T.R., Ghandour, G., Wang, E., Berno, A., Small, P.M., Drobniewski, F., Alland, D., Desmond, E., Holodniy, M.

References

71

72

73

74

75

76

77

and Drenkow, J. (1998) Simultaneous genotyping and species identiﬁcation using hybridization pattern recognition analysis of generic Mycobacterium DNA arrays. Genome Res., 8, 435–448. Golub, T.R., Slonim, D.K., Tamayo, P., Huard, C., Gaasenbeek, M., Mesirov, J.P., Coller, H., Loh, M.L., Downing, J.R., Caligiuri, M.A., et al. (1999) Molecular classiﬁcation of cancer: class discovery and class prediction by gene expression monitoring. Science, 286, 531–537. Grifantini, R., Bartolini, E., Muzzi, A., Draghi, M., Frigimelica, E., Berger, J., Ratti, G., Petracca, R., Galli, G., Agnusdei, M., et al. (2002) Previously unrecognized vaccine candidates against group B meningococcus identiﬁed by DNA microarrays. Nat. Biotechnol., 20, 914–921. Guschin, D., Yershov, G., Zaslavsky, A., Gemmell, A., Shick, V., Proudnikov, D., Arenkov, P. and Mirzabekov, A. (1997) Manual manufacturing of oligonucleotide, DNA, and protein microchips. Anal. Biochem., 250, 203–211. Hacia, J.G. (1999) Resequencing and mutational analysis using oligonucleotide microarrays. Nat. Genet., 21, 42–47. Halperin, A., Buhot, A. and Zhulina, E.B. (2005) Brush effects on DNA chips: thermodynamics, kinetics, and design guidelines. Biophys. J., 89, 796–811. Hardenbol, P., Baner, J., Jain, M., Nilsson, M., Namsaraev, E.A., Karlin-Neumann, G.A., Fakhrai-Rad, H., Ronaghi, M., Willis, T.D., Landegren, U., et al. (2003) Multiplexed genotyping with sequence-tagged molecular inversion probes. Nat. Biotechnol., 21, 673–678. Hardenbol, P., Yu, F., Belmont, J., Mackenzie, J., Bruckner, C., Brundage, T., Boudreau, A., Chow, S., Eberle, J., Erbilgin, A., et al. (2005) Highly multiplexed molecular inversion probe genotyping: over 10 000 targeted SNPs genotyped in a single tube assay. Genome Res., 15, 269–275.

78 Hashsham, S.A., Wick, L.M., Rouillard, J.M., Gulari, E. and Tiedje, J.M. (2004) Potential of DNA microarrays for developing parallel detection tools (PDTs) for microorganisms relevant to biodefense and related research needs. Biosens. Bioelectron., 20, 668–683. 79 Hessner, M.J., Meyer, L., Tackes, J., Muheisen, S. and Wang, X. (2004) Immobilized probe and glass surface chemistry as variables in microarray fabrication. BMC Genomics, 5, 53. 80 Holzman, T. and Kolker, E. (2004) Statistical analysis of global gene expression data: some practical considerations. Curr. Opin. Biotechnol., 15, 52–57. 81 Hughes, T.R., Mao, M., Jones, A.R., Burchard, J., Marton, M.J., Shannon, K.W., Ziman, M., Meyer, M.R., Kobayashi, S., Dai, H., et al. (2001) Expression proﬁling using microarrays fabricated by an ink-jet oligonucleotide synthesizer. Nat. Biotechnol., 19, 342–347. 82 Jaeger, J. and Spang, R. (2006) Selecting normalization genes for small diagnostic microarrays. BMC Bioinformatics, 7, 388. 83 Joyce, E.A., Chan, K., Salama, N.R. and Falkow, S. (2002) Redeﬁning bacterial populations: a post-genomic reformation. Nat. Rev. Genet., 3, 462–473. 84 Kafatos, F.C., Jones, C.W. and Efstratiadis, A. (1979) Determination of nucleic acid sequence homologies and relative concentrations by a dot hybridization procedure. Nucleic Acids Res., 7, 1541–1552. 85 Kajiyama, T., Miyahara, Y., Kricka, L.J., Wilding, P., Graves, D.J., Surrey, S. and Fortina, P. (2003) Genotyping on a thermal gradient DNA chip. Genome Res., 13, 467–475. 86 Kakinuma, K., Fukushima, M. and Kawaguchi, R. (2003) Detection and identiﬁcation of Escherichia coli, Shigella, and Salmonella by microarrays using the gyrB gene. Biotechnol. Bioeng., 83, 721–728.

j27

j 1 Introduction to Microarray-Based Detection Methods

28

87 Kelly, J.J., Chernov, B.K., Tovstanovsky, I., Mirzabekov, A.D. and Bavykin, S.G. (2002) Radical-generating coordination complexes as tools for rapid and effective fragmentation and ﬂuorescent labeling of nucleic acids for microchip hybridization. Anal. Biochem., 311, 103–118. 88 Kessler, N., Ferraris, O., Palmer, K., Marsh, W. and Steel, A. (2004) Use of the DNA ﬂow-thru chip, a three-dimensional biochip, for typing and subtyping of inﬂuenza viruses. J. Clin. Microbiol., 42, 2173–2185. 89 Knight, J. (2001) When the chips are down. Nature, 410, 860–861. 90 Koessler, T., Francois, P., Charbonnier, Y., Huyghe, A., Bento, M., Dharan, S., Renzi, G., Lew, D., Harbarth, S., Pittet, D., et al. (2006) Use of oligoarrays for characterization of community-onset methicillin-resistant Staphylococcus aureus. J. Clin. Microbiol., 44, 1040–1048. 91 Korczak, B., Frey, J., Schrenzel, J., Pluschke, G., Pﬁster, R., Ehricht, R. and Kuhnert, P. (2005) Use of diagnostic microarrays for determination of virulence gene patterns of Escherichia coli K1, a major cause of neonatal meningitis. J. Clin. Microbiol., 43, 1024–1031. 92 Kothapalli, R., Yoder, S.J., Mane, S. and Loughran, T.P. Jr (2002) Microarray results: how accurate are they? BMC Bioinformatics, 3, 22. 93 Kroll, T.C. and Wolﬂ, S. (2002) Ranking: a closer look on globalization methods for normalization of gene expression arrays. Nucleic Acids Res., 30, e50. 94 Li, J., Pankratz, M. and Johnson, J.A. (2002) Differential gene expression patterns revealed by oligonucleotide versus long cDNA arrays. Toxicol. Sci., 69, 383–390. 95 Li, W. (2005) How many genes are needed for early detection of breast cancer, based on gene expression patterns in peripheral blood cells? Breast Cancer Res, 7, e5. 96 Lima, W.F., Monia, B.P., Ecker, D.J. and Freier, S.M. (1992) Implication of RNA structure on antisense oligonucleotide

97

98

99

100

101

102

103

104

105

hybridization kinetics. Biochemistry, 31, 12055–12061. Lindroos, H.L., Mira, A., Repsilber, D., Vinnere, O., Naslund, K., Dehio, M., Dehio, C. and Andersson, S.G. (2005) Characterization of the genome composition of Bartonella koehlerae by microarray comparative genomic hybridization proﬁling. J. Bacteriol., 187, 6155–6165. Lindroos, K., Liljedahl, U., Raitio, M. and Syvanen, A.C. (2001) Minisequencing on oligonucleotide microarrays: comparison of immobilization chemistries. Nucleic Acids Res., 29, e69. Lindroos, K., Sigurdsson, S., Johansson, K., Ronnblom, L. and Syvanen, A.C. (2002) Multiplex SNP genotyping in pooled DNA samples by a four-colour microarray system. Nucleic Acids Res., 30, e70. Lipshutz, R.J., Fodor, S.P., Gingeras, T.R. and Lockhart, D.J. (1999) High density synthetic oligonucleotide arrays. Nat. Genet., 21, 20–24. Lipshutz, R.J., Morris, D., Chee, M., Hubbell, E., Kozal, M.J., Shah, N., Shen, N., Yang, R. and Fodor, S.P. (1995) Using oligonucleotide probe arrays to access genetic diversity. Biotechniques, 19, 442–447. Liu, M., Popper, S.J., Rubins, K.H. and Relman, D.A. (2006) Early days: genomics and human responses to infection. Curr. Opin. Microbiol., 9, 312–319. Liu, W.T., Guo, H. and Wu, J.H. (2007) Effects of target length on the hybridization efﬁciency and speciﬁcity of rRNA-based oligonucleotide microarrays. Appl. Environ. Microbiol., 73, 73–82. Liu, W.T., Mirzabekov, A.D. and Stahl, D.A. (2001) Optimization of an oligonucleotide microchip for microbial identiﬁcation studies: a non-equilibrium dissociation approach. Environ. Microbiol., 3, 619–629. Lockhart, D.J., Dong, H., Byrne, M.C., Follettie, M.T., Gallo, M.V., Chee, M.S., Mittmann, M., Wang, C., Kobayashi, M.,

References

106

107

108

109

110

111

112

113

Horton, H., et al. (1996) Expression monitoring by hybridization to highdensity oligonucleotide arrays. Nat. Biotechnol., 14, 1675–1680. Lodes, M.J., Suciu, D., Elliott, M., Stover, A.G., Ross, M., Caraballo, M., Dix, K., Crye, J., Webby, R.J., Lyon, W.J., et al. (2006) Use of semiconductor-based oligonucleotide microarrays for inﬂuenza a virus subtype identiﬁcation and sequencing. J. Clin. Microbiol., 44, 1209–1218. Loge, F.J., Thompson, D.E. and Call, D.R. (2002) PCR detection of speciﬁc pathogens in water: a risk-based analysis. Environ. Sci. Technol., 36, 2754–2759. Loy, A. and Bodrossy, L. (2006) Highly parallel microbial diagnostics using oligonucleotide microarrays. Clin. Chim. Acta, 363, 106–119. Loy, A., Lehner, A., Lee, N., Adamczyk, J., Meier, H., Ernst, J., Schleifer, K.H. and Wagner, M. (2002) Oligonucleotide microarray for 16S rRNA gene-based detection of all recognized lineages of sulfate-reducing prokaryotes in the environment. Appl. Environ. Microbiol., 68, 5064–5081. Loy, A., Schulz, C., Lucker, S., Schopfer-Wendels, A., Stoecker, K., Baranyi, C., Lehner, A. and Wagner, M. (2005) 16S rRNA gene-based oligonucleotide microarray for environmental monitoring of the betaproteobacterial order Rhodocyclales. Appl. Environ. Microbiol., 71, 1373–1386. Lucchini, S., Liu, H., Jin, Q., Hinton, J.C. and Yu, J. (2005) Transcriptional adaptation of Shigella ﬂexneri during infection of macrophages and epithelial cells: insights into the strategies of a cytosolic bacterial pathogen. Infect. Immun., 73, 88–102. Lucchini, S., Thompson, A. and Hinton, J.C. (2001) Microarrays for microbiologists. Microbiology, 147, 1403–1414. Ludwig, W., Strunk, O., Westram, R., Richter, L., Meier, H., Yadhukumar,

114

115

116

117

118

119

120

121

Buchner, A., Lai, T., Steppi, S., Jobb, G., et al. (2004) ARB: a software environment for sequence data. Nucleic Acids Res., 32, 1363–1371. Lynch, J.L., deSilva, C.J., Peeva, V.K. and Swanson, N.R. (2006) Comparison of commercial probe labeling kits for microarray: towards quality assurance and consistency of reactions. Anal. Biochem., 355, 224–231. Macevicz, S.C. (1995) Nucleic acid sequence determination by multiple mixed oligonucleotide probes, US1988000261702 (US5002867). Malanoski, G.J., Samore, M.H., Pefanis, A. and Karchmer, A.W. (1995) Staphylococcus aureus catheter-associated bacteremia Minimal effective therapy and unusual infectious complications associated with arterial sheath catheters. Arch. Intern. Med., 155, 1161–1166. Marcelino, L.A., Backman, V., Donaldson, A., Steadman, C., Thompson, J.R., Preheim, S.P., Lien, C., Lim, E., Veneziano, D. and Polz, M.F. (2006) Accurately quantifying lowabundant targets amid similar sequences by revealing hidden correlations in oligonucleotide microarray data. Proc. Natl Acad. Sci. USA, 103, 13629–13634. Margulies, M., Egholm, M., Altman, W.E., Attiya, S., Bader, J.S., Bemben, L.A., Berka, J., Braverman, M.S., Chen, Y.J., Chen, Z., et al. (2005) Genome sequencing in microfabricated highdensity picolitre reactors. Nature, 437, 376–380. Maskos, U. and Southern, E.M. (1993) A novel method for the parallel analysis of multiple mutations in multiple samples. Nucleic Acids Res., 21, 2269–2270. Maskos, U. and Southern, E.M. (1993) A study of oligonucleotide reassociation using large arrays of oligonucleotides synthesised on a glass support. Nucleic Acids Res., 21, 4663–4669. Matveeva, O.V., Shabalina, S.A., Nemtsov, V.A., Tsodikov, A.D., Gesteland, R.F. and Atkins, J.F. (2003) Thermodynamic

j29

j 1 Introduction to Microarray-Based Detection Methods

30

122

123

124

125

126

127

128

129

130

calculations and statistical correlations for oligo-probes design. Nucleic Acids Res., 31, 4211–4217. Mikulowska-Mennis, A., Taylor, T.B., Vishnu, P., Michie, S.A., Raja, R., Horner, N. and Kunitake, S.T. (2002) High-quality RNA from cells isolated by laser capture microdissection. Biotechniques, 33, 176–179. Mir, K.U. and Southern, E.M. (1999) Determining the inﬂuence of structure on hybridization using oligonucleotide arrays. Nat. Biotechnol., 17, 788–792. Murray, A.E., Lies, D., Li, G., Nealson, K., Zhou, J. and Tiedje, J.M. (2001) DNA/DNA hybridization to microarrays reveals gene-speciﬁc differences between closely related microbial genomes. Proc. Natl Acad. Sci. USA, 98, 9853–9858. Naef, F. and Magnasco, M.O. (2003) Solving the riddle of the bright mismatches: labeling and effective binding in oligonucleotide arrays. Phys. Rev. E, 68, 011906. Nguyen, H.K. and Southern, E.M. (2000) Minimizing the secondary structure of DNA targets by incorporation of a modiﬁed deoxynucleoside: implications for nucleic acid analysis by hybridization. Nucleic Acids Res., 28, 3904–3909. Pasternack, R.F. and Collings, P.J. (1995) Resonance light scattering: a new technique for studying chromophore aggregation. Science, 269, 935–939. Pawlak, M., Schick, E., Bopp, M.A., Schneider, M.J., Oroszlan, P. and Ehrat, M. (2002) Zeptosens protein microarrays: a novel high performance microarray platform for low abundance protein analysis. Proteomics., 2, 383–393. Pease, A.C., Solas, D., Sullivan, E.J., Cronin, M.T., Holmes, C.P. and Fodor, S.P. (1994) Light-generated oligonucleotide arrays for rapid DNA sequence analysis. Proc. Natl Acad. Sci. USA, 91, 5022–5026. Peplies, J., Glockner, F.O. and Amann, R. (2003) Optimization strategies for DNA

131

132

133

134 135

136

137

138

139

microarray-based detection of bacteria with 16S rRNA-targeting oligonucleotide probes. Appl. Environ. Microbiol., 69, 1397–1407. Perou, C.M., Sorlie, T., Eisen, M.B. Van de, R.M., Jeffrey, S. S., Rees, C. A., Pollack, J. R., Ross, D. T., Johnsen, H., Akslen, L. A., et al. (2000) Molecular portraits of human breast tumours. Nature, 406, 747–752. Perreten, V., Vorlet-Fawer, L., Slickers, P., Ehricht, R., Kuhnert, P. and Frey, J. (2005) Microarray-based detection of 90 antibiotic resistance genes of Grampositive bacteria. J. Clin. Microbiol., 43, 2291–2302. Peterson, A.W., Heaton, R.J. and Georgiadis, R.M. (2001) The effect of surface probe density on DNA hybridization. Nucleic Acids Res., 29, 5163–5168. Phimister, B. (1999) Chipping forecast: going global. Nat. Genet, 21, Suppl 1. Porwollik, S., Frye, J., Florea, L.D., Blackmer, F. and McClelland, M. (2003) A non-redundant microarray of genes for two related bacteria. Nucleic Acids Res., 31, 1869–1876. Pozhitkov, A., Chernov, B., Yershov, G. and Noble, P.A. (2005) Evaluation of gel-pad oligonucleotide microarray technology by using artiﬁcial neural networks. Appl. Environ. Microbiol., 71, 8663–8676. Pozhitkov, A., Noble, P.A., Domazet-Loso, T., Nolte, A.W., Sonnenberg, R., Staehler, P., Beier, M. and Tautz, D. (2006) Tests of rRNA hybridization to microarrays suggest that hybridization characteristics of oligonucleotide probes for species discrimination cannot be predicted. Nucleic Acids Res., 34, e66. Proudnikov, D. and Mirzabekov, A. (1996) Chemical methods of DNA and RNA ﬂuorescent labeling. Nucleic Acids Res., 24, 4535–4542. Puskas, L.G., Zvara, A., Hackler, L., Jr. and Van Hummelen, P. (2002) RNA ampliﬁcation results in reproducible

References

140

141

142

143

144

145

146

147

microarray data with slight ratio bias. Biotechniques, 32, 1330–1340. Raman, R., Raguram, S., Venkataraman, G., Paulson, J.C. and Sasisekharan, R. (2005) Glycomics: an integrated systems approach to structure–function relationships of glycans. Nat. Methods, 2, 817–824. Rhee, S.K., Liu, X., Wu, L., Chong, S.C., Wan, X. and Zhou, J. (2004) Detection of genes involved in biodegradation and biotransformation in microbial communities by using 50-mer oligonucleotide microarrays. Appl. Environ. Microbiol., 70, 4303–4317. Rimour, S., Hill, D., Militon, C. and Peyret, P. (2005) GoArrays: highly dynamic and efﬁcient microarray probe design. Bioinformatics., 21, 1094–1103. Roth, F.P., Hughes, J.D., Estep, P.W. and Church, G.M. (1998) Finding DNA regulatory motifs within unaligned noncoding sequences clustered by whole-genome mRNA quantitation. Nat. Biotechnol., 16, 939–945. Rudi, K., Flateland, S.L., Hanssen, J.F., Bengtsson, G. and Nissen, H. (2002) Development and evaluation of a 16S ribosomal DNA array-based approach for describing complex microbial communities in ready-to-eat vegetable salads packed in a modiﬁed atmosphere. Appl. Environ. Microbiol., 68, 1146–1156. Rudi, K., Treimo, J., Nissen, H. and Vegarud, G. (2003) Protocols for 16S rDNA array analyses of microbial communities by sequence-speciﬁc labeling of DNA probes. Scientiﬁc World Journal, 3, 578–584. Salama, N., Guillemin, K., McDaniel, T.K., Sherlock, G., Tompkins, L. and Falkow, S. (2000) A whole-genome microarray reveals genetic diversity among Helicobacter pylori strains. Proc. Natl Acad. Sci. USA, 97, 14668–14673. Sanguin, H., Herrera, A., Oger-Desfeux, C., Dechesne, A., Simonet, P., Navarro, E., Vogel, T.M., Moenne-Loccoz, Y., Nesme, X. and Grundmann, G.L.

148

149

150

151

152

153

154

155

(2006) Development and validation of a prototype 16S rRNA-based taxonomic microarray for Alphaproteobacteria. Environ. Microbiol., 8, 289–307. SantaLucia, J. Jr (1998) A uniﬁed view of polymer, dumbbell, and oligonucleotide DNA nearest-neighbor thermodynamics. Proc. Natl Acad. Sci. USA, 95, 1460–1465. SantaLucia, J.J., Allawi, H.T. and Seneviratne, P.A. (1996) Improved nearest-neighbor parameters for predicting DNA duplex stability. Biochemistry, 35, 3555–3562. Santos, S.R. and Ochman, H. (2004) Identiﬁcation and phylogenetic sorting of bacterial lineages with universally conserved genes and proteins. Environ. Microbiol., 6, 754–759. Schadt, E.E., Li, C., Ellis, B. and Wong, W.H. (2001) Feature extraction and normalization algorithms for highdensity oligonucleotide gene expression array data. J. Cell Biochem. Suppl., 37, 120–125. Schena, M., Shalon, D., Davis, R.W. and Brown, P.O. (1995) Quantitative monitoring of gene expression patterns with a complementary DNA microarray. Science, 270, 467–470. Scherl, A., Francois, P., Charbonnier, Y., Deshusses, J.M., Koessler, T., Huyghe, A., Bento, M., Stahl-Zeng, J., Fischer, A., Masselot, A., et al. (2006) Exploring glycopeptide resistance in Staphylococcus aureus: a combined proteomics and transcriptomics approach for the identiﬁcation of resistance-related markers. BMC Genomics, 7, 296. Schrenzel, J. and Hibbs, J. (2003) Noncognate hybridization system (NCHS) 00/75377 A2(6,544,777). Sergeev, N., Distler, M., Courtney, S., Al Khaldi, S.F., Volokhov, D., Chizhikov, V. and Rasooly, A. (2004) Multipathogen oligonucleotide microarray for environmental and biodefense applications. Biosens. Bioelectron., 20, 684–698.

j31

j 1 Introduction to Microarray-Based Detection Methods

32

156 Shchepinov, M.S., Case-Green, S.C. and Southern, E.M. (1997) Steric factors inﬂuencing hybridization of nucleic acids to oligonucleotide arrays. Nucleic Acids Res., 25, 1155–1161. 157 Shi, L., Reid, L.H., Jones, W.D., Shippy, R., Warrington, J.A., Baker, S.C., Collins, P.J., De Longueville, F., Kawasaki, E.S., Lee, K.Y., et al. (2006) The MicroArray Quality Control (MAQC) project shows inter- and intraplatform reproducibility of gene expression measurements. Nat. Biotechnol., 24, 1151–1161. 158 Small, J., Call, D.R., Brockman, F.J., Straub, T.M. and Chandler, D.P. (2001) Direct detection of 16S rRNA in soil extracts by using oligonucleotide microarrays. Appl. Environ. Microbiol., 67, 4708–4716. 159 Smyth, G.K. and Speed, T. (2003) Normalization of cDNA microarray data. Methods, 31, 265–273. 160 Sohail, M., Akhtar, S. and Southern, E.M. (1999) The folding of large RNAs studied by hybridization to arrays of complementary oligonucleotides. RNA., 5, 646–655. 161 Southern, E.M. (1989) Analysing polynucleotide sequences 19891100 (PCT89/10977). 162 Southern, E.M. (1975) Detection of speciﬁc sequences among DNA fragments separated by gel electrophoresis. J. Mol. Biol., 98, 503–517. 163 Southern, E.M., Case-Green, S.C., Elder, J.K., Johnson, M., Mir, K.U., Wang, L. and Williams, J.C. (1994) Arrays of complementary oligonucleotides for analysing the hybridization behaviour of nucleic acids. Nucleic Acids Res., 22, 1368–1373. 164 Southern, E.M., Mir, K. and Shchepinov, M. (1999) Molecular interactions on microarrays. Nat. Genet., 21, 5–9. 165 Stenger, D.A., Andreadis, J.D., Vora, G.J. and Pancrazio, J.J. (2002) Potential applications of DNA microarrays in biodefense-related diagnostics.

166

167

168

169

170

171

172

173

Curr. Opin. Biotechnol., 13, 208–212. Stimpson, D.I., Cooley, P.W., Knepper, S.M. and Wallace, D.B. (1998) Parallel production of oligonucleotide arrays using membranes and reagent jet printing. Biotechniques, 25, 886–890. Stowe-Evans, E.L., Ford, J. and Kehoe, D.M. (2004) Genomic DNA microarray analysis: identiﬁcation of new genes regulated by light color in the cyanobacterium Fremyella diplosiphon. J. Bacteriol., 186, 4338–4349. Straub, T.M. and Chandler, D.P. (2003) Towards a uniﬁed system for detecting waterborne pathogens. J. Microbiol. Methods, 53, 185–197. Strezoska, Z., Paunesku, T., Radosavljevic, D., Labat, I., Drmanac, R. and Crkvenjakov, R. (1991) DNA sequencing by hybridization: bases read by a non-gel-based method. Proc. Natl Acad. Sci. USA, 88, 10089–10093. Tamayo, P., Slonim, D., Mesirov, J., Zhu, Q., Kitareewan, S., Dmitrovsky, E., Lander, E.S. and Golub, T.R. (1999) Interpreting patterns of gene expression with self-organizing maps: methods and application to hematopoietic differentiation. Proc. Natl Acad. Sci. USA, 96, 2907–2912. Taroncher-Oldenburg, G., Griner, E.M., Francis, C.A. and Ward, B.B. (2003) Oligonucleotide microarray for the study of functional gene diversity in the nitrogen cycle in the environment. Appl. Environ. Microbiol., 69, 1159–1171. Taylor, S., Smith, S., Windle, B. and Guiseppi-Elie, A. (2003) Impact of surface chemistry and blocking strategies on DNA microarrays. Nucleic Acids Res., 31, e87. Thiyagarajan, S., Karhanek, M., Akhras, M., Davis, R.W. and Pourmand, N. (2006) PathogenMIPer: a tool for the design of molecular inversion probes to detect multiple pathogens. BMC Bioinformatics, 7, 500.

References 174 Tiquia, S.M., Wu, L., Chong, S.C., Passovets, S., Xu, D., Xu, Y. and Zhou, J. (2004) Evaluation of 50-mer oligonucleotide arrays for detecting microbial populations in environmental samples. Biotechniques, 36, 664–665. 175 Treimo, J., Vegarud, G., Langsrud, T., Marki, S. and Rudi, K. (2006) Total bacterial and species-speciﬁc 16S rDNA micro-array quantiﬁcation of complex samples. J. Appl. Microbiol., 100, 985–998. 176 Troesch, A., Nguyen, H., Miyada, C.G., Desvarenne, S., Gingeras, T.R., Kaplan, P.M., Cros, P. and Mabilat, C. (1999) Mycobacterium species identiﬁcation and rifampin resistance testing with high-density DNA probe arrays. J. Clin. Microbiol., 37, 49–55. 177 Tseng, G.C., Oh, M.K., Rohlin, L., Liao, J.C. and Wong, W.H. (2001) Issues in cDNA microarray analysis: quality ﬁltering, channel normalization, models of variations and assessment of gene effects. Nucleic Acids Res., 29, 2549–2557. 178 Urakawa, H., El Fantroussi, S., Smidt, H., Smoot, J.C., Tribou, E.H., Kelly, J.J., Noble, P.A. and Stahl, D.A. (2003) Optimization of single-base-pair mismatch discrimination in oligonucleotide microarrays. Appl. Environ. Microbiol., 69, 2848–2856. 179 Urakawa, H., Noble, P.A., El Fantroussi, S., Kelly, J.J. and Stahl, D.A. (2002) Single-base-pair discrimination of terminal mismatches by using oligonucleotide microarrays and neural network analyses. Appl. Environ. Microbiol., 68, 235–244. 180 Vainrub, A. and Pettitt, B.M. (2002) Coulomb blockage of hybridization in two-dimensional DNA arrays. Phys. Rev. E, 66, 041905. 181 van Bakel, H. and Holstege, F.C. (2004) In control: systematic assessment of microarray performance. EMBO Rep., 5, 964–969. 182 van Leeuwen, W.B., Jay, C., Snijders, S., Durin, N., Lacroix, B., Verbrugh, H.A., Enright, M.C., Troesch, A. and

183

184

185

186

187

188

189

190

Van Belkum, A. (2003) Multilocus sequence typing of Staphylococcus aureus with DNA array technology. J. Clin. Microbiol., 41, 3323–3326. Volokhov, D., Rasooly, A., Chumakov, K. and Chizhikov, V. (2002) Identiﬁcation of listeria species by microarray-based assay. J. Clin. Microbiol., 40, 4720–4728. Voordouw, G., Voordouw, J.K., Karkhoff-Schweizer, R.R., Fedorak, P.M. and Westlake, D.W. (1991) Reverse sample genome probing, a new technique for identiﬁcation of bacteria in environmental samples by DNA hybridization, and its application to the identiﬁcation of sulfate-reducing bacteria in oil ﬁeld samples. Appl. Environ. Microbiol., 57, 3070–3078. Vora, G.J., Meador, C.E., Stenger, D.A. and Andreadis, J.D. (2004) Nucleic acid ampliﬁcation strategies for DNA microarray-based pathogen detection. Appl. Environ. Microbiol., 70, 3047–3054. Wang, D., Coscoy, L., Zylberberg, M., Avila, P.C., Boushey, H.A., Ganem, D. and DeRisi, J.L. (2002) Microarray-based detection and genotyping of viral pathogens. Proc. Natl Acad. Sci. USA, 99, 15687–15692. Wang, R.F., Beggs, M.L., Erickson, B.D. and Cerniglia, C.E. (2004) DNA microarray analysis of predominant human intestinal bacteria in fecal samples. Mol. Cell Probes, 18, 223–234. Wilson, C.L., Pepper, S.D., Hey, Y. and Miller, C.J. (2004) Ampliﬁcation protocols introduce systematic but reproducible errors into gene expression studies. Biotechniques, 36, 498–506. Wilson, K.H., Wilson, W.J., Radosevich, J.L., DeSantis, T.Z., Viswanathan, V.S., Kuczmarski, T.A. and Andersen, G.L. (2002) High-density microarray of small-subunit ribosomal DNA probes. Appl. Environ. Microbiol., 68, 2535–2541. Wong, C.W., Albert, T.J., Vega, V.B., Norton, J.E., Cutler, D.J., Richmond, T.A.,

j33

j 1 Introduction to Microarray-Based Detection Methods

34

191

192

193

194

195

196

197

198

Stanton, L.W., Liu, E.T. and Miller, L. D. (2004) Tracking the evolution of the SARS coronavirus using high-throughput, high-density resequencing arrays. Genome Res., 14, 398–405. Wu, L., Thompson, D.K., Li, G., Hurt, R.A., Tiedje, J.M. and Zhou, J. (2001) Development and evaluation of functional gene arrays for detection of selected genes in the environment. Appl. Environ. Microbiol., 67, 5780–5790. Wu, Y., de Kievit, P., Vahlkamp, L., Pijnenburg, D., Smit, M., Dankers, M., Melchers, D., Stax, M., Boender, P.J., Ingham, C., et al. (2004) Quantitative assessment of a novel ﬂow-through porous microarray for the rapid analysis of gene expression proﬁles. Nucleic Acids Res., 32, e123. Xiang, C.C., Chen, M., Ma, L., Phan, Q.N., Inman, J.M., Kozhich, O.A. and Brownstein, M.J. (2003) A new strategy to amplify degraded RNA from small tissue samples for microarray studies. Nucleic Acids Res., 31, e53. Yang, Y.H., Dudoit, S., Luu, P., Lin, D.M., Peng, V., Ngai, J. and Speed, T.P. (2002) Normalization for cDNA microarray data: a robust composite method addressing single and multiple slide systematic variation. Nucleic Acids Res., 30, e15. Yang, Y.H. and Speed, T. (2002) Design issues for cDNA microarray experiments. Nat. Rev. Genet., 3, 579–588. Ye, R.W., Wang, T., Bedzyk, L. and Croker, K.M. (2001) Applications of DNA microarrays in microbial systems. J. Microbiol. Methods, 47, 257–272. Yershov, G., Barsky, V., Belgovskiy, A., Kirillov, E., Kreindlin, E., Ivanov, I., Parinov, S., Guschin, D., Drobishev, A., Dubiley, S., et al. (1996) DNA analysis and diagnostics on oligonucleotide microchips. Proc. Natl Acad. Sci. USA, 93, 4913–4918. Yue, H., Eastman, P.S., Wang, B.B., Minor, J., Doctolero, M.H., Nuttall, R.L.,

199

200

201

202

203

204 205

206

Stack, R., Becker, J.W., Montgomery, J.R., Vainer, M., et al. (2001) An evaluation of the performance of cDNA microarrays for detecting changes in global mRNA expression. Nucleic Acids Res., 29, e41. Zaigler, A., Schuster, S.C. and Soppa, J. (2003) Construction and usage of a onefold-coverage shotgun DNA microarray to characterize the metabolism of the archaeon Haloferax volcanii. Mol. Microbiol., 48, 1089–1105. Zammatteo, N., Jeanmart, L., Hamels, S., Courtois, S., Louette, P., Hevesi, L. and Remacle, J. (2000) Comparison between different strategies of covalent attachment of DNA to glass surfaces to build DNA microarrays. Anal. Biochem., 280, 143–150. Zhang, L., Hurek, T. and Reinhold-Hurek, B. (2006) A nifH-based oligonucleotide microarray for functional diagnostics of nitrogen-ﬁxing microorganisms. Microb. Ecol., 8, 1721–1735. Zhou, J. and Thompson, D.K. (2002) Challenges in applying microarrays to environmental studies. Curr. Opin. Biotechnol., 13, 204–207. Zwick, M.E., Mcafee, F., Cutler, D.J., Read, T.D., Ravel, J., Bowman, G.R., Galloway, D.R. and Mateczun, A. (2005) Microarray-based resequencing of multiple Bacillus anthracis isolates. Genome Biol., 6, R10. Schena, M. (2005) Protein Microarrays, Jones & Bartlett, Sudbury MA. Talaat, A.M., Hunter, P. and Johnston, S.A. (2000) Genome-directed primers for selective labeling of bacterial transcripts for DNA microarray analysis. Nat Biotechnol., 18, 679–682. Kostic, T., Weilharter, A., Rubino, S., Delogu, G., Uzzau, S., Rudi, K., Sessitsch, A. and Bodrossy, L. (2007) A microbial diagnostic microarray technique for the sensitive detection and identiﬁcation of pathogenic bacteria in a background of non-pathogens. Analytical Biochemistry, 360, 244–254.

j35

Part I: Methods

j37

2 Long Oligonucleotide Microarray-Based Microbial Detection Tanja Kostic and Levente Bodrossy

2.1 Introduction

The advantages and characteristics of the long oligonucleotide probes (50–100mer) and random ampliﬁcation methods were discussed in detail in Chapter 1. Therefore, we only give a brief summary of the main points in Tables 2.1 and 2.2 before describing detailed protocols optimized in our laboratory.

Table 2.1 Long oligonucleotide probes.

Advantages

Higher target binding capacity than short oligonucleotides (15–30mer) Increased sensitivity at the target level (detection of a single target within a total genome/transcriptome background) [2]

Disadvantages

Lower threshold for the differentiation: 85–90% sequence similarity Reduced speciﬁcity of the probes compared to short oligonucleotides

Potential application

Targeting multiple marker genes, each of them speciﬁc to the microorganisms that are to be detected; therefore, the targeted gene primarily deﬁnes the speciﬁcity, rather than the probe itself

Table 2.2 Universal DNA amplification methods.

Advantages

Ampliﬁcation of the input material Yield target representing the entire gene pool of the community; therefore, no reduction in the complexity of the community Very low levels of bias [in comparison to multiplex polymerase chain reaction (PCR)] [3, 4] Enable the parallel detection of multiple genes

Disadvantages

No enrichment of the targeted gene

j 2 Long Oligonucleotide Microarray-Based Microbial Detection

38

2.2 Method

See Figure 2.1 for an overview of the method.

Figure 2.1 Method overview.

2.2.1 DNA Extraction

Essentially, any conventional DNA extraction kit or protocol can be used. Genomic DNA is dissolved in TE buffer pH 8.0 or water. 2.2.2 F29 Amplification

Note: Our protocol was developed using the GenomiPhi DNA Ampliﬁcation Kit (Amersham Biosciences (now GE Life Science) 25-6600-00/01/02). However, this kit was replaced with the newly developed illustra GenomiPhi V2 DNA Ampliﬁcation Kit (Amersham Biosciences 25-6600-30/31/32). We have not tested this new kit, but recommend following the manufacturers protocol since this worked ﬁne for the old kit. Protocol (for the kits 25-6600-00/01/02)

1. 2. 3. 4.

Mix 1 ml genomic DNA (20 ng/ml) with 9 ml sample buffer Incubate 3 min at 95 C Cool on ice Add 9 ml reaction buffer and 1 ml enzyme mix

2.2 Method

5. Mix and spin down 6. Incubate overnight (16–18 h) at 30 C [recommended incubation time for new kits (25-6600-30/31/32) is 1.5 h] 7. Purify using ProbeQuant G-50 Micro Columns (Amersham Biosciences 27-533501) and following the manufacturers instructions 8. Control DNA quality on an agarose gel (1 ml of the product is enough for the quality control since the product is concentrated high-molecular-weight DNA) Note: For ﬁrst-time users we would recommend to additionally perform a control reaction using 1 ml of control DNA provided with the kit as a template – in this case a ﬁnal yield of more than 4 mg DNA is expected. Price per reaction

GenomiPhi ampliﬁcation G-50 puriﬁcation

3.25 D (based on 100-reaction kit) 3.52 D (based on 50-reaction kit)

2.2.3 Klenow Amplification/Labeling Materials . .

BioPrime Array CGH Genomic Labeling System (Invitrogen 18095-011) Cy3-dCTP (Amersham Biosciences PA53021) or Cy5-dCTP (Amersham Biosciences PA55021)

Protocol

1. Dissolve 1.5 mg genomic DNA (or F29 ampliﬁed DNA) in 23 ml water or TE buffer pH 8.0 2. Add 20 ml 2.5 random primers solution 3. Incubate 5 min at 95 C 4. Cool 5 min on ice 5. Add 5 ml 10 dCTP nucleotide mix, 1 ml Cy-dCTP and 1 ml exo-Klenow fragment (40 U/ml) 6. Mix and spin down 7. Incubate 4 h at 37 C 8. Add 5 ml stop buffer 9. Add 45 ml of TE buffer pH 8.0 and purify using the BioPrime Array CGH puriﬁcation module, following the manufacturers instructions 10. Control quality of the target by measuring absorption at wavelengths 260, 280, 310 nm and 550 and 650 nm for Cy3 targets or 650 and 750 nm for Cy5 targets 11. Calculate the yield and the labeling efﬁciency using the following formula (adapted from BioPrime Array CGH Genomic Labeling System manual)

j39

j 2 Long Oligonucleotide Microarray-Based Microbial Detection

40

mðDNAÞ ¼ ðA260 A310 Þ 50 volume nt=dyeðCy3Þ ¼ ½ðA260 A310 Þ=ðA550 A650 Þ 22:7 nt=dyeðCy5Þ ¼ ½ðA260 A310 Þ=ðA650 A750 Þ 37:9 Note: Ampliﬁcation should yield approximately 5 mg of the labeled target (incorporation efﬁciency should be between 50 and 60 nucleotides/dye). Price per reaction

Klenow ampliﬁcation Cy-dCTP

8.67 D (based on 30-reaction kit) 12.44 D (based on 25-nmol pack size)

2.2.4 Probe and Slide Preparation

A wide range of software exists supporting the design of speciﬁc long oligonucleotide probes, the discussion of which is beyond the scope of this chapter (for a comprehensive list, see http://ihome.cuhk.edu.hk/b400559/arraysoft_probe.html). The critical points are: .

. .

Speciﬁcity. Ideally less than 50% sequence identity to any non-targeted sequence that may be present in the analyzed DNA at any point [1]. Avoid probes where all nucleotides with similarity to other sequences are clustered together. No or minimal hairpin and dimer formation. GC% in the range of 40–60% if possible and ideally similar to each other within the probe set.

We use custom synthesized oligonucleotide probes (VBC Biotech; www.vbc-biotech. com) dissolved to a ﬁnal concentration of 100 pmol/ml (100 mM). Different spotting buffers [50% dimethylsulfoxide (DMSO), 40% DMSO þ 12 mM NaxHyPO4 pH 8.0 þ 0.01% sodium dodecylsulfate (SDS) and ArrayIt Micro Spotting Solution] and different end concentrations of the oligonucleotide probes (50, 15 and 5 pmol/ml) were tested. Preliminary hybridizations revealed that the ArrayIt Micro Spotting Solution was the best since it produced homogenous spots of ideal size (approximately 100 mm with SMP3 pins). Spotting from 50% DMSO resulted in extremely small spots, whereas spotting buffer containing SDS caused extensive spreading of the spots on the surface. Regarding concentration of the spotted probes, 15 pmol/ml was sufﬁcient when pure cultures were analyzed. However, in order to enhance the sensitivity for environmental application and to achieve reliable detection of the targets present at the 0.1% relative abundance 50 pmol/ml concentration was needed. Customarily, we spotfrom the ArrayIt MicroSpotting Solutionat55% humidity. Regarding the surfaces, we tested two different amino-coated slides. Amine Silanated Slides from CEL Associates can be used for targets where strong hybridization signal is expected (e.g. pure culture targets). However, in comparison with the Corning

2.2 Method

GAPS II slides, Amine Silanated Slides have signiﬁcantly higher background signals impairing the sensitivity of detection. Therefore, for the detection of the targets present at low abundance, Corning GAPS II slides are recommended. If high sensitivity is not a critical issue, Amine Silanated Slides can be used, their main advantage being a signiﬁcantly lower price. Costs

Exact price per reaction cannot be calculated since it strongly depends on the number of the slides spotted. ArrayIt Micro Spotting Solution 2 (040315)

126.50 US$/50 ml

Oligonucleotide probes

0.40 D /nt (VBC Biotech; 0.01 mmol scale, high-performance liquid chromatography)

Corning GAPS II slide

10.60 D /slide (special prices also available)

CEL Associates Amine Silanated Slides

1.52 US$/slide

Spotted slides should be stored at room temperature (in the dark). The processing is carried out immediately before use. We did not perform any extended studies considering shelf-life of the spotted slides. However, we were using up to 3-month-old batches and did not notice any deterioration of quality. 2.2.5 Slide Processing Protocol (for Amino Surfaces)

Note: Prepare all solutions fresh immediately prior to processing. Materials . . . . . .

Blocking solution 1: dissolve 0.5 g sodium borohydride in 200 ml 2 SSC Wash solution: 5 SSC, 0.1% lauroylsarcosine Blocking solution 2: dissolve completely 3.6 g succinic anhydride in 200 ml methyl-pyrrolidinone and then add 8.96 ml 1 M sodium borate buffer pH 8.0 20 SSC 100 C dH2O –20 C EtOH

Protocol

1. UV cross-linking: 65000 mJ 2. Incubate slides in blocking solution 1 at 42 C (blocking solution should be pre-warmed) for 30 min with agitation 3. Wash 2 times 5 min in 1 SSC (with agitation) 4. Wash 3 times 5 min in 0.2 SSC (with agitation)

j41

j 2 Long Oligonucleotide Microarray-Based Microbial Detection

42

5. Rinse brieﬂy in wash solution 6. Incubate 15 min in blocking solution 2 at room temperature in the dark with agitation 7. Rinse 5 times with dH2O 8. Incubate 2 min in 100 C dH2O (without agitation) 9. Incubate 1 min in 20 C EtOH (without agitation) 10. Dry with oil-free air gun or by centrifugation (3 min at 900 rpm; Sorvall 75006445 rotor) Note: This protocol was optimized for Corning GAPS II slides. In our experience these slides may display an auto-ﬂuorescent background signal in the Cy3 channel. Therefore, an extended processing including additional sodium borohydride blocking may be necessary. However, this (i.e. steps 2 to 4) can be omitted if hybridization is performed with only Cy5-labeled target (and values from Cy3 channel are not taken into consideration). It is also unnecessary when other amino slides are used (i.e. from CEL Associates). Costs

Exact price per reaction is difﬁcult to calculate since it depends on the volume used and the number of slides processed at once (e.g. we use large coplin jars that require 200 ml of each solution and enable processing of up to 20 slides at once). As guidelines we list the prices of the reagents used for the processing: 20 SSC Sodium borohydride N-lauroylsarcosine Na-salt Succinic anhydride 1-Methyl-2-pyrrolidinone Boric acid

Invitrogen 15557-044 Merck 806373 Sigma-Aldrich L5777 Sigma-Aldrich 134414 Sigma-Aldrich M6762 Sigma-Aldrich 15663

18.00 D /l 15.20 D /25 g 45.40 D /50 g 19.50 D /500 g 23.70 D /l 18.60 D /250 g

2.2.6 Hybridization and Slide Washing

Two different hybridization buffers were tested with ﬁnal concentrations of 6 SSC, 1 Denhardts solution, 0.1% SDS and 100 mg/ml salmon sperm DNA, with and without 40% formamide. Both provided stringent hybridization conditions for our probe–target combination. However, the hybridization buffer containing 40% formamide resulted in somewhat lower background which was critical for the highsensitivity detection. The composition of the hybridization buffer, the hybridization temperature and the wash temperature all showed signiﬁcant inﬂuence on both hybridization stringency and background levels. A series of optimization experiments established that the buffer containing 40% formamide was yielding best results when used at 42 C, and buffer without formamide at 55 C.

2.2 Method

Washing was performed in four successive steps for 5 min each, using the following buffers: 2 SSC þ 0.1% SDS, 0.2 SSC (2 times) and 0.1 SSC. Initially, all steps were performed at room temperature. However, this resulted in a low level of cross-hybridization on the negative control spots. Different wash temperatures (50, 55 and 60 C; and 55 C for 15 min) for the ﬁrst stringent wash step (2 SSC þ 0.1% SDS) were tested to solve this problem. Wash at 55 C for 5 min proved to be an ideal compromise reducing the background signal and eliminating the weak falsepositive signals but not signiﬁcantly weakening the expected positive signal of low abundant target. Higher wash temperature (60 C) or longer duration of the wash step (15 min) had negative inﬂuence on the intensity of the positive signal of the low abundance target. The composition of the hybridization buffers given below is calculated for 50 ml target and a ﬁnal hybridization volume of 210 ml (this being optimal for our hybridization chambers). If different system is used, it is important to keep the same ﬁnal concentrations of the reagents (see values given in brackets) – hybridization volumes can be increased or decreased. Hybridization buffer 1 (40% formamide): for hybridization at 42 C

Formamide 20 SSC dH2O 50 Denhardts solution 10% SDS Salmon sperm DNA (10 mg/ml)

84.0 ml (40%) 63.0 ml (6) 4.6 ml 4.2 ml (1) 2.1 ml (0.1%) 2.1 ml (100 mg/ml)

Hybridization buffer 2 (no formamide): for hybridization at 55 C

dH2O 20 SSC 50 Denhardts solution 10% SDS Salmon sperm DNA (10 mg/ml)

88.6 ml 63.0 ml (6) 4.2 ml (1) 2.1 ml (0.1%) 2.1 ml (100 mg/ml)

Wash protocol (for both hybridization buffers/temperatures)

2 SSC þ 0.1% SDS 0.2 SSC 0.2 SSC 0.1 SSC

5 min, 5 min, 5 min, 5 min,

55 C room temperature room temperature room temperature

j43

j 2 Long Oligonucleotide Microarray-Based Microbial Detection

44

Costs

Costs of hybridization and wash buffers are negligible. Formamide Denhardts solution Salmon sperm DNA Hybridization chambers

Sigma-Aldrich 47671 Sigma-Aldrich D9905 Ambion AM9680 Grace Bio Labs HBW2240FL

29.30 D /250 ml 68.20 D /5 ml 228.00 D /10 10 mg 208.00 US$/100 pieces

2.2.7 Comments

The setup and optimization of the above-described protocols were performed using ﬁve test 70mer probes (see below). In our experience, new probe sets are likely to require some further optimization and reﬁnement of the protocol. The combination of the hybridization buffer, hybridization temperature and wash conditions is one of the ﬁrst points that we take into consideration when setting up a new system. Furthermore, if another surface chemistry is chosen (e.g. epoxy slides), retesting of the hybridization buffers and the concentration of the probes is recommended. 2.3 Our Test System and Results

In order to develop and test the above described protocol, a small system was set up containing four Salmonella-speciﬁc probes (targeting genes invA and sopB) and one control probe (targeting 16S rRNA gene of the Enterobacteriaceae family). The system was tested using genomic DNA of Salmonella enterica Senftenberg and Escherichia coli and their mixes in ratios 1 : 1, 1 : 10 and 1 : 100. The aim was to optimize the protocol to enable stringent differentiation of S. enterica Senftenberg and E. coli as well as the detection of S. enterica Senftenberg at the relative abundance of 1%. Furthermore, we wanted to establish a system able to work with relatively low amounts of input material (genomic DNA). The ﬁnal result is summarized in Figure 2.2. Highly speciﬁc (i.e. E. coli not giving any false-positive signals with Salmonella speciﬁc probes; Figure 2.2 C and D) and highly sensitive hybridization results (i.e. Salmonella-speciﬁc signals clearly detectable even in the cases when S. enterica Senftenberg was present at only 1% of relative abundance; Figure 2.2 (A and B) were achieved using the system and conditions described above. 2.4 Conclusions

The protocol described above makes it possible to detect target sequences from bacteria representing 1% of the whole microbial community without application of

2.4 Conclusions

Figure 2.2 Microarray images showing hybridization results with: 1% S. enterica serovar Senftenberg in 99% E. coli (A and B) and E. coli only (C and D). Images were scanned on a GenePix 4000A scanner (Axon Instruments) at 100% laser power, 1000 V photomultiplier tube gain and display results in a single color (A and C) or rainbow color mode (B and D). Each array contains the same probes in

triplicates. Uneven rows (1, 3, 5), left to right: sopB_1C6NH2, sopB_2C6NH2, invA_1C6NH2, invA_5C6NH2, Ent_16SC6NH2, sopB_1, sopB_2, invA_1, invA_5, Ent_16S. Even rows (2, 4, 6), left to right: empty, sopB_2C6NH2, invA_1C6NH2, invA_5C6NH2, Ent_16SC6NH2, empty, sopB_2, invA_1, invA_5, Ent_16S.

additional signal ampliﬁcation methods. Furthermore, the method allows for a minimal input (i.e. 20 ng of total DNA sufﬁce for two microarray hybridizations) and parallel detection of multiple genes without PCR and the bias inherent to it, as well as without a need for the development and optimization of multiplex PCR. We have additionally tested the potential offered by tyramide signal ampliﬁcation. However, our results indicated that there was no signiﬁcant improvement in the overall sensitivity of the array (i.e. 0.1% sensitivity level could not be established). Regarding high costs of such signal ampliﬁcation (approximately 120 D /reaction – based on the PerkinElmer Micromax TSA labeling and detection kit) we do not consider it as a cost-effective improvement and therefore it was not included as part of the protocol. Total (approximate) costs per reaction

45–50 D 38–43 D 35–40 D 28–33 D

(for (for (for (for

Corning slides) Corning slides without GenomiPhi ampliﬁcation) CEL Associates slides) CEL Associates slides without GenomiPhi ampliﬁcation)

j45

j 2 Long Oligonucleotide Microarray-Based Microbial Detection

46

Citing this work

If you use this method as a tool in your published research, we ask that you cite the following reference: T. Kostic, A. Weilharter, A. Sessitsch and L. Bodrossy (2005) High-sensitivity, polymerase chain reaction-free detection of microorganisms and their functional genes using 70-mer oligonucleotide microarray. Analytical Biochemistry 346, 333–335.

References 1 He, Z., Wu, L., LI, X., Fields, M.W. and Zhou, J. (2005) Empirical establishment of oligonucleotide probe design criteria. Appl. Environ. Microbiol., 71, 3753–3760. 2 Bodrossy, L. and Sessitsch, A. (2004) Oligonucleotide microarrays in microbial diagnostics. Curr. Opin. Microbiol., 7, 245–254. 3 Vora, G.J., Meador, C.E., Stenger, D.A. and Andreadis, J.D. (2004) Nucleic acid

ampliﬁcation strategies for DNA microarray-based pathogen detection. Appl. Environ. Microbiol., 70, 3047–3054. 4 Wu, L., Liu, X., Schadt, C.W. and Zhou, J. (2006) Microarray-based analysis of subnanogram quantities of microbial community DNAs by using wholecommunity genome ampliﬁcation. Appl. Environ. Microbiol., 72, 4931–4941.

j47

3 Sequence-Specific End-Labeling of Oligonucleotides Tanja Kostic and Levente Bodrossy 3.1 Introduction

Sequence-speciﬁc end-labeling of oligonucleotides (SSELO) is a novel labeling approach ﬁrst described by Rudi et al. [1]. This approach is based on the conventional short oligonucleotide microarray combined with an alternative labeling protocol. The main features of the SSELO approach are the high speciﬁcity and the high sensitivity of detection. These are ensured by several elements incorporated into this approach. The principle of the SSELO is depicted in Figure 3.1. Capture oligonucleotides are immobilized on the microarray. Reverse complements of the capture oligonucleotides (RC oligonucleotides) are end-labeled in a linear ampliﬁcation reaction, which depends upon the availability of the corresponding target sequence (i.e. PCR product). The entire mixture is then hybridized to the microarray to sort the sequences. During this hybridization event each RC oligonucleotide encounters its perfect binding partner on the array. In this way, using the SSELO approach we observe a shift of the speciﬁcity determining step. Thus, in classical microarray analysis, conditions of the hybridization and washing step predominantly inﬂuence the speciﬁcity of the probe–target binding. In SSELO, speciﬁcity is determined by the labeling reaction. The speciﬁcity of the SSELO probes could be further increased by the utilization of competitive oligonucleotides (COs; Figure 3.2). COs are designed as a variation of an RC oligonucleotide showing false-positive signals with non-target strains. The probe sequence is altered in a way to design a new probe that is a perfect match towards the strain exhibiting false-positive signals. COs should therefore have a higher speciﬁcity towards the sequence giving rise to the false-positive signal than the corresponding RC oligonucleotides. Furthermore, the target is composed only of the RC oligonucleotides, each of which has a perfect binding partner on the microarray. Therefore, the level of non-speciﬁc, background hybridization is decreased. In the case of classical microbial diagnostic microarrays (MDMs), in which the entire PCR product is labeled, non-speciﬁc, background hybridization is the major factor limiting detection sensitivity. Therefore, the current reported sensitivity threshold of the MDMs lies in the range of 1–5%

j 3 Sequence-Specific End-Labeling of Oligonucleotides

48

Figure 3.1 Principle of SSELO.

Figure 3.2 Principle of COs.

3.1 Introduction

j49

j 3 Sequence-Specific End-Labeling of Oligonucleotides

50

relative abundance [2–4]. Using the SSELO approach we were able to demonstrated detection sensitivity in the relative abundance range of 0.1% [5]. 3.2 Probe Design

For the probe design, we customarily use the ARB software package [6]. Probes were analyzed in silico using CalcOligo 2.03 [7] and OligoAnalyzer 3.0 (http://www.idtdna. com/analyzer/Applications/OligoAnalyzer/). The most important features that should be considered during probe design are: 1. Selection of the marker gene. The degree of conservation of the marker gene is the major factor inﬂuencing speciﬁcity of the microarray [8]. Another important issue is the number of available sequences, since the quality of the database will be reﬂected in the quality (speciﬁcity) of the designed probes. Our MDM for detection of pathogenic bacteria utilizes the gyrB gene encoding the subunit B of the bacterial gyrase. The gyrB gene was used because of its universal spread, low rate of horizontal transfer and high resolution. 2. Placement of the diagnostic mismatch(es) as close to the 30 -end as possible. As described before, the extension of the RC oligonucleotide probe by a single, ﬂuorescently labeled dideoxynucleotide (ddNTP) is the speciﬁcity determining step of the SSELO approach. Therefore the most crucial region is the 30 -end of the probe, followed by the central region; consequently, diagnostic mismatch(es) should be placed in these regions. 3. Cytosine as 30 -terminal residue. An additional speciﬁcity conferring feature is the 30 -terminal residue itself. In our case, we used only one residue at the 30 -end (i.e. cytosine). This allowed the application of stringent labeling conditions, using only labeled ddCTP in the combination with the other three unlabeled ddNTPs. This reduces the chance of unexpected labeling by 75%. However, this seriously limits the possibilities of probe design. Initial tests indicated that this rule could be relaxed without compromising the speciﬁcity of the probe set (for more detailed data, see [5]). Therefore, the use of more 30 -terminal residues could be taken into consideration (e.g. two different 30 residues, subsequently used with two different ﬂuorescent labels should not affect the speciﬁcity of the probe set). 4. Similar melting temperature (targeted Tm ¼ 60 2 C). Similar melting temperature of the probes is important as it ensures the uniformity of the probe set under the same experimental conditions [8]. 5. Length between 15 and 30 nucleotides. For the probe design in ARB, the probe length was set at 20 nucleotides. However, this was subsequently adjusted in order to account for other criteria (e.g. melting temperature, 30 cytosine). 6. No or minimal hairpin and dimer formation. A major problem for the SSELO approach is the possible self-priming of the oligonucleotides because it can result in the non-speciﬁc labeling and false-positive results. Analysis of the oligonucleotides

3.3 Slide Preparation (Spotting)

using OligoAnalyzer 3.0 allowed us to predict hairpin formation and self-priming potential, thus avoiding the potentially problematic probes. One of the remaining questions was the potential formation of the hetero-duplexes between different oligonucleotides in the probe set. Unfortunately, we did not ﬁnd any programs that could perform such analysis for a set of probes. As pairwise analysis would be extremely tedious, we chose probes with the lowest DG for hairpin and 30 dimer, and discarded probes which formed 30 dimers that were followed by guanine, as it was considered that this could lead to false labeling. Later on, we found that the Sequencher 4.6 software (Gene Codes Corporation) was able to provide good indication of potential heteroduplex formation and eventual false positive labelling when entire probe set was submitted to contig assembly using non-stringent assembly parameters. Nevertheless, although the above-listed criteria can greatly improve probe design, they are in no way absolutes. The probe set always needs to be thoroughly validated using both targeted strains as well as closely related non-targeted strains. In our experience, all probes with weighted mismatch (wMM) values below 0.5 (for more detailed information about the calculation of the wMM values, see [5]) will yield false-positive results and COs can be designed and applied up front. On the other hand, we have observed somewhat heterogeneous behavior of the probes with wMM values between 0.7 and 2.7, and in this range the necessity for the COs was determined experimentally. 3.3 Slide Preparation (Spotting)

Oligonucleotide probes for spotting were custom synthesized (VBC Biotech; www. vbc-biotech.com) with a 50 primary amino group, followed by a C6 spacer and ﬁve thymidine residues preceding the probe sequence. Probes were spotted from ArrayIt Micro Spotting Solution onto silylated aldehyde slides at a ﬁnal concentration of 50 mM. Spotting was performed at 55% relative humidity and 21 C. After spotting slides were stored in a humidity chamber (200 ml saturated NaCl in a 2-l airtight plastic box) until processing (12–24 h).

Costs

ArrayIt Micro Spotting Solution 2 (040315)

126.50 US$/50 ml

CEL Associates Silylated Aldehyde Slides

38.00 US$/25 slides

Oligonucleotide probes (VBC Biotech)

0.40 D /nt [0.01 mmol scale, high-performance liquid chromatography (HPLC)]

50 -Aminolink C6 modiﬁcation (VBC Biotech)

18.00 D /probe

j51

j 3 Sequence-Specific End-Labeling of Oligonucleotides

52

3.4 Slide Processing Protocol ( for Aldehyde Surfaces)

Aldehyde slides should be processed 12–24 h after spotting. Note: Prepare blocking solution freshly before processing. Materials . . . .

0.2% SDS 100 C dH2O PBS buffer pH 7.4 (for 1 l dissolve 8.0 g NaCl, 2.0 g KCl, 1.44 g Na2HPO4 and 0.24 g KH2PO4; adjust pH to 7.4 with HCl) Blocking solution: dissolve completely 0.5 g sodium borohydride (Merck 806373) in 150 ml PBS pH 7.4 and then add 44 ml ethanol

Protocol

1. 2. 3. 4. 5. 6. 7. 8. 9.

Rinse 2 2 min in 0.2% SDS at room temperature with vigorous agitation Rinse 2 2 min in dH2O at room temperature with vigorous agitation Denature DNA 2 min in 100 C dH2O Cool the slides at room temperature (approximately 5 min) Block for 5 min in the blocking solution at room temperature (perform this step in the fume hood) Rinse 3 1 min in 0.2% SDS at room temperature with vigorous agitation Rinse 1 min in dH2O at room temperature with vigorous agitation Dry with oil-free air gun or by centrifugation (3 min at 900 rpm; Sorvall 75006445 rotor) Store desiccated at room temperature in dark

When using this protocol, please refer to Stralis-Pavese et al. [7].

3.5 DNA Extraction and PCR Amplification of the Targeted Gene

These ﬁrst two basic steps are dependent on the type of organism or environment of interest and on the nature of the targeted marker gene. Any conventional DNA extraction kit or protocol can be used and the PCR reaction should be optimized for the gene of interest. PCR products should be puriﬁed using a commercial PCR puriﬁcation kit (e.g. QIAquick PCR Puriﬁcation Kit; QIAGEN 28106; 1.55 D /sample – based on 250reaction kit). After puriﬁcation, the DNA concentration of the PCR product should be measured (e.g. NanoDrop spectrophotometer; NanoDrop Technologies) and adjusted to a ﬁnal concentration of 50 ng/ml.

3.7 Labeling

3.6 Shrimp Alkaline Phosphatase Treatment

Puriﬁed PCR products are treated with shrimp alkaline phosphatase (SAP) in order to dephosphorylate any remaining nucleotides that could interfere with the speciﬁcity of the following labeling step. Reaction mix

Puriﬁed PCR product (50 ng/ml) 10 Thermo Sequenase buffer SAP (1 U/ml)

20.0 ml 2.0 ml 4.0 ml

Incubate 30 min at 37 C, followed by inactivation for 10 min at 95 C. Costs

10 Thermo Sequenase buffer

Amersham Biosciences 25-0200-69

126 D /10 ml

SAP

Roche 11758250

115 D /1000 U

3.7 Labeling

For labeling, a set of RC oligonucleotides lacking 30 -terminal nucleotide was custom synthesized (VBC Biotech) and dissolved to a ﬁnal concentration of 100 pmol/ml. Competitive oligonucleotides were synthesized with 30 -terminal nucleotide and with additional 30 -phosphate modiﬁcation. This was included in order to disable elongation/labeling and to ensure the silencing feature. The oligonucleotide mix (RC mix) used in the labeling reaction contained each RC oligonucleotide and each CO at a ﬁnal concentration of 1 pmol/ml. However, further tests have shown that using the RC mix at a ﬁnal concentration of 0.1 pmol/ml for each oligonucleotide does not decrease performance of the assay (for more detailed data, see [5]). The use of the silencing and labeled dideoxynucleotides (ddNTPs) is dependent on the setup of the system. In our case we only used one 30 -terminal residue (i.e. cytosine) and therefore for the labeling we used only labeled ddCTP. The silencing mix contained the remaining non-labeled ddNTPs at a ﬁnal concentration of 10 pmol/ml.

j53

j 3 Sequence-Specific End-Labeling of Oligonucleotides

54

Materials/Costs

Tamra-ddCTP ddNTP set Taq polymerase Oligonucleotides 30 -Phosphorylation

PerkinElmer NEL 473 Roche 3732738 Invitrogen 10342-053 VBC Biotech (0.01 mmol scale, HPLC) VBC Biotech (0.01 mmol scale)

170.00 D /2.5 nmol 183.50 D /4 1 mmol 46.00 D /100 U 0.40 D /nt 20.00 D /probe

Reaction mix

SAP-treated PCR product (50 ng/ml) SAP-treated control PCR product (5 ng/ml) 10 Thermo Sequenase buffer RC mix (1 mM for each oligo) ddNTPs-C (10 mM for each ddNTP) Tamra-ddCTP (100 mM) Taq (5 U/ml) dH2O

2.0 ml (100 ng) 2.0 ml (10 ng) 1.0 ml (1 ) 1.0 ml (1 pmol each) 1.0 ml (10 pmol each) 0.1 ml (10 pmol) 0.6 ml (3 U) 2.3 ml

Labeling reaction (25 cycles)

30 s 95 C 75 s 60 C

3.8 Hybridization and Slide Washing

Given that hybridization and washing conditions are not the speciﬁcity determining step in this approach, we have initially tested standard conditions from our other applications. These also worked well for this approach. Variations of hybridization and wash temperature did not result in any signiﬁcant improvement of hybridization and therefore the original conditions were maintained. Hybridization buffer

dH2O 20 SSC 50 Denhardts solution 10% SDS

130.7 ml 63.0 ml (6) 4.2 ml (1) 2.1 ml (0.1%)

Labeled targets (10 ml) were mixed with 200 ml hybridization buffer and hybridization was performed at 55 C for 2 hours or overnight.

3.11 Microarray for Detection of Pathogenic Bacteria

Wash protocol

2 SSC þ 0.1% SDS 0.2 SSC 0.2 SSC 0.1 SSC

5 min, room temperature 5 min, room temperature 5 min, room temperature 5 min, room temperature

Microarrays were scanned immediately after washing at 10 mm resolution using GenePix 4000A scanner (Axon Instruments). Photomultiplier tube gain was adjusted to scan the spots below saturation levels. Images were saved as multilayer tiff and analyzed using the GenePix Pro 6.0 software package (Axon Instruments).

3.9 Data Analysis

In order to normalize and compare the results from different experiments, every sample was spiked with an internal control. In our case, we used a pmoA PCR product from Methylosinus trichosporium OB3b. The labeling mix also contained RC oligonucleotide (Msi_294) targeting this sequence. Microarray hybridization results were normalized to the signal obtained from this internal control expressed as a percentage: 100% equaling the signal of the control probe. Probes were considered to be positive during validation if their normalized signal was at least 10% of the control signal, Msi_294.

3.10 Costs

The exact price per reaction depends on the number of probes used on the array and the number of samples analyzed. Therefore, we can give only an estimation based on a hypothetical probe set containing 80 probes and 10 COs (average probe length 20 nucleotides) and on the analysis of 500 samples (see Table 3.1).

3.11 Microarray for Detection of Pathogenic Bacteria

Using the above described method we have developed MDM for the detection of pathogenic bacteria. Our probe set contains 65 oligonucleotides and 11 COs targeting the gyrB gene of 25 of the most important waterborne pathogens and indicator organisms (manuscript in preparation). Citing this work

If you use this method as a tool in your published research, we ask that you cite this reference: T. Kostic, A. Weilharter, S. Rubino, G. Delogu, S. Uzzau, K. Rudi,

j55

j 3 Sequence-Specific End-Labeling of Oligonucleotides

56

Table 3.1 Estimated costs.

Approximate price/sample DNA extraction (based on average extraction kit) PCR ampliﬁcation PCR puriﬁcation SAP treatment Slides Capture oligonucleotides 80 probes 20 nt 0.40 D /nt ¼ 640 D 80 probes 18.00 D /50 –NH2C6 ¼ 1440 D RC and CO oligonucleotides 80 probes 20 nt 0.40 D /nt ¼ 640 D 10 COs 20 nt 0.40 D /nt ¼ 80 D 10 COs 20.00 D /30 –PO4 ¼ 200 D Labeling (with only one labeled nucleotide) Labeling (with all four labeled nucleotides) Hybridization chamber Total price/sample (including materials for slide processing, hybridization and wash buffers)

2.00–3.00 D 0.50 D 1.55 D 0.50 D 1.52 US$ 4.00 D

1.85 D

1.70 D 3.50 D 2.08 US$ 18–20 D

A. Sessitsch and L. Bodrossy (2007) A microbial diagnostic microarray technique for sensitive detection and identiﬁcation of pathogenic bacteria in a background of nonpathogens. Analytical Biochemistry 360, 244–254.

References 1 Rudi, K., Treimo, J., Nissen, H. and Vegarud, G. (2003) Protocols for 16S rDNA array analyses of microbial communities by sequence-speciﬁc labeling of DNA probes. Scientiﬁc World Journal, 3, 578–584. 2 Bodrossy, L., Stralis-Pavese, N., Murrell, J.C., Radajewski, S., Weilharter, A. and Sessitsch, A. (2003) Development and validation of a diagnostic microbial microarray for methanotrophs. Environ. Microbiol., 5, 566–582. 3 Denef, V.J., Park, J., Rodrigues, J.L., Tsoi, T.V., Hashsham, S.A. and Tiedje, J.M. (2003) Validation of a more sensitive method for using spotted oligonucleotide DNA microarrays for functional genomics studies on bacterial communities. Environ. Microbiol., 5, 933–943.

4 Tiquia, S.M., Wu, L., Chong, S.C., Passovets, S., Xu, D., Xu, Y. and Zhou, J. (2004) Evaluation of 50-mer oligonucleotide arrays for detecting microbial populations in environmental samples. Biotechniques, 36, 664–665. 5 Kostic, T., Weilharter, A., Rubino, S., Delogu, G., Uzzau, S., Rudi, K., Sessitsch, A. and Bodrossy, L. (2007) A microbial diagnostic microarray technique for sensitive detection and identiﬁcation of pathogenic bacteria in a background of nonpathogens. Analytical Biochemistry, 360, 244–254. 6 Ludwig, W., Strunk, O., Westram, R., Richter, L., Meier, H., Yadhukumar, Buchner, A., Lai, T., Steppi, S., Jobb, G., Forster, W., Brettske, I., Gerber, S.,

References Ginhart, A.W., Gross, O., Grumann, S., Hermann, S., Jost, R., Konig, A., Liss, T., Lussmann, R., May, M., Nonhoff, B., Reichel, B., Strehlow, R., Stamatakis, A., Stuckmann, N., Vilbig, A., Lenke, M., Ludwig, T., Bode, A. and Schleifer K.H. (2004) ARB: a software environment for sequence data. Nucleic Acids Res., 32, 1363–1371. 7 Stralis-Pavese, N., Sessitsch, A., Weilharter, A., Reichenauer, T., Riesing, J.,

Csontos, J., Murrell, J.C. and Bodrossy, L. (2004) Optimization of diagnostic microarray for application in analysing landﬁll methanotroph communities under different plant covers. Environ. Microbiol., 6, 347–363. 8 Loy, A. and Bodrossy, L. (2006) Highly parallel microbial diagnostic using oligonucleotide microarrays. Clin. Chim. Acta, 363, 106–119.

j57

j59

4 Non-Cognate Approaches for Pathogen Detection on Microarrays Antoine Huyghe, Patrice Francois, Yvan Charbonnier, David Hernandez, Jonathan Hibbs, and Jacques Schrenzel

4.1 Introduction

Since agents used as bioterror weapons are very infrequent, general knowledge on these organisms remains sparse. In addition, information about speciﬁc marker(s) for molecular epidemiological assessments are lacking for these pathogens. Diagnostic possibilities for detecting these highly pathogenic and potent microorganisms are generally focused at one or possibly a limited number of species. Identiﬁcation results can be rapidly and accurately delivered by using quantitative PCR assays (qPCR). Unfortunately, qPCR-based techniques display several limitations, including the prerequisite knowledge of target sequences for appropriate assay design. Furthermore, they suffer from limited multiplexing capabilities. For instance, current qPCR-based assays allow only detection of a handful of targets at a time. qPCR approaches can therefore be designed and validated when the threat has been a priori deﬁned and well characterized. This approach is ideal when used as a defensive procedure, such as during anthrax threats and attacks. Under these conditions, sampling the environment for the presence of the bio-agent will allow fast decisions for decontamination procedures. However, the simple detection of the organism is not sufﬁcient – one needs also to check for the presence of its toxins [1] when assessing its pathogenic potential. Therefore, when considering a broader panel of potential bio-agents, one is forced to employ numerous PCR assays. Alternatively, the development of microarrays can overcome the low multiplexing capabilities of qPCR assays. Different microarray-based approaches have been suggested [2, 4, 11] and reported useful [9], but none appears amenable to routine diagnosis.

j 4 Non-Cognate Approaches for Pathogen Detection on Microarrays

60

4.2 Non-Cognate Hybridization System 4.2.1 Concept

We suggest here a generic innovative detection strategy named the non-cognate hybridization system (NCHS) to provide strain identiﬁcation without any a priori sequence knowledge. By synthesizing all probes of a given length that can be generated by a combination of four nucleotides, microarrays could detect any single organism. Unfortunately, for generating realistic probe lengths (i.e. permitting unique sequence identiﬁcation as well as adequate hybridization behavior), the total number of permutations of the four nucleotides would yield very large numbers of possible probes (i.e. 413 ¼ 67 108 864 probes for 13mer nucleotides). Synthesis of such large microarrays is currently technically impossible. Furthermore, bioinformatics tools that would be necessary for microarray analysis of billons of probes are also not yet available. Thus, reducing the probe set and the complexity of the analytical approach is warranted. Random reduction of the probe set to a smaller number of probes might result in potentially missing sequences characteristic for given pathogens. An alternative would be the pruning of low informative probes (i.e. probes that have no targets in a collection of selected microorganisms). However, this approach yields ultimately to another cognate design, with the inherent limitations described above. Another way to reduce the probe set is to consider permutations of only two nucleotides for the capture sequences. This divides the probe set by 2(n 2), where n is the length of the probes. This microarray design is truly non-cognate and guarantees that no organism is a priori favored or missed. The ﬁrst challenge is to deﬁne an optimal probe length, providing enough target-speciﬁc information, while allowing array manufacturing and permitting adequate hybridization. 4.2.2 Definition of the Optimal Probe Length

Analysis of all available bacterial genomes revealed that the optimal probe length was 13. This value represents the best balance between the number of recognized targets detecting studied bacterial genomes (assuming perfectly homologous hybridization) and the technological cost of adding extra nucleotides to the probe length. Figure 4.1 shows the average count of probes that can provide perfect detection of bacterial genomes on dinucleotide arrays (red curve) and the percentage of the total amount of probes on the array involved in detection. The red curve shows a maximum using 14mers; however, only 8.5% of the probes on a 14mer chip would be involved in detection (homologous hybridization). For a 13mer chip the total count of matching probes is only slightly lower but the percentage of probes involved in the detection doubles to 16%! The performance of such arrays in ﬂora analysis can be evaluated by computing the worst discrimination capability of the design. The worst discrimination capability is

4.2 Non-Cognate Hybridization System

Figure 4.1 Average probe count and percentage of chip coverage. The red curve shows the average count of probes that identify targets (i.e. that are perfectly homologous) in each of the 92 fully sequenced bacterial genomes subjected to analysis. Data are shown using increasing probe length (x-axis) with the number of

possible dinucleotide permutations in parentheses. The blue curve shows the percentage of array probes that provide genome coverage (i.e. that are perfectly homologous to existing targets). This value is determined by the average count of probes divided by the total number of probes required for all permutations of dinucleotides of that length.

given by the number of microorganisms that can be concurrently hybridized on the array and that would still provide non-completely overlapping patterns. A 13mer array would thus be able to concomitantly identify at least six different microorganisms. In reality, this value should never be achieved as conserved sequences can be found in any microorganism. Thus, the discrimination capability of the design is dramatically increased by this speciﬁcity. 4.2.3 Virtual Assessment of Array Performances (in Silico Experiments)

In silico experiments were performed to validate the NCHS approach for bacterial identiﬁcation and genotyping. Using 92 available bacterial genomes, we extracted every target sequence that would perfectly hybridize to the 13mer NCHS design. Every bacterium produced a complex pattern called virtual hybridization pattern. When compared to each other, the patterns revealed to be partly non-overlapping, even if the bacteria are closely related. To further determine the discrimination capabilities of the NCHS design, we performed virtual hybridization of two Pseudomonas species (Figure 4.2). In accordance with their high GC content, most of the informative probes were indeed composed of G and C nucleotides. Interestingly, 6371 probes discriminated the two species, predicting genotyping capabilities of an ideal NCHS chip. Similar virtual hybridization was obtained using two strains of the

j61

j 4 Non-Cognate Approaches for Pathogen Detection on Microarrays

62

Figure 4.2 Virtual hybridization of two species of Pseudomonas and their differential hybridization pattern. This figure shows in silico hybridization (red) of P. aeruginosa and P. putida on the four different types of NCHS chips (13mer dinucleotides). The differential pattern (green) represents probes that hybridize exclusively to either one of the two bacteria, yielding to a total of 6371 informative probes for the four NCHS chips.

AT-rich Staphylococcus aureus. Strains N315 and Mu50 share very similar genome contents [6], but can be differentiated at the strain level. According to virtual hybridization on 13mer NCHS chips, these two strains should be differentiated by 216 discriminating probes (Figure 4.3). These in silico experiments underline the potential of NCHS arrays as a bacterial identiﬁcation system as well as a generic typing tool. The potential resolution of the method might even differentiate a sequenced strain from that of a genetically engineered derivative.

Figure 4.3 Virtual hybridization of two related strains of S. aureus and their differential hybridization pattern. This figure shows in silico hybridization (red) of the related S. aureus strain N315 (a methicillin-resistant bacteria) and S. aureus strain Mu50 (also methicillin-resistant, but displaying

decreased susceptibility to glycopeptides) on the four different types of NCHS chips (13mer dinucleotides). The differential pattern (green) represents probes that hybridize exclusively to either one of the two strains, yielding to a total of 216 informative probes for the four NCHS chips.

4.2 Non-Cognate Hybridization System

4.2.4 Array Manufacturing and Hybridization (Wet-Lab Experiments)

The 13mer NCHS arrays were manufactured, and consisted of (i) all permutations of AG and (ii) all permutations of AC (i.e. 2 8192 probes). The selection of these two probe compositions (AG and AC) results from further in silico experiments that have shown a quasi-saturation of the AT and GC probes (not shown). These probes were therefore considered as poorly informative. Manufactured arrays (Agilent; www.agilent.com) were hybridized using standard hybridization parameters and yielded very reproducible signals (Figure 4.4), stressing that NCHS arrays might provide reliable identiﬁcation tools in a biothreat context. Nevertheless, signal intensities were strongly biased by GC contents rather that by probe speciﬁcity, as expected when using standard hybridization protocols. Current experiments are underway to minimize the inﬂuence of the GC contents by using tetramethylammonium chloride during washings or hybridization [10]. Hybridization parameters will be adapted to achieve optimal speciﬁcity, i.e. optimal correlation between in silico and wet-lab experiments. In any case, a suboptimal correlation would not preclude adequate identiﬁcation, provided adequate analytical methods are employed [3].

Figure 4.4 Reproducibility of microarray experiments performed on the NCHS chips. Reproducibility of microarray experiments by hybridizing 1 mg of S. aureus genomic DNA were fluorescently labeled with Cy3 or Cy5 using the Klenow method [5] and com-

petitively hybridized on the NCHS chip using the hybridization buffer provided by Agilent. Raw fluorescence intensities are plotted from two representative experiments and show a correlation coefficient of 0.90.

j63

j 4 Non-Cognate Approaches for Pathogen Detection on Microarrays

64

4.2.5 Analysis

Analysis implementation is based on advanced pattern recognition of hybridization signals [8], using self-organizing maps (SOMs). SOM structures the microorganism signatures in a geometrical manner [7]. This ensures that closely related microorganisms (in terms of hybridization patterns) will reside in close map position with respect to related and known organisms. Additionally if the SOM is trained with mixed microorganism patterns, the SOM should be able to identify the bacterial mixture. Finally, the settings of the SOM can be easily updated as new organism sequences become available.

4.3 Perspectives

The ﬁnal goal will be the utilization of NCHS-based microarrays for the detection, identiﬁcation and deﬁnition of bacterial genetic background in the presence of DNA of unknown origin. As separate beneﬁts, NCHS arrays will provide generic typing tools, usable in various epidemiological settings and might prove helpful in metagenomics. Such approaches will give insights into problems such as genome evolution and occupation of particular niches whose study is currently hindered by our inability to culture most microorganisms.

Acknowledgments

This work was supported by grants from the COST B28 program [COST C05.0103 (J.S.)] and the Swiss National Science Foundation [PP00B-103002/1 and 3100A0112370/1 (J.S.), 3100A0-116075/1 (P.F. and J.S.), and 3100A0-116075/1 (P.F.)].

References 1 Bell, C.A., Uhl, J.R., Hadﬁeld, T.L., David, J.C., Meyer, R.F., Smith, T.F. and Cockerill, F.R. III (2002) Detection of Bacillus anthracis DNA by LightCycler PCR. J. Clin. Microbiol., 40, 2897–2902. 2 Bodrossy, L. and Sessitsch, A. (2004) Oligonucleotide microarrays in microbial diagnostics. Curr. Opin. Microbiol., 7, 245–254. 3 Francois, P., Charbonnier, Y., Jaquet, J., Utinger, D., Bento, M., Lew, D.P., Kresbach, G., Schlegel, W. and

Schrenzel, J. (2005) Rapid bacterial identiﬁcation using evanescent waveguide oligonucleotide microarray classiﬁcation. J. Microbiol. Methods, 65, 390–403. 4 Hashsham, S.A., Wick, L.M., Rouillard, J.M., Gulari, E. and Tiedje, J.M. (2004) Potential of DNA microarrays for developing parallel detection tools (PDTs) for microorganisms relevant to biodefense and related research needs. Biosens. Bioelectron., 20, 668–683.

References 5 Koessler, T., Francois, P., Charbonnier, Y., Huyghe, A., Bento, M., Dharan, S., Renzi, G., Lew, D., Harbarth, S., Pittet, D. and Schrenzel, J. (2006) Use of oligoarrays for characterization of community-onset methicillin-resistant Staphylococcus aureus. J. Clin. Microbiol., 44, 1040–1048. 6 Kuroda, M., Ohta, T., Uchiyama, I., Baba, T., Yuzawa, H., Kobayashi, I., Cui, L., Oguchi, A., Aoki, K., Nagai, Y., Lian, J., Ito, T., Kanamori, M., Matsumaru, H., Maruyama, A., Murakami, H., Hosoyama, A., Mizutani-Ui, Y., Takahashi, N.K., Sawano, T., Inoue, R., Kaito, C., Sekimizu, K., Hirakawa, H., Kuhara, S., Goto, S., Yabuzaki, J., Kanehisa, M., Yamashita, A., Oshima, K., Furuya, K., Yoshino, C., Shiba, T., Hattori, M., Ogasawara, N., Hayashi, H. and Hiramatsu, K. (2001) Whole genome sequencing of methicillin-resistant Staphylococcus aureus. Lancet, 357, 1225–1240.

7 Liou, C.Y. and Tai, W.P. (1999) Conformal self-organization for continuity on a feature map. Neural Netw., 12, 893–905. 8 Schrenzel, J. and Hibbs, J. (2003) Noncognate hybridization system (NCHS). 00/75377 A2(6,544,777). 9 Sergeev, N., Volokhov, D., Chizhikov, V. and Rasooly, A. (2004) Simultaneous analysis of multiple staphylococcal enterotoxin genes by an oligonucleotide microarray assay. J. Clin. Microbiol., 42, 2134–2143. 10 Sorg, U., Enczmann, J., Sorg, R. and Wernet, P. (1991) Rapid non-radioactive TMACl hybridization protocol employing enzymatically labeled oligonucleotides. Nucleic Acids Res., 19, 4782. 11 Stenger, D.A., Andreadis, J.D., Vora, G.J. and Pancrazio, J.J. (2002) Potential applications of DNA microarrays in biodefense-related diagnostics. Curr. Opin. Biotechnol., 13, 208–212.

j65

j67

5 Patterning Techniques for Array Platforms Erhan Pi¸skin, Bora Garipcan, G€okhan Demirel, and Og uzhan Çaglayan

5.1 Introduction

Thousands of genes and the products they encode (i.e. proteins) function in a complicated and orchestrated way that creates the mystery of life of all organisms. In genomics, the genes of organisms, their functions and activities are investigated. Genomics is naturally linked to proteomics – the study of the proteins encoded by the organisms genome. Genomics in combination with proteomics resulted in fascinating biomedical research; however, it requires large-scale and high-throughput methodologies [1–3]. High-throughput techniques using DNA and protein microarrays have accelerated the process of understanding gene and protein functions in living organism. Macrochips contain sample spot sizes of about 300 mm or larger, and can be easily imaged by existing gel and blot scanners. The sample spot sizes in a microchip are typically less than 200 mm in diameter. The further improvement and developments in array technology led to the introduction of the term nanoarray technologies in the literature. Nanometric biomolecular arrays may enable high-throughput screening of biomolecules at the single-molecule level. Also, with precise control on position and orientation of individual molecules, such arrays may become powerful tools for studying multivalent and multicomponent molecular interactions in biological systems [4]. To these ends, protein arrays and/or DNA arrays with feature sizes smaller than 100 nm have been fabricated by novel techniques, mostly using dip-pen nanolithography (DPN), conductive atomic force microscopy (cAFM) nanolithography and nanografting [4, 5]. One of the essential elements of the arrays at any size range is of course patterning of the substrate surfaces. Surface patterning is a method to create two-dimensional predesigned heterogeneous surfaces with different/desired morphology, hydrophilic/ hydrophobic characteristics and/or chemical functionality within well-deﬁned regions in the micro- or nanoscale, as schematically exempliﬁed in Figure 5.1. Different techniques (including soft lithography, photolithography, dip-pen lithography, cAFM

j 5 Patterning Techniques for Array Platforms

68

Figure 5.1 Schematic illustration of various surface patterns.

lithography, nanoshaving, nanografting, etc.) have been developed to create novel patterned surfaces, which are brieﬂy described in this chapter together with some recent interesting applications. We focus here mainly on preparation of array platforms for biorecognition of microorganisms.

5.2 Soft Lithography

Soft lithography refers a group of methods, which include several approaches such as microcontact printing (mCP), replica molding, microtransfer molding [6], micromolding in capillaries [7], solvent-assisted micromolding [8] and near-ﬁeld phase shift lithography [9, 10] In general, the common feature in soft lithography approaches is using an elastomeric stamp with pre-patterned structures on its solid surface. This creates patterns and structures with sizes in the range of 30 nm to 100 mm on the target array platform surfaces. As schematically described in Figure 5.2, there are a series of common steps in soft lithography; for instance, mCP includes the following steps [10]. (1) A master pattern surface (usually silicon), which will be used as a mold, is prepared by different techniques (including photolithography [11], e-beam patterning [12], reactive ion etching [13], etc.) in a selected design that will be the ﬁnal pattern on the substrate surface. (2) A liquid pre-polymer (e.g. polydimethylsiloxane) is then cast on the structured master surface, and cured (for cross-linking) by applying heat and/or radiation [i.e. ultraviolet (UV) radiation] to reach a cross-linked elastomeric structure, which is subsequently used as the elastomeric stamp in the further steps. (3) The stamp is then treated (inked) with the solution containing the molecules to be printed. (4) In the ﬁnal step, the molecules are transferred by printing onto the substrate. Once the stamp has been inked and dried, the stamp is then brieﬂy pressed onto a solid substrate via mechanical contact and the ink molecules transfer from the polymeric stamp to

5.2 Soft Lithography

Figure 5.2 Schematic description of the mCP protocol: (1) preparation of the master surface, (2) preparation of the elastomeric stamp, (3) inking the patterned surface of the stamp and (4) printing (transfer of the ink) on the substrate surface.

the solid substrate where they self-assemble into pre-determined patterns by the relief patterns of the stamp. Soft lithography has attracted a lot of research interest and is regarded as one of the most widely used patterning techniques due to the following advantages: low cost, easy to prepare, straightforward to apply and accessible, operates with a wide range of controllable surface chemistry options, many features can be printed simultaneously with one stamp application, a relatively high spatial resolution of features produced (line widths of less than 100 nm), large printing capability to form patterned microstructures on non-planar surfaces and, ﬁnally, it does not need a photoreactive surface to create a nanostructure. Nevertheless, it has also limitations, which will eventually be solved by future technological developments. The disadvantages are the restrictions in the resolution of the resulting patterns (which depends on the material and dimensions of the elastomeric stamp), the deformation and distortion of the elastomeric stamp, and the limitations concerning the reproducibility, which is dependent on the stamps resistance to degradation. There are many reports on the preparation of patterned surfaces for a wide variety of application using soft lithography [14–16], Some examples are given below. Howell et al. designed and constructed patterned microarrays carrying antibody probes for detection of several bacteria (i.e. Escherichia coli and Renibacterium salmoninarum) [17]. They were able to demonstrate the selective binding of the bacteria on the regions containing speciﬁc antibodies. Weibel et al. developed a series of bacterial stamps [18]. They started from an agarose stamp, attached the bacteria on the patterns from bacterial suspensions and then printed (transferred) the bacteria

j69

j 5 Patterning Techniques for Array Platforms

70

onto agar media in culture plates. They were able to deposit Vibrio ﬁscheri colonies onto agar surface with different pattern designs.

5.3 Photolithography

Photolithography is one of the most widely studied lithography techniques and is a non-contact printing approach that has the following two advantages over mechanical contact printing methods: (i) reduced contamination and (ii) higher throughput [19–23]. It uses photoresistive masks and photoactive materials/molecules. In one of the most common approaches, the substrate surface carrying functional groups is ﬁrst coated (usually spin-coated) with an inert but photoactive polymer layer. Then, it is exposed to UV light through a photomask (having patterned openings). As such, the UV-exposed areas are cross-linked, and the non-cross-linked parts are dissolved and removed from the surface. The created openings (patterns) can be used for further probe immobilization steps. Alternatively, substrate surfaces can also be coated ﬁrst with photoactive molecules that carry functional groups for probe immobilization. The photomask is then placed onto the surface and the light activates or inactivates the surface photosensitive molecules through the openings in the mask. Thus, probe molecules are immobilized onto the activated molecules in a patterned way. A similar approach has been applied for manufacturing DNA arrays. Here, oligodeoxynucleotide (ODN) probes are synthesized in situ using photolithographic techniques and modiﬁed ODN synthesis chemistry [24, 25]. As schematically presented in Figure 5.3, the array substrate surface is coated with photosensitive modiﬁed

Figure 5.3 In situ synthesis of DNA arrays by photolithography: (1) attaching first layer of photosensitive nucleotides, (2) activating the desired parts by using photomasks and attaching the second layer of nucleotide via these activated species, and (3) using the second mask, for further similar steps.

5.4 Robotic Printing

nucleotides and then the selected parts are activated by turning the light on using a photomask (Mask 1). Subsequently, the second layer is attached via these activated molecules. The procedure is continued in a similar way until ODNs have reached the desired length. This state-of-the-art methodology provides the possibility to interrogating hundreds of thousands of sequences per assay, and allows wide ﬂexibility in design of custom arrays without preliminary efforts in synthesis and maintenance of a vast library of modiﬁed ODNs [26]. On the other hand, in situ synthesis does not allow quality control and puriﬁcation of generated features, and the less-efﬁcient phosphoramidite monomer coupling gives poor yields for longer ODN probes.

5.4 Robotic Printing

There are several robotic printing techniques able to create patterned surfaces and immobilize probes (bioligands) on patterns. Usually a computer-controlled robotic holder (arm) or motion-control print-head, with different types of pins carrying the probes molecules, moves on the substrate surface and delivers the probes to the designated areas. As the material is deposited in liquid form, water-based chemistry, which is common in biology, can be used. Here, we brieﬂy discuss two common robotic printing technologies – micro-spotting and ink-jet printing. 5.4.1 Micro-Spotting

Micro-spotting technology is widely used for manufacturing nucleic acid and protein arrays [27, 28]. A typical system consists of a motion-control system ﬁtted with a holder or print-head and one or more micro-spotting pins. As depicted schematically in Figure 5.4, the pin is ﬁrst brought to the target point (1); it is moved in the

Figure 5.4 Delivery mechanism of micro-spotting.

j71

j 5 Patterning Techniques for Array Platforms

72

z-direction, usually close to or even on the surface, and leaves a droplet of liquid behind (2). Then, the pin is moved to another point (3). A variety of different pins are available. The volume of the liquid that can be loaded on a pin (loading volume) and the volume of the droplet delivered (print volume) are in the range of 0.2–1.0 ml and 0.5–2.5 nl, respectively. Spot size can vary from 50 to 350 mm and spot density from 500 to 10 000 spots/cm2. All these properties are dependent on the pin design and droplet properties (e.g. surface tension, viscosity, density, etc.). The volume of the droplet delivered cannot be programmed, which is one of the main disadvantages of micro-spotting; however, it is very simple compared to injection units and signiﬁcantly cheaper. The number of pins that can be combined on a printing-head can be up to 64. The pins are usually made from brass and stainless steel, which provide superior gliding properties and durability. Micro-spotting pins can be made by using electrical discharge machining (EDM) to hold tolerances to within a few tenths of a millimeter [29]. Here, the spot diameter is largely dependent on the dimensions of the pin tip, which can be controlled by EDM. There is little or no contact between the pin and the surface in order to deliver the sample, which minimizes or even eliminates the impact forces and increases the lifespan of the pins to several millions of cycles. The high quality of the substrates allows producing precise spot diameters. The diameters are by large independent of the quality of the motion control system. The pins are also capable of printing in any orientation, including horizontal and upside down, enabling custom manufacturing applications. Downward pin movement is controlled by gravity instead of springloading, which minimizes surface forces and allows printing on delicate surfaces (e.g. acrylamide gel layers and silicon wafers). There are many reports in the literature about applications of micro-spotting [27, 28, 30]. One interesting application of micro-spotting allowing the formation of bacteria colonies onto an agar slate was reported by Al-Khaldi et al. [31]. Their microarrays consisted of 40 micro-spots in ﬁve replicates of eight bacteria by using a micro-spotting robotic system. Likewise, they could accurately identify the spotted microorganism by infrared spectroscopy in a timespan of 3 h. 5.4.2 Ink-Jet Printing

Ink-jet printing is a technology best known by its use in low-cost desktop printers [32, 33]. As schematically described in Figure 5.5, it is a drop-on-demand delivery approach, in which a computer-controlled ejection of single microscopic droplet is delivered into computer-deﬁned arrangements (patterns). Typically, piezoelectric dispensers (Figure 5.6A) or syringe-solenoid inject-type dispensers (Figure 5.6B) are used. Both dispensers allow programmable volume delivery, which is an important advantage over micro-spotting technology. Loading volumes of both types of dispenser are much higher than micro-spotting pins and are in the range of 5–10 ml. Print volumes with piezoelectric dispensers can go as low as 0.05 nl which is much better than micro-spotting pins, while in the case of syringe-solenoids it is signiﬁcantly higher (4–100 nl). Droplets ejected by the most sophisticated print

5.4 Robotic Printing

Figure 5.5 Delivery mechanism of ink-jet printing.

Figure 5.6 Dispensers for ink-jet printing: (A) piezoelectric dispenser, (B) syringe-solenoid inject-type dispenser, (C) thermoelectric dispenser.

j73

j 5 Patterning Techniques for Array Platforms

74

heads will leave approximately 20–30 ml. Spot densities of piezoelectric dispensers are comparable with the micro-spotting pins, in contrast to the syringe-solenoid dispensers that can offer only a low density. Ink-jet printing can deposit multiple layers of materials, which is an advantage. However, dispensers are much more complex and expensive (especially piezoelectric ones) than micro-spotting pins. There are many reports in the literature on the application of ink-jet printing [34–36] Recently, Allain et al. [37] developed an assay for Bacillus anthracis detection, based on DNA hybridization, using several dyes (eosin Y, ﬂuorescein and 4-methylumbelliferone).

5.5 Lithography with AFM

AFM is one of the most widely used nanoscale imaging techniques [38–43]. As shown in Figure 5.7, AFM utilizes a molecular or atomically ﬁne tip attached to the bottom of a ﬂexible/reﬂective cantilever. As the tip scans the surface of the sample, the laser beam is deﬂected off the cantilever; therefore its position and the extent of deﬂection of the cantilever can be monitored. Both the lateral position of the cantilever and the distance of the tip to the sample are controlled by piezoelectric crystal tubes. It is possible to change the direction and scanning rate of the tip on the surface. One may also apply a speciﬁc voltage to the substrate surface using cAFM tips. We now discuss two options to use AFM as a lithographic system for surface patterning.

Figure 5.7 Basic concept of AFM.

5.5 Lithography with AFM

Figure 5.8 Schematic description of DPN.

5.5.1 Dip-Pen Lithography with AFM

Similar to classical lithographic methods describe above, AFM can be used to deliver molecules loaded on the tip in the desired patterns. Mirkin et al. in 1999 was the ﬁrst to use AFM in lithography. They named it DPN [44]. The concept of DPN is the printing of molecules with nanoscale writing. As shown in Figure 5.8, the AFM tip carrying the probe is moved in pre-programmed deﬁned patterns. The tip will be in contact with the surface, either allowing the tip to dwell at a certain location or simply by rastering the tip close to the surface at a particular speed. As such, different shapes and sizes (i.e. dots, lines) can be created (patterned) allowing molecules to diffuse onto the solid surface by means of capillary forces. Resolution of dip-pen patterning with AFM is down to line widths of about 15 nm, depending on the type of substrate, contact time (between the tip and substrate), scan speed, relative humidity and relative solubility of the molecules in the water meniscus [45]. Mirkin et al. described the effects of several factors on DPN lithography. Two different types of ODNs (thiol-functionalized and acrylamide-functionalized), were patterned onto gold and SiO2 substrates, respectively. The size of the spots patterned increased signiﬁcantly with both contact time and relative humidity. Much larger spots were observed in the case of acrylamide-functionalized ODNs on SiO2 surfaces [46]. DPN is a suitable method for printing molecules with a variety of functional groups. Many materials can now be patterned using DPN-printed substrates [47–51]. Modiﬁcation of gold-coated substrates [e.g. surface plasmon resonance (SPR) slides] by DPN technology are a good example. N-Alkane thiol molecules (or their derivatives) have been widely used for surface patterning of gold-coated surfaces, in which a gold–sulfur (AuS) bond is formed. The alkene molecules are then directed to the

j75

j 5 Patterning Techniques for Array Platforms

76

surface of the patterned areas and make them hydrophobic. As a further modiﬁcation step, one may backﬁll the bare regions with different thiol-functionalized molecules (e.g. ODNs carrying thiol end-groups as probes in SPR biosensors) simply by exposing the patterned substrate to a solution of other thiol molecules. Additionally, different probes can be deposited at different spots and therefore arrays can be produced. Weinberger et al. were able to deposit 1-octadecanethiol onto a gold surface of a multilayer substrate using DPN. Then they performed a selective etching to form nanostructures as possible protein and DNA nanoarray platforms [52]. Lee et al. developed a protein array by initially patterning 16-mercaptohexadecanoic acid (MHA) on a gold thin-ﬁlm substrate in the form of dots or grids [53]. Proteins were absorbed on the preformed MHA patterns by immersing the substrate in a solution containing the desired protein. 5.5.2 cAFM Lithography

Scanning probe techniques, including cAFM, to form nanometer-scale patterns of organic molecules on silicon substrates have attracted much interest for their potential applications in chemical and biological sensors and molecular electronic device structures including DNA and protein arrays [54–56]. In cAFM, a voltage is applied to the AFM tip, while AFM is performing its normal scan. The tip acts as cathode, and the water meniscus formed between the tip and surface serves as electrolyte. As demonstrated by Hou et al. [57], the strong electric ﬁeld near the tip causes electrochemical reactions in the water column as water decomposes into hydroxyl ions (OH) and radicals (H). This results in breakdown, including ﬁeld-induced ionization, of the water molecules yielding electrons, protons and free radicals (OH) as follows: H2 O þ e ! OH þ H H2 O þ e ! OH þ H þ The OH molecules can be consumed in three ways: they may (i) formed hydrogen peroxide (H2O2) molecules, (ii) couple with H radicals to produce water or (iii) could be electrochemically reduced to hydroxyl ions at the tip surface area [58, 59]: OH þ OH ! H2 O2 OH þ H ! H2 O e þ OH ! OH The chemical environment that may be created during cAFM is schematically drawn in Figure 5.9.

5.5 Lithography with AFM

Figure 5.9 Schematic description of cAFM nanolithography.

The organic monolayer composition and topography can be altered by using cAFM. There are several factors affecting the physical (topology) and chemical changes on the nanopatterned areas with the use of cAFM nanolithography, such as applied voltage, tip velocity, humidity, tip conductivity and tip diameter [58, 59]. cAFM nanolithography is widely used for immobilization of nanostructures including biomolecules to template-guided patterned areas [4, 60]. The technique allows the control of the position and orientation of the patterns that may become powerful tools for biological applications such as protein and DNA arrays. A typical example of the use of cAFM is given by Gu et al., who prepared nanometric protein arrays on robust monolayers of a-hepta(ethylene glycol)methyl w-undecenyl ether associated with conductive silicon (111) [4]. After functionalizing the surface by cAFM lithography, avidin molecules were immobilized to these patterns. Note that avidin on these arrays serves as a template for the attaching proteins labeled with biotin. Recently, nanopatterns were created on alkene silane-functionalized silicon substrates by cAFM lithography (Figure 5.10A). Titanium wafers were ﬁrst hydroxylated through a treatment with 20% HNO3 at 80 C for 30 min. They were then incubated in a trichlorohexeneylsilane (TCHS)/toluene solution (2% v/v) to create double bonds on the substrate. Nanopatterns as lines were created on these surfaces by cAFM lithography. Thus, alkene groups of the TCHS molecules were converted into carboxyl and/or hydroxyl groups on the patterned lines. As seen in Figure 5.10(B), the height of patterns decreased when the tip velocity decreased [61]. 5.5.3 Nanoshaving and Nanografting

TherecentlydevelopedAFM-basednanolithographynanoshavingandnanografting, allowed the fabrication of nanoscale surface structures of alkane thiols, proteins

j77

j 5 Patterning Techniques for Array Platforms

78

Figure 5.10 Creating nanopatterns on alkene-silane-functionalized titanium substrates by cAFM lithography: (A) surface functionalization, (B) an example of nanopatterned lines.

and DNA [54, 62, 63]. In nanoshaving, in a ﬁrst step, self-assembled monolayers (SAMs) are formed. Usually, long-chain alkane thiols having w-terminal functionality (or not) are used on substrate surfaces (usually coated with a gold layer) (Figure 5.11). Then, the desired nanopatterns are formed by nanoshaving by

5.5 Lithography with AFM

Figure 5.11 Schematic description of nanoshaving.

making trenches on the SAM using the AFM tip. Different probes (bioligands) are immobilized on to different trenches and, likewise, nanoarrays are formed. Nanostructures (nanoparticles, nanowires, nanotubes, etc.) may be ﬁrst functionalized for immobilization of probes on their surfaces. These functionalized nanostructures may then be grafted onto pre-patterned surfaces (e.g. nanoshaved). This is referred as nanografting (Figure 5.12). Note that by using this approach the substrate surface area can be extended and therefore resolution (sensitivity) can be increased signiﬁcantly. Zhao et al. recently used nanoshaving for making trenches on the alkylthiol SAM by using an AFM tip [54]. In the ﬁrst step, they deposited mouse IgG on the shaved trenches. In the second step, they shaved another array of trenches and deposited human IgG on these. Finally, Alexa Fluor 546-labeled anti-mouse IgG nanotubes were allowed to interact on the mouse IgG trenches and ﬂuorescein isothiocyanatelabeled anti-human IgG nanotubes onto the human IgG. Recently, we initiated a study in which nanopatterning of alkene-silane-modiﬁed silicon wafers by cAFM nanolithography were used as nanoarray platform. Silicon nanowires were synthesized via a gold-catalyzed chemical vapor deposition method [64] and functionalized using 3-aminopropylmethoxysilane molecules. Immobilization of these functionalized nanowires onto the patterned areas of the nanoarrays created novel nanobioarrays. These can have diverse applications, including detection of pathogenic microorganisms.

j79

j 5 Patterning Techniques for Array Platforms

80

Figure 5.12 Schematic description of nanografting.

5.6 Conclusions

Array technologies offer a number of advantages for the screening of large number of analytes including speed, convenience and high-throughput analysis. Array platforms consist of patterned surfaces and probe molecules immobilized onto these patterns. There are two main approaches for patterning: (1) surfaces are ﬁrst patterned and then different probe molecules are selectively immobilized onto different locations on the surface (a two step process) or (2) different probe molecules are delivered to different locations on the surface to form a patterned surface (two steps together). Several techniques (e.g. soft lithography, photolithography, robotic printing, etc.) have been developed to prepare array platforms, even at the commercial scale, which usually contain micron-size spots (locations) for different probes on the same platform. In this chapter, some approaches have been described and some interesting uses further exempliﬁed. Each method has advantages and limitations when compared to the others. The most suitable one should be selected mainly based on the target use and available facilities. Even very simple techniques may be suitable for the efﬁcient detection of target molecules. Nanopatterns can be created using some novel techniques including dip-pen lithography with AFM, nanoshaving and nanografting, and will certainly trigger the development of novel array platforms. Nanotechnology is a rather new ﬁeld attracting a lot of scientiﬁc and technological interest in the recent years. By nanopatterning one can potentially load millions of different probe molecules on a small surface area and perform millions of parallel

References

measurements. However, one should carefully select the correct size level for the use of an array platform (e.g. in cell arrays, an array platform with nanopatterns would be simply unrealistic or even a wrong approach). In addition, it should be noted that in order to detect any kind of interaction on the surface (with the immobilized probe and target molecules), which is usually realized by using labels (ﬂuorescence dyes, etc.), a minimum amount of emission is needed. Therefore, resolution in the measurements may limit the application of nanopatterned array platforms.

Acknowledgments

E.P. is supported by the Turkish Academy of Sciences as a Full Member. G.D. is also _ supported as a Post-doctoral Fellow by TÜBITAK.

References 1 Eing, A. and Vaupel, M. (2002) Imaging Ellipsometry in Biotechnology, Verlag Axel, Gierspeck, pp. 1–10. 2 Zhu, H. and Snyder, M. (2003) Protein chip technology. Curr. Opin. Chem. Biol., 7, 55. 3 Kim, J.H. (2002) Bioinformatics and genomic medicine. Genet. Med., 4, 62S. 4 Gu, J., Yam, C.M., Li, S. and Cai, C. (2004) Nanometric protein arrays on proteinresistant monolayers on silicon surfaces. J. Am. Chem. Soc., 126, 8098. 5 Kramer, S., Fuierer, R.R. and Gorman, C.B. (2003) Scanning probe lithography using self-assembled monolayers. Chem. Rev., 103, 4367. 6 Zhao, X.M., Xia, Y.N. and Whitesides, G.M. (1996) Fabrication of threedimensional micro-structures: microtransfer molding. Adv. Mater., 8, 837. 7 Kim, E., Xia, Y.N. and Whitesides, G.M. (1996) Micromolding in capillaries: applications in materials science. J. Am. Chem. Soc., 118, 5722. 8 Brehmer, M., Conrad, L. and Funk, L. (2003) New developments in soft lithography. J. Dispers. Sci. Technol., 24, 291. 9 Aizenberg, J., Rogers, J.A., Paul, K.E. and Whitesides, G.M. (1998) Imaging proﬁles of light intensity in the near ﬁeld:

10 11

12

13

14

15

16

applications to phase-shift photolithography. Appl. Optics, 37, 2145. Xia, Y. and Whitesides, G.M. (1998) Soft lithography. Annu. Rev. Mater. Sci, 28, 153. Duffy, D.C., McDonald, J.C., Schueller, O.J.A. and Whitesides, G.M. (1998) Rapid prototyping of microﬂuidic systems in poly(dimethylsiloxane). Anal. Chem., 70, 4974. Schulz, H., Lyebyedyev, D., Scheer, H.C., Pfeiffer, K., Bleidiessel, G., Grutzner, G. and Ahopelto, J. (2000) Master replication into thermosetting polymers for nanoimprinting. J. Vac. Sci. Technol. B, 18, 3582. Graf, H.P., Bosch, M.A., Good, E., Haller, H.R. and Hubbard, W. (1985) Master fabrication for optical-data disks using reactive ion-beam etching. J. Appl. Phys., 58, 3255. Rozhok, S., Shen, C.K.F., Littler, P.L.H., Fan, Z.F., Liu, C., Mirkin, C.A. and Holz, R.C. (2005) Synthesis and characterization of NiO–Ta2O5 nanocomposite electrode for electrochromic devices. Small, 4, 445. Rowan, B., Wheeler, M.A. and Crooks, R.M. (2002) Patterning bacteria within hyperbranched polymer ﬁlm templates. Langmuir, 18, 9914. Morhard, F., Pipper, J., Dahint, R. and Grunze, M. (2000) Immobilization of

j81

j 5 Patterning Techniques for Array Platforms

82

17

18

19

20

21

22

23

24

25

antibodies in micropatterns for cell detection by optical diffraction. Sens. Act. B, 70, 232. Howell, S.W., Inerowicz, H.D., Regnier, F.E. and Reifenberger, R. (2003) Patterned protein microarrays for bacterial detection. Langmuir, 19, 436. Weibel, D.B., Lee, A., Mayer, M., Brady, S.F., Bruzewicz, D., Yang, J., DiLuzio, W.R., Clardy, J. and Whitesides, G.M. (2005) A bacterial printing press that regenerates its ink: contact printing bacteria using hydrogel stamps. Langmuir, 21, 6436. Barbulovic-Nad, I., Lucente, M., Sun, Y., Zhang, M., Wheeler, A.R. and Bussmann, M. (2006) Bio-microarray fabrication techniques – a review. Crit. Rev. Biotechnol., 26, 237. Pirrung, M.C. and Huang, C.Y. (1996) A general method for the spatially deﬁned immobilization of biomolecules on glass surfaces using caged biotin. Bioconjug. Chem., 7, 317. Pritchard, D.J., Morgan, H. and Cooper, J.M. (1995) Patterning and regeneration of surfaces with antibodies. Anal. Chem., 67, 3605. Britland, S., Clark, P., Connolly, P. and Moores, G. (1992) Micropatterned substratum adhesiveness: a model for morphogenetic cues controlling cell behavior. Exp. Cell. Res., 198, 124. Healy, K.E., Thomas, C.H., Rezania, A., Kim, J.E., McKeown, P.J., Lom, B. and Hockberger, P.E. (1996) Kinetics of bone cell organization and mineralization on materials with patterned surface chemistry. Biomaterials, 17, 195. McGall, G.H., Barone, A.D., Diggelmann, M., Fodor, S.P.A., Gentalen, E. and Ngo, N. (1997) The efﬁciency of light directed synthesis of DNA arrays on glass substrates. J. Am. Chem. Soc., 119, 5081. Southern, E.M., Case-Green, S.C., Elder, J.K., Johnson, M., Mir, K.U., Wang, L. and Williams, J.C. (1994) Arrays of complementary oligonucleotides for analysing the hybridization behaviors

26

27

28

29 30

31

32

33

34

35

36

of nucleic acids. Nucleic Acids Res., 22, 1368. Fodor, S.P.A. (1997) DNA sequencing: massively parallel genomics. Science, 277, 393. Sheehan, P.E., Edelstein, R.L., Tamanaha, C.R. and Whitman, L.J. (2003) A simple pen-spotting method for arraying biomolecules on solid substrates. Biosens. Bioelectron., 18, 1455. Balazs, I., Beekman, J., Neuweiler, J., Liu, H., Watson, E. and Ray, B. (2001) Molecular typing of HLA-A, -B, and -DRB using a high throughput micro array format. Human Immun., 62, 850. Schena, M. (2003) Microarray Analysis, Wiley-Liss, Hoboken, NJ, p. 159. Diehl, F., Grahlmann, S., Beier, M. and Hohesiel, J.D. (2001) Manufacturing DNA microarrays of high spot homogeneity and reduced background signal. Nucleic Acid Res., 29, e38. Al-Khaldi, S.F., Mossoba, M.M., Ismail, A.A. and Fry, F.S. (2004) Accelerating bacterial identiﬁcation by infrared spectroscopy by employing microarray deposition of microorganisms. Foodborne Path. Dis., 1, 172. Heinzl, J. and Hertz, C.H. (1985) Ink-jet printing. Adv. Electron Electron Phys., 65, 91. Hebner, T.R., Wu, C.C., Marcy, D., Lu, M.H. and Sturm, J.C. (1998) Ink-jet printing of doped polymers for organic light emitting devices. App. Phys. Let., 72, 519. Barron, J.A., Young, H.D., Dlott, D.D., Darﬂer, M.M., Krizman, D.B. and Ringeisen, B.R. (2005) Printing of protein microarrays via a capillary-free ﬂuid jetting mechanism. Proteomics, 5, 4138. Xu, B.J., Lee, Y.K., Jin, Q.H., Zhao, J.L. and Ho, C.M. (2006) Multilayer SU-8 based microdispenser for microarray assay. Sens. Act. B., 132, 714. Nallani, A.K., Chen, T., Hayes, D.J., Che, W.S. and Lee, J.B. (2006) A method for improved VCSEL packaging using MEMS

References

37

38

39

40

41

42

43

44

45

46

and ink-jet technologies. J. Lightwave Tech., 24, 1504. Allain, L.R., Stratis-Cullum, D.N. and Vo-Dinh, T. (2004) Investigation of microfabrication of biological sample arrays using piezoelectric and bubble-jet printing technologies. Anal Chim. Acta, 518, 77. Binnig, G., Quate, C.F. and Gerber, C.H. (1986) Atomic force microscope. Phys. Rev. Lett., 56, 9, 930. Hoh, J.H. and Hansma, P.K. (1992) Atomic force microscopy for highresolution imaging in cell biology. Trends Cell Biol., 2, 208. Raab, A., Han, W., Badt, D., Smith-Gill, S.J., Lindsay, S.M., Schindler, H. and Hinterdorfer, P. (1999) Antibody recognition imaging by force microscopy. Nature Biotechnology, 17, 901. Kim, J.M., Ohtani, T., Sugiyama, S., Hirose, T. and Muramatsu, H. (2001) Simultaneous topographic and ﬂuorescence imaging of single DNA molecules for DNA analysis with a scanning near-ﬁeld optical/atomic force microscope. Anal. Chem., 73, 5984. Liu, M. and Liu, G. (2005) Hybridization with nanostructures of single-stranded DNA. Langmuir, 21, 1972. Van der Heijden, T., Moreno-Herrero, F., Kanaar, R., Wyman, C. and Dekker, C. (2006) Comment on direct and real-time visualization of the disassembly of a single RecA–DNA–ATPgS complex using AFM imaging in ﬂuid. Nano Lett., 6, 3000. Piner, R.D., Zhu, J., Xu, F., Hong, S.H. and Mirkin, C.A. (1999) Dip-pen nanolithography. Science, 283, 661. Smith, R.K., Lewis, P.A. and Weiss, P.S. (2004) Patterned self-assembled monolayers. Prog. Surface Sci., 75, 1. Demers, L.M., Ginger, D.S., Park, S.J., Li, Z., Chung, S.W. and Mirkin, C.A. (2002) Direct patterning of modiﬁed oligonucleotides on metals and insulators by dip-pen nanolithography. Science, 296, 1836.

47 Maynor, B.W., Li, Y. and Liu, J. (2001) Au ink for AFM dip-pen nanolithography. Langmuir, 17, 2575. 48 McKendry, R., Huck, W.T.S., Weeks, B., Fiorini, M., Abel, C. and Rayment, T. (2002) Creating nanoscale patterns of dendrimers on silicon surfaces with dip-pen nanolithography. Nano Lett., 2, 713. 49 Noy, A., Miller, A.E., Klare, J.E., Weeks, B.L., Woods, B.W. and DeYoreo, J.J. (2002) Fabrication of luminescent nanostructures and polymer nanowires using dip-pen nanolithography. Nano Lett., 2, 109. 50 Agarwal, G., Naik, R.R. and Stone, M.O. (2003) Immobilization of histidine-tagged proteins on nickel by electrochemical dip pen nanolithography. J. Am. Chem. Soc., 125, 7408. 51 Hampton, J.R., Dameron, A.A. and Weiss, P.S. (2006) Double-ink dip-pen nanolithography studies elucidate molecular transport. J. Am. Chem. Soc., 128, 1648. 52 Weinberger, D.A., Hong, S., Mirkin, C.A., Wessels, B.W. and Higgins, T.B. (2000) Combinatorial generation and analysis of nanometer- and micrometer-scale silicon features via dip-pen nanolithography and wet chemical etching. Adv. Mater., 12, 1600. 53 Lee, K.B., Park, S.J., Mirkin, C.A., Smith, J.C. and Mrksich, M. (2002) Protein nanoarrays generated by dip-pen nanolithography. Science, 295, 1702. 54 Zhao, Z., Banerjee, I.A. and Matsui, H. (2005) Simultaneous targeted immobilization of anti-human IgG-coated nanotubes and anti-mouse IgG-coated nanotubes on the complementary antigen patterned surfaces via biological molecular recognition. J. Am. Chem. Soc., 127, 8930. 55 Kinser, C.R., Schmitz, M.J. and Hersam, M.C. (2005) Conductive atomic force microscope nanopatterning of hydrogenpassivated silicon in inert organic solvents. Nano Lett., 5, 91. 56 Maoz, R., Frydman, E., Cohen, S.R. and Sagiv, J. (2000) Constructive nanolithography: inert monolayers as patternable templates for in-situ

j83

j 5 Patterning Techniques for Array Platforms

84

57

58

59

60

61

nanofabrication of metal–semiconductor– organic surface structures – a generic approach. Adv. Mater., 12, 725. Hou, S., Li, Z., Li, Q. and Liu, Z.F. (2004) Poly(methyl methacrylate) nanobrushes on silicon based on localized surface-initiated polymerization. Appl. Surf. Sci., 222, 338. Park, J. and Lee, H. (2004) Effect of surface functional groups on nanostructure fabrication using AFM lithography. Mater. Sci. Eng. C, 24, 311. Lyuksyutov, S.F., Paramonov, P.B., Dolog, I. and Ralich, R.M. (2003) Peculiarities of an anomalous electronic current during atomic force microscopy assisted nanolithography on n-type silicon. Nanotechnology, 14, 716. Zheng, J., Chen, Z. and Liu, Z. (2000) Atomic force microscopy-based nanolithography on silicon using colloidal au nanoparticles as a nanooxidation mask. Langmuir, 16, 9673. Garipcan, B., Winters, J., Atchison, J.S., Cathell, M.D., Schiffman, J.D., Leaffer,

O.D., Nonnenmann, S.S., Schauer, C.L., Pi¸skin, E., Nabet, B., Spanier, J.E. (2008) Conductive-AFM Nanolithography on Alkene-silane Modiﬁed TiO2 Substrates. Langmuir, 24, 8944. 62 Wang, X., Zhou, D., Rayment, T. and Abel, C. (2003) Systematic manipulation of surface chemical reaction on the nanoscale: a novel approach for constructing three-dimensional nanostructures. Chem. Commun., 4, 474. 63 Zhou, D., Sinniah, K., Abel, C. and Rayment, T. (2003) Label-free detection of DNA hybridization at the nanoscale: a highly sensitive and selective approach using atomic-force microscopy. Angew. Chem, 115, 5084. 64 Cao, L., Garipcan, B., Atchison, J.S., Ni, C., Nabet, B. and Spanier, J.E. (2006) Instability and transport of metal catalyst in the growth of tapered silicon nanowires. Nano Lett., 6 (9), 1852.

j85

6 Probe Immobilization Techniques in Array Technologies Erhan Pis¸kin, Bora Garipcan, and Memed Duman

6.1 Introduction

In array technologies one of the most important steps is the immobilization of the probe molecules or so-called bioligands [e.g. oligodeoxynucleotides (ODNs), oligopeptides/proteins, etc.) effectively on the array platforms, in an active conﬁguration. Selection of the substrate material is the ﬁrst step, and this mainly depends on the detection technique that will be used for interaction between the probe molecules immobilized on the substrate and target molecules in solution. Glass slides, silicon wafers, gold-coated slides and gold particles, quantum dots (QDs), polymer-coated slides, and polymeric ﬁlms and particles have all been proposed as probe support materials, and some of them are already being marketed for use in commercial array systems [1–6]. Slides made of the different materials mentioned above, are used routinely as substrates in surface-based array technologies (Figure 6.1A) [7–9]. Patterns, usually as spots or lines, but also with different two- or even three-dimensional structural forms are created on the substrate surface using several techniques. Then, different probe molecules are positioned (immobilized) into different areas (spots, lines, etc.) to generate the ﬁnished outline of the array platform. Alternatively, immobilization and patterning of the probe can be applied at the same time (patterning and delivery of the probe molecules to patterned areas are described in Chapter 5). Suspension-based array technologies have also been developed, in which metal micro (gold, silver, etc.) or inorganic nanoparticles (magnetite, QDs, etc.), or polymers and polymer-coated particles (usually in the form of spherical particles but also irregular shapes, rod-like structures, etc.) are used as substrates (Figure 6.1B). These suspension arrays are usually mixtures of different particles and probe molecules suspended in an aqueous medium [10, 11]. Particles (particularly nanoparticles) have also been used in surface-based array technologies. They are adsorbed onto surfaces, usually as monolayers to increase the surface area (also called as nanografting). This in turn increases the surface density of the probe molecules, thus increasing the sensitivity or resolution of downstream

j 6 Probe Immobilization Techniques in Array Technologies

86

Figure 6.1 Array platforms: (A) surface-based, (B) suspension-based.

processes [1, 12, 13]. In some cases probe molecules can be directly immobilized onto substrate surfaces (e.g. if it is a gold surface) or several modiﬁcations can be applied to native substrate surfaces to create functional groups, which are then used for probe immobilization (brieﬂy discussed below). The ﬁnal, but very important step, is immobilization of the probe molecules onto substrate surfaces; here there are different points that should be strictly and carefully considered. Physical attachment of the probe molecules onto surfaces is possible and has been utilized. However, covalent bond formation between the probe molecules and surface functional groups is usually much more successful, and results in stable probe immobilization with regeneration capabilities. However, in the later case, the surface that is not occupied with the probe molecules may cause some undesirable/ non-speciﬁc adsorption, which may reduce the speciﬁcity of the array system. This should be eliminated or signiﬁcantly reduced. Orientation is critical for effective interaction of the probe molecules immobilized on the substrate surface with the complementary (target) molecules. Probes are usually biomolecules, such as ODNs or oligopeptides/proteins, and therefore they may loose their activity if their threedimensional conformations (especially in the case of proteins) are disrupted. This can occur as a result of immobilization procedures. If the probe is not presented in its active form on the surface there will be no good signal generation or response, therefore this should be prevented. Correct and incorrect probe orientations are outlined in Figure 6.2.

6.2 Support Material 6.2.1 Glass

Glass is one of the most preferred solid supports, due primarily to its low cost and low intrinsic ﬂuorescence, its transparency, resistance to high temperature, and relatively homogeneous chemical surface. In addition, glass offers a number of practical

6.2 Support Material

Figure 6.2 Orientation of the probe molecules (ODNs and antibodies) on substrate surfaces: (A) correct and (B) incorrect.

advantages over porous membranes and gel pads. Liquids can not penetrate the surface of the support and target nucleic acids have direct access to the probe without internal diffusion. Microscope slides are commonly used in laboratories because they are easy to handle and adaptable to automatic readers. Table 6.1 exempliﬁes some commercially available glass slides and their manufacturers. Note that some Table 6.1 Commercially available glass slides and manufacturers.

Manufacturer

Tradename

Modification

Sigma, St Louis, MO, USA

AminoPrep

CEL Associates, Houston, TX, USA Corning Life Sciences, Corning, NY, USA

CMT-GAPS

aminopropylsilane coated silanated, amine aminopropylsilane coated untreated untreated untreated untreated and isothiocyanate modiﬁed mercaptosilane derivatized polylysine coated aminoalkylsilane derivatized nylon coated amino modiﬁed aldehyde modiﬁed

Corning Life Sciences, Corning, NY, USA Erie Scientiﬁc, Portsmouth, NH, USA Knittel Glaser, Braunschweig, Germany Menzel Glaser, Braunschweig, Germany Orchid Bioscience, Princeton, NJ, USA Sigma, St Louis, MO, USA Sigma-Aldrich, Steinheim, Germany Schleicher & Schuell, Dassel, Germany SurModics, Eden Prairie, MN, USA TeleChem International, Sunnyvale, CA, USA

PolyPrep SilanePrep

j87

j 6 Probe Immobilization Techniques in Array Technologies

88

manufacturers market untreated glass slides while others have treated ones which carry different functional groups for direct immobilization of the probe molecules. Glass surfaces can be functionalized using several methods. The ﬁrst step is usually hydroxylation of surfaces (carrying hydroxyl groups), created simply by incubating the glass slides with several basic or acidic solutions including Piranha solution (the most widely used one); 25% NH4OH; KOH solution pH 11; NaOH solution (10%); H2O2 (30%), 1 N HNO3; HCl (saturated):H2O (3:1); and H2SO4 (70%). In the second step the glass surface is modiﬁed, usually with various types of silane compounds via the surface hydroxyl groups. Several silane compounds are used for this silane chemistry (Table 6.2). Most of them carry di- or trimethoxy (or sometimes ethoxy) groups, which react with surface hydroxyl groups, leading to covalent attachment of the silane compound onto the surface. Note that these compounds carry different functional groups, such as amino, epoxy, carboxylic acid, isocyanate, sulfhydryl and aldehyde, which are then used for the immobilization of probe molecules via similar functional groups. Electrophilic glass coatings such as epoxysilane or phenylisothiocyanate can be prepared by direct silanation [14]. These are very reactive and immobilization of the probe molecules occurs very rapidly. However, these strongly electrophilic groups are chemically unstable, have a poor shelf-life and it is difﬁcult to precisely control the reactions. Surfaces treated with milder electrophiles such as aliphatic aldehydes or carboxylic acids are more stable. However, these probe immobilization reactions are slow, and require several activating and stabilization agents. For instance, probe molecules carrying amino groups can be covalently bound to aldehyde-modiﬁed supports, which require post-arraying reduction of the unstable Schiff base linkages with sodium borohydride. Use of carboxylic acid supports requires the use of carbodiimide coupling agents. More commonly, amine-modiﬁed slides are activated with homo- or heterobifunctional coupling agents to provide an electrophilic surface just prior to arraying [15, 16]. Thiol- or disulﬁde-modiﬁed probes can be grafted onto aminosilane via a hetero-bifunctional cross-linker or on 3-mercaptopropylsilane. (Note that use of coupling agents is difﬁcult for routine use by researchers, as the conditions are difﬁcult to reproduce and reagents are hazardous.) Table 6.2 Silane compounds used for silanization.

Silane compound

Functional group created

3-Aminopropyltrimethoxysilane p-Aminophenyltrimethoxysilane Diethoxymethylaminopropylsilane Triethoxyaminopropylsilane Glycidyloxopropyltrimethoxysilane 3-Mercaptopropyltrimethoxysilane 3-Isocyanatepropyltrimethoxysilane

amino amino amino amino epoxy sulfhdryl isocyanate

6.2 Support Material

Figure 6.3 Surface functionalization by ozonization: (A) ozonide formation, (B) converting surface ozonide groups into different functional groups.

Glass slide surfaces can be also functionalized by ozonization, using an ozone generator and also conductive atomic force microscopy (see Chapter 5). The chemistry of ozonization reactions are depicted schematically in Figure 6.3. Hydroxyl groups on the modiﬁed glass slides are silanized, then converted into ozonide by ozonization. These unstable groups are then linked to different functional groups.

j89

j 6 Probe Immobilization Techniques in Array Technologies

90

6.2.2 Silicon

The use of silicon wafers as a solid support for array technologies has several additional advantages compared with the commonly used glass slides. Unoxidized crystalline silicon offers advantages as a substrate for immobilization because of its high purity, highly organized and deﬁned crystalline structure, robustness, and its ubiquitous use in microelectronics industry [4]. Native silicon surfaces react with air under ambient conditions to form a thin surface layer of silicon oxide. This oxidized silicon surface is chemically similar to glass and suffers from some of the same drawbacks (i.e. non-homogeneity and variability in the relative number of SiOSi and SiOH linkages). This non-homogeneity can lead to difﬁculties in the reproducibility and homogeneity of the subsequent ligand immobilized surfaces [18, 19]. Recently, chemical pathways for direct functionalization of silicon substrates without an oxide layer has opened up new possibilities for highly controlled probe immobilization which involves direct carbon–silicon bonds and have resulted in methyl, chlorine, ester or acid terminated substrates [4]. Silicon wafers also have less surface roughness, which increases the uniformity of probe deposition, at a higher density and generates smaller array spot size. The ﬂat surface allows high-density arrays to be analyzed with confocal laser-based scanners. Silicon surfaces provide a better signal-to-noise ratio as they show less background ﬂuorescence and the dark, non-transparent surface absorbs excitation light. Finally, silicon wafer technology is readily available which facilitates the fabrication of chip-based devices [17]. 6.2.3 Gold

Gold-coated glass slides are the basic elements of novel surface plasmon resonance optical biosensors and arrays [20, 21]. In the mid-1980s, the spontaneous adsorption of organosulfur compounds on gold was a widespread method for the preparation of self-assembling monolayers (SAMs) [32–35]. If an alkyl chain is long enough, these architectures show a molecular orientation nearly perpendicular to the surface and are quite stable under controlled conditions [33, 36–38]. The alkyl chains can be terminated by reactive head groups, such as carboxylic acid and amino groups, which are used for formation of functionalized SAMs. These alkyl chains may be covalently bond to gold surfaces to act as spacer arms for further immobilization steps [41, 42]. These sublayers are capable of supporting the immobilization of other biomolecules via covalent chemical coupling, electrostatic physiosorption or supramolecular interactions [43]. It should also be noted that it is possible to create SAMs on gold surfaces with desired patterns for the preparation of array platforms using dip-pen lithography and even atomic force microscopy (see Chapter 5) [44, 45]. Gold nanoparticles (spherical, regular in shape and also rod-like) have also been utilized as support matrices for suspension array platforms due to their unexpected optical properties at nanosize dimensions [10, 22–26]. They increase the surface area required for high resolution micro and nanografting array technologies [27, 28].

6.2 Support Material

Thiol end-groups are sufﬁcient to covalently bond molecules to these gold surfaces [29, 30]. It is quite straightforward, easy and quick reaction, and requires no other chemicals. It is possible to couple molecules carrying thiol end-group on gold surfaces simply by treating the gold surface in 1 M KH2PO4 solutions at pH 3.8, even at room temperature for about 120 min [31]. However, one should consider that thiolbonded monolayers can be destabilized by chemicals and other factors in the environment [39, 40]. 6.2.4 Polymers

Polymeric membranes and ﬁlms (such as nylon, cellulose, etc.), polymer-coated surfaces (polycations, e.g. polyethyleneimine and polylysine, hydrogels such as carboxylmethylated dextran, polyamidoamine (PAMAM) starburst dendrimers, polymeric particles (polystyrene, polyacrylates, dextran-based, etc.) have all been utilized/proposed as substrates in the preparation of both surface-based and suspension-based array platforms [46–51]. Polymeric substrates can be produced in almost any desired shape from a wide variety of polymers carrying almost all type of functional groups, which may then be used for immobilization of probe molecules, using similar chemistries to those that are applied to other substrate materials, as mention above. Porous polymeric materials (membranes, particles or as coatings) may be used in order to increase the surface area, which in turn may result in a higher density of the probe molecules or unit mass of the substrate. However, since the probe molecules are also immobilized within macropores, they are not readily available (i.e. the time taken for the target molecules to reach these probes may be longer). In other terms the pore diffusion resistance is quite high, which in turn results in a very long response time of the detection system, which is not desirable. Some of the polymeric substrates do absorb water and swell (such as hydrogels); however, this may cause changes in the structural properties and create some instabilities as a result. Some of the polymer surfaces (mostly hydrophobic ones) also exhibit some non-speciﬁc interactions (especially protein targets). This is also an uncontrollable parameter and can lead to false-positive readings in some detection units. Most of the polymeric materials (especially the porous ones) are not transparent, which is a limitation to their use with some detection systems in which transparency is necessary.

6.2.5 QDs

QDs, which are also referred to as artiﬁcial atoms, are monosize and nanocrystalline nanoparticles, and are of a size in the range of typically 2–10 nm (10–50 atoms) [52–54]. QDs can be synthesized from various types of semiconductor materials (such as CdS, CdSe, CdTeI, InP, InAsI and PbSeI), therefore they also act as semiconductors.

j91

j 6 Probe Immobilization Techniques in Array Technologies

92

Figure 6.4 QDs emitting light in a variety of colors. Colloidal CdSe QDs dispersed in hexane. (From http://web.mit.edu/felicef/ with permission).

They exhibit unique light absorption and emission properties [55, 56]. Individual QDs are too small to see with the naked eye; however, they signal their presence by emitting light in a variety of colors depending on both their size and composition, as depicted in Figure 6.4. For instance CdSe QDs (about 2 nm diameter) are ﬂuorescent in the green spectrum, while larger ones (about 5 nm in diameter) are ﬂuorescent in the red. By altering the QD size and chemical composition, the ﬂuorescence emission may be tuned from the near-UV, throughout the visible, and into the near-infrared (IR), spanning a broad wavelength range of 400–2000 nm [57]. For example, CdS and ZnSe QDs emit blue to near-UV light, different sized CdSe QDs emit light across the visible spectrum, and InP and InAs QDs emit in the far-red and near-IR [58]. These novel/ unique properties have positioned them as very strong candidates for future use in revolutionizing ﬂuorescence-based detection protocols [56, 59]. QDs have several advantages compared with traditional organic dyes (e.g. rhodamine 6G, ﬂuorescein) [3, 60, 61]: .

. .

.

.

.

QDs exhibit symmetric and narrow emission bandwidths (full width at half maximum of around 30–45 nm) that span the visible spectrum, which allows for simultaneous excitation of several particle sizes at a single wavelength. They have a broader absorption proﬁle. QDs reduce cross-talk between crystals of different colors because each color is emitted as a narrow and symmetrical, ﬂuorescence peak, and the peak excitation wavelength does not overlap with emission wavelength. All samples are induced to emit their respective colors even though a single source was used to excite them, which facilitates simultaneous detection, imaging and quantiﬁcation. They are very stable. As mentioned above, they are composed of inert inorganic compounds and are further stabilized with the shell layer that increases resistance to photochemical damage. QDs ﬂuorescence intensely and exhibit high quantum yields, therefore their brightness is comparable to or greater than traditional organic dyes.

6.2 Support Material .

.

. .

CdSe/ZnS QDs have ﬂuorescence lifetimes of 15–20 ns – an order of magnitude greater than conventional organic dyes and even greater than the auto-ﬂuorescence lifetime of organic dyes. CdSe–ZnS core–shell QDs are luminescent inorganic ﬂuorophores that have the potential to overcome some of the functional limitations encountered by organic dyes in ﬂuorescence labeling applications. Luminescence emission from QDs is detected at concentrations comparable to organic dyes by using conventional ﬂuorescence methods. Individual QDs and QD bioconjugates are easily observable by confocal microscopy.

The size and shape of QDs can be precisely controlled by the duration, temperature, and ligand molecules used in the synthesis [62]. The ﬁrst step in synthesizing QDs is the preparation of a core, which determines the color and is composed of nanocrystals. CdS, CdSe and CdTe are typical core nanocrystals having, respectively, UV-blue, visible, far-red and near-IR emissions. In earlier attempts at QDs synthesis, wide size distribution, structural defects and low ﬂuorescence quantum yields were observed. The ﬁrst successful QDs, synthesized by Murray et al., had quite a narrow size distribution, excellent crystallinity and almost no defects [63]. Here, appropriate metallic or organometallic precursors (zinc, cadmium or mercury) with a corresponding chalcogen precursor (sulfur, selenium or tellurium) were used for synthesis in coordinating organic solvents at high temperatures. Tri-n-octylphosphine oxide (TOPO) in conjunction with other surfactants or co-solvents such as tri-n-octylphosphine, hexadecylamine or stearic acid were commonly used due to their high boiling points, and ability to coordinate both metal and chalcogen elements. In order to increase the intensity of the ﬂuorescence and reduce photodegradation of QDs, a shell layer is formed around the core nanocrystals, which is composed of semiconductor material of wider band gap than the core material (typically ZnS). The core–shell QDs obtained are highly ﬂuorescent, photostable and sufﬁciently monodisperse. However, these QDs are not soluble in aqueous solution and do not have functional groups for conjugation to biomolecules for further steps. There are several techniques to overcome these limitations, where commonly the surface TOPO molecules are replaced or surfaces are modiﬁed with linker molecules carrying functional groups (carboxylic acid, amine, thiol, etc.). These are then used for conjugation of biomolecules (e.g. DNA ODN or aptamer, oligopeptide, antibody, etc.) using well-established protocols [57, 64–67]. Interaction of target molecules with counter bioligand (probe), immobilized to the surface of QDs, results in a change in their ﬂuorescent signal, which is then used for detection of the target molecules. The most widespread use of QDs in nanobiotechnology is as novel label alternatives to organic dyes in a variety of applications for detection of biological interactions with a target, including diagnostic kits, biosensors, in vivo targeting and imaging [58, 68–73]. Gerion et al. used functionalized QDs for the detection of pathogens and also for targeting peptide-conjugated QDs to the nucleus of living cells [74]. They also utilized QDs conjugated to single-stranded ODNs to detect single-base polymorphisms of genomic DNA in cDNA microarray platforms.

j93

j 6 Probe Immobilization Techniques in Array Technologies

94

Sun et al. were able to develop a sandwich immunoassay using CdSe/ZnS QDs conjugated with bioligand antibodies using a laser confocal scanner [75]. Goldman et al. utilized avidin-carrying CdSe/ZnS QDs conjugated to biotinylated antibodies for the detection of protein toxins in ﬂuoro-immunoassays [76]. They were also able to apply a sandwich immunoassay approach for simultaneous detection of four protein toxins (i.e. cholera toxin, ricin, shiga-like toxin 1 and staphylococcal enterotoxin B) using QDs with four different emission maxima (510, 555, 590 and 610 nm) carrying speciﬁc antitoxin antibodies [77]. Dwarakanath et al. also developed CdSe/ZnS QDs carrying antibodies or DNA aptamers for detection of bacteria, where detection was achieved by observing shifts in the ﬂuorescence emission when bacteria were bound [78]. Ho et al. exhibited multiplex detection of three target sequences derived from B. anthracis genes using multicolor ODN-functionalized QDs as nanoprobes with high sensitivity and speciﬁcity [61]. These QD nanoprobes were prepared by surface-functionalizing QDs with target-speciﬁc ODN probes. In this study, two target-speciﬁc QD nanoprobes with different emission wavelengths were used to form a sandwich with the target, creating a QD probe–target nanoassembly. The nanoassembly was then distinguished from unbound QDs as a blended color due to co-localization of both the QD nanoprobes. Nie et al. embedded multicolor CdSe QDs with different intensities within polystyrene beads as carriers. These were utilized for conjugation with several bioligands for detection of target (complementary) molecules within a medium [73]. It was possible to read all the QD-encoded beads using a single light source for successful detection of the target molecules. 6.3 Immobilization

An effective immobilization technique should exhibit the following properties: .

.

.

.

. .

Immobilization of the probe molecules in the correct orientation and in an active form (Figure 6.2), allowing efﬁcient interaction reactions between the probe and the target. Orientation of the probe molecules so as to not cause deformation of the threedimensional structures (especially in the case of protein-based probes) during the immobilization procedure. A relatively high (or better optimized) density of probe molecules on the surface of the array substrate. Note that low-density surface coverage will yield a correspondingly low detection signal, while a high density may cause steric interference (i.e. reduce probe–target interaction). Minimal, or preferably no, non-speciﬁc interactions of the probe molecules with the substrate surface, immobilization only via speciﬁc functional groups (amino, carboxylic acid, etc.), preferentially using covalent bonds. Rapid immobilization reactions that allow the use of low concentrations of immobilization reagents. Well-controlled kinetics between the probe and target molecules on the surface that can be easily monitored with conventional spectroscopic methods.

6.3 Immobilization .

.

Little, if any, post-synthetic modiﬁcation of the probes before immobilization, to maximize the number of compounds that can be generated by solution or solidphase synthesis and minimize the cost of these reagents. Low background, high stability/reproducibility/sensitivity and low cost.

Immobilization techniques are primarily chosen according to the type of probe. There are two important classes of probes used in bioarrays that will be considered here: ODNs (which may also be cDNAs) and proteins (also oligopeptides). ODNs are simple oligomers of four different nucleotides. Each nucleotide has a negatively charged phosphate group, a sugar molecule and a base carrying amino groups which form hydrogen bonds with other similar groups. Single-stranded ODNs are often used as probes; they are usually designed and synthesized from the target sequences. The probe is constructed as the complementary of this target and hybridization (binding two complementary ODNs) on the substrate surface is the basis of the detection method. A number of chemical modiﬁcation strategies have been employed by researchers for the attachment of ODNs to substrate surfaces (e.g. glass, silicon and polymeric). There are advantages and disadvantages to these approaches. Substrate surfaces can be coated with positively charged polycations (e.g. polyethyleneimine and polylysine). Then negatively charged probe ODNs are immobilized using non-speciﬁc ionic interactions, which are non-covalent. It is rather difﬁcult to have the correct orientation of the probe ODNs on the substrate surface, due to relatively weak bonds between the probe molecules and the support using this method. These arrays are often susceptible to detachment of the probe molecules from the surface when exposed to stringent hybridization conditions (high salt and/or high temperatures). Note also that downstream hybridization experiments using these substrates have resulted in experimental inconsistencies that have lead to uncertain results and difﬁculties in data interpretation. Covalent coupling chemistries for probe immobilization are preferred, because they generate stable and reproducible immobilization probe/substrate coupling. The substrate surface usually has functional groups, which are created in several ways, as explained in Section 6.2. Modiﬁcation of nucleic acids by silane compounds allows molecules to be attached covalently to even unmodiﬁed glass surfaces directly. Different procedures have been developed to covalently conjugate an active silane moiety on the ODNs or cDNA in solutions to form a new class of modiﬁed nucleic acid: silanized nucleic acids. Silanized ODNs and cDNA have been shown to immobilize readily on glass slides upon deposition. The immobilization is fast and arrays produced with this method gave strong hybridization signals with negligible background [79]. As exempliﬁed in Table 6.3, functionalized ODNs can be covalently immobilized onto substrate surfaces carrying different functional groups. In some cases, reactions are fast and require no other chemicals, such as reactions between isocyanate or epoxy residues [80]. While, some others (e.g. carboxylic acid, amino, hydroxyl, aldehyde) require extra chemicals, such as coupling agents (usually several types of carbodiimides) and stabilizers. Gold surfaces are usually used without any functionalization step, as mentioned previously [41, 42]. Thiol modiﬁed-ODNs

j95

j 6 Probe Immobilization Techniques in Array Technologies

96

Table 6.3 Some examples of functionalized surfaces and ODNs.

Surface

ODNs

Amino carrying

carboxylated phosphorylated

Carboxylic acid, isocyanate, epoxy or aldehyde carrying

aminated hydroxylated

Aminosilane via hetero-bifunctional cross-linker or mercaptosilane modiﬁed

thiol modiﬁed bisulﬁde modiﬁed

interact easily with gold surfaces and form covalent bonds, which is of course the most time efﬁcient method [81, 82]. In order to make probe ODNs freely available for the target, so-called spacer arms are used, which are located between the probe ODN molecule and the functional endgroup [83]. For instance in the case, where ODNs are immobilized onto gold surfaces, thiolated hydrocarbon chains with approximately six carbon atoms [e.g. SH(CH2)6] are usually used as spacer arms [84]. Longer homonucleic acid chains (e.g. the oligo timin of about 15 base pairs in length) may also be used. However, it should be noted that longer chains do bend and may cause loss of orientation (therefore activity) of the probe. Ideally, spacer arm-modiﬁed ODN probes should interact with the surface through the functional end-groups (e.g. the sulfur atom of the thiol group). However, some studies have shown that, nitrogen containing nucleoside side chains of the probe and also the homonucleic acid type of spacer arms may bend and interact directly with the surface. This is undesirable and should be prevented. Glutaraldehyde (GA) is a bifunctional linker, and has two reactive aldehyde endgroups. These have been very widely used therefore to cross-link amino-carrying molecules to surfaces (e.g. proteins) [85, 86]. In this procedure amino-modiﬁed surfaces are incubated with GA aqueous solution (2.5%) at 4 C for about 2 h, to convert surface amino groups to aldehyde groups. These are then treated further with amino-modiﬁed probe molecules (e.g. ODNs), using approximately the same conditions for immobilization [86]. Here, six carbon atoms carrying GA molecules act not only as a biofunctional linker, but also as an effective spacer arm. One should be careful in the GA treatment to reduce cross-linking reactions and GA homopolymerization on the surface or in the solution. Helper molecules (modulator spacer molecules) are also used to control the surface conﬁguration of probe molecules (i.e. to orient them vertically on the surface) [87]. A typical example is oriented immobilization of probe ODNs onto gold surfaces – a method which is routinely utilized in the authors laboratory. The most commonly used helper molecule is mercaptohexanol (MCH), which can form nice SAMs on gold surfaces. In order to simulate its structure, probe ODNs are modiﬁed by attaching a thiolated hydrocarbon chain with approximately six carbon atoms [SH(CH2)6] at the end-group, which acts as a spacer arm [31]. Here, it should be noted that the optimal length of the helper molecule should be equal to the length

6.3 Immobilization

Figure 6.5 Helper molecules increase surface orientation: (A) without helper molecules, (B) with helper molecules (MCH).

of the spacer group of the probe ODN. These thiol-modiﬁed single-stranded ODN probe molecules interact with a surface not only through their thiol end-groups but also via nitrogen atom-containing nucleotide bases and lose their surface orientation, as schematically depicted in Figure 6.5(A). However, when MCH molecules are put on the surface they help/push the ODN molecules into a re-oriented position, which makes them available for hybridization with target single-stranded ODNs. The MCH: probe ODNs ratio is critical for optimization of the probe molecules surface concentration (density), as this eliminates (or reduces) steric interferences. Alternatively, substrate surfaces (mainly gold) can be covered ﬁrst with SAMs with usually long-chain alkanethiols as a linker layer, which serves as a functionalized structure for further modiﬁcations at the surface. They create a barrier to prevent probe molecules (especially proteins) from coming into contact with the substrate surfaces (especially metallic substrates) [45]. These SAM-forming linkers have suitable reactive groups on one end of the molecule and a gold-complexing thiol on the other [88]. Alkanethiol molecules can be both chemisorbed and physiosorbed [89]. The probe molecules can then be attached directly to the alkanethiol monolayer. Phosphite-triester chemistry is another method for directly attaching the ﬁrst nucleotide to a glass plate by covalent bonding between the hydroxyl group of the glass surface and the phosphate group of the protected deoxyribonucleotide. This covalent bond is similar in structure and strength to the phosphodiester bonds within the DNA molecule [90]. An alternative simple immobilization strategy for attaching ODNs onto substrate surfaces has been proposed by the authors group, in which the probe and its opposite strand are synthesized with a particular complementary sequence and hybridized to form a double-stranded ODN [86]. Note that the probe contains one extra nucleotide at the 50 -end and after complete hybridization of these two complementary single strands the double strand contains a free base at one end. This means that the amino

j97

j 6 Probe Immobilization Techniques in Array Technologies

98

groups of this base are available for covalent binding of the probe ODN onto the substrate (e.g. carrying amino groups), using GA as a linker and spacer arm. Dendritic linkers, initially developed by Tomalia et al. in the early 1980s [91], can be attached to silica surfaces, enabling the highly efﬁcient immobilization of aminomodiﬁed nucleic acids and other probes [46, 91, 92]. These surfaces are coated with a thin layer of covalently immobilized and cross-linked PAMAM starburst dendrimers. The PAMAM moieties provide a high density of terminal amino groups on the outer substrate surface, which increases signiﬁcantly the binding capacity and homogeneity of probe molecules, and ensures their stability against regeneration procedures. Avidin and streptavidin have also been used to immobilize bioligands tagged with biotin (so-called biotinylated probes) onto substrate surfaces in highly oriented conformations [93, 94]. Both are tetrameric proteins with a molecular weight of around 60 kDa and have a very speciﬁc and strong binding ability for biotin (note that avidin is obtained from egg-white and is a glycoprotein, and that carbohydrate moieties reduce its speciﬁcity in afﬁnity applications) [95]. Streptavidin produced from Streptomyces avidinii contains no carbohydrate moieties and therefore exhibits much higher speciﬁcity. Its isoelectric point is also around 5–6, which is an extra advantage in use compared to avidin [96]. A schematic protocol for immobilization of biotinlyated ODNs onto surfaces using streptavidin, which has been used successfully by the authors group, is given in Figure 6.6 [97–99]. The ﬁrst step is to modify the substrate surface to create suitable functional groups (e.g. carboxylic acid groups). Streptavidin is then immobilized via its amino-groups using carboxylic acid moieties on the surface and a coupling agent [e.g. 1-ethyl-3-(3-dimethylaminopropyl)

Figure 6.6 Immobilization of biotinylated probe ODNs via streptavidin: (A) streptavidin immobilization onto carboxylated substrate surfaces and (B) oriented immobilization of biotinylated probe ODNs. EDC ¼ 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide.

References

carbodiimide]. The surface is then incubated with solutions containing the probe biotinlyated ODNs, which then become immobilized. Proteins, especially antibodies and also in some cases oligopeptides, are also used as bio-speciﬁc probes in protein arrays. Almost all immobilization protocols described above for ODNs can also be applied using protein probes. Unlike ODNs, proteins are much more sensitive to environmental conditions (temperature, pH, ionic strength, etc.) and can become easily denatured by physical or chemical effects. Their speciﬁcities are due to their three-dimensional structure, therefore any deformation (denaturation) may result lost of activity (this also means any biospeciﬁc recognition ability). Proteins prefer hydrophobic surfaces and adsorb on these types of surfaces by non-speciﬁc interactions. These non-speciﬁc interactions may lead to spurious results and should therefore be prevented, not only at the immobilization step of the protein probes, but also in detection of the target proteins. In many detection systems target molecules are labeled, usually with ﬂuorescence molecules (markers). Marker binding may also adversely affect the natural activity of proteins, therefore precautionary measures should also be taken at this step.

Acknowledgments

E.P. is supported by the Turkish Academy of Sciences as a Full Member.

References 1 Li, H., Park, S.H., Reif, J.H., LaBean, T.H. and Yan, H. (2004) DNA-templated self-assembly of protein and nanoparticle linear arrays. J. Am. Chem. Soc., 126, 418. 2 Zhang, G., Yan, X., Hou, X., Lu, G., Yang, B., Wu, L., and Shen, J. (2003) Binary DNA arrays on heterogeneous patterned surfaces. Langmuir 19, 9850. 3 Robelek, R., Niu, L., Schmid, E.L. and Knoll, W. (2004) Multiplexed hybridization detection of quantum dot-conjugated DNA sequences using surface plasmon enhanced ﬂuorescence microscopy and spectrometry. Anal. Chem., 76, 6160. 4 Strother, T., Hamers, R.J. and Smith, L.M. (2000) Covalent attachment of oligodeoxyribonucleotides to amine-modiﬁed Si (001) surfaces. Nucleic Acids Res., 28, 3535.

5 Afanassiev, V., Hanemann, V. and W€olﬂ, S. (2000) Preparation of DNA and protein micro arrays on glass slides coated with an agarose ﬁlm. Nucleic Acids Res., 28, e66. 6 Le Berre, V., Trevisiol, E., Dagkessamanskaia, A., Sokol, S., Caminade, A.M., Majoral, J.M., Meunier, B. and François, J. (2003) Dendrimeric coating of glass slides for sensitive DNA microarrays analysis. Nucleic Acids Res., 31, e88. 7 Park, S., Lee, M., Pyo, S.J. and Shin, I. (2004) Carbohydrate chips for studying high-throughput carbohydrate–protein interactions. J. Am. Chem. Soc., 12, 4812. 8 Falsey, J.R., Renil, M., Park, S., Li, S. and Lam, K.S. (2001) Peptide and small molecule microarray for high throughput cell adhesion and functional assays. Bioconjugate Chem., 12, 346.

j99

j 6 Probe Immobilization Techniques in Array Technologies

100

9 Duburcq, X., Olivier, C., Malingue, F., Desmet, R., Bouzidi, A., Zhou, F., Auriault, C., Gras-Masse, H. and Melnyk, O. (2004) Peptide–protein microarrays for the simultaneous detection of pathogen infections. Bioconjugate Chem., 15, 307. 10 Yu, C. and Irudayaraj, J. (2007) Multiplex biosensor using gold nanorods. Anal. Chem., 79, 572. 11 Wang, L., Zhao, W., ODonoghue, M.B. and Tan, W. (2007) Fluorescent nanoparticles for multiplexed bacteria monitoring. Bioconjugate Chem., 18, 297. 12 Zhou, D., Sinniah, K., Abell, C. and Rayment, T. (2003) Label-free detection of DNA hybridization at the nanoscale: a highly sensitive and selective approach using atomic-force microscopy. Angew. Chem., 115, 5084. 13 Wang, Z., Levy, R., Fernig, D.G. and Brust, M. (2005) The peptide route to multifunctional gold nanoparticles. Bioconjugate Chem., 16, 497. 14 Podyminogin, M.A., Lukhtanov, E.A. and Reed, M.W. (2001) Attachment of benzaldehyde-modiﬁed oligodeoxynucleotide probes to semicarbazide-coated glass. Nucleic Acids Res., 15, 5090. 15 Guo, Z., Guilfoyle, R.A., Thiel, A.J., Wang, R. and Smith, L.M. (1994) Direct ﬂuorescence analysis of genetic polymorphisms by hybridization with oligonucleotide arrays on glass supports. Nucleic Acids Res., 22, 5456. 16 Beier, M. and Hoheisel, J.D. (1999) Versatile derivatization of solid support media for covalent bonding on DNAmicrochips. Nucleic Acids Res., 27, 1970. 17 Lenigk, R., Carles, M., Ip, N.Y. and Sucher, N.J. (2001) Surface characterization of a silicon-chip-based DNA microarray. Langmuir, 17, 2497. 18 Waddell, T.G., Leyden, D.E. and DeBello, M.T. (1981) The nature of organosilane to silica-surface bonding. J. Am. Chem. Soc., 103, 5303. 19 Linford, M.R., Fenter, P., Eisenberger, P.M. and Chidsey, C.E.D. (1995) Alkyl monolayers on silicon prepared from

20

21

22

23

24

25

26

27

28

29

1-alkenes and hydrogen-terminated silicon. J. Am. Chem. Soc., 117, 3145. Thiel, A.J., Frutos, A.G., Jordan, C.E., Corn, R.M. and Smith, L.M. (1997) In situ surface plasmon resonance imaging detection of DNA hybridization to oligonucleotide arrays on gold surfaces. Anal. Chem, 69, 4948. Wegner, G.J., Lee, H.J., Marriott, G. and Corn, R.M. (2003) Fabrication of histidinetagged fusion protein arrays for surface plasmon resonance imaging studies of protein–protein and protein–DNA interactions. Anal. Chem., 75, 4740. Haes, A.J. and Van Duyn, R.P. (2004) A uniﬁed view of propagating and localized surface plasmon resonance biosensors. Anal. Bioanal. Chem., 379, 920. Zhao, J., Zhang, X., Yonzon, C.R., Hales, A.J. and Van Duyne, R.P. (2006) Localized surface plasmon resonance biosensors. Nanomedicine, 1, 219. Hutter, E. and Fendler, J.H. (2004) Exploitation of localized surface plasmon resonance. Adv. Mater., 19, 1685. Cao, Y.C., Jin, R., Thaxton, C.S. and Mirkin, C.A. (2005) A two-color-change, nanoparticle-based method for DNA detection. Talanta, 67, 449. Storhoff, J.J., Lucas, A.D., Garimella, V., Bao, Y.P. and M€ uler, U.R. (2004) Homogeneous detection of unampliﬁed genomic DNA sequences based on colorimetric scatter of gold nanoparticle probes. Nat. Biotechnol., 22, 883. Nath, N. and Chilkoti, A. (2004) Label-free biosensing by surface plasmon resonance of nanoparticles on glass: optimization of nanoparticle size. Anal. Chem., 76, 5370. Lin, H.Y., Chen, C.T. and Chen, Y.C. (2006) Detection of phosphopeptides by localized surface plasma resonance of titania-coated gold nanoparticles immobilized on glass substrates. Anal. Chem., 78, 6873. Habb, B.B. (2001) Advances in protein microarray technology for protein expression and interaction proﬁling. Curr. Opin. Drug Discov., 4, 116.

References 30 Cahill, D.J. (2001) Protein and antibody arrays and their medical applications. J. Immunol. Methods, 250, 81. 31 Herne, T.M. and Tarlov, M.J. (1997) Characterization of DNA probes immobilized on gold surfaces. J. Am. Chem. Soc., 119, 8916. 32 Nuzzo, R.G. and Allara, D.L. (1983) Adsorption of bifunctional organic disulﬁdes on gold surface. J. Am. Chem. Soc., 105, 4481. 33 Nuzzo, R.G., Dubois, L.H. and Allara, D.L. (1990) Fundamental studies of microscopic wetting on organic surfaces. 1. Formation and structural characterization of a self-consistent series of polyfunctional organic monolayers. J. Am. Chem. Soc., 112, 558. 34 Porter, M.D., Bright, T.B., Allara, D.L. and Chidsey, C.D. (1987) Spontaneously organized molecular assemblies. 4. Structural characterization of n-alkyl thiol monolayers on gold by optical ellipsometry, infrared spectroscopy, and electrochemistry. J. Am. Chem. Soc., 109, 3559. 35 Strong, L. and Whitesides, G.M. (1988) Structures of self-assembled monolayer ﬁlms of organosulfur compounds adsorbed on gold single crystals: electron diffraction studies. Langmuir, 4, 546. 36 Arnold, R., Terfort, A. and Woll, C. (2001) determination of molecular orientation in self-assembled monolayers using IR absorption intensities: the importance of grinding effects. Langmuir, 17, 4980. 37 Houssiau, L., Graupe, M., Colorado, R., Kim, H.I., Lee, T.R., Perry, S.S. and Rabalais, J.W. (1998) Characterization of the surface structure of CH3 and CF3 terminated n-alkanethiol monolayers self assembled on Au{111}. J. Chem. Phys., 109, 9134. 38 Buscher, C.T., McBranch, D. and Li, D.Q. (1996) Understanding the relationship between surface coverage and molecular orientation in polar self-assembled monolayers. J. Am. Chem. Soc., 118, 2950. 39 Schlenoff, J.B., Li, M. and Ly, H. (1995) Stability and self-exchange in

40

41

42

43

44

45

46

47

48

alkanethiol monolayers. J. Am. Chem. Soc., 117, 12528. Bain, C.D., Troughton, E.B., Tao, Y.T., Evall, J., Whitesides, G.M. and Nuzzo, R.G. (1989) Formation of monolayer ﬁlms by the spontaneous assembly of organic thiols from solution onto gold. J. Am. Chem. Soc., 111, 321. Wang, J. and Bard, A.J. (2001) monitoring DNA immobilization and hybridization on surfaces by atomic force microscopy force measurements. Anal. Chem., 73, 2207. Mark, S.S., Sandhyarani, N., Zhu, C., Campagnolo, C. and Batt, C.A. (2004) Dendrimer-functionalized self assembled monolayers as a surface plasmon resonance sensor Surface. Langmuir, 20, 6808. Riepl, M., Enander, K., Liedberg, B., Schaferling, M., Kruschina, M. and Ortigao, F. (2002) Functionalized surfaces of mixed alkanethiols on gold as a platform for oligonucleotide microarrays. Langmuir, 18, 7016. Weinberger, D.A., Hong, S., Mirkin, C.A., Wessels, B.W. and Higgins, T.B. (2000) Combinatorial generation and analysis of nanometer- and micrometer-scale silicon features via dip-pen nanolithography and wet chemical etching. Adv. Mater., 12, 1600. Lee, K.B., Park, S.J., Mirkin, C.A., Smith, J.C. and Mrksich, M. (2002) Protein nanoarrays generated by dip-pen nanolithography. Science, 295, 1702. Benters, R., Niemeyer, C.M., Drutschmann, D., Blohm, D. and W€ohrle, D. (2002) DNA microarrays with PAMAM dendritic linker systems. Nucleic Acids Res., 30, e10. Taylor, S., Smith, S., Windle, B. and Elie, A.G. (2003) Impact of surface chemistry and blocking strategies on DNA microarrays. Nucleic Acids Res., 31, 16. Bertucci, F., Bernard, K., Loriod, B., Chang, Y.C., Granjeaud, S., Birnbaum, D., Nguyen, C., Peck, K. and Jordan, B.R. (1999) Sensitivity issues in DNA

j101

j 6 Probe Immobilization Techniques in Array Technologies

102

49

50

51

52

53

54

55

56

57

array-based expression measurements and performance of nylon microarrays for small samples. Human Mol. Genet., 8 (9), 1715. Matsuura, K., Oda, R., Kitakouji, H., Kiso, M., Kitajima, K. and Kobayashi, K. (2004) Surface plasmon resonance study of carbohydrate–carbohydrate interaction between various gangliosides and Gg3carrying polystyrene. Biomacromolecules, 5, 937. Pirri, G., Damin, F., Chiari, M., Bontempi, E. and Depero, L.E. (2004) Characterization of a polymeric adsorbed coating for DNA microarray glass slides. Anal. Chem., 76, 1352. Rajeshwar, K., de Tacconi, N.R. and Chenthamarakshan, C.R. (2001) Semiconductor-based composite materials: preparation, properties, and performance. Chem. Mater., 13, 2765. Kyo, M., Usui-Aoki, K. and Koga, H. (2005) Label-free detection of proteins in crude cell lysate with antibody arrays by a surface plasmon resonance imaging technique. Anal. Chem., 77, 7115. Costa-Fernandez, J.M., Pereiro, R. and Sanz-Medel, A. (2006) The use of luminescent quantum dots for optical sensing. Trends Anal. Chem., 25, 207. Ozkan, M. (2004) Quantum dots and other nanoparticles: what can they offer to drug discovery? Drug Discov. Today, 9, 1065. Michalet, X., Pinaud, F.F., Bentolila, L.A., Tsay, J.M., Doose, S., Li, J.J., Sundaresan, G., Wu, A.M., Gambhir, S. S. and Weiss, S. (2005) Quantum dots for live cells, in vivo imaging, and diagnostics. Science, 307, 538. Tully, E., Hearty, S., Leonard, P. and OKennedy, R. (2006) The development of rapid ﬂuorescence-based immunoassays, using quantum dot-labeled antibodies for the detection of Listeria monocytogenes cell surface proteins. Int. J. Biol. Macromol., 39, 127. Bailey, R.E., Smith, A.M. and Nie, S. (2005) Quantum dots in biology and medicine. Physica E, 25, 1.

58 Chan, W.C.W., Maxwell, D.J., Gao, X., Bailey, R.E., Han, M. and Nie, S. (2002) Luminescent quantum dots for multiplexed biological detection and imaging. Curr. Opin. Biotechnol., 13, 40. 59 Mattoussi, H., Mauro, J.M., Goldman, E.R., Anderson, G.P., Sundar, V.C., Mikulec, F.V. and Bawendi, M.G. (2000) Self-assembly of CdSe–ZnS quantum dot bioconjugates using an engineered recombinant protein. J. Am. Chem. Soc., 12, 12142. 60 Gerion, D., Parak, W.J., Williams, S.C., Zanchet, D., Micheel, C.M. and Alivisatos, A.P. (2002) Sorting ﬂuorescent nanocrystals with DNA. J. Am. Chem. Soc., 124, 7070. 61 Ho, Y.P., Kung, M.C., Yang, S.T. and Wang, H. (2005) Multiplexed hybridization detection with multicolor colocalization of quantum dot nanoprobes. Nano Lett., 5, 1693. 62 Alivisatos, A.P. (1996) Semiconductor clusters, nanocrystals, and quantum dots. Science, 271, 933. 63 Murray, C.B., Norris, D.J. and Bawendi, M.G. (1993) Synthesis and characterization of nearly monodisperse CdE (E ¼ S, Se, Te) semiconductor nanocrystallites. J. Am. Chem. Soc., 115, 8706. 64 Dubertret, B., Skourides, P., Norris, D.J., Noireaux, V., Brivanlou, A.H. and Libchaber, A. (2002) In vivo imaging of quantum dots encapsulated in phospholipid micelles. Science, 298, 1759. 65 Akerman, M.E., Chan, W.C.W., Laakkonen, P., Bhatia, S.N. and Ruoslahti, E. (2002) Nanocrystal targeting in vivo. Proc. Natl Acad. Sci. USA, 99, 12617. 66 Mitchell, G.P., Mirkin, C.A. and Letsinger, R.L. (1999) Programmed assembly of DNA functionalized quantum dots. J. Am. Chem. Soc., 121, 8122. 67 Patolsky, F., Gill, R., Weizmann, Y., Mokari, T., Banin, U. and Willner, I. (2003) Lighting-up the dynamics of telomerization and DNA replication by

References

68

69

70

71

72

73

74

75

76

77

CdSe–ZnS quantum dots. J. Am. Chem. Soc., 125, 13918. Mitchell, G.P., Mirkin, C.A. and Letsinger, R.L. (1999) Programmed assembly of DNA functionalized quantum dots. J. Am. Chem. Soc., 121, 8122. Chan, W.C.W. and Nie, S.M. (1998) Quantum dot bioconjugates for ultrasensitive nonisotopic detection. Science, 281, 2016. Dahan, M., Laurence, T., Pinaud, F., Chemla, D.S., Alivisatos, A.P., Sauer, M. and Weiss, S. (2001) Time-gated biological imaging by use of collloidal quantum dots. Opt. Lett., 26, 825. Weissleder, R., Tung, C.H., Mahmood, U. and Bogdanov, A. Jr. (1999) In vivo imaging of tumors with protease-activated nearinfrared ﬂuorescent probes. Nat. Biotechnol., 17, 375. Bremer, C., Tung, C.H. and Weissleder, R. (2001) In vivo molecular target assessment of matrix metalloproteinase inhibition. Nat. Med., 7, 743. Han, M.Y., Gao, X., Su, J.Z. and Nie, S.M. (2001) Quantum-dot tagged microbeads for multiplexed optical coding of biomolecules. Nat. Biotechnol., 19, 631. Daniele, G.Qdots technology for detection available at LLNL http://physci.llnl.gov/ Research/qsg-090205/graphics/people/ danieleG_qdotdoc.pdf. Sun, B., Xie, W., Yi, G., Chen, D., Zhou, Y. and Cheng, J. (2001) Microminiaturized immunoassays using quantum dots as ﬂuorescent label by laser confocal scanning ﬂuorescence detection. J. Immunol. Methods, 249, 85. Goldman, E.R., Balighian, E.D., Mattoussi, H., Kuno, M.K., Mauro, J.M., Tran, P.T. and Anderson, G.P. (2002) Avidin: a natural bridge for quantum dot–antibody conjugates. J. Am. Chem. Soc., 124, 6378. Goldman, E.R., Clapp, A.R., Anderson, G.P., Uyeda, H.T., Mauro, J.M., Medintz, I.L. and Mattoussi, H. (2004) Multiplexed toxin analysis using four colors of quantum dot ﬂuororeagents. Anal. Chem., 76, 684.

78 Dwarakanath, S., Bruno, J.G., Shastry, A., Phillips, T., John, A., Kumar, A. and Stephenson, L.D. (2004) Quantum dot–antibody and aptamer conjugates shift ﬂuorescence upon binding bacteria. Biochem Biophys. Res. Commun., 325, 739. 79 Kumar, A., Larsson, O., Parodi, D. and Linag, Z. (2001) Silanized nucleic acids: a general platform for DNA immobilization. Nucleic Acids. Res., 28, e71. 80 Kieffer, A., Hartwig, A., Schmidt-Naake, G. and Hennemann, O.D. (1998) Reactions at the interphase between an isocyanate and epoxy resins – infrared spectroscopic investigations. Acta Polymerica, 49, 12, 720. 81 Moa, X.T., Zhoub, Y.P., Leia, H. and Denga, L. (2002) Microbalance-DNA probe method for the detection of speciﬁc bacteria in water. Enzyme Microbial Technol., 30, 583. 82 Herne, T.M. and Tarlov, M.J. (1997) Characterization of DNA probes immobilized on gold surfaces. J. Am. Chem. Soc, 119, 8916. 83 Nelson, B.P., Grimsrud, T.E., Liles, M.R., Goodman, R.M. and Corn, R.M. (2001) Surface plasmon resonance imaging measurements of DNA and RNA hybridization adsorption onto DNA microarrays. Anal. Chem., 73, 1. 84 Mannelli, I., Minunni, M., Tombelli, S., Wang, R., Spiriti, M.M. and Macsini, M. (2005) Direct immobilization of DNA probes for the development of afﬁnity biosensors. Bioelectrochemistry, 66, 129. 85 Saber, R., Mutlu, S. and Pi¸skin, E. (2002) Glow/discharge treated piezoelectric quartz crystals as immunosensors for HSA detection. Biosens. Bioelectron., 17, 727. 86 Duman, M., Saber, R. and Piskin, E. (2003) A new approach for immobilization of oligonucleotides onto piezoelectric quartz crystal for preparation of a nucleic acid sensor for following hybridization. Biosens. Bioelectron., 18, 1355. 87 Satjapipat, M., Sanedrin, R. and Zhou, F. (2001) Selective desorption of alkanethiols in mixed self assembled monolayers for subsequent oligonucleotide attachment

j103

j 6 Probe Immobilization Techniques in Array Technologies

104

88

89

90

91

92

93

and DNA hybridization. Langmuir, 17, 7644. Duschl, C., Sevin-Landais, A. and Vogel, H. (1996) Surface engineering: optimization of antigen presentation in self-assembled monolayers. Biophys. J., 70, 1985. Peterlinz, K.A. and Georgiadis, R. (1996) In situ kinetics of self-assembly by surface plasmon resonance spectroscopy. Langmuir, 12, 4731. Cohen, G., Deutsch, J., Fineberg, J. and Levine, A. (1997) Covalent attachment of DNA oligonucleotides to glass. Nucleic Acids. Res., 25, 911. Tomalia, D.A., Naylor, A.M. and Goddard, W.A. (1990) Starburst dendrimers: molecular-level control of size, shape, surface chemistry, topology and ﬂexibility from atoms to macroscopic matter. Angew. Chem. Int. Ed. Engl., 29, 138. Benters, R., Niemeyer, C.M. and Wohrle, D. (2001) Dendrimer-activated solid supports for nucleic acid- and protein-microarrays. Chem. Biol. Chem., 2, 686. Dontha, N., Nowall, W.B. and Kuhr, W.G. (1997) Generation of biotin/avidin/ enzyme nanostructures with maskless photolithography. Anal. Chem., 69, 2619.

94 Chen, H.Y., Elkasabi, Y. and Lahann, J. (2006) Surface modiﬁcation of conﬁned microgeometries via vapor-deposited polymer coatings. J. Am. Chem. Soc., 128, 1, 374. 95 Schriemer, D.C. and Li, L. (1996) Combining avidin–biotin chemistry with matrix-assisted laser desorption/ ionization mass spectrometry. Anal. Chem., 68, 19, 3382. 96 Green, N.M. (1990) Avidin and streptavidin. Methods Enzymol., 184, 51. 97 Niemeyer, C.M., Ceyhan, B. and Blohm, D. (1999) Functionalization of covalent DNA–streptavidin conjugates by means of biotinylated modulator components. Bioconjugate Chem., 10, 708. 98 Wang, J., Li, J., Baca, A.J., Hu, J., Zhou, F., Yan, W. and Pang, D.W. (2003) Ampliﬁed voltammetric detection of DNA hybridization via oxidation of ferrocene caps on gold nanoparticle/ streptavidin conjugates. Anal. Chem., 75, 3941. 99 Erdem, A., Sayar, F., Karadeniz, H., Guven, G., Ozsoz, M. and Piskin, E. (2007) development of streptavidin carrying magnetic nanoparticles and their applications in electrochemical nucleic acid sensor systems. Electroanalysis, 19, 798.

j105

Part II: Identification

j107

7 Low-Cost and Low-Density Microarrays – A Novel Technique for Identification and Typing of Microorganisms Dimitrios Frangoulidis, Volker Heiser, Olfert Landt, and Hermann Meyer

7.1 Introduction

During the last decade, a variety of different microarray platforms have been described for identiﬁcation and typing of bacteria and viruses. Today, high-density oligonucleotide-based whole-genome microarrays can be used for genomic analysis beyond simple gene expression proﬁling. However, when looking at hundreds of thousands of analytes, the need to focus on subsets of those analytes is the basis for mid- and low-density panels built around disease states or genes of interest. Furthermore, DNA microarray technology allows a comparison of different levels of gene expression in all kinds of specimens, permitting comparisons between different infection status, treatments and physiological status. So far, microarray technology has relied on well-equipped laboratories and welltrained personnel. However, research is directed to simplify this technique for use in the ﬁeld. In order to compete with or supplement so-called immunological based point-of-care assays, the microarray technology has to be as simple, reliable and robust as possible. Here, we describe a novel and fast low-cost and low-density (LCD) DNA microarray principle that could be used for direct identiﬁcation and presumptive typing of pathogens.

7.2 Chip Design/Array Description

The LCD microarray uses a so-called hybridization assay format that reacts with products of a pre-performed speciﬁc polymerase chain reaction (PCR). It consists of a transparent, pre-structured polymer support with the size of a standard slide (50 50 mm), as shown in Figure 7.1. It contains eight identical microarrays in rectangular reaction chambers that can be addressed individually. Each array contains the same set of different immobilized oligonucleotides able to hybridize with an amplicon. The oligonucleotides (usually 25–30mers) are immobilized on

j 7 Low-Cost and Low-Density Microarrays – A Novel Technique for Identification

108

Figure 7.1 LCD DNA microarray: (1 þ 2 and 3 þ 4) examples of different oligonucleotides spotted in duplicates, (5 þ 6) position of internal controls and (C) hybridization and staining controls.

spots with an average diameter of 300 mm and are arranged in the form of a square consisting of seven rows and seven lanes, resulting in 49 spots. Three dots (probes with an unrelated sequence motif – fungal origin) serve as a control to monitor correct hybridization and staining. They are positioned in both upper corners and in the lower corner on the right-hand side. In this way the user can position the array grid correctly for data analysis. Agent-speciﬁc oligonucleotides are dotted in duplicates. The slide-like format of the array and the use of a precipitating substrate (biotin–avidin complex) to demonstrate successful hybridization allows an easy interpretation of results even by the naked eye. By using a simple transmission light scanning device, grayscale images of 10-mm resolution are generated for data analysis. An internal control system (l DNA) is added to control the PCR reaction.

7.3 Protocol for a LCD Array Experiment 7.3.1 Amplification

An aliquot of 1 ml of primer mix (biotinylated) should be used to prime a 25-ml PCR reaction. If the mixes are used in combinations (duplex or multiplex PCR), 1 ml of each mix should be used to prime a 25-ml PCR reaction. Standard PCR protocols are required; however, depending on the quality and concentration of the template, the cycle number might be increased or/and the annealing and elongation times prolonged.

7.3 Protocol for a LCD Array Experiment

7.3.2 Short Protocol for Hybridization and Labeling

The buffers are provided (e.g. hybridization buffer, modulator, etc.). 1. Water bath at 35 C 2. 22 ml hybridization buffer (equilibrated to room temperature) combined with 2 ml of modulator buffer and 10 ml of PCR reaction (hybridization mix); this mix is pipetted to a single set of the microarray, which is subsequently incubated at 35 C for 30 min in a humidity chamber (Figure 7.2) 3. Prepare three wash containers with wash solution 4. Rinse slide in wash container 1 and 2 for 10 s each, and incubate in wash container 3 for 1 min 5. Dry the slide by spinning at around 1000 rpm 6. Prepare the labeling mix by combining 270 ml of dilution buffer, 30 ml of modulator and 2 ml of label (avidin complex) (for eight reactions) 7. Apply 30 ml of label mix to each ﬁeld of the slide and incubate for 5 min 8. Replace the wash solution in all containers and repeat the wash procedure from steps (5) þ (6) 9. Dry the slide as in step (5) 10. Apply 30 ml of stain to each ﬁeld of the slides. Avoid contamination of the stain solution 11. Carefully observe the staining process and stop the process by rinsing the slide in the last wash container for 10 s 12. Dry the slide as in step (5) 13. Slide analysis See also Figure 7.3.

Figure 7.2 Adding solutions.

j109

j 7 Low-Cost and Low-Density Microarrays – A Novel Technique for Identification

110

Figure 7.3 Work flow for hybridization, washing and staining steps.

7.5 Conclusions

Figure 7.4 Appropriate scanner and software.

7.4 Results and Discussion

Results can be analyzed by a scanner device and software (Figure 7.4). In addition, the reactions can be seen by holding the slides into light or in front of a slide viewer (Figure 7.5). Hybridizations control dots should be easily distinguished in the three corners of the array ﬁeld. Speciﬁc reaction dots are clearly stained and no unspeciﬁc background is seen.

7.5 Conclusions

This new technique provides a very fast (45 min with 15 min hands-on time only) economic, sensitive and speciﬁc detection of PCR ampliﬁcation products. In addition, only minimal laboratory equipment is required, allowing this method to be

j111

j 7 Low-Cost and Low-Density Microarrays – A Novel Technique for Identification

112

Figure 7.5 Results of the reactions: visual inspection.

introduced into ﬁeld situations, and clinical and epidemiological studies in animals and humans to be conducted. It could be designed as a multiplex assay for identiﬁcation of several independent pathogens causing a similar clinical entity, such as respiratory or neurological disorders. Alternatively, it could be used for single nucleotide polymorphism analysis. The latter is of advantage for differentiation of closely related microorganisms. In this case the LCD-Chip method can compete with real-time PCR assays based on hybridization probes because up to 20 variants can be detected in one single assay, thereby reducing time and avoiding a variety of costly hybridization probes. This assay is suitable for many differently equipped laboratories and results can be compared without any additional devices. It is low cost and its simplicity favors application in developing countries.

j113

8 DNA Microarray Technique for Detection and Identification of Viruses Causing Encephalitis and Hemorrhagic Fever Henrik Nordstr€om, Kerstin I. Falk, Peter Nilsson, and Åke Lundkvist

8.1 Introduction

Our group at the Swedish Institute for Infectious Disease Control works with research and diagnostics for Biosafety Level 3 and 4 pathogens including viruses causing hemorrhagic fever and severe encephalitis. A major goal of our work is to develop microarray-based methods for detection and identiﬁcation of viral nucleic acid from these viruses, with assistance of our collaborators at the Royal Institute for Technology. The main advantage the technique provides is an ability to screen a sample for nucleic acid from several different viruses in one test.

8.2 Principle of Applying Microarray Technology for Virus Detection and Identification

The principle of a microarray-based test is illustrated in Figure 8.1. The most common sample type is serum, but could be various human clinical samples. Since most of these viruses are of the zoonotic type, bird or animal samples could also be expected. Soil or other environmental samples such as bioterrorism-associated powder samples are less likely to be encountered due to the high level of instability outside the host of these viruses. Most of the viruses of interest are RNA viruses, but for later inclusion of more viruses or other pathogens to the assays, extraction of both RNA and DNA equally well in the same protocol is preferred. For these RNA viruses, for which it is very important that they are completely killed, phenol/guanidine thiocyanate/chloroformbased methods, such as the Trizol protocol (Invitrogen; www.invitrogen.com) or TriPure (Roche; www.roche.com), are currently used; however, these are not optimal for DNA extraction.

j 8 DNA Microarray Technique for Detection and Identification of Viruses

114

Figure 8.1 Principle of the microarray assay for virus detection and identification.

Usually there are rather low amounts of viral nucleic acid in the sample, requiring a nucleic acid ampliﬁcation step before the microarray hybridization. This is really the key step of the whole procedure where the group of viruses targeted must be selected. For a larger group of distantly related viruses of different families (e.g. Flaviviridae and Bunyaviridae), with low or non-existent nucleic acid sequence similarity, but sometimes similar disease symptoms, a broad ampliﬁcation is needed. Ultimately a random ampliﬁcation targets all the nucleic acid in a sample, but is inefﬁcient and requires signal ampliﬁcation, making the whole procedure more time-consuming. For a group of more closely related viruses of the same family and genera (e.g. a group of ﬂaviviruses), a more speciﬁc and efﬁcient ampliﬁcation with degenerated primers targeting common nucleic acid sequence motifs can be used. Depending on the design of the method, the DNA can be labeled with ﬂuorescent dyes directly during ampliﬁcation or in an additional step. The labeled nucleic acid is puriﬁed and hybridized to the microarray. On the microarray slides, virus-speciﬁc DNA probes are attached. Depending on the design, commercial alternatives from Affymetrix (www.affymetrix.com), Agilent (www.agilent.com) and others are available, and give a higher reproducibility compared to the in-house spotted variants. However, the in-house variants are signiﬁcantly less expensive and more ﬂexible designs are possible, which is more suitable for these viruses. The hybridization is performed in a hybridization station that allows both mixing of the sample during incubation as well as a controlled stringency in terms of temperature, incubation times and amount of wash buffers used. A controlled stringency is critical for these assays, to pick up new strains or to achieve a highly selective identiﬁcation. Finally, the slide is scanned in a laser scanner and hybridization signals are quantiﬁed from the produced image for subsequent numerical analysis. Virusspeciﬁc hybridization patterns are looked for by importing the data into a Microsoft Excel template that produces a result visualized in a histogram.

8.4 Advantages and Drawbacks of Using the Microarray Technique

8.3 Viruses and the Importance of Rapid Diagnostics

The group of interesting viruses includes, from the Bunyaviridae family, hantaviruses such as Hantaan and Sin Nombre, Crimean-Congo hemorrhagic fever virus, a nairovirus, and Rift Valley fever virus, a phlebovirus. From the Filoviridae family, both Ebola viruses and Marburg virus are included, and from the Arenaviridae, Lassa virus. From the Flaviviridae family there are several mosquito-borne ﬂaviviruses, including Dengue viruses, Yellow fever virus, Japanese encephalitis virus and West Nile virus. Due to the severity of disease and need for both supportive care and patient isolation it is of utmost importance to make a rapid detection and identiﬁcation of the disease-causing pathogen. Nucleic acid-based methods are suitable since often no antibodies have developed in early disease. Identiﬁcation of viral nucleic acid in a sample is considered a strong indication of the presence of a certain pathogen. Ideally a one-test-for-all would be desired; however, as mentioned above, the ampliﬁcation step is a limiting factor since these viruses are mostly single-stranded RNA viruses with little or no mutual sequence motifs in between. New sequence variation is frequently encountered due to error-prone replication and the emergence of new strains. The considerable sequence variation could obstruct reverse transcription-polymerase chain reaction (RT-PCR) ampliﬁcation by mutations in primer sites, making detection together with species-level identiﬁcation the main emphasis for the development of new methods. Species-level identiﬁcation is usually sufﬁcient for clinical purposes for many of these viruses. Often there is little known about different diseases associated with certain strains or certain nucleic acid sequence motifs.

8.4 Advantages and Drawbacks of Using the Microarray Technique

Different virus-speciﬁc protocols of PCR, RT-PCR or real-time PCR/RT-PCR are widely used for detection and identiﬁcation of virus nucleic acid in clinical samples. The key factor is the annealing of a primer pair to the viral nucleic acid. This can give relatively high speciﬁcity, but is also the principal limitation (as described above). The microarray technique presents some potential advantages compared to the PCR-based protocols. Mainly, the high number of possible probes on the microarray provides a better multiplexing capacity by allowing investigation of more DNA fragments. This means that both more viruses and several parts of the genomes can be targeted in one test. A broader test is valuable, saving time and effort, as well as sample, in cases of an unclear clinical picture or for a broader screening of a set of samples. In addition, by combining a broad ampliﬁcation from a sample and microarray probes for different parts of the viral genome, there is a higher possibility to detect and identify viral nucleic acid and avoid false-negative results, even from

j115

j 8 DNA Microarray Technique for Detection and Identification of Viruses

116

new strains with new sequence variations. At the same time, hybridization to several probes for each virus is informative and increases the accuracy of virus identiﬁcation, to avoid false-positive results. The manner of hybridization further allows for detection and distinction of multiple infections by different viruses, which could be very difﬁcult to identify by PCR. Drawbacks with the technique are, primarily, the still rather undeveloped and complicated format, and the non-quantitative result. As for most assays, further veriﬁcation by other methods is needed, but a microarray test can efﬁciently function as a screening tool to assist in selecting a speciﬁc PCR.

8.5 Key Factors for Development of a Microarray-Based Test

By applying random nucleic acid ampliﬁcation, both a wide range of viruses as well as new and diverged strains could be ampliﬁed. The hybridization to the microarray extracts the viral sequence from the randomly ampliﬁed mixture of nucleic acid. Still, the drawback is the ampliﬁcation of parts of the viral genome not targeted by the microarray probes and non-viral nucleic acid that consumes reagents and makes the ampliﬁcation less efﬁcient. To achieve a sufﬁcient lower limit of detection, signal ampliﬁcation involving more ampliﬁcation (e.g. re-ampliﬁcation), and longer incubations times for labeling and hybridization are needed. If instead a group of closely related viruses with sequence sites in common is targeted, the ampliﬁcation can be more speciﬁc and accordingly more efﬁcient. Thereby, a more rapid test can be developed. Different approaches have been tested for the viruses of interest. First, by targeting smaller groups of related viruses such as hantaviruses and ﬂaviviruses. The extensive sequence variation among the ﬂaviviruses still requires semirandom approaches to target all strains. Later, a random protocol can be applied for a group of viruses with little or no mutual sequence, but causing similar hemorrhagic fever symptoms. The other factor that will inﬂuence the ability of the method to detect and discover new strains or to make a speciﬁc identiﬁcation of a certain strain is the probe length (Figure 8.2). With 20mer probes sometimes even single point mutation variations can be identiﬁed, providing highly speciﬁc sequence veriﬁcation. The drawback is that even with a large number of short probes for a certain virus there could still be problems ﬁnding new strains with more than 10% of new sequence variation evenly distributed. Longer, 60–70mer oligonucleotide probes increase the chances of ﬁnding new strains and 500mer probes can allow detection of new strains of a certain virus species that differs by 20–30% in nucleic acid sequence. We have mainly used 500mer probes for the viruses of interest due to an expected high sequence variation in new strains while still having virus identiﬁcation on the species level. The drawback is the time-consuming synthesis of the probe fragments by PCR, requiring access to viral RNA. For a broad screening of many different viruses the commercially produced long oligonucleotides are a more practical alternative, taking into account the possibility to miss out on some highly diverged new strains.

8.7 Flavivirus Microarray

Figure 8.2 Short or long probe strands on microarrays modulate specificity and ability to detect new strains. A new strain of virus B might not be detected based on mismatches with short probes.

8.6 Hantavirus Microarray

The initial project [1] involved hantaviruses, of which one species, Puumala virus, is endemic in Northern Sweden. These rodent-borne viruses have been discovered on the Eurasia, in North and South America, and recently in West Africa. The Eurasian species mainly causes renal dysfunction, but sometimes also hemorrhagic fever in humans, while infection by the species discovered in the Americas lead to cardiac and pulmonary dysfunction. A microarray was constructed containing overlapping 500 nucleotide PCR fragments covering the S and M genome segments of a group of hantaviruses (Figure 8.3). Viral RNA was ampliﬁed from cell culture and wild rodents using hantavirus universal primer sets before subsequent ﬂuorescent labeling and hybridization. The results showed a distinction of Puumala virus strains up to 90% similar in nucleic acid sequence identity in parallel with an ability to detect new strains, differing up to 30%, by cross-hybridization.

8.7 Flavivirus Microarray

The ﬂavivirus microarray assay [2] included seven mosquito-borne ﬂaviviruses: West Nile virus, Japanese encephalitis virus, Yellow fever virus and Dengue 1–4 viruses. These are predominantly endemic in tropical and subtropical regions, causing hundreds of millions of cases of disease every year, mainly Dengue virus infections.

j117

j 8 DNA Microarray Technique for Detection and Identification of Viruses

118

Figure 8.3 Hantavirus microarray constructed with 500-nucleotide fragment probes [1]. Closely related strains of Puumala virus could be identified and distinguished.

However, travelers, the animal trade and migrating birds spread virus infections all over the globe. The viruses are single positive-stranded RNA viruses with a high sequence variation between strains, sometimes leading to false-negative RT-PCR results. A microarray-based test was developed, targeting ﬁve genome regions of these ﬂaviviruses. For each of the ﬁve genome positions, 500 nucleotide PCR fragment probes were synthesized for all seven viruses and attached to the microarray surface (Figure 8.4). A multiplex RT-PCR was designed, targeting the same ﬁve positions of all seven viruses for ampliﬁcation of viral RNA from a sample. Five primer pairs were designed with a highly degenerated 30 part targeting the same position in all seven viruses and with an artiﬁcial 50 -tag similar for all 10 primers (Figure 8.5). The virus part targeted the viral RNA in the cDNA and second strand synthesis to produce ﬁve doublestranded DNA complexes having the artiﬁcial tag sequence. The tag could then be targeted for a subsequent ampliﬁcation of the ﬁve fragments using only one primer. This facilitated multiplex ampliﬁcation of the ﬁve different regions in parallel and the use of highly degenerated primers. The method was demonstrated on cell-cultured virus and on clinical samples from Dengue virus infections. A lower limit of detection of about 10 viral genome copies was determined on Dengue 3 virus and overall the performance of the method was comparable to the different routinely used RT-PCR methods. Further, it was shown

8.7 Flavivirus Microarray

Figure 8.4 Design of the flavivirus microarray assay [2]. The amplicons generated from a sample by the highly degenerated multiplex amplification are shown in yellow and the probe fragments attached on the microarray slide surface are shown in red.

Figure 8.5 Strategy for amplification from sample for the flavivirus microarray assay. The tag on the 50 part of the primers is shown in red.

j119

j 8 DNA Microarray Technique for Detection and Identification of Viruses

120

Figure 8.6 Testing of an unknown sample with the flavivirus microarray assay.

that a mixture of Dengue 1 and Dengue 3 viral RNA, about 75% similar in nucleic acid sequence identity, could be resolved in spite of more than 100-fold differences in the quantity of RNA. The method demonstrated its practical usefulness when a sample taken early from a patient with hemorrhagic fever symptoms was tested. Based on the origin of the patient from southwest India, RT-PCRs targeting Crimean-Congo hemorrhagic fever virus and Dengue viruses were selected and proved to be negative. In parallel, the ﬂavivirus microarray test was performed and produced a positive result for Dengue 2 viral RNA (Figure 8.6). A Dengue 2 virus real-time RT-PCR subsequently conﬁrmed the result and samples taken at later clinical stages were serologically positive for Dengue. The reason for the negative result of the ﬁrst Dengue RT-PCR was most probably mutations in the primer sites or a poor lower limit of detection, but this was not further investigated. The ﬂavivirus method is currently under development to improve the capacity to target new strains, to get a better distinction between different West Nile virus strains, and to simplify and obtain a more rapid test. West Nile virus is a typical example of an emerging virus with a dramatic spread to and over a new continent, the Americas, since 1999. In addition, new strains with highly diverged genomic sequences have been discovered during the last decade (e.g. the Rabensburg virus in central Europe). Among the known West Nile virus strains, there is both a signiﬁcant sequence variation, presently suggested for division in ﬁve lineages (Figure 8.7), and a large

8.8 Hemorrhagic Fever Viruses

Figure 8.7 Generalized West Nile virus phylogeny with representative strains. Lineages 3–5 have been recently suggested. Lineage 3 includes Rabensburg virus (RabV) recently discovered in central Europe.

difference in the diseases caused. These two factors are not necessarily connected, but Lineage 1 strains in general have been associated with more severe disease, in particular the strain spread to the Americas. Lineage 1 strains are mainly divided into three clades, one including the Egypt 101 (Eg51) strain associated with milder disease, the next including most of the Lineage 1 strains found in Europe and Russia (Rus99a and It98) associated with somewhat more severe disease, and the last including the high-pathogenic North American strain with closely related strains discovered in Israel, Hungary, Tunisia and Russia (NY99a and Tu97). The vast sequence variation in newly discovered strains in combination with different diseases in closely related strains makes West Nile virus a very suitable model for which to develop a new virus detection and identiﬁcation technique.

8.8 Hemorrhagic Fever Viruses

A microarray assay for a group of hemorrhagic fever viruses is under development. The group includes some hantaviruses and ﬂaviviruses, but also Marburg virus and Ebola viruses, Crimean-Congo hemorrhagic fever virus, Lassa virus and Rift Valley fever virus. In the ﬁrst-generation microarray, 500 nucleotide probes were synthesized from the glycoproteins. These viruses are not closely related, and a random ampliﬁcation protocol was applied and tested successfully for cell-cultured virus (Figure 8.8).

j121

Figure 8.8 First-generation hemorrhagic fever virus microarray assay – preliminary results with cell-cultured virus. Hybridization signals for the Marburg Musoke glycoprotein gene probes (GP1–GP3) are indicated on the left microarray image. Signals for the Sin Nombre glycoprotein gene (M2, M3) and for the nucleocapsid gene (S2) are indicated on the right image.

122

j 8 DNA Microarray Technique for Detection and Identification of Viruses

References

8.9 Conclusions

The microarray technique offers some advantages compared to other nucleic-based virus identiﬁcation methods, both in terms of multiplexing capacity and ability to ﬁnd new strains. The key factors for designing a microarray-based method are the ampliﬁcation and the probe length. These will decide how many different viruses the method should target and the ability to distinguish virus strains as well as to detect new strains. For setting up these methods, standard microarray equipment was used, including hybridization machines for mixing during hybridization and controlled stringency. Further, the development is focused on gradual simpliﬁcation and shortening down of the protocols. The hantavirus project demonstrated the usefulness of long 500mer probes on the microarray for distinction of different viruses and detection of new strains. The ﬂavivirus method was tested and evaluated on Dengue clinical samples, and performed with a lower limit of detection compared with the routinely used RT-PCRs.

References 1 Nordstr€ om, H., Johansson, P., Li, Q.G., Lundkvist, Å., Nilsson, P. and Elgh, F. (2004) Microarray technology for identiﬁcation and distinction of Hantaviruses. J. Med. Virol., 72, 646–655.

2 Nordstr€om, H., Falk, K.I., Lindegren, G., Mouzavi-Jazi, M., Walden, A., Elgh, F., Nilsson, P. and Lundkvist, Å. (2005) DNA microarray technique for detection and identiﬁcation of seven Flaviviruses pathogenic for man. J. Med. Virol., 77, 528–540.

j123

j125

9 Microarrays for Genomotyping of Pathogens Jasper Kieboom, Ingrid Voskamp, and Martien P. Broekhuijsen

9.1 Aim and Approach

Rapid typing of potential biological weapons is of great importance for protection against these agents. The efﬁcacy of medical countermeasures is partly dependent on rapid biotyping and identiﬁcation. By means of molecular tools it is possible to identify bacteria to the subspecies level. However, not all known types can yet be identiﬁed by means of a single rapid molecular method. To overcome this problem, whole-genome microarray experiments were used to identify new genomic markers, in the genomotyping of Francisella and Brucella. Genomotyping basically uses hybridization of genomic DNA of a strain of interest to the microarray, along with the genomic DNA of a reference strain. The latter is typically the strain of which the genome sequence was employed to establish the array. By comparing the ratios of the signal intensities obtained from the spots on the microarray, one can predict whether a given gene is present or divergent in the strain of interest. Here, divergent refers to either absence of a gene or to a gene that has poor hybridization properties. Thus, the genome of one organism is screened in reference to the genome of a chosen reference organism. A number of studies have investigated genome composition using DNA microarrays. Species such as Campylobacter jejuni [1], Helicobacter pylori [2], Staphylococcus aureus [3], Vibrio cholerae [4], and Streptococcus [5] have been examined by the DNA microarray genomotyping technique. One limitation of such an approach is that only absent or divergent genes are identiﬁed in comparison to the reference genome and that neither DNA insertions nor DNA translocations can be detected with this method.

9.2 Francisella Genomotyping

Tularemia is a zoonotic disease affecting a multitude of mammalian species, most notably humans, rabbits, hares and many rodents, including beavers [6]. Humans may contract the disease by direct contact with infected animals, by inhalation of

j 9 Microarrays for Genomotyping of Pathogens

126

infected material, through ingestion of contaminated water or food, or by bites from vectors such as biting ﬂies, mosquitoes or ticks [7]. The genus Francisella contains two known species, F. tularensis and F. philomiragia. Of these, F. philomiragia is an opportunistic pathogen, rarely causing human disease and often associated with water [8]. At present, four subspecies of F. tularensis are recognized, F. tularensis subspecies tularensis (also referred to as type A), F. tularensis subspecies holarctica (type B), F. tularensis subspecies mediaasiatica and F. tularensis subspecies novicida. The four subspecies show marked variations in their virulence and originate from different regions in the world but still display a very close phylogenetic relationship. Moreover, the subspecies of F. tularensis are antigenically similar. The F. tularensis microarray was constructed using 1832 clones from a shotgun library of the highly virulent strain F. tularensis Schu S4 with an estimated genome coverage of more than 95%. After hybridization with DNA from the homologous strain, 98% of the probes showed a clear signal (see Figure 9.1). As a control, 16 probes containing various mouse DNA sequences were included in the microarray. In all experiments, these mouse probes did not hybridize to F. tularensis DNA. Chromosomal DNA from Yersinia pestis showed hybridization with only 10 probes (out of 1832). Twenty-seven strains of F. tularensis that represented each of the four subspecies and that were isolated in different parts of the world over a period of more than 70 years were selected for the study. When DNA from the strains was analyzed with the microarray, spots that showed reduced hybridization signal intensities compared with that for reference F. tularensis strain Schu S4 were detected. The differences in the hybridization patterns detected among strains of the same subspecies were smaller than those detected among strains of different subspecies. Five distinct hybridization patterns were apparent. Overall, the strains clustered according to their subspecies and the clustering supported the present subspecies division. The whole-genome approach revealed some additional information. Interestingly, the subspecies mediaasiatica strains, which comprise strains from the Central Asian republics of the former USSR, and the subspecies tularensis strains, which represent strains from North America, clustered together in the analysis. The subspecies mediaasiatica strains exhibit a moderate degree of virulence for mammals, but most of their biochemical characteristics are common with those of strains of the highly virulent subspecies tularensis [9]. Thus, virulence appears to be the most important criterion for discrimination of these two subspecies. A whole-genome sequence analysis of a subspecies mediaasiatica strain could possibly pinpoint the cause of this difference in virulence. The microarray analysis also revealed that strains of subspecies holarctica showed a greater degree of heterogeneity than strains belonging to the subspecies mediaasiatica and subspecies tularensis. Although subdivisions of the subspecies holarctica have been proposed [10], strains of the subspecies display very similar phenotypic characteristics and there is little evidence supporting this proposal. The present investigation and previous studies based on genetic analyses have demonstrated that Japanese strains show genetic patterns distinct from those of European or American strains of subspecies holarctica [11, 12]. Compared to strain Schu S4, Japanese strains

9.2 Francisella Genomotyping

Figure 9.1 Cladogram of F. tularensis strains using hierarchical cluster analysis on numerical data from the microarray hybridizations.

lacked some spots found in other strains of subspecies holarctica. One speculative reason for this may be that Japanese strains represent an intermediate between the subspecies tularensis or mediasiatica and subspecies holarctica. Since the Japanese strains appear to be less virulent than their European or American counterparts, further studies are warranted to determine if these strains constitute a separate subspecies. Among the 27 F. tularensis strains analyzed, only 0.1–3.7% of the probes showed differential hybridization. Comparing these data with data from intraspecies variation of other bacterial species, as was demonstrated with Pseudomonas aeruginosa, where the variation was in the range of 3% [13], H. pylori strains, where 22% of the open reading frames were absent from at least one strain [2], and S. aureus isolates

j127

j 9 Microarrays for Genomotyping of Pathogens

128

where up to 12% of the open reading frames were absent [3], the microarray analysis of the different subspecies of F. tularensis indicates a high degree of genetic conservation within the species. Since the F. tularensis microarray also includes intergenic regions, a higher degree of genetic variability could be expected compared to that found with an array with genes only. In this context, it should be noted that sequence duplications as well as insertions relative to the sequence of the reference strain might be undetected in any comparative DNA microarray analysis. Moreover, very similar sequences such as homologous genes may cross-hybridize in microarray experiments. However, our results indicate that there are indeed few deletion events and few major sequence differences within the species F. tularensis. This is similar to recent ﬁndings in a microarray study with Mycobacterium tuberculosis – a species considered to be clonal and of recent global dissemination [14, 15]. A possible cause for the genetic homogeneity of F. tularensis could be related to the intracellular lifestyle of the bacterium. According to a recently proposed model, bacterial lifestyle correlates with genomic stability [16]. According to this model, obligate intracellular symbionts show the highest degree of genomic stability; on the other end of the scale, free-living species undergo frequent rearrangement due to continuous ﬂuctuations in their ambient environment [17].

9.3 Brucella Genomotyping

The Brucella family of bacteria was named after Sir David Bruce who, in 1886, ﬁrst isolated the organism from the post-mortem spleen of a soldier who died of a disease then called Malta fever. The genus Brucella currently contains six species: B. abortus, B. melitensis, B. suis, B. canis, B. ovis and B. neotomae, which vary in their ability to infect different hosts. According to the Food and Agriculture Organization of the United Nations/World Health Organization Collaborating Center for Reference and Research on Brucellosis, B. abortus primarily infects cattle, but is transmitted to buffaloes, camels, deer, dogs, horses, sheep and humans. B. melitensis causes a highly contagious disease in sheep, goats and, occasionally, in cattle. B. melitensis is the most important species in terms of potential transmission to humans, causing human brucellosis. B. suis covers a wider host range than the other Brucella species. B. suis has give biovars, of which biovars 1 and 3 primarily infect swine, biovar 2 infects European wild hares, biovar 4 infects reindeer and wild caribou, and biovar 5 was isolated from infected rodents in the USSR. All B. suis biovars can be transmitted to humans with the possible exception of biovar 2. B. canis causes epididymo-orchitis in male dogs, and abortion and metritis in bitches. It has not been reported to infect other animal species than dog, except humans. B. ovis is responsible for epididymitis in rams and occasional abortion in ewes, but it does not infect other animals or humans. Goats were susceptible to the B. ovis caused disease through experimental infection. B. neotomae is only known to infect the desert wood rat under natural conditions and no other cases of infection have been reported to date.

9.3 Brucella Genomotyping

Figure 9.2 False-color overlay of one subgrid (1/8 of the microarray), revealing present, divergent and absent genes. B. melitensis 16M was labeled with Cy5 (red) and the B. suis reference strains were labeled with Cy3 (green).

The microarray used in the Brucella genomotyping study was designed and constructed on the basis of the B. melitensis 16M and B. suis 1330 genome sequence. A total of 3670 DNA oligonucleotides representative for the B. melitensis 16M and the B. suis 1330 genome sequences were selected and synthesized, according to the following criteria: 100% homology, length of 50–55 base pairs and a melting temperature of 79 2 C. An assignment of each oligonucleotide was given using a script in the Biopython scripting language, using batchwise BLAST database matching of all oligonucleotide sequences. A second script was then developed to link the initially retrieved homology information to phenotypical information, such as gene function and gene position of the oligonucleotide. This established a link between general oligonucleotide information from the DNA microarray to the speciﬁc sequence information at gene level. The Brucella microarray was designed in such a way that each slide consisted of triplicate microarrays. Each of these three microarrays contained eight subarrays

j129

j 9 Microarrays for Genomotyping of Pathogens

130

Table 9.1 Total number of absent (ratio cut-off below 3) and present (ratio cut-off above 3) genes (the ratio was determined by comparing the fluorescence intensity of each spot of the tested strains to that of B. melitensis 16M).

B. melitensis 16M B. suis biovar 1 B. suis biovar 2 B. suis biovar 3 B. suis biovar 4

B. melitensis 16M

B. suis biovar 1

B. suis biovar 2

B. suis biovar 3

B. suis biovar 4

Unique oligonucleotides

0 142 65 143 79

142 0 125 109 91

65 125 0 112 72

143 109 112 0 92

79 91 72 92 0

– 47 8 40 7

of 20 24 spots. Although some of these spots were used as blanks, the 3670 oligonucleotides were spotted in triplicate on one microarray slide. The width between spots (resolution) of the single microarrays was 150 mm. In an initial experiment, four different biovars of B. suis were compared to the reference strain B. melitensis 16M (Figure 9.2). Hybridization results were quickly evaluated using Microsoft Excel. For this purpose, raw data underwent normalization of total signal intensity (corrected for background) of each spot to the median total signal intensity. The total number of absent (ratio below 3) and present genes (ratio above 3) was scored as shown in Table 9.1. DNA microarray analysis of these four B. suis biovars revealed the high potential of the designed and implemented genomotyping method for Brucella. In all, 47 unique oligonucleotides have been identiﬁed for B. suis biovar 1, eight for B. suis biovar 2, 40 for B. suis biovar 3 and seven for B. suis biovar 4. In further experiments the applicability of these unique sequences need to be evaluated by testing more Brucella strains.

9.4 Conclusions

Using DNA microarray technology for genomotyping, it is possible to discover new genetic markers for the discrimination of species and subspecies. The approach is feasible when genomic sequence data and representatives of the species are available. The results of such an approach produce new insights in inter- and intra-species relatedness, and allow discovery of new genetic markers that can be used for identiﬁcation methods, like PCR.

Acknowledgments

We greatly acknowledge K. Walravens of the CODA-CERVA culture collection in Belgium for kindly providing the Brucella strains.

References

References 1 Dorrell, N., Mangan, J.A., Laing, K.G., Hinds, J., Linton, D., Al-Ghusein, H., Barrell, B.G., Parkhill, J., Stoker, N.G., Karlyshev, A.V., Butcher, P.D. and Wren, B.W. (2001) Whole genome comparison of Campylobacter jejuni human isolates using a low-cost microarray reveals extensive genetic diversity. Genome Res, 11, 1706–1715. 2 Salama, N., Guillemin, K., McDaniel, T.K., Sherlock, G., Tompkins, L. and Falkow, S. (2000) A whole-genome microarray reveals genetic diversity among Helicobacter pylori strains. Proc Natl Acad Sci USA, 97, 14668–14673. 3 Fitzgerald, J.R., Sturdevant, D.E., Mackie, S.M., Gill, S.R. and Musser, J.M. (2001) Evolutionary genomics of Staphylococcus aureus: insights into the origin of methicillin-resistant strains and the toxic shock syndrome epidemic. Proc Natl Acad Sci USA, 98, 8821–8826. 4 Dziejman, M., Balon, E., Boyd, D., Fraser, CM., Heidelberg, J.F. and Mekalanos, J.J. (2002) Comparative genomic analysis of Vibrio cholerae: genes that correlate with cholera endemic and pandemic disease. Proc Natl Acad Sci USA 99, 1556–1561. 5 Smoot, J.C., Barbian, K.D., Van Gompel, J.J., Smoot, L.M., Chaussee, M.S., Sylva, G.L., Sturdevant, D.E., Ricklefs, S.M., Porcella, S.F., Parkins, L.D., Beres, S.B., Campbell, D.S., Smith, T.M., Zhang, Q., Kapur, V., Daly, J.A., Veasy, L.G. and Musser, J.M. (2002) Genome sequence and comparative microarray analysis of serotype M18 group A Streptococcus strains associated with acute rheumatic fever outbreaks. Proc Natl Acad Sci USA, 99, 4668–4673. 6 Hopla, C.E. (1974) The ecology of tularemia. Adv Vet Sci Comp Med, 18, 25–53. 7 Dennis, D.T., Inglesby, T.V., Henderson, D.A., Bartlett, J.G., Ascher, M.S., Eitzen, E., Fine, A.D., Friedlander, A.M., Hauer, J.,

8

9

10

11

12

13

14

Layton, M., Lillibridge, S.R., McDade, J.E., Osterholm, M.T., OToole, T., Parker, G., Perl, T.M., Russell, P.K. and Tonat, K. (2001) Tularemia as a biological weapon: medical and public health management. J Am Med Ass, 285, 2763–2773. Hollis, D.G., Weaver, R.E., Steigerwalt, A.G., Wenger, J.D., Moss, C.W. and Brenner, D.J. (1989) Francisella philomiragia comb. nov (formerly Yersinia philomiragia) and Francisella tularensis biogroup novicida (formerly Francisella novicida) associated with human disease. J Clin Microbiol, 27, 1601–1608. Olsufjev, N.G. and Meshcheryakova, I.S. (1983) Subspeciﬁc taxonomy of Francisella tularensis. Int J Syst Bacteriol, 33, 872–874. Olsufjev, N.G. (1970) Taxonomy and characteristic of the genus Francisella Dorofeev. J Hyg Epidemiol Microbiol Immunol, 14, 67–74. Ibrahim, A., Gerner-Smidt, P. and Sj€ostedt, A. (1996) Ampliﬁcation and restriction endonuclease digestion of a large fragment of genes coding for rRNA as a rapid method for discrimination of closely related pathogenic bacteria. J Clin Microbiol, 34, 2894–2896. Johansson, A., Ibrahim, A., G€oransson, I., Eriksson, U., Gurycova, D., Clarridge, J.E. 3rd, and Sj€ostedt, A. (2000) Evaluation of PCR-based methods for discrimination of Francisella species and subspecies and development of a speciﬁc PCR that distinguishes the two major subspecies of Francisella tularensis. J Clin Microbiol, 38, 4180–4185. Liang, X., Pham, X.Q., Olson, M.V. and Lory, S. (2001) Identiﬁcation of a genomic island present in the majority of pathogenic isolates of Pseudomonas aeruginosa. J Bacteriol, 183, 843–853. Kato-Maeda, M., Rhee, J.T., Gingeras, T.R., Salamon, H., Drenkow, J., Smittipat, N. and Small, P.M. (2001) Comparing genomes within the species

j131

j 9 Microarrays for Genomotyping of Pathogens

132

Mycobacterium tuberculosis. Genome Res, 11, 547–554. 15 Sreevatsan, S., Pan, X., Stockbauer, K.E., Connell, N.D., Kreiswirth, B.N., Whittam, T.S. and Musser, J.M. (1997) Restricted structural gene polymorphism in the Mycobacterium tuberculosis complex indicates evolutionarily recent global dissemination. Proc Natl Acad Sci USA, 94, 9869–9874. 16 Mira, A., Klasson, L. and Andersson, S. (2002) Microbial genome evolution:

sources of variability. Curr Opin Microbiol, 5, 506. 17 Broekhuijsen, M., Larsson, P., Johansson, A., Bystrom, M., Eriksson, U., Larsson, E., Prior, R.G., Sjostedt, A., Titball, R.W. and Forsman, M. (2003) Genome-wide DNA microarray analysis of Francisella tularensis strains demonstrates extensive genetic conservation within the species but identiﬁes regions that are unique to the highly virulent F. tularensis subsp. tularensis. J Clin Microbiol, 41, 2924–2931.

j133

Part III: Typing

j135

10 Single Nucleotide Polymorphisms as Targets for DNA-Based Identification and Typing of Biosafety Level 3 Bacteria Pierre Wattiau, Pieter Vos, and David Fretin

10.1 Introduction

In the modern microbiological laboratory, there is an increasing need for powerful genetic markers to perform clear-cut identiﬁcation and typing of pathogenic organisms. Moreover, since identiﬁcation of a single marker is usually not sufﬁcient for the accurate diagnostic of a suspect pathogen, modern methods must ideally screen for several markers at once. A marker can be a speciﬁc DNA sequence of a deﬁned length (e.g. a virulence factor typically found in the target pathogen). Alternatively, this can be a variant within a deﬁned gene family (e.g. an allele of a ubiquitous housekeeping gene). The latter, in its simplest interpretation, can be viewed as a short sequence stretch (10–20 base pairs in length) containing at least one polymorphic nucleotide. This chapter will concentrate on the use of such single nucleotide polymorphisms (SNPs) for the identiﬁcation and typing of some Biosafety Level (BSL) 3 bacteria through a number of diagnostic applications. 10.1.1 Real-Time Polymerase Chain Reaction: More Than a Simple Polymerase Chain Reaction

In the conventional polymerase chain reaction (PCR), the ampliﬁed product (amplicon) is detected by an end-point analysis, by running DNA on an agarose gel after the reaction has ﬁnished. In contrast, real-time PCR allows the accumulation of ampliﬁed product to be detected and measured as the reaction progresses (i.e. in real time). Real-time detection of PCR products is made possible by including in the reaction a ﬂuorescent molecule that reports an increase in the amount of DNA with a proportional increase in ﬂuorescent signal. The ﬂuorescent chemistries employed for this purpose include DNA-binding dyes and ﬂuorescently labeled sequencespeciﬁc primers or probes. Specialized thermal cyclers equipped with ﬂuorescence detection modules are used to monitor the ﬂuorescence as ampliﬁcation occurs. The measured ﬂuorescence reﬂects the amount of ampliﬁed product in each cycle.

j 10 Single Nucleotide Polymorphisms as Targets for DNA-Based Identification

136

The main advantage of real-time PCR over conventional PCR is that real-time PCR allows us to determine the starting template copy number with accuracy and high sensitivity over a wide dynamic range. Real-time PCR results can either be qualitative (presence or absence of a sequence) or quantitative (number of copies of DNA). In contrast, conventional PCR is at best semiquantitative. Additionally, real-time PCR data can be evaluated without gel electrophoresis, resulting in reduced experiment time and increased throughput. As reactions are run and data are evaluated in a closed-tube system, opportunities for contamination are reduced and the need for post-ampliﬁcation manipulation is eliminated. Finally, thanks to the discriminating power of some probe types, SNP detection is possible with real-time PCR. Several real-time PCR options are available to type SNPs in a target DNA sequence [1–3]. The most commonly used assays are the 50 -exonuclease (TaqMan; Applied Biosystems; www.appliedbiosystems.com) assay, the hybridization probe (also called the ﬂuorescence energy transfer or FRET assay) and the molecular beacon assay.

10.2 Results of Different SNP Typing Methods 10.2.1 Locked Nucleic Acid-containing TaqMan Probes as SNP Typing Tools

While conventional TaqMan probes usually have little or no SNP discrimination capacity, locked nucleic acid (LNA)-containing TaqMan probes perform fairly well in SNP typing. Compared to conventional nucleotides, LNAs are modiﬁed in such a way that they confer higher afﬁnity to the probes in which they are incorporated (they contain a 20 O, 40 C CH2 bridge in the deoxyribose backbone). Since LNAcontaining probes have a higher Tm, their length can be reduced without affecting the hybridization efﬁciency. When a polymorphic nucleotide is LNA-substituted in a TaqMan probe, its hybridization efﬁciency will be drastically different on a mutant allele compared to a wild-type (100% complementary) allele. Roughly, a DTm of 5–10 C is observed. 10.2.1.1 Use of TaqMan LNA Probes for the Typing of Brucella suis Subspecies Important differences exist within B. suis biovars described to date regarding the risk of animal-to-human transmission and disease severity: while biovar 2 is not pathogenic for humans, the other biovars are considered dangerous pathogens. Fast and accurate B. suis identiﬁcation is thus an important issue for the public health authorities. To identify suitable markers for the biovar typing of B. suis, DNA sequence alignments of B. suis, B. melitensis and B. abortus total genomes were computed using a dedicated algorithm. Potential discriminatory markers were short-listed based on (i) their location in the coding sequence of genes with housekeeping function and (ii) the presence of at least four SNPs within 400-base-pair sequence stretches. Five candidates were kept for further study, PCR-ampliﬁed and sequenced. Reference strains of the ﬁve B. suis biovars described to date were used for initial SNP validation.

10.2 Results of Different SNP Typing Methods Table 10.1 SNPs distribution in two housekeeping genes relevant for B. suis biovar typing.

Gene

Species

Open reading frame or accession no.

ptsP

B. suis B. suis B. suis B. suis B. suis

biovar 1 biovar 2 biovar 3 biovar 4 biovar 5

BR1870 DQ865114 DQ865115 DQ865116 DQ993291

pyrH

B. suis B. suis B. suis B. suis B. suis

biovar 1 biovar 2 biovar 3 biovar 4 biovar 5

BR1160 DQ865121 DQ865122 DQ865123 DQ993293

a)

SNP position

Polymorphic group

1403a)

1512a)

1573a)

1677a)

C C C C C 798a) C C C C C

C C C C C 811a) A A A A A

G G G G G 816 G A A A A

A G A A G 817 T G G G G

I (¼biovar 1) II (¼biovar 2 þ 5) III (¼biovar 3 þ 4) III (¼biovar 3 þ 4) II (¼biovar 2 þ 5)

These positions are polymorphic in the other Brucella species.

Polymorphisms were observed in some genes, including ptsP and pyrH. The observed SNPs deﬁned three allelic groups: B. suis biovar 1, biovar 2/5 and biovar 3/4 (Table 10.1) The biovar-speciﬁc allelism observed on single-type strains was validated by real-time PCR typing as described below. Two real-time PCR assays were set up that could discriminate three SNPs critical for the identiﬁcation of B. suis biovars (ptsP-1677, pyrH-816 and pyrH-817). The probes structure is given at Table 10.2. Although such assays cannot differentiate biovar 2 from biovar 5 nor biovar 3 from biovar 4, they are sufﬁcient for the rapid typing of B. suis strains isolated from Suidae (swine) and hares – biovar 4 and biovar 5 being never found in these animals. Four LNA positions were substituted in these probes. Each real-time PCR assay was performed with 10 ng of template DNA, two PCR primers (0.4 mM each) and two probes (0.1 mM each) matching either ptsP or pyrH in a ﬁnal volume of 50 ml using 2-fold concentrated premixed reagents (BioRad; www.bio-rad.com). Dimethylsulfoxide (10% v/v) was added to improve the probe speciﬁcity and Tris (25 mM, pH 7) was added to optimize the stability of the Cy5 label

Table 10.2 Sequence of TaqMan LNA probes used for B. suis typing by real-time PCR.

Gene

Primer/probe

Nucleotide sequence

ptsP

PTSP-1 PTSP-2 PYRH -1 PYRH -2

Cy5-CACCCGCTGCCTTC-BHQ3 TxRd-CACCCGCCGCCTTC-BHQ3 FAM-CCACAAGGAGTTTCTCGATC-BHQ1 HEX-CCACAAGGAAGTTCTCGATC-BHQ1

pyrH

Underlined nucleotides are LNA-substituted, while bold nucleotides are polymorphic. TxRd, Texas Red; FAM, 6-carboxyﬂuorescein; HEX, hexachloroﬂuorescein.

j137

j 10 Single Nucleotide Polymorphisms as Targets for DNA-Based Identification

138

Figure 10.1 Real-time PCR typing of B. suis. DNA (10 ng) from the type strains B. suis biovar 1 strain 1330 (*), biovar 2 strain Thomsen (&), biovar 3 strain 686 (), biovar 4 strain 40 (~) and biovar 5 strain 513 ( þ ) was assayed by real-time PCR using allele-specific TaqMan LNA probes. Fluorescence signal is given as

a function of the PCR cycle for the amplification of pyrH monitored in optical windows optimized for 6-carboxyfluorescein (A) and hexachlorofluorescein (B), and for ptsP in optical windows optimized for Cy5 (C) and Texas Red (D). No-template controls are shown in plain lines.

in samples assayed with the ptsP probes. Thermal cycling conditions were as follows: 3 min at 95 C followed by 40 cycles of 30 s at 95 C, 30 s at 59 C and 45 s at 72 C. Fluorescence was measured during the extension step at 72 C. As shown in Figure 10.1, each probe proved to hybridize speciﬁcally to the target alleles, showing little or no cross-reactivity with markers differing only by one (ptsP) or two (pyrH) nucleotides. Real-time PCR SNP typing results were reproducible and perfectly matched those obtained by DNA sequencing for all tested strains [4].

10.2 Results of Different SNP Typing Methods

10.2.2 Typing SNPs with Molecular Beacons

Molecular beacons are oligonucleotides coupled to a ﬂuorophore at the 50 -end and to a quencher at the 30 -end. They form a hairpin structure with a stem and a loop designed to hybridize speciﬁcally to a section of the target sequence. On either side of the loop are bases that are complementary to each other, and that form a stem structure that serves to bring the reporter and quencher together. In the hairpin structure, no ﬂuorescence is detected from the reporter due to its physical proximity to the quencher. During the annealing step of the ampliﬁcation reaction, the molecular beacon binds to its target sequence, separating the reporter and quencher such that the reporter is no longer quenched. Unlike TaqMan assays, molecular beacons are displaced but not destroyed during ampliﬁcation. Molecular beacons can be adapted for allelic discrimination experiments. 10.2.2.1 Use of Molecular Beacons for Typing Burkholderia pseudomallei Subspecies Close relatedness and genomic plasticity characterizing the high-threat pathogens B. pseudomallei and B. mallei render the molecular diagnosis of these species hard to guarantee with a maximal conﬁdence level. This section describes fast molecular assays derived from compiled sequences of two housekeeping genes determined in more than 1000 strains. The assays proved to be robust and appropriate for general detection as well as species identiﬁcation purposes. To allow differentiation between B. mallei and B. pseudomallei, polymorphic nucleotides in narK (narK-139) and gltB (gltB-144) were selected. All B. mallei display the same two-locus allelic proﬁle (narK-139A and gltB-144T) while three different speciﬁc proﬁles are displayed by B. pseudomallei. The hallmark of the closely related B. thailandensis and B. oklahomensis is a C at position 285 of narK. Three pairs of molecular beacons were selected to discriminate the critical nucleotides (Table 10.3). The presence of a stable, GC-rich hairpin structure in the DNA ﬂanking nucleotide gltB-144 was overcome by selecting a customized beacon pair whose properties are described elsewhere [5]. Real-time PCR assays were conducted in 50 ml using premixed reagents (BioRad). Final concentrations were 0.1 mM for molecular beacons and 0.4 mM for PCR primers. Each pair of molecular beacons was assayed separately using the same thermal cycling conditions: 3 min at 95 C followed by 45 cycles of 30 s at 95 C, 15 s at 60 C, 30 s at 65 C and 20 s at 72 C. Fluorescence was measured during the temperature step at 65 C. When assayed by real-time PCR, all tested B. pseudomallei, B. mallei and B. thailandensis strains gave ﬂuorescence signals matching the expected two-locus allelic proﬁles. The real-time ﬂuorescence proﬁles were very similar to the one obtained with TaqMan LNA probes shown in Figure 10.1 and are not shown here for space constraints, but can be retrieved from [6]. 10.2.3 Typing Large Number of SNPs by Ligation Detection Reaction

While some authors reported the use of up to 20 primer pairs in a single PCR reaction, real-time PCR can at best identify four or ﬁve different SNPs at once.

j139

BG0075 BG0106

c)

b)

a)

175–296 (285-C)

175–296 (285-T)

130–149 (139-A)

41–57 475–493 130–149 (139-G)

130–152 (144-C) 130–152 (144-T)

442–458

108–128

Coordinatesb)

þ þ

þ

þ

II

þ

þ

I

þ

þ

þ

III

B.pseudomallei c)

þ

þ

þ

IV

B. mallei

þ

þ

þ

B. thailandensis/ oklahomensis V

YaYe, Yakima Yellow; DABCYL, 4-dimethylaminophenylazobenzoic acid; TxRd, Texas Red; FAM, 6-carboxyﬂuorescein; HEX, hexachloroﬂuorescein. Primer coordinates are given according to multilocus sequence typing sequence numbering, discriminatory nucleotides are mentioned into brackets. Three two-locus allelic proﬁles can be found in B. pseudomallei. Most strains (above 95%) display proﬁle I.

BG0194

BG0152

BG0185

TCGCGTGGCGCAATCTC CGAACTGCACGACCGACAC FAM-CAGCTGCCGATCGCGCGGGCTTTCACTT GCAGCTG-DABCYL YaYe-CAGCTGCCGATCGCGCGAGCTTTCACTT GCAGCTG-DABCYL HEX-CGCGATCGCTGCTGATTCCCGCGCTCG GATCGCG-DABCYL FAM-CGCGATCGCTGCTGATCCCCGCGCTCG ATCGCG-DABCYL

YaYe-TAGCCGCTTCGGCGTGACGGCTA-DABCYL TxRd-TAGCCGCTTCGGTGTGACGGCTA-DABCYL

gltB-REV

nark-FWD nark-REV BG0076

ACGTGATCGGCCTTCGC

gltB-FWD

gltB

narK

CTCGAAGATCAAGCAGGTCGC

Primer/ probe name

Gene

Sequence (50 –30 )a)

Table 10.3 Primers, molecular beacons and typing results of B. pseudomallei subspecies.

140

j 10 Single Nucleotide Polymorphisms as Targets for DNA-Based Identification

10.2 Results of Different SNP Typing Methods

Figure 10.2 Principle of the LDR methodology coupled to a low-density DNA microarray as developed by Check-Points. (A) When properly hybridized to a target sequence, the nick lying between two adjacent LDR probe arms is ligated, so that a single circular fragment is generated. (B) Critical mismatches in the target sequence will cause ligation to fail, leaving the probe ends apart. (C) Successful ligation products are amplified by PCR using a single pair of amplimers annealing to complementary sequences

included in the LDR probes (white boxes). (D) Unique ZIP codes (hashed box) assigned to each LDR probe will be specifically captured by complementary oligonucleotides (cZIP codes, reverse-hashed box) spotted on the microarray. (E) Detection occurs thanks to a label incorporated at the 50 -end of one of the PCR primer. The system can be multiplexed with many different LDR probes, each bearing a unique ZIP code (black-filled boxes). The successive reactions are processed in a single tube.

Therefore, if many SNPs need to be screened in a single experiment, efﬁcient multiplexing methods need to be used. The so-called ligation detection reaction (LDR) is a recent method used to generate large collections of integral DNA molecules from adjacent partial fragments that can be either separately synthesized or borne on a single padlock-shaped probe (Figure 10.2). Correct matching of the nucleotides hybridizing to the distal part of the probe is critical for the success of the ligation. The target genetic markers are selected to give little or no intra- and inter-probe interferences. After ligation, just the successfully sealed DNA molecules can be PCR ampliﬁed by means of a single pair of amplimers, whose target sequence is incorporated in every ligation probe [7]. The PCR amplicons are subsequently analyzed, either by capillary electrophoresis (based on the length of the ampliﬁed products) or by hybridization (based on speciﬁc sequence identiﬁers included in the padlock probes). In the latter option, developed commercially by the biotech company Check-Points (www.check-points.com) [8], hybridization can be performed on commercially available, low-density microarray platforms (e.g. ClonDiag, www.clondiag.com; PamGene, www.pamgene.com). Positive

j141

j 10 Single Nucleotide Polymorphisms as Targets for DNA-Based Identification

142

hybridizations are visualized thanks to a label incorporated in one of the PCR primer (Figure 10.2). 10.2.3.1 Use of LDR Probes and Low-Density Microarrays for the Multiplex Diagnostic of BSL3 Bacteria In this example, a single test was tentatively designed to identify any out of 13 different genetic markers at once. Padlock-shaped probes were designed that match (1) the selected genetic markers, (2) the reverse and forward PCR primer sequences, and (3) a unique probe identiﬁer called ZIP code (Figure 10.3). If required, probes were adjusted to the target sequences so that polymorphic nucleotides matched the distal part of the arms. As a result, no ligation can occur if the target sequence is not 100% complementary to the probe.

Figure 10.3 Structure of a typical ligation probe of the padlock type. For successful performance in SNP typing, critical nucleotides in the target sequence should hybridize to the distal part of the arms (i.e. 50 -P and/or 30 OH).

All 13 probes were pooled in a single ligation mix supplied by Check-Points together with premixed kit components. An aliquot of 10 ml of a 1 ng/ml solution of sample DNA was added to 8 ml of a proprietary ligation mix containing all 13 ligation probes together with thermostable ligase and buffer in a 0.2-ml microtube. The sample was heated in a PCR instrument for 3 min at 95 C followed by 24 cycles of 0.5 min at 95 C and 5 min at 65 C followed by a ﬁnal denaturation at 98 C for 2 min. Next, 15 ml of a proprietary exonuclease mix (Check-Points) was added and the sample was incubated for 45 min at 37 C and subsequently for 10 min at 95 C to remove unreacted LDR probes. Next, 15 ml of a proprietary ampliﬁcation mix (Check-Points) containing PCR primers, deoxynucleotide triphosphates, thermostable polymerase and buffer was added and heated for 10 min at 95 C followed by 30 cycles of 0.5 min at 95 C, 0.5 min at 55 C, 1 min at 72 C and a ﬁnal denaturation step of 2 min at 98 C. The ampliﬁed ligation products (ﬁnal volume 48 ml) were then subjected to DNA hybridization in a PamStation DNA microarray platform (PamGene) using custom microarrays spotted with cZIP code oligonucleotides complementary to the ZIP codes included in the LDR probes. Hybridization was performed using 10 ml of each ampliﬁed reaction. Positive hybridizations were visualized by ﬂuorescent detection thanks to a 6-carboxyﬂuorescein-label incorporated in one of the PCR primers. Ten DNA samples, each representing a particular target organism, were assayed as summarized in Table 10.4. Each DNA sample gave a typical DNA microarray proﬁle shown in Figure 10.4. No interference was observed between the various LDR probes and each organism was correctly identiﬁed based on its genetic signatures. A universal 23S rDNA control was included in A5. It matched any bacterial DNA but the one of Coxiella, which diverges substantially from the canonical sequence. Markers

A1

DNA array spot

Burkholderia mallei NCTC10229

Burkholderia pseudomallei NCTC1688

Coxiella burnettii reference strain

Francisella tularensis EQAE9

Brucella suis biovar2

plasmid-cloned SPV

plasmid-cloned GPV

plasmid-cloned SGPV

10.4.c

10.4.d

10.4.e

10.4.f

10.4.g

10.4.h

10.4.i

10.4.j

a)

þ

Bacillus anthracis EQAE2

10.4.b

þ

A4

pla

þ

þ

þ

þ

þ

þ

A5

23S

þ

A6

narK18

þ

þ

B2

sctV

þ

B3

IS1111

þ

B4

fopA

þ

B5

IS711

þ

C0

Vir 2

þ

C1

Vir 3

þ

C3

Vir 4

þ

C4

Vir 5

Genetic markers and their corresponding capture position on the array are indicated in the ﬁrst two rows. Positive hybridizations are indicated in bold. The associated experimental results obtained for each sample are listed in the ﬁrst column. SPV and GPV indicate sheep and goat pox virus, respectively, while SGPV is a hybrid virus.

þ

A2

plcR-313T

Yersinia pestis NCTC8789

DNA source

CAP

Genetic marker

10.4.a

Figure

Table 10.4 Set up of the low-density DNA microarray used to identify 10 different organisms.

10.2 Results of Different SNP Typing Methods

j143

j 10 Single Nucleotide Polymorphisms as Targets for DNA-Based Identification

144

Figure 10.4 Microarray pictures obtained with each of the DNA sample listed in Table 10.4. Positive hybridization controls are found at positions A0, A7 and H6.

plcR-313T and narK-18 were able to discriminate alleles differing by one SNP only. Bacterial genomic DNA (Figure 10.4a–d, f and g), PCR fragments (Figure 10.4e) and plasmid-cloned DNA corresponding to viral genes (Figure 10.4h–j) could be evenly detected with the multiplexed system. Although the sensitivity threshold was not identical for each probe, a minimal amount of about 10 ng DNA was required. Although the system performs very well on pure strains and puriﬁed DNA, its performance on raw biological material and composite samples is not known, and will be investigated in the future.

10.3 Conclusions

The typing of multiple SNPs for the molecular characterization of pathogenic microorganisms is a promising and accurate technique. Depending on the application, the discrimination capacity of multiple SNP typing can be adjusted as to allow differentiation at the genus, species or subspecies (biovar, serovar, etc.) level. The development of powerful multiplexing PCR methods such as LDR, on the one hand, and of user-friendly screening methods such as low-density microarrays, on the other hand, is likely to boost the success of these methods in the future.

Acknowledgments

The authors wish to thank Mieke Van Hessche (Veterinary and Agrochemical Research Center) and Thijs Weyers (Check-Points) for expert technical assistance, and Frank Van den Bussche (Veterinary and Agrochemical Research Center) for the supply of pox virus sequences and samples.

References

References 1 Edwards, K. and Logan, J. (2004) Mutation detection by real-time PCR, In Real-Time PCR: An Essential Guide (eds K. Edwards, J. Logan, and N. Saunders), Horizon Bioscience, London. 2 Mhalanga, M.M. and Malmberg, L. (2001) Using molecular beacons to detect singlenucleotide polymorphisms with real-time PCR. Methods (San Diego, Calif.), 25, 463–471. 3 Tapp, I., Malmberg, L., Rennel, E., Wik, M. and Syvanen, A.C. (2000) Homogeneous scoring of single-nucleotide polymorphisms: comparison of the 50 -nuclease TaqMan assay and Molecular Beacon probes. BioTechniques, 28, 732–738. 4 Fretin, D., Whatmore, A.M., Al Dahouk, S., Neubauer, H., Garin-Bastuji, B., Albert, D., Van Hessche, M., Menart, M., Godfroid, J., Walravens, K. and Wattiau, P. (2008) Brucella suis identiﬁcation and biovar

5

6

7

8

typing by real-time PCR. Vet Microbiol, 131, 376–385. Wattiau, P. and Fretin, D. (2006) Real-time PCR typing of single nucleotide polymorphism in DNA containing inverted repeats. BioTechniques, 41, 544–546. Wattiau, P., Van Hessche, M., Neubauer, H., Zachariah, R., Wernery, U. and Imberechts, H. (2007) Identiﬁcation of Burkholderia pseudomallei and related bacteria by multiplelocus sequence typing-derived PCR and realtime PCR. J Clin Microbiol, 45, 1045–1048. Schouten, J.P., McElgunn, C.J., Waaijer, R., Zwijnenburg, D., Diepvens, F. and Pals, G. (2002) Relative quantiﬁcation of 40 nucleic acid sequences by multiplex ligationdependent probe ampliﬁcation. Nucleic Acids Res, 30, e57. Andreoli, P., Thijssen, J., Anthony, R., Vos, P. and De Levita, W. (2004) Fast method for detecting micro-organisms in food samples. WO2004/10654.

j145

j147

11 Use of a Microchip to Detect Antibiotic Resistance Genes in Bacillus anthracis Vincent Perreten and Joachim Frey

11.1 Introduction

Bacillus anthracis, the etiologic agent of anthrax (a severe zoonotic disease), is a member of the Bacillus cereus group, which further includes the closely related B. cereus and B. thuringiensis [1]. Anthrax commonly occurs in wild and domestic animals, including sheep, cattle, horses, pigs and goats. Animal anthrax is mostly due to spores of B. anthracis that survive in the soil for many decades and favor transmission to cattle when grazing after strong rainfall or when conditions of the soil reach pH above 6.0 and high contents of organic matter. Humans can develop anthrax infection after exposure to B. anthracis via infected animals, animal products such as skin, hair and wool, meat or by direct exposure to B. anthracis [2–4]. Sporadic outbreaks of anthrax in humans have occurred as a result of both agricultural problems and military actions. Due to the long periods where B. anthracis can survive as spores in a non-vegetative state, evolution of B. anthracis seems to be very slow and the genomic structure of B. anthracis is rather stable. Nevertheless, efﬁcient methods based on appropriate markers for different evolutionary scales were developed to subtype strains of B. anthracis and for molecular epidemiological purposes [5–10]. Studies that were carried out to investigate the in vitro susceptibility of B. anthracis to various antibacterial agents revealed variations in the minimal inhibitory concentrations (MICs) due to the different methods used. However, it was generally found that ﬁeld strains of B. anthracis were susceptible to the most commonly used antibiotics [10, 11–16]. Resistance to antibiotics is described in B. cereus and B. thuringiensis. In B. anthracis, resistance can be developed experimentally to most of the current antibacterial agents by selection of spontaneous mutants in serial passage studies. In vitro B. anthracis strains resistant to rifampicin were reported after ultraviolet mutagenesis and found to be due to mutations in the rpoB gene [17]. Fluoroquinolone resistance was introduced in vaccine strain Sterne by sequential subculturing in subinhibitory concentrations of ciproﬂoxacin trovaﬂoxacin or gatiﬂoxacin [18, 19]. Furthermore, strains were made resistant in vitro to doxycycline by transfer of the tetracycline resistance plasmid pBC16, and induced

j 11 Use of a Microchip to Detect Antibiotic Resistance Genes in Bacillus anthracis

148

resistance to the macrolides erythromycin and clarithromycin was reported [18, 20]. The objective of the current study was to obtain information on the mechanisms of antimicrobial resistance and transfer of resistance genes to B. anthracis, and to study the use of microarrays for rapid detection of antibiotic resistance genes in Bacillus anthracis.

11.2 Conjugal Transfer of Antibiotic Resistance Genes Between Enterococcus Species and Avirulent Strains of B. anthracis

Three avirulent strains of B. anthracis were used as recipients for conjugation: B. anthracis 4230 lacking the virulence plasmid pXO1 and containing a deletion of the capsule genes on pXO2 [Dcap::specr (around 7 kb), acpA pXO2 þ ; pXO1] [21], B. anthracis BANT2 lacking the capsule plasmid pXO2 and containing a deletion of the virulence genes on plasmid pXO1 [Dlef-cya::specr (25 kb), pXO1 þ ; pXO2] [22], and B. anthracis RP930 lacking the edema factor gene on plasmid pXO1 and the capsule genes on pXO2 [Dcya::kanr, pXO1 þ ; Dcap::specr (around 7 kb) pXO2 þ ] [21]. Conjugations were performed using Enterococcus faecalis RE25 [23] and Enterococcus sp. RE39 [24] as donor strains, and the three different B. anthracis recipient strains mentioned above. E. faecalis RE25 harbored the 50 kb conjugative plasmid pRE25 which carries ﬁve resistances genes including erm(B), catpIP501, aph(30 )-III, sat4 and ant(60 )-Ia. Enterococcus sp. RE39 harbored an erm(B)-carrying conjugative plasmid (pRE39). The transconjugants were selected on LB agar plates containing trimethaxole and 10 mg of erythromycin/ml. The multidrug-resistant plasmid pRE25 was transferred from E. faecalis RE25 into B. anthracis 4230. However, it could not be conjugated into B. anthracis BANT2 and B. anthracis RP930 probably due to an incompatibility between the enterococcal plasmid pRE25 and the B. anthracis virulence plasmid pXO1. Three transconjugants resulting from conjugation between E. faecalis and B. anthracis 4230 were selected, and named B. anthracis 4251, 4252 and 4253. The plasmid pRE39 was transferred from Enterococcus sp. RE39 into B. anthracis 4230, BANT2 and RP930.

11.3 Determination of the MICs of Different Antibiotics for the B. anthracis Transconjugants

The MICs of different antibiotics were determined for the donor strains E. faecalis RE25 and Enterococcus sp. RE39, the recipient strains B. anthracis 4230 and BANT2, and for the resulting B. anthracis transconjugants by broth microdilution testing (Table 11.1). Both plasmids pRE25 and pRE39 confer resistance to erythromycin and tylosin in B. anthracis. However, pRE25 did not confer resistance to the aminoglycoside antibiotics kanamycin and streptomycin in B. anthracis even in the presence of the aminoglycoside resistance genes aph(30 )-III and ant(60 )-Ia as detected in the

11.4 Detection of Antibiotic Resistance Genes in B. anthracis by Microchip-Based Table 11.1 Susceptibility of Enterococcus sp. and B. anthracis strains and transconjugants to different antibiotics as determined by microdilution testing.

Strains

Erythromycin

Tylosin

Kanamycin

Streptomycin

Enterococcus faecalis RE25 Bacillus anthracis 4230 Bacillus anthracis 4230/pRE25 Enterococcus sp. RE39 Bacillus anthracis 4230 Bacillus anthracis 4230/pRE39 Bacillus anthracis BANT2 Bacillus anthracis BANT2/pRE39 Bacillus anthracis RP930 Bacillus anthracis RP930/pRE39

>128 1 >128 >128 1 >128 1 >128 ND ND

>128 1 64 >128 1 >128 1 >128 ND ND

>128 1 1 ND 1 1 1 1 ND ND

>128 1 1 ND 1 1 1 1 ND ND

ND, not determined.

B. anthracis transconjugants by DNA–DNA hybridization using a microarray (Table 11.1 and Figure 11.1).

11.4 Detection of Antibiotic Resistance Genes in B. anthracis by Microchip-Based Hybridization System (ArrayTube)

A microarray was developed in the framework of this project for the detection of 90 antibiotic resistance genes in Gram-positive bacteria [25]. This array allows rapid identiﬁcation of antibiotic resistance genes using genomic DNA from bacteria. When performed with B. anthracis, the DNA extraction necessary for the labeling procedure presents hazardous critical points such as the centrifugation steps. These critical steps could be avoided by using, for PCR ampliﬁcation, ﬁltered DNA released after a simple cell lysate. The labeling procedures were then appropriately adapted for small amount of DNA. The microarray analysis of B. anthracis 4230 showed the presence of two b-lactamase genes bla1 and bla2, and the spectinomycin resistance gene ant(90 )-I present on plasmid pXO2 instead of the cap genes (Figure 11.1). In the transconjugant strain 4253, the microarray shows the acquisition of the genes sat4, erm(B), aph(30 )-III and ant(60 )-Ia. B. anthracis is susceptible to penicillin although it harbors two b-lactamase genes bla1 and bla2 as conﬁrmed in our analysis. The same phenomenon was observed when the aminoglycoside resistance genes aph(30 )-III and ant(60 )-Ia present on plasmid pRE25 were transferred into B. anthracis by conjugation: the strains remained susceptible to kanamycin and streptomycin (Table 11.1). In a separate study, B. anthracis ﬁeld strains from Chad were found to be susceptible to all major classes of antibiotics as determined by the broth dilution method (MIC) and were screened for the presence of silent genes using microarray

j149

j 11 Use of a Microchip to Detect Antibiotic Resistance Genes in Bacillus anthracis

150

Figure 11.1 Microphotograph of microarrays hybridized with DNA of avirulent B. anthracis [pXO2 Dcap::ant(90 )-Ia; pXO1] before and after transformation with plasmid pRE25. B. anthracis: bla1 and bla2, chromosomal b-lactamase genes of B. anthracis; ant(90 )-Ia, spectinomycin resistance gene used to delete the capsule gene cap on plasmid pXO2 from B. anthracis. Plasmid pRE25, aph(30 )-III,

aminoglycosides resistance gene; erm(B), macrolides, lincosamides and stretogramin B (MLSB) resistance gene; sat4, streptothricin resistance gene; c, biotin position marker. (The Custom ArrayTubes were manufactured by Clondiag Chips Technologies, Jena, Germany, and are distributed by IP Pragmatics Ltd, The London Bioscience Innovation Centre, London, UK).

technology. In these strains, only the b-lactamase genes bla1 and bla2 were present. No other antibiotic resistance genes could be detected. B. anthracis causing animal death in Africa was shown to be free of known transmissible antibiotic resistance genes [10].

11.5 Conclusions

A novel microarray-based technology was tested with recombinant B. anthracis strains for the detection of antibiotic resistance genes. It has been shown that B. anthracis is able to acquire antibiotic resistance genes from other bacteria by conjugation. Field isolates of B. anthracis causing animal death in Africa were shown to be susceptible to the major class of antibiotics and to be free of known transmissible antibiotic resistance genes. The microarray method allows the rapid detection of antibiotic resistance genes in Gram-positive bacteria, thus ﬁlling an important gap and providing a very useful tool for an application in different domains. In particular, in clinical microbiology laboratories, it can be applied to slow-growing bacteria, for which standard antibiotic resistance determinations are difﬁcult, and to rapid antimicrobial testing of highly pathogenic organisms. In particular, the method can detect silent antibiotic resistance genes which might be turned on in vivo or spread to other bacteria.

References

Acknowledgments

This work was supported by grant 4049-067448 of the National Research Programme NRP49 on antibiotic resistance of the Swiss National Science Foundation.

References 1 Helgason, E., Okstad, O.A., Caugant, D.A., Johansen, H.A., Fouet, A., Mock, M., Hegna, I. and Kolsto (2000) Bacillus anthracis, Bacillus cereus, and Bacillus thuringiensis–one species on the basis of genetic evidence. Appl. Environ. Microbiol., 66, 2627–2630. 2 Dixon, T.C., Meselson, M., Guillemin, J. and Hanna, P.C. (1999) Anthrax. N. Engl. J. Med., 341, 815–826. 3 Hanna, P. (1998) Anthrax pathogenesis and host response. Curr. Top. Microbiol. Immunol., 225, 13–35. 4 Mock, M. and Fouet, A. (2001) Anthrax. Annu. Rev. Microbiol., 55, 647–671. 5 Keim, P., Kalif, A., Schupp, J., Hill, K., Travis, S.E., Richmond, K., Adair, D.M., Hugh Jones, M., Kuske, C.R. and Jackson, P. (1997) Molecular evolution and diversity in Bacillus anthracis as detected by ampliﬁed fragment length polymorphism markers. J. Bacteriol., 179, 818–824. 6 Keim, P., Price, L.B., Klevytska, A.M., Smith, K.L., Schupp, J.M., Okinaka, R., Jackson, P.J. and Hugh-Jones, M.E. (2000) Multiple-locus variable-number tandem repeat analysis reveals genetic relation-ships within Bacillus anthracis. J. Bacteriol., 182, 2928–2936. 7 Keim, P., Van Ert, M.N., Pearson, T., Vogler, A.J., Huynh, L.Y. and Wagner, D.M. (2004) Anthrax molecular epidemiology and forensics: using the appropriate marker for different evolutionary scales. Infect. Genet. Evol., 4, 205–213. 8 Keim, P. and Smith, K.L. (2002) Bacillus anthracis evolution and epidemiology. Curr. Top. Microbiol. Immunol., 271, 21–32. 9 Levy, H., Fisher, M., Ariel, N., Altboum, Z. and Kobiler, D. (2005) Identiﬁcation of

10

11

12

13

14

15

strain speciﬁc markers in Bacillus anthracis by random ampliﬁcation of polymorphic DNA. FEMS Microbiol. Lett., 244, 199–205. Maho, A., Rossano, A., H€achler, H., Holzer, A., Schelling, E., Zinsstag, J., Hassane, M.H., Toguebaye, B.S., Akakpo, A.J., Van Ert, M., Keim, P., Keneﬁc, L., Frey, J. and Perreten, V. (2006) Antibiotic susceptibility and molecular diversity of Bacillus anthracis strains in Chad: detection of a new phylogenetic subgroup. J. Clin. Microbiol., 44, 3422–3425. Athamna, A., Massalha, M., Athamna, M., Nura, A., Medlej, B., Ofek, I., Bast, D. and Rubinstein, E. (2004) In vitro susceptibility of Bacillus anthracis to various antibacterial agents and their time-kill activity. J. Antimicrob. Chemother., 53, 247–251. Bakici, M.Z., Elaldi, N., Bakir, M., D€okmetas, I., Erandaç M. and Turan, M. (2002) Antimicrobial susceptibility of Bacillus anthracis in an endemic area. Scand. J. Infect. Dis., 34, 564–566. Bryskier, A. (2002) Bacillus anthracis and antibacterial agents. Clin. Microbiol. Infect., 8, 467–478. Cavallo, J.D., Ramisse, F., Girardet, M., Vaissaire, J., Mock, M. and Hernandez, E. (2002) Antibiotic susceptibilities of 96 isolates of Bacillus anthracis isolated in France between 1994 and 2000. Antimicrob. Agents Chemother., 46, 2307–2309. Frean, J., Klugman, K.P., Arntzen, L. and Bukofzer, S. (2003) Susceptibility of Bacillus anthracis to eleven antimicrobial agents including novel ﬂuoroquinolones and a ketolide. J. Antimicrob. Chemother., 52, 297–299.

j151

j 11 Use of a Microchip to Detect Antibiotic Resistance Genes in Bacillus anthracis

152

16 Jones, M.E., Goguen, J., Critchley, I.A., Draghi, D.C., Karlowsky, J.A., Sahm, D.F., Porschen, R., Patra, G. and DelVecchio, V.G. (2003) Antibiotic susceptibility of isolates of Bacillus anthracis, a bacterial pathogen with the potential to be used in biowarfare. Clin. Microbiol. Infect., 9, 984–986. 17 Vogler, A.J., Busch, J.D., Percy-Fine, S., Tipton-Hunton, C., Smith, K.L. and Keim, P. (2002) Molecular analysis of rifampin resistance in Bacillus anthracis and Bacillus cereus. Antimicrob. Agents Chemother., 46, 511–513. 18 Brook, I., Elliott, T.B., Pryor, H.I., Sautter, T.E., Gnade, B.T., Thakar, J.H. and Knudson, G.B. (2001) In vitro resistance of Bacillus anthracis Sterne to doxycycline, macrolides and quinolones. Int. J. Antimicrob. Agents, 18, 559–562. 19 Grohs, P., Podglajen, I. and Gutmann, L. (2004) Activities of different ﬂuoroquinolones against Bacillus anthracis mutants selected in vitro and harboring topoisomerase mutations. Antimicrob. Agents Chemother., 48, 3024–3027. 20 Pomerantsev, A.P., Shishkova, N.A. and Marinin, L.I. (1992) [Comparison of therapeutic effects of antibiotics of the

21

22

23

24

25

tetracycline group in the treatment of anthrax caused by a strain inheriting tetgene of plasmid pBC16]. Antibiot. Khimioter., 37, 31–34. Fouet, A. and Mock, M. (1996) Differential inﬂuence of the two Bacillus anthracis plasmids on regulation of virulence gene expression. Infect. Immun., 64, 4928–4932. Guignot, J., Mock, M. and Fouet, A. (1997) AtxA activates the transcription of genes harbored by both Bacillus anthracis virulence plasmids. FEMS Microbiol. Lett., 147, 203–207. Schwarz, F.V., Perreten, V. and Teuber, M. (2001) Sequence of the 50-kb conjugative multiresistance plasmid pRE25 from Enterococcus faecalis RE25. Plasmid, 46, 170–187. Teuber, M., Perreten, V. and Wirsching, F. (1996) Antibiotikumresistente Bakterien: eine neue Dimension in der Lebensmittelmikrobiologie. LebensmittelTechnologie, 29, 182–198. Perreten, V., Vorlet-Fawer, L., Slickers, P., Ehricht, R., Kuhnert, P. and Frey, J. (2005) Microarray-based detection of 90 antibiotic resistance genes of Grampositive bacteria. J. Clin. Microbiol., 43, 2291–2302.

j153

Part IV: Quality Control

j155

12 Progress Towards Development of Microarrays for Routine Diagnostic Use Karen Kempsell, Sonal Shah, Susanna Sherwin, Richard Vipond, and Nigel Silman

12.1 Introduction

The Department of Novel and Dangerous Pathogens, in particular the Special Pathogens Reference Laboratory, plays a major role in the Strategic Response capability of the Department of Health in the UK. We are tasked with providing a rapid and robust diagnostic service and to this end are constantly seeking new means by which we can deliver a speciﬁc, accurate and speedy public health resource. This is part of an ongoing core service provision, but can also be called upon to provide the Department of Health with an effective diagnostic capability for responding rapidly to natural and deliberate infectious disease outbreaks (e.g. the foot and mouth epidemic of 2001). In addition to providing vital support services for diagnosis of exotic infectious diseases, we undertake research into novel and developing technologies with a view to their development and validation for implementation into routine diagnostic use. We currently utilize a number of block-based and realtime polymerase chain reaction (PCR) assays for molecular diagnosis of infectious disease in referred clinical samples. We are constantly seeking to reﬁne these procedures and have begun investigating the use of DNA microarrays as an adjunct to our existing technologies. Microarrays have become the focus of a considerable amount of attention in recent years as a means of interrogating complex probes sets in miniaturized format. They have been used for a variety of purposes (e.g. differential gene expression, ChIP-on-Chip analysis); however, their utility for use in diagnostics has lagged somewhat behind these types of studies [2, 30], as there have been some issues with establishing reproducible methodologies. Microarrays in general are still subject to ongoing validation as there have been many concerns about the reproducibility of array data [1, 7, 13, 36] and there is still some debate as to the most sensible means of handling data outputs from microarray systems [6, 11, 31]. Many groups have sought to establish methodologies for quality assessment of microarray platforms [24, 33] and there are still considerable efforts being invested in developing these combined technologies, including robust systems for data management [15],

j 12 Progress Towards Development of Microarrays for Routine Diagnostic Use

156

normalization of microarray outputs and appropriate statistic analyses [18], and in validating different microarray platforms [21, 23]. These quests continue unabated and it is the collective desire of the microarray-users community to delineate sources of variation, implement strategies to minimize these deviations by reﬁning experimental procedures to reduce experimental noise and further increase conﬁdence in data outputs. Lack of reproducibility is one of the reasons that microarrays have been considered to be limited in their utility as a routine diagnostic tool. However, some groups (including our own) have begun investigations into the use of diagnostic microarrays and have made good progress. Wang et al. [35] were the ﬁrst to develop microarrays for the purposes of viral detection and more recently others have conducted similar analyses [12, 17]. Call et al. [10] developed an array for detection of bacterial pathogens in environmental samples and other workers have developed microarrays for differentiation of bacterial species based on small ribosomal subunit RNAs [14, 25–28, 37, 39]. A number of groups (including our own) [8, 9] have developed small lowcomplexity arrays for detection and diagnosis of select organisms based on a combination of probes to both general housekeeping and pathogen-speciﬁc genes [3, 4, 19, 20, 22, 29, 32, 34, 38]. These have been successful in identifying and discriminating pathogen sequences from closely related organisms, and for detection of organisms in environmental and other samples. However, more complex pan-pathogen multiplex microarrays for potential use in diagnosis of infectious disease have always been an extremely attractive concept, due to their amenability to multiplexing and high throughput. The potential application of such arrays in the realm of human health protection is enormous. For example, in the event of presentation of an infected individual with disease symptoms of a diffuse and non-speciﬁc nature, a rapid and accurate diagnosis could be made in the absence of culture or any a priori knowledge of the infectious agent. Samples could be processed identically irrespective of the nature of the infectious agent and testing could be conducted for many pathogens simultaneously. We have recently developed a multiplex pan-pathogen DNA oligonucleotide microarray (manuscript in preparation), containing approximately 2200 oligonucleotides to some 130 different bacterial and viral species and strains of clinical signiﬁcance. The oligonucleotide set has been designed primarily to pathogens of particular threat for use as biowarfare agents; however, other agents of select community-acquired infections have been included (e.g. Neisseria meningitidis). The array consists of a mix of 50 and 70mer oligonucleotides, into which was built a degree of redundancy (i.e. multiple probes per organism). These were designed using a variety of design algorithms and manual methods. The array has been assessed using randomly ampliﬁed ﬂuorescently labeled targets synthesised from puriﬁed pathogen nucleic acids, according to the method of Wang et al. [35]. Our objective was to develop this microarray into a robust technology which could be used for pathogen detection in patient and environmental samples, and would be amenable to high throughput in the event of an adverse incident or event. Minimum requirements for a diagnostic assay array would be speciﬁcity, sensitivity and reproducibility; a desired characteristic would be rapidity, with results ideally

12.2 Materials and Methods

being obtained within 1 day of receipt. This would make it comparable with other existing traditional block-based and real-time PCR tests. However, this technology has not been subject to rigorous validation in a diagnostic context and we wished to investigate the performance of our diagnostic array in correct identiﬁcation of ampliﬁed pathogen nucleic acids. During the course of development of this diagnostic array, we encountered sources of experimental variation that had not been precisely delineated prior to onset of the study. To this end we sought to dissect out these potential sources of error. We have used as the basis of our analysis, parameters set out according to the principles of quality assessment for validation of assays (e.g. precision, intermediate precision, robustness, etc.). Here, we describe a three-way operator study we have conducted using ﬂuorescently labeled targets from two Biosafety Level 3 pathogens, Yersinia pestis and Francisella tularensis. We attempt to quantify and describe some of the sources of experimental variation encountered and propose remedial measures which could be applied to many microarray technology platforms.

12.2 Materials and Methods 12.2.1 Bacterial Strains, Culture and Nucleic Acid Purification

All bacterial strains used were obtained from the culture collection of the Special Pathogens Reference Unit (Heath Protection Agency, Porton Down, UK). Y. pestis Porton was grown on Brain Heart Infusion medium at 37 C with shaking overnight, pathogen genomic DNA from both species was prepared as described by Green et al. [16]. DNA from Francisella tularensis subspecies SHU S4 was provided by one of the authors (R.V.). 12.2.2 Design and Printing of Oligonucleotide Probes

The pan-pathogen array was designed for detection of high-risk pathogens from the Australia Group list (see Table 12.1). The oligonucleotides are a mixture of 50 and 70mers. Genus-speciﬁc probes were designed from alignments of 16S rRNA sequences as described previously [8]. All other oligonucleotide probes were designed from alignments of bacterial species or strain-speciﬁc gene sequences. In short, 50–70mer regions of low sequence conservation between related species and strains were selected which were discriminatory. The speciﬁcity of these probes were conﬁrmed by database probing using search algorithms (e.g. BLAST; http://www. ncbi.nlm.nih.gov/blast). Oligonucleotides probes with greater than 90% sequence similarity with other related bacterial sequences were excluded. The relative melting temperatures (Tm) of each probe were calculated and those with a Tm of 60 8 C were selected. The suitability of selected oligonucleotides for use as discriminatory

j157

j 12 Progress Towards Development of Microarrays for Routine Diagnostic Use

158

Table 12.1 High-risk pathogen species represented on the

pan-pathogen array (multiple species-specific oligonucleotides for each species or strain have been designed where possible, in addition to generic probes). Pathogen

Disease

Bacterial Bacillus anthracis Brucella abortus Brucella melitensis Brucella suis Burkholderia mallei Burkholderia pseudomallei Chlamydia psittaci Clostridium botulinum Coxiella burnetii Francisella tularensis Rickettsia prowaseckii Rochalimaea (Rickettsia) quintana Salmonella typhi Shigella dysenteriae Vibrio cholerae Yersinia pestis

anthrax brucellosis brucellosis swine brucellosis glanders melioidosis psittacosis botulism Q fever tularemia epidemic typhus trench fever typhoid fever dysentery cholera bubonic/pneumonic plague

Viral Chikungunya virus Congo-Crimean hemorrhagic fever virus Dengue fever virus Eastern equine encephalitis virus Ebola virus Guanarito virus Hantaan virus Japanese encephalitis virus Junin virus Lassa fever virus Lymphotropic choriomeningitis virus Machupo virus Marburg virus Monkey pox virus Mopeia Rift Valley fever virus Sabia virus Tick-borne encephalitis virus Variola virus Venezuelan equine encephalitis virus West Nile virus Western equine encephalitis virus White pox Yellow fever virus

Chikungunya fever Congo-Crimean hemorrhagic fever Dengue fever viral encephalitis Ebola hemorrhagic fever Venezuelan hemorrhagic fever Korean hemorrhagic fever Japanese encephalitis Argentine hemorrhagic fever Lassa fever lymphotropic choriomeningitis virus meningitis/encephalitis Bolivian hemorrhagic fever Marburg hemorrhagic fever Monkey pox no human disease Rift Valley fever Brazilian hemorrhagic fever tick-borne encephalitis smallpox variola major Venezuelan equine encephalitis West Nile fever, encephalitis, meningitis Western equine encephalitis pox variola minor Yellow fever

12.2 Materials and Methods

probes with respect to self-complimentarity was assessed using Oligonucleotide Calculator (http://www.basic.northwestern.edu/biotools/oligocalc. html). All in silico validated 50 and 70mer oligonucleotide probes were synthesized by Illumina (www. illumina.com) and printed in quadruplicate onto epoxy-coated Nexterion E slides using a BioRobotics Microgrid II gridder www.genomicsolutions.com. Oligonucleotides were diluted to a ﬁnal concentration of 20 mM in 1.5 mM SSC, 25% dimethylsulfoxide and 0.005% sodium dodecylsulfate (SDS) prior to printing. The slides were then air-dried, baked at 80 C for 2 h and stored dry at ambient temperature in the dark prior to use. 12.2.3 Random Amplification and Cy3-labeling of Nucleic Acid

All hybridizations in this study were conducted using Cy3-labeled randomly ampliﬁed DNA targets from puriﬁed pathogen genomic DNA. The method used to generate randomly ampliﬁed DNA targets is a modiﬁcation of the method published by Bohlander et al. [5] and Wang et al. [35] and has been described previously [8]. 12.2.4 Hybridization of Labeled Targets and Data Processing

Microarray slides were hybridized with target DNAs as published previously [8], with minor modiﬁcations. After pre-hybridization in 5 SSC, 0.1% SDS and 4 Denhardts solution, the slides were washed in sterile, nuclease-free water, 100% propan-2-ol and air-dried. Cy3-labeled randomly ampliﬁed targets DNAs were denatured at 95 C for 3 min, then diluted to a ﬁnal concentration of 80 mg/ml in 5 SSC, 0.1% SDS, 4 Denhardts solution at 50 C. An aliquot of 40 ml of hybridization mix was applied to the microarray slide, using either standard 24 50-mm microscope coverslips or 24 50 mm side-gasket lifterslips (Erie Scientiﬁc; www.eriesci.com). The slide was then incubated in a humidiﬁed multi-slide chamber (Genetix; www.genetix.com) for 16 h and then washed once for 2 min in each of the following buffers: (A) 1 SSC, 0.2% SDS, 50 C, (B) 0.1 SSC, 0.2% SDS, 50 C, (C) 0.1 SSC, 0.2% SDS, 20 C. These were centrifuged to dryness at 1000 rpm for 5 min and scanned using an Affymetrix 428 scanner (www.affymetrix. com). Image ﬁles were saved in Tiff format then quantiﬁed using the scanning software Bluefuse (BlueGnome; www.bluegnome.co.uk). The output ﬁle was processed through a script written in the statistical package R which allowed calculation of the median, mean and inter-quartile range of all four quadruplicate spots. This output was then processed using a macro written in Microsoft Excel, which generated ranked median and mean ﬂuorescence intensities and sorted the data into groups. A cut-off of above 1000 was arbitrarily selected as positive for any given hybridized target. Data were further sorted into disease-relevant groups and analyzed for target–probe hybridization speciﬁcity compared with non-speciﬁc pathogen signals.

j159

j 12 Progress Towards Development of Microarrays for Routine Diagnostic Use

160

12.3 Results 12.3.1 Random Amplification and Hybridization of Cy3-labeled Pathogen DNA Targets

Pathogen DNAs from F. tularensis and Y. pestis were ampliﬁed and labeled with Cy3 using the random ampliﬁcation method; puriﬁed high-molecular-weight target was then hybridized to the array as described above. Results for a F. tularensis Schu S4 hybridization (using standard glass coverslips) are given in Figures 12.1 and 12.2. Many of the F. tularensis-speciﬁc oligonucleotides (highlighted in black; Figure 12.2) hybridize speciﬁcally with F. tularensis Cy3-labeled target. However, some probes bind non-speciﬁcally (highlighted in white; Figure 12.2). In general, most ampliﬁed pathogen targets hybridize with greatest ﬂuorescence intensity to their speciﬁc oligonucleotide probes. However, there remains a certain degree of cross-hybridization to non-speciﬁc probes with all targets. There appear to be three classes of non-speciﬁc probe binding: (i) hybridization to highly conserved genes due to sequence similarity (e.g. 16S rRNA sequences), (ii) hybridization due to poor oligonucleotide synthesis quality issues which can be removed by re-synthesis and high-performance liquid chromatography puriﬁcation, and (iii) hybridization due to inherent oligonucleotide characteristics of unknown origin which cannot be removed by re-synthesis and HPLC puriﬁcation. This was an unexpected ﬁnding and we are currently working on methodologies to minimize these variations. However, there appear to be other sources of experimental error associated with practical aspects of the hybridization technology. One such source is operator variation,

Figure 12.1 Image of the pan-pathogen array hybridized with Cy3-labeled F. tularensis Schu S4 DNA target. The 4 4 subarray configuration is highlighted in green. There are two complete copies of the array printed on the slide, thus e generating four replicates of each oligonucleotide.

12.3 Results

Figure 12.2 Graphic depiction of ranked oligonucleotide fluorescence intensities hybridized with Cy3-labeled F. tularensis Schu S4 DNA target, only the top 50 are shown. Oligonucleotides depicted in black are Francisella species-specific, other nonspecific oligonucleotides are shown in white.

which can clearly be seen in Figure 12.4. For this reason we sought to delineate the sources of experimental variation with a view to improving our methodologies, increasing our conﬁdence in our data output and subsequent identiﬁcation of speciﬁc pathogen hybridization signatures. 12.3.2 Quality Assessment of Intra- and Inter-Operator Variation

To assess some of the sources and extent of experimental variation with the panpathogen microarray, we conducted a series of experiments using quality validation procedures as guidelines (http://www.vam.org.uk/biomeasurement/biomeasurement_quality.asp). The parameters investigated in this study so far are: (i) precision (inherent variation within one operator) and (ii) intermediate precision (inherent variation with more than one operator or on more than 1 day). Three Operators A, B and C synthesized randomly ampliﬁed target DNAs using 10 ng of pathogen DNA from a common stock (see Figure 12.3). Targets were synthesized independently by all three operators for Y. pestis pathogen DNA and hybridized to the pan-pathogen array on 3 non-consecutive days, using standard glass cover-slips. Additionally, Operator A hybridized three slides on 1 day using both coverslips and lifterslips. The ﬂuorescence intensities of all data were compiled, and the mean and standard deviation of the median ﬂuorescence values for all four spot replicates calculated. The data for Yersinia genus and Y. pestis species-speciﬁc oligonucleotide probe sets are shown.

j161

j 12 Progress Towards Development of Microarrays for Routine Diagnostic Use

162

Figure 12.3 Schematic representation of the experimental strategy adopted for quality assessment of the pan-pathogen array.

12.3.2.1 Sources of Intra-Operator Experimental Variation In order to assess precision, Operator A hybridized the same Cy3-labeled Y. pestis target DNA to three slides of the pan-pathogen array on 1 day, using either standard glass microscope coverslips or lifterslips (Figures 12.4 and 12.5). The coefﬁcient of variation (%CV) was calculated for these data and is presented in Figure 12.6. It can be seen that in the main Y. pestis Cy3-labeled target binds preferentially with high intensity to Y. pestis oligonucleotides or those of the closely related Y. pseudotuberculosis (boxed in red). These are found high in the top 50 ranked

Figure 12.4 Graphic depiction of Yersinia species-specific oligonucleotide fluorescence intensities hybridized with Cy3labeled Y. pestis DNA target, using a standard glass coverslip. Oligonucleotides boxed in red are either generic (e.g. Yersinia 1 and 2) or Y. pestis/pseudotuberculosis species-specific.

12.3 Results

Figure 12.5 Graphic depiction of Yersinia species-specific oligonucleotide fluorescence intensities hybridized with Cy3labeled Y. pestis DNA target, using an Erie scientific lifterslip. Oligonucleotides boxed in red are either generic (e.g. Yersinia 1 and 2) or Y. pestis/pseudotuberculosis species-specific.

ﬂuorescence intensity values (data not shown) for all oligonucleotides on the array. This data is of sufﬁcient quality to identify the pathogen proﬁle as Y. pestis. It can also be seen from these results that data generated using coverslips, appears more random between different slides than that generated using lifterslips. This is conﬁrmed by calculation of the %CV, which gives an indication of inter-replicate error (Figure 12.6). The %CV values for lifterslip hybridizations are lower for most

Figure 12.6 Graphic depiction of Yersinia species-specific oligonucleotide fluorescence intensities hybridized with Cy3labeled Y. pestis DNA target, using either standard glass coverslips (&) or lifterslips (&). The %CV is shown for each oligonucleotide for all three slides hybridized in this experiment.

j163

j 12 Progress Towards Development of Microarrays for Routine Diagnostic Use

164

speciﬁc oligonucleotides than those hybridized using coverslips, suggesting less variability between the replicates for the lifterslip data. This implies that lifterslips promote improved speciﬁcity, perhaps by allowing more freedom of movement of the labeled target across the surface of the slide and improving hybridization efﬁciency. 12.3.2.2 Sources of Inter-Operator Experimental Variation In order to assess intermediate precision, Operators A, B and C hybridized their own randomly ampliﬁed targets to three slides of the pan-pathogen array on 3 nonconsecutive days using standard glass microscope coverslips. The data generated for slides from three different operators on three different days generates a pathogen-speciﬁc proﬁle. However, the signal intensity values for each oligonucleotide vary signiﬁcantly from one other, i.e. there is a lot of noise in the data. It was hypothesized that part of this experimental error could be derived from differences in speciﬁc activity of the labeled target [ frequency of incorporation (FOI)]. In order to assess the impact of FOI on comparability of data between different operators, the data were normalized to the FOI of their labeled target (all data are depicted graphically with and without correction for FOI in Figure 12.7). Correcting the data to the FOI removes some data-point variability and makes the oligonucleotide ﬂuorescence intensities more comparable; however, a signiﬁcant amount of residual variability remains. Some of this can be ameliorated through the use of lifterslips as discussed previously; however, there are other sources of variability

Figure 12.7 Graphic representation of Yersinia species-specific oligonucleotide fluorescence intensities hybridized with Cy3-labeled Y. pestis DNA target, using standard glass coverslips. Raw data for median fluorescence values from Yersinia-specific oligonucleotides from three

slides from three operators on 1 day, hybridized with each operators individually synthesized target are shown. Uncorrected values are depicted in pale grey; values corrected to the respective FOI for that target are depicted in dark grey.

References

as yet not fully quantiﬁed. Work is in progress to further dissect and assess the remaining sources of technical variation, with a view to developing improved methodologies.

12.4 Discussion

Microarrays are currently in development for diagnostic purposes by a number of groups including our own. However, they require considerable validation prior to their use as a routine diagnostic tool. We have developed a pan-pathogen array in house for use in detection of a wide variety of mostly high-risk infectious agents. During the course of this study we have encountered various sources of experimental variation when applying quality procedures. Quality assessment guidelines have proved extremely useful in providing a framework for delineating some of these sources of variation. When assessing precision, signiﬁcant ﬂuorescence intensity variation could be seen when one operator applied the same labeled target to three slides hybridized on the same day. This was unexpected and is appears that a source of intra-operator experimental error arises as a consequence of general hybridization methodologies (i.e. in application of labeled target to the slide and during the course of hybridization). Some of this can be ameliorated by the use of lifterslips, suggesting that coverslips do not facilitate adequate mixing and freedom of movement of the target across the surface of the slide, thus signiﬁcantly inﬂuencing target–probe binding. This suggests that mixing efﬁciency is an important parameter in signal speciﬁcity. When assessing intermediate precision using multiple operators it can be seen that there is a signiﬁcant amount of variation between oligonucleotide ﬂuorescence intensities. Some of this experimental noise can be compensated by correction of ﬂuorescence intensity values to target FOI; however, some residual experimental error remains. A proportion of this error can be resolved through improved target mobility, although this does not remove entirely error between replicates. We are currently furthering our studies to delineate these additional parameters of residual variation with a view to overall improvement of our microarray technologies. Our expectation is that this will lead to development of robust routine microarray diagnostic methods that can be used for pathogen detection with conﬁdence.

References 1 Ball, C.A., Sherlock, G., Parkinson, H., Rocca-Sera, P., Brooksbank, C., Causton, H.C., Cavalieri, D., Gaasterland, T., Hingamp, P., Holstege, F., Ringwald, M., Spellman, P., Stoeckert, C.,J. Jr.

Stewart, J.E., Taylor, R., Brazma, A. and Quackenbush, J. (2002) Microarray Gene Expression Data (MGED) Society. Standards for microarray data. Science, 298, 539.

j165

j 12 Progress Towards Development of Microarrays for Routine Diagnostic Use

166

2 Bodrossy, L. and Sessitsch, A. (2004) Oligonucleotide microarrays in microbial diagnostics. Curr Opin Microbiol, 7, 245–254. 3 Bodrossy, L., Stralis-Pavese, N., Konrad-Koszler, M., Weilharter, A., Reichenauer, T.G., Schofer, D. and Sessitsch, A. (2006) mRNA-based parallel detection of active methanotroph populations by use of a diagnostic microarray. Appl Environ Microbiol, 72, 1672–1676. 4 Bodrossy, L., Stralis-Pavese, N., Murrell, J.C., Radajewski, S., Weilharter, A. and Sessitsch, A. (2003) Development and validation of a diagnostic microbial microarray for methanotrophs. Environ Microbiol, 5, 566–582. 5 Bohlander, S.K., Espinosa, R. 3rd., Le Beau, M.M., Rowley, J.D., and Diaz, M.O. (1992) A method for the rapid sequence-independent ampliﬁcation of microdissected chromosomal material. Genomics, 13, 1322–1324. 6 Brazma, A., Hingamp, P., Quackenbush, J., Sherlock, G., Spellman, P., Stoeckert, C., Aach, J., Ansorge, W., Ball, C.A., Causton, H.C., Gaasterland, T., Glenisson, P., Holstege, F.C., Kim, I.F., Markowitz, V., Matese, J.C., Parkinson, H., Robinson, A., Sarkans, U., Schulze-Kremer, S., Stewart, J., Taylor, R., Vilo, J. and Vingron, M. (2001) Minimum information about a microarray experiment (MIAME) – toward standards for microarray data. Nat Genet, 29, 365–371. 7 Burgoon, L.D. (2006) The need for standards, not guidelines, in biological data reporting and sharing. Nat Biotechnol, 24, 1369–1373. 8 Burton, J.E., Oshota, O.J., North, E., Hudson, M.J., Polyanskaya, N., Brehm, J., Lloyd, G. and Silman, N.J. (2005) Development of a multi-pathogen oligonucleotide microarray for detection of Bacillus anthracis. Mol Cell Probes, 19, 349–357. 9 Burton, J.E., Oshota, O.J. and Silman, N.J. (2006) Differential identiﬁcation of

10

11

12

13

14

15 16

17

Bacillus anthracis from environmental Bacillus species using microarray analysis. J Appl Microbiol, 101, 754–763. Call, D.R., Borucki, M.K. and Loge, F.J. (2003) Detection of bacterial pathogens in environmental samples using DNA microarrays. J Microbiol Methods, 53, 235–243. Canales, R.D., Luo, Y., Willey, J.C., Austermiller, B., Barbacioru, C.C., Boysen, C., Hunkapiller, K., Jensen, R.V., Knight, C.R., Lee, K.Y., Ma, Y., Maqsodi, B., Papallo, A., Peters, E.H., Poulter, K., Ruppel, P.L., Samaha, R.R., Shi, L., Yang, W., Zhang, L. and Goodsaid, F.M. (2006) Evaluation of DNA microarray results with quantitative gene expression platforms. Nat Biotechnol, 24, 1115–1122. Chou, C.C., Lee, T.T., Chen, C.H., Hsiao, H.Y., Lin, Y.L., Ho, M.S., Yang, P.C. and Peck, K. (2006) Design of microarray probes for virus identiﬁcation and detection of emerging viruses at the genus level. BMC Bioinformatics, 7, 232. Couzin, J. (2006) Genomics. Microarray data reproduced, but some concerns remain. Science, 13, 1559. Desantis, T.Z., Brodie, E.L., Moberg, J.P., Zubieta, I.X., Piceno, Y.M. and Andersen, G.L. (2007) High-density universal 16S rRNA microarray analysis reveals broader diversity than typical clone library when sampling the environment. Microb Ecol, 53, 371–383. Galbraith, D.W. (2006) The daunting process of MIAME. Nature, 444, 31. Green, B.D., Battisti, L. and Thorne, C.B. (1989) Involvement of Tn4430 in transfer of Bacillus anthracis plasmids mediated by Bacillus thuringiensis plasmid pXO12. J Bacteriol, 171, 104–113. Hsia, C.C., Chizhikov, V.E., Yang, A.X., Selvapandiyan, A., Hewlett, I., Duncan, R., Puri, R.K., Nakhasi, H.L. and Kaplan, G.G. (2007) Microarray multiplex assay for the simultaneous detection and discrimination of hepatitis B, hepatitis C, and human immunodeﬁciency type-1

References

18

19

20

21

22

23

viruses in human blood samples. Biochem Biophys Res Commun, 356, 1017–1023. Klebanov, L., Qiu, X., Welle, S. and Yakovlev, A. (2007) Statistical methods and microarray data. Nat Biotechnol, 25, 25–26. Kostic, T., Weilharter, A., Rubino, S., Delogu, G., Uzzau, S., Rudi, K., Sessitsch, A. and Bodrossy, L. (2007) A microbial diagnostic microarray technique for the sensitive detection and identiﬁcation of pathogenic bacteria in a background of nonpathogens. Anal Biochem, 360, 244–254. Kostic, T., Weilharter, A., Sessitsch, A. and Bodrossy, L. (2005) High-sensitivity, polymerase chain reaction-free detection of microorganisms and their functional genes using 70-mer oligonucleotide diagnostic microarray. Anal Biochem, 346, 333–335. Lee, N.H. (2007) Saeed AI. Microarrays: an overview. Methods Mol Biol, 353, 265–300. Loy, A. and Bodrossy, L. (2006) Highly parallel microbial diagnostics using oligonucleotide microarrays. Clin Chim Acta, 363, 106–119. MAQC Consortium; Shi, L., Reid, L.H., Jones, W.D., Shippy, R., Warrington, J.A., Baker, S.C., Collins, P.J., de Longueville, F., Kawasaki, E.S., Lee, K.Y., Luo, Y., Sun, Y.A., Willey, J.C., Setterquist, R.A., Fischer, G.M., Tong, W., Dragan, Y.P., Dix, D.J., Frueh, F.W., Goodsaid, F.M., Herman, D., Jensen, R.V., Johnson, C.D., Lobenhofer, E.K., Puri, R.K., Schrf, U., Thierry-Mieg, J., Wang, C., Wilson, M., Wolber, P.K., Zhang, L., Amur, S., Bao, W., Barbacioru, C.C., Lucas, A.B., Bertholet, V., Boysen, C., Bromley, B., Brown, D., Brunner, A., Canales, R., Cao, X.M., Cebula, T.A., Chen, J.J., Cheng, J., Chu, T.M., Chudin, E., Corson, J., Corton, J.C., Croner, L.J., Davies, C., Davison, T.S., Delenstarr, G., Deng, X., Dorris, D., Eklund, A.C., Fan, X.H., Fang, H., Fulmer-Smentek, S., Fuscoe, J.C., Gallagher, K., Ge, W., Guo, L., Guo, X., Hager, J., Haje, P.K., Han, J.,

24 25

26

27

28

Han, T., Harbottle, H.C., Harris, S.C., Hatchwell, E., Hauser, C.A., Hester, S., Hong, H., Hurban, P., Jackson, S.A., Ji, H., Knight, C.R., Kuo, W.P., LeClerc, J.E., Levy, S., Li, Q.Z., Liu, C., Liu, Y., Lombardi, M.J., Ma, Y., Magnuson, S.R., Maqsodi, B., McDaniel, T., Mei, N., Myklebost, O., Ning, B., Novoradovskaya, N., Orr, M.S., Osborn, T.W., Papallo, A., Patterson, T.A., Perkins, R.G., Peters, E.H., Peterson, R., Philips, K.L., Pine, P.S., Pusztai, L., Qian, F., Ren, H., Rosen, M., Rosenzweig, B.A., Samaha, R.R., Schena, M., Schroth, G.P., Shchegrova, S., Smith, D.D., Staedtler, F., Su, Z., Sun, H., Szallasi, Z., Tezak, Z., Thierry-Mieg, D., Thompson, K.L., Tikhonova, I., Turpaz, Y., Vallanat, B., Van, C., Walker, S.J., Wang, S.J., Wang, Y., Wolﬁnger, R., Wong, A., Wu, J., Xiao, C., Xie, Q., Xu, J., Yang, W., Zhang, L., Zhong, S., Zong, Y., and Slikker, W. Jr. (2006) The MicroArray Quality Control (MAQC) project shows inter- and intraplatform reproducibility of gene expression measurements. Nat Biotechnol, 24, 1151–1161. Minor, J.M. (2006) Microarray quality control. Methods Enzymol, 411, 233–255. Peplies, J., Glockner, F.O. and Amann, R. (2003) Optimization strategies for DNA microarray-based detection of bacteria with 16S rRNA targeting oligonucleotide probes. Appl Environ Microbiol, 69, 1397–1407. Relman, D.A. (1993) The identiﬁcation of uncultured microbial pathogens. J Infect Dis, 168, 1–8. Reyes-Lopez, M.A., Mendez-Tenorio, A., Maldonado-Rodriguez, R., Doktycz, M.J., Fleming, J.T. and Beattie, K.L. (2003) Fingerprinting of prokaryotic 16S rRNA genes using oligodeoxynucleotide microarrays and virtual hybridization. Nucleic Acids Res, 31, 779–789. Schmidt, T.M. and Relman, D.A. (1994) Phylogenetic identiﬁcation of uncultured pathogens using ribosomal RNA sequences. Methods Enzymol, 235, 205–222.

j167

j 12 Progress Towards Development of Microarrays for Routine Diagnostic Use

168

29 Sessitsch, A., Hackl, E., Wenzl, P., Kilian, A., Kostic, T., Stralis-Pavese, N., Sandjong, B.T. and Bodrossy, L. (2006) Diagnostic microbial microarrays in soil ecology. New Phytol, 171, 719–735. 30 Simon, R. (2003) Using DNA, microarrays for diagnostic and prognostic prediction. Expert Rev Mol Diagn, 3, 587–595. 31 Smith, G.W. and Rosa, G.J. (2007) Interpretation of microarray data: trudging out of the abyss towards elucidation of biological signiﬁcance. J Anim Sci, 85 (13 Suppl): E20–E23. 32 Stralis-Pavese, N., Sessitsch, A., Weilharter, A., Reichenauer, T., Riesing, J., Csontos, J., Murrell, J.C. and Bodrossy, L. (2004) Optimization of diagnostic microarray for application in analysing landﬁll methanotroph communities under different plant covers. Environ Microbiol, 6, 347–363. 33 Uno, K. and Ueda, H.R. (2007) Microarrays: quality control and hybridization protocol. Methods Mol Biol, 362, 225–244. 34 Volokhov, D., Pomerantsev, A., Kivovich, V., Rasooly, A. and Chizhikov, V. (2004) Identiﬁcation of Bacillus anthracis

35

36

37

38

39

by multiprobe microarray hybridization. Diagn Microbiol Infect Dis, 49, 163–171. Wang, D., Coscoy, L., Zylerberg, M., Avila, P.C., Boushey, H.A., Ganem, D. and DeRisi, J.L. (2002) Microarray-based detection and genotyping of viral pathogens. Proc Natl Acad Sci USA, 99, 15687–15692. Wilkes, T., Laux, H. and Foy, C.A. (2007) Microarray data quality – review of current developments. OMICS: A Journal of Integrative Biology, 11, 1–13. Wilson, K.H., Wilson, W.J., Radosevich, J.L., DeSantis, T.Z., Viswanathan, V.S., Kuczmarski, T.A. and Andersen, G.L. (2002) High density microarray of smallsubunit ribosomal DNA probes. Appl Environ Microbiol, 68, 2535–2541. Wilson, W.J., Strout, C.L., DeSantis, T.Z., Stilwell, J.L., Carrano, A.V. and Andersen, G.L. (2002) Sequence-speciﬁc identiﬁcation of 18 pathogenic microorganisms technology. Mol Cell Probes, 16, 119–127. Zimmermann, K., Eiter, T. and Scheiﬂinger, F. (2003) Consecutive analysis of bacterial PCR samples on a single electronic microarray. J Microbiol Methods, 55, 471–474.

j169

Index a accessory genes 19 acrylamide gel layers 72 aldehyde-coated slides 9 aldehyde end-groups 96 aliphatic aldehydes 88 alkene silane-functionalized silicon 77 – substrates 78 alkyl chains 90 amine silanated slides 40 amino-modiﬁed probe molecules, see ODNs amino silane surface chemistry 9 ampliﬁcation methods 11 – GenomiPhi 11 – MessageAmp 11 ampliﬁcation sequences 20 anthrax 59, 147, 158 antibiotic resistance genes 149 antigen-antibody interactions 1 antitoxin antibodies 94 ARB software package 49 array platforms 67 array technologies 80, 85 atomic force microscopy (AFM) 74, 90 – tip 75, 76 avidin molecules 77 – CdSe/ZnS QDs 94

b

Bacillus anthracis 21, 147 bacterial genomes 60, 144 bacterial identiﬁcation system 62 bacterial pathogens 2 bacterial species 156, 157 bead-based arrays 21 biofunctional linker 96 bioinformatics tools 60 bioligands 85, 93

– antibodies 94 – oligodeoxynucleotides 85 biological interactions 93 biological material 144 biological weapons 125 biomolecules 93 – DNA 93 – ODN 93 bioterror weapons 59 biotinlyated ODNs 98

c carboxylic acids 88 CCD camera 15 cDNA fragments 18 cDNA microarray platforms 93 CEL associates 40 cell-cultured virus 118 CGH approaches 19 chip-based devices 90 conductive atomic force microscopy 67, 77, 79, 89 Crimean-Congo hemorrhagic fever virus 120 custom-designed resequencing array 21

d dendritic linkers 98 dengue viruses 120 dideoxynucleotides (ddNTPs) 53 dinucleotide arrays 60 – permutations 61 dip-pen nanolithography 67, 80 – schematic description 75 DNA-based targets 10 DNA-binding dyes 135 DNA aptamers 94 DNA arrays 70

j Index

170

DNA extraction kit/protocol 38, 52 DNA fragments 9, 115 DNA hybridization 74, 142 DNA microarray 1, 5, 18, 107, 125, 129, 155 – analysis 128, 130 – microbial screening 10 – platform 142 – technology 107, 130 DNA probes 15 DNA sequence 20, 126, 135, 141 – alignments 136 double-stranded DNA complexes 118 double-stranded ODN 97 downward pin movement 72 droplet properties 72 – density 72 – surface tension 72 – viscosity 72

g GA aqueous solution 96 GA treatment 96 gene expression 10, 153 – analysis 3 – proﬁling 107 generic arrays 5, 19 genetic markers 143 genetic patterns 126 genomic DNA 125 – sequences 120 Gibbs free energies 5 Glass slide surfaces 89 gold-catalyzed chemical vapor deposition method 79 gold-coated glass slides 90 gold-coated substrates 75 gold nanoparticles 90

h e elastomeric stamp 69 electrical discharge machining (EDM) 72 electrochemical reaction 14 electrophilic glass coatings 88 – epoxysilane 88 – phenylisothiocyanate 88 electrophilic groups 88 electrostatic interactions 9 electrostatic physiosorption/supramolecular interactions 90 environmental samples 113 – bioterrorism-associated 113 enzymatic ampliﬁcation 22 enzyme-assisted hybridization strategies 22 epidemiological markers 2 epoxy-coated substrates 9 Escherichia coli 69 Eurasian species 117

f

F. tularensis strains 127 – cladogram of 127 false-positive signals 44, 47 ﬂavivirus microarray assay 117, 119, 120 ﬂuorescence-based methods 14 ﬂuorescence signals 139 ﬂuorochrome-labeled tyramides 22 – horseradish peroxidase-mediated deposition 22 ﬂuorophore molecules 15

Hantavirus microarray 117, 118 hemorrhagic fever viruses 121 – ﬁrst-generation microarray 122 – symptoms 116, 120 hetero-bifunctional cross-linker 96 high-density microarrays 90 – oligonucleotide-based whole-genome 107 high-performance liquid chromatography puriﬁcation 160 high-throughput techniques 21, 67 – approaches 21 – DNA/protein microarrays 67 – schematic description 69 house-keeping genes 17 – atpD 17 – groEL 17 – gyrB 17 – recA 17 – rpoB 17 hybridization assay format 107 hybridization buffer 42, 54 – parameters 63 hybridization technology 160 hydrophobic barriers 13 hydroxyl groups 89 hydroxyl ions 76

i immobilization technique 94 immobilize tagged bioligands 98 in silico experiments 61 in situ oligonucleotide synthesis 13 ink-jet nozzles 13

Index ink-jet printing 71, 73 – delivery mechanism 73 – process 13 inter-operator experimental variation 164 intracellular symbionts 128

j Japanese strains 127

l large-scale/high-throughput methodologies 67 LCD array experiment 107, 108 LCD-Chip method 112 ligation detection reaction (LDR) 141 – probes 142 lithographic system 74 – techniques 70 locked nucleic acid (LNA) 136, 137 long-chain alkanethiols 97 long vs. short oligonucleotide probes 6 low-cost and low-density (LCD) 107 low-density surface coverage 94 low-quality coatings 9

m Marburg virus 121 marker gene 49 MCH molecules 97 metallic/organometallic precursors 93 – cadmium 93 – mercury 93 – zinc 93 micro-spotting 72 – pins 74 – robotic system 72 – technology 71, 72 micro-spotting/ink-jet printing 71 microarray assay 22, 121 microarray-based test 113, 116, 118 – analysis 126 – applications 15 – design 20 – detection sensitivity 14 – experiments 128 – hybridizations 127 – images 45, 50 – platforms 3, 107 – probes 1, 116 – protocols 22 – slides 159 – system 20, 155 microarray quality control (MAQC) consortium 17

microarray technology 20, 22, 113, 115 – principle 113 microbial diagnostic microarrays (MDMs) 47 microcontact printing 68 microelectronics industry 90 Microsoft Excel 130 – template 114 microsphere-based ﬁber-optic arrays 16 minimal inhibitory concentrations 147 MIP technology 20 modulator spacer molecules 96 molecular barcodes 20 molecular beacons 139 molecular epidemiological assessments 59 molecular inversion probe (MIP) technology 20 motion-control print-head 71 mRNA-based targets 10 multiplex assay 112 Mycobacterium tuberculosis 128

n

N-alkane thiol molecules 75 nanoarray technologies 67, 79 NanoDrop spectrophotometer 52 nanometric protein arrays 77 nanoscale imaging techniques 74 non-cognate hybridization system 19, 60 – approach 61 – based microarrays 62, 64 – design 61 – utilization 64 non-contact printing approach 70 non-speciﬁc interactions 91 non-targeted sequence 40 novel techniques 80 nucleic acid 11, 114, 115, 116 – ampliﬁcation methods 2, 114, 116 – derivatives 10 – sequence 114

o oligodeoxynucleotides (ODNs) 96, 97 – arrays 5, 17 – probes 71, 96 – signals 17 – single-stranded 97 oligonucleotide primer 20 oligonucleotide probes 3, 10, 12, 13, 37, 40, 51, 116, 157 – advantages 37

j171

j Index

172

– characteristics 37 optical biosensors 90 optical techniques 14 optimal probe length 60 – deﬁnition 60

p Padlock-shaped probes 142 pathogenic bacteria 55 patterning techniques 65, 67 photolithography 70, 80 photosensitive nucleotides 70 phylogenetic marker 16 piezoelectric crystal tubes 74 piezoelectric dispensers 72 piezoelectric ink-jet printers 13 planar-waveguide technology (PWT) 14 polymer-coated particles 85 polymer-coated surfaces 91 polymerase chain reaction (PCR) 2, 107, 135, 155 – ampliﬁcation 10, 16, 52, 141 – assays 59, 137 – fragment probes 118 – instrument 142 – LDR 144 – methods 144 – primers 17 – protocols 108 – puriﬁcation kit 52 – reaction 52, 108 – real-time detection 135 polymeric ﬁlms 85 polymeric materials 91 polymeric membranes/ﬁlms 91 – cellulose 91 – nylon 91 polymeric substrates 91 polystyrene microspheres 15 pore diffusion resistance 91 porous polymeric materials 91 positive-stranded RNA 118 positive hybridizations 143 probe density 13 probe design 5, 49 probe immobilization reactions 88 probe length 49 probe molecules, see bioligands problematic probes 51 prokaryotic RNAs 10 proprietary surface chemistry 9 protein arrays 76 – probes 94 Pseudomonas aeruginosa 127

pulmonary dysfunction 117 PWT technology-based microarrays 14, 15 – schematic representation 15

q QD bioconjugates 93 QD nanoprobes 94 qPCR-based techniques 59

r rapid immobilization reactions 94 RC oligonucleotide 47 Renibacterium salmoninarum 69 resonance-light scattering (RLS) 14 – principle 14 reverse transcription-polymerase chain reaction (RT-PCR) 115 Rift Valley fever virus 121 RNA sequences 11 RNA viruses 113, 115 robotic printing 71 rRNA fragments 11

s Salmonella-speciﬁc probes 44 salt concentrations 11 sandwich immunoassay approach 94 scanning probe techniques 76 Schiff base linkages 88 self-assembled monolayers (SAMs) 78, 90 – linkers 97 self-organizing maps (SOM) 64 – structure 64 semiconductor industry 13 semiconductor material 93 sequence-speciﬁc end-labeling of oligonucleotides (SSELO) 47 – approach 47, 49 – probes 47 shotgun libraries 3 shrimp alkaline phosphatase (SAP) 53 – treatment 53 signal-to-noise ratio 15, 90 signal ampliﬁcation 22, 114, 116 silicon nanowires 79 silicon wafers 72 – technology 90 single-labeled dideoxynucleotide 12 single nucleotide polymorphisms 135 sodium borohydride 88 soft lithography 68, 69, 80

Index solid-state nucleic acid sequencing 5 Streptomyces avidinii 98 substrate surfaces 95 – advantages 95 – disadvantages 95 surface plasmon resonance 90 – slides 75 suspension-based array technologies 85

t TaqMan probes 136 target-speciﬁc QD nanoprobes 94 TCHS molecules 77 tetramethylammonium chloride 11 thiol end-groups 91 – ODN probe molecules 97 thiolated hydrocarbon chain 96 three-dimensional microarray systems 16 – gel-pads 16 TOPO molecules 93 transcript sequences 20 two-probe proximal chaperon detection system 11

u UV-exposed areas 70

v viral nucleic acid 114 virus-speciﬁc DNA probes 114

w Wash protocol 43 wave-guiding ﬁlm 14 weighted mismatch (wMM) values 51 well-controlled kinetics 94 well-designed probes 5 whole-genome ampliﬁcation methods 10 – approach 126 – expression analysis 2 whole-genome microarray experiments 125 whole-genome sequencing 18

y

Yersinia species-speciﬁc oligonucleotide ﬂuorescence intensities 164

z ZIP codes 142 zoonotic disease, see anthrax

j173