METHODS
IN
M O L E C U L A R B I O L O G Y TM
Series Editor John M. Walker School of Life Sciences University of Her...
45 downloads
564 Views
8MB 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
METHODS
IN
M O L E C U L A R B I O L O G Y TM
Series Editor John M. Walker School of Life Sciences University of Hertfordshire Hatfield, Hertfordshire, AL10 9AB, UK
For other titles published in this series, go to www.springer.com/series/7651
Single Nucleotide Polymorphisms Methods and Protocols Second Edition Edited by
Anton A. Komar Center for Gene Regulation in Health and Disease, Department of Biological, Geological, and Environmental Sciences, Cleveland State University, Cleveland, OH, USA
Editor Anton A. Komar Center for Gene Regulation in Health and Disease, Department of Biological, Geological and Environmental Sciences, Cleveland State University Cleveland, OH 44115 USA
ISSN 1064-3745 e-ISSN 1940-6029 ISBN 978-1-60327-410-4 e-ISBN 978-1-60327-411-1 DOI 10.1007/978-1-60327-411-1 Library of Congress Control Number: 2009933280 # Humana Press, a part of Springer ScienceþBusiness Media, LLC 2003, 2009 All rights reserved. This work may not be translated or copied in whole or in part without the written permission of the publisher (Humana Press, c/o Springer ScienceþBusiness Media, LLC, 233 Spring Street, New York, NY 10013, USA), except for brief excerpts in connection with reviews or scholarly analysis. Use in connection with any form of information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed is forbidden. The use in this publication of trade names, trademarks, service marks, and similar terms, even if they are not identified as such, is not to be taken as an expression of opinion as to whether or not they are subject to proprietary rights. While the advice and information in this book are believed to be true and accurate at the date of going to press, neither the authors nor the editors nor the publisher can accept any legal responsibility for any errors or omissions that may be made. The publisher makes no warranty, express or implied, with respect to the material contained herein. Cover illustration: Chapter 14, Figure 1 Printed on acid-free paper springer.com
Preface One of the major challenges in modern molecular biology is to understand how phenotypic differences between species or individual representatives of the same species are encoded in their genomes. Single nucleotide polymorphisms (SNPs) are known to be the major contributors to the genetic variations. They comprise more than 80% of all known polymorphisms and are assumed to be primarily responsible for phenotypic differences between individuals of the same species. SNPs were also suggested to affect the likelihood of the development and progression of many diseases in humans as well as to determine the response of different individuals to drug treatment and/or environmental stress. In addition, SNPs can serve as valuable biological markers. Thus, genetic tests and methods allowing for rapid and accurate determination of a defined SNP, or a set of SNPs and/or of a complete individual’s SNP profile, as well as methods allowing for complete and rapid sequencing of one’s individual genome are considered of immense importance in personalized medicine and drug treatment. With reduction of the cost of genotyping, personal SNP profiles might be extremely helpful (in a not so distant future) in determining one’s best living and working conditions. However, much work has yet to be done in this direction before we will have a complete understanding of how exactly each and every particular SNP might (or might not) affect the health of an individual and/or her/his reaction to a drug. The second edition of Single Nucleotide Polymorphisms in the Methods in Molecular BiologyTM series further aims to provide an overview of a variety of techniques currently used for SNP detection and analysis. Since the first edition of this book (published in 2003), substantial progress has been made in increasing the accuracy, efficiency, and automation of SNP genotyping, the development of high-throughput genotyping approaches, as well as understanding of the impact of SNPs on gene function. Consequently, the total number of chapters has increased from 17 (in the first edition) to 28 (in the second one). It is becoming apparent that SNPs occurring in ‘‘junk DNA’’ as well as silent SNPs, previously assumed to be neutral, can have effects on gene function. Therefore, a separate chapter (Chapter 2), describing the effects of silent (synonymous) SNPs, has been added to the current edition. With the availability of millions of SNPs found in genomes, resources containing sequence and mapping data, search, browsing, and retrieval systems are also becoming extremely valuable. A novel chapter (Chapter 3) provides a comprehensive overview of SNP databases. Tools and strategies are outlined in this chapter that can help researchers to obtain the most appropriate information needed for their research aims. In general, the first part of the book aims to address the fundamental aspects of the impact of SNPs on gene function and phenotype (Chapters 1 and 2). SNP databases and methods applied for SNP bioinformatics discovery and analysis are discussed further (Chapters 3 and 4, respectively). The second part of the book is primarily devoted to the advanced experimental approaches used for SNP detection (Chapters 5–28). SNP genotyping usually involves the generation of a specific DNA product of a selected region of the genome, followed by direct (e.g., by sequencing) v
vi
Preface
or indirect SNP detection. Thus, the vast majority of the methods involve a polymerase chain reaction (PCR) amplification step of the targeted genome fragment. However, alternative approaches, such as isothermal smart amplification process (Chapter 28), as well as strategies including advanced (next-generation) whole genome sequencing methods (Chapters 5 and 6) are also addressed and discussed. I have tried to group various methods according to the base principles they utilize for SNP detection, as well as their throughput (and plexing) capabilities. Therefore, prescreening (melting- and conformation based) approaches are described first (Chapters 8–15), followed by the high-throughput applications (Chapters 16–24). Very simple methods, such as PCR–restriction fragment length polymorphism (Chapter 25), that require minimum equipment and resources are also discussed. Advances in modern technology allow the rapid development of many new techniques aimed at SNP detection and the improvement of the existing ones. It is almost apparent that by the time this volume will be published, many new applications will be available to researches; however, I hope that the second edition of this book together with the first one will further serve as a valuable source of information for individual researchers as well as institutions and companies working in the field. I am indebted to all the contributors, who kindly agreed to share their knowledge of SNP detection and identification strategies and without whom the second edition of this book would not be possible. Anton A. Komar
Contents Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
v xi
SECTION I: INTRODUCTION TO SINGLE NUCLEOTIDE POLYMORPHISMS 1.
SNPs: Impact on Gene Function and Phenotype . . . . . . . . . . . . . . . . . . . . . . . . . . Barkur S. Shastry
3
2.
Silent (Synonymous) SNPs: Should We Care About Them? . . . . . . . . . . . . . . . . . . Ryan Hunt, Zuben E. Sauna, Suresh V. Ambudkar, Michael M. Gottesman, and Chava Kimchi-Sarfaty
23
SECTION II: BIOINFORMATIC ANALYSIS OF SNPS 3.
SNP Databases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Christopher Phillips
4.
Mining SNPs from DNA Sequence Data; Computational Approaches to SNP Discovery and Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Jan van Oeveren and Antoine Janssen
43
73
SECTION III: SNP IDENTIFICATION AND DETECTION STRATEGIES: WHOLE GENOME SEQUENCING AND RE-SEQUENCING 5.
6.
7.
Next-Generation Sequencing Methods: Impact of Sequencing Accuracy on SNP Discovery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95 Eugene Y. Chan Scanning Probe and Nanopore DNA Sequencing: Core Techniques and Possibilities. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113 John Lund and Babak A. Parviz Pyrosequencing for SNP Genotyping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123 Jose Luis Royo and Jose Jorge Gala´n
SECTION IV: PRESCREENING (MELTING BASED) METHODS FOR SNP DISCOVERY AND ANALYSIS 8.
9.
Single Nucleotide Polymorphism Screening with Denaturing Gradient Gel Electrophoresis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137 Leslie A. Knapp Temporal Temperature Gradient Electrophoresis for Detection of Single Nucleotide Polymorphisms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153 Bethan M. Jones and Leslie A. Knapp
SECTION V: PRESCREENING (CONFORMATION BASED) METHODS FOR SNP DISCOVERY AND ANALYSIS 10. Zn(II)–Cyclen Polyacrylamide Gel Electrophoresis for SNP Detection . . . . . . . . . 169 Emiko Kinoshita-Kikuta, Eiji Kinoshita, and Tohru Koike
vii
viii
Contents
11. Phosphate-Affinity Polyacrylamide Gel Electrophoresis for SNP Genotyping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183 Eiji Kinoshita, Emiko Kinoshita-Kikuta, and Tohru Koike 12. Estimation of SNP Allele Frequencies by SSCP Analysis of Pooled DNA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 193 Tomoko Tahira, Yoji Kukita, Koichiro Higasa, Yuko Okazaki, Aki Yoshinaga, and Kenshi Hayashi 13. Phenylethynylpyrene Excimer Forming Hybridization Probes for Fluorescence SNP Detection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 209 Igor A. Prokhorenko, Irina V. Astakhova, Kuvat T. Momynaliev, Timofei S. Zatsepin, and Vladimir A. Korshun 14. The Chemical Cleavage of Mismatch for the Detection of Mutations in Long DNA Fragments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 223 Tania Tabone, Georgina Sallmann, and Richard G. H. Cotton 15. Mismatch Oxidation Assay: Detection of DNA Mutations Using a Standard UV/Vis Microplate Reader. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 235 Tania Tabone, Georgina Sallmann, and Richard G. H. Cotton
SECTION VI: TOWARDS HIGH-THROUGHPUT METHODS OF SNP GENOTYPING 16. High-Throughput Methods for SNP Genotyping . . . . . . . . . . . . . . . . . . . . . . . . . 245 Chunming Ding and Shengnan Jin 17. High-Throughput SNP Genotyping: Combining Tag SNPs and Molecular Beacons . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 255 Luis B. Barreiro, Ricardo Henriques, and Musa M. Mhlanga 18. SNP Genotyping by the 50 -Nuclease Reaction: Advances in High-Throughput Genotyping with Nonmodel Organisms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 277 James E. Seeb, Carita E. Pascal, Ramesh Ramakrishnan, and Lisa W. Seeb 19. The TaqMan Method for SNP Genotyping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 293 Gong-Qing Shen, Kalil G. Abdullah, and Qing Kenneth Wang 20. Qualitative and Quantitative Genotyping Using Single Base Primer Extension Coupled with Matrix-Assisted Laser Desorption/Ionization Time-of-Flight Mass Spectrometry (MassARRAY1). . . . . . . . . . . . . . . . . . . . . . . . 307 Paul Oeth, Guy del Mistro, George Marnellos, Tao Shi, and Dirk van den Boom 21. SNP Detection Using Trityl Mass Tags . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 345 Klara R. Birikh, Pablo L. Bernad, Vadim V. Shmanai, Andrei D. Malakhov, Mikhail S. Shchepinov, and Vladimir A. Korshun 22. Putting the Invader1 Assay to Work: Laboratory Application and Data Management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 363 Yi Zhang, Edward Smith, and Michael Olivier 23. SNP Genotyping Using Multiplex Single Base Primer Extension Assays . . . . . . . . 379 Daniele Podini and Peter M. Vallone 24. High-Throughput SNP Detection Based on PCR Amplification on Magnetic Nanoparticles Using Dual-Color Hybridization . . . . . . . . . . . . . . . . . . . . . . . . . . . 393 Nongyue He, Song Li, and Hongna Liu
Contents
ix
SECTION VII: OTHER METHODS 25. Restriction Enzyme Analysis of PCR Products . . . . . . . . . . . . . . . . . . . . . . . . . . . . 405 Masao Ota, Hideki Asamura, Takahito Oki, and Masaharu Sada 26. Allele-Specific PCR in SNP Genotyping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 415 Muriel Gaudet, Anna-Giulia Fara, Isacco Beritognolo, and Maurizio Sabatti 27. Modified Multiple Primer Extension Method. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 425 Toshihide Yanagawa and Hisashi Koga 28. Detection of SNP by the Isothermal Smart Amplification Method . . . . . . . . . . . . . 437 Alexander Lezhava and Yoshihide Hayashizaki Subject Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 453
Contributors KALIL G. ABDULLAH • Department of Molecular Cardiology, Lerner Research Institute, Cleveland Clinic, Cleveland, OH, USA SURESH V. AMBUDKAR • Laboratory of Cell Biology, National Cancer Institute, National Institutes of Health, Bethesda, Maryland, USA HIDEKI ASAMURA • Department of Legal Medicine, Shinshu University School of Medicine, Asahi, Matsumoto, Japan IRINA V. ASTAKHOVA • Shemyakin-Ovchinnikov Institute of Bioorganic Chemistry, Miklukho-Maklaya, Moscow, Russia LUIS B. BARREIRO • Department of Human Genetics, The University of Chicago, Chicago, IL, USA ISACCO BERITOGNOLO • Department of Forest Resources and Environment (DiSAFRi), University of Tuscia, Viterbo, Italy PABLO L. BERNAD • Tridend, Division of Oxford Gene Technology Ltd, Yarnton, Oxford, UK KLARA R. BIRIKH • Shemyakin-Ovchinnikov Institute of Bioorganic Chemistry, MiklukhoMaklaya, Moscow, Russia EUGENE Y. CHAN • The DNA Medicine Institute, Boston, MA, USA RICHARD G. H. COTTON • Genomic Disorders Research Centre, Carlton South; The University of Melbourne, Department of Medicine, Melbourne, Victoria, Australia CHUNMING DING • Stanley Ho Centre for Emerging Infectious Diseases and Li Ka Shing Institute of Health Sciences, The Chinese University of Hong Kong, Prince of Wales Hospital, Shatin, New Territories, Hong Kong Special Administrative Region, China ANNA-GIULIA FARA • Department of Forest Resources and Environment (DiSAFRi), University of Tuscia, Viterbo, Italy JOSE JORGE GALA´N • Department of Structural Genomics, Neocodex SL. Avda. Charles Darwin, Seville, Spain MURIEL GAUDET • Department of Forest Resources and Environment (DiSAFRi), University of Tuscia, Viterbo, Italy MICHAEL M. GOTTESMAN • Laboratory of Cell Biology, National Cancer Institute, National Institutes of Health, Bethesda, Maryland, USA KENSHI HAYASHI • Division of Genome Analysis, Research Center for Genetic Information, Medical Institute of Bioregulation, Kyushu University, Higashi-ku, Fukuoka, Fukuoka, Japan YOSHIHIDE HAYASHIZAKI • Genome Exploration Research Group, Genomic Sciences Center (GSC), RIKEN Yokohama Institute, Tsurumi-ku, Yokohama, Kanagawa, Japan NONGYUE HE • State Key Laboratory of Bioelectronics, School of Biological Science and Medical Engineering, Southeast University, Nanjing, China RICARDO HENRIQUES • Gene Expression and Biophysics Unit, Institute for Molecular Medicine, Faculty of Medicine of the University of Lisbon, Lisbon, Portugal xi
xii
Contributors
KOICHIRO HIGASA • Division of Genome Analysis, Research Center for Genetic Information, Medical Institute of Bioregulation, Kyushu University, Higashi-ku, Fukuoka, Fukuoka, Japan RYAN HUNT • Laboratory of Hemostasis, Division of Hematology, Center for Biologics Evaluation and Research, Food and Drug Administration, Bethesda, Maryland, USA ANTOINE JANSSEN • Division of Bioinformatics, Keygene NV, AE Wageningen, The Netherlands SHENGNAN JIN • Institute of Digestive Diseases, The Chinese University of Hong Kong, Prince of Wales Hospital, Shatin, New Territories, Hong Kong Special Administrative Region, China BETHAN M. JONES • National Oceanography Centre, School of Ocean and Earth Science, University of Southampton, Southampton, UK CHAVA KIMCHI-SARFATY • Laboratory of Hemostasis, Division of Hematology, Center for Biologics Evaluation and Research, Food and Drug Administration, Bethesda, Maryland, USA EIJI KINOSHITA • Department of Functional Molecular Science, Graduate School of Biomedical Sciences, Hiroshima University, Hiroshima, Japan EMIKO KINOSHITA-KIKUTA • Department of Functional Molecular Science, Graduate School of Biomedical Sciences, Hiroshima University, Hiroshima, Japan LESLIE A. KNAPP • Primate Immunogenetics and Molecular Ecology Research Group, Department of Biological Anthropology, University of Cambridge, Cambridge, UK HISASHI KOGA • Laboratory of Medical Genomics, Department of Human Genome Technology, Kazusa DNA Research Institute, Kazusa-Kamatari, Kisarazu, Chiba, Japan TOHRU KOIKE • Department of Functional Molecular Science, Graduate School of Biomedical Sciences, Hiroshima University, Hiroshima, Japan VLADIMIR A. KORSHUN • Shemyakin-Ovchinnikov Institute of Bioorganic Chemistry, Miklukho-Maklaya, Moscow, Russia YOJI KUKITA • Division of Genome Analysis, Research Center for Genetic Information, Medical Institute of Bioregulation, Kyushu University, Higashi-ku, Fukuoka, Fukuoka, Japan ALEXANDER LEZHAVA • Genome Exploration Research Group, Genomic Sciences Center (GSC), RIKEN Yokohama Institute, Tsurumi-ku, Yokohama, Kanagawa, Japan SONG LI • State Key Laboratory of Bioelectronics, School of Biological Science and Medical Engineering, Southeast University, Nanjing, China HONGNA LIU • State Key Laboratory of Bioelectronics, School of Biological Science and Medical Engineering, Southeast University, Nanjing, China JOHN LUND • Department of Electrical Engineering, University of Washington Paul Allen Center, Seattle, WA, USA ANDREI D. MALAKHOV • Shemyakin-Ovchinnikov Institute of Bioorganic Chemistry, Miklukho-Maklaya, Moscow, Russia GEORGE MARNELLOS • Research and Development, Sequenom, Inc., San Diego, CA, USA MUSA M. MHLANGA • Gene Expression and Biophysics Unit, Institute for Molecular Medicine, Portugal and Gene Expression and Biophysics Group, CSIR Biosciences, Pretoria, South Africa
Contributors
xiii
GUY DEL MISTRO • Research and Development, Sequenom, Inc., San Diego, CA, USA KUVAT T. MOMYNALIEV • Research Institute of Physico-Chemical Medicine, MiklukhoMaklaya, Moscow, Russia PAUL OETH • Research and Development, Sequenom, Inc., San Diego, CA, USA YUKO OKAZAKI • Division of Genome Analysis, Research Center for Genetic Information, Medical Institute of Bioregulation, Kyushu University, Higashi-ku, Fukuoka, Fukuoka, Japan TAKAHITO OKI • Department of Legal Medicine, Shinshu University School of Medicine, Asahi, Matsumoto, Japan MICHAEL OLIVIER • Biotechnology and Bioengineering Center, Medical College of Wisconsin, Milwaukee, WI, USA MASAO OTA • Department of Legal Medicine, Shinshu University School of Medicine, Matsumoto, Japan BABAK A. PARVIZ • Department of Electrical Engineering, University of Washington Paul Allen Center, Seattle, WA, USA CARITA E. PASCAL • School of Aquatic and Fishery Sciences, University of Washington, Seattle, WA, USA CHRISTOPHER PHILLIPS • Forensic Genetics Unit, Institute of Legal Medicine, University of Santiago de Compostela, Galicia, Spain DANIELE PODINI • Department of Forensic Sciences, George Washington University, Washington, DC, USA IGOR A. PROKHORENKO • Shemyakin-Ovchinnikov Institute of Bioorganic Chemistry, Miklukho-Maklaya, Moscow, Russia RAMESH RAMAKRISHNAN • Fluidigm Corporation, South San Francisco, CA, USA JOSE LUIS ROYO • Department of Structural Genomics, Neocodex SL.; currently Centro Andaluz de Biologı´a del Desarrollo. Consejo Superior de Investigaciones Cientı´ficasUniversidad Pablo de Olavide, Seville, Spain MAURIZIO SABATTI • Department of Forest Ressources and Environment (DiSAFRi), University of Tuscia, Viterbo, Italy MASAHARU SADA • Department of Reproduction Medicine, National Cardiovascular Center, Suita, Osaka, Japan GEORGINA SALLMANN • Monash University, Department of Medicine, Central and Eastern Clinical School, Melbourne, Victoria, Australia ZUBEN E. SAUNA • Laboratory of Hemostasis, Division of Hematology, Center for Biologics Evaluation and Research, Food and Drug Administration, Bethesda, Maryland, USA JAMES E. SEEB • School of Aquatic and Fishery Sciences, University of Washington, Seattle, WA, USA LISA W. SEEB • School of Aquatic and Fishery Sciences, University of Washington, Seattle, WA, USA BARKUR S. SHASTRY • Department of Biological Sciences, Oakland University, Rochester, MI, USA MIKHAIL S. SHCHEPINOV • Tridend, a division of Oxford Gene Technology Ltd, Sandy Lane, Yarnton, Oxford, UK GONG-QING SHEN • Department of Molecular Cardiology, Lerner Research Institute, Cleveland Clinic, Cleveland, OH, USA
xiv
Contributors
TAO SHI • Research and Development, Sequenom, Inc., San Diego, CA, USA VADIM V. SHMANAI • Institute of Physical Organic Chemistry, Minsk, Belarus EDWARD SMITH • Department of Biological Sciences, University of Warwick, Coventry, UK TANIA TABONE • Ludwig Institute for Cancer Research, Parkville, Victoria, Australia TOMOKO TAHIRA • Division of Genome Analysis, Research Center for Genetic Information, Medical Institute of Bioregulation, Kyushu University, Higashi-ku, Fukuoka, Fukuoka, Japan PETER M. VALLONE • National Institute of Standards and Technology, Biochemical Science Division, Gaithersburg, MD, USA DIRK VAN DEN BOOM • Research and Development, Sequenom, Inc., San Diego, CA, USA JAN VAN OEVEREN • Division of Bioinformatics, Keygene NV, Wageningen, The Netherlands QING KENNETH WANG • Center for Cardiovascular Genetics, Department of Cardiovascular Medicine, Cleveland Clinic; Department of Molecular Medicine, Cleveland Clinic Lerner College of Medicine of Case Western Reserve University, Cleveland, OH, USA TOSHIHIDE YANAGAWA • Laboratory of Medical Genomics, Department of Human Genome Technology, Kazusa DNA Research Institute, Kisarazu, Chiba, Japan AKI YOSHINAGA • Division of Genome Analysis, Research Center for Genetic Information, Medical Institute of Bioregulation, Kyushu University, Higashi-ku, Fukuoka, Fukuoka, Japan TIMOFEI S. ZATSEPIN • Department of Chemistry, M. V. Lomonosov Moscow State University; Central Research Institute of Epidemiology, Novogireevskaya, Moscow, Russia YI ZHANG • Biotechnology and Bioengineering Center, Medical College of Wisconsin, Milwaukee, WI, USA
Chapter 1 SNPs: Impact on Gene Function and Phenotype Barkur S. Shastry Abstract Single nucleotide polymorphism (SNP) is the simplest form of DNA variation among individuals. These simple changes can be of transition or transversion type and they occur throughout the genome at a frequency of about one in 1,000 bp. They may be responsible for the diversity among individuals, genome evolution, the most common familial traits such as curly hair, interindividual differences in drug response, and complex and common diseases such as diabetes, obesity, hypertension, and psychiatric disorders. SNPs may change the encoded amino acids (nonsynonymous) or can be silent (synonymous) or simply occur in the noncoding regions. They may influence promoter activity (gene expression), messenger RNA (mRNA) conformation (stability), and subcellular localization of mRNAs and/or proteins and hence may produce disease. Therefore, identification of numerous variations in genes and analysis of their effects may lead to a better understanding of their impact on gene function and health of an individual. This improved knowledge may provide a starting point for the development of new, useful SNP markers for medical testing and a safer individualized medication to treat the most common devastating disorders. This will revolutionize the medical field in the future. To illustrate the effect of SNPs on gene function and phenotype, this minireview focuses on evidences revealing the impact of SNPs on the development and progression of three human eye disorders (Norrie disease, familial exudative vitreoretinopathy, and retinopathy of prematurity) that have overlapping clinical manifestations. Key words: Single nucleotide polymorphism, dominant, genomics, genotype, mutation, phenotype, recessive.
1. Introduction It is generally believed that the genomes between two randomly selected individuals contain approximately 0.1% difference or variation. This variation is called ‘‘polymorphism’’ and it arises because of mutations. The simplest form of DNA variation among individuals is the substitution of one single nucleotide for another. This type of change is called ‘‘single nucleotide polymorphism’’ (SNP) A.A. Komar (ed.), Single Nucleotide Polymorphisms, Methods in Molecular Biology 578, DOI 10.1007/978-1-60327-411-1_1, ª Humana Press, a part of Springer Science+Business Media, LLC 2003, 2009
3
4
Shastry
and it is found to be more common than other types of polymorphisms. It is estimated that SNPs occur at a frequency of approximately one in 1,000 base pairs (bp) throughout the genome (1) and more than three million SNPs have been charted so far. These simple changes are believed to be stable and not deleterious to organisms. According to a published report, 50% of SNPs occur in the noncoding regions, 25% lead to missense mutations, and the remaining 25% are silent mutations (2). These silent SNPs are also called ‘‘synonymous SNPs’’ because they do not change the encoded amino acids. Silent SNPs were largely assumed previously to have no effect on gene function and phenotype. However, recent findings show that this might not be the case (see Chapter 2). Nonsynonymous SNPs (altering amino acids) may often produce disease and therefore may be subjected to natural selection. SNPs can be observed between individuals in a population, and may, for example, influence promoter activity (gene expression), messenger RNA (mRNA) conformation (stability), and translational efficiency. Therefore, they may be responsible for the susceptibility of an individual to many common diseases, medicinal drug metabolism, and genome evolution. They may also play a direct role with or without other factors in the phenotypic expression of diseases or traits such as tallness, curly hair, and individuality (3–6). In recent years, application of clinicogenetic knowledge has revolutionized our ability to understand the effects of nucleotide substitutions and genetic basis of several complex and common disorders. In this article, an attempt has been made to correlate the effect of SNPs with gene function and phenotype, although such genotype–phenotype correlation is not straightforward. To illustrate this, three human eye disorders [Norrie disease, familial exudative vitreoretinopathy (FEVR), and retinopathy of prematurity (ROP)] that have overlapping clinical phenotypes have been selected. As can be seen from the following discussion, all of these disorders involve alterations in the same group of genes that are responsible for the same signal transduction pathway, providing evidence that SNPs alter gene function (producing cellular signaling defect) and hence may be responsible for the change in phenotype. Because of the brief nature of this article, readers are requested to consult other reviews (7–10) for detailed description of other aspects of SNPs.
2. Norrie Disease Before we attempt to understand the impact of SNPs on gene function and phenotype, it is important to start with clinical descriptions of some of the disorders we will be considering. For instance, Norrie
SNPs: Impact on Gene Function and Phenotype
5
disease, pseudoglioma (NDP) is a severe, X-linked recessive neurodevelopmental disorder affecting mostly males (11). However, manifesting female carriers, which could be due to skewed X-chromosome inactivation, have also been reported (12–14). The condition has been observed in all ethnic groups and is characterized by bilateral congenital blindness due to retinal detachment, retrolental mass, cataract, vitreous hemorrhage, optic nerve atrophy, glaucoma, and atrophy of the globe (11). It is also often associated with some form of mental retardation and progressive sensorineural deafness in the second decade of life. Additionally, in a number of patients, more complex phenotypes and intrafamilial variability in the onset and severity have been described. Although it is a rare disease, it has received attention because it affects the neuroectoderm involving brain, retina, and the inner ear. 2.1. Norrie Disease Gene
DNA linkage analysis (see Notes 1–4) has mapped the NDP gene to the short arm of the X chromosome at Xp11.3–p11.4 and a candidate gene has been isolated by positional cloning (15–16). The genomic size of the NDP gene is 28 kb and the gene consists of two introns and three exons of 201, 380, and 1,257 bp, respectively (Fig. 1.1a). The first exon is an untranslated region containing a potential TATA box and CT dinucleotide repeats that are shown to be involved in controlling the expression (17). An open reading frame of 399 bp is contained within exons 2 and 3 and the encoded protein contains 133 amino acids with a molecular mass of 15 kDa. The third exon is the most conserved region of the gene.
2.2. Functions of NDP Gene
The NDP gene product, norrin, is a highly evolutionarily conserved (18) cysteine-rich protein. The presence of a small leader sequence at the N-terminal end suggests that it is a secreted protein (19). The C-terminal cysteine-rich domain shares homology with mucins, immediate gene, and von Willebrand factor (19, 20). Computer-generated three-dimensional modeling of norrin reveals a cystine knot motif that is present in growth factor families such as nerve growth factor, transforming growth factor b2, and platelet derived growth factor BB (21, 22). On the basis of these observations as well as knockout and in vitro transfection experiments (23–25), it has been suggested that norrin may function as a dimeric growth factor in the retina and may have a role in vascular development. In addition, norrin is a high-affinity ligand to the frizzled-4 (FZD4) receptor (see below) and its binding to the receptor induces activation of the classical Wnt/b-catenin signaling pathway (26, 27). The gene is expressed in many tissues, including eye, ear, brain, placenta, muscle, and lung (15, 16, 20, 24). In situ hybridization experiments revealed abundant hybridization signals in outer nuclear, inner nuclear, and ganglion cell layers of the retina in mice, rabbits, and humans (25, 28). There is
6
Shastry
Fig. 1.1. The Norrie disease, pseudoglioma (NDP) (panel a), frizzled-4 (FZD4) (panel b) and low density lipoprotein receptor protein (LRP5) (panel c) genes showing the locations of some of the mutations within the protein domain. FZD4 is a seven transmembrane receptor and LRP5 is a single-pass coreceptor. The horizontal box in panel B denotes the cysteine-rich domain (CRD) in the N-terminal region of FZD4 protein. Black boxes in panel A represent the nontranslated regions. Maps are not drawn to scale.
also a significant expression in spiral ganglion cells, stria vascularis, and spiral ligament cells of mouse ear. In a mouse model of NDP, the primary lesion was localized to the stria vascularis and the mice exhibited an abnormal vasculature and eventual loss of vessels in this region. These results are consistent with its expression pattern (29). High expression levels have also been observed in the cerebellar granular layer, hippocampus, and the olfactory bulb of the rabbit brain. These data support the idea that norrin could play an important role in vascular development and differentiation or maintenance of the differentiated state of the cochlea, brain, and the retina. 2.3. Mutational Analysis and Phenotype
Several mutations in the NDP gene have been identified in affected individuals (Table 1.1) (see Note 5). These include deletions, insertions, and missense, nonsense, and frameshift mutations (30). Although NDP is a single gene disorder, considerable intra- and interfamilial variability in phenotypic expression has been described. For instance, two affected patients carrying the same (nonsense) mutation exhibited a different cognitive
SNPs: Impact on Gene Function and Phenotype
7
Table 1.1 A partial list of the single base pair mutations in the Norrie disease, pseudoglioma gene and their clinical manifestations Mutation
Clinical phenotype
Hearing impairment
Mental disturbance
Reference
M1V
ND
–
–
(45)
C39R
ND
+
–
(46)
S57X
ND
+
mild
(44)
K58N
ND (severe)
+
–
(38)
L61F
ND
+
–
(44)
L61P
ND
+
–
(31)
A63D
ND
+
–
(31)
C65Y
ND
+
+
(36)
C65W
ND
?
?
(31)
C69S
ND
+
+
(13)
S73X
ND (severe)
–
–
(40)
R74C
ND
+
–
(40)
S75C
ND
+
+
(44)
R90P
ND
–
–
(44)
C96Y
ND
–
+
(30)
S101F
ND (less severe)
–
–
(40)
K104Q
ND (less severe)
–
–
(37)
R109X
ND
+
+
(31)
C110X
ND
–
–
(44)
R121W
ND (severe)
–
–
(37)
R121Q
ND
–
–
(37, 38)
I123N
ND (severe)
+
+
(12, 31)
C128X
ND
–
+
(42, 31)
R41K
XL FEVR
–
–
(57)
H42R
XL FEVR
–
–
(57)
C110G
XL FEVR
–
–
(39)
Y120C
XL FEVR
–
–
(57)
R121L
XL FEVR
–
–
(60)
R121W
XL FEVR
–
–
(56, 58)
L124F
XL FEVR
–
–
(55)
ND Norrie disease, XL X-linked recessive, FEVR familial exudative vitreoretinopathy
8
Shastry
phenotype (31). Interestingly, at the same time, a patient having a 150-kb deletion, including the entire NDP gene, has been found to exhibit a mild phenotype (15). Although the comparison between the deletion size and the severity of the disorder may not be straightforward, large submicroscopic deletions, in general, appear to be associated with a more severe neurological syndrome. Mutations in exon 2 of the NDP gene (e.g., L16P, V45E, K58N, V60Q, L61P, L62P, C65Y, S73X, R74C, S75P) that are closer to amino acid 60 or located near the cystine knot motif are reported to cause a severe NDP phenotype with mental retardation or deafness or both (32–38). This could be due to conformational change of norrin because of substitution in highly conserved residues. For instance, codon 65 cysteine is thought to be critical for the formation of the disulfide bridge, which holds the cystine knot tertiary structure together. Disruption of this motif by substitution with tyrosine would be expected to result in loss of activity of the protein (36). Similarly, a substitution mutation (G ! A) of the first nucleotide in intron 1 of the NDP gene destroys the highly conserved splice donor site and produces typical NDP features with mental disturbance and hearing impairment. The message instability or translational inefficiency or both (35) could explain this phenotype. However, a splice donor site mutation in exon 2/intron 2 also exhibits a typical NDP phenotype but without mental retardation or deafness. This suggests that alterations at certain location have a more significant effect on the function of the gene. In contrast to exon 2 mutations, missense mutations at the Cterminus of the NDP gene (Fig. 1.1a) appear to produce much less severe phenotypes without mental retardation and deafness (34, 35, 37–40). For instance, S101F, A105T, C110G, C110R, and R115L alterations are associated with blindness only. Although alanine 105 is conserved and is located in the cysteinerich domain, its substitution with threonine is expected to alter the correct protein folding; but it only produced blindness (39). Similarly, C110G, which alters one of the conserved cysteines, produced mild symptoms such as retinal traction and intra- and subretinal exudates. However, a different substitution in the same codon (C110R) produces phthisis bulbi along with bilateral sensorineural deafness, suggesting a more severe functional abnormality in the protein. Interestingly, mutations at codons 18 (I18K) and 112 (G112E) seem to produce a significant phenotypic heterogeneity (34, 41). In one individual there was only a mild peripheral retinal pigmentary change with normal vision even at the age of 79, whereas in another individual with the same mutation, a unilateral subtotal retinal detachment was described. Similarly, a nonsense mutation at codon 126 (C126X) exhibits retinal fold in one eye with severe traction and vitreous opacity (35). This mutant protein lacks the last eight amino acids of the wild-type
SNPs: Impact on Gene Function and Phenotype
9
norrin, including three conserved cysteine residues involved in the formation of the cystine knot motif. The importance of conserved cysteine at codon 128 is demonstrated by the mutation (C128X) that produced congenital blindness and mental retardation but not auditory function (42). Thus, it appears that mutations affecting the cystine knot motif produce a severe effect and cause severe retinal dysplasia, suggesting the importance of this structure in the normal functioning of the gene/ protein in retinal development (43). On the other hand, missense mutations that do not involve the cystine knot motif within the C-terminal part of norrin cause a much less severe phenotype of the disease, with few exceptions. These amino acid changes may cause a partially impaired function of norrin (37). Additionally, the intra- and interfamilial phenotypic variation suggests that the ocular symptoms, mental retardation, and deafness are all pleiotrophic effects of mutations in the NDP gene with other factors. Similarly, several other symptoms, such as epilepsy, contraction of muscle, and growth retardation, that have been described in some patients could be due to other genes or nongenetic effects (39, 44). The examples above clearly show that that the observed phenotypes arise owing to a number of various mutations (in particular, SNPs); however, it should be noted that the exact manifestation of these phenotypes depends on the complex and intricate balance of many genetic factors/ interactions within the organism as well as interactions of the organism and the environment.
3. Familial Exudative Vitreoretinopathy
Another example to illustrate the impact of SNPs on the gene function and phenotype is the disorder called ‘‘familial exudative vitreoretinopathy’’ (FEVR), which has some clinical features similar to those of NDP. It is a rare bilateral eye disorder affecting fullterm infants, with a high penetrance and highly variable expressivity even within the same family. The disease usually strikes young patients and severely affected individuals may be legally blind during the first decade of life. The disorder involves primarily the vitreous body and the retina without any systemic association in most cases. The condition is characterized by the premature arrest of the vascularization of the peripheral retina. Minimally affected patients are generally free of any visual problems, but more severely affected eyes exhibit a variety of clinical signs, including progressive dragging of the macula, intraretinal and subretinal hemorrhage, retinal fold (Fig. 1.2), exudates, and retinal detachment
10
Shastry
Fig. 1.2. Fundus picture of (a) normal and (b) FEVR patient. The affected eye (b) shows retinal fold extending towards the retinal periphery with dragging of the optic nerve head.
(30). FEVR is a genetically heterogeneous disorder and is inherited as an X-linked recessive (47–49), autosomal dominant (50) and autosomal recessive trait (51–52). Additionally, genetic heterogeneity has been reported within both autosomal dominant and X-linked disorders (53, 54). 3.1. X-Linked FEVR Gene and Mutational Analysis
X-linked FEVR has been mapped to the same region as the NDP gene (Xp11.4) and both X-linked FEVR and NDP are now allelic disorders. Mutations in the NDP gene have been identified (Table 1.1) in several X-linked families (39, 55–60). Affected individuals exhibited variable expressivity with the same mutation (61, 62). For instance, in a family with R38C mutation, one patient became blind at the age of 5, while his brother retained good vision in the right eye. This same mutation in a French family produced NDP-like symptoms and in another population a patient has not developed any symptoms (61). This could be due to an incomplete penetrance, although the disorder is considered to be fully penetrant. In general, most of the Xlinked FEVR patients carrying mutations in the NDP gene exhibit some degree of retinal detachment or exudation. All of the missense mutations identified in X-linked FEVR are expected to affect the secondary structure of norrin and the severity of the phenotype depends on the type and location of substitution(s). For instance, mutations affecting codons 39, 65, 69, 96, 126, and 128 may interfere with the stability or folding of norrin by disrupting key disulfide bonds and may produce a more severe retinal dysgenesis. Mutations outside these amino acids may produce much milder phenotypes. It is also possible that there might be other environmental or systemic factors that may contribute to the phenotypic variability.
SNPs: Impact on Gene Function and Phenotype
11
3.2. Autosomal Dominant FEVR Gene and Its Product
Autosomal dominant FEVR, a major form of FEVR, was assigned to the chromosome 11q13-23 locus (63–67) and a second locus was mapped to chromosome 11p (68). Recently, a gene encoding Wnt receptor FZD4 from the 11q13 region was isolated and a spectrum of mutations has been reported in several affected families (69–75). The FZD4 gene encodes a 537 amino acid protein and contains two exons. It is widely expressed throughout the body and is an integral membrane protein belonging to the frizzled family receptors that bind secreted Wnt proteins. The Wnt signaling pathway plays a major role in embryonic development, tissue and cell polarity, and regulation of proliferation and eye development (76, 77). The frizzled receptors contain a cysteine-rich domain at the Nterminus that acts as the binding site for Wnts. The cysteinerich domain contains ten conserved cysteines that are involved in disulfide bond formation. A motif located at the C-terminus after the seventh hydrophobic domain (transmembrane domain) is also highly conserved and is needed for the activation of the Wnt/b-catenin pathway. Apart from this region, the rest of the C-terminal tail is not well conserved among frizzled proteins (78). Recently, it was also found that Wnts require interaction with low density lipoprotein receptor protein (LRP5), which is a single transmembrane receptor (78).
3.3. Mutational Analysis
Many affected autosomal dominant FEVR families have been found to contain a spectrum of mutations in the FZD4 gene. A deletion mutation at codon 501 (Fig. 1.1b) of FZD4 produces a widespread peripheral retinal avascularity, extraretinal neovascularization, and temporal dragging of the macula and optic nerve (74). This truncated protein traps wild-type frizzled protein in the endoplasmic reticulum by oligomerization and inhibits its signaling (79). However, a heterozygous mutation (M342V) produced only a milder phenotype (71), while H69Y, M105V, M157V, C181R, R417Q, G488D, and Q505X produced either mild to severe abnormality (72) or highly variable phenotypes (R417Q and G488D) without retinal detachment (70, 74) or with retinal detachment (Q505X). Interestingly, M105V did not produce retinal detachment but M105T produced bilateral retinal detachment at a very young age, resulting in blindness. These mutations are located in the region of the surface of the cysteine-rich domain (Fig. 1.1b) that is most strongly indicated in Wnt binding. Similarly, patients harboring S497F had the classic features of FEVR with dragged disc, hemorrhage, and gliosis. This mutation is located in the highly conserved C-terminal motif and is likely to interfere with the activation of the Wnt/b-catenin pathway. In short, the locations of several mutations suggest that they may result in loss of functions of FZD4 protein.
12
Shastry
3.4. LRP5 Gene Alterations Cause Autosomal Dominant and Autosomal Recessive FEVR
Heterozygous and homozygous missense mutations in the LRP5 gene have recently been reported as the underlying cause of the autosomal dominant and the autosomal recessive forms of FEVR, respectively (80, 81). Patients having LRP5 mutations exhibit reduced bone mineral density (69, 81) along with typical clinical symptoms of FEVR. This is the only systemic association that has been reported in some cases of FEVR. The LRP5 gene consists of 23 exons and encodes a protein of 1,615 amino acids. It contains four epidermal growth factor (EGF) precursor spacer domains (Fig. 1.1c), each comprising five YWTD repeats and one EGF repeat. These together form a bpropeller structure. This is followed by the three low density lipoprotein receptor like ligand binding domains, a single transmembrane domain, and a cytoplasmic tail (82). The gene is widely expressed in all tissues, including the retina. Mutations in LRP5 have also been associated with osteoporosis-pseudoglioma, a syndrome characterized by abnormal retinal vascularization (83). Mutations Y1168H and C1361G are in the conserved residues of LRP5 and produce classic FEVR features, including retinal fold and retinal detachment. However, mutation T173M produced abnormal retinal vasculature and retinal fold, although this amino acid is not well conserved. The insertion, deletion, and splicing mutations may lead to premature termination of protein translation, while the homozygous mutations (G550R, R570Q, R752G, and E1367K) that have been described for the autosomal recessive FEVR may cause destabilization of LRP5 (75, 80).
4. Retinopathy of Prematurity Retinopathy of prematurity (ROP) is an example to illustrate that SNPs in the noncoding regions can impair the functions of genes and change the phenotype. ROP is a leading cause of blindness in children. It is a retinal vascular disorder that affects infants with low birth weight and short gestational age (84, 85). The condition is characterized by the abnormal vessel growth in the vitreous humor that can lead to vitreoretinal traction, retinal detachment, and other complications resulting in blindness. The disorder in its most severe form can lead to retinal detachment that may result in severe visual impairments or blindness. Morphologically the disorder is similar to FEVR that occurs in full-term infants. 4.1. Pathogenesis and Predisposing Genetic Risk Factors
The pathogenesis of advanced ROP and its cause are currently unknown. In the past, many causative factors, such as length of time exposed to supplemental oxygen, excessive ambient light exposure, and hypoxia, were suggested but evidence for these as independent risk factors in recent years is not compelling. It is not clear why ROP in
SNPs: Impact on Gene Function and Phenotype
13
a subset of infants with low birth weight progresses to a severe stage (retinal detachment) despite timely intervention whereas in other infants with similar clinical characteristics ROP regresses spontaneously. Recent research with a candidate gene approach has identified mutations in NDP and FZD4 genes in a small number of ROP patients, which suggests a strong genetic predisposition to ROP besides environmental factors (86–95). This notion has been further supported by the higher concordance rate of the disorder in monozygotic twins (96), clinical and experimental animal studies (97), race (98), and strain-dependent differences in oxygen-induced ROP in the inbred rats (99, 100). However, variations observed in the NDP and FZD4 genes in ROP were present in a small percentage of patients or were limited geographically to specific ethnic populations (101–105). One possible explanation for the lower frequency or inconsistency among studies is the genetic heterogeneity of the condition that has been observed in many other inherited retinal disorders, including FEVR. It should also be remembered that ROP is a nonfamilial disease (similar to the sporadic cases of inherited diseases) and in such cases it is possible that many genes may be involved in disease pathogenesis. Interestingly, many mutations in the NDP gene that have been identified in ROP patients in the studies mentioned above are polymorphic changes (in the noncoding regions) and there are no functional experiments for every mutation that has been described that conclusively demonstrate their involvement in the disease processes. Although many of the mutations may not have any effect on the protein structure and functions, they can, however, influence the gene structure and function in vivo. For instance, mutations in the promoter and 50 - and 30 -untranslated regions may affect the stability of RNA, the transcriptional and translational efficiency of a gene, and localization of either mRNA or protein. It is not always necessary that only those SNPs that produce amino acid change in the coding sequence exhibit phenotypic differences. The abnormal expression of a gene can also have a similar influence on the phenotype. Interestingly, it has been found recently that the NDP gene is also mutated in certain cases of persistent fetal vasculature (PFV) and Coats disease (62, 106–107) in addition to the X-linked FEVR, ROP, and Norrie disease. These diseases also have a clinical appearance overlapping that of FEVR, NDP, and ROP.
5. The Wnt Signaling Pathway and Results of Change in Gene Function
The studies described in the previous sections show that NDP, FEVR, ROP, PFV, and Coats disease, which have overlapping clinical features, involve alterations in genes encoding the Wnt signaling pathway proteins. The Wnt signaling pathway (Fig. 1.3)
14
Shastry
Fig. 1.3. The canonical Wnt signaling pathway. Norrin and Wnt act as ligands to bind FZD4, which interacts with LRP5. In the absence of Wnt signaling, b-catenin is phosphorylated and subjected to proteosomal degradation. In the presence of Wnt signal, b-catenin accumulates in the cytoplasm and enters the nucleus. Its subsequent interaction with a member of Tcf/Lef family activates the transcription of Wnt target genes. AD autosomal dominant, AR autosomal recessive, XL Xlinked recessive, FEVR familial exudative vitreoretinopathy, ROP retinopathy of prematurity, NDP Norrie disease, pseudoglioma.
is highly conserved and regulated among many species. It plays a key role in embryonic development, including eye development (76, 77, 108, 109). Wnts are secreted cysteine-rich glycoproteins that act as ligands and bind frizzleds and coreceptors LRP5/LRP6. A variety of responses could be initiated from Wnt–frizzeld interactions. The norrin/FZD4 receptor pair activates the well-known cascade, the canonical Wnt/b-catenin pathway in the presence of LRP5/LRP6 that regulates the expression of Wnt target genes through b-catenin/T-cell receptor. The above-mentioned studies illustrate that SNPs in either ligand (norrin) or receptor genes (FZD4 and LRP5) of the same signaling pathway alter gene functions that may produce changes in the signaling which in turn produce complex and highly variable phenotypes. Because the same signaling pathway is defective in all of the above-mentioned disorders, it is not surprising that all of them have some degree of overlapping clinical manifestations. Although there is no direct correlation between genotype and phenotype, alterations in highly conserved amino acids (110) may affect the structure of receptors, ligands, localization of mRNA or proteins, and/or the levels of proteins. This may inactivate or alter the pathway, resulting in the
SNPs: Impact on Gene Function and Phenotype
15
inhibition or abnormality of vascular development in the abovementioned disorders. Mutations in the 50 - and 30 -untranslated regions (seen in NDP gene in ROP patients) may also alter the regulation of gene expression at the level of transcription, translation, and mRNA stability (111–113). This may result in profound changes in the signal transduction (reduction in signaling) and hence produce changes in vascular development (114). Mice lacking FZD4 and LRP5 further demonstrate the importance of wildtype genes in capillary maturation as well as of this pathway in vasculogenesis and normal retinal development (26, 115–116). Further understanding of this pathway in ocular disease may lead to a novel therapeutic approach to treat or prevent these potentially blinding disorders in the future.
6. Concluding Remarks In this article, with use of human diseases as examples, the effects of SNPs on the gene functions and phenotype have been presented. Depending on the location of a codon and the type of substitution, structural and functional changes in the encoded proteins may result, and this may change the phenotype. The phenotypic changes may vary from individual to individual depending on the degree of functional changes in the gene product. We must also not forget that phenotypic variations could be due to the prolonged interactions of genetic and environmental factors. These types of changes in the gene are also responsible for every human trait, such as curly hair, individuality, and interindividual difference in drug response and genome evolution. On the basis of these arguments, by cataloging the DNA polymorphisms in different populations and species, it may be possible to (a) develop genome-based knowledge of the susceptibility of an individual to many common diseases, (b) manufacture a safer and effective individualized medication, and (c) understand the evolutionary processes. In the future, it is hoped that research will uncover methods of making SNP markers as useful tags for medical testing. Finding out how SNPs affect the health of an individual and drug reaction at the gene level rather than the protein level, and then transferring this knowledge to the development of a new medicine, will undoubtedly revolutionize the treatment of most common devastating disorders. If this concept of personal genomics becomes realistic, every newborn child in the neonatal unit may be genotyped in a routine procedure (similar to the finding of a blood type before blood transfusion) for disease susceptibility and improved treatment. Thus, for patients, the future looks bright, but there are many challenges.
16
Shastry
7. Notes
1. Many disease-causing genes can be mapped by linkage analysis without prior knowledge of the function and product of the gene. A successful genetic linkage analysis, however, requires family data, application of statistical analysis of laboratory experiments, and interpretation of these data by using the tools of genetic epidemiology. 2. A correct diagnosis of disease phenotype plays an important role in linkage analysis. Misdiagnosis may mask the real linkage or demand more data to obtain the same power to detect a linkage. 3. In selecting the genetic markers, scientists use markers that are polymorphic and provide sufficient coverage of the entire genome. The most useful probes in this respect are the microsatellite repeat polymorphisms. It is now possible to select such highly polymorphic probes covering most of the genome because genetic maps of these markers spaced at 10-cM intervals are now available for all human chromosomes. 4. Many times in linkage analysis, a blind interpretation of genotypic informationbyanindividual whodoesnot knowthediseasestatus of the individual being genotyped becomes helpful in preventing the overinterpretation of questionable genotyping results. 5. Although a variety of methods are available for mutational analysis of candidate genes, cycle sequencing has proved to be a fast and efficient method. It involves only a small amount of DNA template and does not require additional cloning. In addition, multiple samples can be simultaneously analyzed and compared. This eliminates artifacts of sample preparation and increases the accuracy of the results.
Acknowledgements My apologies to those whose work or original publications could not be cited in this short article because of space limitations. References 1. Brookes, A. J. (1999) The essence of SNPs. Gene 234, 177–186. 2. Halushka, M. K., Fan, J. B., Bentley, K., Hsie, L., Shen, N. P., Weder, A., Cooper, R., Lipshutz, R. and Chakravarti, A. (1999) Patterns of single nucleotide
polymorphisms in candidate genes for blood pressure homeostasis. Nat. Genet. 22, 239–247. 3. Martin, N., Boomsma, D. and Machin, G. (1997) A twin pronged attacks on complex traits. Nat. Genet. 17, 387–392.
SNPs: Impact on Gene Function and Phenotype 4. Krawezak, M., Reiss, J. and Cooper, D. N. (1992) The mutational spectrum of single base-pair substitutions in mRNA splice junctions of human genes: causes and consequences. Hum. Genet. 90, 41–54. 5. Lohrer, H. D. and Tangen, U. (2000) Investigations into the molecular effects of single nucleotide polymorphism. Pathobiology 68, 283–290. 6. LeVan, T. D., Bloom, J. W., Bailey, T. J., Karp, C. L., Halonen, M., Martinez, F. D. and Vercelli, D. (2001) A common single nucleotide polymorphism in the CD 14 promoter decreases the affinity of Sp protein binding and enhances transcriptional activity. J. Immunol. 167, 5838–5844. 7. Shastry, B. S. (2007) SNPs in disease gene mapping, medicinal drug development and evolution. J. Hum. Genet. 52, 871–880. 8. Shastry, B. S. (2006) Pharmacogenetics and the concept of individualized medicine. Pharmacogenomics J. 6, 16–21. 9. Shastry, B. S. (2005) Genetic diversity and new therapeutic concepts. J. Hum. Genet. 50, 321–328. 10. Shastry, B. S. (2004) Role of SNP/haplotype map in gene discovery and drug development: an overview. Drug Dev. Res. 62, 143–150. 11. Warburg, M. (1966) Norrie’s disease: a congenital progressive oculo-acustico-cerebral degeneration. Acta Ophthalmol. 89, 1–147. 12. Sims, K. B., Irvine, A. R. and Good, W. V. (1997) Norrie disease in a family with a manifesting female carrier. Arch. Ophthalmol. 115, 517–519. 13. Chen, Z. Y., Battinelli, E. M., Woodruff, G., Young, I., Breakefield, X. O. and Craig, I. W. (1993) Characterization of mutation within the NDP gene in a family with a manifesting female carrier. Hum. Mol. Genet. 2, 1727–1729. 14. Shastry, B. S., Hiraoka, M., Trese, D. C. and Trese, M. T. (1999) Norrie disease and exudative vitreoretinopathy in families with affected female carries. Eur. J. Ophthalmol. 9, 238–242. 15. Berger, W., Meindl, A., van de Pool, T. J., Cremers, F. P., Ropers, H.-H., Doerner, C., Monaco, A., Bergen, A. A., Lebo, R., Warburg, M., Zergollern, L., Lorenz, B., Gal, A., Bleeker-Wagemakers, E. M. and Meitinger, T. (1992) Isolation of a candidate gene for Norrie’s disease by positional cloning. Nat. Genet. 1, 199–203. 16. Chen, Z. Y., Hendricks, R. W., Jobling, M. T. A., Powell, J. F., Breakefield, X. O., Sims, K. B.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
17
and Criag, I. W. (1992) Isolation and characterization of a candidate gene for Norrie disease. Nat. Genet. 1, 204–208. Kenyon, J. R. and Criag, I. W. (1999) Analysis of the 5’-regulatory region of the human Norrie disease gene: evidence that non-translated CT dinucleotide repeat in exon-1 has a role in controlling expression. Gene 277, 181–188. Katoh, M. and Katoh, M. (2005) Comparative genomics on Norrie disease gene. Int. J. Mol. Med. 15, 885–889. Meindl, A., Berger, W., Meitinger, T., van de Pool, D., Achatz, H., Dorner, C., Haasemann, M., Hellebrand, G. A., Cremers, F. and Ropers, H.-H. (1992) Norrie disease is caused by mutations in an extracellular protein resembling C-terminal globular domains of mucins. Nat. Genet. 2, 139–143. Chen, Z. Y., Bettinelli, E. M., Hendricks, R. W., Powell, J. F., Middleton-Price, H., Sims, K. B., Breakefield, X. O. and Craig, I. W. (1993) Norrie disease gene: characterization of deletions and possible functions. Genomics 16, 533–535. Meitinger, T., Meindl, A., Bork, P., Rost, B., Sander, C., Haasemann, M. and Markrn J. (1993) Molecular modeling of Norrie disease protein predicts a cystine knot growth factor tertiary structure. Nat. Genet. 5, 376–380. McDonald, N. G. and Hendrickson, W. A. (1993) A structural superfamily of growth factors containing a cystine knot motif. Cell 73, 421–424. Perez-Vilar, J. and Hill, R. L. (1997) Norrie disease protein (norrin) forms disulphide linked oligomers. J. Biol. Chem. 272, 33410–33415. Luhmann, U. F., Meunier, D., Shi, W., Luttges, A., Pfarrer, C., Fundele, R. and Berger, W. (2005) Fetal loss in homozygous mutant Norrie disease mice: a new role of norrin in reproduction. Genesis 42, 253–262. Berger, W., van de Pool, D., Bachner, D., Oerlemans, F., Winkens, H., Hameister, H., Wieringa, B., Hendricks, W. and Roper, H.-H. (1996) An animal model for Norrie disease: gene targeting of mouse ND gene. Hum. Mol. Genet. 5, 51–59. Xu, O., Wang, Y. S., Dabdoub, A., Smallwood, P. M., Williams, J., Woods, C., Kelley, M. W., Jiang, L., Tasman, W., Zhang, K. and Nathans, J. (2004) Vascular development in the retina and inner ear: control by norrin and frizzled 4, a high affinity ligand receptor pair. Cell 116, 883–895.
18
Shastry
27. Luhmann, U. F., Lin, J., Acar, N., Lammel, S., Feils, S., Grimm, C., Seeliger, M. W., Hammes, H. P. and Berger, W. (2005) Role of the Norrie disease pseudoglioma gene in sprouting angiogenesis during development of the retinal vasculature. Invest. Ophthalmol. Vis. Sci. 46, 3372–3382. 28. Hartzer, M. K., Cheng, M., Liu, X. and Shastry, B. S. (1999) Localization of the Norrie disease gene mRNA by in situ hybridization. Brain Res. Bull. 49, 355–358. 29. Rehm, H. L., Zhang, D. S., Brown, M. C., Burgess, B., Halpin, C., Berger, W., Morton, C., Corey, D. P. and Chen, Z., Y. (2002) Vascular defects and sensorineural deafness in a mouse model of Norrie disease. J. Neurosci. 22, 4286–4292. 30. Shastry, B. S. and Hiraoka, M. (2000). Molecular genetics of familial exudative vitreoretinopathy and Norrie disease. Curr. Genomics 1, 259–269. 31. Schuback, D. E., Zheng, Y. C., Criag, I. W., Breakefield, X. O. and Sims, K. B. (1995) Mutations in the Norrie disease gene. Hum. Mutat. 5, 285–292. 32. Lev, D., Weigi, Y., Hasan, M., Gak, E., Davidovich, M., Vinkler, C., Leshinsky-Silver, E., Lerman-Sagie, T. and Watemberg, N. (2007) A novel missense mutation in the NDP gene in a child with Norrie disease and severe neurological involvement including infantile spasms. Am. J. Med. Genet. 143, 921–924. 33. Yamada, K., Limprasert, P., Ratanasukon, M., Tengtrisorn, S., Yingchareonpukdee, J., Vasiknanonte, P., Kitaoka, T., Ghadami, M., Niikawa, N. and Kishino, T. (2001) Two Thai families with Norrie disease (ND): association of two novel missense mutations with severe ND phenotype, seizures and a manifesting carrier. Am. J. Med. Genet. 100, 52–55. 34. Kondo, H., Qin, M., Kusaka, S., Tahira, T., Hasebe, H., Hayashi, H., Uchio, E. and Hayashi, K. (2007) Novel mutations in Norrie disease gene in Japanese patients with Norrie disease and familial exudative vitreoretinopathy. Invest. Ophthalmol. Vis. Sci. 48, 1276–1282. 35. Fuchs, S., van de Pool, D., Beudt, U., Kellner, U., Meire, F., Berger, W. and Gal, A. (1996) Three novel and two recurrent mutations of the Norrie disease gene in patients with Norrie syndrome. Hum. Mutat. 8, 85–88. 36. Strasberg, P, Liede, H. A., Stein, T., Warren, I., Sutherland, J. and Ray, P. N. (1995) A
37.
38.
39.
40.
41.
42.
43.
44.
45.
novel mutation in the Norrie disease gene predicted to disrupt the cystine knot growth factor motif. Hum. Mol. Genet. 4, 2179–2180. Meindl, A., Lorenz, B., Achatz, H., Hellebrand H., Schmitz-Vakkenberg, P. and Meitinger, T. (1995) Missense mutation in the NDP gene in patients with a less severe course of Norrie disease. Hum. Mol. Genet. 4, 489–490. Fuentes, J. J., Volpini, V., Fernandez-Toral, F., Coto, E. and Estivill, X. (1993) Identification of two new missense mutations (K58N and R121Q) in the Norrie disease (ND) gene in two Spanish families. Hum. Mol. Genet. 2, 1953–1955. Torrente, I., Mangino, M., Gennarelli, M., Novelli, G., Giannotti, A., Vadala, P. and Dallapiccola, B. (1997) Two new missense mutations (A105T and C110G) in the Norrie gene in two Italian families with Norrie disease and familial exudative vitreoretinopathy. Am. J. Med. Genet. 72, 242–244. Walker, J. L., Dixon, J., Fenton, C. R., Hungerford, J., Lynch, S. A., Stenhouses, S. A. K., Christian, A. and Craig, I. W. (1997) Two new mutations in exon 3 of the NDP gene: S73X and S101F associated with severe and less severe ocular phenotypes respectively. Hum. Mutat. 9, 53–56. Allen, R. C., Russell, S. R., Streb, L. M., Alsheikheh, A. and Stone, E. M. (2006) Phenotypic heterogeneity associated with a novel mutation (Gly 112 Glu) in the Norrie disease protein. Eye 20, 234–241. Wang, F., Goldberg, M. F. and Hao, Y. (1993) Identification of a nonsense mutation at codon 128 of the Norrie’s disease gene in a male infant. Arch. Ophthalmol. 111, 1553–1557. Drenser, K. A., Fecko, A., Dailey, W. and Trese, M. T. (2007) A characteristic phenotypic retinal appearance in Norrie disease. Retina 27, 243–246. Berger, W., van de Pool, D., Warburg, M., Gal, A., Bleeker-Wagemakers, L., de Silva, H., Meindl, A., Meitinger, T., Cremers, F. and Ropers, H-H. (1992) Mutations in the candidate gene for Norrie disease. Hum. Mol. Genet. 1, 461–465. Isashiki, Y., Ohba, N., Yanagita, T., Hokita, N., Doi, N., Nakagawa, M., Ozawa, M. and Kuroda, N. (1995) Novel mutation at the initiation codon in the Norrie disease gene in two Japanese families. Hum. Genet. 95, 105–108.
SNPs: Impact on Gene Function and Phenotype 46. Joos, K. M., Kimura, A. E., Vandenburgh, K., Bartley, J. A. and Stone E. M. (1994) Ocular findings associated with a cys39Arg mutation in the Norrie disease gene. Arch. Ophthalmol. 112, 1574–1579. 47. Plager, D. A., Orgel, I. K., Forest, D. E., Hartzer, M., Trese, M. T. and Shastry, B. S. (1992) X-linked recessive familial exudative vitreoretinopathy. Am. J. Ophthalmol. 114, 145–148. 48. Fullwood, P., Jones, J., Bundey, S., Dudgeon, J., Fielder, A. R. and Kilpatrick, M. W. (1993) X-linked exudative vitreoretinopathy: clinical features and genetic linkage analysis. Br. J. Ophthalmol. 77, 168–170. 49. Clement, F., Beckford, C. A., Corral, A. and Jimenez, R. (1995) X-linked familial exudative vitreoretinopathy. Retina 15, 141–145. 50. Van Nouhuys, C. E. (1982) Dominant exudative vitreoretinopathy and other vascular developmental disorders of the peripheral retina. Doc. Ophthalmol. 54, 1–414. 51. Shastry, B. S. and Trese, M. T. (1997) Familial exudative vitreoretinopathy: further evidence for genetic heterogeneity. Am. J. Med. Genet. 69, 217–218. 52. De Crecchio, G., Simonelli, F., Nunziata, G., Mazzeo, S., Greco, G. M., Rinaldi, E., Ventruto, V., Ciccodicola, A., Miano, M. G., Testa, F., Curci, A., D’Urso, M., Rinaldi, M. M., Cavaliere, M. L. and Castelluccio, P. (1998) Autosomal recessive familial exudative vitreoretinopathy: evidence for genetic heterogeneity. Clin. Genet. 54, 315–320. 53. Shastry, B. S., Liu, X., Hejtmancik, J. F., Plager, D. A. and Trese, M. T. (1997) Evidence for genetic heterogeneity in X-linked familial exudative vitreoretinopathy. Genomics 44, 247–248. 54. Bamashmus, M. A., Downey, L. M., Inglehearn, C. F., Gupta, S. R. and Mansfield, D. C. (2000) Genetic heterogeneity in familial exudative vitreoretinopathy: exclusion of the EVR1 locus on chromosome 11q in a large autosomal dominant pedigree. Br. J. Ophthalmol. 84, 358–363. 55. Chen, Z. Y., Bettinelli, E. M., Fielder, A., Bundey, S., Sims, K., Breakefield, X. O. and Craig, I. W. (1993) A mutation in the Norrie disease gene associated with X-linked familial exudative vitreoretinopathy. Nat. Genet. 5, 180–183. 56. Shastry, B. S., Hejtmancik, J. F., Plager, D. A., Hartzer, M. T. and Trese, M. T. (1995) Linkage and candidate gene analysis of Xlinked familial exudative vitreoretinopathy. Genomics 27, 341–344.
19
57. Shastry, B. S., Hejtmancik, J. F. and Trese, M. T. (1997) Identification of novel missense mutations in the Norrie disease gene associated with one X-linked and four sporadic cases of familial exudative vitreoretinopathy. Hum. Mutat. 9, 396–401. 58. Fuchs, S., Kellner, U., Wedemann, H. and Gal, A. (1995) Missense mutation (Arg121Trp) in the Norrie disease gene associated with X-linked exudative vitreoretinopathy. Hum. Mutat. 6, 257–259. 59. Drenser, K. A., Dailey, W., Capone, A. and Trese, M. T. (2006) Genetic evaluation to establish the diagnosis of X-linked familial exudative vitreoretinopathy. Ophthalmic Genet. 27, 75–78. 60. Johnson, K., Mintz-Hittner, H. A., Conley, Y, P. and Ferrell, R. E. (1996) X-linked exudative vitreoretinopathy caused by an arginine to leucine substitution (R121L) in the Norrie disease protein. Clin. Genet. 50, 113–115. 61. Riveiro-Alvarez, R., Trujillo-Tiebas, M. J., Gimenez-Pardo, A., Garcia-Hoyos, M., Cantalapiedra, D., Lorda-Sanchez, I., de Alba, M. R., Ramos, C. and Ayuso, C. (2005) Genotype-phenotype variations in five Spanish families with Norrie disease or X-linked FEVR. Mol. Vis. 11, 705–712. 62. Wu, W.-C., Drenser, K., Trese, M. T., Capone, A. and Dailey, W. (2007) Retinal genotype-phenotype correlation of pediatric patients expressing mutations in the Norrie disease gene. Arch. Ophthalmol. 125, 225–230. 63. Li, Y., Fuhrmann, C., Schwinger, E., Gal, A. and Laqua, H. (1992) The gene for autosomal dominant exudative vitreoretinopathy (Criswick-Schepens) is on the long arm of chromosome 11. Am. J. Ophthalmol. 113, 712–713. 64. Muller, B., Orth, U., van Nouhuys, C. E., Durigneau, C., Fuhrmann, C., Schwinger, E., Laqua, H. and Gal, A. (1994) Mapping of the autosomal dominant exudative vitreoretinopathy locus (EVR1) by multipoint linkage analysis in four families. Genomics 20, 317–319. 65. Price, S. M., Periam, N., Humphries, A., Woodruff, G. and Trembath, R. C. (1996) Familial exudative vitreoretinopathy linked to D11S533 in a large Asian family with consanguinity. Ophthalmic Genet. 17, 53–57. 66. Shastry, B. S., Hejtmancik, J. F., Hiraoka, M., Ibaraki, N., Okubo, Y., Okubo, A., Han, D. P. and Trese, M. T. (2000) Linkage and candidate gene analysis of autosomal
20
67.
68.
69.
70.
71.
72.
73.
74.
75.
Shastry dominant familial exudative vitreoretinopathy. Clin. Genet. 58, 329–332. Kondo, H., Ohno, K., Tahira, T., Hayashi, H., Oshima, K. and Hayashi, K. (2001) Delineation of the critical interval for the familial exudative vitreoretinopathy gene by linkage and haplotype analysis. Hum. Genet. 108, 368–375. Downey, L. M., Keen, T. J., Roberts, E., Mansfield, D. C., Bamashmus, M. and Inglehearn, C. F. (2001) A new locus for autosomal dominant familial exudative vitreoretinopathy maps to chromosome 11p12-13. Am. J. Hum. Genet. 68, 778–781. Qin, M., Hayashi, H., Oshima, K., Tahira, T., Hayashi, K. and Kondo, H. (2005) Complexity of genotype-phenotype correlation in familial exudative vitreoretinopathy with mutations in the LRP5 and/or FZD 4 genes. Hum. Mutat. 26, 104–112. Kondo, H., Hayashi, H., Oshima, K., Tahira, T. and Hayashi, K. (2003) Frizzled 4 gene (FZD 4) mutations in patients with familial exudative vitreoretinopathy with variable expressivity. Br. J. Ophthalmol. 87, 1291–1295. Yoshida, H., Arita, R., Yoshida, A., Tada, H., Emori, A., Noda, Y., Nakao, S., Fujisawa, K. and Ishibashi, T. (2004) Novel mutation in FZD 4 gene in a Japanese pedigree with familial exudative vitreoretinopathy. Am. J. Ophthalmol. 138, 670–671. Omoto, S., Hayashi, T., Kitahara, K., Takeuchi, T. and Ueoka, Y. (2004) Autosomal dominant familial exudative vitreoretinopathy in two Japanese families with FZD 4 mutations (H69Y and C181R). Ophthalmic Genet. 25, 81–90. Robitaille, J., MacDonald, M. L. E., Kaykas, A., Sheldahi, L. C., Zeisler, J., Dube, M. P., Zhang, L. H., Singaraja, R. R., Guernsey, D. L., Zheng, B., Siebert, L. F., Hoskin-Mott, A., Trese, M. T., Pimstone, S. N., Shastry, B. S., Moon, R. T., Hayden, M. R., Goldberg, Y. P. and Samuels, M. E. (2002) Mutant frizzled 4 disrupts retinal angiogenesis in familial exudative vitreoretinopathy. Nat. Genet. 32, 326–330. Toomes, C., Bottomley, H. M., Scott, S., Mackey, D. A., Craig, J. E., Appukuttan, B., Stout, J. T., Faxel, C. J., Zhang, K., Black, G. C. M., Fryer, A., Downey, L. M. and Inglehearn, C. F. (2004) Spectrum and frequency of FZD 4 mutations in familial exudative vitreoretinopathy. Invest. Ophthalmol. Vis. Sci. 45, 2083–2090. Toomes, C., Bottomley, H. M., Jackson, R. M., Towns, K. V., Scott, S., Mackey,
76.
77.
78.
79.
80.
81.
82.
83.
84.
D. A., Craig, J. E., Jiang, L., Yang, Z., Trembath, R., Woodruff, G., GregoryEvans, C. Y., Gregory-Evans, K., Parker, M. J., Black, G. C. M., Downey, L. M., Zhang, K. and Inglehearn, C. F. (2004) Mutations in LRP5 or FZD 4 underlie the common familial exudative vitreoretinopathy locus on chromosome 11q. Am. J. Hum. Genet. 74, 721–730. Huang, H.-C. and Klein, P. S. (2004) The frizzled family: receptors for multiple signal transduction pathways. Genome Biol. 5, 234–239. De Iongh, R. U., Abad, H. E. and Hime, G. R. (2006) Wnt/frizzled signaling in eye development and disease. Frontiers Biosci. 11, 2442–2464. Pinson, K. L., Brennan, J., Monkley, S., Avery, B. J. and Sharnes, W. C. (2000) An LDL – receptor related protein mediates Wnt signaling in mice. Nature 407, 535–538. Kykas, A., Yang-Snyder, J., Aeroux, M., Shah, K. V., Bouvier, M. and Moon, R. T. (2004) Mutant frizzled 4 associated with vitreoretinopathy traps wild-type frizzled in the endoplasmic reticulum by oligomerization. Nat. Cell. Biol. 6, 52–58. Jiao, X., Ventruto, V., Trese, M. T., Shastry, B. S. and Hejtmancik, J. F. (2004) Autosomal recessive familial exudative vitreoretinopathy is associated with mutations in LRP5. Am. J. Hum. Genet. 75, 878–884. Downey, L. M., Bottomley, H. M., Sheridan, E., Ahmed, M., Gilmour, D. F., Inglehearn, C. F., Reddy, A., Agrawal, A., Brodbury, J. and Toomes, C. (2006) Reduced bone mineral density and hyaloid vasculature remnants in a consanguineous recessive FEVR family with a mutation in LRP5. Br. J. Ophthalmol. 90, 1163–1167. Hey, P. J., Twells, R. C., Phillips, M. S., Yusuke, N., Brown, S. D., Kawaguchi, Y., Cox, R., Gouchun, X., Dugan, V., Hammond, H., Metzker, M. L., Todd, J. A. and Hess, J. F. (1998) Cloning of a novel member of the low-density lipoprotein receptor family. Gene 216, 103–111. Gong, Y., Slee, R. B., Fukai, N., Rawadi, G., Roman-Roman, S., Reginato, A. M. and Wang, H. et al. (2001) LDL receptor related protein 5 (LRP5) affects bone accrual and eye development. Cell 107, 513–523. Plamer, E. A. (1996) The continuing threat of retinopathy of prematurity. Am. J. Ophthalmol. 122, 420–423.
SNPs: Impact on Gene Function and Phenotype 85. Gilbert, C., Rahi, J., Eckstein, M., O’Sullivan, J. and Foster, A. (1997) Retinopathy of prematurity in middle-income countries. Lancet 350, 12–14. 86. Shastry, B. S., Pendergast, S. D., Hartzer, M.K., Liu, X. and Trese, M. T. (1997) Identification of missense mutations in the Norrie disease gene associated with advanced retinopathy of prematurity. Arch. Ophthalmol. 115, 651–656. 87. Vannay, A., Dunai, G., Banyasz, I.,Szabo, M., Vamos, R., Treszl, A., Hajdu, J., Tulassay, T. and Vasarhelyi, B. (2005) Association of genetic polymorphisms of vascular endothelial growth factor and risk of proliferative retinopathy of prematurity. Pediatr. Res. 57, 396–398. 88. Cooke, R. W., Drury, J. A., Mountford, R. and Clark, D. (2004) Genetic polymorphism and retinopathy of prematurity. Invest. Ophthalmol. Vis. Sci. 45, 1712–1715. 89. Haider, M. Z., Devarajan, L. V., Al-Essa, M. and Kumar, H. A. (2002) A C597 ! A polymorphism in the Norrie disease gene is associated with advanced retinopathy of prematurity in premature Kuwaiti infants. J. Biomed. Sci. 9, 365–370. 90. Haider, M. Z., Devarajan, L.V., Al-Essa, M. and Kumar, H. (2002) Angiotensin – converting enzyme gene insertion/deletion polymorphism in Kuwaiti children with retinopathy of prematurity. Biol. Neonate. 82, 84–88. 91. Talks, S. J., Ebenezer, N., Hykin, P., Adams, A., Yang, F., Schulenberg, E., GregoryEvans, K. and Gregory-Evans, C. Y. (2001) De novo mutations in the 5’ regulatory region of the Norrie disease gene in retinopathy of prematurity. J. Med. Genet. 38, e46. 92. Hiraoka, M., Berinstein, D. M., Trese, M. T. and Shastry, B. S. (2001) Insertion and deletion mutations in the dinucleotide repeat region of the Norrie disease gene in patients with advanced retinopathy of prematurity. J. Hum. Genet. 46, 178–181. 93. Hutcheson, K. A., Paluru, P. C., Bernstein, S. L., Koh, J., Rappaport, E. F., Leach, R. A. and Young, T. L. (2005) Norrie disease gene sequence variants in an ethnically diverse population with retinopathy of prematurity. Mol. Vis. 11, 501–508. 94. Dickinson, J. L., Sale, M. M., Passmore, A., FitzGerald, L. M., Wheatley, C. M., Burdon, K. P., Franzco, J. E. C., Tengtrisorn, S., Franzco, S. M. C., Franzco, H. M. and Franzco, D. A. M. (2006) Mutations in the NDP gene: contribution to Norrie disease, familial exudative
95.
96.
97.
98.
99.
100.
101.
102.
103.
104.
21
vitreoretinopathy and retinopathy of prematurity. Clin Exp. Ophthalmol. 34, 682–688. MacDonald, M. L. E., Goldberg, Y. P., MacFarlane, J., Samuels, M. E., Trese, M. T. and Shastry, B. S. (2005) Genetic variants of frizzled-4 gene in familial exudative vitreoretinopathy and advanced retinopathy of prematurity. Clin Genet. 67, 363–366. Bizzarro, M. J., Hussain, N., Jonsson, B., Feng, R., Ment, L. R., Gruen, J. F., Zhang, H. and Bhandari, V. (2006) Genetic susceptibility to retinopathy of prematurity. Pediatrics 118, 1858–1863. Holmstrom, G., Van Wijngaarden, P., Coster, D. J. and Williams, K. A. (2007) Genetic susceptibility to retinopathy of prematurity: the evidence from clinical and experimental animal studies. Br. J. Ophthalmol. 97, 1704–1708. Lang, D. M., Backledge, J. and Arnold, R. W. (2005) Is specific race a retinopathy prematurity risk factor. Arch. Pediatr. Adolesc. Med. 159, 771–773. Floyd, B. N., Leske, D. A., Wren, S. M., Mookadam, M., Fautsch, M. P. and Holmes, J. M. (2005) Differences between rat strains in models of retinopathy of prematurity. Mol. Vis. 11, 524–530. Van Wijngaarden, P., Coster, D. J., Brereton, H. M., Gibbins, I. L. and Williams, K. A. (2005) Strain-dependent differences in oxygen-induced retinopathy of prematurity in the inbred rats. Invest. Ophthalmol. Vis. Sci. 46, 1445–1452. Shastry, B. S. (2007) Assessment of the contribution of insulin-like growth factor I receptor 3174 G ! A polymorphism to the progression of advanced retinopathy of prematurity. Eur. J. Ophthalmol. 17, 950–953. Shastry, B. S. and Qu, X. (2007) Lack of association of the VEGF gene promoter (–634G ! C and –460C ! T) polymorphism and the risk of advanced retinopathy of prematurity. Graefes Arch. Clin. Exp. Ophthalmol. 245, 741–743. Balogh, A., Derzbach, L., Vannay, A. and Vasarhelyi, B. (2006) Lack of association between insulin-like growth factor I 3174G ! A polymorphism and retinopathy of prematurity. Graefes Arch. Clin. Exp. Ophthalmol. 244, 1035–1038. Kwinta, P., Mitkowska, Z., Tomasik., T., Bik-Multanowski, M. and Pietrzyk, J. J. (2004) Vascular endothelial growth factor gene polymorphism and the risk of retinopathy of prematurity. Pediatr. Res. 56, 488.
22
Shastry
105. Kim, J. H., Yu, Y. S., Kim, J. and Park, S. S. (2002) Mutation of the Norrie disease gene in Korean ROP infants. Korean J. Ophthalmol. 16, 93–96. 106. Dhingra, S., Shears, D. J., Blake, V., Stewart, H. and Patel, C. K. (2006) Advanced bilateral persistent fetal vasculature associated with a novel mutation in the Norrie gene. Br. J. Ophthalmol. 90, 1324–1325. 107. Black, G. C. M., Perveen, R., Bonshek, R., Cahill, M., Clayton-Smith, J., Lloyd, C. and McLeod, D. (1999) Coats’ disease of the retina (unilateral retinal telangiectasis) caused by somatic mutation in the NDP gene: a role for norrin in retinal angiogenesis. Hum. Mol. Genet. 8, 2031–2035. 108. Takahashi-Yanaga, F. and Sasaguri, T. (2007) The Wnt/b-catenin signaling pathway as a target in drug discovery. J. Pharmacol. Sci. 104, 293–302. 109. Moon, R. T., Kohn, A. D., DeFerrari, G. V. and Kaykas, A. (2004) Wnt and b-catenin signaling: diseases and therapies. Nat. Rev. Genet. 5, 691–701. 110. Smallwood, P. M., Williams, J., Xu, Q., Leahy, D. J. and Nathans, J. (2007) Mutation analysis of norrin-frizzled 4 recognition. J. Biol. Chem. 282, 4057–4068. 111. Chen, J.-M., Ferec, C. and Cooper, D. N. (2006) A systematic analysis of disease-associated variants in the 3’ regulatory regions of human protein coding genes: general principles and overview. Hum. Genet. 120, 1–21.
112. Wong, P. M. C., Yuan, Q., Chen, H., Sultzer, B. M. and Chung, S.-W. (2001) A single point mutation at the 3’ untranslated region of Ran mRNA leads to profound changes in lipopolysaccharide endotoxin mediated responses. J. Biol. Chem. 276, 33129–33138. 113. Zhang, W., Duan, S., Kistner, E. O., Bleibel, W. K., Huang, R. S., Clark, T. A., Chen, T. X., Schweitzer, A. C., Blume, J. E., Cox, N. J. and Dolan, M. E. (2008) Evaluation of genetic variation contributing to differences in gene expression between populations. Am. J. Hum. Genet. 82, 631–640. 114. Qin, M., Kondo, H., Tahira, T. and Hayashi, K. (2008) Moderate reduction of norrin signaling activity associated with the causative missense mutations identified in patients with familial exudative vitreoretinopathy. Hum. Genet. 122, 615–623. 115. Kato, M., Patel, M. S., Levasseur, R., Lobov, I., Chang, B.-H. J., Glass, D. A., Hartmann, C., Li, L., Hwang, T.-H., Brayton, C. F., Lang, R. A., Karsenty, G. and Chan, L. (2002) Cbfa1 – independent decrease in osteoblast proliferation, osteopenia and persistent embryonic eye vascularization in mice deficient in LRP5, a Wnt coreceptor. J. Cell Biol. 157, 303–314. 116. Xia, C. H., Liu, H., Cheung, D., Wang, M., Cheng, C., Du, X., Chang, B., Beutler, B. and Gong, X. (2008) A model for familial exudative vitreoretinopathy caused by LRP5 mutations. Hum. Mol. Genet. 17, 1605–1612.
Chapter 2 Silent (Synonymous) SNPs: Should We Care About Them? Ryan Hunt, Zuben E. Sauna, Suresh V. Ambudkar, Michael M. Gottesman, and Chava Kimchi-Sarfaty Abstract One of the surprising findings of the Human Genome Project was that single nucleotide polymorphisms (SNPs), which, by definition, have a minor allele frequency greater than 1%, occur at higher rates than previously suspected. When occurring in the gene coding regions, SNPs can be synonymous (i.e., not causing a change in the amino acid) or nonsynonymous (when the amino acid is altered). It has long been assumed that synonymous SNPs are inconsequential, as the primary sequence of the protein is retained. A number of studies have questioned this assumption over the last decade, showing that synonymous mutations are also under evolutionary pressure and they can be implicated in disease. More importantly, several of the mechanisms by which synonymous mutations alter the structure, function, and expression level of proteins are now being elucidated. Studies have demonstrated that synonymous polymorphisms can affect messenger RNA splicing, stability, and structure as well as protein folding. These changes can have a significant effect on the function of proteins, change cellular response to therapeutic targets, and often explain the different responses of individual patients to a certain medication. Key words: Single nucleotide polymorphism, messenger RNA splicing, messenger RNA stability, messenger RNA structure, protein folding, synonymous mutations, nonsynonymous mutations, codon frequency, codon usage.
1. Introduction A large portion of intraspecies phenotypic variation can be attributed to single nucleotide polymorphisms (SNPs), single base changes occurring at a frequency greater than 1% within a population. In humans, for instance, it is estimated that about 90% of sequence variation can be credited to SNPs (1). It is believed that development and progression of cancer, diabetes, cardiovascular disease, psychiatric issues, and other common diseases are dependent on inputs from multiple loci and environmental cues, which A.A. Komar (ed.), Single Nucleotide Polymorphisms, Methods in Molecular Biology 578, DOI 10.1007/978-1-60327-411-1_2, ª Humana Press, a part of Springer Science+Business Media, LLC 2003, 2009
23
24
Hunt et al.
make them markedly more challenging to study than rare, highly penetrant Mendelian diseases (2) (see also Chapter 1 in this volume). Association studies have uncovered SNPs at the basis of many disease traits. The development of robust technologies to detect SNPs promises to bolster the library of these functional SNPs. Even if a disease-associated SNP is not discovered to be the causative agent, it often exists in linkage disequilibrium (see Note 1) with the disease-causing allele, and can lead investigators to the actual functional risk variant, which may be hundreds of kilobases away (1, 3). Thus, SNPs are major contributors to variability seen in drug response and disease susceptibility, rendering them key players in advancing pharmacogenomics and the development of personalized medicine. SNPs exist throughout the entire genome, both within coding regions as well as outside coding sequences. SNPs residing outside coding regions can occur in intergenic sequences, 50 - or 30 -untranslated regions, intronic regions, and associated noncoding regions such as promoters and transcription factor binding sites. Those falling within coding regions can be further categorized into two groups: synonymous, or ‘‘silent,’’ (see Note 2) and nonsynonymous. A coding-region SNP which alters the transcribed codon such that a different amino acid is incorporated into the polypeptide is referred to as a ‘‘nonsynonymous SNP.’’ Owing to wobble transfer RNA (tRNA) base pairing and redundancy in the genetic code, some (synonymous) SNPs within coding regions do not lead to a change in the amino acid incorporated at the site of their occurrence. Much of the lack of appreciation for this class of SNPs derives from faulty nomenclature. The ubiquitously popular term ‘‘silent SNP’’ leads one to the view that such a change in deoxynucleotide has no consequence for molecular, cellular, or physiological phenomena. The use of this terminology can easily mislead many audiences, yet many continue to opt for ‘‘silent SNP,’’ and the term remains prevalent in publications to this date (4, 5). With a growing body of evidence invalidating this presupposition, it is prudent to lay this historic term to rest while promoting terminology that reflects current understanding. Here, we advocate for the use of ‘‘synonymous’’ to categorize those SNPs that do not lead to a change in primary polypeptide sequence. Although the concept of similarity is still present, ‘‘synonymous SNP’’ directs one’s attention to the fact that these single base pair changes require different tRNAs to decode them; tRNAs that are, however, charged with the same amino acid. Moreover, this title avoids the notion that these SNPs are biologically silent or inconsequential. Here, we argue that studying synonymous SNPs is a worthwhile endeavor. We also hope to defeat any notion that synonymous SNPs are trivial, as this belief is flawed on many fronts: messenger RNA (mRNA) splicing, mRNA stability, mRNA structure, protein translation and cotranslational protein folding.
Silent (Synonymous) SNPs
25
2. The Impact of Synonymous SNPs 2.1. mRNA Splicing
The most well documented way in which synonymous SNPs exert their impact on gene function is via perturbations of mRNA splicing. It has been estimated that 92–94% of all multiexonic human genes are alternatively spliced (6). With the modest number of protein-coding genes revealed by the Human Genome Project, creating sufficient proteome diversity is achieved, in part, through alternative splicing. Ensuring the correct relative abundance of transcript variants is critical for correct proteostasis, as protein isoforms can have null, distinct, antagonizing, or agonistic functions (7, 8). Therefore, synonymous SNPs generating ectopic mRNA splicing can bring about distinct phenotypes, including human disease. Schizophrenia is a complex neurological disorder stemming from frontal and temporal lobe dysfunction. Although complicated by the multifactorial nature of the disease, the genetic underpinnings giving rise to schizophrenia are beginning to be understood. Glutamate neurotransmission, a process tightly linked to schizophrenia (among many other neurological disorders), is largely regulated by metabotropic glutamate receptors (GRMs). Activation of group II GRMs, composed of GRM2 and GRM3, moderates glutamate synaptic levels. A synonymous SNP within GRM3 was the first SNP discovered showing strong ties to the development of schizophrenia (9). Until recently, the mechanisms underlying this association have remained unclear and negative findings have added doubt to the validity of this SNP’s schizophrenic ties (10, 11). A splicing variant of GRM3 lacking the fourth exon transmembrane domain, GRM34, was known to be present in both schizophrenic and control patients. Recent investigation into the aforementioned SNP located within the third exon of GRM3 revealed that its presence markedly increased (about 20%) levels of the truncated splicing variant GRM34. However, this transcript was not detected at statistically significant levels in schizophrenic patients (12). The authors did note that the power of their testing was insufficient, and that this synonymous SNP is still likely to interact with other SNPs to generate schizophrenia. Advancing our understanding of psychophysiological phenomena, such as the development of schizophrenia, requires a holistic analysis of biological inputs. This finding provides evidence that SNPs, notably synonymous SNPs, should fall under that realm. Synonymous SNPs can also influence the splicing of precursor mRNA (pre-mRNA) without giving rise to a disease phenotype or predisposition. In fact, one study highlights a synonymous mutation’s ability to silence the effects of a deleterious mutation.
26
Hunt et al.
The medium-chain acyl-CoA dehydrogenase (MCAD) gene encodes for MCAD, and deficiency in MCAD can be brought about by a missense mutation in exon 5. An exonic splicing enhancer, which normally antagonizes an exonic splicing silencer, becomes inactivated in the presence of this mutation, thus leading to nonfunctioning MCAD. While it has no effect on MCAD splicing by itself, the presence of a single synonymous SNP in MCAD exon 5 renders the exonic splicing silencer inactive, thus restoring levels of the functional splice product in the presence of the missense mutation (13). Synonymous SNPs also contribute to the intricateness of infectious disease. There has been long-standing interest in identifying genetic factors giving rise to individual variability in infectious disease susceptibility and progression. The flavivirus West Nile virus is causing a growing global epidemic and has been a point of increasing scientific attention in recent years. In an effort to identify genetic indicators of viral susceptibility, researchers sequenced oligoadenylate synthetase (OAS) genes and ribonuclease L (RNASEL) genes, two groups of innate viral resistance genes previously identified by mouse models, in 33 individuals hospitalized with West Nile virus. Surprisingly, they found no deletions, insertions, or nonsense mutations in any patient. But through analysis of coding-region sequences, they identified a single synonymous polymorphism in the 20 ,50 -oligoadenylate synthetase like (OASL) gene occurring at a significantly high rate in affected individuals. Moreover, because this polymorphism is present within an exonic splice enhancer, it was suggested that the polymorphic transcript undergoes increased splicing, increasing the yield of a truncated OASL product, which may impair innate viral immunity (14). The role of auxiliary elements within exons (exonic splicing enhancers and exonic splicing silencers) in human disease has been the subject of increasing attention in recent years (15). Owing to their capacity to affect mRNA splicing, synonymous SNPs are know to be a part of numerous multifaceted diseases, from multiple sclerosis (16) to autism (17). Synonymous base pair changes in these splicing motif regions can change the splicing patterns of mRNA transcripts directly, or they can alter the penetrance of concurring mutations elsewhere in the gene. 2.2. mRNA Stability
The stability of mRNA is intimately linked to gene expression. The ability to monitor and rapidly adjust the half life of mRNA with specificity is critical at both the cellular and the physiological levels. The factors which determine the stability of a given mRNA fall into two categories: characteristics of the transcript itself (cis factors) and determinants originating outside of the primary transcript (trans factors). The latter group includes elements like growth factors, RNases, RNA binding proteins, ions, dissolved oxygen,
Silent (Synonymous) SNPs
27
and irradiation (for a review see (18)). Here, we direct attention to cis factors, particularly synonymous SNPs and their impact on mRNA stability. While most cis factors that determine mRNA stability fall within the 30 -untranslated region of a transcript, various influences stemming from coding sequences have been identified. At least nine susceptibility loci exist for the chronic skin disorder psoriasis (PSORS1–PSORS9). The CDSN gene exists within the PSORS1 region and encodes corneodesmosin, an extracellular adhesion protein found in the major component of the epidermis, keratinocytes. Because corneodesmosin is often overexpressed in skin lesions of psoriasis patients, one hypothesis conjectured that risk alleles of CDSN predict an mRNA transcript with a heightened half life, thus leading to the overexpression of corneodesmosin. Upon investigation of a risk haplotype, it was revealed that the transcript bearing the haplotype exhibited a twofold increase in mRNA stability. What is more, site-directed mutagenesis revealed that this synonymous SNP was solely responsible for the observed increase in mRNA stability. Further examination revealed that the synonymous SNP altered the transcript’s affinity for a cytoplasmic RNA binding protein (19). ATP-binding-cassette (ABC) transporters comprise a family of integral membrane proteins that function in the ATP-dependent efflux of numerous compounds from cells (20). Studies of ABCC2 (mutations in this gene have been observed in patients with Dubin–Johnson syndrome) revealed that a synonymous SNP within this peptide may lead to increased mRNA stability and concurrent high levels of ABCC2 expression. In accord with these findings, individuals heterozygous for this SNP display distinct pharmacokinetics when administered the cholesterolreducing drug pravastatin, a substrate of ABCC2 (21). The six known synonymous SNPs of the human dopamine receptor D2 (DRD2) were investigated for their possible functional consequences. One SNP was determined to decrease mRNA stability and translation. Another, while having no detectable effects on the receptor by itself, voided the influence of the former synonymous SNP in haplotype analysis. This study, while adding to the evidence that synonymous SNPs impact mRNA stability, reminds us that the additive effects of two synonymous SNPs cannot be foreseen by assessment of the function of each SNP individually. Ultimately, these cases demonstrate that gene expression can be significantly impacted by synonymous SNPs via perturbations of mRNA stability. While SNPs within regulatory regions of 50 - and 30 -untranslated regions, on a global view, manipulate mRNA stability to a greater extent, synonymous SNPs should still remain a keen point of interest in mRNA stability studies.
28
Hunt et al.
2.3. mRNA Structure
In conjunction with determinants stemming from the cellular environment, the structure of any mRNA is largely determined by its primary nucleotide sequence. It is known that mRNA structure is integral to a diverse array of biological processes: splicing characteristics and processing of pre-mRNA, translation control and rhythm, and other regulatory functions dependent on specific mRNA structure. The impact of synonymous SNPs on mRNA structure is well documented. The minor allelic forms of alanyltRNA synthetase and replication protein A both contain a single synonymous SNP. In 1999, a study into these two minor alleles revealed marked differences in mRNA secondary structure relative to their major allele counterparts. Distinct nuclease S1 digestion patterns were observed for both polymorphic mRNA transcripts. Additional enzymatic probing showed different protection characteristics achieved using nuclease-resistant phosphorothioate oligonucleotides targeted to regions flanking the SNP site, further highlighting the synonymous-SNP-mediated changes in mRNA structure (22). While cellular or physiological implications were not explored in this instance, more recent work on polymorphisms of the cathechol-O-methyltransferase (COMT) gene has successfully demonstrated that synonymous SNPs can modulate mRNA structure and have downstream effects on protein expression and phenotype. COMT inactivates catecholamine neurotransmitters and its genetic variants show strong associations to individual variation in pain perception (23). Until recently, the molecular mechanisms by which one COMT haplotype conferred high sensitivity to experimental pain, while another desensitized pain sensation remained unclear. Prior research had focused on a nonsynonymous SNP that showed reduced enzymatic activity, but the results of this work were often inconsistent. However, when this common SNP was analyzed in concurrence with two other synonymous SNPs, the correlation between haplotype and pain sensitivity became more obvious and a causative molecular mediator emerged: mRNA structure. Specifically, the haplotype linked to high pain responsiveness (HPS) transcribed the most stable stem loop structure in the region containing the SNPs on the mRNA transcript. Conversely, individuals bearing the low pain responsiveness (LPS) haplotype produced a local stem loop structure with the highest Gibbs free energy. Expression levels were inversely related to the folding potentials, as the HPS haplotype exhibited the lowest protein levels and enzymatic activity (24). Moreover, this work revealed that the three major haplotypes of the COMT gene are composed of SNPs that assume an unexpected hierarchy of influence. The nonsynonymous SNP val158met is found in the average pain responsiveness haplotype, but is absent in the HPS and LPS haplotypes. Thus, the synonymous polymorphisms of COMT are the
Silent (Synonymous) SNPs
29
foremost determinants of phenotype. This study demonstrates that haplotypes composed of synonymous SNPs can have profound effects on gene function, and in some cases their effects can be stronger than those of their nonsynonymous counterparts. It was hypothesized that the variation in COMT expression levels among haplotypes was due to differences in protein translation efficiency, which marks the final manner in which synonymous SNPs can exert their influence: protein translation and cotranslational protein folding. 2.4. Protein Folding
The ways in which mRNA splicing, structure, and stability are perturbed by the presence of synonymous SNPs is relatively well established. Recent work has begun to shed light on how synonymous SNPs have an influence on protein folding and ultimate protein function. Although the hypothesis that ‘‘silent’’ SNPs impact protein folding characteristics has been widespread for some time, hard evidence validating this suggestion has been hard to come by until recently. This idea directly challenges two concepts that have dominated thinking about the role (or lack thereof) of synonymous mutations in biology. The first arises from Anfinsen’s principle (25), which holds that the amino acid sequence of a protein alone determines the threedimensional structure of that protein. Therefore, it stands to reason that mutations which do not alter amino acid residues would not affect the tertiary structure of the protein or its function. The second is the application of this idea to evolutionary theory. Mutations in the genome are selected for on the basis of the effect on the fitness of the organism. Consequently, synonymous mutations have been thought of as being evolutionarily silent (26). Computational and experimental studies in recent years suggest that neither of these perceptions may be entirely true. There is now mounting evidence for selection against synonymous mutations in several organisms (for reviews see (15, 27)). A very comprehensive recent study (28) used a combination of data mining and computer simulation to determine the selection pressures on synonymous mutations in six model organisms (Escherichia coli, Saccharomyces cerevisiae, Caenorhabditis elegans, Drosophila melanogaster, Mus musculus, and Homo sapiens) belonging to distinct taxonomic groups. What sets this study apart from earlier studies is the taxonomic range of the organisms studied and the fact that the authors simultaneously assessed a set of ten independent parameters. On the basis of principal component analysis, translational fidelity is the single underlying phenomenon that explains correlations with evolutionary rate. Moreover, the view that the amino acid sequence of a protein alone determines the three-dimensional structure has also been challenged over the last two decades (29–31).
30
Hunt et al.
A central precept in protein folding has been that the native conformation of a protein represents the free-energy minimum, i.e., protein folding is under thermodynamic control. In 1987 Alistair Brown and colleagues made a theoretical case for the kinetic control of protein folding (29). They suggested that rare codons could provide translational pauses which may be vital for correct protein folding. In addition, subsequent studies have shown that synonymous mutations may alter RNA secondary structure, which in turn influences the rate of translation (32, 33). Translational pauses at critical sites on the protein permit individual domains of a protein to fold correctly by curtailing disruptive interactions between unfolded regions of the protein. Experimental validation of these ideas came from the laboratory of Anton Komar (30). Consecutive rare codons were substituted with synonymous frequent codons in chloramphenicol acetyltransferase. This resulted in the enzyme having 20% lower specific activity, which was attributed to a change in the kinetics of protein folding. These studies were carried out using an in vitro translation system, raising the question: is this phenomenon physiologically significant? A significant impediment to designing similar studies in vivo is that proteins that are incorrectly folded are eliminated by extremely efficient quality control mechanisms (34). Evidence for the biological significance of the phenomena described in the preceding sections was provided by an analysis of the SNPs in the Multidrug Resistance1 (MDR1) gene (31), the product of which (P-glycoprotein, P-gp) is an ABC transporter. P-gp is an ATP-driven efflux pump that extrudes numerous compounds from cells, including chemotherapeutic agents, and has been implicated in the multidrug resistance of many human cancers (for a review see (35)). A large proportion of several populations (e.g., Chinese, Indian, and Malay) carry a haplotype of three SNPs: C1236T-G2677T-C3435T (36, 37). Of these three, G2677T is a nonsynonymous change, while the other two are synonymous. To understand the functional significance of this haplotype that might confer a selective advantage, the three SNPs individually and as a haplotype were studied (31). There was a significant reduction in the extent of reversal of transport by the P-gp modulators verapamil and cyclosporine A in the haplotype, but not in the individual SNPs. Thus, this change was not a consequence of the single nonsynonymous SNP. The altered transport implicated a structural change which was demonstrated using the conformation-sensitive monoclonal antibodies UIC2, and trypsin digestion patterns supported such a conclusion. Changes in the stability of the mRNA as well as splice variants could also have explained these results. However, a fulllength mRNA was found and its levels, P-gp levels, and localization were unchanged in the haplotype. This study thus suggested proteins with identical primary sequences could nonetheless exhibit different conformations leading to a change in function.
Silent (Synonymous) SNPs
31
The hypothesis proposed by Brown (see above and (29)) provides a simple framework to understand these observations, and a more detailed hypothesis has recently been advanced (38, 39). A complex protein is composed of several structural motifs, not all of which have the same folding kinetics. Thus, for example, a-helices are ultrafast folders (39) and fold much faster than, say, b-sheets. Studies have also shown that proteins fold while being translated (39–43) and that ‘‘cotranslational folding’’ leads to intrapeptide contacts during folding different from those that occur during refolding of the covalently intact but unfolded polypeptide (39, 43–45). Furthermore, as has been stated elsewhere, the rate of translation is not uniform and is often controlled by codon bias, mRNA structure, etc. (46). The rate of translation affects protein conformation most significantly when translation occurs between different structural motifs in a multidomain protein. If these domains are composed of fast folders, a change in the rate of translation is unlikely to have a significant effect. If, however, a slow folding domain follows a fast folding domain (see, e.g., Fig. 2 in (38)) the slow folder is likely to have competing conformations, one of which may be ‘‘stabilized’’ by the fast folder. However, ribosome stalling could permit the slow folder to be completely folded prior to formation of the subsequent motifs. Thus, the protein could potentially follow alternative branches in a folding pathway, leading to minor conformational changes. In retrospect, several distinctive properties of P-gp make it an ideal experimental system to experimentally observe alterations in protein conformation as a consequence of synonymous SNPs that alter the rate of translation. Multidrug transporters utilize a single large cavity comprising a ‘‘drug-binding pocket’’ where individual ‘‘drug-binding sites’’ are generated by subtle alterations in the accessibility of different subset(s) of residues for drug-binding (47). Also, multiple drug resistance proteins can alter substrate specificity fairly easily and even subtle changes in the conformation of the drug-binding pocket can result in changes in substrate specificity (see (35) for examples). Also, the absence of a single distinctive correctly folded state of the protein could potentially allow the protein to escape the quality control machinery of the cell. Although evidence from P-gp suggests that these SNPs predominantly exert their influence though an altered rate in translation and cotranslational folding, a holistic investigation, assessing these phenomena and those mentioned previously, is necessary to reveal the functional consequences of any synonymous SNP.
3. Frequency of SNPs Table 2.1 shows that of the 896,454 SNPs documented to date in the human genome (all numbers are from NCBI’s SNP
3701
29,049
1594
4416
31,706
3329
75
60
65
78
1735
75,808
A,C
A,G
A,T
C,G
C,T
G,T
A/C/G
A/C/T
A/G/T
C/G/T
All others (including deletions)
Total
107,748
13,542
160
146
114
220
9018
28,366
11,620
4924
30,783
8855
Coding nonsynonymous
6,411,225
1,328,009
2444
2047
2166
2485
440,620
1,692,412
447,120
360,280
1,694,999
438,464
Intron
a
mRNA UTR messenger RNA untranslated region The change in nucleotide, which could occur in either direction
Coding synonymous
Nucleotide changea
1077
300
0
3
0
4
98
326
68
40
176
62
Splice site
182,361
34,160
135
86
108
130
13,170
48,426
14,732
9361
48,035
13,992
mRNA UTR
360,737
74,582
255
193
191
247
25,745
93,000
30,048
18,367
91,977
26,104
Flanking 2000 up/ 500 down
10,757,498
1,806,707
5806
5556
5570
5885
810,560
2,929,112
753,613
695,762
2,928,878
810,028
Outside coding regions
17,896,454
3,259,035
8878
8096
8209
9046
1,302,540
4,823,348
1,261,617
1,090,328
4,823,897
1,301,206
Total
Table 2.1 Frequency of various types of single nucleotide polymorphisms (generated using the NCBI database; total number of cases are presented)
32 Hunt et al.
Silent (Synonymous) SNPs
33
database, reference human genome), about 60% appear on regions outside the coding regions. Of the remaining ones, about 91% of SNPs are in introns, and within the coding regions about 1.5% are nonsynonymous changes, while about 1.1% are synonymous changes. Interestingly, the common changes are cytosine to thymine (or thymine to cytosine) and adenine to guanine (or guanine to adenine), which are transitions and are expected to occur at higher frequencies (48). An interesting feature of this analysis is that when we consider the coding region of the genome alone, the frequencies of synonymous and nonsynonymous SNPs are comparable. It has often been argued that as the synonymous mutations are ‘‘silent’’ these would show higher rates of mutations; however, this does not appear to be so, at least within the coding regions of the human genome. Moreover the SNPs showing transitional substitutions appear at about twice the rate among nonsynonymous SNPs compared with synonymous SNPs. We endeavored to determine whether this could be explained on the basis of the frequency of each nucleotide in each position within a codon. Table 2.2 describes the frequency of each nucleotide within the codon. The nucleotides guanine, cytosine, and adenine have about the same frequency, but thymine is significantly less used overall. The frequency is given as a number out of 1,000 codons within the coding region (based on http://www.kazusa.or.jp/codon/cgi-bin/ showcodon.cgi?species=9606&aa=1&style=GCG).
Table 2.2 Frequency of each nucleotide within the codon Nucleotide
1st position
2nd position
3rd position
Total
A
268
310
192
770
C
246
233
298
777
G
311
192
287
790
T
175
265
222
662
Although guanine is the most common base, it is significantly the rarest in the second position and although thymine is significantly less used overall, it is the second most frequent in the second position. Adenine is significantly rarer in the third position among the four bases. However, as synonymous codons most often differ only in their third position, we analyzed the codon frequency from a different perspective with different conclusions. To generate a specific amino acid, there is a need to use one of its codons. Therefore, the frequency of each nucleotide in each codon
34
Hunt et al.
position depends primarily on the frequency of the amino acids in the genome. It may well be that the low frequency of thymine in the first position derives directly from the protein sequence. Since synonymous codons most often differ in their third position, it is this position that can be subject to selection on nucleotide preference. However, not all amino acids exhibit a redundancy in codon usage. When we examine only those amino acids that harbor synonymous codons with 3rd position changes, the frequency of adenine is raised from about 19% (as appears in Table 2.2) to about 25% (Table 2.3).
Table 2.3 Frequency of each nucleotide within the third codon position, when the change does not alter the amino acid (i.e., synonymous substitution) Nucleotide
A
C
G
T
Number of appearances within the 3rd position
7,766,153
7,769,044
10,237,953
5,366,408
Percentage
25
25
33
17
Comparing the data in Tables 2.2 and 2.3 demonstrates that guanine appears at a higher frequency in the 3rd position than the other nucleotides and thymine has lower frequency, in situations where more than one codon codes for the same amino acid. These results show that the frequency of nucleotide change in SNPs cannot be attributed to the statistical probability of a particular base pair occurring more frequently at a particular place. It is more likely that natural selection determines the frequency with which the nucleotide changes (depicted in Table 2.1) occur.
4. Concluding Remarks SNPs contribute significantly to genetic variation among humans—on average, one SNP exists per 1.3 kb of human genomic DNA (49). Unlike the onset and progression of rare Mendelian diseases (see Note 3), those of common diseases may be influenced substantially by a host of common alleles within the human population (50, 51). Individual variation in drug response is another complex, polygenic phenomenon that arises from common allele determinants. Thus, SNPs are a critical part of overcoming the alarming number of deaths due to fatal drug responses seen in hospitalized patients (52).
Silent (Synonymous) SNPs
35
There is a consensus that nonsynonymous SNPs and SNPs within regulatory regions have a greater propensity, relative to their synonymous SNP counterparts, to affect gene behavior when analyzed in isolation, i.e., on a wild-type background (48). This understanding is not disputable and is unlikely to be overturned as we continue to document functional SNPs and their cellular implications. Moreover, it is likely that the majority of synonymous SNPs within the human genome do not produce a significant change in gene expression or gene product function, which may account for the widespread reluctance to invest time in studying them. But overlooking synonymous SNPs would be a costly mistake in the quest to see genetics at the forefront of unraveling human disease. Even if only a small percentage of synonymous SNPs generate functional consequences, with the large number of synonymous SNPs across the human genome, these SNPs undoubtedly warrant our attention. While most changes attributed to synonymous SNPs are only subtle, when they are considered in concert with other SNPs and environmental inputs they may become critically important. With that said, instances also abound in which synonymous SNPs act alone to produce phenotypic variation. As we have argued, a growing body of evidence is revealing that synonymous SNPs can indeed perturb cellular functions and elicit distinct clinical phenotypes. Synonymous SNPs falling within splicing consensus sequences can perturb the normal splicing characteristics of a gene. Synonymous SNPs, often through a change in mRNA structure, can alter the stability of a transcript by changing the ability of RNA binding proteins to recognize the transcript. The secondary structure of mRNA is integral to splicing and processing of pre-mRNA, as well as translation control and rhythm. Therefore, a synonymous SNP altering mRNA structure can influence a host of cellular functions. Finally, these SNPs affect protein translation and cotranslational protein folding, a concept not easily accepted, as it invalidates Anfinsen’s principle and demonstrates that a simple thermodynamic model for protein folding is too simplistic. It is debated whether codon usage drives tRNA expression or vice versa, with many recent investigations showing that they coevolve (53). Regardless of the cause, we know organisms show a preference for certain codons/tRNAs over others, with expression profiles mirroring these biases. As a result, the rate at which an amino acid is incorporated into a growing peptide can change significantly from one synonymous codon to the next. During translation, some proteins require programmed translational pauses at specific sites to ensure correct cotranslational folding, which can be accomplished through a stretch of rare codons. Synonymous polymorphisms can alter conventional protein folding kinetics and can lead to a protein product of distinct conformation and activity.
36
Hunt et al.
The widespread belief that little or no selective pressure exists on synonymous SNPs has prejudiced our thinking for quite some time. A case in point lies in a common method to assess the evolutionary pressure acting on protein-coding sequences: investigators often calculate the ratio of the rate of nonsynonymous substitutions (Ka) to the rate of synonymous substitutions (Ks) (54). While Ka/Ks ratios may be applicable for a majority of coding sequences, the value of this ratio breaks down in areas containing functional synonymous SNPs, as the calculation relies on the assumption that synonymous substitutions are evolutionarily neutral. As we have contended throughout this chapter, the idea that synonymous SNPs are exempt from selective pressure is inconsistent with numerous findings. Some have even noted extraordinary synonymous codon conservation in mammalian coding regions, where synonymous preservation takes precedence over nonsynonymous substitutions (55). We have presented strong evidence stressing the importance of synonymous SNPs in a span of biological phenomena. With the rapidly growing list of synonymous polymorphisms associated with human diseases (for a review see (56)), these SNPs cannot be ignored by ascribing to them the term ‘‘silent.’’ Rather, synonymous SNPs should be the recipients of increasing attention, especially as we continue to advance novel techniques with which to detect their presence and assay their potential role in human disease and personalized medicine.
5. Notes
1. Linkage disequilibrium stands for the nonrandom association between two or more alleles (haplotype) in a population, such that combinations of these alleles are more likely to occur together than would be expected by chance. Linkage disequilibrium analysis is a useful method in genetic mapping studies. 2. Owing to the fact that the primary polypeptide sequence appears unchanged, synonymous SNPs have garnered the label ‘‘silent’’ under the belief that their presence is inconsequential to the gene and its corresponding product. Recent evidence shows that this is not really the case and the term ‘‘silent’’ should be generally avoided when referring to synonymous SNPs. 3. Mendelian disease is a popular/common term for a genetic disorder which follows simple Mendelian patterns of inheritance and results from a mutation at a single locus. Most
Silent (Synonymous) SNPs
37
Mendelian diseases (e.g., cystic fibrosis) are rare, but the frequency may vary depending on the genetic background of a particular population.
Acknowledgments This research was supported, in part, by the Intramural Research Program of the National Institutes of Health, National Cancer Institute. Special thanks are expressed to George Leiman, NCI, NIH, and Geetha S., CBER, FDA, for editorial assistance. References 1. Collins, F. S., Brooks, L. D. and Chakravarti, A. (1998) A DNA polymorphism discovery resource for research on human genetic variation. Genome Res. 8, 1229–1231. 2. Glazier, A. M., Nadeau, J. H. and Aitman, T. J. (2002) Finding genes that underlie complex traits. Science 298, 2345–2349. 3. Goldstein, D. B. and Weale, M. E. (2001) Population genomics: Linkage disequilibrium holds the key. Curr. Biol. 11, R576–579. 4. Gumus-Akay, G., Rustemoglu, A., Karadag, A. and Sunguroglu, A. (2008) Genotype and allele frequencies of MDR1 gene C1236T polymorphism in a Turkish population. Genet. Mol. Res. 7, 1193–1199. 5. Sauvage, C., Bierne, N., Lapegue, S. and Boudry, P. (2007) Single nucleotide polymorphisms and their relationship to codon usage bias in the pacific oyster crassostrea gigas. Gene 406, 13–22. 6. Wang, E. T., Sandberg, R., Luo, S., Khrebtukova, I., Zhang, L., Mayr, C., Kingsmore, S. F., Schroth, G. P. and Burge, C. B. (2008) Alternative isoform regulation in human tissue transcriptomes. Nature 456, 470–476. 7. Hart, M. C. and Cooper, J. A. (1999) Vertebrate isoforms of actin capping protein beta have distinct functions in vivo. J. Cell. Biol. 147, 1287–1298. 8. Xing, Y., Xu, Q. and Lee, C. (2003) Widespread production of novel soluble protein isoforms by alternative splicing removal of transmembrane anchoring domains. FEBS Lett. 555, 572–578. 9. Egan, M. F., Straub, R. E., Goldberg, T. E., Yakub, I., Callicott, J. H., Hariri, A. R.,
10.
11.
12.
13.
Mattay, V. S., Bertolino, A., Hyde, T. M., Shannon-Weickert, C., Akil, M., Crook, J., Vakkalanka, R. K., Balkissoon, R., Gibbs, R. A., Kleinman, J. E. and Weinberger, D. R. (2004) Variation in GRM3 affects cognition, prefrontal glutamate, and risk for schizophrenia. Proc. Natl. Acad. Sci. U.S.A. 101, 12604–12609. Marti, S. B., Cichon, S., Propping, P. and Nothen, M. (2002) Metabotropic glutamate receptor 3 (GRM3) gene variation is not associated with schizophrenia or bipolar affective disorder in the German population. Am. J. Med. Genet. 114, 46–50. Norton, N., Williams, H. J., Dwyer, S., Ivanov, D., Preece, A. C., Gerrish, A., Williams, N. M., Yerassimou, P., Zammit, S., O’Donovan, M. C. and Owen, M. J. (2005) No evidence for association between polymorphisms in GRM3 and schizophrenia. BMC Psychiatry 5, 23. Sartorius, L. J., Weinberger, D. R., Hyde, T. M., Harrison, P. J., Kleinman, J. E., and Lipska, B. K. (2008) Expression of a GRM3 splice variant is increased in the dorsolateral prefrontal cortex of individuals carrying a schizophrenia risk SNP. Neuropsychopharmacology 33, 2626–2634. Nielsen, K. B., Sorensen, S., Cartegni, L., Corydon, T. J., Doktor, T. K., Schroeder, L. D., Reinert, L. S., Elpeleg, O., Krainer, A. R., Gregersen, N., Kjems, J. and Andresen, B. S. (2007) Seemingly neutral polymorphic variants may confer immunity to splicing-inactivating mutations: A synonymous SNP in exon 5 of MCAD protects from deleterious mutations in a flanking exonic splicing enhancer. Am. J. Hum. Genet. 80, 416–432.
38
Hunt et al.
14. Yakub, I., Lillibridge, K. M., Moran, A., Gonzalez, O. Y., Belmont, J., Gibbs, R. A. and Tweardy, D. J. (2005) Single nucleotide polymorphisms in genes for 2’-5’-oligoadenylate synthetase and RNAse L in patients hospitalized with West Nile virus infection. J. Infect. Dis. 192, 1741–1748. 15. Cartegni, L., Chew, S. L. and Krainer, A. R. (2002) Listening to silence and understanding nonsense: Exonic mutations that affect splicing. Nat. Rev. Genet. 3, 285–298. 16. Fedetz, M., Matesanz, F., Caro-Maldonado, A., Fernandez, O., Tamayo, J. A., Guerrero, M., Delgado, C., Lopez-Guerrero, J. A. and Alcina, A. (2006) OAS1 gene haplotype confers susceptibility to multiple sclerosis. Tissue Antigens 68, 446–449. 17. Solis-Anez, E., Delgado-Luengo, W., BorjasFuentes, L., Zabala, W., Arraiz, N., Pineda, L., Portillo, M. G., Gonzalez-Ferrer, S., Chacin, J. A., Pena, J., Montiel, C., Morales, A., Rojas de Atencio, A., Canizales, J., Gonzalez, R., Miranda, L. E., Abreu, N., and Delgado, J. (2007) [Molecular analysis of the GABRB3 gene in autistic patients: An exploratory study]. Invest Clin. 48, 225–242. 18. Ross, J. (1995) mRNA stability in mammalian cells. Microbiol. Rev. 59, 423–450. 19. Capon, F., Allen, M. H., Ameen, M., Burden, A. D., Tillman, D., Barker, J. N. and Trembath, R. C. (2004) A synonymous SNP of the corneodesmosin gene leads to increased mRNA stability and demonstrates association with psoriasis across diverse ethnic groups. Hum. Mol. Genet. 13, 2361–2368. 20. Jones, P. M. and George, A. M. (2004) The abc transporter structure and mechanism: Perspectives on recent research. Cell. Mol. Life. Sci. 61, 682–699. 21. Niemi, M., Arnold, K. A., Backman, J. T., Pasanen, M. K., Godtel-Armbrust, U., Wojnowski, L., Zanger, U. M., Neuvonen, P. J., Eichelbaum, M., Kivisto, K. T. and Lang, T. (2006) Association of genetic polymorphism in ABCC2 with hepatic multidrug resistance-associated protein 2 expression and pravastatin pharmacokinetics. Pharmacogenet. Genomics 16, 801–808. 22. Shen, L. X., Basilion, J. P. and Stanton, V. P., Jr. (1999) Single-nucleotide polymorphisms can cause different structural folds of mRNA. Proc. Natl. Acad. Sci. U.S.A. 96, 7871–7876. 23. Diatchenko, L., Slade, G. D., Nackley, A. G., Bhalang, K., Sigurdsson, A., Belfer, I., Goldman, D., Xu, K., Shabalina, S. A., Shagin, D., Max, M. B., Makarov, S. S., and Maixner, W. (2005) Genetic basis for individual variations
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
in pain perception and the development of a chronic pain condition. Hum. Mol. Genet. 14, 135–143. Nackley, A. G., Shabalina, S. A., Tchivileva, I. E., Satterfield, K., Korchynskyi, O., Makarov, S. S., Maixner, W. and Diatchenko, L. (2006) Human catechol-omethyltransferase haplotypes modulate protein expression by altering mRNA secondary structure. Science 314, 1930–1933. Anfinsen, C. B. (1973) Principles that govern the folding of protein chains. Science 181, 223–230. Kimura, M. (1977) Preponderance of synonymous changes as evidence for the neutral theory of molecular evolution. Nature 267, 275–276. Chamary, J. V., Parmley, J. L. and Hurst, L. D. (2006) Hearing silence: Non-neutral evolution at synonymous sites in mammals. Nat. Rev. Genet. 7, 98–108. Drummond, D. A. and Wilke, C. O. (2008) Mistranslation-induced protein misfolding as a dominant constraint on codingsequence evolution. Cell 134, 341–352. Purvis, I. J., Bettany, A. J., Santiago, T. C., Coggins, J. R., Duncan, K., Eason, R. and Brown, A. J. (1987) The efficiency of folding of some proteins is increased by controlled rates of translation in vivo. A hypothesis. J. Mol. Biol. 193, 413–417. Komar, A. A., Lesnik, T. and Reiss, C. (1999) Synonymous codon substitutions affect ribosome traffic and protein folding during in vitro translation. FEBS Lett. 462, 387–391. Kimchi-Sarfaty, C., Oh, J. M., Kim, I. W., Sauna, Z. E., Calcagno, A. M., Ambudkar, S. V. and Gottesman, M. M. (2007) A ‘‘Silent’’ Polymorphism in the MDR1 gene changes substrate specificity. Science 315, 525–528. Ivanov, I. G., Saraffova, A. A. and Abouhaidar, M. G. (1997) Unusual effect of clusters of rare arginine (AGG) codons on the expression of human interferon alpha 1 gene in Escherichia coli. Int. J. Biochem. Cell. Biol. 29, 659–666. Parmley, J. L. and Hurst, L. D. (2007) How do synonymous mutations affect fitness? Bioessays 29, 515–519. Bukau, B., Weissman, J. and Horwich, A. (2006) Molecular chaperones and protein quality control. Cell 125, 443–451. Ambudkar, S. V., Dey, S., Hrycyna, C. A., Ramachandra, M., Pastan, I. and Gottesman, M. M. (1999) Biochemical, cellular, and pharmacological aspects of the multidrug
Silent (Synonymous) SNPs
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
transporter. Annu. Rev. Pharmacol. Toxicol. 39, 361–398. Pauli-Magnus, C. and Kroetz, D. L. (2004) Functional implications of genetic polymorphisms in the multidrug resistance gene MDR1 (ABCB1). Pharm. Res. 21, 904–913. Kimchi-Sarfaty, C., Marple, A. H., Shinar, S., Kimchi, A. M., Scavo, D., Roma, M. I., Kim, I. W., Jones, A., Arora, M., Gribar, J., Gurwitz, D., and Gottesman, M. M. (2007) Ethnicity-related polymorphisms and haplotypes in the human ABCB1 gene. Pharmacogenomics 8, 29–39. Tsai, C. J., Sauna, Z. E., Kimchi-Sarfaty, C., Ambudkar, S. V., Gottesman, M. M. and Nussinov, R. (2008) Synonymous mutations and ribosome stalling can lead to altered folding pathways and distinct minima. J. Mol. Biol. 383, 281–291. Komar, A. A. (2009) A pause for thought along the co-translational folding pathway. Trends Biochem. Sci. 34, 16–24. Clarke, D. T., Doig, A. J., Stapley, B. J. and Jones, G. R. (1999) The alpha-helix folds on the millisecond time scale. Proc. Natl. Acad. Sci. U.S.A. 96, 7232–7237. Kiho, Y. and Rich, A. (1964) Induced enzyme formed on bacterial polyribosomes. Proc. Natl. Acad. Sci. U.S.A. 51, 111–118. Fedorov, A. N. and Baldwin, T. O. (1995) Contribution of cotranslational folding to the rate of formation of native protein structure. Proc. Natl. Acad. Sci. U.S.A. 92, 1227–1231. Fedorov, A. N. and Baldwin, T. O. (1997) Cotranslational protein folding. J. Biol. Chem. 272, 32715–32718. Batey, S., Scott, K. A. and Clarke, J. (2006) Complex folding kinetics of a multidomain protein. Biophys. J. 90, 2120–2130. Kowarik, M., Kung, S., Martoglio, B. and Helenius, A. (2002) Protein folding during cotranslational translocation in the endoplasmic reticulum. Mol. Cell. 10, 769–778. Sauna, Z. E., Kimchi-Sarfaty, C., Ambudkar, S. V. and Gottesman, M. M. (2007) The sounds of silence: Synonymous
47.
48.
49.
50.
51.
52.
53.
54.
55.
56.
39
mutations affect function. Pharmacogenomics 8, 527–532. Schumacher, M. A. and Brennan, R. G. (2003) Deciphering the molecular basis of multidrug recognition: Crystal structures of the staphylococcus aureus multidrug binding transcription regulator QacR. Res. Microbiol. 154, 69–77. Keller, I., Bensasson, D. and Nichols, R. A. (2007) Transition-transversion bias is not universal: A counter example from grasshopper pseudogenes. PLoS Genet. 3, e22. Cargill, M., Altshuler, D., Ireland, J., Sklar, P., Ardlie, K., Patil, N., Shaw, N., Lane, C. R., Lim, E. P., Kalyanaraman, N., Nemesh, J., Ziaugra, L., Friedland, L., Rolfe, A., Warrington, J., Lipshutz, R., Daley, G. Q. and Lander, E. S. (1999) Characterization of single-nucleotide polymorphisms in coding regions of human genes. Nat. Genet. 22, 231–238. Risch, N. and Merikangas, K. (1996) The future of genetic studies of complex human diseases. Science 273, 1516–1517. Lander, E. S. (1996) The new genomics: Global views of biology. Science 274, 536–539. Lazarou, J., Pomeranz, B. H. and Corey, P. N. (1998) Incidence of adverse drug reactions in hospitalized patients: A meta-analysis of prospective studies. JAMA 279, 1200–1205. Higgs, P. G. and Ran, W. (2008) Coevolution of codon usage and tRNA genes leads to alternative stable states of biased codon usage. Mol. Biol. Evol. 25, 2279–2291. Hurst, L. D. (2002) The ka/ks ratio: Diagnosing the form of sequence evolution. Trends Genet. 18, 486. Schattner, P. and Diekhans, M. (2006) Regions of extreme synonymous codon selection in mammalian genes. Nucleic Acids Res. 34, 1700–1710. Charmary, J. V. and Hurst, L. D. (2009) How Trivial DNA Changes Can Hurt Health. Sci Am. 30, 46–53.
Chapter 3 SNP Databases Christopher Phillips Abstract Researchers interested in obtaining detailed information on SNPs now work in a golden age of online database availability: never has so much data and such a wealth of information been freely accessible for such a substantial proportion of the 18 million single nucleotide polymorphism (SNP) loci currently characterized in the human genome. This chapter describes the major SNP databases available for human genetics studies. Tools and strategies are outlined that can help researchers properly formulate a database query to be able to access the most appropriate information needed for their research aims, including medical or population genetics analysis – an approach that is getting increased attention given the expanding scale of online SNP data. Key words: Single nucleotide polymorphism, database, search, query, National Center for Biotechnology Information, dbSNP Entrez, HapMap.
1. Introduction In silico research as a part of the preparation for a genetics study is now an essential preamble to the choice of genomic regions to analyze and markers to use, the design of genotyping approaches, and the listing of appropriate samples to characterize. This chapter provides a simple guide to the structure and use of the major online SNP databases, adapted to Sections 2 and 3, by linking each database to a particular research planning task: finding sets of single nucleotide polymorphisms (SNPs) that share common characteristics (NCBI Entrez); obtaining detailed information on a SNP locus and collating other genetically relevant data (dbSNP); exploring SNPs in coding regions (SNPper and PupaSuite); performing simple scrutiny of linkage A.A. Komar (ed.), Single Nucleotide Polymorphisms, Methods in Molecular Biology 578, DOI 10.1007/978-1-60327-411-1_3, ª Humana Press, a part of Springer Science+Business Media, LLC 2003, 2009
43
44
Phillips
disequilibrium (LD) block structure and choosing SNP markers to tag chromosome regions (HapMap); and assessing population genetics parameters from online SNP data (Haplotter and SPSmart). Some straightforward, common sense advice is given about Internet browsing (see Notes 1 and 2), processing of SNP data, once obtained, and direct use of generic search engines such as Google – to look across the Web space before focusing on known SNP databases. The latter approach can yield interesting results, but otherwise this chapter assumes the user will go directly to a particular SNP database gateway (see Table 3.1) to initiate a directed search of online data.
Table 3.1 The major online single nucleotide polymorphism (SNP) databases Database
Host organization
Gateway URL for initiating SNP data searches
dbSNP
NCBI
http://www.ncbi.nlm.nih.gov/sites/entrez?db=snp
HapMap
The HapMap Consortium
http://www.hapmap.org/cgi-perl/gbrowse/
Ensembl
EMBL-EBI/Sanger Center
http://www.ensembl.org/Homo_sapiens/index.html
Santa Cruz
University of California, Santa Cruz
http://genome.ucsc.edu/cgi-bin/hgGateway
Perlegen
Perlegen Sciences
http://genome.perlegen.com/browser/index_v2.html
Assays-onDemand
Applera (Applied Biosystems)
https://products.appliedbiosystems.com/ab/en/US/adirect/ab? cmd=ABGTKeywordSearch&catID=600769
SeattleSNPs
US NHLBI (PGA)
http://gvs.gs.washington.edu/GVS/
NCBI National Center for Biotechnology Information, NHLBI, National Heart, Lung, and Blood Institute, PGA Program for Genomic Applications
2. Materials 2.1. The Major SNP Databases
Suggesting a definitive list of the major online SNP databases runs the risk of becoming out of date once done, but dbSNP, the SNP database of the National Center for Biotechnology Information (NCBI), and HapMap head the list in Table 3.1, which is otherwise not intended to indicate an order of size or usefulness. NCBI continues to host by far the most important and comprehensive set of genomic databases available, while the HapMap project has an
SNP Databases
45
ever-closer relationship with dbSNP in collating human SNP data. To summarize a complex and far-reaching project, HapMap was intended to concentrate global resources on the characterization of the variant part of the genome as a natural extension of the work of the original Human Genome Mapping Project in establishing the invariant sequence common to everyone and held in NCBI (1, 2). An important part of the initial work of HapMap was to check the efficiency of dbSNP, i.e., how well did the dbSNP catalogue represent the true extent of SNP variability in humans? This was achieved by resequencing ten ENCODE regions (detailed in Table 3.2 of (1) and at http://www.hapmap.org/ downloads/encode1.html.en) and extrapolating the SNP variability found to the genome as a whole. Two findings emerged from this comparison: firstly the false-negative rate of dbSNP (i.e., how often SNPs were present but not detected) although very low was significant for rare SNPs – loci with allele frequencies around 1% (0.01) or less; secondly the overriding majority of common variation SNPs had been captured by dbSNP or if absent had proxies in the same region in tight correlation and listed by dbSNP. It is
Table 3.2 HapMap study populations for phase I/phase II (rows 1–4) and populations added to phase III (rows 5–11). Many published studies, including those of HapMap (1), merge CHB and JPT to a ‘‘population panel’’ abbreviated to ASN Abbreviation
Samples
Full description
Group
1
YRI
180
Yoruba in Ibadan, Nigeria
African
2
CEU
180
Utah residents with northern and western European ancestry
European
3
CHB
90
Han Chinese in Beijing, China
East Asian (ASN)
4
JPT
90
Japanese in Tokyo, Japan
East Asian (ASN)
5
ASW
90
African ancestry in southwest USA
African
6
CHD
100
Chinese in metropolitan Denver, Colorado, USA
East Asian
7
GIH
100
Gujarati Indians in Houston, Texas, USA
South Asian
8
LWK
100
Luhya in Webuye, Kenya
African
9
MEX
90
Mexican ancestry in Los Angeles, California, USA
Native American
10
MKK
180
Maasai in Kinyawa, Kenya
African
11
TSI
100
Tuscans in Italy
European
46
Phillips
interesting that estimates of false-positive rates in dbSNP (i.e., incorrectly listing a nucleotide position as a SNP) were not detailed by HapMap, indicating that these were negligible and therefore dbSNP had developed very efficient systems for confirming that SNPs were real (see Note 3). In summary, dbSNP has proved to be both a comprehensive and a reliable catalogue of human SNP variability with an efficient system to cross-reference multiple submissions of the same SNPs from centers outside NCBI (see Note 4). Since 2003, HapMap has been the major contributor of SNP data to dbSNP. The other databases listed in Table 3.1 both parallel and feed data into dbSNP, so they either provide an alternative system of browsing and searching the core human genome SNP data (Ensemble and Santa Cruz), or list the SNPs generated by their own independent genotyping initiatives with stand-alone browser systems dedicated to the data they have generated (Perlegen, Assays-on-Demand, and Seattle SNPs). 1. dbSNP. The strength of NCBI lies in the breadth of genomic databases held under the single umbrella. This means that queries to any of the NCBI databases can tap into the relationships that exist between the subject of interest and each of some twenty or more major databases within NCBI. So genetics research involving SNPs is easily set in the context of supporting information that details published studies of the SNP, context sequence of the SNP, gene structure and function (if this is where the SNP is sited), and how the SNP variation is expressed as a phenotype. These data are handled in NCBI by PubMed, GenBank, Gene, and Online Mendelian Inheritance in Man (OMIM) databases, respectively (see Note 5). In addition, NCBI benefits from a unified approach to constructing database queries, so once the user is familiar with the way to query one NCBI database, the same rules will apply to all other queries made. When accessing the most extensive NCBI data, comprising SNP, gene, protein, publications, phenotype, and sequence, one can execute data queries directly from a menu of choices in a global system termed ‘‘Entrez’’ (outlined in detail in Section 3.1). The SNP Entrez system EntrezSNP has a homepage menu listing the principal SNP criteria (http://www.ncbi.nlm.nih.gov/sites/ entrez?db=snp) that help to define a search. This is the main starting point of EntrezSNP and this SNP-focused menu differs from those of other Entrez databases such as EntrezGene and EntrezProtein. Therefore, dbSNP can be accessed in three ways: by using EntrezSNP; by following hyperlinks embedded in other NCBI databases, and by direct access to SNP summary pages, termed ‘‘Cluster Reports’’ – forming the core data page for each locus in dbSNP. A Cluster Report can be thought of as the SNP ‘‘homepage’’ listing a full set of the key parameters in a standardized format.
SNP Databases
47
Reference to a SNP within NCBI, within all other SNP databases, and now, almost universally, in the scientific literature is made using a unique identifier: the rs-number, consisting of a number prefixed with ‘‘rs.’’ As an open database, dbSNP receives submissions from genotyping centers and collates the data into a merged reference set (see Note 4). Since different centers routinely report identical SNPs to dbSNP, the submissions are clustered into reference SNPs (termed ‘‘refSNPs’’) based on genome-wide comparisons of the context sequence submitted. For these reasons the distinction between reference SNPs and submitted SNPs is made by rs and ss, respectively: prefixing a unique SNP identification number with rs, while creating a number for each submission prefixed with ss. All rs-numbers are displayed throughout NCBI as hyperlinks returning the Cluster Report. 2. SNP-related databases in NCBI: PubMed, GenBank, Gene, and OMIM. PubMed is the NCBI bibliographic database that provides the starting point for researchers to assess the current published state of the art in their chosen area of study. Data comprise ten million published articles from about 5,000 peer-review journals. PubMed is by default predominantly text-oriented so it works by matching text recognized in the query to the text in the data records, including key words in the article body text itself. Therefore, to work efficiently the system needs to carefully regulate vocabulary, which is done by a separate database of words used to index PubMed known as MeSH (i.e., medical subheadings) – searchable itself using the search menu at the top left of each NCBI homepage. It can be an important check to clarify the vocabulary relating to a trait or disease of interest before performing PubMed searches by subject. Searches using rs-numbers can be an efficient way to find studies related to a research aim, but note that the habit persists in many publications of identifying SNPs in genes by the amino acid substitution they create (e.g., shorthand such as MC1R V60L), so these will not be returned from queries (see Note 6). GenBank is the nucleotide sequence database of NCBI. This simple description belies the scale of the information held – a collection of sequences comprising 60 gigabases of data from more than 130,000 species updated daily. Despite the complexity of GenBank, most users interested in SNP analysis will simply require a specific context sequence segment of about 120–200 nucleotides to design a genotyping assay for the SNPs of interest. As explained later the
48
Phillips
application of RepeatMasker and a neighbor SNP scan of –100 bases of context sequence makes it more advantageous to collect it directly from the SNP Cluster Report. Gene is the gene catalogue of NCBI and like dbSNP presents a single summary page format of relevant information for coding regions, including function summaries, transcription structure, genome maps, bibliography, protein data, sequences, and related links to supporting data. NCBI uses the near-standard gene identifiers that take the form of the letter/number combination standardized by the Human Genome Organization (HUGO) (http://www.gene.ucl.ac. uk/nomenclature/) or throughout NCBI by a GeneID number (see Note 7). Working with data that include a large proportion of text-based information can be difficult, so to view the context of a SNP or list of SNPs sited in genes it may be preferable to use a purely graphical display of SNP positions aligned with intronic, exonic, and 50 /30 -flanking region sequences such as that given by SNPper (see Section 3.3.1). OMIM taking text-based data even further towards an article format, the OMIM database has summary pages written as articles describing a phenotype, trait, or disorder with a known or suspected genetic basis. As such, both Gene and OMIM are best consulted along with PubMed during the initial stages of a study design to gain an overview of the current understanding of a disease process. Luckily, OMIM is highly readable and can be described as an online textbook expanded and updated as knowledge of a trait or condition is consolidated. Searches of OMIM just provide the descriptive text and a list of articles without the benefit of the search items highlighted within the text; publications must then be read to gather the links to the area of interest. Similarly, rs-numbers are not regularly listed in the OMIM article body text. 3. HapMap. The original stated aim of the HapMap Project – to determine the haplotype structure of the human genome – has expanded to encompass the characterization of all common human sequence variation. The inclusion of copy number variation and the broadening of ENCODE resequencing efforts to capture rare variation will extend this even further, but HapMap remains dominated by common SNPs and their haplotypes: the correlated arrangement of loci in segments defined by highly variable recombination rates. HapMap data have been structured into study phases I–III with different ranges of SNPs, SNP characteristics, and study populations. It is not always easy to find how each phase was defined but, in short, phase I encompassed about one million SNPs in four populations to give one common SNP per 5,000 bases, phase II consolidated SNP coverage with a further 2.5 million
SNP Databases
49
markers, and phase III has added another seven study populations. Current study population details are outlined in Table 3.2 and at the time of writing phase III data have become publicly available. The HapMap Web site provides a wide range of data, but of principal interest will be the genome browser, the SNP summary pages, and the HapMap data mart. HapMap has taken the view that the vast majority of users will start with a graphical overview of a chromosome segment and work outwards from there, so HapMap presents perhaps the best graphical genome browser for SNP variability currently available. Although the default map details (tracks) are relatively sparse, this provides clarity, while numerous other tracks can be added and kept as the user’s default arrangement for future browsing. The chromosome coordinates and SNP positions stay as fixed tracks throughout. This representation usefully complements dbSNP since any SNP not characterized by HapMap has a hyperlink rs-number in position to gain the Cluster Report. SNPs characterized by HapMap are linked to their own summary pages, which are briefer in content, so again linking out to dbSNP can be the best approach here too. The HapMap graphical browser really becomes informative when used to study the haplotype structure around the sites of interest (see Section 3.4) – originally mainly coding regions, but increasingly including intergenic regions identified by genome-wide association studies. The methods of graphical representation of haplotype structure can be a challenge to the first-time visitor to HapMap and it is recommended that users familiarize themselves with approaches used by HapMap and in key papers to display LD and that they understand the characteristics of the principal SNP association metrics of r2 and D 0 (3, 4). 4. Ensembl, Santa Cruz, Perlegen, and Assays-on-Demand. Ensembl and Santa Cruz genome databases largely provide alternatives to NCBI to access most of the same SNP and genome data. Ensemble specializes in the analysis of genome features and sequence to best identify and annotate genes and has a large range of species under study. This provides the most informative approach for users interested in comparative genomic approaches: where commonality of nucleotide or protein sequence can be identified by comparing different species. Ensembl has had a pivotal role in the complex task of gene identification and characterization, pioneering automated gene annotation techniques. Hosted in Ensembl, the Vertebrate Genome Annotation (VEGA) database provides a range of genome browsers (5). The main aim of VEGA is in providing the high-quality manual annotation of vertebrate
50
Phillips
genome sequence. Lastly, Ensembl provides close integration with the high-quality protein sequence database of Swiss-Prot/ UniProt (http://www.ebi.ac.uk/uniprot/and http://www. expasy.org/sprot/). This comprises manually annotated protein sequences with content that is fully linked with the Ensembl gene annotation pipeline. Santa Cruz has several features that can provide easier ways than NCBI to obtain information for SNP analysis – for example, the simple process of collecting extended context sequence for a SNP is more straightforward in Santa Cruz than from within dbSNP (see Note 7). Therefore, on occasions, working with two Web pages with different genome data browsers (essentially accessing the same underlying information) can be the optimum approach. The guide to Santa Cruz queries is at http://genome.ucsc.edu/goldenPath/ help/hgTracksHelp.html. Perlegen and Applied Biosystems’s Assays-on-Demand are private databases of SNP variability information that has been submitted to dbSNP and is publicly available, but can also be accessed from each company’s Web site with dedicated filtered search pages. Filters parallel the query process of Entrez by offering a choice of criteria that reduce the data set returned to a small, more manageable group of items meeting the criteria. Both databases elected to study US European, US AfricanAmerican, and US Chinese population panels that to a large extent mirror those of HapMap’s CEU, YRI, and CHB, so data obtained can be combined from different sources to allow meaningful comparisons of population variability or less often directly between different samples but originating from the same population group (although comparing YRI Africans with African-Americans highlights the about 20–30% European admixture in the latter). The easiest way to compare SNP data from similar populations in different databases is to use SPSmart (see Section 3.5.2). Note that Perlegen uses an internal SNP identifier with the format ‘‘PS+8 digit no.’’ (e.g., PS04631975) but accepts rs-number queries, while SPSmart provides a list of these numbers in its returned data. Assays-on-Demand SNP data are in large part based on the Celera SNP database generated during the private genome annotation performed by Celera after the human sequence had been completed in parallel to the completion of the public Human Genome Mapping Project in 2000. Celera genome data were available on a subscription basis (as Celera Discovery System, or CDS) between 2002 and 2006, but now all Celera’s SNP data have been incorporated into dbSNP and can be individually filtered in a search in Entrez with the inclusive term ‘‘AND Celera’’ or the exclusive term ‘‘NOT
SNP Databases
51
Celera’’ (see Section 3.3.1). Accessing Celera SNP data is also possible through Assays-on-Demand; users in the latter case can utilize a stand-alone tool called ‘‘SNPbrowser’’ comprising five million SNPs from public and CDS sources. This allows access to some of the original CDS SNP and gene annotation but is of most use as an alternative to HapMap for the definition of haplotype structure in a particular chromosome segment (see Section 2.2.2). Particularly in the population genetics field, Assays-on-Demand allows a simple system to review a large data set of SNP allele frequency variability from three major population groups and so has provided a core search step for many studies seeking to isolate and develop ancestry informative marker SNPs (6). 5. SeattleSNPs. The SeattleSNPs initiative is funded as part of the US National Heart, Lung, and Blood Institute (NHLBI) Program for Genomic Applications (PGA) – the latter abbreviation is used by dbSNP to reference SeattleSNPs SNP submissions. The project has undertaken the resequencing of more than 300 genes identified as primarily important in the inflammatory response, but also including cardiovascular disease and immunity (a full list of completed genes is at http://pga.gs.washington.edu/finished_genes.html). Although it is important to stress that the gene list mentioned above is not prescriptive – users are encouraged to nominate candidates for consideration. Therefore, SeattleSNPs provides a key opportunity to capture and characterize low-frequency SNPs from whole sequence data that would otherwise escape detection or be subject to acquisition bias (see Note 9). As sequencing technology has recently undergone one of the periodic quantum leaps in throughput, the chance to properly discover and catalogue new low-frequency SNPs by resequencing sufficiently large sample groups or individuals with a particular disorder will form the next major phase of SNP databasing. The extended ENCODE studies and the SeattleSNPs initiative stand at the vanguard of this work, with the 1,000 Genomes Project poised at the time of publication to take resequencing to the next level of resolution: that of full individual genomes. The evident drawback of SeattleSNPs comes from a focus on targeting a subset of genes or the pathways they occupy with the bias this might represent in attempting to understand the disease process. This is mainly due to the need to direct resources to the best areas for detailed SNP genotyping, and the fact that SeattleSNPs is actively engaged in association studies allows it to combine the knowledge this generates with new targets for resequencing in the genome. The SNP data from resequencing is fed
52
Phillips
into a database known as the Genome Variation Server (GVS) and users are encouraged to access the tutorials that explain optimum use of SeattleSNPs and GVS at http://www.open helix.com/downloads/seattlesnps/seattlesnps_home.shtml. 2.2. A Selection of Tools To Aid Analysis of SNP Data
The following tools are available to use as Web-based search systems or stand-alone programs that can help to make directed searches of the databases outlined previously. 1. NCBI tools: dbSNP-announce, MyNCBI, MapViewer, and Genome Workbench. Although not strictly online tools, dbSNP-announce (http:// www.ncbi.nlm.nih.gov/About/news/announce_submit. html) and MyNCBI (http://www.ncbi.nlm.nih. gov/entrez/ login.fcgi) are important subscription-based adjuncts to any use of dbSNP. Subscribing to dbSNP-announce provides automatic reports to the user’s e-mail address of dbSNP updates. As well as reporting the release of each new build, announcing newly added features, and outlining corrections or discovered problems with past or present builds, dbSNP-announce has an archive for referencing possible problems with, or qualifications to, previously obtained search data MyNCBI, requires a single subscription step to provide a search workspace for the user that provides a clipboard permitting combined searches from stored results obtained at different times (see Section 3.1.7). MapViewer integrates the bulk of the NCBI databases into a customizable genome map of aligned components termed ‘‘map elements.’’ The SNP data map element, termed ‘‘Variation,’’ can be included with any other genome feature in a custom map. A simple, clean icon set against each SNP marker positioned on the map showing a chromosome segment provides a clear summary description of the locus. Map browsing offers an intuitive way to review large numbers of SNPs in one session. Exploring a chromosome segment as a map is the best way to scrutinize the position and characteristics of nearby genome features of importance such as neighbor SNPs, genes, and their transcripts. Furthermore, the features around each SNP can be scrutinized easily through a series of hyperlinks embedded in many of the key map elements such as Genes. NCBI Genome Workbench (http://www.ncbi.nlm.nih.gov/projects/gbench/) is a stand-alone program that works locally, i.e., independently of individual online access to NCBI. Once installed, it can access and display genomic data from NCBI and combine this with the user’s own data in a series of graphical representations. The program is available for download and installation in any operating system format,
SNP Databases
53
and offers considerable flexibility in how the user chooses to align and compare genomic data. This extends to a range of alignment views, phylogenetic tree views, and tabular views of data. It can also align user’s data to those of public databases, and retrieve BLAST results. A full guide is beyond the scope of this chapter, so users are encouraged to explore this tool and the five tutorials (http://www.ncbi.nlm.nih.gov/projects/gbench/tutorial. html) for themselves. 2. Checking SNP assay primer designs: BLAST and Santa Cruz In Silico PCR. BLAST is a tool for assessing/calculating sequence similarity between a query sequence and the target sequence(s) available in the NCBI GenBank nucleotide databases. Users interested in developing SNP assay designs will query Nucleotide BLAST in two ways: (1) finding the location of a submitted sequence that includes the SNP, as the query ‘‘does the submitted sequence exist in a GenBank database?’’, and (2) checking for coincidental similarity in a sequence, normally a PCR primer, the query being ‘‘what is the degree of specificity of the submitted sequence?’’ These sequence comparisons can be made by choosing the standard BLAST (termed ‘‘blastn’’) and Search for short and near exact matches options, respectively. As a simple and quick alternative to BLAST, the Santa Cruz In Silico PCR tool (http://genome.ucsc.edu/ cgi-bin/hgPcr?command=start) offers a straightforward system that indicates the expected PCR product sequence from primer designs submitted by the user from comparisons to the current human reference nucleotide sequence. This tool is highly recommended since it provides a simple check before committing to primer purchases. 3. Exploring haplotype block structure maps: Haploview and SNPBrowserTM. Haploview (http://www.broad.mit.edu/mpg/haploview/) is an essential adjunct to HapMap browsing comprising a Java applet tool that permits the analysis and visualization of haplotype block patterns in HapMap data, choosing tagSNPs (7, 8), and estimating haplotype frequencies (see Section 3.4). The Applied Biosystems SNPBrowserTM tool (http:// marketing.appliedbiosystems.com/mk/get/snpb_landing) provides a stand-alone database of five million Celera SNPs that is downloaded to the user’s PC and can therefore be accessed offline or in parallel to online searches. The SNP data are presented as a chromosome segment map showing haplotype block distributions defined by Celera’s own pairwise analysis of 160,000 SNPs (termed ‘‘backbone validated
54
Phillips
SNPs’’), so it provides an alternative to HapMap in the annotation of human haplotype blocks, although it can display both HapMap and Celera haplotype maps. Additionally, SNPBrowserTM is easily configured to tailor haplotype block annotation displayed, SNP type, population studied, and size of the region shown. SNPBrowserTM works along the same lines as Assayson-Demand in providing a shopping list of SNPs based on user’s criteria that can then be ordered as commercial singleplex (Applied Biosystems TaqManTM) or multiplex (Applied Biosystems SNPlexTM) SNP genotyping assays. 4. Mapping SNPs and mutations in genes: SNPper. SNPper provides a tool for the extraction and re-presentation of SNP data from public databases focused on coding regions, offering the clearest system for scrutinizing SNP positions in and around genes (9). Once the user has provided the gene identifier, SNPper will list exonic, intronic, and 50 /30 -regions, plus embedded SNP positions within these, either as a a plain nucleotide sequence or as triplet code groups with their amino acid codes. Although the same output can be achieved with GenBank and Santa Cruz nucleotide browsers, SNPper is a much quicker and simpler system for listing SNPs in a gene of interest with a clean and intuitive graphical summary of the gene. This particularly suits the cataloguing of mutation sites in coding regions since these are usually defined by the amino acid changes they produce and SNPper allows their identification in relation to the SNP landscape that surrounds them, providing a straightforward way to develop genotyping assays. 5. Exploring the effect of SNPs on gene action: PupaSuite, Polyphen, and ESEfinder. PupaSuite (‘‘Pupa’’ stands for putative phenotype alterations) encompasses two tools – PupaSNP and SNPeffect – that aid the identification of SNPs effecting the processing of genes (10, 11), namely, sites of intron/exon boundaries or exonic splicing enhancers (ESEs), predicted transcription factor binding sites, and amino acid sequence changes. PupaSuite works with the Ensembl gene annotation and SNP database and can process an uploaded SNP list, but the user can also provide individually identified SNP sites with the aim of exploring their effect on gene action. The utility of PupaSuite is the ability to explore the effect of SNPs on transcriptional activity and splicing as well as protein sequence – an increasingly important step when analyzing coding regions. PolyPhen (http://genetics.bwh.harvard.edu/pph/data/ index.html) is a tool that usefully predicts the possible impact of an amino acid sequence change on the properties of a
SNP Databases
55
protein (12). Although it will not accept nucleotide input directly as it holds a nonsynonymous SNP database comprising about 50,000 SNPs from dbSNP, PolyPhen can check whether a SNP is nonsynonymous or not, using the site tool SNP2Prot. Effects on proteins are tentatively defined as unknown, benign, possibly damaging, and probably damaging. Users can input rs-numbers directly for comparison against the PolyPhen data, but they are advised to go directly to PupaSuite for novel coding SNPs discovered in their study. As with SNPper, this tool is particularly applicable to the characterization of mutations that are, by definition, SNPs at very low frequency. Of the three tools that help define SNP effects, the most specialized is ESEfinder (http://exon.cshl.edu/ESE/), a tool dedicated to identifying precursor RNA splice site changes from SNPs sited at exonic splicing enhancers (ESEs) (13). As such, SNPs at the ESE positions of proteins that routinely undergo alternative splicing can profoundly affect the final protein structure. ESEfinder makes use of databases of different ESE sequence motifs to help identify putative SNPs influencing splice patterns. 6. Using SNP haplotypes to detect signatures of selection: Haplotter and SWEEPTM. Compared with the tools available for studying gene and genome structure described above, population genetics tools are latecomers to SNP database analysis. Data of genome-wide patterns of polymorphic marker variation provide a powerful tool for studying the history of migration, bottlenecks/expansions, and adaptation in human populations. For those interested in analyzing such events, a major advantage in using SNP data is the distribution of SNPs at much higher densities compared with microsatellite or insertion–deletion variation and in the advanced characterization of SNP-based haplotypes. Therefore, SNPs are obvious candidate markers for the analysis of patterns of haplotype structure that can indicate signatures of past natural selection. Positive selection will amplify the frequency of a particular haplotype surrounding a favorable, novel gene variant because the haplotype accompanying the variant on the same chromosome strand also rises rapidly in frequency throughout the population. Before recombination disrupts this association, much higher SNP homozygosity is seen, as identical haplotypes are more likely to be found on each chromosome. Therefore, homozygosity is raised in the immediate vicinity of the selected gene variant and diminishes with distance, as recombination increasingly breaks up associations. This is the basis of the extended haplotype homozygosity (EHH) test that aims to
56
Phillips
detect signatures of recent selection by analyzing uncharacteristically long haplotype homozygosity patterns (14). Two tools are available for EHH analysis: Haplotter uses HapMap data and is accessed online, while SWEEP is a stand-alone program that can use data from any source that has been phased (i.e., allele combinations assigned to one of two strands). Haplotter (http://hg-wen.uchicago.edu/selection/haplotter.htm) measures a value iHS (15) that expresses the contrast between haplotypes with changed frequencies and the surrounding genome landscape, so it can reveal frequency rises in ancestral alleles (positive contrasts as the allele increases in frequency) as well as in variant alleles (negative contrasts). Haplotter can work from gene identifiers or a single SNP landmark (slower and varied in coverage). The program returns plots of iHS, plus standard selection signature or population diversity measures H, D, and Fst, followed by a table of adjacent genes, colored light blue when showing significant evidence of selection. The major advantage of Haplotter is it allows an unbiased approach to finding regions with indications of recent selection, so in use it is likely to reveal interesting new candidates for more detailed study. This can enable studies to focus on the phenotypes such loci exhibit as a way to explore differences in susceptibility to disease between populations. An advantage of using HapMap data is that the study populations will be extended to allow examination of more widely distributed patterns of local selection. The stand-alone program SWEEPTM (http://www. broad.mit.edu/mpg/sweep/) acts like Haplotter to measure the rate of decay of homozygosity with distance from putative regions subject to selection (14). Although it requires time and care to become familiar with use of the program, the graphical output, particularly diagrams termed ‘‘bifurcation plots,’’ provides very good representations of results summarizing extended homozygosity versus genomic distance to the core haplotypes. SPSmart (http://spsmart.cesga.es/) is a tool that performs the simple task of re-presenting SNP allele frequency data from multiple sources as pie charts identical to those of the HapMap browser. So SPSmart allows the user to review SNP variability across a wider range of populations than is feasible from single databases accessed one by one. This appears to offer little extra value if, for example, the study populations of HapMap phase II and Perlegen are considered, with only a comparison of YRI Africans and African-Americans of potential interest. However, SPSmart also processes data from the Stanford and Michigan University initiatives that have
SNP Databases
57
genotyped some 650,000 SNPs in the CEPH human genome diversity panel (HGDP) comprising almost 1,000 samples from 51 global populations. Incorporation of HapMap phase III populations has also boosted the scope of global variability that can be accessed with SPSmart.
3. Methods 3.1. Finding Sets of SNPs That Share Particular Characteristics: NCBI Entrez and Boolean Rules of Database Searching
1. The NCBI Entrez system uses Boolean terms or operators to define searches. These include the principal operators: AND, OR, and NOT, summarized in Fig. 3.1. Operators are the key parameters that define the relationship between criteria that describe database entries. In Entrez these descriptive details or criteria are put in groups termed ‘‘fields’’ that are defined by tags (alternatively qualifiers). Field details can be written in lowercase letters (but following an appropriate format, or syntax) ahead of their tags, which are always given in capital letters with fixed syntax within square brackets, for example, to define search criteria ‘‘SNPs on chromosome 2200 the field would be 22 denoted by the tag [CHR] written as 22[CHR], the chromosome field syntax being a number or X or Y. Users can either manually construct their own search with any combination of fields/tags and operators or simply choose tags from a menu on the EntrezSNP homepage (http:// www.ncbi.nlm.nih.gov/sites/entrez?db=snp) and provide fields to make a query using a default AND operator. For users unfamiliar with searches in NCBI, the latter option of choice from a menu can be easier to start with. The principal fields and their tags provided in EntrezSNP are given in Table 3.3. Fields separated by spaces alone also default to AND, e.g., query ‘‘HERC2[GENE] coding nonsynon[FUNC]’’ finds the nonsynonymous SNPs in HERC2 exons.
A
OR (union)
B
AND (intersection)
NOT (difference)
Fig. 3.1. Boolean operators. OR applies to all items in A or B, AND applies to items found in both A and B, and NOT applies to items in A not found in B.
58
Phillips
Table 3.3 Key EntrezSNP fields and their tags Description
Tag
Search field used
Example query
Observed alleles
[ALLELE]
IUPAC allele code (see Table 3.4)
R[ALLELE] find SNPs with A/G substitutions
Chromosome
[CHR]
Number/X, Y
21[CHR] OR 22[CHR] find SNPs on chromosomes 21 & 22
Base position
[BPOS]
Ranged number & AND & [CHR]
18000:28000[BPOS] AND Y[CHR] find SNPs in 10 kb segment of Y chromosome
Heterozygosity
[HET]
Ranged number
30:50[HET] find SNPs with heterozygosity value in range 30–50%
Function Class
[FUNC]
Locus region, intron, etc. (8 in total)
Coding nonsynon[FUNC]
Build
[CBID]
Number
125[CBID] search build 125
Gene location
[GENE]
Gene symbol
DARC[GENE] search for SNPs in Duffy blood group, chemokine receptor
Genotyping method
[METHOD]
Description as listed at URL below
Hybridize[METHOD] search for SNPs found by chip hybridization
(http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=snp – METHOD) Map weight
[MPWT]
Number: 1=once, 2=twice, 3=3–9 times
NOT (2[HIT] OR 3[HIT]) exclude SNPs mapping twice or more in genome
SNP single nucleotide polymorphism.
2. As a worked example, a search could be written longhand: ‘‘find unique SNPs in dbSNP on human chromosome 22 that are AC substitutions and show heterozygosity of more than 45%.’’ To perform a manual EntrezSNP search, place the following search description in the search box: 1[MPWT] AND human[ORGN] AND 22[CHR] AND M[ALLELE] AND 45:50[HET]. Note the field/tag items follow the same order as the longhand query, but this is not essential. The IUPAC allele codes applicable to the [ALLELE] field/tag are listed in Table 3.4. To perform the same search using the Entrez menu system, go to the limits menu – http:// www.ncbi.nlm.nih.gov/sites/entrez?db=snp&TabCmd= Limits – and choose tick boxes in (respectively) Map Weight, Organism, Chromosome, Variation Allele, and Heterozygosity. 3. Note that the heterozygosity tag in the above example search uses ranging: a range of values to define the field, described by a colon (:) in the middle of range limits. The menu-based
SNP Databases
59
Table 3.4 IUPAC SNP substitution codes used in Entrez as the field with the [ALLELE] tag. In addition A, C, G, and T can also be used with [ALLELE] to select all SNPs showing that base as an allele Code
Substitution
Code
Substitution
M
A or C
V
A or C or G
R
A or G
H
A or C or T
W
A or T
D
A or G or T
S
C or G
B
C or G or T
Y
C or T
N
A or C or G or T
K
G or T
N can also denote an indeterminate base.
system only allows heterozygosity ranges of 10%, so finetuned searches such as 49–50% heterozygosity require manual construction. Two other modifiers of operator function exist for manual searches: parentheses and the wild-card asterisk (*). Parentheses group search terms into logical sets to obtain items that further operations can search. To a large extent, the logic follows that used in a normal sentence, for example, in a PubMed search ‘‘find articles on the effects of heat and humidity on multiple sclerosis’’ is (heat OR humidity) AND multiple sclerosis, while ‘‘find articles on the effects of heat as well as the effects of humidity on multiple sclerosis’’ is heat OR (humidity AND multiple sclerosis). The wild-card asterisk in place of missing text allows a partial entry to be used as a query term, e.g., using BRC*[GENE] will find both BRCA1 and BRCA2 genes. 4. Each SNP in the EntrezSNP list that is returned from a query defaults to a summary graphic with components that describe the key parameters of the SNP, outlined in Fig. 3.2a. For the above example, query EntrezSNP lists 911 SNPs in order of chromosome position. If multiple chromosomes are listed, these are in order: Y, X, 22, 21, etc. A useful feature is the ability to change the default listing order amongst six options, including SNP ID (ascending rs-number) and heterozygosity. When assessing the role of particular SNPs in a disease process or by association with a candidate region, a particularly useful feature is the ‘‘Cited in PubMed’’ tab. Click the ‘‘Links/Pubmed (SNP Cited)’’ hyperlinks in this list to obtain each publication abstract.
60
Phillips
a +/– 25 base flanking sequence
OMIM entry (*607040) Hyperlink to GenBank sequence (4kb segment) Hyperlink to dbSNP coding SNP summary page
Heterozygosity scale (average 0.48)
Hyperlink to MapViewer
Hyperlink to submitter genotypes (both icons) V=validated by genotyping, G=sited in or near a gene
Chromosome (green spot underneath denotes unique SNP)
b Hyperlink to Cluster Report page
Validation success 1-2-3 stars, 80%–90%–95%
Genotype data available
Heterozygosity scale (showing range of estimates) Locus: SNP in gene, Transcript, Coding sequence Maps to a unique position (multiple positions = two yellow triangles)
Fig. 3.2. Key to summary graphics for single nucleotide polymorphisms (SNPs). (a) Key to summary graphics for EntrezSNP search return of example SNP rs17822931. (b) Key to summary graphics for SNPs shown in the chromosome view in NCBI MapViewer of example SNP rs17822931.
5. It is important to note that not all possible search fields can be accessed from the EntrezSNP limits menu: some 14 from a total of 24 are available (the full list and details are at http:// www.ncbi.nlm.nih.gov/sites/entrez?db=snp). By far the most useful search field for medical genetics studies omitted from the limits menu is Gene. This allows the listing of SNPs found within and close to a gene (or multiple genes using the OR operator) described by the query. SNPs at the 50 -end and the 30 -end (that order) of the gene are also listed, but it is prudent to extend the search using chromosome/base position tags – [CHR] AND [BPOS] – to capture potential promoter SNPs further afield. 6. An Entrez search can be initiated from the NCBI Entrez homepage by selecting ‘‘all databases’’ from the Search drop-down menu or specifically from Entrez SNP (‘‘SNP’’
SNP Databases
61
from the same menu). While searches from the EntrezSNP homepage return just a list of SNPs meeting the criteria, using the all databases option creates an NCBI-wide search with hyperlinked numbers of entries from 35 different databases that can be explored individually: making a good starting point in the early stages of a genetic study. The cross-database returns page groups six text-based databases, 26 genomic databases, and three catalogues (books, journals, and MeSH vocabulary) into three separate boxes. A cross-database search can be made using specific or general terms to obtain, respectively, a focused query of the broadest possible coverage within NCBI or a more open ended survey. For instance, ‘‘rs2293855’’ as a query returns a single PubMed reference to a possible role of this SNP in obesity, however with no reference to the gene MTMR9, where it resides, while ‘‘obesity’’ as a query lists no specific SNPs, but more than 111,000 publications and 681 genes, including MTMR9. 7. EntrezSNP gives the most efficient system for progressive searches as the lists generated can be stored in a clipboard and then sent to MyNCBI (avoiding an 8 h inactivity delete step for the clipboard), exported as a text file, combined with new searches, or re-searched itself. This uses the clipboard and history tabs at the top of the EntrezSNP page. The clipboard is a workspace for holding up to 500 items, while history lists the database search activity as numbers prefixed by a hash (#). Making Entrez searches at different times by exporting to MyNCBI allows the user to monitor the number of returns obtained with different search term combinations. Previous searches can be combined as hash fields with operators (e.g., #1 AND #2 gives items common to both searches). It is also possible to use hash fields together with normal fields, helping to build a stepwise record of the search process as it is modified in incremental stages. 8. To automatically reduce SNP numbers returned by a search, certain fields are best used as filters with fixed values including the organism and map weight. Therefore use of the human[ORGN] field/tag ensures only human SNPs are listed and 1[MPWT] ensures all SNPs are unique (i.e., single map weight). The SNP validation tag also provides a system to filter out SNPs detected by sequence comparisons alone, using by frequency[VALIDATION]. . 3.2. Obtaining Detailed Information on a SNP: dbSNP Cluster Reports
1. The Cluster Report page of dbSNP provides most, if not all, the information needed to assess the characteristics of a SNP and design a genotyping assay if required. Each page is broken down into seven sections with largely self-explanatory
62
Phillips
headings: Submitter records; Fasta Sequence; GeneView; Integrated Maps; NCBI Resource Links; Population Diversity; and Validation Summary. This section of the chapter outlines the steps required to (1) obtain sufficient context sequence to design a genotyping assay, (2) scrutinize the map position of the SNP with accompanying genomic features, and (3) begin analysis of the population variation shown by the SNP. 2. The Fasta section is named after the fast-all sequence similarity program used by dbSNP to detect identical SNP submissions given in the Submitter section (see Note 4). Fasta lists the flanking sequence surrounding the SNP position – the quantity of nucleotides listed is variable in extent but always arranged as groups of ten bases, six groups per line, with the SNP positioned on a separate line as an IUPAC code base. The single Fasta header line contains summary locus details explained in the ‘‘Legend’’ hyperlink in the title above. Reliance on dbSNP Fasta sequence alone for primer designs can cause problems (see Note 10); however, one clear advantage of dbSNP Fasta is the inclusion of information from the initial submitter and from RepeatMasker analysis (see Note 11) which predicts the likely genomic uniqueness of the SNP context sequence. This can help avoid certain sequence segments that may occur in multiple genomic locations and therefore reduce the specificity of any genotyping assay designs. 3. An essential additional aid to the primer design process is the neighbor SNP detection tool found in the Integrated Maps section. Clicking on the ‘‘View’’ hyperlink under the Neighbor SNP heading of each assembly (for ref_assembly: i.e., the reference sequence is recommended) gives all SNPs within – 100 bp of the reported SNP, including at 0 bp, the target substitution itself. This permits easy location and masking of variable sites that can interfere with the predictable binding of primers designed for the assay. 4. GeneView gives a graphical overview of the SNP if it is located in a gene, giving a position mark using nine color codes for one of 16 predicted functions (hyperlink guide: color legend). The cSNP radio button directs one to the coding SNP listing page for each gene (see Note 6). 5. The Integrated Maps section provides a ‘‘snapshot’’ genome view of the SNP position as a red mark-point by clicking the chromosome number hyperlink. This whole genome view can be used to map a series of SNPs by using the OR operator between each rs-number in the query string in the search box. Including ‘‘NOT Celera’’ eliminates the double mark-points and position listing (see Note 7). From this overall map any
SNP Databases
63
chromosome can be viewed in detail in NCBI MapViewer, centered on the SNP position by clicking the number hyperlink of each chromosome showing a mark-point. Once in MapViewer, configure the view by clicking ‘‘maps and options’’ to show the ‘‘Variation’’ map as the master to ensure a summary icon set accompanies each SNP position (outlined in Fig. 3.2b). Any number of other tracks (map components) can be selected as part of the genome landscape around the SNP, although the single most useful of these is Gene. 6. The Population Diversity section summarizes SNP allele frequency distributions in different populations using gold (reference allele) and blue (alternative allele) bars. Clicking on ‘‘Genotype Detail’’ provides the Genotype and Allele Frequency Report (still in beta status at the time of publication). It is possible to obtain the individual SNP genotypes submitted from each contributing laboratory – data that can be particularly useful when using standard DNA samples, such as Coriell panels (http://ccr.coriell.org/), as genotyping controls in an assay. The easiest way to achieve this directly from HapMap, the major source of SNP population variability data in dbSNP, is to go to the individual SNP information page in HapMap (http://www.hapmap.org/ cgi-perl/snp_details? name=rsnumber) and click the ‘‘retrieve genotypes’’ hyperlinks. Genotypes are always listed in the same sample-ID order and so can be downloaded directly to Excel and correctly ordered as rows per population per SNP using the text to columns option (see Note 1). 3.3. Exploring SNPs in Coding Regions: SNPper and PupaSuite
1. SNPper requires a subscription before the user can explore a gene of interest which can be found using ‘‘Gene Finder’’ by inputting standard identifiers (see Note 8). Once the gene has been obtained, click on the ‘‘Annotated’’ sequence hyperlink to obtain the nucleotide listing marked as follows: green, 50 /30 ; black lowercase, exonic noncoding; black uppercase, coding; gray, intronic; blue underlined, SNPs. A useful approach for the reliable detection of mutations or scrutiny of coding SNPs is to click ‘‘View amino acid sequence’’ to obtain the coding nucleotides as triplet codes above their accompanying amino acids. To locate a novel mutation from the standard ‘‘wild-type amino acid/codon/variant amino acid’’ format as normally reported in the literature (e.g., V60L in MC1R) it is necessary to carefully count the relevant nucleotide and codon numbers from the leftmost reference numbers (pencil annotations of a printout are recommended).
64
Phillips
2. PupaSuite can accept a list of genes using Ensembl or GeneID identifiers or can review a defined chromosome segment to search for SNPs and suggest an effect. PupaSuite is of particular interest if novel or uncharacterized SNPs are being studied as there is the opportunity to apply the same predictive tools to these loci. To explore the above three options, upload the relevant data to ‘‘Upload/paste file of genes,’’ ‘‘Search a region,’’ and ‘‘Have you got new SNPs?’’, respectively. There is an option to define gene flanking regions as numbers of nucleotides upstream of the translation start site to find SNPs that may affect transcription factor binding sites. Therefore, PupaSuite is a particularly useful tool for the identification of SNP sites associated with changes to intron/exon boundaries or transcription factor binding. Lastly, additional functional annotations are provided to help assess the impact of the uploaded SNPs, including gene ontology, homology data, and OMIM references. 3.4. Simple Reviewing of SNP Haplotype Block Structure: HapMap
1. Users new to SNP analysis may hesitate before undertaking the process of analyzing human haplotype block structure in regions of interest. The accurate mapping of haplotype blocks, interpretation of D’ and r2 values, selecting tag SNPs to track blocks (3, 4, 7, 8), and assessment of genome-wide patterns of association are all specialist tasks needing care and experience (16). However, all current genetic analysis approaches require an understanding of the likely patterns of association between a set of SNPs and correlating genes or regions of interest; therefore, using HaploView within the HapMap database browser can provide a simple overview to start this process. Once HaploView has been installed on the user’s own PC as a Java applet, it is possible to work directly on data from HapMap or Perlegen, but it is easier to start by configuring and viewing LD maps in the HapMap genome browser. 2. Add a gene (or region) of interest to the ‘‘Landmark or Region’’ search box and tick the three ‘‘Analysis’’ tracks: Phased Haplotype Display, LD Plot, and tag SNP Picker. Clearer graphics can be obtained by initially selecting one population at a time by selecting both ‘‘Annotate LD Plot and Phased Haplotype Display’’ and clicking ‘‘Configure. . .,’’ then choosing a single population radio button. 3. The phased haplotype display presents the alternative haplotype blocks as blue and yellow segments matching the chromosome lengths occupied. The ease with which the user can interpret these depends on the number and length of the haplotypes in the region displayed. As an example, a very
SNP Databases
65
simple pattern is shown by ATM: a large but highly conserved gene (strong selective constraints apply to ATM, OMIM: 607585). The phased haplotype plot clearly shows that two haplotypes account for almost two equal halves of the CEU sample. No fewer than 27 of the 31 blocks define this division and the pattern is underlined by a series of identical equalsegment CEU pie charts for the genotyped SNPs in ATM. Note that at blocks 7 (left to right) and 13 a third and fourth common haplotype can be discerned and the third haplotype is characterized by different SNP alleles at blocks 17, 23, 25, 29, and 30. Finally, a singleton (literally a single CEU sample) and a minor-frequency haplotype can be seen in blocks 25 and 29, respectively. The pattern shown by ATM is, in fact, relatively common in the human genome and is termed ‘‘yin– yang haplotypes’’ (17). 4. The LD plots represent the extent of LD between SNPs in the region queried shown as inverted pyramid graphics. The default color scale, also in widespread use in the literature, shows maximum LD as dark red blocks and minimum LD as light gray blocks. Two example genes, CAPG and DTNBP1, illustrate how these plots can summarize both simple and complex predicted LD patterns: showing, respectively, a single, simple pyramid and multiple overlapping pyramids with heterogeneous LD values within each pyramid (checkerboards of red and gray blocks). While this partly reflects gene size and therefore SNP density (note the sevenfold difference between each gene), LD plots can provide a summary overview of recombination and SNP association patterns in the region. 5. The Tag SNP display, once configured, updates automatically between genes and many users may wish to rely just on this system to collect tag SNPs to combine with other core loci (nonsynonymous coding SNPs and translation/transcriptionmodifying SNPs identified by PupaSuite) to construct simple directed association studies. Although this process has largely been replaced by a standard two-stage approach of whole-genome scans then follow-up directed SNP genotyping, HapMap browsing can give a simple system for assessing the transportability, i.e., the applicability of a tag in multiple populations (8), power, and positioning of the tag SNPs that now form the core battery of markers in whole-genome analyses. 3.5. Assessing Population Genetics Parameters from Online SNP Data: Haplotter and SPSmart
1. Haplotter provides a useful way to begin exploring the population genetics parameters of iHS (outlined in Section 2.2.5), H, D, and Fst, in a genomic region. Queries are initiated by chromosome region, gene, or SNP and this will return four graphics which summarize the
66
Phillips
above-mentioned parameters in the same order, with plots for each of the three HapMap panels (i.e., CHB and JPT populations are combined as panel ASN). The Fst graphic plots the three population comparisons to give a useful overview of genomic divergence – in particular the outliers plotted as single points can highlight those SNPs that show very strong interpopulation diversity. A table is given of iHS values around the region of interest with levels diagnostic of EHH highlighted in blue. An oftenquoted example that users can investigate for themselves is the gene LCT (gene-ID 3938), this shows a very broad peak of elevated iHS in the CEU population extending well beyond the LCT chromosome region, underlined by high CEU-YRI and CEU-ASN Fst values and blue-labeled iHS levels in the accompanying table. Both the original study of selection patterns in LCT (18) and the Haplotter paper (14) ably explain these patterns. 2. In a fashion identical to Haplotter, SPSmart reviews a region, gene, or SNP list with the primary aim of summarizing the population variability found in multiple SNP databases as pie charts and key population metrics: observed H (heterozygosity), expected H, Fs, Fst, and divergence (In). Usefully SPSmart also pulls from dbSNP chromosome and position, validation status, reference and ancestral allele, and the minor allele frequency, providing alongside the population metrics a succinct one-line summary of each SNP. To explore the population variability of a set of SNPs, choose the SNP databases from HapMap phase II, HapMap phase III, Perlegen, and Stanford/Michigan CEPH-HGDP (4, 4+7, 3, and 51 populations, respectively) and provide the rs-numbers or locations. Clicking ‘‘metasearch’’ permits selection of a population or any combination from each of the five databases (but note the overlap between HapMap phase II and HapMap phase III) prior to uploading the SNPs of interest. For example, to review European frequency variability for the SNP rs12075, click each of the databases, tick the populations of interest, (e.g., CEU, TSI, European American, Italy-Sardinian, France Basque), add the rs12075 query to the ‘‘Search by SNPs’’ box, then (after choosing optional filters) click ‘‘search.’’ Pie charts and population metrics (and their downloadable data) are returned as separate tabs, while missing data are clearly labeled. The evident Basque divergence for rs12075 demonstrates the simplicity but informative value of HapMap style pie charts as an aid to reviewing SNP variation across multiple population-based databases. .
SNP Databases
67
4. Notes 1. Several approaches to database searching using a PC can help the user considerably when manipulating the data obtained from a query. Using tabbed Web page holders in the Web browser of choice (Internet Explorer; Firefox; Safari) allows simple switching between pages. While much SNP data is numerical, all information can be uploaded to a simple individual database in Excel, which now also offers sophisticated text-handling capacity, for offline processing. Although it is rarely recommended by specialists, Excel can offer a simple stand-alone database system by adapting cells to use functions such as LOOKUP, COUNTIF, or those specifically geared to database searches prefixed with ‘‘D,’’ such as DGET. Excel compensates for a lack of power by providing a simple and easily mastered set-up of small-scale personal databases suiting many SNP studies. The ‘‘Text to Columns’’ tool in the Excel ‘‘Data’’ menu is a straightforward way to directly process plain text files downloaded from the Web, while preserving the structure of data items separated by spaces, semicolons, or other delimiters. Simple text editors themselves are highly efficient systems for holding and searching data. For example, it is possible to find a single SNP amongst a list of 650,000 in real time using the ‘‘Find’’ function available in all text editors. 2. Google can be used directly to search for specific items such as rs-numbers or mitochondrial substitution sites – the latter being a particularly fruitful approach to finding medical or population studies reporting diagnostic mitochondrial haplotypes (19). For example, entering the search string ‘‘human mtDNA G6261A’’ into Google provides a list of papers reporting this mutation and a supposed role as a cancer risk factor. Care should be taken to ensure full use of the adjacency function of Google searches (known as the Boolean operator NEAR), which is not part of most genome database search engines. Therefore, to avoid very long lists of returns, it is advisable to include terms such as human mtDNA alongside the standard Cambridge Reference Sequence descriptions. As a compliment to PubMed queries, Google Scholar (http:// scholar.google.com/advanced_scholar_search) should also be part of every researcher’s online SNP query bookmarks. 3. HapMap experienced minor problems when collating project data generated in different genotyping centers for the same SNP sites, for example, SNP rs1355497 was amongst 37 SNPs reported as showing fixed-difference allele frequencies (1) but has since been shown to be an invariant, monomorphic SNP (also see Note 9).
68
Phillips
4. Since a SNP is characterized by the context sequence each side of the nucleotide substitution site, it should be possible to uniquely define a SNP by referencing organism, chromosome, and base-pair position. However, a small but significant proportion of SNPs are nonunique, so the context sequence and its likelihood of repetition in multiple locations become critical in identifying whether a SNP is unique or not. The submission criteria of dbSNP are very effective at detecting nonunique SNPs, with a process that uses the FASTA program to check a minimum 100 bp flanking sequence to assess if the SNP can be positioned uniquely in the genome and can be matched with other submissions of the same SNP. The proportion of nonunique SNPs remains very small in dbSNP at about 5% and is much more common in certain regions, e.g., pericentromeric areas of each chromosome. 5. Very useful and readable guides to the routine use of the NCBI sites are detailed in a PDF handbook that can be downloaded chapter by chapter (http://www.ncbi.nlm.nih. gov/books/bv.fcgi?rid=handbook.part.1). Chapters particularly relevant to SNP research include Chapters 2 (PubMed), 5 (dbSNP), 7 (OMIM), 15 (Entrez), 16 (BLAST), and 20 (Map Viewer). Download them by clicking the PDF icon on each chapter summary page. 6. SNP sites are still routinely described by the amino acid substitution they create rather than an rs-number, particularly if they are mutations or rare enough to escape detection by dbSNP. The easiest way to obtain the rs-number (if it exists) is to record the gene identification number from NCBI Gene (e.g., query ‘‘MC1R AND human’’ gives GeneID 4157) then go to the coding SNP part of dbSNP using the following URL finishing with the ID number to obtain the listing of known coding substitutions and their affected amino acid residues: http://www.ncbi.nlm.nih.gov/SNP/ snp_ref.cgi?locusId= number. For example, in gene MC1R, R151C is amongst the most commonly described coding SNPs, and was revealed to be rs1805007 using the procedure described above. 7. Human genes are consistently identified across different databases by a gene symbol (sometimes termed the ‘‘gene name’’) comprising a series of uppercase letters and numbers (http:// www.genenames.org/), provided by HUGO. A gene ID consists of a number alone and refers to the number codes given to each gene by NCBI Gene. These can be used both within NCBI as the main point of reference for a gene (e.g., when reviewing coding SNPs) and in certain other databases. A useful gene ID converter tool is provided at http://idcon verter.bioinfo.cnio.es/.
SNP Databases
69
8. Use of the single map weight filter in EntrezSNP does not lead to the exclusion of all the SNPs in dbSNP with different Celera locations; however, the list of returns is headed by SNPs that carry the warning ‘‘Mapped unambiguously on non-reference assembly only.’’ Note, however, that when one uses the NCBI MapViewer (see Section 3.2.5) both reference sequence and Celera locations are marked on the genomewide chromosome map. Therefore, it is advisable to include the term NOT Celera at the end of a multiple SNP list in MapViewer queries. 9. Resequencing is the only method of SNP characterization that avoids acquisition bias. This is the phenomenon where the characteristics of the SNP itself affect the chances of its detection by genotyping methods. Examples of SNP features that mean the loci are either not detected or incorrectly genotyped by large-scale projects such as HapMap include triallelic SNPs (the medically important CRP promoter rs3091244 being a notable example), SNPs with very low frequency minor alleles (also missed by resequencing if insufficient samples are typed), and SNPs with very dense arrays of closely neighboring SNPs such as those of the hypervariable major histocompatibility complex. Acquisition bias can also describe the process of selecting SNPs from databases using criteria which bias the SNP lists produced from a query. 10. Often the Fasta section lists less than 100 bp of context sequence each side of a SNP (e.g., rs1805009) – often owing to the fact that a submitting laboratory only provided short segments. The easiest way to obtain – 100 bp of context sequence for assay primer design purposes is to use the Santa Cruz genome assembly. Add the SNP rs-number to the URL http://genome.ucsc.edu/cgi-bin/hgTracks?position=rs number (several dbSNP builds available), click the highlighted SNP in the map, click ‘‘view DNA for this feature,’’ then opt for 100 bases upstream/downstream. The SNP base is the reference allele and is not marked so it is best to use 100+0 and 0 +100 in two separate sequence dumps. Another potential problem in the Cluster Report Fasta section is the occasional (and apparently ad hoc) listing of neighbor SNPs as IUPAC codes (see Table 3.4). For example, rs1805006 includes no fewer than six other SNPs in – 100 bp of sequence, given as K, R, R, Y, R, R (in that order), that may cause problems once the sequence is inserted into primer design software. Processing the SNP context sequence directly from Santa Cruz avoids this problem. 11. Fasta section nucleotides are presented in two ways: in uppercase/lowercase letetrs and in black/green. Uppercase letters denote a normal, unique, genomic sequence, while lowercase letters are used for a sequence identified by RepeatMasker
70
Phillips
(http://repeatmasker.genome.washington.edu/cgi.bin/Re peatMasker) as a low-complexity or repetitive element sequence. Green denotes a sequence identified by the submitter during the SNP assay process (a single green SNP base signifying identification by sequence comparison), while black denotes a flanking sequence used by NCBI from the nucleotide databases as part of the SNP submission checks.
Acknowledgements The author would like to thank Maviky Lareu, Antonio Salas, and Angel Carracedo, University of Santiago de Compostela, for useful discussions in the preparation of this chapter. The work was in part supported by funding from Xunta de Galicia: PGIDTIT06PXIB228195PR and the Spanish MEC: BIO2006-06178. References 1. The International HapMap Consortium. (2005) A haplotype map of the human genome. Nature 437, 1299–1320. 2. The International HapMap Consortium. (2007) A second generation human haplotype map of over 3.1 million SNPs. Nature 449, 851–861. 3. Gabriel, S. B., Schaffner, S. F., Nguyen, H., Moore, J. M., Roy, J., Blumenstiel, B. et al. (2002) The structure of haplotype blocks in the human genome. Science 296, 2225–2229. 4. Wall, J. D. and Pritchard, J. K. (2003) Haplotype blocks and linkage disequilibrium in the human genome. Nat. Rev. Genet 4, 587–597. 5. Ashurst, J. L., Chen, C. K., Gilbert, J. G., Jekosch K., Keenan S., Meidl P. et al. (2005) The Vertebrate Genome Annotation (Vega) database. Nucleic Acids Res. 33, D459–465. 6. Yang, N., Li, H., Criswell, L. A., Gregersen, P. K., Alarcon-Riquelme, M. E., Kittles, R. et al. (2005) Examination of ancestry and ethnic affiliation using highly informative diallelic DNA markers. Hum. Genet. 118, 382–392. 7. de Bakker, P. I., Yelensky, R., Pe’er, I., Gabriel, S. B., Daly, M. J. and Altshuler, D. (2005) Efficiency and power in genetic association studies. Nat. Genet. 37, 1217–1223. 8. de Bakker, P. I. W., Noel, N. P., Burtt, N. P., Graham, R. R., Guiducci, C., Yelensky, R., Drake, J.A. et al. (2006) Transferability of tag
9.
10.
11.
12.
13.
14.
15.
SNPs in genetic association studies in multiple populations. Nat. Genet. 38, 1298–1303. Riva, A. and Kohane, I. S. (2002) SNPper: retrieval and analysis of human SNPs. Bioinformatics 18, 1681–1685. Conde, L., Vaquerizas, J. M., Santoyo, J., Shahrour, F., Ruiz-Llorente, S., Robledo, M. et al. (2004) PupaSNP Finder: a web tool for finding SNPs with putative effect at transcriptional level. Nucleic Acids Res. 32, W242–248. Conde L., Vaquerizas J.M., Dopazo H., Arbiza L., Reumers J., Rousseau F. et al. (2006) PupaSuite: finding functional single nucleotide polymorphisms for large-scale genotyping purposes. Nucleic Acids Res. 34, W621–625. Ramensky, V., Bork, P., and Sunyaev, S. (2002) Human non-synonymous SNPs: server and survey. Nucleic Acids Res 30, 3894–3900. Cartegni, L., Wang, J., Zhu, Z., Zhang, M. Q., and Krainer, A. R. (2003) ESEfinder: A web resource to identify exonic splicing enhancers. Nucleic Acids Res. 31, 3568–3571. Voight, B. F., Kudaravalli, S., Wen, X. and Pritchard, J. K. (2006) A map of recent positive selection in the human genome. PLoS Biol 4, 446–458. Sabeti, P. C., Reich, D. E., Higgins, J. M., Levine, H. Z. P., Richter, D. J., Schaffner, S. F. et al. (2002) Detecting recent positive selection in the human genome from haplotype structure. Nature 419, 832–837.
SNP Databases 16. Haiman, C. A. and Stram, D. O. (2008) Utilizing HapMap and tagging SNPs. Methods Mol. Med. 141, 37–54. 17. Zhang, J., Rowe, W.L., Clark, A.G. and Buetow, K.H. (2003) Genomewide distribution of high-frequency, completely mismatching SNP haplotype pairs observed to be common across human populations. Am. J. Hum. Genet. 73, 1073–1081.
71
18. Bersaglieri, T., Sabeti, P. C., Patterson, N., Vanderploeg, T., Schaffner, S.F., Drake J.A. et al. (2004) Genetic signatures of strong recent positive selection at the lactase gene. Am. J. Hum. Genet. 74, 1111–1120. 19. Bandelt, H. J., Salas, A. and Bravi, C. M. (2006) What is a ‘novel’ mtDNA mutation– and does ‘novelty’ really matter? J. Hum. Genet. 51, 1073–1082.
Chapter 4 Mining SNPs from DNA Sequence Data; Computational Approaches to SNP Discovery and Analysis Jan van Oeveren and Antoine Janssen Abstract Single nucleotide polymorphisms (SNPs) are the most abundant form of genetic variation and are the basis for most molecular markers. Before these SNPs can be used for direct sequence-based SNP detection or in a derived SNP assay, they need to be identified. For those regions or species where no validated SNPs are available in the public databases, a good alternative is to mine them from DNA sequences. The alignment of multiple sequence fragments originating from different genotypes representing the same region on the genome will allow for the discovery of sequence variants. The corresponding nucleotide mismatches are likely to be SNPs or insertions/deletions. A large amount of sequence data to be mined is present in the public databases (both expressed sequence tags and genomic sequences) and is free to use without having to do large-scale sequencing oneself. However, with the appearance of the next-generation sequencing machines (Roche GS/454, Illumina GA/Solexa, SOLiD), high-throughput sequencing is becoming widely available. This will allow for the sequencing of polymorphic genotypes on specific target areas and consequent SNP identification. In this paper we discuss the bioinformatics tools required to analyze DNA sequence data for SNP mining. A general approach for the consecutive steps in the mining process is described and commonly used SNP discovery pipelines are presented. Key words: Single nucleotide polymorphism discovery, single nucleotide polymorphism mining, de novo single nucleotide polymorphism mining, expressed sequence tag, pipeline tools, DNA sequencing, next-generation sequencing, base-calling quality, alignment, reference sequence.
1. Introduction Single nucleotide polymorphisms (SNPs), as the main form of genetic variation, are very important for molecular marker development. In addition to the already validated SNPs in public-domain databases, there exist many sources of data with the potential to reveal novel polymorphisms. The combined databanks of NCBI/EMBL/DDJB A.A. Komar (ed.), Single Nucleotide Polymorphisms, Methods in Molecular Biology 578, DOI 10.1007/978-1-60327-411-1_4, ª Humana Press, a part of Springer Science+Business Media, LLC 2003, 2009
73
74
van Oeveren and Janssen
(1–3), for instance, contain millions of sequenced fragments often representing multiple samples from varied genetic sources. Among the most abundant data types in these databanks are sequences from expressed sequence tags (ESTs). ESTs derived from different genetic sources can be aligned and compared, revealing putative polymorphisms where sequence differences are seen. These sequence differences are likely SNPs or insertions/deletions (indels) (4, 5). Alternatively, SNPs can be mined from genomic sequence data, e.g., bacterial artificial chromosomes (BACs) (6) or random shotgun fragments (7). Naturally, these sequence data sources are not limited to the public domain. Next-generation sequencing (NGS) technologies (Roche GS/454 (8), Illumina GA/Solexa (9), SOLiDTM (10)) allowing high-throughput (HT) sequencing will likely to bring many new SNPs to the attention of researchers (see Chapter 5). Generating large amounts of sequence data can therefore become a routine procedure performed in-house. While sequencing an entire genome inhouse may still be way off, sequencing for deep coverage on a specific target or set of targets on a genome is certainly in reach and will enable the identification of a considerable number of sequence variants. Polymorphisms identified by this process can be further used to develop cost-effective marker assays. Some early SNP discovery studies on the human genome used a resequencing approach on specific amplicons (11) or reduced representation shotgun sequencing (12). Unfortunately, sequence data often suffer from sequencing errors (see Chapter 5); therefore, it is critical to take into consideration the number of sequences aligned on the region of interest and the quality of the bases called at a candidate SNP position (the base-calling quality indicates the probability that the nucleotide is correctly identified). Increasing the number of aligned sequence fragments during SNP discovery increases the accuracy and confidence of correctly identifying polymorphisms, and reduces the chance that a sequencing error is identified as a polymorphism. The methods used for SNP discovery will be very similar for both (public) database-derived and HT-sequencing-derived data. However, specific methods may vary with different source data and varying complexities of the approach. In this paper we present an overview of the methodology for SNP mining from DNA sequence data and discuss our preferred software tools.
2. Materials 2.1. Sequence Data Classification
SNP mining can be performed on various types of sequence data. These sequence data can be classified according to criteria which affect the method of analysis. Generally, all methods encompass the following steps (also depicted in Fig. 4.1):
Computational Approaches to SNP Discovery and Analysis
De novo
75
Reference
Shotgun fragments RE-fragments ESTs
STSs
Shotgun fragments RE-fragments ESTs
Chromatograms or sequence reads
Clustering
Mapping
Assembly and Alignment
Sequence variant identification
SNP filter criteria
Putative SNPs
Fig. 4.1. Flowchart of the general approach to single nucleotide polymorphism (SNP) mining from DNA sequence data.
1. Grouping sequence reads according to their sequence similarity to identify reads covering the same part of the genome or having the same transcript origin. 2. Aligning the reads. 3. Identifying and classifying sequence variants as potential polymorphisms. Step 1 may considerably differ depending on the availability of reference sequence information. Therefore, we will differentiate the methodology into reference-based and de novo SNP mining. 2.1.1. Reference Sequence Data
All sequence data generated from fully sequenced species belong to the reference sequence data group. The human genome and many microbial species are examples of the reference group, but as an increasing number of species are rapidly sequenced by nextgeneration machines, this group will grow exponentially. Publicly available Web sites such as NCBI’s http://www.ncbi.nlm.nih. gov/Genomes list available fully sequenced and partially sequenced genomes. Even when the sequencing has not reached a finished status, it can be used to map newly created sequence data. Therefore, the ‘‘reference’’ label can also be applied to sequence data that map to a
76
van Oeveren and Janssen
partially completed genome. These partially completed genomes can consist of a set of BAC sequences or complementary DNA (cDNA) data. Such data sets can be found at Unigene (13) or Gene Indices (14). Note that NCBI Unigene and Gene Indices differ in that the Gene Indices set is clustered more strictly, often resulting in multiple clusters for a specific set of transcripts compared with a single cluster in NCBI Unigene. A third group of reference data are those sequences generated from a known PCR product (amplicon) where primers have been developed for a specific sequenced region. The PCR products can be used in screening a large set of genotypes or mutants specific to the amplicon, resulting in ultradeep sequencing. In such a monoplex situation, no grouping or mapping has to be done; in a situation of multiple amplicons, a simple mapping is all that is necessary. 2.1.2. De Novo Sequence Data
For many species, none of the aforementioned conditions are applicable. The first step in SNP mining procedures for de novo sequences thus involves the clustering of all sequence reads into homologous groups. The source material for these sequenced fragments can be either genomic DNA or cDNA. Since the de novo assembly of the complete genome sequence is a fairly complex topic on its own, we will focus on methods dealing with a relatively small portion of the genome. These methods include the transcriptome methods (cDNA, EST), methods for the systematic reduction of genome complexity (e.g., AFLP1 (15), CRoPS1 (16)), and methods targeting a unique fraction of the genome (e.g., Cot Filtration (17)). All of these methods will result in either a set of relatively short fragments (cDNA and AFLP) or a set of nonrepetitive fragments that should be suitable for assembly and linking to complementary sequence reads. The only requirements of these methods are that sequence coverage is deep and wide enough to allow for ample overlap of putative polymorphic regions and that clustered genotypes can be merged into individual assemblies.
2.2. Base-Calling Quality
An additional characteristic of sequence reads that should be taken into consideration is the quality of base-calling scores. When sequence information is generated, the assignment of a nucleotide to a certain position will be assigned a probability value. For Sanger sequencing (see Note 1), this is typically a ‘‘phred score,’’ named after the base-calling software tool Phred (18, 19). The phred score can be calculated by the equation –10 log10 P(the base calling is false). Often a phred score of 20 is taken as a reliable threshold, equivalent to a false-positive rate of 1%. Some of the new sequencing machines have their own quality measure, but it often resembles the phred score (e.g., Illumina GA has a score range of –5 to 40).
Computational Approaches to SNP Discovery and Analysis
77
An alternative to the phred score is a likelihood value for each of the four bases. Getting information for all bases allows for the prediction of a second possible allele in the case of a heterozygote genotype. This will only apply when multiple fragments (e.g., from the homologous chromosomes) form the basis for a single sequence read, as in Sanger sequencing. For many of the NGS machines, reads are derived from single fragments and cannot result in heterozygote reads. Although the main sequencing platforms used at the moment can yield large amounts of high-quality data, there are some shortcomings: Sanger sequencing suffers from low base quality at the terminal ends of both the 50 and the 30 sides (17, 18); the Roche/ 454 technology (8) has a problem in accurately detecting the length of homopolymer regions and the Illumina/Solexa machines produce only very short reads as additional base calling at the end becomes unreliable (9). It is wise to use the base-calling scores where possible, as they will provide additional information from which to determine the quality of putative polymorphisms. When sequence variants are observed which are based on high-quality base scores, they are likely to represent true polymorphisms. On the other hand, low-quality scores for mismatches have a high chance of being sequencing errors. In situations where quality scores are not present, e.g., when retrieving sequence data from the public domain, this lack of information can be compensated by getting enough coverage such that putative polymorphisms are confirmed by replicated reads. Additionally, one must be aware that some sequencing errors are easily reproduced (20) once they have been linked to a specific sequence pattern (e.g., short repeats). Flaws in the process of sample preparation (e.g. reverse transcription to cDNA) can also cause sequence errors, but these will not be filtered by sequence quality as they are correctly called (they are ‘‘sequence,’’ not ‘‘sequencing’’ errors). From this we can conclude that both base-calling quality as well as read redundancy increase confidence in detecting SNPs and reduce false-positive rates.
3. Methods Different analysis steps apply to the two types of sequence data: reference sequence data and de novo sequence data. The consecutive steps are depicted in Fig. 4.1 and are explained in detail below.
78
van Oeveren and Janssen
3.1. Reference Sequence Data 3.1.1. Mapping
When sequence data are obtained from species for which a reference sequence is available, a homology search tool is required to map the new sequence reads to the reference. In the case of a whole genome sequence or partial genomic data (e.g., BACs, fosmids), we can use either a global alignment tool or a local alignment tool (e.g., BLAST (21), SSAHA (22)). Special tools are being developed for Illumina’s short sequence reads (SOAP (23), MAQ (24)). When transcript data are involved, it is easiest to map the data against a (Uni)gene set, as this should result in an ungapped alignment. When such a set is not available, it can be mapped to genomic data using a spliced alignment tool (e.g., Spidey (25), BLAT (26)). If multiple locations are found as candidate regions, the read should be tagged as unreliable or put aside. Optionally, a tool can be used to ignore repeated sequence regions in the query sequence read, as these will likely result in spurious candidate matches on the reference sequence. A wellknown tool to perform this task is RepeatMasker (27), which requires a list of known repeat sequences as input.
3.1.2. Alignment
The result of the previous step is the aligned position of the new sequence read on the reference. Either a multiple alignment or a pairwise alignment can be used for the evaluation of base constitution on each position and the consequent SNP identification. The alignment tool should deal with gaps to enable indels. Such an alignment is visualized in Fig. 4.2. The best-known and most widely used software tools for multiple alignment are Phrap (28) and CAP3 (29). Recent studies based on Illumina GA data report the development of new tools specifically aimed at the alignment of short sequences (30).
3.1.3. SNP Identification and Analysis
The alignment allows for the identification of sequence variants at each base position represented by multiple reads. The following criteria apply to the classification of these sequence variants as putative polymorphisms: l
Number of reads in the alignment The more reads are available in representing a certain region, the higher the chance of finding a polymorphism and the more accurate a candidate polymorphism can be assessed.
l
Number of reads per allele A sequence variant (allele) can be distinguished from a sequencing error when it is confirmed by multiple reads. The higher the number of reads per allele, the higher the probability of it being a true polymorphism.
l
Sample origin of the reads Sequence variants should be consistent between reads from the same sample. This again allows for the confirmation of a certain allele. There is an exception for heterozygote
Computational Approaches to SNP Discovery and Analysis
79
Fig. 4.2. Example of an alignment of tomato expressed sequence tags. Mismatches are marked by a square box and indicate putative polymorphisms (Keygene Cluster Viewer).
individuals when they are screened on a platform where the sequence read is the result from multiple fragments (e.g., Sanger sequencing). In this case one sample can produce two different alleles – or even more alleles in the case of polyploid samples. l
Base quality per read (when available) Another way to discriminate sequencing errors from true polymorphisms is to look at the sequence quality of the specific base called. If the assessment of the base is made with high confidence, then the probability of a true polymorphism over a sequencing error is high. In the case of heterozygote reads, also the score of the second most likely base is informative.
80
van Oeveren and Janssen l
Distance to the nearest neighboring SNP For both high confidence in SNP calling and possible conversion into a SNP assay for detection, an isolated position of the SNP is favorable. When a candidate SNP is flanked with other candidate SNPs, such an abundant SNP region is likely to originate from an erroneous alignment, e.g., repetitive sequence regions or paralogues. Such regions should be discarded from further SNP mining or attempts can be made to split the cluster with more stringent settings.
l
Length of flanking sequence Beginning-of-sequence and end-of-sequence reads often suffer from low quality base calling; therefore, candidate polymorphisms from these regions have a high probability of being sequencing errors. Setting a minimum length for the flanking region of a candidate SNP can reduce this risk. Additionally this criterion will secure the required flanking sequence information for the conversion of a SNP to a SNP assay (e.g., for primer development).
An example of such an alignment and the origin of some of the parameters specified above are given in Fig. 4.3. 3.2. De Novo Sequence Data 3.2.1. Clustering
3.2.2. Alignment and Assembly
With de novo sequence data, an additional step has to be made to group sequence reads which belong to the same region on the genome. Several sequence and assembly tools, as mentioned in the next section, can perform this task, as they split up input data sets which cannot be assembled into a single contig. However, when the number of reads becomes too large, this will be a time-consuming step. Therefore, specialized tools have been developed to perform an initial segregation of sequence fragments into more or less homologous groups, which can be further decomposed into clusters of unique origin by assembly tools. To prevent the erroneous creation of large clusters, it is wise to perform a preanalysis step of masking low-complexity and repeat regions, usually by using RepeatMasker (27). Examples of cluster tools are d2cluster (31), TeraClu (TimeLogic1), and TGICL, a cluster tool developed by TIGR to create Gene Indices (32). After the clustering step has been completed, each cluster has to be processed to align all reads within the cluster. All bases from different reads representing the same position on the gene or genome are aligned and can be easily compared. If some fragments cannot be properly aligned, they do not belong to a single cluster and are split into a second cluster. See Section 3.1.2 for the list of alignment tools.
Computational Approaches to SNP Discovery and Analysis
81
Fig. 4.3. Example of a multiple sequence alignment (MSA) with putative SNPs and sample related properties from a CRoPS analysis. (Reproduced by permission from (16)).
3.2.3. Polymorphism Identification and Analysis
After individual reads have been clustered into homologous groups aligned, the final step of polymorphism identification is identifying variations in the alignment and applying a scoring scheme (see Section 3.1.2).
82
van Oeveren and Janssen
3.3. Published SNP Mining Tools
The following sections give an overview of the main software tools for SNP discovery that have been released (see Notes 2 and 3). They can be split into two classes: 1. Tools for analyzing known (resequenced) fragments, often starting from trace files or using other types of base-quality scores in addition to the nucleotide sequence. They include the commonly used PolyBayes (33) and PolyPhred (34) packages. These will be discussed in Section 3.3.1 along with more recent alternatives. 2. Tools aimed at the de novo SNP mining approach, where no reference sequence is available. These tools require a cluster step to group homologues. They include AutoSNP (35), QualitySNP (36), and MAVIANT (37) and are discussed in Section 3.3.2. As these tools do not make any assumptions concerning the origin of the data, they can also be applied to the reference-based sequence data. These tools will be suboptimal when fragments from multiple regions are involved, but are good alternatives to the tools mentioned in Section 1 when a single amplicon is considered. A special section describes the CRoPS technology (see Section 3.3.3), which offers the unique combination of selective target preparation for de novo HT sequencing on a Roche GS platform and additional SNP discovery.
3.3.1. Reference Sequence Based Tools 3.3.1.1. PolyBayes
PolyBayes was developed by Gabor Marth (33) and was one of the first tools to systematically and in an automated way exploit the abundance of (Sanger) sequence data for SNP mining. It uses both reference sequence data for mapping and base-quality scores (as produced by Phred). The ‘‘Bayes’’ part in the name refers to the Bayesian approach in calculating at first the probability of a sequence cluster representing paralogue or true allelic variants and secondly the likelihood of the site being a true SNP (PSNP). The latter is calculated from the coverage redundancy, base-quality values, and a priori estimate of polymorphism rate. Further requirements are prior estimates of the level of paralogous genes and the expected polymorphism rates between versus within paralogue sets. The PolyBayes method was tested on human EST data, mapping to a 1-Mbp region. This yielded 54 SNP candidates from 1,365 ESTs covered by 147 contigs. Fifty-six percent of the candidates could be confirmed in a validation experiment on 36 candidate SNPs. Good results were obtained for clusters with very shallow coverage, even for one single EST mapping to the reference sequence (33).
Computational Approaches to SNP Discovery and Analysis
83
A study in EST data from white spruce reported a validation rate of 85% for SNPs with a PSNP of 0.95 or more and a falsenegative rate of only 1% (38). PolyBayes has also been used on NGS data from both 454/ Roche (39) and Solexa/Illumina (30) machines. As PolyBayes was developed for Sanger sequencing, adaptations are necessary to optimize the tool for the characteristics of these new sequencing machines. PyroBayes (40) has been developed for the Roche GS-20 machine and another adaptation of the PolyBayes algorithm has been created to cope with the large number of short reads from the Illumina GA machine ((30), GigaBayes/PolyBayesShort (41)). PolyFreq is an improved version of PolyBayes, aimed at SNP mining in clusters with deep coverage (42). The main difference from PolyBayes is the way it aligns: whereas PolyBayes creates a multiple alignment of all clustered reads, PolyFreq performs only pairwise alignments of ‘‘new’’ reads with the reference region. In a validation trial the numbers of false negatives and false positives were lower for PolyFreq than for PolyBayes. 3.3.1.2. PolyPhred
Another of the early methods was developed by Nickerson et al. (34) for fluorescence-based detection and genotyping of SNPs, particularly for heterozygous diploid samples. It starts from chromatograms which result from Sanger sequencing of targeted amplicons. In contrast to PolyBayes, it does not work from basecalling quality alone and especially identifies sites with high second base peaks. These sites have low base-quality scores but are indications for a heterozygote locus, thus a SNP. It has been further improved by Stephens et al. (43) to achieve significantly lower false discovery rates.
3.3.1.3. NovoSNP
As an alternative to PolyPhred, novoSNP was developped by Weckx et al. (44) for resequenced regulatory regions to improve false detection rates. The algorithm has two major steps: a detection and a validation supported by a graphical user interface. The input is a reference sequence combined with a set of trace files, which are aligned by BLAST. Mismatches are identified and rated in a cumulative way by four different scores, using peak heights, differences between peaks, and agreement between forward and reverse reads. The algorithm was applied to a set of over 10,000 trace files originating from two human gene regions and the performance was compared with that of both PolyPhred and PolyBayes. It outperformed the two alternative methods in both false-negative and false-positive rates. The latest release (45) includes a database to keep track of the status of sequence variants and annotation.
84
van Oeveren and Janssen
3.3.1.4. SNPdetector
SNPdetector (20) is aimed at automating the manual review step in the SNP mining procedure, as an alternative to PolyPhred. It starts from base calling quality scores (Phred) on specific amplicon fragments. These are aligned using SIM (46), allowing for substantial sequence variation. Next, polymorphisms are identified in regions of high base quality (neighborhood quality standard (12)). Finally, heterozygote genotypes are identified and SNPs evaluated by means of comparing forward and reverse sequencing data. Experimental data in mouse and human resulted in 1,178 and 11,000 SNPs respectively. Validation in mouse showed a falsepositive rate of less than 1% and a false-negative rate of 2.6%. These values are much lower than those obtained with PolyPhred and novoSNP (20).
3.3.1.5. Other Tools and Pipelines for ReferenceBased SNP Mining
SNP-PHAGE (47) covers a complete pipeline for the analysis of Sanger resequenced sequence tagged site fragments and combines Phred, Phrap, PolyBayes, and PolyPhred with storage in a database. It provides a Web interface for viewing alignments and putatitve SNPs. InSNP (48) is similar to PolyPhred and novoSNP. It reports a higher sensitivity for mining indels. SsahaSNP (49) is designed to map large data sets from whole genome shotgun sequencing by using a fast search algorithm (22) and consequently identifies SNPs and indels. It has been used in the international SNP consortium for identifying and mapping human SNPs (50). The authors are working on a version which is suited to analyzing NGS data and which can manage spliced alignments (51).
3.3.2. De Novo Sequence Based Tools
AutoSNP is one of the first tools for SNP discovery aimed at exploiting the large number of ESTs available in the public domain (35). The input consists of large sets of ESTs of often unknown gene origin and without trace files or base quality data. The ESTs are first clustered with d2cluster and additionally aligned and assembled with CAP3. Two parameters are used for putative SNP identification: l SNP redundancy score is the minimum number of reads per allele (two by two).
3.3.2.1. AutoSNP
SNP cosegregation score is the percentage of other SNPs with an identical segregation pattern. The algorithm has been tested on over 100,000 maize ESTs originating from five genotypes. These ESTs were clustered in 13,000 groups, of which 6,000 contained at least four reads. A total of 3,500 candidate SNPs were mined. From a test set of 264 candidates, 91% were confirmed. l
Computational Approaches to SNP Discovery and Analysis
85
AutoSNP was built into a pipeline called SNPServer (52). It can either take EST sets or single fragments as input. For the latter case, the fragment is blasted against a nucleotide database to get matching sequences. SNPServer is available as a Web-based tool and delivers HTML output files. 3.3.2.2. QualitySNP
This tool was developed to cope with EST data sets without basecalling information and to handle the paralogue issue (36). It uses CAP3 for clustering and alignment. All polymorphisms are identified and haplotypes are assessed. A metric for the variation in the number of SNPs between haplotypes is used to discriminate between true allelic variants and paralogues. Additionally, a SNP confidence score is determined according to the number of reads per allele and the base quality. With use of open reading frame prediction tools, SNPs are classified as synonymous or nonsynonymous. Finally, the results are put in a database. The algorithm has been tested on human, chicken, and potato ESTs and yielded thousands of SNPs. Validation in human data resulted in a 35% confirmation level from dbSNP data and it outperformed AutoSNP by a factor 4. QualitySNP is also available in a Web-based approach in the form of HaploSNPer (53).
3.3.2.3. MAVIANT
MAVIANT (37) combines PolyBayes with a clustering step for de novo sequence fragments. Like AutoSNP, it is aimed at ESTs from unknown gene origins, which are grouped by TeraClu into sets originating from the same transcript, and uses Phrap for alignment. The algorithm has been tested on over 800,000 pig ESTs from which 4,700 nonsynonymous SNPs were identified with a validation rate of 59%(37).
3.3.2.4. Other Tailor-Made Pipeline Tools for SNP Mining
Useche et al. (54) created a SNP pipeline to obtain candidate SNPs from maize EST data. Their tool is based upon Phrap and CAT for alignment and also uses PolyBayes for the SNP mining process. Over 2,000 SNPs were mined from a set of 68,000 maize ESTs with PSNP > 0.99 and the aid of visual inspection. All current tools for SNP mining from public ESTs are intended for Sanger sequence data as these have been the predominant type. With the appearance of the NGS technologies we expect a shift towards tools which are fit to handle the new type of sequence data.
3.3.3. CRoPS
CRoPS (Complexity Reduction of Polymorphic Sequences) (16) combines AFLP (15) with the HT sequencing technology on the Genome Sequencer (GS-20/GS-FLX) from Roche Applied Science. It is an example of how NGS techniques can be successfully applied for SNP discovery.
86
van Oeveren and Janssen
AFLP is a technology for selective amplification of restriction fragments. With CRoPS, two or more genetically diverse samples are prepared by AFLP and each fragment is sequenced to a fivefold to tenfold average coverage. This coverage can be controlled by variation in the choice of restriction enzymes and selective bases. Through the addition of different sample ID tags, many samples can be pooled and the resultant sequences can be easily identified and separated. The CRoPS bioinformatics pipeline recognizes the sample tag, removes the AFLP adaptor and sample ID tag, restores the restriction site(s) in silico, and then clusters and assembles the fragments with TGICL. This results in multiple alignments in which polymorphic positions can be identified (Fig. 4.3). Several mining options are used to identify candidate polymorphisms (both SNPs and indels). These include: l
Allele frequency per sample (different thresholds for homozygotes and heterozygotes)
l
Minimum flank length on both sides of the polymorphic base
l
Minimum length of flanks without any other (neighbor) SNPs
l
Quality of a polymorphic base to be considered as a SNP (quality is defined as a value between 0 and 1 and depends on the ratio between the major and the minor allele and the depth of the multiple alignment
l
Maximum number of reads per SNP position (this setting allows one to filter out potential repeats/duplicates/paralogues) (see Note 3)
The CRoPS technology can be applied on both genomic DNA and cDNA. Although CRoPS is mainly applied in plants species, it is widely applicable.
4. Notes
1. Most tools are based on Sanger sequences as this still constitutes the bulk of the sequence data available. New tools are being developed specifically aimed at the analysis of NGS data. 2. In this paper we provide an overview of the available methods for SNP discovery from DNA sequence data. Although we do not intend to be comprehensive, the most common procedures and tools are presented. For a schematic overview, the methods and their key characteristics are displayed in Table 4.1.
Sequence platform
Sanger
Sanger/ Illumina/ Roche
Sanger
Sanger
Sanger
Sanger
Sanger
Sanger/ Illumina/ Roche
Sanger
PolyBayes
GigaBayes
PolyPhred
PolyFreq
novoSNP
SNP-PHAGE
SNPdetector
ssahaSNP/ ssaha_pileup
InSNP
Reference-based tools
SNP mining tool
Yes
Yes
Seq + qual files Trace files
Yes
Yes
Yes
Yes
Trace files
Trace files
Trace files
Trace files
Yes
Yes
Seq + qual files Trace files
Yes
Base quality required
Trace files
Input
No
Yes
No
Yes
No
Yes
No
Yes
Yes
Paralogue identification
Yes
No
No
Yes
No
No
No
No
No
Visual check required
http://www.mucosa.de/insnp/ (continued)
http://www.sanger.ac.uk/Software/ analysis/SSAHA2/
ftp://ftp1.nci.nih.gov/pub/SNPdetector
http://bfgl.anri.barc.usda.gov/ML/ snp-phage
http://www.molgen.ua.ac.be/bioinfo/ novosnp/
http://deepc2.psi.iastate.edu/aat/ PolyFreq//
http://droog.gs.washington.edu/ PolyPhred.html/
http://bioinformatics.bc.edu/marthlab/
http://bioinformatics.bc.edu/marthlab/ PolyBayes
URL
Table 4.1 Overview of the single nucleotide polymorphism (SNP) mining tools and their characteristics
Computational Approaches to SNP Discovery and Analysis 87
Sanger
MAVIANT
CRoPS
Roche GS
Any/Sanger
QualitySNP/ HaploSNPper
Technology
Any
Sequence platform
AutoSNP/ SNPserver
De novo tools
SNP mining tool
Table 4.1 (continued)
Sequence
Trace files
Sequence and traces
Sequence
Input
No
Yes
Opt
No
Base quality required
No
No
Yes
Yes
Paralogue identification
No
Yes
No
No
Visual check required
http://www.keygene.com/key-gene/techsapps/technologies_crops.php
http://snp.agrsci.dk/maviant/
http://www.bioinformatics.nl/tools/ haplosnper
http://hornbill.cspp.latrobe.edu.au/ snpdiscovery.html/
URL
88 van Oeveren and Janssen
Computational Approaches to SNP Discovery and Analysis
89
3. We would like to mention that the methods presented herein for SNP discovery represent powerful means to exploit sequence data that are often already present and freely available or are becoming available at high speed with NGS. Different tools may be favorable, depending on the data source. Some problems will remain, however, which can hamper the flawless detection of SNPs, such as low minor allele frequencies and SNPs in repetitive or paralogous regions. However, by setting appropriate SNP filter criteria, one can always reliably mine a significant fraction of SNPs.
Acknowledgements We thank Jifeng Tang, John Smith, Mike Cariaso, and Michiel J.T. van Eijk for their comments on the manuscript. The AFLP1 and CRoPS1 technologies are covered by patents and patent applications owned by Keygene NV. AFLP and CRoPS are registered trademarks of Keygene NV. Other (registered) trademarks are the property of the respective owners. References 1. 2. 3. 4.
5.
6.
7.
8.
http://www.ncbi.nlm.nih.gov/ http://www.ebi.ac.uk/embl/ http://www.ddbj.nig.ac.jp/ Buetow, K. H., Edmonson, M. N. and Cassidy, A. B. (1999) Reliable identification of large numbers of candidate SNPs from public EST data. Nat. Genet. 21, 323–325. Picoult-Newberg, L., Ideker, T. E., Pohl, M. G., Taylor, S. L., Donaldson, M. A., Nickerson, D. A. and Boyce-Jacino, M. (1999) Mining SNPs from EST databases. Genome Res. 9, 167–174. Taillon-Miller, P., Gu, Z., Li, Q., Hillier, L. and Kwok, P. Y. (1998) Overlapping genomic sequences: a treasure trove of singlenucleotide polymorphisms. Genome Res. 8, 748–754. Sachidanandam, R., Weissman, D., Schmidt, S. C., Kakol, J. M., Stein, L. D., Marth, G. et al. (2001) A map of human genome sequence variation containing 1.42 million single nucleotide polymorphisms. Nature 409, 928–933. Margulies, M., Egholm, M., Altman, W. E., Attiya, S., Bader, J. S., Bemben, L. A., Berka, J. et al. (2005) Genome sequencing
9.
10.
11.
12.
13. 14.
in microfabricated high-density picolitre reactors. Nature 437, 376–380. Bentley, D.R. (2006) Whole-genome resequencing. Curr. Opin. Genet. Dev. 16, 545–552. Fu, Y., Peckham, H. E., McLaughlin, S. F., Ni, J. N., Rhodes, M. D., Malek, J. A., McKernan, K. J. and Blanchard, A. P. (2008) SOLiDTM system sequencing and 2 base encoding. Cold Spring Harbor, Biology of Genomes 2008. Wang, D. G., Fan, J. B., Siao, C. J., Berno, A., Young, P., Sapolsky, R. et al. (1998) Large-scale identification, mapping, and genotyping of single-nucleotide polymorphisms in the human genome. Science 280, 1077–1082. Altshuler, D., Pollara, V. J., Cowles, C. R., Van Etten, W. J., Baldwin, J., Linton, L. and Lander, E. S. (2000) An SNP map of the human genome generated by reduced representation shotgun sequencing. Nature 407, 513–516. http://www.ncbi.nlm.nih.gov/sites/ entrez?db=unigene http://compbio.dfci.harvard.edu/tgi
90
van Oeveren and Janssen
15. Vos, P., Hogers, R., Bleeker, M., Reijans, M., van de Lee, T., Hornes, M., Frijters, A., Pot, J., Peleman, J., Kuiper, M. et al. (1995) AFLP: a new technique for DNA fingerprinting. Nucleic Acids Res. 23, 4407–4414. 16. Van Orsouw, N. J., Hogers, R. C. J., Janssen, A., Yalcin, F., Snoeijers, S., Verstege, E., Schneiders, H., Van der Poel, H., Van Oeveren, J., Verstegen, H. and Van Eijk, M. J. T. (2007) Complexity Reduction of Polymorphic Sequences (CRoPSTM): a novel approach for large-scale polymorphism discovery in complex genomes. PLoS ONE 11, e1172. 17. Peterson, D. G., Wessler, S. R. and Paterson, A. H. (2002) Efficient capture of unique sequences from eukaryotic genomes. Trends Genet. 18, 547–550. 18. Ewing, B., Hillier, L., Wendl, M. C. and Green, P. (1998) Base-calling of automated sequencer traces using phred. I. Accuracy assessment. Genome Res. 8, 175–185. 19. Ewing, B. and Green, P. (1998) Base-calling of automated sequencer traces using phred. II. Error probabilities. Genome Res. 8, 186–194. 20. Zhang, J., Wheeler, D. A., Yakub, I., Wei, S., Sood, R., Rowe, W., Liu, P. P., Gibbs, R. A. and Buetow, K. H.. (2005) SNPdetector: a software tool for sensitive and accurate SNP detection. PLoS Comput. Biol. 1, 395–404. 21. Altschul, S. F., Gish, W., Miller, W., Myers, E. W. and Lipman, D. J. (1990) Basic local alignment search tool. J Mol. Biol. 215, 403–410. 22. Ning, Z., Cox, A. J. and Mullikin, J. C. (2001) SSAHA: A fast search method for large DNA databases. Genome Res. 11, 1725–1729. 23. Li, R., Li, Y., Kristiansen, K. and Wang, J. (2008) SOAP: short oligonucleotide alignment program. Bioinformatics 24, 713–714. 24. Li, H. (2008) Maq. http://maq.sourceforge.net/maq-man.shtml#intro 25. Wheelan, S. J., Church, D. M. and Ostell, J. M. (2001) Spidey: A Tool for mRNA-toGenomic Alignments. Genome Res. 11, 1952–1957. 26. Kent, J. K. (2002) BLAT—The BLAST-Like Alignment Tool. Genome Res. 12, 656–664. 27. Smit, A. F. A., Hubley, R. and Green, P. RepeatMasker Open-3.0. 1996–2004, http://www.repeatmasker.org 28. Green, P. http://www.phrap.org 29. Huang, X. and Madan, A. (1999) CAP3: A DNA sequence assembly program. Genome Res. 9, 868–877.
30. Hillier, L. W., Marth, G. T., Quinlan, A. R., Dooling, D., Fewell, G. et al. (2008) Wholegenome sequencing and variant discovery in C. elegans. Nat. Methods 1179, 1–6. 31. Burke, J., Davison, D. and Hide, W. (1999) d2_cluster: a validated method for clustering EST and full-length cDNA sequences. Genome Res. 9, 1135–1142. 32. Pertea, G., Huang, X., Liang, F., Antonescu, V., Sultana, R., Karamycheva, S., Lee, Y., White, J., Cheung, F., Parvizi, B., Tsai, J. and Quackenbush, J. (2003) TIGR Gene Indices clustering tools (TGICL): a software system for fast clustering of large EST datasets. Bioinformatics 19, 651–652. 33. Marth, G. T., Korf, I., Yandell, M. D., Yeh, R. T., Gu, Z., Zakeri, H. et al. (1999) A general approach to single-nucleotide polymorphism discovery. Nat. Genet. 23, 452–456. 34. Nickerson, D. A., Tobe, V. O. and Taylor, S. L. (1997) PolyPhred: automating the detection and genotyping of single nucleotide substitutions using fluorescence-based resequencing. Nucleic Acids Res. 25, 2745–2751. 35. Batley, J., Barker, G., O’Sullivan, H., Edwards, K.J. and Edwards, D. (2003) Mining for single nucleotide polymorphisms and insertions/deletions in maize expressed sequence tag data. Plant Physiol. 132, 84–91. 36. Tang, J., Vosman, B., Voorrips, R. E., van der Linden, C. G. and Leunissen, J. A. (2006) QualitySNP: a pipeline for detecting single nucleotide polymorphisms and insertions/deletions in EST data from diploid and polyploid species. BMC Bioinformatics 7, 438. 37. Panitz, F., Stengaard, H., Hornshøj, H., Gorodkin, J., Hedegaard, J., Cirera, S. et al. (2007) SNP mining porcine ESTs with MAVIANT, a novel tool for SNP evaluation and annotation. Bioinformatics 23, 387–391. 38. Pavy, N., Parsons, L. S., Paule, C., MacKay, J. and Bousquet, J. (2006) Automated SNP detection from a large collection of white spruce expressed sequences: contributing factors and approaches for the categorization of SNPs. BMC Genomics 7, 174. 39. Barbazuk, W. B., Emrich, S. J., Chen, H. D., Li, L. and Schnable, P. S. (2007) SNP discovery via 454 transcriptome sequencing. Plant J. 51, 910–918. 40. Quinlan, A. R., Stewart, D. A., Strømberg, M. P. and Marth, G. T. (2008) Pyrobayes:
Computational Approaches to SNP Discovery and Analysis
41. 42.
43.
44.
45.
46.
47.
48.
an improved base caller for SNP discovery in pyrosequences. Nat. Methods 5, 179–181. Marth, G. T. et al. (2008) http://bioinformatics.bc.edu/marthlab/PbShort Wang, J. and Huang, X. (2005) A method for finding SNPs with allele frequencies in sequences of deep coverage. BMC Bioinformatics 6, 220. Stephens, M., Sloan, J. S., Robertson, P. D., Scheet, P. and Nickerson, D. A. (2006) Automating sequence-based detection and genotyping of SNPs from diploid samples. Nat. Genet. 38, 375–381. Weckx, S., Del-Favero, J., Rademakers, R., Claes, L., Cruts, M., De Jonghe, P., Van Broeckhoven, C. and De Rijk, P. (2005) novoSNP, a novel computational tool for sequence variation discovery. Genome Res. 15, 436–442. De Rijk, P. and Del-Favero, J. (2007) novoSNP3: variant detection and sequence annotation in resequencing projects. Methods Mol. Biol. 396, 331–344. Huang, X. Q., Hardison, R. C. and Miller, W. (1990) A space-efficient algorithm for local similarities. Comput. Appl. Biosci. 6, 373–381. Matukumalli, L. K., Grefenstette, J. J., Hyten, D. L., Choi, I. Y., Cregan, P. B. and Van Tassell, C. P. (2006) SNPPHAGE – high throughput SNP discovery pipeline. BMC Bioinformatics 23, 468. Manaster, C., Zheng, W., Teuber, M., Wa¨chter, S., D¨oring, F., Schreiber, S. and
49.
50.
51.
52.
53.
54.
91
Hampe, J. (2005) InSNP: a tool for automated detection and visualization of SNPs and InDels. Hum. Mutat. 26, 11–19. Ning, Z., Caccamo, M. and Mullikin, J. C. (2005) ssahaSNP – a polymorphism detection tool on a whole genome scale. 2005 IEEE Computational Systems Bioinformatics Conference – Workshops (CSBW’05) 251–254. The International SNP Map Working Group (2001) A map of human genome sequence variation containing 1.42 million single nucleotide polymorphisms. Nature 409, 928–933. Ning, Z., Gu, Y., Blackburne, B., Ponstingl, H. and Cox, A. (2008) Alignment and de novo assembly of transcriptome reads from Solexa sequencing. ISMB2008 poster P08. Savage, D., Batley, J., Erwin, T., Logan, E., Love, C. G., Lim, G. A., Mongin, E., Barker, G., Spangenberg, G. C. and Edwards, D. (2005) SNPServer: a real-time SNP discovery tool. Nucleic Acids Res. 33, 493–495. Tang, J., Leunissen, J.A., Voorrips, R.E., van der Linden, C.G. and Vosman B. (2008) HaploSNPer: a web-based allele and SNP detection tool. BMC Genet. 9, 23. Useche, F. J., Gao, G., Hanafey, M. and Rafalski, A. (2001) High-throughput identification, database storage and analysis of SNPs in EST Sequences. Genome Inform. 12,194–203.
.
Chapter 5 Next-Generation Sequencing Methods: Impact of Sequencing Accuracy on SNP Discovery Eugene Y. Chan Abstract The advent of next-generation sequencing technologies has spurred remarkable progress in the field of genomics. Whereas traditional Sanger sequencing has yielded the first complete human genome sequence, next-generation methods have been able to resequence several human genomes. In this manner, nextgeneration approaches have powerful capabilities for understanding human variation. The throughput for these approaches is often measured in billions of base pairs per run, astounding numbers when compared with the millions of base pairs per day generated by automated capillary DNA sequencers. However, unlike traditional Sanger dideoxy sequencing, these methods have lower accuracy and shorter read lengths than the dideoxy gold standard. Are these limitations offset by the higher throughputs? An in-depth look at the single read and composite accuracy of these methods is presented. The stringent requirements for single nucleotide polymorphism (SNP) discovery utilizing these approaches is discussed along with a review of studies that have successfully employed next-generation sequencing methods for large-scale SNP discovery. Ultimately, the application of these ultra-high-throughput sequencing methods for SNP discovery will open up new horizons for understanding human genomic variation. Key words: Next-generation sequencing, sequencing technology, single nucleotide discovery, accuracy, Sanger sequencing, polony sequencing, pyrosequencing, cycle extension, 454, Illumina, Helicos, SOLiD.
1. Introduction Next-generation sequencing methods are those that employ massively parallel approaches in sequencing several hundred thousand to millions of reads simultaneously. This is an enormous increase in the number of reads compared with existing capillary sequencers that employ 96-capillary arrays. Currently, the four commercialized next-generation sequencing methods are parallel pyrosequencing in A.A. Komar (ed.), Single Nucleotide Polymorphisms, Methods in Molecular Biology 578, DOI 10.1007/978-1-60327-411-1_5, ª Humana Press, a part of Springer Science+Business Media, LLC 2003, 2009
95
.
96
Chan
picoliter reactors (1), Illumina sequencing (2–4), SOLiD or polony sequencing (5), and single molecule sequencing (6). These methods have enormous breadth in applications, including new applications such as ChIP-seq (7), transcript sequencing (8), microRNA discovery (9, 10), whole genome resequencing (11), and single nucleotide polymorphism (SNP) discovery (2, 11, 12). The use of these methods for SNP discovery is by far the most demanding application since it requires a high accuracy. In contrast to ChIPseq or transcript sequencing applications, which rely on sequence identification for various reads, SNP discovery requires identification of individual base pairs and their differences. Any error in sequencing leads to incorrect SNP identification since SNPs occur at a high frequency, at every few hundred to thousand base pairs (11, 13, 14). Furthermore, the abundance of other variations in the human genome, including copy number variations (15), insertions/deletions (indels), and rearrangements, makes the task even more challenging. The shorter read lengths of next-generation methods and the abundance of repeated sequences in the human genome create unique issues in being able to accurately cover all parts of the genome for SNP discovery. Nonetheless, some nextgeneration sequencing methods have been able to bypass these challenges and have been able to add many new SNPs to dbSNP (see Note 1) (16). The following describes the accuracy, error, and coverage of next-generation sequencing methods as required for SNP discovery.
2. Sequencing Accuracy What is the required accuracy for SNP discovery? Another way to think of the question is to understand how many errors can be tolerated in a given stretch of DNA. Since single base changes can give rise to significant phenotypic change, errors cannot be missed. One error in 100,000 bp would yield an error rate of 99.999%, a value set by the Archon Genomics X Prize (see Note 2) (17). Shendure et al. (5), in their polony paper, suggested an even more stringent number, a consensus error rate of one in 1,000,000 bp, or an error rate 99.9999%. Any error in sequencing can give rise to false positives or false negatives, leading to challenging downstream studies. The accuracies of the next-generation sequencing methods are compared with the accuracy of Sanger dideoxy sequencing. The universally accepted accuracy metric for Sanger dideoxy sequencing is given in terms of terms of Phred values (18, 19). Since each sequencing method has its own quality scores, the raw accuracy of
.
Next-Generation Sequencing Methods
97
each sequencing method needs to be presented in a standard manner. Each method’s performance relative to different errors, including in relation to homopolymers, base substitutions, and deletion–insertion polymorphisms, needs to be assessed. The massively parallel capabilities of next-generation methods allows for vast oversampling to correct for errors. Their short read lengths, however, lead to some parts of the genome being underrepresented. The inability to cover certain parts of the genome leads to large unanalyzed stretches of SNPs. The degree of this problem is explored for each of the methods, as is the redundancy required for calling SNPs in genome-scale efforts.
3. Raw Sequencing Accuracy The accuracy of a single-pass sequencing read is its raw sequencing accuracy. Any accuracy figure that is a result of a composite or requires averaging is not a raw read. Raw accuracy allows the methods to be compared directly with Sanger dideoxy sequencing. It is an excellent measurement of each method’s chemistry, fluorescence readout, process, and base-calling software. Ultimately, the more robust the raw reads, the fewer the required redundant reads. The limit to raw accuracy is the fidelity attained by nature’s polymerases, which is on the order of 105–107 per base pair for commercially available polymerases (20, 21). Polymerases with 30 !50 exonuclease activity have fidelity values closer to 107 errors per base pair. Given that all existing commercially available methods utilize a polymerase for readout or amplification, it is reasonable to say that this represents the ideal sequencing raw accuracy. Sanger sequencing, as performed by existing capillary sequencers with fluorescent dye terminators, sets a high bar for raw accuracy. All base pairs that are accepted as reads have an accuracy greater than Phred Q20, or 99% accuracy (18, 19). For instance, the Applied Biosystems 3730xl in its long read mode can generate more than 800 bp of sequence at or above the Phred Q20 threshold (22). In fact the majority of base pairs in any given long read run, especially those between 100 and 700 bp, have Phred scores that approach Q50, or 99.999% (22, 23). Its performance is fairly constant across a broad range of different templates (23). The high confidence in the capillary readout makes the approach robust for sequencing and SNP discovery. PCR is utilized in Sanger sequencing and the majority of nextgeneration sequencing methods. It is important to understand the impact of this step on overall sequencing accuracy. After PCR, a population of different molecules are generated, a fraction of
.
98
Chan
which contain errors. The final fraction of DNA containing at least one PCR-derived mutation, given as F, determines the significance of the error. For instance, a PCR that has F ¼ 0.50, with an error propagated at one position early in the reaction, would yield a heterozygote call upon Sanger sequencing. In contrast, the same reaction with F ¼ 0.1 would yield a correct call since 90% of the signal would be from the original wild-type sequence. The relationship between F and polymerase fidelity (f), PCR product size (b), and number of DNA doublings (d) is given by the equation F ¼ 1 – e-bfd, as outlined by Keohavong et al. (24). For a 1,000-bp target and 20 doublings, a polymerase with fidelity equal to 107 would yield F 0.2%. For 106, F 2.0%; for 105, F 18%; and for 104, F 86%. For high-fidelity polymerases, such as Phusion polymerase, which performs with 4.4 107 errors per base pair, about 0.9% of the fragments would yield an error. Although the use of a high-fidelity polymerase is advantageous, errors are still present, especially in next-generation sample preparation methods that have two different PCR steps. In these methods, the first PCR step generates linkers on the ends of genomic DNA fragments. A second PCR step then utilizes these linker-amplified molecules to clonally amplify each one of them. Errors in the first PCR step are propagated in the second step. For Phusion polymerase, about 0.9% of the 1,000-bp fragments have an error, corresponding to 105 errors per base pair. From this calculation, it is clear that sample preparation errors are present when high-fidelity polymerases are used. However, this error is small when compared with readout errors, which can be between 10–1 and 10–2 for some methods. Table 5.1 summarizes the raw accuracy for the four next-generation sequencing methods compared with Sanger sequencing. The values represent data from single reads and exclude those derived from averaging, bidirectional reads, and filtered data. For each method, a range is reported. The range corresponds to raw accuracy
Table 5.1 Comparison of raw accuracies and tabulation of read lengths Method
Raw accuracy range
Sanger
99.0% to > 99.999% (22, 23)
454
96.0% (1, 31) to 97.0% (25)
Illumina
96.2% to 99.7 (28)
SOLiD
99.0% to > 99.9% (22)
Helicos
93.0 to 97.0% (6)
.
Next-Generation Sequencing Methods
99
values within a sequencing read. For instance, the accuracy in automated dideoxy sequencing is greater than Phred Q50, or 99.999%, for most of its bases. Dideoxy sequencing accuracy at positions less than 100 bp and greater than 700 bp decreases to Phred Q20, or 99.0%. Errors increase with base position in all the methods. For Sanger sequencing, larger DNA fragments are more difficult to resolve, leading to decreased quality scores at higher positions, and ultimately limiting read length. For the next-generation sequencing methods, errors accumulate because of asynchronous chemistries or imperfect fluorescence detection, as illustrated in Fig. 5.1 for the Illumina method. For instance, a 98% readout efficiency at each base pair leads to difficult calls with increasing length. At the 35th base pair, the readout for fewer than half the molecules would be correct, since 0.9835 ¼ 0.49. Higher readout efficiencies are challenging because, in fact, there are numerous steps in the read out of each base pair. The Helicos and 454 methods require at least eight steps to read out each base pair, including multiple wash and nucleotide addition steps. The large number of steps that need to be optimized makes it difficult to get more accurate results at longer reads. The read length limit for these methods is ultimately dictated by the drop off in base pair quality, which is 250 bp for parallel pyrosequencing, 36 bp for Illumina, 36 bp for SOLiD, and 25 bp for Helicos.
Fig. 5.1. Error rate increases with base pair position for Illumina Genome Analyzer. Data are from 27-mer Bacillus vulgaris reads and 32-mer Helicobacter acinonychis genome sequencing. (From Dohm et al. (28); reprinted with permission from Oxford University Press).
Both Sanger sequencing and SOLiD sequencing have the highest raw accuracy values, at greater than 99% for all the base pairs, corresponding to a minimum of Phred Q20. The higher accuracy for SOLiD is achieved by redundantly
.
100
Chan
reading each base pair twice by a two-base encoding scheme. While it can be argued that it is in fact two separate reads for the overlapping two-base reads, the read being performed on one sequence makes it an individual read. It is interesting to note that the uncorrected raw accuracy for SOLiD without two-base encoding is between 90 and 99% (22), which is similar to the single pass error for the Helicos method, which utilizes a second pass to attain accuracies of 0.1–0.3%. In the Helicos two-pass sequencing approach, the inefficiencies of single fluorophores are bypassed with bidirectional reading.
4. Homopolymer Errors A problem unique to next-generation sequencing methods is contiguous runs of the same base pair, called ‘‘homopolymer repeats.’’ For instance, the sequence 50 AAAAA 30 is an A5 homopolymer. Large stretches of homopolymers exist in the human genome. Homopolymer runs greater than 5 bp in length or more comprise more than 4% of the genome, according to human bacterial artificial chromosome data (25). Up to now, the majority of these have been placed in GenBank via Sanger sequencing, which resolves homopolymers readily in a gel matrix. The ability to read these sequences accurately allows more complete coverage of the genome and thus more complete SNP discovery efforts. The inability to read these sequences leads to a biased set of SNPs that may exclude some potentially important information. The extent of this problem among the next-generation sequencing methods is examined. Both parallel pyrosequencing and Helicos single molecule sequencing have significant difficulties with homopolymer stretches (1, 6). For instance, it has been estimated that 39% of parallel pyrosequencing’s errors are from incorrect analysis of homopolymer regions (26). An analysis of error at each homopolymer length is revealing, as documented by Wicker et al. (27). For A5 motifs, the 454 method has a 3.3% error rate. The error rate increases with longer motifs. For A8 motifs, the method has a 50% error rate. For the Helicos method, Harris et al. (6) described being able to call between 45 and 65% of four-mer homopolymers in single pass reads. The method, however, addresses these issues with its two-pass sequencing approach that employs a voting scheme. In this manner, more than 80% of A, G, and T four-mers are called correctly (Fig. 5.2).
.
Next-Generation Sequencing Methods
101
Fig. 5.2. Two-pass homopolymer error rate for the Helicos single molecule sequencing method. The fraction of homopolymers called correctly decreases with homopolymer length. (From Harris et al. (6); reprinted with permission from AAAS).
Homopolymer errors are unique to next-generation sequencing methods that read signal intensity as a measure of homopolymer length. In parallel pyrosequencing (see Chapter 7 for the details of the method), when a homopolymer stretch is encountered, multiple incorporations of that base occur, releasing a proportionate amount of pyrophosphate that is detected as an intensity signal. The accuracy of the light intensity signal is diminished by asynchronous molecules and by photon shot noise. Photon shot noise follows a Poisson distribution, with the standard deviation equal to the square root of the average photon signal. As longer homopolymers are read, the probability of their Poisson distributions overlapping increases owing to the greater photon shot noise. These two impediments together lead to the errors observed for parallel pyrosequencing. Helicos distinguishes homopolymers on the basis of the inverse of their intensity since adjacent incorporated fluorescent nucleotides have increased quenching. Although the single molecule method does not have the problems of asynchronous reads, it does face a challenge in that it does not read more than a 3-bp homopolymer motif at a time. Longer homopolymers are stitched together in separate read cycles. In contrast, Illumina and SOLiD address the homopolymer challenge through reversible terminators and two-base encoding. Illumina sequencing utilizes chemistry that caps the 30 -OH of modified fluorescent bases so that upon encountering homopolymer stretches only one nucleotide is incorporated at a time. In this manner, bases are only read out one at a time, allowing homopolymer stretches to be counted positionally as opposed to their being measured via measurement intensity. In a detailed analysis of Illumina’s method, Dohm et al. (28) concluded that the method
.
102
Chan
handled homopolymer stretches without increased errors. Given the recent commercial introduction of the SOLiD system, confirmatory studies of the system’s homopolymer accuracy have yet to be performed. Conceptually, and according to the manufacturer (22), the approach should not have any issues. The SOLiD system’s chemistry positionally interrogates two bases at each cycle so there is no ambiguity about homopolymer length during a particular run.
5. Base Substitution Errors and Indels
Knowing the specific type of error introduced by the methods allows selection of the appropriate next-generation sequencing method for a given application. For SNP discovery, the method’s accuracy with respect to base substitution errors is the most important. For discovery of deletion–insertion polymorphisms, microsatellites, and other length-based mutations, accuracy with respect to indels is paramount. The various different methods perform differently with respect to types of errors generated. Illumina derives the majority of its errors from base substitutions. Dohm et al. (28) studied the accuracy of the method across 2.8 million 27-mer reads from Bacillus vulgaris and 12.3 million 36-mer reads from Helicobacter acinonychis. Base substitutions accounted for more than half of the errors. In particular, these were the transversions A ! C, G ! T, and A ! T. Two possible explanations for these base substitution biases are possible. First, the particular polymerase and chemistry utilized by the method could favor the incorporation of pyrimidines in these contexts. Second, as suggested by Dohm et al. (28), there is insufficient spectral discrimination between the fluorophores, leading to bleed-through and miscalled bases. On the other hand, indel errors were infrequent for Illumina sequencing, occurring at a rate of less than 0.01% (28). On the rare chance that a base insertion occurred, it was most likely to be correlated with homopolymer runs of greater than four nucleotides. The addition of a base at a time with reversible terminator chemistry contributes to the accuracy of the method with respect to indels. For parallel pyrosequencing and the Helicos method, the predominant errors are indel errors that arise from incorrect knowledge about readout position. For 454, 63% of total errors were indels, 36% from insertions and 27% from deletions (26). The high rate of indel errors, when compared with the 39% of homopolymer errors (26), suggests that indel errors occur anywhere and this is a fundamental challenge for parallel pyrosequencing’s readout method. The method did better with mismatches, with 16% of total errors from this category (26). The remaining 21% errors were ambiguous base calls and could have led to indel or base substitution errors. For the Helicos
.
Next-Generation Sequencing Methods
103
single molecule sequencing method, the major source of error is deletions, which account for 3–7% of errors on single pass reads. Insertion errors occur at a rate of 0.02–1.1% and base substitution errors at a rate of 0.01–1.0% (6). Of the deletion errors, 3–4% arise from nonemitting nucleotides and undetected events, suggesting that detection efficiency is an area that can be improved through brighter dyes and better chemistries. SOLiD sequencing appears to perform equally well with respect to both types of errors, although independently verified studies are required to fully assess the capabilities of the method. At present, data are sparse for detailed accuracy analyses. Since the two-base encoding strategy of SOLiD sequencing gives good results with homopolymers, the majority of its errors are most likely to arise from base substitution errors.
6. Consensus Accuracy, Redundancy and Coverage
The throughput of the next-generation sequencers allows for more reads and thus more opportunities to correct innate errors in raw accuracy. Overall consensus accuracy is the composite accuracy from sequencing a genomic region many times. Even a sequencing method with 90% raw accuracy can attain 99.999% consensus accuracy by providing fivefold redundancy. In practice, attaining sufficient redundancy with short reads provides challenges since half of the human genome is repeat sequences (13, 14). Repeat sequences prevent short reads from being accurately mapped to the genome. Consequently, areas of the genome remain without adequate coverage, excluding potentially important areas of the genome from analysis. In practice, the actual redundancy required to attain high consensus accuracies (greater than 99.99%) is higher than the theoretically predicated number. For instance, for Illumina sequencing, the required redundancy to attain close to no errors per kilobase pair is 20-fold for bacterial genomes (28). The SOLiD method requires 12-fold sequence redundancy in human germline genome sequencing (22). The Helicos method requires at least 20-fold redundancy for good mutation detection. In their M13 viral genome study, they attained 150-fold redundancies (6). Parallel pyrosequencing requires 7.4-fold redundancy to sequence a diploid human genome (11). For Sanger sequencing, it is commonly accepted that threefold redundancy is sufficient to attain the required accuracies for diploid genomes. It should be noted that these values for parallel pyrosequencing and Sanger sequencing are for diploid genomes, whereas for the others, they are for haploid genomes.
.
104
Chan
Genome coverage is perhaps the most significant issue for the next-generation sequencing methods. Short read length methods, those with 20–30-bp reads, are at a significant disadvantage in analyzing complex genomes and may leave parts of the genome inadequately covered. Short reads do well with smaller, haploid genomes, as seen from the use of the Helicos method with the M13 viral genome, where it attained 100% coverage (6). With bacterial-sized genomes, coverage for short read next-generation methods starts to decrease. In the sequencing of a strain of Escherichia coli with polony technology and a mate-paired library, 30.1 Mb of sequence was generated, with 83.3% of the genome having a twofold or greater redundancy (5). The numbers quickly dropped to 66.9% of the genome with fourfold or greater redundancy (5). The coverage issue for small bacterial genomes can be bypassed by higher throughput. Dohm et al. (28) showed complete coverage utilizing the Illumina platform, but some regions of the genome ended up having more than 350-fold redundancy to ensure full coverage given the GC-rich read biases. Ultimately, the average redundancy required for full bacterial genome coverage will be dictated by sample preparation biases, which are yet to be determined for the SOLiD and Helicos methods. In a whole-genome sequencing effort of Caenorhabditis elegans by Hillier et al. (2), the genome was repeat-masked for short repeats. In this manner, approximately 23% of the genome was excluded from analysis (Fig. 5.3).
Fig. 5.3. Repetitive content in Caenorhabditis elegans. The genome has a large percentage of microrepeats and sequences identified by RepeatMasker. (From Hillier et al. (2); reprinted by permission from Macmillan Publishers Ltd).
.
Next-Generation Sequencing Methods
105
Genome coverage is improved with longer read lengths. Parallel pyrosequencing, with average read lengths of 250 bp, sequenced a human genome with average 7.4-fold redundancy (11). Unlike for biased short reads, the method gave fairly uniform coverage of the human genome in a Poisson distribution (Fig. 5.4). Average redundancy was 3.7-fold for the X chromosome. Twenty nine megabases in 110,000 contigs did not match back to the genome, with 65% of these sequences being identified as satellite DNA and the others in repeat-rich regions such as ALU, LINE1, and LINE2 repeat sequences. When compared with the reference genome and excluding centromere sequences, the data showed that more than 98% of chromosome 1 was covered, showing that average read lengths of 250 bp could be utilized for whole human genome resequencing.
Fig. 5.4. Average human genome sequencing redundancy for parallel pyrosequencing. For all chromosomes, average redundancy was 7.4-fold, and for the X chromosome, it was 3.7-fold. (From Wheeler et al. (11); reprinted by permission from Macmillan Publishers Ltd).
7. Sensitivity and Specificity of SNP Discovery
The sensitivity and specificity of SNP discovery, particularly compared with those of current accepted technologies, ultimately reflect the performance of each of the next-generation sequencing methods. These are the ultimate metrics since they are the product of each method’s inherent error and coverage. A direct comparison with existing SNP analysis technologies allows next-generation methods to be thoroughly evaluated. Existing SNP analysis technologies include dideoxy sequencing, Affymetrix genotyping arrays, and Illumina genotyping bead arrays. Of the existing next-generation sequencing methods, parallel pyrosequencing has made the greatest strides in being able to accurately document its performance in human SNP discovery
.
106
Chan
through sequencing of an individual genome (11). In comparing the parallel pyrosequencing method with validated Affymetrix genotyping arrays, the approach showed 99.4% specificity in homozygous reference calls, 95.1% with homozygous variants, and 75.8% with heterozygous calls. In their analysis, the authors concluded that heterozygotes require at least a 13-fold redundancy to accurately call 99% of heterozygous SNPs (Fig. 5.5). On the basis of the method’s Poisson distribution, close to 20fold mean redundancy would need to be attained to ensure that most of the human genome has at least 13-fold redundancy, more than doubling the cost of the $1 million genome. The specificity was 93.3% for homozygotes and 97.8% for heterozygotes. Utilizing this approach, parallel pyrosequencing was able to identify 3.32 million SNPs, with 606,797 of those as novel SNPs. This compared well with the shotgun-sequenced Venter genome that had 3.47 million SNPs, with 647,767 of those being novel (11).
Fig. 5.5. Percentage of heterozygotes called corrected versus redundancy in parallel pyrosequencing. At least 13-fold redundancy is required to correctly call 99% of heterozygotes. (From Wheeler et al. (11); reprinted by permission from Macmillan Publishers Ltd).
The short read methods, having arrived at the marketplace later than parallel pyrosequencing, are showing that they can be effective for SNP discovery. The SOLiD method has been utilized to sequence a Yoruba germline genome and the data are being analyzed to demonstrate its performance in SNP discovery (22). The Illumina method has been utilized for a smaller diploid C. elegans genome for the N2 Bristol and CB4858 strains (2). For the published C. elegans study, a detailed analysis was performed to assess the capabilities of the platform for SNP discovery. Putative SNPs identified with the approach were verified by a PCR/dideoxy sequencing approach. A SNP validation rate of 93.8% was attained. The conversion rate, or the number of confirmed variants divided by the number of submitted variants, was lower at 87.7%,
.
Next-Generation Sequencing Methods
107
presumably owing to difficulties with the PCR/dideoxy validation step. With use of this approach, a total of 45,539 C. elegans SNPs were identified across nonrepetitive regions of the genome. Additional studies with these approaches will solidify next-generation sequencing’s performance in SNP discovery.
8. Complete Polymorphism Discovery
9. Outlook for NextGeneration Sequencing in SNP Discovery
Next-generation methods have a potential for broad utility in understanding human variation through SNPs and other polymorphisms, such as copy number variations and indels. Insight into the capabilities of these methods with respect to these polymorphisms are beginning to emerge. To assess copy number variations properly, the sequencing method needs to be able to have unbiased coverage of sequences in the genome. Parallel pyrosequencing has relatively unbiased coverage (except for the X chromosome) and was shown to match comparative genome hybridization studies in 18 of 23 regions in the human genome (11). Deletions were also identified using this method, including sequencing of breakpoint regions. The Illumina method has been shown to have coverage biases based on GC content (2, 28), which would undoubtedly impact copy number variation measurements. Despite this GC bias, Hillier et al. (2) were able to find that there was correlation between the number of ribosomal DNA sequences and the number of 32-mer reads. As for indels, the underlying homopolymer error rate for each method dictates its performance in being able to identify indels, an important emerging class of polymorphisms (29). Parallel pyrosequencing identified over 12.5 million one-base indels in the human genome sequencing data set. A total of 10.4 million of these were associated with homopolymeric runs two to 20 bases in length, leading to considerable ambiguity about the accuracy of these events (11). The Helicos method showed between 92 and 100% success in calling mock indels in the smaller 7.2 M13 genome. From these studies, it is clear that the potential is there for next-generation sequencing methods to play a role in understanding any human variation at its most fundamental level.
SNP discovery will undoubtedly benefit from the enormous data sets generated by next-generation sequencing technologies. However, there are stringent accuracy requirements that need to be met
.
108
Chan
for these methods to be effective. Errors with respect to raw data, homopolymer stretches, indels, redundancy, and coverage ultimately dictate the utility of each technology for SNP discovery. Since no technology is error-free, it is important to understand the limitations of each so that effective SNP discovery can be performed. Next-generation sequencing methods can attain adequate consensus accuracies through adequate depth of coverage for SNP discovery. An average of approximately 20-fold redundancy is required for the 454 and Illumina methods to accurate call SNPs in a diploid genome, which includes calling homozygous and heterozygous SNPs. Throughput and cost ultimately factor into attaining these levels of redundancy. For the 454 method, it would cost more than $2 million and close to 6 months to attain a 20-fold redundancy. For the Illumina method, sequencing 60 Gb of DNA would cost approximately $160,000 in reagents at $4000 per run. This estimate is simplistic because its reads are biased towards GC-rich sequences (28), meaning that a much higher average redundancy would need to be attained to have AT-rich sequences represented at least 20-fold. Furthermore, the method has short read lengths, which predisposes it to limitations when sequencing complex genomes. The predominate error in short read length methods, defined as those with 20–30-bp reads, is genome coverage. Even for a moderately complex genome such as that of C. elegans, about 25% of the genome is excluded from analysis owing to repeats (2). Given that the human genome is about 50% repeats, short read methods would not be able to discover SNPs for a significant portion of the genome. Improvements in raw base and homopolymer accuracy would only allow those regions accessible to short reads to be analyzed. The only way to improve upon genome coverage is to attain longer reads. Doing so would be challenging given that the limiting factor for these methods is their innate chemistry. Sample preparation contributes a small amount of error to the overall workflow. The major difference in sample preparation techniques is the use of PCR. In amplification with PCR, errors are propagated, especially when there is more than one PCR step involved. In other words, clonal amplification of a PCR-amplified sample leads to errors. With a high-fidelity polymerase, sample preparation leads to an error rate of 105 for clonally amplified molecules. Sample preparation with one PCR step, such as for some Sanger sequencing reactions, would have no errors since the predominant molecules in the reaction are the correct sequence. Single molecule approaches, such as the Helicos method, utilize direct reads off single molecules to bypass sample preparation errors. Homopolymers and indel errors present challenges for SNP discovery, particularly for parallel pyrosequencing and single molecule sequencing. Even with oversampling, the inherent errors in
.
Next-Generation Sequencing Methods
109
these methods limit their utility in these genomic motifs. These methods measure homopolymer length via measured intensity, which is prone to errors due to photon shot noise, fluorescence quenching, and asynchronous molecules. Indel errors also arise because of missed calls regarding base incorporations. Deletions are prominent for the Helicos method since single molecules are particularly prone to detection errors. In contrast, the Illumina and SOLiD technologies interrogate each base pair in a positional manner, with the base pair positions always recorded. It is clear from this accuracy analysis that SNP discovery is the most stringent application for next-generation sequencing methods. The 454 and Illumina technologies have been first in studies demonstrating their use in SNP discovery (2, 11). As for Helicos and SOLiD methods, their use in SNP discovery will be born out with time since they were only recently introduced into the marketplace. An abundance of applications exist for next-generation sequencing methods, most of which require less accuracy. Among these are transcriptome sequencing, ChIP-seq, and microRNA studies. While these are the applications that will enjoy rapid adoption, there is significant interest to fully utilize next-generation sequencing technologies for SNP discovery. One such effort is the 1,000 Genomes Project, which is the most significant ongoing project to characterize human genomic variation. Next-generation sequencers will play a key role in data generation (30). The project’s goal is to create a new map of the human genome that will catalog human variation more comprehensively than existing maps. In addition to an international consortium of scientists, Illumina, 454, and Applied Biosystems will participate by contributing their technologies to the effort. Each company will sequence at least 75 Gb, with Applied Biosystems contributing a total of 275 Gb. The vast amounts of data that will arise from this project will likely lead to the discovery of many SNPs and other important human variations. It will rigorously test the capability of each of these approaches for large-scale SNP discovery. More importantly, the completed map will be a composite of the various approaches, potentially allowing the diversity of errors specific to each approach to be averaged out. Exciting developments in the field of next-generation sequencing has led to a deeper understanding of variation across individual genomes. These approaches leverage off the successful completion of the Human Genome Project to resequence entire genomes, enabling large-scale SNP discovery on an unprecedented scale. Data from individual genomes have already shown that nextgeneration sequencers are up to the demands of SNP discovery. As progress is made towards routine human genome resequencing, further improvements in next-generation sequencing raw accuracy, homopolymer reads, and coverage are inevitable and will undoubtedly change how we understand SNPs in the years to come.
.
110
Chan
10. Notes
1. The SNP database (dbSNP) is a public-domain archive for a collection of SNPs provided by the National Center for Biotechnology Information (NCBI). For more information on various SNP databases, see Chapter 3. 2. The private X-Prize Foundation is offering US $10 million to the first team of researchers to sequence 100 human genomes in 10 days for less than $10,000 a genome (17). References 1. Margulies, M., Egholm, M., Altman, W. E., Attiya, S., Bader, J. S. et al. (2005) Genome sequencing in microfabricated high-density picolitre reactors. Nature 437, 376–380. 2. Hillier, L. W., Marth, G. T., Quinlan, A. R., Dooling, D., Fewell, G. et al. (2008) Wholegenome sequencing and variant discovery in C. elegans. Nat. Methods 5, 183–188. 3. Bennett, S. T., Barnes, C., Cox, A., Davies, L. and Brown, C. (2005) Toward the 1,000 dollars human genome. Pharmacogenomics 6, 373–382. 4. Bennett, S. (2004) Solexa Ltd. Pharmacogenomics 5, 433–438. 5. Shendure, J., Porreca, G. J., Reppas, N. B., Lin, X., McCutcheon, J. P. et al. (2005) Accurate multiplex polony sequencing of an evolved bacterial genome. Science 309, 1728–1732. 6. Harris, T. D., Buzby, P. R., Babcock, H., Beer, E., Bowers, J. et al. (2008) Singlemolecule DNA sequencing of a viral genome. Science 320, 106–109. 7. Mardis, E. R. (2007) ChIP-seq: welcome to the new frontier. Nat. Methods 4, 613–614. 8. Marioni, J. C., Mason, C. E., Mane, S. M., Stephens, M. and Gilad, Y. (2008) RNAseq: An assessment of technical reproducibility and comparison with gene expression arrays. Genome Res. 18, 1509–1517. 9. Chen, X., Ba, Y., Ma, L., Cai, X., Yin, Y. et al. (2008) Characterization of microRNAs in serum: a novel class of biomarkers for diagnosis of cancer and other diseases. Cell Res. 10, 997–1006. 10. Glazov, E. A., Cottee, P. A., Barris, W. C., Moore, R. J., Dalrymple, B. P. et al. (2008) A microRNA catalog of the developing
11.
12.
13.
14.
15.
16. 17. 18.
19.
20.
chicken embryo identified by a deep sequencing approach. Genome Res. 18, 957–964. Wheeler, D. A., Srinivasan, M., Egholm, M., Shen, Y., Chen, L. et al. (2008) The complete genome of an individual by massively parallel DNA sequencing. Nature 452, 872–876. Ng, P. C., Levy, S., Huang, J., Stockwell, T. B., Walenz, B. P. et al. (2008) Genetic variation in an individual human exome. PLoS Genet. 4, e1000160. Lander, E. S., Linton, L. M., Birren, B., Nusbaum, C., Zody, M. C. et al. (2001) Initial sequencing and analysis of the human genome. Nature 409, 860–921. Venter, J. C., Adams, M. D., Myers, E. W., Li, P. W., Mural, R. J. et al. (2001) The sequence of the human genome. Science 291, 1304–1351. McCarroll, S. A. and Altshuler, D. M. (2007) Copy-number variation and association studies of human disease. Nat. Genet. 39, S37–42. http://www.ncbi.nlm.nih.gov/projects/ SNP/ http://genomics.xprize.org/ Ewing, B., Hillier, L., Wendl, M. C. and Green, P. (1998) Base-calling of automated sequencer traces using phred. I. Accuracy assessment. Genome Res. 8, 175–185. Ewing, B. and Green, P. (1998) Base-calling of automated sequencer traces using phred. II. Error probabilities. Genome Res. 8, 186–194. Paez, J. G., Lin, M., Beroukhim, R., Lee, J. C., Zhao, X. et al. (2004) Genome coverage and sequence fidelity of phi29 polymerase-based multiple strand displacement whole genome amplification. Nucleic Acids Res. 32, e71.
.
Next-Generation Sequencing Methods 21. Huang, H. and Keohavong, P. (1996) Fidelity and predominant mutations produced by deep vent wild-type and exonucleasedeficient DNA polymerases during in vitro DNA amplification. DNA Cell Biol. 15, 589–594. 22. http://www.appliedbiosystems.com 23. Heiner, C. R., Hunkapiller, K. L., Chen, S. M., Glass, J. I. and Chen, E. Y. (1998) Sequencing multimegabase-template DNA with BigDye terminator chemistry. Genome Res. 8, 557–561. 24. Keohavong, P. and Thilly, W. G. (1989) Fidelity of DNA polymerases in DNA amplification. Proc. Natl. Acad. Sci. U. S. A. 86, 9253–9257. 25. Brockman, W., Alvarez, P., Young, S., Garber, M., Giannoukos, G. et al. (2008) Quality scores and SNP detection in sequencing-bysynthesis systems. Genome Res. 18, 763–770. 26. Huse, S. M., Huber, J. A., Morrison, H. G., Sogin, M. L. and Welch, D. M. (2007) Accuracy and quality of massively parallel DNA pyrosequencing. Genome Biol. 8, R143.
111
27. Wicker, T., Schlagenhauf, E., Graner, A., Close, T. J., Keller, B. et al. (2006) 454 sequencing put to the test using the complex genome of barley. BMC Genomics 7, 275. 28. Dohm, J. C., Lottaz, C., Borodina, T. and Himmelbauer, H. (2008) Substantial biases in ultra-short read data sets from highthroughput DNA sequencing. Nucleic Acids Res. 36, e105. 29. Bhangale, T. R., Stephens, M. and Nickerson, D. A. (2006) Automating resequencing-based detection of insertion-deletion polymorphisms Nat. Genet. 38, 1457–1462. 30. http://www.1000genomes.org 31. Chen, F., Alessi, J., Kirton, E., Singan, V. and Richardson, P. (2006) Comparison of 454 sequencing platform with traditional Sanger sequencing: a case study with de novo sequencing of Prochlorococcus marinus NATL2A genome. Poster LBNL 59003. Plant & Animal Genome XIV Conference, January 14–18, 2006 (San Diego, CA).
Chapter 6 Scanning Probe and Nanopore DNA Sequencing: Core Techniques and Possibilities John Lund and Babak A. Parviz Abstract We provide an overview of the current state of research towards DNA sequencing using nanopore and scanning probe techniques. Additionally, we provide methods for the creation of two key experimental platforms for studies relating to nanopore and scanning probe DNA studies: a synthetic nanopore apparatus and an atomically flat conductive substrate with stretched DNA molecules. Key words: DNA, nanopore, sequencing, scanning probe, electronic.
1. Introduction While next-generation sequencing approaches have shown great promise (see Chapter 5 in this volume), and will most likely lead to substantial reductions in both the time and the cost of sequencing genomes, every next-generation technique involves numerous rinsing and solute addition steps (1–3). These unavoidable material costs and time-consuming steps suggest existing next-generation techniques may face difficulties in sequencing entire genomes inexpensively without the development of radically new ways to conserve reagents. A handful of future-generation approaches to sequencing DNA are currently being developed by researchers around the globe. These techniques do away with the need for PCR to amplify DNA by performing sequencing on a single DNA molecule. These techniques also generally use electronic technologies to identify the individual bases along the DNA molecule. Two of the most prominent future-generation approaches are nanopore A.A. Komar (ed.), Single Nucleotide Polymorphisms, Methods in Molecular Biology 578, DOI 10.1007/978-1-60327-411-1_6, ª Humana Press, a part of Springer Science+Business Media, LLC 2003, 2009
113
114
Lund and Parviz
sequencing and scanning probe sequencing. An appealing feature of these methods is their ability to leverage the sizable infrastructure already in place in the electronics industry and to take advantage of highly sophisticated technologies that have already been developed in this area for other applications. 1.1. Nanopore Sequencing
The basic idea behind nanopore sequencing is to pass a DNA molecule through a small hole, and identify each base as the molecule passes through the pore. This establishes two unique problems that need to be solved to make nanopore sequencing a reality: building a small, well-defined pore, and developing a method to identify nucleotides on a DNA molecule that is confined within a pore. One simple approach researchers have used to build nanopores adopts a technique used in nature to produce nanoscale pores in cell walls. A very small hole between two reservoirs is covered by a lipid bilayer. A small concentration of the protein a-hemolysin is then added to one of the reservoirs. The a-hemolysin protein selfassembles into a pore structure in a lipid bilayer (4). The pore has an inner diameter of approximately 1.5 nm and a length of approximately 10 nm (5). Once a pore forms in the lipid bilayer, the two reservoirs connected by the pore are filled with an ionic solution containing DNA molecules. When a small bias (about 100 mV) is applied between the two reservoirs, the negatively charged DNA molecules begin to migrate through the pore. As a molecule passes through the pore, it blocks the normal flow of ions, leading to a decrease in measured ionic current, referred to as ‘‘ionic blockade.’’ This measurement technique has proven useful for identifying many features of DNA molecules. Long strands of purine and pyrimidine bases exhibit slightly different ionic currents (6) and the length of the DNA molecule can be determined (roughly) on the basis of the time needed for translocation through the pore (7). However, despite these successes, a-hemolysin nanopores have not been successfully used to identify individual DNA bases. Because of the length of the pore, many bases of the DNA molecule occupy the pore simultaneously, making identification of any individual base unlikely. Protein nanopores are also very sensitive to pH and electric fields, which limits the types of measurements and experimental conditions that may be used with a-hemolysin pores. To solve some of the problems associated with protein nanopores, many researchers have turned to synthetic nanopores to provide a better platform for identifying individual bases. Synthetic nanopores can be engineered to almost any size and are far more stable under high electric fields than protein-based pores (8). One of the most common methods for fabricating synthetic nanopores involves boring a hole through a thin silicon oxide or silicon nitride membrane using a focused beam of
Scanning Probe and Nanopore DNA Sequencing
115
high-energy ions. This technique uses commercially available focused ion beam (FIB) systems. Once the beam of ions breaks through the membrane, the beam is turned off, leaving a hole of several nanometers in diameter (9). For a pore with more precisely controlled dimensions, the FIB-bored pore can be precisely closed using a transmission electron microscope (TEM), which images the pore with a beam of high-energy electrons. The bombardment of the pore with these electrons will cause the pore to close slowly, and as soon as the pore has closed to the appropriate dimension, the beam can be turned off, stopping the pore-closing process (10). Synthetic nanopores have been used to detect several DNA features, including hairpins in the DNA molecule (11) and molecule length (12). However, no synthetic nanopore experiment has been able to detect individual DNA bases. One technique used by many researchers to improve the sensitivity of synthetic nanopores is transverse electrodes. Transverse electrodes measure electron tunneling and ionic currents across the pore, rather than through the pore as is the case in the ionic blockade measurements. Simulations of transverse electrode synthetic nanopores indicate they may be able to differentiate between individual DNA bases, but as of the date of this publication, no fabricated transverse electrode nanopore has achieved individual base detection (13, 14). It is expected that thermal noise will play an important role in these experiments. Thermal fluctuations introduce uncertainty in the location of the molecule in the pore and with respect to electrodes. We also note that most sensitive electronic measurements are performed under cryogenic conditions to reduce noise; however, such an approach is not applicable to nanopore experiments as a solution of mobile DNA molecules must be maintained to ensure the passage of the molecules through the pore. 1.2. Scanning Probe Sequencing
Unlike nanopore sequencing, in which electrodes are fixed and a molecule is induced to move with respect to their location, scanning probe sequencing attempts to use a moving probe to sequence a DNA molecule fixed on surface. This eliminates any noise due to movement of the DNA molecule, and allows DNA analysis outside a fluid environment, which permits ultra-highvacuum and cryogenic temperature conditions. One of the most common scanning probe instruments used for DNA sequencing experiments is the scanning tunneling microscope (STM). An STM images a surface by maintaining a constant tunneling current between an atomically sharp tip and a conductive substrate while scanning the tip across the surface. The benefit of this tool is it images using electron tunneling, and this electron tunneling may be useful for identifying the individual bases of the DNA molecule. The idea behind STM-based DNA sequencing is
116
Lund and Parviz
to first stretch a DNA molecule on a conductive substrate, then image the DNA molecule with an STM. Once the molecule has been located, the STM tip is positioned on top of the molecule, and traverses the molecule while identifying the DNA’s constituent bases using electron spectroscopy. Stretching a DNA molecule on atomically flat conductive substrates necessary to conduct these experiment is a nontrivial task. One approach is to modify the conductive surface with a selfassembled monolayer (SAM) of molecules that attract the negatively charged DNA. When a droplet containing DNA is dragged across the surface, the DNA molecules are attracted to the surface but are pulled straight by the receding meniscus of the droplet. This leaves aligned and well-stretched molecules on the conductive surface (15). Imaging DNA with an STM is also a difficult task, but has been achieved by many research groups (16, 17). Unfortunately, imaging DNA with an STM appears to reveal insufficient detail to identify individual DNA bases, and it is likely that more complicated electron tunneling techniques will be necessary to achieve reliable differentiation. Some researchers have produced excellent preliminary results identifying individual bases by performing a voltage sweep while keeping the STM tip over a single base (18). Despite these promising results, as of the date of this publication, no published research has demonstrated the ability to identify individual DNA bases (A, G, C, T) in a stretched molecule reliably using scanning probe techniques. A variety of next-generation DNA sequencing techniques aim to drive the time and cost to sequence genomes to extreme lows. While none of these techniques have proven successful for sequencing applications just yet, they are constantly improving and are actively being pursued by numerous research groups. To perform experiments using next-generation techniques, a researcher must first be able to create fundamental experimental platforms. The following experimental procedures detail techniques to develop two fundamental next-generation sequencing platforms: a synthetic nanopore and a stretched DNA molecule on an atomically flat conductive substrate.
2. Materials 2.1. Synthetic Nanopore Fabrication
1. Silicon nitride membrane window grid with a 200-mm-thick silicon frame and a 0.5-mm-wide low-stress silicon nitride window with 100-nm thickness (Structured Probe, part number 4122SN-BA).
Scanning Probe and Nanopore DNA Sequencing
117
2. Self-closing tweezers 3. Teflon polytetrafluoroethylene (PTFE) block, virgin grade (Professional Plastics) 4. High-vacuum grease (Dow Corning) 5. 1 M nitric acid (Fluka). 2.2. DNA Stretching on a Conductive Substrate
1. Hydrogen-flame-annealed atomically flat gold (111) on mica (Agilent). 2. Nonmethylated -phage DNA (New England Biolabs). 3. Mercaptoundecylamine (Sigma-Aldrich). 4. Dodecanethiol (Sigma-Aldrich). 5. Ethanol (Sigma-Aldrich). 6. TE buffer: 10 mM tris(hydroxymethyl)aminomethane–HCl, pH 7.4, 1 mM EDTA (Fluka). 7. Glass slide cover slip.
2.3. Equipment
1. FIB microscope (FEI Quanta 3D 200i, open to academic and industrial users through the NNIN). 2. TEM (Hitachi H-9500) 3. STM (Agilent model 5500 with atomic-range STM scanner).
3. Methods 3.1. Synthetic Nanopore Fabrication
To perform electrical experiments with DNA molecules passing through synthetic nanopores, it is important to first develop a reliable and repeatable platform for synthetic nanopore formation and incorporation into an apparatus that contains two fluidic wells separated by the synthetic pore. The following procedure outlines a method for first producing a thin membrane with a nanoscale pore, and then incorporating this membrane structure between two fluid wells micromachined into a Teflon block: 1. The nitride membrane is placed (nitride side up) in the chamber of a sputter coater and is coated with a suitable thin metal film (Au, Pt, or Au–Pt). The purpose of this thin metallic layer is to provide a conductive layer on the membrane to avoid electron-beam and ion-beam distortion due to charging (see Notes 1 and 2).
118
Lund and Parviz
2. The membrane is mounted (nitride side up) in a FIB microscope. A well-focused beam of Ga+ ions accelerated at 30 keV with a beam current of approximately 25 pA is directed at the nitride membrane for approximately 200 ms to drill an approximately 100 nm wide hole through the nitride film (see Note 3). 3. The membrane is placed in a TEM, preferably with an accelerating voltage of at least 200 keV. After the pore on the membrane has been located, the microscope accelerating voltage is increased to the highest available setting. The imaging window is set to a width and height of approximately 500 nm. The pore should remain centered in the microscope’s field of view. The beam current is increased incrementally until a noticeable shrinking of the pore occurs. The pore closing commences very slowly (about 1 nm/min) as the atoms move in the membrane structure. Once the pore has reached the desired size (about 5 nm), the beam is blanked and the sample is removed (see Notes 4 and 5). 4. The Teflon PTFE block is micromachined with two 200-mL wells separated by a 2-mm barrier. A 1-mm-wide hole is drilled through the separating barrier to connect the two reservoirs (see Note 6). 5. The Teflon block is thoroughly washed with soap and water following micromachining. The cleaned Teflon is then cleaned further by filling the reservoirs with 1 M nitric acid, followed by a rinse with deionized water (see Note 7). 6. A sterile cotton swab of high-vacuum grease is rubbed around the 1-mm hole separating the two reservoirs, with one making sure not to plug the hole. The nitride membrane is then carefully placed over the hole (nitride side out). The highvacuum grease forms a tight seal around the membrane and affixes it to the wall between the chambers. 7. The apparatus is now complete. The reservoirs can now be filled with electrolytic solution and DNA molecules so one can perform nanopore experiments (Fig. 6.1). 3.2. DNA Stretching on a Conductive Substrate
Before a DNA molecule is examined using an STM, the DNA must be stretched on a conductive substrate with atomically flat terraces. The following procedure details a method for covering an atomically flat gold surface with well-stretched -DNA molecules: 1. A mixed solution for making a SAM is prepared by pipetting 5 mL of mercaptoundecylamine and 100 mL of dodecanethiol into a beaker containing 100 mL of ethanol (see Note 8).
Scanning Probe and Nanopore DNA Sequencing
119
Fig. 6.1. The completed synthetic nanopore structure (not drawn to scale). Note there is a round hole connecting the two wells, which is covered completely by the nitride membrane grid.
2. The atomically flat gold is placed (gold side up) in the beaker and the beaker is covered with a looking glass. The gold substrate is allowed to incubate for 3 h to produce a highly ordered mixed SAM (see Note 9). 3. The gold substrate is removed from the solution and vigorously rinsed five times with ethanol, and then dried under a stream of nitrogen. 4. -phage DNA is diluted to 2.5 mg/mL with TE buffer (see Note 10). 5. A 5-mL droplet of 2.5 mg/mL DNA solution is placed at a corner of the gold surface, and is slowly dragged across the gold using a thin glass cover slip. It should take at least 20 s to move the droplet from one side of the gold to the other. 6. The remaining droplet should be wicked off the gold surface using the edge of a piece of filter paper. 7. The gold surface is now covered with well-stretched DNA molecules. The density and orientation of the molecules are easily confirmed using a tapping-mode atomic force microscope if desired (Fig. 6.2).
120
Lund and Parviz
Fig. 6.2. Atomic force microscope image of double-stranded DNA molecules stretched on an atomically flat gold surface modified with a mixed self-assembled monolayer of mercaptoundecylamine and dodecanethiol. The dark arrows point towards the molecules.
4. Notes
1. Nitride membranes are extremely fragile and should be handled with care. Self-closing tweezers are recommended for safely manipulating the membranes, as excessive pressure will cause the membranes to break and insufficient pressure will likely lead to the membrane being dropped. 2. Sputter coaters are ubiquitous in electron microscopy sample preparation, and the appropriate procedures for the coater of choice should be followed. If available, an inert gas should be used for sputtering instead of air. 3. FIB systems vary greatly. It is unlikely one system will be able to match precisely the performance of another. For this reason, a dose test may be necessary to establish a precise exposure time to produce an appropriately sized nanopore. A dose test involves using a range of exposure times at different
Scanning Probe and Nanopore DNA Sequencing
121
locations on the membrane (spaced at least 10 mm apart) to identify the precise exposure time for producing an optimal pore. 4. TEMs pose a variety of safety issues, including X-ray radiation, high-voltage power, and cryogenic liquids. These hazards are generally well controlled in modern TEM systems, but all system operation and safety procedures should be followed meticulously. 5. Sample loading and handling is typically a challenging aspect of TEM operation. The nitride membranes are sized to be compatible with most TEMs, but it may be necessary to mount them on a separate conductive sample holder specific to the TEM in use. In this case, the nitride membrane should remain faceup at all times and should be handled carefully with self-closing tweezers. 6. The recommended dimensions for the micromachined wells are 4 mm wide by 10 mm long by 5 mm deep. The 4-mm width is intended to ensure the nitride membrane has little room to move side to side, so it may be aligned more easily when it is affixed to the separating barrier between the two reservoirs. The hole between the two wells may need to be drilled at an angle as a drill bit has little room to work within the small wells. The well size is kept small because larger volumes of liquid will lead to an increase in ionic current noise. 7. Nitric acid is a hazardous chemical. Proper laboratory attire should be used, including goggles and a laboratory coat. Butyl gloves should be worn when working with nitric acid, and the acid and Teflon block should be kept under a fume hood at all times during the cleaning process. Nitric acid should never be rinsed down the drain, but should be disposed of in a manner compliant with local regulations. 8. All solutions should be prepared inside a well-ventilated hood using proper laboratory attire, including goggles and nitrile gloves. 9. The atomically flat gold samples are highly prone to contamination. Even in a sealed vial containing an inert gas, the gold fouls after about 1 month, and is nearly impossible to clean with renewed flame annealing. Always use fresh gold samples. It can be difficult to identify the gold and mica sides of the substrate. Agilent typically scratches a date on the mica side of its samples, which can be helpful in identifying both the age of the sample as well as the orientation. 10. DNA molecules may break down at room temperature. All DNA solutions should be stored below 0C (–20C preferred). An effort should be made to keep DNA solutions
122
Lund and Parviz
free from contamination. If a DNA solution has undergone more than eight freezing and thawing cycles, it should be replaced with fresh DNA. References 1. Mashayekhi, F. and Ronaghi, M. (2007) Analysis of read length limiting factors in Pyrosequencing chemistry. Anal. Biochem. 363, 275–287. 2. Shendure, J., Porreca, G. J., Reppas, N. B., Lin, X., McCutcheon, J. P., Rosenbaum, A. M., Wang, M. D., Zhang, K., Mitra, R. D. and Church, G. M. (2005) Accurate multiplex polony sequencing of an evolved bacterial genome. Science 309, 1728–1732. 3. Mikkelsen, T. S., Ku, M., Jaffe, D. B., Issac, B., Lieberman, E., Giannoukos, G., Alvarez, P., Brockman, W., Kim, T. -K., Koche, R. P., Lee, W., Mendenhall, E., O’Donovan, A., Presser, A., Russ, C., Xie, X., Meissner, A., Wernig, M., Jaenisch, R., Nusbaum, C., Lander, E. S. and Bernstein, B. E. (2007) Genome-wide maps of chromatin state in pluripotent and lineage-committed cells. Nature 448, 553–560. 4. Kasianowicz, J. J., Brandin, E., Branton, D. and Deamer, D. W. (1996) Characterization of individual polynucleotide molecules using a membrane channel. Proc. Natl. Acad. Sci. U.S.A. 93, 13770–13773. 5. Song, L., Hobaugh, M. R., Shustak, C., Cheley, S., Bayley, H. and Gouaux, J. E. (1996) Structure of Staphylococcal alphahemolysin, a heptameric transmembrane pore. Science 274, 1859–1865. 6. Deamer, D. W. and Branton, D. (2002) Characterization of nucleic acids by nanopore analysis. Acc. Chem. Res. 35, 817–825. 7. Meller, A., Nivon, L., Brandin, E., Golovchenko, J. and Branton, D. (2000) Rapid nanopore discrimination between single polynucleotide molecules. Proc. Natl. Acad. Sci. U.S.A. 97, 1079–1084. 8. Fologea, D., Gershow, M., Ledden, B., McNabb, D. S., Golovchenko, J. A. and Li, J. (2005) Detecting single stranded DNA with a solid state nanopore. Nano Lett. 5, 1905–1909. 9. Li, J., Stein, D., McMullan, C., Branton, D., Aziz, M. J. and Golovchenko, J. A. (2001)
10.
11.
12.
13.
14.
15.
16.
17.
18.
Ion-beam sculpting at nanometre length scales. Nature 412, 166–169. Storm, A. J., Chen, J. H., Ling, X. S., Zandbergen, H. W. and Dekker, C. (2003) Fabrication of solid-state nanopores with single-nanometre precision. Nat. Mater. 2, 537–540. Li, J., Gershow, M., Stein, D., Brandin, E. and Golovchenko, J. A. (2003) DNA molecules and configurations in a solid-state nanopore microscope. Nat. Mater. 2, 611–615. Storm, A. J., Storm, C., Chen, J., Zandbergen, H., Joanny, J. F. and Dekker, C. (2005) Fast DNA translocation through a solid-state nanopore. Nano Lett. 5, 1193–1197. Lagerqvist, J., Zwolak, M. and Di Ventra, M. (2007) Comment on ‘‘Characterization of the tunneling conductance across DNA bases’’. Phys. Rev. E Stat. Nonlinear Soft Matter Phys.) 76, 013901–013903. Chen, Y. C., Zwolak, M. and DiVentra, M. (2004) Inelastic current-voltage characteristics of atomic and molecular junctions. Nano Lett. 4, 1709–1712. Mehta, R., Rahimi, M., Lund, J. A. and Parviz, B. A. (2007) Rapid extension of single and double stranded DNA on atomically flat conductive surfaces. IEEE Trans. Nanotechnol. 6, 734–736. Shapir, E., Sagiv, L., Borovok, N., Molotski, T., Kotlyar, A. B. and Porath, D. (2008) High-resolution STM imaging of novel single G4-DNA molecules. J. Phys. Chem. B 112, 9267–9269. Tanaka, H. and Kawai, T. (2003) Visualization of detailed structures within DNA. Surf. Sci. 539, L531–L536. Yoshida, Y., Nojima, Y., Tanaka, H. and Kawai, T. (2007) Scanning tunneling spectroscopy of single-strand deoxyribonucleic acid for sequencing. J. Vac. Sci. Technol. B Microelectron. Nanometer Struct. 25, 242–246.
Chapter 7 Pyrosequencing for SNP Genotyping Jose Luis Royo and Jose Jorge Gala´n Abstract Pyrosequencing is a real-time DNA sequencing method. It is based on the transformation of pyrophosphates, released during DNA elongation by DNA polymerase, into measurable light. During DNA elongation, a single pyrophosphate molecule is released following incorporation of a single nucleotide. In the pyrosequencing reaction, released pyrophosphates are then rapidly converted by sulfurylase to adenosine triphosphate, which in turn is utilized by luciferase to produce light. Within standardized conditions, this reaction is accomplished in a few milliseconds and the light produced can be registered with a CCD camera. Therefore, it becomes possible to quantitatively measure the nucleotides incorporated. This approach has been automated in different platforms and can be used for a wide variety of applications, such as single-nucleotide polymorphism (SNP) genotyping, DNA sequencing, loss of heterozygosity analysis, and CpG methylation studies. Here we describe the entire process, focusing our attention on SNP genotyping, and giving examples of some other applications. Key words: Pyrosequencing, genotyping, sequencing, real time.
1. Introduction Pyrosequencing is a real-time sequencing method based on the conversion of pyrophosphate groups, released during DNA elongation, into measurable light. This process relies on the fact that the generated light is directly proportional to and reflects the nucleotides incorporated into DNA by DNA polymerase at each given moment in time. Thus, researchers can quantitatively read from one up to 100 nucleotides of each DNA product. The procedure can be split into three main steps. First, the genomic region of interest must be amplified using the polymerase chain reaction (PCR). This is a critical step for pyrosequencing technology since the reliability and reproducibility of this A.A. Komar (ed.), Single Nucleotide Polymorphisms, Methods in Molecular Biology 578, DOI 10.1007/978-1-60327-411-1_7, ª Humana Press, a part of Springer Science+Business Media, LLC 2003, 2009
123
124
Royo and Gala´n
technique largely depends on the quality and quantity of the PCR products. Second, the PCR products are converted into singlestranded DNA (ssDNA) fragments to which a sequencing primer is annealed. To this end, amplicons must be generated with a biotinylated primer and immobilized, while being incubated under denaturing conditions, which would keep the DNA in a single-stranded form. Third, specific nucleotides are added to the sequencing-primer-bound ssDNA in a user-defined order. When a user-supplied nucleotide is incorporated into the sequencing primer 30 -end, a pyrophosphate group is released, and sulfurylase then converts it into ATP, which is used by luciferase to generate light, subsequently recorded by a CCD camera (1, 2). Meanwhile, an apyrase eliminates the excess of nucleotides, leaving the reaction volume prepared for the injection of a second/next nucleotide. The entire process is illustrated in Fig. 7.1. 1
CCD Record
GTP
Light
GTP
2 G C TA G C
G AC G C T GC
5
Polymerase 3A 3B
Pyrophosphate ATP
GTP
Luciferase
4
Apyrase
Sulfurolyase
Fig. 7.1. Pyrosequencing reaction scheme. During pyrosequencing, the cartridge delivers nucleotides according to the dispensation order specified by the researcher (1). When DNA polymerase elongates (2), incorporating the nucleotide, the pyrophosphate molecule is released (3A). The excess nucleotides are degraded by apyrase (3B). Sulfurolyase converts pyrophosfate into ATP (4), which is then utilized by luciferase to produce light (5).
Pyrosequencing is especially versatile for single-nucleotide polymorphism (SNP) genotyping. Since the raw data are simply real-time sequencing data of the region of interest, pyrosequencing can be used to genotype not only diallelic polymorphisms, but also triallelic variants and small insertions/deletions (indels) (3). In addition, and given the accuracy of pyrosequencing technology (see also Chapter 5 in this volume), it is also possible to quantitatively determine the amount of each allele in the DNA sample, and this can be used to estimate the allelic frequency of a SNP in a population, if one pools all DNAs equimolarly (4). Pyrosequencing can be also applied to detect loss of heterozygosity (LOH) regions, allelic variation in gene expression analyses, as it can be also used to define the extent of DNA methylation within a CpG island (5–7).
Pyrosequencing for SNP Genotyping
125
In this chapter we will describe the methodological procedure of pyrosequencing technology focused on SNP genotyping, and we will also refer to some other useful applications.
2. Materials 2.1. Polymerase Chain Reaction
1. 96-well PCR thermocycler.
2.1.1. Instruments
1. Biotinylated sense amplification primer at 10 mM. 2. Antisense amplification primer at 10 mM.
2.1.2. Reagents
3. PCR buffer with MgCl2 available from many commercial sources. 4. Deoxynucleotide triphosphates (dNTPs) at 1.25 mM. 5. Taq DNA polymerase.
2.2. ssDNA Generation and Sequencing Primer Annealing 2.2.1. Instruments
1. Vacuum Prep Tool (Biotage, Uppsala, Sweden). 2. Vacuum source (water jet or vacuum pump). 3. Orbital shaker for microtiter plates. 4. PSQTM96 plate (Biotage). 5. Heating block (set up at 80C).
2.2.2. Reagents
All buffers and solutions should be prepared with high-purity water (Milli-Q): 1. Streptavidin–Sepharose HP (GE Healthcare/Amersham Biosciences). Storage conditions as recommended by the manufacturer. 2. Binding buffer: 10 mM tris(hydroxymethyl)aminomethane (Tris)–HCl, 2 M NaCl, 1 mM EDTA, 0.1% Tween 20. Adjust the pH to 7.6 with 1 M HCl at 22 – 1C. Store at 4C (for a maximum of 1 month). 3. Denaturation solution: 0.2 M NaOH. Store at 4C (for a maximum of 1 month). 4. Washing buffer: 10 mM Tris–acetate. Adjust the pH to 7.6 with 4 M acetic acid at 22 – 1C. Store at 4 C (for a maximum of 1 month). 5. Ethanol 70%. This solution may be stored at room temperature. 6. Annealing buffer: 20 mM Tris–acetate, 2 mM magnesium acetate. Adjust the pH to 7.6 with 4 M acetic acid at 22 – 1C. Store at 4C (for a maximum of 1 month). 7. Antisense sequencing primer at 10 mM.
126
Royo and Gala´n
2.3. Sequencing Reaction
1. PSQTM96 instrument (Biotage, Uppsala, Sweden). 2. PSQTM96 reagent cartridge (Biotage, Uppsala, Sweden). 3. Pyro Gold reagents (Biotage, Uppsala, Sweden).
3. Methods 3.1. PCR Amplification
1. Design a PCR protocol according to the technology requirements (see Note 1). In the case of analyzing a large number of SNPs in a few DNA samples, it is highly advisable to employ indirect labeling of the amplification primers (see Note 2). 2. Prepare a 96-well plate with 10–20 ng of DNA per sample in each well (see Note 3). 3. Add the PCR cocktail to each well. The total volume for each well should not be greater than 40 mL. 4. Seal the plate and introduce it into the thermocycler. 5. Run the PCR program. 6. Once the PCR has finished, store the PCR plate at 4C. If you are not going to analyze the PCR products, in less than 24 h, store them at –20C.
3.2. ssDNA Preparation and Sequencing Primer Annealing
1. Let the binding, annealing, denaturing, and washing buffers reach room temperature. 2. Prepare a PSQTM96 plate by filling the corresponding wells with 0.4 mM sequencing primer in 40 mL annealing buffer (see Note 4). 3. Add high-purity water to the PCR products to reach a final volume of 40 mL (per sample). 4. Vortex vigorously the stock of streptavidin–Sepharose beads and make a mixture with 4 mL of the beads plus 36 mL of the binding buffer (per sample). 5. Add 40 mL of the above mixture to each well with PCR products and seal the plate (see Note 5). 6. Place the plate in an orbital shaker for microtiter plates and shake it constantly (1,000–1,200 rpm) for 10 min at room temperature. Caution: Weak mixing may lead to bead sedimentation. 7. Prepare four troughs with 150 mL of high-purity water (Milli-Q or similar), 150 mL of 70% ethanol, 100 mL of denaturation solution, and 150 mL of washing solution, respectively.
Pyrosequencing for SNP Genotyping
127
8. Place the Vacuum Prep Tool in the water trough. Connect the Vacuum pump and let the water flushes through the filters for 20 s. 9. Capture the beads containing immobilized PCR products on the filter probes by lowering the Vacuum Prep Tool into the 96 well PCR plate (see Note 6). 10. Move the Vacuum Prep Tool to the trough containing 70% ethanol and let the solution flush through the filters for 5 s. 11. Move the Vacuum Prep Tool to the trough containing denaturation solution and let the solution flush through the filters for 5 s (see Note 7). 12. Move the Vacuum Prep Tool to the trough with washing buffer and let the solution flush through the filters for 5 s. 13. Pick up the Vacuum Prep Tool vertically and let it dry for a few seconds. Return it to the horizontal position (see Note 8). 14. Switch off the vacuum pump and place the Vacuum Prep Tool in the PSQTM96 plate (see Note 9). 15. Manually shake the Vacuum Prep Tool and let the filters rest for a few seconds on the bottom of the wells while the beads are released (see Note 10). 16. Once the ssDNA has been released into the PSQTM96 plate, introduce it into a heating block at 80C for 2 min. 17. Let the PSQTM96 plate reach room temperature and place it in the PSQTM96 instrument. Samples are now ready for the pyrosequencing reaction. 18. Move the Vacuum Prep Tool to the water trough to wash the filters. It is advisable to remove possible bead excess from the filters by introducing the filter probes into a sonication bath. 3.3. Sequencing Reaction
The sequencing reaction basically relies on the sequential addition of dNTPs to the sequencing-primer-bound ssDNA following a user-specified order (dispensation order). As dNTPs are being incorporated into the sequencing product, light is produced in each well, registered, and reflected by the system in real time as light peaks. 1. Design the dispensation order (see Note 11). For an example of a dispensation order, see Note 12. 2. Select the instrument parameters and edit the names for each well. 3. Place the PSQTM96 plate in the PSQTM96 instrument. 4. Load the PSQTM96 reagent cartridge with a suitable amount of Pyro Gold reagents as recommended by the manufacturer.
128
Royo and Gala´n
5. Introduce the PSQTM96 reagent cartridge into the PSQTM96 instrument. 6. Start the run. 3.4. Quality Control and Data Analysis
1. Available software usually allows automatic genotyping with a color-based legend to show those reliable or unreliable samples. However, it is always recommended to go back to the raw data to check peak intensities (signal greater than 20) and shape (better sharp than wide) before exporting the results to an analysis database (see Note 13). 2. Samples can also be analyzed using allelic quantification, this is especially useful for applications such as methylation studies (see Note 14), LOH (see Note 15), and allele frequency determination in DNA pools (see Note 16).
4. Notes
1. The PCR amplification of the DNA segment of interest is a critical step for pyrosequencing technology. Low-quality PCR products negatively affect the reliability and reproducibility of the technique. To potentially avoid such problems, PCR products should ideally fulfill the following requirements: – – – –
Length between 80 and 200 bp. G+C content below 60%. Free from unspecific products and/or primer–dimer. Free from self-annealing problems.
One of the two amplification primers (sense primer) must be labeled at its 50 -end with biotin. This molecule will allow the isolation of ssDNA. The manufacturer offers a valuable application to help in oligo designing. However, there are some Web-based tools such as Primer3 (http://fokker.wi. mit.edu/primer3/input.htm) and SOP3 (8) that can be very helpful in PCR primer design. In addition, it is also advisable to carry out quality controls of PCR products before proceeding with the technique. For instance, you can introduce in the PCR plate some wells with the PCR cocktail but free from DNA (amplification controls). Once the PCR has finished, randomly select 5% of the PCR products and load 5 mL of each one and 5 mL of the
Pyrosequencing for SNP Genotyping
129
contamination controls on a 2% agarose gel. After electrophoresis and gel staining, a unique and clear band of the expected size without smear and/or any primer–dimer band(s) should be observed in the case of all selected PCR products. Contamination controls must appear without any visible band. If this is not the case, the reaction must be repeated under different conditions, since contamination clearly exists and thus can affect the results of the experiment. 2. The main disadvantage of pyrosequencing is the cost attributable to biotinylated primers, especially for projects involving too many SNPs and too little DNA samples. For these projects, the use of indirect biotin labeling is recommended. One of the amplification primers in the latter case may be designed with a 50 -DNA tail (such as M13 or the equivalent). When PCRs are set up, a biotinylated M13 primer is incorporated together with the forward and reverse primers. Thus, during amplification, the biotinylated primer will be incorporated in the amplicons, labeling them in an indirect fashion (9, 10). 3. The 96-well PCR plate should have a shape that will allow the filter probes of the Vacuum Prep Tool to reach the bottom of the wells and aspirate all the volume of the PCR and the binding mixture. 4. The sequencing primer must be complementary to and the reverse of the biotinylated ssDNA. This primer should be about 20–25-nt long and should also be free from potential self-annealing and unspecific amplicon annealing problems. The software mentioned in Note 1 may be very helpful for the proper primer design. 5. The mixture provides streptavidin–Sepharose beads with the ideal conditions to bind biotinylated PCR products. 6. This step must be carried out quickly since streptavidin– Sepharose beads sediment within a few seconds. 7. In this step the PCR strands without biotin are separated from the complementary immobilized strands. Therefore, ssDNAs are generated and kept bound to filter probes, while nonbiotinylated strands are swept out to the waste bottle by the flush. 8. This step is aimed at removing the excess of liquid from filter probes and the Vacuum Prep Tool. 9. Caution: If the pump is still on, it will aspirate the annealing buffer and the sequencing primer. 10. To check whether the beads have been correctly released, place the annealing plate on a dark surface. A white shadow at the bottom of each well should be observed.
130
Royo and Gala´n
11. The dispensation order must coincide with the complementary and reverse DNA sequence of the immobilized ssDNA. In the case of SNPs, the polymorphic locus is interrogated by consecutively adding the nucleotides complementary to each allele. The dispensation order should also include control nucleotides to enable testing of the specificity of the sequencing product generated. These control nucleotides must be different from those that would be bound to the sequencing product. Therefore, no light peaks are expected for those control nucleotides. If light peaks are detected, it would mean that these control nucleotides have been incorporated into the sequencing product. This may be due to the presence of unspecific PCR products, amplicon self-annealing, and/or unspecific sequencing primer annealing to the PCR product. In the latter case it is advisable to change some parts of the protocol; for instance, new amplification and/or sequencing primers can be designed, biotin labeling can be applied to the antisense amplification primer (in this case the sequencing primer and the dispensation order must also change), and more astringent PCR conditions can be applied. 12. Here we present an example of a dispensation order for the genotyping of the rs8179176 polymorphism located at intron 1 of the human ESR1 gene (Fig. 7.2a). In this case the dispensation order was designed as GCTAGTATG. Note that the first and the fourth nucleotides are the control nucleotides. The second and the third nucleotides interrogate the corresponding alleles for this polymorphism. Finally, the last five nucleotides are aimed at testing the specificity of the sequencing product. Figure 7.2b shows the pyrograms resulting from the genotyping of the rs8179176 polymorphism in the wild-type, heterozygous, and homozygous DNA samples. 13. Since pyrosequencing is a real-time sequencing method, one can design a dispensation order according to the wild-type sequence and check putative discordances if the template does not appear to possess the correct sequence (9). This can be used for already-characterized deletions, but needs to be designed case by case as mentioned in Note 12. 14. In mammals, methylation occurs almost exclusively at the CpG dinucleotide. Treatment with bisulfite converts cytosine to uracil unless the cytosine has a methyl group. Thus, treated DNA can be analyzed using PCR and pyrosequencing, since methylated cytosines will be observed as C/A alleles (7, 11). 15. As mentioned in this chapter, there is a direct correspondence between a single nucleotide incorporation and the light produced/recorded in the course of the reaction.
Pyrosequencing for SNP Genotyping
131
A)
B)
Fig. 7.2. Dispensation order for the genotyping of rs8179176 polymorphism. (a) DNA sequence. Forward and reverse (biotinylated) amplification primers are shown singleunderlined and double-underlined, respectively. The sequencing primer is highlighted with white letters in black shadow. The polymorphic locus is represented with the letter Y (change C to T). (b) Pyrograms derived from the genotyping of rs8179176 polymorphism in homozygous (CC), heterozygous (TC), and wild-type (TT) DNA samples. The dispensation order is indicated at the bottom of each panel. E and S stand for enzyme and substrate reagents, respectively. Notice that all light peaks have the same height (corresponding to only one nucleotide incorporated), with the exception of the second thymine (corresponding to three nucleotides incorporated), see the sequence in a. In the case of the TC sample, the nucleotides interrogating both alleles of the rs8179176 polymorphism present half of the height than in CC and TT samples, since the amount of both alleles are identical in this heterozygote DNA sample.
This can be easily verified/checked by every researcher simply by mixing (at different ratios) of already-genotyped samples and using the allelic quantification analysis option to check the linearity between the real genotype and the allelic quantification obtained (Fig. 7.3). Comparing the pattern of the series of heterozygous SNPs from germline DNA and tumor DNA, one can detect LOH regions (Fig. 7.4) (12). 16. Pyrosequencing has been used to estimate allelic frequencies in a population simply by pooling the DNAs equimolarly and performing allelic quantification of each SNP under study (4). This approach takes advantage of the strong linearity between the relative ratios of each allele present in the DNA template and allele quantification obtained using pyrosequencing.
132
Royo and Gala´n
Allele 1 quantification (%)
100
y = –9,51x + 108 r2 = 0,99
90 80 70 60 50
100 90 80 70 60 50 DNA alele 1 concentration from pooled DNAs
Fig. 7.3. Standard quantification curve of PAK1 rs538670. Different ratios were obtained by mixing homozygote and heterozygote samples in different proportions. The samples were subjected to PCR–pyrosequencing and analyzed using allelic quantification.
C:45%
C:49%
C:24%/T:76%
C:65%/T:35%
C:72%/T:28%
C:29%/T:71%
Fig. 7.4. Loss of heterozygosity detection using pyrosequencing. The panels represent pyrograms obtained when genotyping normal heterozygous PAK1 rs2844336, (upper left) or EMSY rs4300410 (upper right). When tumor DNA samples are analyzed, it is possible to detect an allelic imbalance affecting one of the two alleles (middle panels). This can be observed in any of the two alleles (lower panels), depending on the haplotype that is being amplified or deleted. This phenomenon is what we call ‘‘loss of heterozygosity.’’
Acknowledgements The authors would like to thank Ana Salinas, Maria del Carmen Rivero, and Juan Velasco for their technical support. References 1. Ronaghi, M., Uhlen, M. and Nyren, P. (1998) A sequencing method based on realtime pyrophosphate. Science 281, 363–365.
2. Ronaghi, M. (2001) Pyrosequencing sheds light on DNA sequencing. Genome Res. 11, 3–11.
Pyrosequencing for SNP Genotyping 3. Guo, D. C., Qi, Y., He, R., Gupta, P. and Milewicz, D. M. (2003) High throughput detection of small genomic insertions or deletions by Pyrosequencing. Biotechnol. Lett. 25, 1703–1707. 4. Lavebratt, C. and Sengul, S. (2006) Single nucleotide polymorphism (SNP) allele frequency estimation in DNA pools using Pyrosequencing. Nat. Protoc. 1, 2573–2582. 5. Loeuillet, C., Weale, M., Deutsch, S., Rotger, M., Soranzo, N., Wyniger, J., Lettre, G., Dupre´, Y., Thuillard, D., Beckmann, J. S., Antonarakis, S. E., Goldstein, D. B. and Telenti, A. (2007) Promoter polymorphisms and allelic imbalance in ABCB1 expression. Pharmacogenet. Genomics 17, 951–959. 6. Lee, E. S., Issa, J. P., Roberts, D. B., Williams, M. D., Weber, R. S., Kies, M. S. and El-Naggar, A. K. (2008) Quantitative promoter hypermethylation analysis of cancer-related genes in salivary gland carcinomas: comparison with methylation-specific PCR technique and clinical significance. Clin. Cancer Res. 14, 2664–2672. 7. Marsh, S. (2007) Pyrosequencing applications. Methods Mol. Biol. 373, 15–24. 8. Alexander, A. M., Pecoraro, C., Styche, A., Rudert, W. A., Benos, P. V., Ringquist, S. and Trucco, M. (2005) SOP3: a web-based
9.
10.
11.
12.
133
tool for selection of oligonucleotide primers for single nucleotide polymorphism analysis by Pyrosequencing. Biotechniques 38, 87–94. Royo, J. L., Pascual, M. H., Salinas, A., Tello, F. J., Rivero, M. del C., Herrero, E. F., Real, L. M. and Ruiz, A. (2006) Pyrosequencing protocol requiring a unique biotinylated primer. Clin. Chem. Lab. Med. 44, 435–341. Royo, J. L., Hidalgo, M. and Ruiz, A. (2007) Pyrosequencing protocol using a universal biotinylated primer for mutation detection and SNP genotyping Nat. Protoc. 2, 1734–1739. Korshunova, Y., Maloney, R. K., Lakey, N., Citek, R. W., Bacher, B., Budiman, A., Ordway, J. M., McCombie, W. R., Leon, J., Jeddeloh, J. A. and McPherson, J. D. (2008) Massively parallel bisulphite pyrosequencing reveals the molecular complexity of breast cancer-associated cytosinemethylation patterns obtained from tissue and serum DNA. Genome Res. 18, 19–29. Pascual, M. H., Royo, J. L., Martı´nez-Tello, F. J., Crespo, C., Salinas, A., Herrero, E. F., Lopez-Garcı´a, M., Real, L.M., Ruiz, A. and Ramirez-Lorca, R. (2006) Exploring allelic imbalance within paraffin-embedded tumor biopsies using pyrosequencing technology. Clin. Chem. Lab. Med. 44, 1076–1081.
Chapter 8 Single Nucleotide Polymorphism Screening with Denaturing Gradient Gel Electrophoresis Leslie A. Knapp Abstract Denaturing gradient gel electrophoresis (DGGE) is a powerful technique for identifying DNA sequencebased differences. The method relies on the fact that double-stranded DNA molecules have unique denaturation rates that are based upon the specific nucleotide composition of the DNA sequence(s). While DGGE is typically used to screen for polymorphisms that vary by multiple nucleotides, it is equally useful for screening single nucleotide polymorphisms (SNPs). For most applications, it is possible to use computer software in advance to determine if SNPs can be differentiated using DGGE. The software can also model the effect of attaching a GC-rich clamp to the PCR primer to improve detection of SNPs. Once feasibility has been confirmed, a perpendicular DGGE can be used to identify the optimal denaturing gradient for the sequences of interest. Parallel gels can then be used to screen large numbers of samples at one time, eliminating the need for cloning and sequencing or direct sequencing of PCR products. This chapter provides step-by-step instructions on the use of DGGE and illustrates its application for detection of SNPs, as well as multiple nucleotide polymorphisms, in the major histocompatibility complex. Key words: Denaturing gradient gel electrophoresis, WinMelt, major histocompatibility complex.
1. Introduction Denaturing gradient gel electrophoresis (DGGE) is based on the principle that double-stranded DNA molecules have unique denaturation rates that are based upon the specific nucleotide composition of the DNA sequence (1). When subjected to a denaturing environment, DNA strands will separate in discrete regions that are called ‘‘melting domains.’’ The temperature at which the melting domain denatures is called the ‘‘melting temperature.’’ Differences in the extent of denaturation within the melting domain can be detected by subjecting DNA molecules to acrylamide gel electrophoresis (2). Urea and A.A. Komar (ed.), Single Nucleotide Polymorphisms, Methods in Molecular Biology 578, DOI 10.1007/978-1-60327-411-1_8, ª Humana Press, a part of Springer Science+Business Media, LLC 2003, 2009
137
138
Knapp
formamide are added to the acrylamide gel to create a denaturing environment and when their concentrations are varied, a wide range of denaturing gradients can be produced. Electrophoresis at a constant temperature of 60C also assists with denaturation of the DNA molecules. When DNA molecules begin to denature, their mobility in the acrylamide gel will be altered. Molecules with the greatest extent of denaturation will migrate more slowly than molecules that are less denatured. Differences in the mobility will result in the DNA molecules appearing at different positions on a gradient gel and they can be visualized as separate bands (Fig. 8.1). A
0%
80% denaturant
Each line corresponds to a different SNP
B 40%
denaturant
60%
Each line, and lane, corresponds to different SNPs
Fig. 8.1. Denaturing gradient gel electrophoresis (DGGE) can be used to screen single nucleotide polymorphisms (SNPs) in PCR products that are the same size, but possess different sequence(s). (A) The optimal denaturing gradient for screening is determined using a perpendicular (i.e., horizontal) DGGE. (B) Multiple SNPs, from different individuals, are separated on the appropriate parallel denaturing gradient. DNA sequences that denature rapidly will move through the gel at a slower rate than sequences that denature more slowly.
Single Nucleotide Polymorphism Screening with DGGE
139
DGGE can be used to analyze DNA molecules of lengths ranging from 100 base pairs (bp) to 1,000 bp. While larger fragments can usually be separated using 6–8% acrylamide gels, smaller fragments often require a higher concentration of acrylamide to slow down DNA mobility and maximize denaturing conditions. Therefore, shorter PCR products (e.g., 300–400 bp) are better separated using acrylamide concentrations as high as 12% (3). In practice, DGGE has been, for example, used for screening CC chemokine receptor 2 gene single nucleotide polymorphisms (SNPs) (CCR2 V64I) (4) and for scanning polymorphism in exon 8 of the putative tumor suppressor gene fragile histidine triad (FHIT) gene (5). We frequently employ DGGE for high-resolution major histocompatibility complex genotyping (3, 6, 7), which typically involves resolution of greater nucleotide polymorphism than SNPs. Thus, DGGE is equally suitable for screening polymorphism confined to a single location in a DNA sequence (i.e., SNPs) and for identifying polymorphism distributed across one or more exons. An additional reason for the widespread popularity of DGGE is that it can be combined with DNA sequencing to determine the precise nucleotide sequences (see Note 1).
2. Materials 2.1. Optional Testing of the Feasibility of DGGE for Known DNA Sequences and Determining the Placement of GC Clamps Using Computer Software
1. WinMeltTM or MacMeltTM software tools (Bio-Rad).
2.2. PCR Prior to DGGE
1. Template DNA. This needs to be previously extracted, highquality DNA that is known to amplify under the experimental conditions. 2. Typical reagents for PCR, including 10X PCR buffer without magnesium chloride (MgCl2), 10 mM deoxyribonucleotide triphosphates (dNTPs) mix, Taq polymerase, and MgCl2 solution (all obtained from Invitrogen, Paisley, UK). All reagents should be stored at –20C. Ultrapure (DNAse- and RNAse-free) PCR water is also required.
140
Knapp
3. PCR primers. Typically, one primer will have a 30–40-bp GC clamp (see Note 2). Primers are diluted in PCR-grade water and stored in 500-mL aliquots at –20C (see Note 3). 2.3. Denaturing Gradient Gel Electrophoresis
1. 50X tris(hydroxymethyl)aminomethane–acetate–EDTA (TAE) pH 8.0 buffer (store at room temperature). 2. 40% acrylamide (37.5:1 acrylamide/bisacrylamide, 2.6% C; Bio-Rad). Store at 4C. Caution must to be exercised, because unpolymerized acrylamide is a neurotoxin and can be easily absorbed through skin. 3. Molecular biology tested formamide (Sigma). Use cautiously, as formamide is a teratogen. Exposure should be limited. 4. Molecular biology grade urea. 5. Cellulose nitrate vacuum filter (0.22-mm pore size) with receiving bottle (e.g., Corning vacuum filter/bottle storage system). 6. Dye solution for gradient visualization (Bio-Rad) (see Note 4). 7. Polymerization agents: N,N,N0 ,N0 -tetramethylethylenediamine (TEMED; Bio-Rad) and 10% ammonium persulfate (APS). Prepare 10% APS in sterile water (see Note 5), divide into 500-mL aliquots for single use, and store at –20C (see Note 6). 8. Sigmacote (Sigma), stored at 4C. Follow the manufacturer’s instructions and use in a chemical hood to avoid inhalation of fumes. 9. Gradient delivery system (Bio-Rad). 10. 2X loading dye: 2 mM EDTA pH 8.0, 70% glycerol, 0.05% xylene cyanol, and 0.05% bromophenol.
2.4. Visualization, and Optional Excision, of DNA Molecules from DGGE Gels
1. Staining solution: SYBR Gold (Invitrogen) and 50X TAE. SYBR Gold is placed as single-use aliquots. Tubes have to be wrapped in foil (stain is light-sensitive) and stored at –20C (see Note 7).
2.5. Optional PCR Amplification for Direct Sequencing of SNPs Identified Using DGGE
1. Template DNA is derived from excised bands on DGGE. DNA is eluted by incubating excised bands in 50 mL double-distilled water (ddH2O) at room temperature overnight. Eluates can be stored at room temperature for 1 week or frozen at –20C for longer storage periods.
2. Wide-bore pipette tips previously autoclaved and stored in sterile environment until use.
2. Typical reagents for PCR, including 10X PCR buffer without MgCl2, dNTPs mix, Taq polymerase, and MgCl2 solution (Invitrogen). All reagents are stored and handled as described in Section 2.2.2.
Single Nucleotide Polymorphism Screening with DGGE
141
3. PCR primers. Primer sequences can be the same as those used prior to DGGE or, more optimally, primers can be heminested within the region previously amplified by PCR (see Note 8). Primers are stored and handled as described in Section 2.2.3.
3. Methods DGGE can be performed in a perpendicular or parallel direction (Fig. 8.1). The denaturing gradient runs horizontally across the gel when a perpendicular gel is poured and the gradient runs vertically from top to bottom in parallel denaturing gels. As a rule, perpendicular gels, with a gradient range of 0–80%, are used to determine the optimal denaturing gradient for the particular size and type of DNA molecule of interest. A sample is electrophoresed through the perpendicular gel and the resulting melting pattern of the molecule(s) determines what gradient will reveal the greatest difference in mobility between molecules. This optimal gradient can then be employed for subsequent parallel DGGEs. Parallel gels have the advantage of a narrower gradient, which increases mobility differences between DNA molecules, and of the possibility of electrophoresing many samples at the same time. With parallel gels, it is beneficial to conduct a time study to experimentally determine the optimal running time of samples passing through the denaturing gradient that will maximize separation of different sequences (see Note 9). Under ideal conditions, samples should only be partially denatured to maximize differences in mobility. Complete denaturation usually results in poor resolution since fully denatured samples have equivalent mobilities. To ensure that DNA molecules do not fully denature under DGGE conditions, a long string of Gs and Cs, called a ‘‘GC clamp’’ (8) may be incorporated. Since Gs and Cs have a higher melting temperature than DNA sequences with a more equivalent distribution of nucleotides, the DNA fragment of interest will denature, while the GC clamp will remain annealed. Since most DNA molecules used in DGGE are generated using PCR, the addition of a GC clamp is made possible by including the GC-rich string at the 30 -end of one of the PCR primers. The optimal placement of the GC clamp (i.e., if it is added to the forward or reverse primer sequence) can be determined using computer software that models the melting patterns of DNA sequences. Placement of a GC clamp can also be tested empirically, but this approach is not as cost-effective as modeling with different lengths and sequences of GC clamps using WinMeltTM.
142
Knapp
3.1. Testing the Feasibility of DGGE for Known DNA Sequences and Determining the Placement of GC Clamps Using Computer Software
1. WinMeltTM can be used to model the effect of attaching a GC clamp to the forward or reverse PCR primer, to identify the melting domains of different DNA sequences, and, most importantly, to determine the feasibility of using DGGE for SNP screening (Fig. 8.2). This step requires knowledge of the DNA sequences of interest, so that any, or all, SNPs are incorporated in the analysis.
Fig. 8.2. WinMeltTM, or MacMeltTM for Macintosh users, can be used to determine if two different DNA sequences can be differentiated using DGGE and to model the effect of attaching a GC clamp to the PCR primer. In this example, two HLA-DRB1 sequences that differ by a single nucleotide (position 41), indicated by black lines and gray lines, are compared with WinMeltTM to determine if the sequence differences will yield unique denaturation profiles and to evaluate the effect of a 40-bp GC clamp on the 30 -end of the reverse primer (299–346 bp). The sequences have different melting properties within the first melting domain (1–70 bp, see inset), so it should be possible to visualize differences using DGGE. Note, however, that once the samples are heated above 75C, both sequences have the same melting profile and any differences in mobility will be lost. In this case, a perpendicular gel was required to determine the optimal denaturing gradient to reduce the chance of overdenaturing the sequences, thereby failing to detect the differences in the first melting domain.
2. Start the WinMeltTM program by opening a new project from the File option. Using the Sequence option, paste or import each DNA sequence of interest, including both forward and reverse primer sequences. For each sequence, choose a 50% melting probability and choose to view it on a graph. 3. Once two or more sequences have been added to a project, the user will be able to view the temperature ranges at which 50% of the sequence’s double helix would split apart and thus move through the DGGE gel. Each graph shows base pairs against
Single Nucleotide Polymorphism Screening with DGGE
143
melting temperature. If two sequences have different mobilities, this will be apparent in the Graph View generated from the sequences. It is also possible to estimate the value and optimal placement of a GC clamp by adding the GC-rich sequence (see Note 2) to either side of the DNA and primer sequence. 3.2. PCR Prior to DGGE
1. Amplification of template DNA by PCR can be carried out according to user-defined conditions (see Note 10).
3.3. Denaturing Gradient Gel Electrophoresis
1. These instructions assume that the Bio-Rad DCode system and Bio-Rad gradient delivery system are used for DGGE. Alternative DGGE equipment from Hoefer or INGENY and gradient makers from CBS Scientific or Teledyne Isco can be used with little alteration of materials and methods (see Note 11). 2. Prepare 100 mL each of 0 and 80% denaturing acrylamide solutions (see Note 12), which will be used as stock reagents for either perpendicular or parallel denaturing gradient gels. A 12% acrylamide, 0% denaturing solution is made by mixing 32.4 g acrylamide solution with 1.8 g 50X TAE and approximately 66.24 g ddH2O, to bring the final volume to 100 mL. A 12% acrylamide, 80% denaturing solution is made by first mixing 30.5 g urea with 33.12 g ddH2O in a glass bottle/beaker and gently heating the solution to aid in dissolving the urea (see Note 13), which should take approximately 10 min. The use of a magnetic stir plate and heater, plus a magnetic flea for stirring, can assist in more even heating of the solution. The magnetic flea should be removed using a magnetic wand, 32.4 g acrylamide solution and 1.8 g 50X TAE should be added, and the volume should be brought to 100 mL by adding ddH2O. It is recommended that both solutions are filtered and degassed for 15 min using a vacuum filter system. Once they have cooled, the denaturants can be wrapped in aluminum foil and stored at 4C for up to 4 weeks (see Note 14). 3. It is recommended that before the gels are poured the electrophoresis apparatus is set up and the buffer is warmed. To heat the buffer, add 140 mL 50X TAE to the buffer chamber and fill the chamber with approximately 7 L deionized water to a level half way between the ‘‘fill’’ and ‘‘run’’ markers on the outside of the apparatus. Place a cover on the buffer chamber, turn on the heater and the pump, and set the heat at 64C (see Note 15). Gel heating will take approximately 1 h. 4. Typically, a 0.75- or 1-mm gel is optimal for detection of SNPs using DGGE (see Note 16). To make a gel of either thickness, select a pair of glass plates (see Note 17) and wash each side with detergent. Rinse the plates with sterile water and treat the inside of each plate with silicon solution following the manufacturer’s instructions. Rinse both plates with sterile water and then with 10% ethanol and allow the plates to air-dry before proceeding.
144
Knapp
5. These instructions relate to the preparation of a parallel DGGE. To prepare a perpendicular DGGE gel, the user must change the gel pouring orientation and the denaturants are delivered through tubing attached to the bottom of the gel ‘‘sandwich’’ (see Note 18). Select a set of spacers and place these on the larger plate in the proper orientation. Position the smaller glass plate directly on top, put a clamp on each side of this gel plate ‘‘sandwich,’’ and fasten loosely. 6. Place the plates in the gel pouring platform, insert the alignment card, and align the spacers as necessary. Ensure the spacers are flush with the edge of the glass otherwise spillage will occur during casting. Once one is satisfied with the alignment, remove the insert and take the plates out of the platform to tighten the clamps. 7. Place a dry foam gasket on the base of the gel pouring platform and position the gel plates on top. Insert and rotate the cams to secure the plates. Place a wide-bore needle tip, fitted with a large Luer lock adaptor, into the space at the upperright corner of the plates. Push a clean well-forming comb into the space between the two glass plates. 8. Before preparing the desired denaturing solutions, ensure that the gradient delivery system is set up, that two clean syringes are ready to collect the denaturants, and that all parts of delivery apparatus are unblocked and ready to use. 9. To pour a denaturing gradient gel, remove stock denaturants from the refrigerator and 10% APS from the freezer. A typical 8 cm 10 cm, 0.75-mm thick, gel requires 28 mL of acrylamide, made using 14 mL of each denaturing solution. If a perpendicular gel is to be poured, then 14 mL of 0% and 14 mL of 80% denaturing solutions will be used without dilution (see Note 19). If a parallel gel is to be poured, the optimal gradient can be determined empirically by first using a perpendicular gradient gel. The 0 and 80% denaturants are used to make any gradient desired. For example, to prepare 14 mL of 40% denaturing solution, 7 mL of the 80% stock solution is mixed with 7 mL of the 0% stock solution. A 60% solution requires 10.5 mL of 80% stock solution and 3.5 mL of 0% stock solution for a final volume of 14 mL. 10. Place two 25-mL polypropylene screw-top tubes in a rack. Label them with the desired denaturant concentrations and pipette appropriate volumes of 0% and 80% denaturants into each tube. Add 250 mL dye solution to the heaviest denaturant. If 14 mL of each denaturant is used (sufficient to pour a 0.75-mm gel), add 140 mL thawed 10% APS and 14 mL TEMED to each tube and gently invert the tube. Immediately transfer each denaturant to a separate syringe by drawing the solution through Tygon tubing attached to the syringe by a large Luer lock.
Single Nucleotide Polymorphism Screening with DGGE
145
11. Gently expel air and prime the tubing with acrylamide. Once the tubing is full and without air bubbles, quickly attach each syringe to the appropriate side of the delivery system and gently introduce the acrylamide solutions into the space between the two gel plates. The delivery should be rapid enough to finish pouring the gel before polymerization, but slow enough to prevent bubbles forming (see Note 20). 12. Completely fill the space between the two plates. Note that the dye will be darkest at the bottom of the gel, where most of the heavier denaturant is found. Remove the needle, tube, and syringe (see Note 21). 13. Allow the gel to polymerize for 30–45 min. Monitor the polymerization of the gel using the remaining acrylamide in the syringe. We suggest that the researcher uses this time to prepare the samples for DGGE by mixing them with an equal volume of 2X loading buffer. A maximum volume of 45–50 mL can be loaded per well; however, it is often easier to load volumes under 40 mL. 14. When the gel is set, remove the plates and gel from the gel pouring platform and gently remove the comb (see Note 22). Fit plates into the buffer chamber (see Note 23). 15. Turn off the heater and pump and remove approximately 500 mL of warm buffer before submerging the gel in the buffer tank. When doing this, partially dip the gel into the buffer several times before it is placed in the tank. This should allow the temperature of the glass to equalize with that of the buffer slowly, thereby preventing the glass breaking. 16. Rinse the wells vigorously with running buffer and load samples into the wells using a capillary pipette tip. 17. Reattach the top and turn on the heater. If the temperature has dropped below 60C, allow the buffer to be reheated to this temperature before beginning electrophoresis. Once the buffer has returned to 60C, electrophorese samples at 300 V for 3–5 h, depending on the optimal conditions (see Note 24). 3.4. Visualization, and Optional Excision, of DNA Molecules on DGGE Gels
1. Approximately 15 min before end of the run, prepare 150 mL SYBR Gold stain by adding 15 mL SYBR Gold and 3 mL 50X TAE to sterile water in a large, flat glass dish. Cover this entirely with foil to prevent light degradation of the stain. 2. Once the DGGE run is complete, turn off the system at the mains and remove the gel from the upper buffer chamber, ensuring that the hot buffer is drained away. 3. Place the upper buffer system on the plastic-lined paper and detach the plates containing the gel.
146
Knapp
4. With the dish containing the SYBR Gold at hand, separate the glass plates by removing the spacers and carefully easing the plates apart. Typically, the gel will remain on one plate. Hold that plate over the stain and use a spacer to detach a corner of the gel from the plate. The gel should then roll down into the stain. Replace the foil once the gel is in the staining solution. 5. Place the staining dish on a slow rocker, replace the foil cover, and stain at room temperature for 20 min. 6. Visualize over UV light using a transilluminator and record the image using a camera system (Fig. 8.3). 7. If desired, excise bands of interest with sterile wide-bore pipettes using a cookie-cutter action (see Note 25). Place these gel-containing pipettes into prelabeled sterile 1.5-mL
1
2
3
4
5
6
7
Fig. 8.3. A SYBR Gold stained 40–60% parallel DGGE gel showing the physical separation of the two HLA-DRB1 sequences analyzed in Fig. 8.2. Lanes 1, 2, 3, and 5 contain cloned DNA with each sequence alone and lane 4 contains both sequences together. Lanes 6 and 7 represent typical results when PCR samples with multiple sequences, differing by three or more nucleotides across the entire 346-bp sequence, are screened on a 40– 60% parallel DGGE gel.
Single Nucleotide Polymorphism Screening with DGGE
147
tubes and resuspend the gel plugs from the tips into 100 mL PCR water. Remove the tips and store tubes at room temperature. 3.5. Optional PCR Amplification for Direct Sequencing of SNPs Identified Using DGGE
1. If DNA sequencing of excised gel bands is to be undertaken, heat the samples at 37C for 10 min to elute the DNA from the gel. 2. PCRs to amplify specific sequences can be accomplished with a standard 25-mL PCR that includes 10 mL of gel eluate mixed with 1X PCR buffer, 3 mM MgCl2, 200 mM dNTPs, 0.1 mM forward primer, 0.1 mM reverse primer, 0.5 U Taq polymerase, and sterile water to bring the final volume to 25 mL. 3. A typical thermal cycler program for these PCRs would be one round of initial denaturation at 94C, 40 cycles of denaturation at 94C for 20 s, annealing at 60C for 30 s, and extension at 72C for 36 s, followed by a 10-min extension step at 72C. 4. The success of the reaction and the concentration of PCR products can be evaluated by electrophoresing of a 5-mL aliquot(s) on a 1% agarose gel and the remaining product can be incubated with 2 mL ExoSap-IT at 37C for 15 min, followed by 80C for 15 min if the bands appear distinct and are at least 30 ng/mL in concentration (see Note 26). The treated PCR products can then be used in whatever DNA sequencing reaction is preferred by the user.
4. Notes 1. While DGGE can be used to screen for the presence of previously identified SNPs, it has the added benefit of providing the researcher with the ability to identify new polymorphisms that were previously unidentified. 2. We typically attach the following GC-rich sequence (50 -CGCCCGCCGCGCCCCGCGCCCGGCCCGCCGCC GCCCCCGCCCGC-30 ). When primers for DGGE are being designed, GC clamps must be attached to the 50 -end of either the forward or the reverse primer. The GC clamp is only attached to one primer, so if the forward primer contains the clamp it should not be attached to the reverse primer. It is possible to determine which primer enables optimal amplification by trial and error or by estimation using either WinMeltTM or MacMeltTM. Bear in mind that modeling various lengths and sequences of GC clamps with these programs is often most cost-effective.
148
Knapp
3. It is important to dilute PCR reagents in sterile PCR water to avoid exposure to previously amplified DNA that would contaminate PCRs. 4. A blue dye solution allows the researcher to assess if a relatively even gradient from high to low denaturants has been poured. If an even gradient has been prepared, the color of the gel will be darkest at the bottom (highest concentration of the denaturant) and lightest at the top (lowest concentration of the denaturant). The dye solution can also be prepared from scratch by mixing 0.05 g bromophenol blue and 0.05 g xylene cyanol with 10 mL 1X TAE. The solution is stored at room temperature. 5. In all instances, unless stated otherwise, reagents should be prepared in water with a resistivity of 18.2 M -cm. This is referred to as ‘‘sterile water’’ in the text. 6. APS should be ‘‘fresh’’ whenever it is used to polymerize acrylamide. Rather than preparing small volumes before each DGGE experiment, we prepare larger volumes in advance and store small aliquots at –20C until needed. 7. We find that SYBR Gold is more sensitive and generally safer, for visualization of bands on DGGE gels, than ethidium bromide. It also does not inhibit successful DNA elution from acrylamide gel samples when DNA sequencing is performed following DGGE. Silver staining is even more sensitive, but we have found this type staining to be more laborintensive and less optimal if DNA sequencing is used after DGGE. 8. We typically design our PCR primers for post-DGGE amplification by attaching a universal M13Forward sequence to the 50 -end of one primer (3) and a universal M13Reverse sequence to the 50 -end of the other primer to simplify priming in our DNA sequencing reactions. We also choose to make one of the two primers 2–3 bp nested within the original DGGE primer sequence, so that we minimize amplification of primer dimers. 9. The optimal conditions for separation of different DNA sequences using DGGE are best determined by conducting a time study that involves loading samples at multiple time points. We usually find that it is best to choose three time points that vary by 30 min each (e.g., 3, 3.5, and 4 h), to determine the optimal electrophoresis time. In our time studies, we load three different samples in separate wells and electrophorese them for 30 min before loading another aliquot of the same samples in adjacent lanes and electrophoresing for another 30 min before loading a final aliquot of the same samples and electrophoresing all samples for the final 3 h.
Single Nucleotide Polymorphism Screening with DGGE
149
10. We usually electrophorese 5 mL PCR product on a 1 or 1.5% agarose gel with ethidium bromide staining to check for successful amplification and to estimate PCR yield. We use a molecular mass standard (Bio-Rad) to estimate the yield and if 50 mg/mL or more DNA is present, we use this sample for DGGE. Care must be taken when using ethidium bromide as it is a potent carcinogen; therefore, relevant safety guidelines for this product must be followed. 11. If your laboratory has multiple systems, even from the same manufacturer, we recommend using the same piece of equipment for DGGE for a particular sequence. We have found that identical DGGE systems produce different results, possibly owing to slight thermostat variations. In addition, we place a thermometer inside the tank to compare the temperature of the buffer with the reading on the thermostat. 12. Although many of the reagents used in the preparation of denaturing solutions are liquid, we find that weighing all the reagents (including solutions) allows us to produce more consistent denaturing solutions and, therefore, we have more reproducible results. 13. We recommend that urea is dissolved in TAE, before adding the formamide. 14. During storage, acrylamide and bisacrylamide are slowly converted to acrylic acid and bisacrylic acid, respectively. As this reaction is catalyzed by light, we wrap our solutions in foil to prolong the life of the solution. 15. Although DGGE is typically performed at 60C, we set the starting temperature to 64C because sample loading can be slow and the buffer cools below the desired running temperature of 60C. We also find that it helps to start heating the buffer before casting the gel. By the time the gel has polymerized, the buffer should be at the set temperature. 16. We typically use spacers that are either 0.75- or 1.0-mmthick. The Bio-Rad set includes one spacer with a short groove and one with a long groove, which only has relevance for the user when perpendicular DGGE gels are prepared. Ensure that the spacers are the same height, as spacers shrink when used repeatedly. 17. The Bio-Rad DCode system employs glass plates that measure 8 cm 10 cm. One plate is slightly shorter than the other, to allow for contact between the buffer and acrylamide gel during DGGE. The shorter plate will be placed on the inside of the unit during electrophoresis.
150
Knapp
18. Pouring perpendicular gels is significantly more complicated than pouring parallel gels, as the denaturants must be poured with the gradient running across the glass plates and all four sides of the plates must be completely secured before delivering the denaturants. The Bio-Rad system allows the researcher to secure all plate edges with clamps and the denaturants are then delivered through asmall holeinoneof the side clamps. Step-by-stepinstructions are included in the manual and users are encouraged to become familiar with the steps, using water instead of acrylamide, before pouring a perpendicular gel. If the Bio-Rad, or a similar, system is employed, users should note that the type and position of the spacers differs when pouring perpendicular gels. Also, it is recommended that any visible gaps at the edges of the glass plates are filled with 1% agarose to prevent leakage during pouring. 19. An initial perpendicular DGGE with the gradient ranging from 0 to 80% is typically used to identify the optimal denaturing gradient for the sequences of interest. 20. If any bubbles are formed in the gel during pouring, it may be possible to remove them by very gently tapping the gel plates. 21. Syringes should be replaced after they have been used for five gels, to ensure that delivery of the acrylamide is smooth during gel pouring. 22. We recommend that after the comb has been removed the wells are carefully, but vigorously, flushed with sterile water to remove any unpolymerized acrylamide. 23. If you are only running one gel, place a dummy gel into the other side to create a complete buffer chamber. 24. The optimal electrophoresis time will be determined by the length and sequence of the samples and also by the presence and position of the GC clamp. When our samples are between 300 and 400 bp, including the GC clamp, we typically electrophorese them at 300 V for approximately 4.5 h. Longer DNA fragments will require a longer electrophoresis time at 300 V. 25. Care must be taken because of the UV light on the transilluminator, so wear a laboratory coat with elasticated sleeves and tuck these into gloves. Also wear a visor, if necessary. 26. It is easy to carry out this procedure in a thermocycler with the reagents and PCR product in thin-walled PCR tubes.
Acknowledgement Thanks to J. Osborn for technical assistance.
Single Nucleotide Polymorphism Screening with DGGE
151
References 1. Fisher, S. G. and Lerman, L. S. (1983) DNA fragments differing by single base pair substitutions are separated in denaturing gradient gels: correspondence with melting theory. Proc. Natl. Acad. Sci. U.S.A. 80, 1579–1583. 2. Myers, R. M., Sheffield, V. C. and Cox, D. R. (1988) Detection of single base changes in DNA: ribonuclease cleavage and denaturing gradient gel electrophoresis. In: Genome Analysis: A practical approach. (Davies, K., ed.). Oxford: IRL Press, pp. 95–139. 3. Knapp, L. A., Cadavid, L. F., Eberle, M. E., Knechtle, S.J., Bontrop, R. E. and Watkins, D. I. (1997) Identification of new MamuDRB alleles using DGGE and direct sequencing. Immunogenetics 45, 171–179. 4. Peterse, D. C., Laten, A., Zeier, M. D., Grimwood, A., van Rensburg, E. J. and Hayes, V.M. (2002) Novel mutations and SNPs identified in CCR2 using a new comprehensive denaturing gradient gel electrophoresis assay. Hum. Mutat. 20, 253–259.
5. Suh, Y. and Vijg, J. (2005) Single Nucleotide Polymorphisms (SNPs): Detection, Interpretation, and Applications. Mutat. Res. 573, 41–53. 6. Knapp, L. A., Lehmann, E., Hennes, L., Eberle, M. E. and Watkins, D. I. (1997) High-resolution HLA-DRB typing using DGGE and direct sequencing. Tissue Antigens 50, 170–177. 7. Huchard, E., Cowlishaw, G., Raymond, M., Weill, M. and Knapp, L. A. (2006) Molecular study of Mhc-DRB in wild chacma baboons reveals high variability and evidence for trans-species inheritance. Immunogenetics 58, 805–816. 8. Sheffield, V. C., Cox, D. R., Lerman, L. S. and Meyers, R. M. (1989) Attachment of a 40-base-pair G+C rich sequence (GCclamp) to genomic DNA fragments by the polymerase chain reaction results in improved detection of single-base changes. Proc. Natl. Acad. Sci. U.S.A. 86, 232–236.
Chapter 9 Temporal Temperature Gradient Electrophoresis for Detection of Single Nucleotide Polymorphisms Bethan M. Jones and Leslie A. Knapp Abstract The presence of single nucleotide polymorphisms (SNPs) in nuclear DNA and mitochondrial DNA (mtDNA) can be detected using a range of electrophoretic techniques, of which temporal temperature gradient electrophoresis (TTGE) is often the most user-friendly and reproducible. The technique operates on the same principle as denaturing gradient gel electrophoresis, but does not require a chemical gradient in the gel. Instead, TTGE relies on a steady and gradual increase in temperature during electrophoresis to denature and separate DNA sequences that differ by as little as one base pair. TTGE can be easily accomplished using DNA of high quality and it is a rapid-throughput method for SNP screening once conditions have been optimized. Detection of SNPs is, for example, important for the diagnosis of mitochondrial disorders such as heteroplasmy, the presence of more than one type of mitochondria within a cell or tissue. Here we describe the basic steps for TTGE and illustrate its utility for the detection of heteroplasmy in mtDNA control region sequences. Key words: Temporal temperature gradient electrophoresis, mutation detection, mitochondria, D-loop, control region, HV1, heteroplasmy.
1. Introduction A variety of electrophoresis techniques are available for the detection and screening of single nucleotide polymorphisms (SNPs). Both nuclear and mitochondrial SNPs can be studied using acrylamide gel electrophoresis methods that include single-stranded conformation polymorphism (see Chapter 12), denaturing gradient gel electrophoresis (see Chapter 8), and temporal temperature gel electrophoresis (TTGE). We have tried all of these approaches for the study of polymorphisms in the mitochondrial genome and have A.A. Komar (ed.), Single Nucleotide Polymorphisms, Methods in Molecular Biology 578, DOI 10.1007/978-1-60327-411-1_9, ª Humana Press, a part of Springer Science+Business Media, LLC 2003, 2009
153
154
Jones and Knapp
found that TTGE is one of the most straightforward, informative, and reproducible methods for routinely detecting SNPs in mitochondrial DNA (mtDNA) in vertebrate species, including humans. The detection and screening of human mitochondrial SNPs is especially important in a medical context. Since the mammalian mitochondrial genome has a high copy number, numbering several thousand per cell (1), and has a high mutation rate compared with the nuclear genome (2), it is possible for mitochondria with different sequence types to coexist within a cell, tissue, or individual. This phenomenon, known as ‘‘heteroplasmy,’’ can be a major cause of disease in humans. Clinical expression of disease is often dependent on the proportion of mutant to wild-type mtDNA present in tissues. While DNA sequencing can be used to identify mutations, heteroplasmic mtDNA sequences that are present in a low proportion or that differ from wild-type mtDNA by a single base pair may be difficult to detect from background signals. Additionally, it is not always practical to routinely sequence the entire mtDNA genome for every individual suspected of an undiagnosed mitochondrial disorder. Using denaturing gradient gel electrophoresis (DGGE), Tully et al. (3) found common heteroplasmy for the hypervariable region 1 (HV1) of the D-loop or control region of human mtDNA, a noncoding region where heteroplasmy is linked to cyclic vomiting syndrome and migraines (4). However, DGGE can be technically demanding owing to the requirement for denaturing chemical gradients, making results difficult to reproduce (see Chapter 8 in this volume). To circumvent these problems, TTGE has been used to sensitively detect mitochondrial heteroplasmy in many regions, including the control region (5, 6), and has been proven as a suitable mutation detection method for the entire mtDNA genome (7). The technique operates on the same principle as DGGE; however, instead of the application of a chemical gradient for DNA denaturing, the denaturing environment in a TTGE gel is formed by a constant concentration of urea and a gradual increase of temperature during electrophoresis (8). The procedure also differs from temperature gradient gel electrophoresis, where specialized equipment is required to apply a temperature gradient from the top to the bottom of the gel in real time. Instead, during TTGE, the temperature of the gel is the same at any one location, changing through time (temporally) and uniformly through the gel to exploit the sequence-specific melting behavior of DNA sequences at different temperatures. TTGE is sensitive enough to separate and compare sequences that differ by just one nucleotide and it is applicable to both nuclear DNA and mtDNA. This chapter summarizes the methods used to screen SNPs with TTGE, using the example of control region mtDNA heteroplasmy. The steps outlined can be adapted to the study of SNPs in both mitochondrial and nuclear regions.
Temporal Temperature Gradient Electrophoresis
155
2. Materials 2.1. Estimation of Optimal Gradient for TTGE
1. WinMeltTM or MacMeltTM software tools (Bio-Rad) (see Note 1).
2.2. Polymerase Chain Reaction Prior to TTGE
1. Template DNA of high quality, known to amplify under the experimental conditions. 2. Typical reagents for polymerase chain reaction (PCR), including 10X PCR buffer without magnesium chloride (MgCl2), 10 mM deoxyribonucleotide triphosphates (dNTPs) mix, 1 mg/mL bovine serum albumin, Taq polymerase (see Note 2), and MgCl2 solution (all obtained from Invitrogen). All reagents should be stored at –20C. Ultrapure (DNAse and RNAse-free) PCR water is also required. 3. PCR primers. Typically, one primer will have a 30–40-bp GC clamp (see Note 3). Primers are diluted in PCR water and stored in 500-mL aliquots at –20C (see Note 4).
2.3. Preparation of Acrylamide Solution for TTGE
1. 40% acrylamide solution (37.5:1 acrylamide/bis acrylamide, Bio-Rad). Store at 4C. Caution must to be exercised, since unpolymerized acrylamide is a neurotoxin and can be easily absorbed through skin. 2. Molecular biology grade urea and 50X tris(hydroxymethyl) aminomethane–acetate–EDTA (TAE), stored at room temperature. 3. Cellulose nitrate vacuum filter (0.22-mm pore size) with a receiving bottle (e.g., Corning vacuum filter/bottle storage system).
2.4. Casting and Running TTGEs
1. Sigmacote (Sigma), stored at 4C. Follow the manufacturer’s instructions and use in a chemical hood to avoid inhalation of fumes. 2. Polymerization agents: N,N,N 0 ,N 0 -tetramethylethylenediamine (TEMED; Bio-Rad) and 10% ammonium persulfate (APS). Prepare 10% APS in sterile water (see Note 5), divide into 500-mL aliquots for single use and store at –20C (see Note 6). 3. Running buffer: 50X TAE. 4. 2X loading dye: 2 mM EDTA pH 8.0, 70% glycerol, 0.05% xylene cyanol, and 0.05% bromophenol.
156
Jones and Knapp
2.5. Visualization, and Optional Excision, of DNA Molecules on TTGE Gels
1. Staining solution: SYBR Gold (Invitrogen) and 50X TAE. SYBR Gold is placed in single-use aliquots that have been wrapped in foil (stain is light-sensitive) and stored at –20C. 2. Wide-bore pipette tips and 1.5-mL polypropylene tubes that have been autoclaved and stored in a sterile environment.
2.6. Optional PCR Amplification for Direct Sequencing of SNPs Identified Using TTGE
1. PCR reagents: 10X PCR buffer without MgCl2, dNTPs mix, Taq polymerase, MgCl2, (as in Section 2.1). 2. PCR primers. Primer sequences can be the same as those used prior to TTGE or, more optimally, primers can be heminested within the region previously PCR-amplified (see Note 7). 3. ExoSAP-IT (GEHealthcare) for degrading unused primers and dNTPs following PCR amplification for direct sequencing. ExoSAP-IT is stored at –20C.
3. Methods An important step in TTGE is the preparation of the DNA of interest. PCR must amplify DNA from individuals or from clones consisting of the sequence of interest. Using TTGE, one can detect DNA sequences with the same length but different nucleotide composition, differing by as little as a single nucleotide. During electrophoresis, the temperature is increased gradually and uniformly and the increased temperature causes the DNA to denature and decreases its mobility in the gel. The complete denaturation of the products can be prevented by adding a 40-bp G- and C-rich sequence (‘‘GC clamp’’) to the extremity of one PCR product (see Note 4). While TTGE does not specifically require the use of a GC clamp on one of the primer pairs, we attach a clamp on either the forward or the reverse primer when generating PCR products for TTGE. We find that this maximizes band separation on the gels when sequences differ by only a few base pairs and prevents complete denaturation of our samples. Computer simulations with WinMeltTM, or MacMeltTM, can be used to determine the placement of the optional GC clamp and to estimate the melting temperature of sequences prior to TTGE. The WinMeltTM, or MacMeltTM, program is based on the melting algorithm developed by Lerman and Silverstein (9).
Temporal Temperature Gradient Electrophoresis
3.1. Estimation of Melting Temperature Using Computer Software
157
1. WinMeltTM or MacMeltTM can be used to determine which combination of primers would produce an optimal melting domain for screening SNPs by TTGE. This step requires knowledge of the DNA sequences of interest, so that SNPs are incorporated in the WinMeltTM analysis. 2. Start the WinMeltTM program by opening a new project from the File option. Using the Sequence option, paste or import each DNA sequence of interest, including both forward and reverse primer sequences. For each sequence, choose a 50% melting probability and choose to view it on a graph. 3. Once two or more sequences have been added to a project, the user will be able to view the temperature ranges at which 50% of the sequence’s double helix would split apart. Each graph shows base pairs against melting temperature. The graph can also be used to evaluate the importance and optimal placement of a GC clamp by adding the GC-rich sequence (see Note 4) to either side of the DNA and primer sequence. 4. Optimal temperature ranges, at which the remainder of the sequences would denature, are determined by the minimum and maximum temperatures from the range of variation seen between sequences. Subtracting 14–16C from these values gives the optimal temperature range for the TTGE procedure, as 7 M urea gels are used (each mole of urea will lower the melting temperature by 2C) (see Note 8) (Fig. 9.1).
3.2. PCR Prior to TTGE
1. Amplification of template DNA by PCR can be carried out according to the user-defined conditions (see Note 9). 2. Individual thermal cycling conditions will also be determined by the user. We have found that the GC-clamped sequence is best amplified when a final 10-min extension step at 72C is added to the final program.
3.3. Preparation of Acrylamide Solution for TTGE
1. 200 mL of 6% acrylamide solution with 7 M urea can be prepared as follows. Add 84 g urea to 100–150 mL sterile water and pour the mixture into a glass receptacle of suitable size. Add a magnetic flea and place the receptacle on a magnetic stirring and heating plate for approximately 10 min or until all the urea has dissolved. Remove the magnetic flea using a magnetic wand and add 30 mL 40% acrylamide solution and 4 mL 50X TAE. Bring the volume to 200 mL with sterile water. 2. Using the vacuum filtration/bottle storage system, filter and degas the solution for 15 min. Once the solution has cooled, wrap the bottle in foil (see Note 10) and store it at 4C (for up to 1 month).
158
Jones and Knapp
A
B
Fig. 9.1. WinMeltTM, or MacMeltTM, can be used to determine the melting domains of different DNA sequences and to model the effect of attaching a GC clamp to the PCR primer. In this example, two mitochondrial control region sequences, differing by a single nucleotide (indicated by black lines and gray lines), are compared using WinMeltTM to determine differences in melting profiles (see positions 70–150 bp) and to evaluate the effect of a 40-bp GC clamp on either the 5’-end of the forward primer or the 5’-end of the reverse primer. (a) If the GC clamp is placed on the forward primer, the two sequences will denature at different rates between 64 and 67C and neither sequence will fully denature until it is heated to 95C. (b) When the GC clamp is placed on the reverse primer, there is a narrow temperature range between the point when both sequences begin to denature and the point when both sequences are fully denatured. On the basis of this WinMeltTM result, we chose to place the GC clamp on the forward primer and estimated that the optimal temperature range for screening the two sequences using temporal temperature gradient electrophoresis (TTGE) would be 48–53C (see Note 8).
Temporal Temperature Gradient Electrophoresis
3.4. Casting and Running TTGEs
159
1. These instructions assume the use of a Bio-Rad DCode mutation detection system (see Note 11). 2. Add 140 mL 50X TAE to the chamber and then fill the chamber with approximately 7 L deionized water, to a level half way between the ‘‘fill’’ and ‘‘run’’ lines. 3. Place the cover on the buffer chamber and turn on the heater and the pump. Set the heat to approximately 2C higher than desired starting temperature (see Note 12). Heat for approximately 1 h, until the desired temperature has been reached. 4. Select a pair of glass plates (see Note 13) and wash each side with detergent. Rinse the plates with sterile water and treat the inside of each plate with silicon solution (e.g., Sigmacote), following the manufacturer’s instructions. Rinse both plates with sterile water and then with 10% ethanol and allow the plates to air-dry before proceeding. 5. Select a set of spacers (see Note 14) and place these on the larger plate with the grooves up and the ridge out. Position the smaller glass plate directly on top, put a clamp on each side of this gel plate ‘‘sandwich,’’ and fasten loosely. 6. Place the plates in the gel pouring platform, insert the alignment card, and align the spacers as necessary. Ensure the spacers are flush with the edge of the glass or spillage will occur during casting. Once one is satisfied with the alignment, remove the insert and take the plates out of the platform to tighten the clamps. 7. Place a dry foam gasket on the base of the gel pouring platform and position the gel plates on top. Insert and rotate the cams to secure the plates. Place a wide-bore needle tip fitted with a large Luer lock adaptor into the space at the upperright corner of the plates. 8. Pipette 25 mL acrylamide for TTGE (prepared as described in Section 2.3) into a polypropylene screw-top tube and add 220 mL 10% APS and 22 mL TEMED. Mix by gentle inversion and immediately transfer the solution to a syringe by drawing the solution through Tygon tubing that is attached to the syringe by a large Luer lock. 9. Gently expel air and prime the tubing with acrylamide. Once the tubing is full and without air bubbles, attach it to the needle. Pour the gel by gently expelling the acrylamide into the space between the two gel plates, slowly to prevent bubbles forming (see Note 15) and to completely fill the space between the two plates. Remove the needle, tube, and syringe (see Note 16). 10. Push a clean well-forming comb into the gel and ensure no air bubbles are trapped. Allow the gel to polymerize for 30 min.
160
Jones and Knapp
11. Monitor the polymerization of the gel using the remaining acrylamide in the syringe and when the gel has set, remove the plates and gel from the gel pouring platform. Gently remove the comb, flush the wells with sterile water to remove any unpolymerized acrylamide in the empty wells and fit the gel into the upper buffer chamber (see Note 17). 12. Turn off the heater and pump on top of the buffer chamber and remove approximately 500 mL of warm buffer before submerging the gel in the buffer tank. When doing this, partially dip the gel into the buffer several times before it is placed in the tank. This should allow the temperature of the glass to equalize with that of the buffer slowly, thereby preventing the glass breaking. Pour the remaining 500 mL of warm buffer into the upper buffer chamber before proceeding. 13. Vigorously flush the wells with running buffer using a plastic Pasteur pipette. 14. Prepare samples with an equal volume of 2X loading buffer and load them into the wells using a capillary pipette tip. A maximum volume of 45–50 mL can be loaded per well; however, it is often easier to load volumes under 40 mL. 15. Reattach the top and turn on the mains supply and heater. Wait 5 min before turning on the pump so that samples in the wells are not displaced. 16. Once the initial temperature for the TTGE has been reached, set the thermostat to ramp from the lowest to the highest temperature at the desired rate (see Note 18). Electrophorese at the chosen voltage for the appropriate amount of time (see Note 19), monitoring the temperature every 30 min. 3.5. Visualization, and Optional Excision, of DNA Molecules on TTGE Gels
1. Approximately 15 min before end of the run, prepare 150 mL SYBR Gold stain by adding 15 mL SYBR Gold and 3 mL 50X TAE to sterile water in a large, flat glass dish. Cover this entirely with foil to prevent light degradation of the stain. 2. Once the end temperature of the TTGE run has been reached, turn off the system at the mains and remove the upper buffer chamber containing the gel, ensuring that the hot buffer is drained away. 3. Place the upper buffer system on the plastic lined paper and detach the plates containing the gel. 4. With the dish containing the SYBR Gold at hand, separate the glass plates by removing the spacers and carefully easing the plates apart. Typically, the gel will remain on one plate. Hold that plate over the strain and use a spacer to detach a corner of the gel from the plate. The gel should then roll down into the stain. Replace the foil once the gel is in the staining solution.
Temporal Temperature Gradient Electrophoresis
161
5. Place the staining dish on a slow rocker, replace the foil cover, and stain at room temperature for 20 min. 6. Visualize over UV light using a transilluminator and record the image using a camera system (Fig. 9.2). 7. If desired, excise bands of interest with sterile wide-bore pipettes using a cookie-cutter action (see Note 20). Place these gel-containing pipettes into prelabeled sterile 1.5-mL tubes and resuspend the gel plugs from the tips in 100 mL PCR water. Remove the tips and store the tubes at room temperature.
1
2
3
4
5
h
Fig. 9.2. TTGE can be used to separate mitochondrial control region sequences that are the same size, but differ by a single nucleotide. WinMeltTM predicted that the optimal temperature gradient for separation of these two sequences was 48–53C (Fig. 9.1). By electrophoresing PCR-amplified DNA from five individuals for 4.5 h at a rate of 1.2C/h at 130 V, we were able to identify individuals with heteroplasmy (h) in the mitochondrial control region (lanes 3, 4, and 5). The samples in lanes 1 and 2 illustrate results from individuals without heteroplasmy in the control region.
3.6. Optional PCR Amplification for Direct Sequencing of SNPs Identified Using TTGE
1. If DNA sequencing of excised gel bands is to be undertaken, heat samples at 37C for 10 min to elute the DNA from the gel. 2. PCRs to amplify specific sequences can be accomplished with a standard 25 mL PCR that includes 10 mL of gel eluate mixed with 1X PCR buffer, 3 mM MgCl2, 200 mM dNTPs, 0.1 mM forward primer, 0.1 mM reverse primer, 0.5 U Taq polymerase, and sterile water to bring the final volume to 25 mL. 3. A typical thermal cycler program for these PCRs would be one round of initial denaturation at 94C, 40 cycles of denaturation at 94C for 20 s, annealing at 60C for 30 s, and extension at 72C for 36 s, followed by a 10-min extension at 72C. 4. The success and concentration of PCR products can be evaluated by electrophoresing 5 mL of the product on a 1% agarose gel and the remaining product can be incubated with 2 mL ExoSAP-IT at 37C for 15 min, followed by 80C for 15 min if the bands appear distinct and are at least
162
Jones and Knapp
30 ng/mL in concentration (see Note 21). The treated PCR products can then be used in whatever DNA sequencing reaction is preferred by the user.
4. Notes
1. This software can be used to determine the melting temperature of amplified sequences and therefore the temperature range for TTGE. 2. A wide range of Taq polymerases are available but we recommend the use of high-quality Taq polymerases in these reactions to prevent the introduction of PCR errors. Taq polymerase should be removed from the freezer only just before use, otherwise it will degrade. Additionally, all reagents are added to tubes that are on ice, either in a PCR tube freezer block or in an ice bucket. 3. We typically attach the following GC-rich sequence (50 CGCCCGCCGCGCCCCGCGCCCGGCCCGCCGCCGCGGCCGC-30 ). When primers for TTGE are being designed, GC clamps must be attached to the 50 -end of either the forward or the reverse primer. The GC clamp is only attached to one primer, so if the forward primer contains the clamp it should not be attached to the reverse primer. It is possible to determine which position enables optimal amplification by trial and error or by estimation using either WinMeltTM or MacMeltTM. Bear in mind that modeling various lengths and sequences of GC clamps with these programs is often most cost-effective. 4. It is important to dilute PCR reagents in sterile PCR water to avoid exposure to previously amplified DNA that would contaminate PCRs. 5. In all instances, unless stated otherwise, reagents should be prepared in water with a resistivity of 18.2 M -cm. This is referred to as ‘‘sterile water’’ in the text. 6. APS should be ‘‘fresh’’ whenever it is used to polymerize acrylamide. Rather than preparing small volumes before each TTGE experiment, we prepare larger volumes in advance and store small aliquots at –20C until needed. 7. We typically design our PCR primers for post-TTGE amplification by attaching a universal M13Forward sequence to the 50 -end of one primer (10) and a universal M13Reverse sequence to the 50 -end of the other primer to simplify priming in our DNA sequencing reactions. We also choose to make
Temporal Temperature Gradient Electrophoresis
163
one of the two primers 2–3 bp nested within the original TTGE primer sequence, so that we minimize amplification of primer dimers. 8. Typically, the optimal temperature range for TTGE is calculated by subtracting 14C from the ranges estimated by WinMeltTM. However, in practice, we have found that visualization of different sequences is improved if 16C is subtracted from the lowest, initial, temperature estimation. 9. We usually electrophorese 5 mL PCR product on a 1 or 1.5% agarose gel with ethidium bromide staining to check for successful amplification and to estimate the PCR yield. We use a molecular mass standard (Bio-Rad) to estimate the yield and when 20 mg/mL or more of the PCR product is present, we use the sample for TTGE. Care must be taken when using ethidium bromide as it is a potent carcinogen; therefore, relevant safety guidelines for this product need to be followed. 10. During storage, acrylamide and bisacrylamide are slowly converted to acrylic acid and bisacrylic acid, respectively. As this reaction is catalyzed by light, we wrap our solutions in foil to prolong the life of the solution. 11. If your laboratory has multiple systems, even from the same manufacturer, we recommend using the same piece of equipment for TTGE for a particular sequence. We have found that identical TTGE systems produce different results, possibly owing to slight thermostat variations. In addition, we place a thermometer inside the tank to compare the temperature of the buffer with the reading on the thermostat. We also adjust the thermostat accordingly, to ensure the system is operating at the temperature required by WinMeltTM or MacMeltTM. 12. The gel and samples are loaded only when the buffer is heated. Removal of the cover during these processes allows the buffer to cool below the set temperature, so we initially heat our buffer to a few degrees above the conditions required for TTGE. This will ensure the buffer is at the correct temperature for the TTGE run to commence after loading. We also find that it helps to start heating the buffer before casting the gel. By the time the gel has polymerized, the buffer should be at the required temperature. 13. The Bio-Rad DCode system employs glass plates that measure 8 cm 10 cm. One plate is slightly shorter than the other, to allow for contact between the buffer and the acrylamide gel during TTGE. The shorter plate is placed on the inside of the unit during electrophoresis.
164
Jones and Knapp
14. We typically use spacers that are either 0.75- or 1.0-mm thick. The Bio-Rad set includes one spacer with a short groove and one with a long groove. Ensure they are the same height, as spacers shrink when used repeatedly. 15. If any bubbles form during gel pouring, it may be possible to remove them by very gently tapping the gel plates. 16. Syringes should be replaced after they have been used to pour five gels, to ensure that delivery of the acrylamide is smooth during gel pouring. 17. If you are only running one gel, place a dummy gel into the other side to create a complete buffer chamber. 18. If optimal denaturing conditions are determined using WinMeltTM, two temperatures will be identified. The lower temperature will be the initial starting temperature and the higher temperature will be achieved at the end of the TTGE experiment. For example, sequences may be best differentiated between 48 and 53.5C. For this relatively narrow range, we would choose to increase the temperature at a rate of 1.2C/h and this could be achieved by electrophoresing for 4.5 h. 19. We find that the optimal voltage for PCR products that range from 300 to 600 bp is 130 V. 20. Care must be taken because of the UV light on the transilluminator, so wear a laboratory coat with elasticated sleeves and tuck these into gloves. Also wear a visor, if necessary. 21. It is easy to carry out this procedure in a thermocycler with the reagents and PCR product in thin-walled PCR tubes.
Acknowledgements We thank Jo Osborn and Ellie Cannell for technical assistance. References 1. Bogenhagen, D. and Clayton, D. A. (1974) The number of mitochondrial deoxyribonucleic acid genomes in mouse L and human HeLa cells. Quantitative isolation of mitochondrial deoxyribonucleic acid. J. Biol. Chem. 249, 7991–7995. 2. Lynch, M., Koskella, B. and Schaack, S. (2006) Mutation pressure and the evolution of organelle genomic architecture. Science 311, 1727–1730.
3. Tully, L. A., Parsons, T. J., Steighner, R. J., Holland, M. M., Marino, M. A. and Prenger, V. L. (2000) A sensitive denaturing gradientgel electrophoresis assay revels a high frequency of heteroplasmy in hypervariable region 1 of the human mtDNA control region. Am. J. Human. Genet. 67, 432–443. 4. Wang, Q., Ito, M., Adams, K., Li, B. U., Klepstock, T., Maslim, A., Higashimoto, T., Herzog, J. and Boles, R. G. (2004)
Temporal Temperature Gradient Electrophoresis Mitochondrial DNA control region sequence variation in migraine headache and cyclic vomiting syndrome. Am. J. Med. Genet. A 131, 50–58. 5. Chen, T.-J., Boles, R. G. and Wong, L.-J. C. (1999) Detection of mitochondrial DNA mutations by temporal temperature gradient gel electrophoresis. Clin. Chem. 45, 1162–1167. 6. Ito, M., Tran Le, S., Chaudhari, D., Higashimoto, T., Maslim, A. and Boles, R. G. (2001) Screening for mitochondrial DNA heteroplasmy in children at risk for mitochondrial disease. Mitochondrion 1, 269–278. 7. Wong, L.-J. C., Chen, T.-J. and Tan, D.-J. (2004) Detection of mitochondrial DNA
165
mutations using temporal temperature gradient gel electrophoresis. Electrophoresis 25, 2602–2610. 8. Yoshino, K., Nishigaki, K. and Husimi, Y. (1991) Temperature sweep gel electrophoresis: a simple method to detect point mutations. Nucleic Acids Res. 19, 3153. 9. Lerman, L. S. and Silverstein, K. (1987) Computational simulation of DNA melting and its application to denaturing gradient gel electrophoresis. Methods Enzymol. 155, 482–501. 10. Knapp, L. A., Cadavid, L. F., Eberle, M. E., Knechtle, S. J., Bontrop, R. E. and Watkins, D. I. (1997) Identification of novel MamuDRB alleles using DGGE and direct sequencing. Immunogenetics 45, 171–179.
Chapter 10 Zn(II)–Cyclen Polyacrylamide Gel Electrophoresis for SNP Detection Emiko Kinoshita-Kikuta, Eiji Kinoshita, and Tohru Koike Abstract We introduce a method for the detection of single-nucleotide polymorphisms (SNPs) by polyacrylamide gel electrophoresis (PAGE) with an additive Zn2+–cyclen complex (cyclen is 1,4,7,10-tetraazacyclododecane), called ‘‘Zn2+–cyclen–PAGE.’’ The method is based on the difference in mobility of mutant DNA (of the same length) in PAGE, which is due to Zn2+–cyclen binding to thymine bases accompanying a total charge decrease and a local conformation change of target DNA. In combination with a polymerase chain reaction based heteroduplexing technique, the method is more accurate than many other conventional PAGE-based methods, as shown by clear gel-shifting bands because of the separation of heteroduplex and homoduplex DNAs. We demonstrate SNP mapping and heterozygosity screening in a human cardiac sodium channel gene (i.e., SCN5A) that relates to inherited arrhythmia syndromes using Zn2+–cyclen– PAGE. Key words: Single nucleotide polymorphism mapping, heterozygosity screening, heteroduplexing, heteroduplex, homoduplex, polyacrylamide gel electrophoresis, Zn2+–1,4,7,10tetraazacyclododecane.
1. Introduction In this chapter, we describe a gel-electrophoresis-based single nucleotide polymorphism (SNP) detection method, Zn2+–cyclen– PAGE, for the small-scale screening of various disease-causing mutations. The principle of this method is based on the property of Zn2+–cyclen (cyclen is 1,4,7,10-tetraazacyclododecane). Previously, Shionoya et al. (1) reported that Zn2+–cyclen selectively and reversibly binds to an imide-containing nucleoside, deoxythymidine (dT), in an aqueous solution with a dissociation constant Kd = [free dT][free Zn2+–cyclen]/[dT––Zn2+–cyclen] = 0.3 mM at pH 8 (Fig. 10.1a). In the resulting 1:1 complex, the nucleoside is an A.A. Komar (ed.), Single Nucleotide Polymorphisms, Methods in Molecular Biology 578, DOI 10.1007/978-1-60327-411-1_10, ª Humana Press, a part of Springer Science+Business Media, LLC 2003, 2009
169
170
Kinoshita-Kikuta, Kinoshita, and Koike
Fig. 10.1. Equilibrium for Zn2+–cyclen (cyclen is 1,4,7,10-tetraazacyclododecane) binding to deoxythymidine (a) and a proposed mechanism of Zn2+–cyclen binding to doublehelical DNA (b). Native DNA (1) and Zn2+–cyclen-bound DNA (2) in (b). (Reproduced with permission from (8), copyright 2002, Oxford University Press).
imide-deprotonated species (dT–) that binds with the Zn2+ ion, where the total charge of the dT molecule increases from 0 to +1. Furthermore, we reported that Zn2+–cyclen derivatives selectively bind dT-rich regions and change the local conformation in doublestranded DNA (e.g., a bulbous structure, as shown in Fig. 10.1b), as proven by nuclease footprinting experiments and gel mobility shift assays (2–4). The dissociation of A:T hydrogen bonds is promoted by Zn2+–cyclen, as observed by lowering the melting temperature with an increase in the concentration of Zn2+–cyclen (2, 5–7). We extended such a dT-recognizing property of Zn2+– cyclen to the polyacrylamide gel electrophoretic separation of various DNA fragments (Zn2+–cyclen–PAGE) (8–11). The combination of a PCR-based heteroduplex method and Zn2+–cyclen–PAGE enabled a more accurate detection of single mutations introduced artificially even for less detectable substitutions, such as A/T to T/A and G/C to C/G. This combination procedure does not require any radioisotopes or fluorescent probes and basically detects heteroduplex bands on the electrophoresis gel, which arise owing to the annealing of complementary strands, one from mutant DNA and one from wild-type DNA (heterozygosity), during PCR. Our approach is based on three principles: (1) a single-base mismatch
Zn(II)–Cyclen Polyacrylamide Gel Electrophoresis for SNP Detection
171
can produce a local conformational change in the double-stranded DNA, leading to differential migration of heteroduplex and homoduplex bands; (2) the addition of Zn2+–cyclen in the gel, which selectively binds to the thymine bases and disrupts the double strands, can intensify the local conformational change, resulting in increased differential migration of both duplexes; (3) binding of the Zn2+– cyclen to the thymine bases decreases the total charge of the target DNA, helping the detection further. The appearance of slow or differentially migrating bands on the gel indicates the presence of heteroduplex bands, which are suggestive of the existence of mutations or polymorphisms. Furthermore, a homoduplex of a certain mutant allele is separated from a homoduplex of the homologous wild-type allele by using Zn2+–cyclen–PAGE. In this case, four distinct migration bands per DNA sample are detectable at maximum (7). The principles for SNP detection using Zn2+–cyclen–PAGE are schematically summarized in Fig. 10.2. The Zn2+–cyclen–PAGE has been applied to the comprehensive screening of heterozygous mutations scattered throughout a human cardiac sodium channel gene, SCN5A, that is related to inherited arrhythmia syndromes (9, 10).
Fig. 10.2. The principles of the combination procedure of PCR-based heteroduplexing and Zn2+–cyclen–PAGE for single nucleotide polymorphism (SNP) detection.
172
Kinoshita-Kikuta, Kinoshita, and Koike
Recently, we determined the potency of this thymine-dependent electrophoresis method for the detection of the G/C-to-C/G single substitutions in some artificial G/C-lined sequences derived from the SCN5A gene (11). All G/C-to-C/G substitutions in the samples tested, which are one to ten bases away from the nearest T/A, were successfully detected by determining the appropriate DNA length (176–241 bp). In addition, we revealed the DNA binding properties of the Zn2+–cyclen moiety by sequencing analyses of the bands of DNA eluted from the Zn2+–cyclen–PAGE gel. We found that the slowest migration of a C:C heteroduplex in the presence of Zn2+– cyclen contributes to accurate detection when analyzing the G/C-toC/G single substitutions in the G/C-lined sequences. The disruption of A:T base pairing due to the binding of Zn2+–cyclen affects even a mismatched site that is ten bases away, causing a detectable local conformation change of target DNA in the Zn2+–cyclen–PAGE gel. Therefore, it is worthwhile considering using Zn2+–cyclen–PAGE for the discovery of various SNPs or disease-causing mutations in the genomic coding region from a small number of samples. As a practical example, we demonstrate SNP mapping and heterozygosity screening in the human SCN5A gene, which is related to inherited arrhythmia syndromes, using Zn2+–cyclen–PAGE.
2. Materials 2.1. Preparation of Genomic DNA
1. Peripheral blood was obtained from 18 unrelated Japanese subjects: ten patients diagnosed as having Brugada syndrome, one as having idiopathic ventricular fibrillation without ST segment elevation in the right precordial electrocardiogram leads, and one as having long QT syndrome type 3, and six healthy individuals without family histories of syncope or sudden death. Written informed consent for participation was obtained from all subjects. 2. The genomic DNA was extracted from the leukocytes according to the standard protocol using the QIAamp DNA blood maxi kit (Qiagen, Hilden, Germany). 3. The DNA concentration was determined by using an ethidium bromide fluorescent quantitative method (12).
2.2. Preparation of PCR Primers for Amplification of All SCN5A Exons and Their Splice Sites
1. The optimum primers used (98 pairs; the primer sets are listed in Table 10.1) were originally designed by referring to the sequences of DDBJ GenBank accession nos. AP006241 (chromosome 3) and M77235 (complementary DNA of SCN5A) so that the length of each PCR product would not exceed about 200 bp for accurately detecting mutations.
Zn(II)–Cyclen Polyacrylamide Gel Electrophoresis for SNP Detection
173
Table 10.1 Optimum primer sets used to amplify SCN5A exons and exon–intron boundaries Exon no.
Forward primer (50 –30 )
Reverse primer (50 –30 )
Size of product (bp)
1
gccgctgagcctgcgcccag
ggaaagttgggcggcggcag
172
2-1
ggcccattgtctgtgtcttc
ggtgccccgaggtaatagga
105
2-2
ggatgagaagatggcaaact
tgcttctctgccatgcgctt
105
2-3
agtccctggcagccatcgag
ggccggggagcctcctcctc
106
2-4
agagccgagaggggctgccc
atgagctcttggggtggatt
106
2-5
agctgccagatctctatggc
cctcttccccctctgctccatt
161
3-1
taatgctactgtctgtcccc
gaagggactgaggacataca
100
3-2
gttcagtgccaccaacgcct
ggtactcagcaggtattaactgcaa
113
4-1
ggtagcactgtcctggcagtgat
ggtcgtgctgggccatgaac
166
4-2
accatcctcaccaactgcgt
aagaggccctgaagatactc
87
5-1
tcactccacgtaaggaacctg
aggaaagtgaacgcgtgcag
161
5-2
ttctggctcgaggcttctgc
atgtggactgcagggaggaagc
203
6-1
ctgtgaatttcccttgttac
cccggaggactcggaaggtg
6-2
ggcaatgtctcagccttacg
ggtattctggtgacaggcacattc
7-1
gtgcttgttcttgccttccc
cagaagactgtgaggaccat
99
7-2
tgaagaagctggctgatgtg
acttgtgccttaggttgccc
99
7-3
atcggcctgcagctcttcat
caagccgtcggcctccacgg
99
7-4
gctcaacggcaccaacggct
gtctgcggtctcacaaagtcttc
124
8-1
ccagcatgatgtttctctta
ggagactcccctggcaggacaa
134
9-1
gggagacaagtccagcccagcaa
agaaaggcccaggcaaagga
173
9-2
acggctacaccagcttcgat
agcccacacttgctgtcccttg
129
10-1
cctctgcccccttgctcccc
gatcaggttcaccaggtaga
107
10-2
tgtcatcttcctggggtcct
ccttctcctcggtctcagcg
107
10-3
gagcaaaaccaagccaccat
cacctataggcaccatcagtcag
150
11-1
ccgttcctccactcttttcc
tcatggctgtttactggggc
102
11-2
ccttggagatgtcccctttg
ctgtcctccccacactcctc
100
11-3
aacggatgtcttcaggaact
ctagaaccacagctgggatt
102
12-1
gccagtggctcaaaagacaggct
gcgaaaggtgaaaatgctcc
159
12-2
gaagccacgttccagccgcg
ctctcccccgctgtgctgtt
102
87 182
(continued)
174
Kinoshita-Kikuta, Kinoshita, and Koike
Table 10.1 (continued) Exon no.
Forward primer (50 –30 )
Reverse primer (50 –30 )
Size of product (bp)
12-3
cagattttgcagatgatgaa
cacagtgctgttctttttgc
164
12-4
tcctggccacgccctccatg
cttcctggggatgtggcctc
102
12-5
tactgggggcaggcgaccca
ggaactgctgatcagtttgggaga
147
13-1
cccttttccccagctgacgcaaa
tgccgtgctcctggctcctc
130
13-2
ctccctgtgtagatggcttc
gcttaaatgacctggggttg
109
14-1
aactcattggctgtcccctc
ggcagcactcccagatcagg
100
14-2
aaccgtctcgcccagcgcta
ggtcagtaaacgggtccatg
100
14-3
cagggagtgaagttggtggt
tgttgtagtgctccagcgcc
100
14-4
gtactcaacacactcttcat
gccagctgcaggcagccctt
103
15-1
ctttcctatcccaaacaatacct
ccctgttggaagtagtagta
113
15-2
agatcattgccctcgacccc
ggacaggcccagctccatga
98
15-3
catcatcgtcatccttagcc
ccccaccatcccccatgcagt
116
16-1
gagccagagaccttcacaaggtcccct
aggttccccagtgcccccac
135
16-2
tcaagatcatcgggaactca
gtagttcttgccaaagagct
110
16-3
ctttgctgtggtgggcatgc
gaggaaggcatgaaagaagt
112
16-4
gcctcgctggcacatgatgg
actgccccgacacctccatg
110
16-5
atcgagaccatgtgggactg
gtgtggcccttggccaactt
112
17-1
gggactggatggcttggcatggt
ttcatctctctgtcctcatc
143
17-2
cagacaacctcacagcccct
caggtggtccgcttgacaaa
103
17-3
gcatccagaggggcctgcgc
ccctgggcggcaagggctgc
103
17-4
cctgcggcagcggcctcaga
tgggaggcaccttctccgtc
113
17-5
ccccctactccccgccaccc
tcttcttcttggtcatctgt
166
17-6
tcgctgtggccgagtcagac
ctgtatatgtaggtgccttatacatg
126
18-1
ctcataggctggggtctttt
gctgccgccagtcggcctga
147
18-2
gaggccgaggccagtgcatc
cccagctggcttcagggacaaa
112
19-1
tgtcccacacccctgtccat
aggtcagggatctgctccag
102
19-2
tgaccaacaccgctgagctc
cctgtcccctctgggtggaact
136
20-1
tgaccctcactctctcccat
aaccgccaccagaccttccc
92
20-2
tggacaccacacaggcccca
tgatgaatgtctcgaaccag
93
20-3
taccacatcgtggagcacag
tgacctgactttccagctggaga
164 (continued)
Zn(II)–Cyclen Polyacrylamide Gel Electrophoresis for SNP Detection
175
Table 10.1 (continued) Exon no.
Forward primer (50 –30 )
Reverse primer (50 –30 )
Size of product (bp)
21-1
tccaggcttcactgtccacctgtct
tatgtgaacatcttgtcggc
116
21-2
cttgtctggccttgtctgca
tatgtgaacatcttgtcggc
99
21-3
ctgctcaagtgggtggccta
cccttcgggtgcccacactc
100
22-1
agtggggagctgttcccatcct
gtgcacgcagcgtccgcagt
130
22-2
gagatgggccccatcaagtc
ggaaggcagccacctctctt
103
23-1
ttgaaaaggaaatgtgctctggg
atgatgctgaagatgagcca
150
23-2
tcctcgtctgcctcatcttc
tagttcaaaggcaagtctcc
112
23-3
ggtgcatcaaccagacagag
ttcaccttggtccagtacaa
112
23-4
ccttgaacttgaccggagaa
aacatcatgggtgatggccat
161
24-1
ctcaagcgaggtacagaattaaatga
gggctttcagatgcagacactgat
202
25-1
gcctgtctgatctccctgtgtga
gaaagacccaaagatgatga
115
25-2
gtacatctattttgtcattt
tggggagctggtgctctacg
123
26-1
ccatgctggggcctctgagaac
ggagcccagcttcttcatgg
126
26-2
gcagaagaagtactacaatg
ggctctgatggctggccatgtg
162
27-1
cccagcgagcactttccatttg
caccatattcaagcagatca
122
27-2
tgacgtcaccatcatgtttc
agcaggttgatcttggccaa
108
27-3
gtcctgagaaaatcaacatc
tggtgaagtagtagtggcgc
108
27-4
attgtcaagctggctgccct
gcttctccgtccagctgacttgta
119
28-1
tgcacagtgatgctggctggaa
ggccaggcggatgactcgga
135
28-2
cttcttctccccgacgctct
aaagagcagcgtgcggatcc
100
28-3
actgatccgaggggccaagg
catgacgaggaagagcagca
100
28-4
tgccctcttcaacatcgggc
gatgccagcctcccacttga
100
28-5
catggccaacttcgcttatg
ggtgatctggaagaggcaca
100
28-6
gaccttcgccaacagcatgc
gcagtagggcggcccagtgt
100
28-7
cctcctcagccccatcctca
gcccacggctgggctcccgc
100
28-8
caatggctctcggggggact
atgtacatgttgaccacgat
99
28-9
acatcatcatctccttcctc
cactcaggggctcggtgctc
99
28-10
ttcagcgtggccacggagga
ctgagtggcctctgggtcaa
99
28-11
ctatgagatctgggagaaat
atacggagtggctcagacag
99
28-12
tgtctgactttgccgatgcc
ggtccccactcaccatgggc
99 (continued)
176
Kinoshita-Kikuta, Kinoshita, and Koike
Table 10.1 (continued) Exon no.
Forward primer (50 –30 )
Reverse primer (50 –30 )
28-13
agcctcatcaacatggacct
cccagactcccccaggaccc
99
28-14
tctctttgccttcaccaaaa
ttggatgggttggctgccat
99
28-15
tccagatggaggagaagttc
tggccgacacctcttcgtgc
99
28-16
accaccacactccggcgcaa
ggaggcatgcttcaaagagc
99
28-17
ccgcaggcacctgctgcaac
tctcgctcaggggcatcctc
99
28-18
cgggcagcggcctctccgaa
tggagggtgggccaaggggt
99
28-19
atgagtgagaacttctcccg
ggtggctctagtgacactgt
99
28-20
ttccttcccaccctcctatg
aagtcggcgagatcttcact
99
28-21
gggggtctgactacagccac
ctggccagccaggccgaggc
Size of product (bp)
100
The 98 pairs of primers were synthesized by referring to the sequences of GenBank accession nos. AP006241 (chromosome 3) and M77235 (complementraty DNA of SCN5A). Reproduced with permission from (9), copyright 2005, American Association for Clinical Chemistry, Inc.
2. Primers obtained from Texas Genomics Japan (Tokyo, Japan) and Espec Oligo Service (Tsukuba, Japan) were used. 2.3. Polymerase Chain Reaction
1. KOD -Plus- DNA polymerase (Toyobo, Osaka, Japan). 2. PCR mixture: 0.20 mM forward and reverse primers, 2.5 ng genomic DNA template, dNTPs (each at 0.20 mM), 1.0 mM MgSO4, 0.10 U KOD -Plus- DNA polymerase, its special buffer (Toyobo) in a volume of 5.0 mL. Prepare just before use.
2.4. Zn2+–Cyclen–PAGE and Visualization of DNA Bands
1. Thirty percent (w/v) acrylamide/bisacrylamide solution (99:1 ratio of acrylamide to N,N’-methylenebisacrylamide) for the separating gel. Store at room temperature in the dark (see Note 1). 2. Thirty percent (w/v) acrylamide/bisacrylamide solution (29:1 ratio of acrylamide to N,N’-methylenebisacrylamide) for the stacking gel. Store at room temperature in the dark. 3. Separating gel buffer (4): 1.5 M tris(hydroxymethyl)aminomethane (Tris)–HCl, pH 8.8. Store at room temperature. 4. Stacking gel buffer (4): 0.5 M Tris–HCl, pH 6.8. Store at room temperature. 5. Zn2+–cyclen solution: 500 mM Zn2+–cyclen (prepared as the dinitrate salt, see Note 2) (8) in distilled water. Adjust to pH 7.9 using 10 M NaOH. Store at room temperature in the dark.
Zn(II)–Cyclen Polyacrylamide Gel Electrophoresis for SNP Detection
177
6. Zinc nitrate solution: 500 mM Zn(NO3)2 6H2O in distilled water. Store at room temperature. 7. Ammonium persulfate solution and N,N,N’,N’-tetramethylethylenediamine (TEMED) (see Note 3): 10% (w/v) ammonium persulfate in distilled water. Prepare just before use. 8. Electrophoresis running buffer (1): 25 mM Tris, 192 mM glycine. Store at room temperature (see Note 4). 9. Sample loading dye solution: 50 mM Na2EDTA, pH 8.0, 30% (v/v) glycerol, 0.05% (w/v) bromophenol blue. Store at room temperature. 10. Gel staining solution: 10,000-fold diluted SYBR green I (Cambrex Bio Science Rockland, Rockland, ME, USA) staining (15 mL per gel) (see Note 5). Prepare just before use.
3. Methods We introduce the procedure of Zn2+–cyclen–PAGE for simple and effective genetic testing using autosomal dominant inherited disorder samples. The Zn2+–cyclen–PAGE is based on the principle that the binding of Zn2+–cyclen [i.e., a mononuclear zinc(II) complex] to the thymine bases changes the local DNA conformation, resulting in differences in the electrophoresis migration of a mutant DNA. In combination with a heteroduplexing technique, the method enables a more accurate detection of single-nucleotide mutations for SNP mapping and heterozygosity screening without using radioisotopes, expensive fluorescent primers, or a special apparatus for real-time PCR. Since this procedure just requires a general PCR apparatus, a mini-slab PAGE system, and an additive of Zn2+–cyclen, it represents a very simple and convenient tool for SNP discovery and analysis. 3.1. PCR for Amplification of All SCN5A Exons and Their Splice Sites
1. PCR for amplification of all SCN5A exons and the splice sites was performed with the PCR mixture described above (final volume, 5.0 mL) using 98 pairs of forward and reverse primers (each at 0.20 mM, see Table 10.1), 2.5 ng of genomic template from 18 individuals, 1.0 mM MgSO4, 0.1 U KOD -Plus- DNA polymerase and its special buffer, and dNTPs (each at 0.20 mM). 2. After initial denaturation at 95C for 3 min, the PCR amplification was performed for 30 cycles of 30-s denaturation at 95C, 30-s annealing at 55–68C, and 30-s extension at 68C (see Note 6). 3. Each PCR mixture (5.0 mL) was dissolved in a half amount of the sample loading dye solution described above. 4. Each sample solution (0.5 mL, 0.5–1.0 ng DNA per well) was applied to the polyacrylamide gel containing Zn2+–cyclen (see Note 7).
178
Kinoshita-Kikuta, Kinoshita, and Koike
3.2. Zn2+–Cyclen–PAGE for SNP Analysis
1. This instruction assumes the use of an Atto model AE-6500 mini-slab gel system (Tokyo, Japan). This will be easily adaptable to other formats, including large-type gels. It is critical that the glass plates (1 mm thick, 9 cm wide, and 9 cm long) for making gels are scrubbed clean with a proper detergent after use and rinsed extensively with distilled water. 2. Set up the casting apparatus. 3. Twenty percent (w/v) separating gel solution (6.5 mL) (see Note 8) containing 5.0 mM Zn2+–cyclen is prepared by mixing 4.33 mL of 30% (w/v) acrylamide/bisacrylamide solution (99:1 ratio of acrylamide to N,N’-methylenebisacrylamide), 1.63 mL of separating gel buffer (4), 65 mL of Zn2+–cyclen solution, and 0.37 mL of distilled water. 4. Degas the mixed separating gel solution under vacuum for 15 min. 5. Add 15 mL of TEMED and 90 mL of ammonium persulfate solution to the degassed solution and mix gently without bubbling. 6. Transfer the separating gel solution (6.3 mL) between the glass plates, pore distilled water (1 mL) on top of the separating gel solution, and then allow the gel to polymerize for about 30 min. 7. The stacking gel solution [4.5% (w/v), 2.0 mL] is prepared by mixing 0.30 mL of 30% (w/v) acrylamide/bisacrylamide solution (29:1 ratio of acrylamide to N,N ’-methylenebisacrylamide), 0.50 mL of stacking gel buffer (4), and 1.15 mL of distilled water. 8. Degas the mixed stacking gel solution under vacuum for 15 min. 9. Add 10 mL of TEMED and 40 mL of ammonium persulfate solution to the degassed solution and mix gently without bubbling. 10. Rinse the top of the separating gel polymerized with distilled water and remove the residual liquid with a paper towel. 11. Pour the stacking gel solution (1.8 mL) on top of the separating gel and then insert a sample-well comb. Allow the gel to polymerize for 1 h. 12. Assemble the electrophoresis apparatus and fill the electrode lower chambers with 150 mL of the electrophoresis running buffer (1) and 1.5 mL of zinc nitrate solution described above (see Note 9). 13. Gently remove the comb from the stacking gel and place the gel in the electrophoresis apparatus. Fill the electrode upper chambers with the electrophoresis running buffer (1). 14. Apply the samples (0.5 mL, 0.5–1.0 ng DNA per well) to the wells.
Zn(II)–Cyclen Polyacrylamide Gel Electrophoresis for SNP Detection
179
15. Attach the leads to a power supply (e.g., ATTO AE-8750 Power Station 1000XP). Run the gels under a constantcurrent condition (25 mA per gel) and at room temperature until the bromophenol blue reaches the bottom of the separating gel (for about 100 min). 16. After running is over, remove the gel from the apparatus and stain it with gel staining solution (described above) to visualize the DNA bands. Typical results of Zn2+–cyclen–PAGE are shown in Fig. 10.3.
Fig. 10.3. Zn2+–cyclen–PAGE for heterozygosity screening. The screening method revealed that nine patients and two healthy individuals have various heterozygosities in the SCN5A gene. Lanes 1–10, 11, 12, and 13–18 in a–d indicate Brugada, idiopathic ventricular fibrillation (IVF), long QT syndrome type 3 (LQT3), and healthy individuals as controls, respectively. (a) PCR products (named as exon 10-2 in Table 10.1; 107 bp with primers 50 -tgtcatcttcctggggtcct and 50 -ccttctcctcggtctcagcg) containing the sequences of exon 10 were analyzed by using Zn2+–cyclen–PAGE. (b) Two additional differentially migrating bands representing heteroduplexes in lane 1 were clearly confirmed. Subsequent direct sequencing of the DNA fragment including exon 10-2 of patient 1 revealed heterozygous nucleotide substitution (G1212A), which does not affect codon L404. PCR products (named as exon 2-3 in Table 10.1; 106 bp with primers 50 -agtccctggcagccatcgag and 50 -ggccggggagcctcctcctc) containing the sequences of exon 2 were analyzed. (c) A similar synonymous SNP (i.e., G87A in codon A29) was detected in exon 2 of five Brugada patients and two control subjects by subsequent direct sequencing. PCR products (named as exon 21-3 in Table 10.1; 100 bp with primers 50 ctgctcaagtgggtggccta and 50 -cccttcgggtgcccacactc) containing the sequences of exon 21 were analyzed. (d) Subsequent direct sequencing identified a heterozygous G to A transversion in the 50 splice junction of the intron between exons 21 and 22, suggesting abnormal splicing linked to the Brugada syndrome (lane 2). PCR products (named as exon 20-2 in Table 10.1; 92 bp: 50 -tggacaccacacaggcccca and 50 -tgatgaatgtctcgaaccag) containing the sequences of exon 20 were analyzed. Subsequent direct sequencing identified a common nucleotide alteration (G3575A) leading to a codon mutation of R1192Q in exon 20 of IVF and LQT3 patients (lanes 11 and 12). This mutation has been reported to be associated with the Brugada (13) and LQT3 (14) syndromes. No heterozygous mutation resulting in changes in the codon was detected in the control subjects. The direct sequencing revealed no mutation in the patient samples showing a single DNA band, indicating that all heterozygous mutations in SCN5A gene tested were detected by our screening. (Reproduced with permission from (9), copyright 2005, American Association for Clinical Chemistry, Inc).
180
Kinoshita-Kikuta, Kinoshita, and Koike
17. Only PCR products that show a heteroduplex pattern in the Zn2+–cyclen–PAGE (multiple DNA bands) are applied to DNA sequencing to identify the SNPs.
4. Notes
1. Since the unpolymerized acrylamide is a neurotoxin, care should be taken to avoid exposure. 2. Zn2+–cyclen is commercially available as an advanced material by consignment synthesis from the Phos-tag consortium (Japan; http://www.phos-tag.com/english/index.html. 3. TEMED is stored in a desiccator at room temperature. Buy small bottles as the quality of TEMED may decline (gels will take longer to polymerize) after the bottle has been opened. 4. By adopting a discontinuous buffer system with Tris–HCl as a separating gel buffer and Tris–glycine as an electrophoresis running buffer, we can increase the detection sensitivity. 5. It is possible to use 10 mg/mL ethidium, bromide in distilled water. Prepare the solution just before use. 6. Determine the best annealing temperatures in the range 55– 68C according to the calculated melting temperature values of the primer sets used. 7. When a large amount of sample solution (more than 2.0 mL) is applied, the tailing of DNA bands will be often observed. In such a case, we need to make a stacking gel as described in this article. By adopting the stacking gel, we can observe DNA bands more sharply and increase the detection sensitivity. 8. Determine the best electrophoresis condition of the percentage of acrylamide gel. A 20% (w/v) gel is appropriate for the DNA size range from 85 to 200 bp in the mini-slab gel system. 9. Take care not to produce bubbles at the bottom surface of the gel set. When bubbles are observed, they should be removed completely.
Acknowledgements The authors would like to express their sincere gratitude to the subjects who provided blood samples for this study. We gratefully acknowledge helpful discussions with Mitsuhiko Shionoya (University of Tokyo), Kazuaki Chayama (Hiroshima University), and
Zn(II)–Cyclen Polyacrylamide Gel Electrophoresis for SNP Detection
181
Yukiko Nakano (Hiroshima University). We also wish to thank the Research Center for Molecular Medicine and the Analysis Center of Life Science, Hiroshima University, Japan, for the use of their facilities. This work was supported by Grants-in-Aid for Scientific Research (B) (19390011) and (C) (19590040) from the Japan Society of the Promotion of Science (JSPS), a Grant-inAid for Young Scientists (B) (18790120) from the Ministry of Education, Culture, Sports, Science, and Technology (MEXT), a research grant for Promoting Technological Seeds from the Japan Science and Technology Agency (JST), and research grants from the Ube Foundation (The Watanabe Memorial Award), the Iketani Science and Technology Foundation, and the Takeda Science Foundation. References 1. Shionoya, M., Kimura, E. and Shiro, M. (1993) A new ternary zinc(II) complex with [12]aneN4 (=1,4,7,10-tetraazacyclododecane) and AZT (= 3’-azido-3’-deoxythymidine). Highly selective recognition of thymine and its related nucleotides by a zinc(II) macrocyclic tetraamine complex with novel complementary associations. J. Am. Chem. Soc. 115, 6730–6737. 2. Kikuta, E., Aoki, S. and Kimura, E. (2002) New potent agents binding to poly(dT) sequence in double-stranded DNA: bis(Zn2+–cyclen) and tris(Zn2+–cyclen) complexes (cyclen = 1,4,7,10-tetraazacyclododecane). J. Biol. Inorg. Chem. 4, 473–482. 3. Kikuta, E., Murata, M., Katsube, N., Koike, T. and Kimura, E. (1999) Novel recognition of thymine base in double-stranded DNA by zinc(II)–macrocyclic tetraamine complexes appended with aromatic groups. J. Am. Chem. Soc. 121, 5426–5436. 4. Kikuta, E., Koike, T. and Kimura, E. (2000) Controlling gene expression by zinc(II)– macrocyclic tetraamine complexes. J. Inorg. Biochem. 79, 253–259. 5. Kimura, E., Ikeda, T., Aoki, S. and Shionoya, M. (1998) Macrocyclic zinc(II) complexes for selective recognition of nucleobases in single- and double-stranded polynucleotides. J. Biol. Inorg. Chem. 3, 259–267. 6. Kikuta, E., Katsube, N. and Kimura, E. (1999) Natural and synthetic doublestranded DNA binding studies of macrocyclic tetraamine zinc(II) complexes appended with polyaromatic groups. J. Biol. Inorg. Chem. 4, 431–440.
7. Kinoshita, E., Kinoshita-Kikuta, E. and Koike T. (2005) A heteroduplex-preferential Tm depressor for the specificity-enhanced DNA polymerase chain reactions. Anal. Biochem. 337, 154–160. 8. Kinoshita-Kikuta, E., Kinoshita, E. and Koike T. (2002) A novel procedure for simple and efficient genotyping of single nucleotide polymorphisms by using the Zn2+–cyclen complex. Nucleic Acids Res. 30, e126. 9. Kinoshita, E., Kinoshita-Kikuta, E., Kojima, H., Nakano, Y., Chayama, K. and Koike, T. (2005) Reliable and cost-effective screening of inherited heterozygosity by Zn2+–cyclen polyacrylamide gel electrophoresis. Clin. Chem. 51, 2195–2197. 10. Nakano, Y., Tashiro, S., Kinoshita, E., Kinoshita-Kikuta, E., Takenaka, S., Miyoshi, M., Ogi, H., Sakoda, E., Oda, N., Suenari, K., Tonouchi, Y., Okimoto, T., Hirai, Y., Miura, F., Yamaoka, K., Koike, T. and Chayama, K. (2007) Non-SCN5A related Brugada syndromes: Verification of normal splicing and trafficking of SCN5A without exonic mutations. Ann. Hum. Genet. 71, 8–17. 11. Kinoshita, E., Kinoshita-Kikuta, E., Yoshimoto, M. and Koike, T. (2008) Detection of the Gua/Cyt-to-Cyt/Gua mutation in a Gua/Cyt-lined sequence using Zn2+–cyclen polyacrylamide gel electrophoresis. Anal. Biochem. 380, 122–127. 12. Sambrook, J., Fritsch, E. F. and Maniatis, T. (1989) Molecular cloning 3: a laboratory manual, 2nd edition, Cold Spring Harbor Laboratory Press, New York, pp. E.5–E.6.
182
Kinoshita-Kikuta, Kinoshita, and Koike
13. Vatta, M., Dumaine, R., Varghese, G., Richard, T. A., Shimizu, W., Aihara, N., Nademanee, K., Brugada, R., Brugada, J., Veerakul, G., Li, H., Bowles, N. E., Brugada, P., Antzelevitch, C. and Towbin, J. A. (2002) Genetic and biophysical basis of sudden unexplained nocturnal death syndrome (SUNDS), a disease allelic to Brugada syndrome. Hum. Mol. Genet. 11, 337–345.
14. Wang, Q., Chen, S., Chen, Q., Wan, X., Shen, J., Hoeltge, G. A., Timur, A. A., Keating, M. T. and Kirsch, G. E. (2004) The common SCN5A mutation R1193Q causes LQTS-type electrophysiological alterations of the cardiac sodium channel. J. Med. Genet. 41, e66.
Chapter 11 Phosphate-Affinity Polyacrylamide Gel Electrophoresis for SNP Genotyping Eiji Kinoshita, Emiko Kinoshita-Kikuta, and Tohru Koike Abstract We introduce a genotyping method which relies on the use of a 1:1 mixture of 50 -phosphate-labeled and nonlabeled allele-specific primers for polymerase chain reaction (PCR). The method is based on the difference in mobility of the phosphorylated and nonphosphorylated PCR products (possessing the same number of base pairs) during phosphate-affinity polyacrylamide gel electrophoresis (PAGE). The phosphate-affinity site in the gel is represented by an immobilized phosphate-binding tag molecule [i.e., a polyacrylamide-bound dizinc(II) complex], which selectively captures the 50 -phosphate-labeled allelespecific product compared with the corresponding nonlabeled one. The DNA migration bands obtained can be visualized by ethidium bromide staining. We demonstrate the genotyping of a single-nucleotide polymorphism reported in a human cardiac sodium channel gene, SCN5A, using the phosphate-affinity PAGE. Key words: Single-nucleotide polymorphism genotyping, allele-specific PCR, phosphate-affinity polyacrylamide gel electrophoresis, zinc, Phos-tag.
1. Introduction In this article, we describe a gel-electrophoresis-based method for single nucleotide polymorphism (SNP) genotyping. The principle of this method is based on our recent findings on a dinuclear metal complex (i.e., 1,3-bis[bis(pyridin-2-ylmethyl)amino]propan-2olato dizinc(II) complex, Phos-tag) (1), which acts as a phosphate-binding tag molecule, in an aqueous solution at a neutral pH (e.g., Kd ¼ 25 nM for phenyl phosphate dianion) (see the structure of the R-OPO32–-bound Zn2+–Phos-tag in Fig. 11.1.). The Phos-tag has a vacancy on two metal ions that is suitable for A.A. Komar (ed.), Single Nucleotide Polymorphisms, Methods in Molecular Biology 578, DOI 10.1007/978-1-60327-411-1_11, ª Humana Press, a part of Springer Science+Business Media, LLC 2003, 2009
183
184
Kinoshita, Kinoshita-Kikuta, and Koike
Fig. 11.1. Structure of acrylamide-pendant Phos-tag ligand and scheme of the reversible capturing of a phosphomonoester dianion (R-OPO32–) by Zn2+–Phos-tag.
the access of a phosphomonoester dianion as a bridging ligand. The resulting 1:1 phosphate-binding complex, ROPO32––Phostag3+, has a total charge of +1. The anion selectivity indexes of the phenyl phosphate dianion against SO42–, CH3COO–, Cl–, and the bisphenyl phosphate monoanion at 25C are 5.2 103, 1.6 104, 8.0 105, and more than 2 106, respectively. These findings have recently been employed for the development of procedures for the matrix-assisted laser desorption/ionization time-of-flight mass spectrometry of phosphorylated compounds (e.g., phosphopeptides and phospholipids) (2), the immobilized metal affinity chromatography based separation and enrichment of phosphoproteins and phosphopeptides (3, 4), the surface plasmon resonance analysis of phosphopeptides (5), and the western blotting analysis of phosphoproteins on a poly(vinylidene difluoride) membrane (6). Furthermore, we have demonstrated that the phosphate-affinity gel electrophoresis using a manganese(II) homologue can be employed for the mobility shift detection of phosphoproteins from their nonphosphorylated counterparts (6–9). In the latter case, for the separating gel in the affinity sodium dodecyl sulfate polyacrylamide gel electrophoresis (PAGE), we utilized a polyacrylamide-bound Mn2+–Phos-tag as a copolymer. Recently, we adapted the electrophoretic separation method for SNP genotyping (10); the separation of phosphate-labeled and nonlabeled PCR products (i.e., allele-specific DNA in the same number of base pairs) was conducted by a similar phosphate-affinity PAGE but with a Phos-tag complex with two zinc(II) ions (Zn2+– Phos-tag PAGE). This method has the following features: (1) the procedure is almost identical to that of the general PCR or PAGE
Phosphate-Affinity Polyacrylamide Gel Electrophoresis
185
system; (2) a radioactive label is avoided; (3) it enables the separation of the 50 -OPO32–-labeled DNA fragment as a slower migration band from the nonlabeled counterpart; (4) it enables the determination of the genotype of the heterozygote or homozygote owing to the difference in the migration degree of allele-specific PCR products; (5) it enables the application of the procedure for gene diagnosis of both dominant and recessive alleles. As a practical example, we demonstrate here the SNP genotyping of a silent mutation reported in the SCN5A gene (11) using Zn2+–Phos-tag PAGE.
2. Materials 2.1. Preparation of Genomic DNA
1. Peripheral blood was obtained from seven healthy blood donors. Written informed consent for participation was obtained from all the donors. 2. The genomic DNA was extracted from the leukocytes according to the standard protocol using the QIAamp DNA blood maxi kit (Qiagen, Hilden, Germany). 3. The DNA concentration was determined by using an ethidium bromide fluorescent quantitative method (12).
2.2. Preparation of Primers for AlleleSpecific PCR
1. Allele-specific PCR primers containing a single nucleotide variation of G and A in exon 2 of the SCN5A gene (a G-allele-specific primer, 50 -GCCGCGGGCTTGCTTCTCCG-30 , and an Aallele-specific primer, 50 -GCCGCGGGCTTGCTTCTCTG-30 , underlined at the variation site) and a reverse PCR primer, 50 -GGTCTGCCCACCCTGCTCTC-30 , were designed by referring to the sequence of DDBJ GenBank accession no. M77235 (complementary DNA of SCN5A) for amplification of the target genome region (220 bp) (see Notes 1 and 2). 2. Primers were obtained from Invitrogen Japan (Tokyo, Japan). 3. T4 polynucleotide kinase (New England Biolabs, Beverly, MA, USA) reaction buffer: 70 mM tris(hydroxymethyl)aminomethane (Tris)–HCl, pH 7.6, 10 mM MgCl2, 5.0 mM dithiothreitol, 1.0 mM ATP.
2.3. Allele-Specific PCR
1. KOD -Plus- DNA polymerase (Toyobo, Osaka, Japan). 2. PCR mixture: a 1:1 mixture of allele-specific primers (each at 0.10 mM), 0.20 mM reverse primer, various concentrations of the genomic DNA template (500, 250, 125, 63, or 32 pg), dNTPs (each at 0.20 mM), 0.80 mM MgSO4, 0.10 U KOD -Plus- DNA polymerase, its special buffer (Toyobo) in a volume of 5.0 mL. Prepare just before use.
186
Kinoshita, Kinoshita-Kikuta, and Koike
2.4. Zn2+–Phos-tag PAGE and Visualization of DNA Bands
1. Thirty percent (w/v) acrylamide/bisacrylamide solution (99:1 ratio of acrylamide to N,N ’-methylenebisacrylamide). Store at room temperature in the dark (see Note 3). 2. Separating gel buffer (4 ): 1.5 M Tris–HCl, pH 8.8. Store at room temperature. 3. Phos-tag solution: 5.0 mM acrylamide-pendant Phos-tag ligand obtained from the Phos-tag consortium (Japan; http://www.phos-tag.com/english/index.html) in distilled water. Store at room temperature in the dark. (see Notes 4 and 5). 4. Zinc nitrate solution: 10 mM Zn(NO3)2 6H2O in distilled water. Store at room temperature. 5. Ammonium persulfate solution and N,N,N ’,N ’-tetramethylethylenediamine (TEMED) (see Note 6): 10% (w/v) ammonium persulfate in distilled water. Prepare just before use. 6. Electrophoresis running buffer (1 ): 25 mM Tris, 192 mM glycine. Store at room temperature (see Note 7). 7. Sample loading dye solution: 50 mM Na2EDTA, pH 8.0, 30% (v/v) glycerol, 0.05% (w/v) bromophenol blue. Store at room temperature. 8. Gel staining solution: 10 mg/mL ethidium bromide in distilled water. Prepare just before use.
3. Methods We introduce here the procedure for the phosphate-affinity gel electrophoresis, Zn2+–Phos-tag PAGE, for a simple and accurate analysis of DNA mutations. The Zn2+–Phos-tag PAGE is based on the principle that the binding of Zn2+–Phos-tag immobilized in the gel to the phosphate group at the 50 -end of DNA fragment results in slower electrophoresis mobility of the 50 -OPO32–labeled DNA fragment compared with the corresponding nonlabeled one. The combination of the allele-specific PCR technique and Zn2+–Phos-tag PAGE enables simple and accurate typing of SNPs without using radioisotopes, expensive fluorescent primers, or a special apparatus for real-time PCR. Since this procedure requires a general PCR apparatus, a mini-slab PAGE system, and a gel-bound Zn2+–Phos-tag molecule, it represents a very convenient tool for clinical researchers and physicians to obtain useful SNP data from a small number of patients. We have reported a single nucleotide variation of G or A at the 87 position of a human cardiac sodium channel gene, SCN5A, in healthy individuals, which causes no change in the amino acid
Phosphate-Affinity Polyacrylamide Gel Electrophoresis
187
sequence (11). PCR was performed to amplify the 220-bp DNA fragment containing this SNP position using a 1:1 mixture of 50 OPO32–-labeled G-allele-specific primer and nonlabeled A-allelespecific primer, the reverse nonlabeled primer, and each genomic template from seven healthy individuals with the following three allele genotypes: the G/A heterozygote in two individuals, the G/ G homozygote in three individuals, and the A/A homozygote in two individuals. Subsequently, the resultant PCR products were analyzed using 20 mM Zn2+–Phos-tag PAGE. As a result, the 50 OPO32–-labeled G-allele product should be observed as a slower migration band and nonlabeled A-allele product as a faster one. The procedure for SNP genotyping using the 50 -OPO32–-labeled and nonlabeled PCR product is schematically summarized in Fig. 11.2. It is important for genotyping to obtain a significant difference in quantity between the allele-specific PCR products in the simultaneous amplification of all individuals. However, it is difficult to distinguish the kinetics of allele-specific amplification from the individual genomic templates of different quality, such as the concentration and/or the level of purification, under the fixed thermal cycle protocol (i.e., fixed annealing temperature and amplification cycle numbers). To avoid this problem, we performed five PCRs using 500, 250, 125, 63, and 32 pg of the genomic template in each individual genotyping. 3.1. Allele-Specific PCR for Genotyping
1. The G-allele-specific primer (50 mM) was phosphorylated with T4 polynucleotide kinase (5.0 U) in the T4 polynucleotide kinase reaction buffer described above (final volume, 20 mL) at 37C for 30 min (see Note 8). 2. To stop the kinase reaction, the mixture was incubated at 75C for 5 min. 3. PCR for amplification of the target region (220 bp) was performed with the PCR mixture described above (final volume, 5.0 mL) using a 1:1 mixture of 50 -OPO32–-labeled G-allele-specific primer (0.10 mM) and nonlabeled A-allelespecific primer (0.10 mM), the reverse nonlabeled primer (0.10 mM), and 500, 250, 125, 63, and 32 pg of each genomic template from seven healthy individuals. 4. After initial denaturation at 95C for 3 min, the PCR amplification was carried out for 30 cycles of 15-s denaturation at 95C and 30-s annealing/extension at 68C to obtain 50 OPO32–-labeled and/or nonlabeled DNA fragments (see Note 9). 5. Each PCR mixture (5.0 mL) was dissolved in 5.0 mL of the sample loading dye solution described above. 6. Each sample solution (2.0 mL) was applied to the polyacrylamide gel containing gel-bound Zn2+–Phos-tag.
188
Kinoshita, Kinoshita-Kikuta, and Koike
Fig. 11.2. Single nucleotide polymorphism (SNP) genotyping by using a two-step procedure of allele-specific PCR and Zn2+–Phos-tag PAGE. The allele-specific PCR is performed to amplify the target DNA fragment containing a SNP position using a 1:1 mixture of 50 -OPO32–-labeled G-allele-specific primer and nonlabeled A-allele-specific primer, and the reverse nonlabeled primer. The Zn2+–Phos-tag polyacrylamide gel electrophoresis (PAGE) is based on the principle that the binding of Zn2+–Phos-tag immobilized in the gel to the phosphate group at the 50 -end of the DNA fragment results in slower electrophoresis mobility of the 50 -OPO32–-labeled DNA fragment compared with the corresponding nonlabeled one. (Reproduced with permission from (10), copyright 2006, Elsevier Inc).
3.2. Zn2+–Phos-tag PAGE for Genotyping
1. This instruction assumes the use of an Atto model AE-6500 mini-slab gel system (Tokyo, Japan). This will be easily adaptable to other formats, including large-type gels. It is critical that the glass plates (1 mm thick, 9 cm wide, and 9 cm long) for making gels are scrubbed clean with a proper detergent after use and rinsed extensively with distilled water. 2. Set up the casting apparatus. 3. Eighteen percent (w/v) separating gel solution (10 mL) (see Note 10) containing 20 mM polyacrylamide-bound Zn2+– Phos-tag (see Note 11) is prepared by mixing 6.0 mL of
Phosphate-Affinity Polyacrylamide Gel Electrophoresis
189
30% (w/v) acrylamide/bisacrylamide solution, 2.5 mL of separating gel buffer (4 ), 40 mL of Phos-tag solution, 40 mL of zinc nitrate solution (2 equiv against Phos-tag), and 1.3 mL of distilled water. 4. Degas the mixed separating gel solution under vacuum for 15 min. 5. Add 20 mL of TEMED and 100 mL of ammonium persulfate solution to the degassed solution and mix gently without bubbling. 6. Transfer the separating gel solution between the glass plates, and then insert a sample-well comb. Allow the gel to polymerize for 1 h.
Genome (pg) 500 250 125 63 32
Genotype G/A
G-allele product (5'-OPO32–) A-allele product (5'-OH)
G/A
G/G
G/G
G/G
A/A
A/A
Fig. 11.3. Zn2+–Phos-tag PAGE of the allele-specific PCR products (220 bp) amplified from three SCN5A genotypes in seven healthy individuals. Each panel shows the result of genotyping in each individual. The allele genotype (G/A heterozygote, G/G homozygote, and A/A homozygote) of each individual is represented on the left side of each panel. The amounts of genomic template DNA used (500, 250, 125, 63, and 32 pg) are shown above each lane. The Zn2+–Phos-tag PAGE gels [20 mM polyacrylamide-bound Zn2+–Phos-tag and 18% (w/v) polyacrylamide] were subjected to ethidium bromide staining. (Reproduced with permission from (10), copyright 2006, Elsevier Inc).
190
Kinoshita, Kinoshita-Kikuta, and Koike
7. Assemble the electrophoresis apparatus and fill the electrode chambers with the electrophoresis running buffer (1) described above (see Note 12). 8. Gently remove the comb from the separating gel and apply the samples (2.0 mL) to the wells. 9. Attach the leads to the power supply (e.g., ATTO AE-8750 Power Station 1000XP). Run the gels under a constant-current condition (15 mA per gel) and at room temperature until the bromophenol blue reaches the bottom of the separating gel (for about 70 min). 10. After running is over, remove the gel from the apparatus and stain it with gel staining solution (described above) to visualize the DNA bands. The data are shown in Fig. 11.3.
4. Notes 1. PCR using allele-specific primers has been well established as the mutant-allele-specific amplification method for the detection of single-base mutation (13). 2. Allele-specific PCR primer is generally designed as follows: the SNP or mutation site is located at the second nucleotide base from the 30 -end in the primer sequence. 3. Since the unpolymerized acrylamide is a neurotoxin, care should be taken to avoid exposure. 4. Because the Phos-tag ligand is an oily product with high viscosity, it should be completely dissolved by intensive pipetting. 5. Phos-tag solution should be used within 3 months. 6. TEMED is stored in a desiccator at room temperature. Buy small bottles as the quality of TEMED may decline (gels will take longer to polymerize) after the bottle has been opened. 7. By adopting a discontinuous buffer system with Tris–HCl as a separating gel buffer and Tris–glycine as an electrophoresis running buffer, we can increase the detection sensitivity. 8. The 50 -OPO32–-labeled primer is commercially available as a custom primer. 9. Determine the best PCR condition for the target DNA amplification. 10. Determine the best electrophoresis condition of the percentage of acrylamide gel. The 18% (w/v) gel is appropriate for the DNA size range of 100–250 bp in the mini-slab gel system.
Phosphate-Affinity Polyacrylamide Gel Electrophoresis
191
11. Determine the best condition of the concentration of Zn2+– Phos-tag for sufficient separation between the 50 -OPO32–labeled and nonlabeled PCR products. Polyacrylamidebound Zn2+–Phos-tag at 20 mM concentration is appropriate for the DNA size range of 100–1,000 bp in the mini-slab gel system. 12. Take care not to produce bubbles at the bottom surface of the gel set. When bubbles are observed, they should be removed completely.
Acknowledgements We would like to express our sincere gratitude to the subjects who provided blood samples for this study. We also wish to thank the Research Center for Molecular Medicine and the Analysis Center of Life Science, Hiroshima University, Japan, for the use of their facilities. This work was supported by Grants-in-Aid for Scientific Research (B) (19390011) and (C) (19590040) from the Japan Society of the Promotion of Science (JSPS), a Grant-in-Aid for Young Scientists (B) (18790120) from the Ministry of Education, Culture, Sports, Science, and Technology (MEXT), a research grant for Promoting Technological Seeds from the Japan Science and Technology Agency (JST), and research grants from the Iketani Science and Technology Foundation and the Takeda Science Foundation. References 1. Kinoshita, E., Takahashi, M., Takeda, H., Shiro, M. and Koike, T. (2004) Recognition of phosphate monoester dianion by an alkoxide-bridged dinuclear zinc(II) complex. Dalton Trans. 8, 1189–1193. 2. Takeda, H., Kawasaki, A., Takahashi, M., Yamada, A. and Koike, T. (2003) Matrixassisted laser desorption/ionization timeof-flight mass spectrometry of phosphorylated compounds using a novel phosphate capture molecule. Rapid Commun. Mass Spectrom. 17, 2075–2081. 3. Kinoshita, E., Yamada, A., Takeda, H., Kinoshita-Kikuta, E. and Koike, T. (2005) Novel immobilized zinc(II) affinity chromatography for phosphopeptides and phosphorylated proteins. J. Sep. Sci. 28, 155–162. 4. Kinoshita-Kikuta, E., Kinoshita, E., Yamada, A., Endo, M. and Koike, T. (2006) Enrichment of phosphorylated
proteins from cell lysate using a novel phosphate-affinity chromatography at physiological pH. Proteomics 6, 5088–5095. 5. Inamori, K., Kyo, M., Nishiya, Y., Inoue, Y., Sonoda, T., Kinoshita, E., Koike, T. and Katayama, Y. (2005) Detection and quantification of on-chip phosphorylated peptides by surface plasmon resonance imaging techniques using a phosphate capture molecule. Anal. Chem. 77, 3979–3985. 6. Kinoshita, E., Kinoshita-Kikuta, E., Takiyama, K. and Koike, T. (2006) Phosphate-binding tag, a new tool to visualize phosphorylated proteins. Mol. Cell. Proteomics 5, 749–757. 7. Kinoshita-Kikuta, E., Aoki, Y., Kinoshita, E. and Koike T. (2007) Label-free kinase profiling using phosphate affinity polyacrylamide gel electrophoresis. Mol. Cell. Proteomics 6, 356–366.
192
Kinoshita, Kinoshita-Kikuta, and Koike
8. Yamada, S., Nakamura, H., Kinoshita, E., Kinoshita-Kikuta, E., Koike, T. and Shiro, Y. (2007) Separation of a phosphorylated histidine protein using phosphate affinity polyacrylamide gel electrophoresis. Anal. Biochem. 360, 160–162. 9. Kinoshita, E., Kinoshita-Kikuta, E., Matsubara, M., Yamada, S., Nakamura, H., Shiro, Y., Aoki, Y., Okita, K. and Koike, T. (2008) Separation of phosphoprotein isotypes having the same number of phosphate groups using phosphate-affinity SDS-PAGE. Proteomics 8, 2994–3003. 10. Kinoshita, E., Kinoshita-Kikuta, E. and Koike, T. (2007) A single nucleotide polymorphism genotyping method using phosphate-affinity polyacrylamide gel
electrophoresis. Anal. Biochem. 361, 294–298. 11. Kinoshita, E., Kinoshita-Kikuta, E., Kojima, H., Nakano, Y., Chayama, K. and Koike, T. (2005) Reliable and cost-effective screening of inherited heterozygosity by Zn2+–cyclen polyacrylamide gel electrophoresis. Clin. Chem. 51, 2195–2197. 12. Sambrook, J., Fritsch, E. F. and Maniatis, T. (1989) Molecular cloning 3: a laboratory manual, 2nd edition, Cold Spring Harbor Laboratory Press, New York, pp. E.5–E.6. 13. Takeda, S., Ichii, S. and Nakamura, Y. (1993) Detection of K-ras mutation in sputum by mutant-allele-specific amplification (MASA). Hum. Mutat. 2, 112–117.
Chapter 12 Estimation of SNP Allele Frequencies by SSCP Analysis of Pooled DNA Tomoko Tahira, Yoji Kukita, Koichiro Higasa, Yuko Okazaki, Aki Yoshinaga, and Kenshi Hayashi Abstract The single strand conformation polymorphism (SSCP) method is a sensitive technique used to detect subtle sequence differences in PCR-amplified DNA fragments as separated peaks in electrophoretic analysis. In this chapter, we focus on SSCP analysis for quantifying polymorphic alleles rather than scanning for mutations. Short fragments carrying single nucleotide polymorphisms are amplified from individual and pooled DNA samples, then the products are labeled with fluorescent dyes and analyzed by automated capillary electrophoresis under nondenaturing conditions. Dedicated software, QSNPlite, interprets trace data of the electrophoresis to identify alleles of individuals and quantify these alleles in the pool. The software can also incorporate sequencing data to assign alleles at the nucleotide level. The procedures described here are being used in association studies that compare allele frequencies between cases and controls to identify genes responsible for common diseases. Key words: Allele frequency, capillary electrophoresis, pooled DNA, single nucleotide polymorphism, single strand conformation polymorphism.
1. Introduction The International HapMap Project has succeeded in constructing haplotype maps of extremely high resolution for major populations. The information obtained from these maps is expected to be useful in large-scale association studies to locate genes responsible for traits such as disease susceptibility. Association studies require an accurate determination of the allele frequencies of many single nucleotide polymorphisms (SNPs) in both case and control groups, each consisting of several hundred individuals or more. This means that the cost and labor for association studies A.A. Komar (ed.), Single Nucleotide Polymorphisms, Methods in Molecular Biology 578, DOI 10.1007/978-1-60327-411-1_12, ª Humana Press, a part of Springer Science+Business Media, LLC 2003, 2009
193
194
Tahira et al.
are considerable, even when recently developed low-cost genotyping techniques are used, if the allele frequencies are determined by individual genotyping and allele counting. Clearly, relative quantification of alleles in pooled DNA has obvious advantages over allele counting, in terms of both cost- and labor-efficiency (1). Single strand conformation polymorphism (SSCP) analysis is a sensitive method for detecting sequence differences of both SNPs and insertions/deletions (indels) in PCR-amplified DNA fragments of one to several hundred nucleotides (2, 3). We have further developed a nonradioactive method, called ‘‘postlabeling method with separation by automated capillary electrophoresis under SSCP conditions’’ (PLACE-SSCP), in which the PCR products are postlabeled with fluorescent dyes and analyzed by automated capillary electrophoresis under nondenaturing conditions (4, 5). The advantage of employing the postlabeling strategy over amplification using labeled primers is that it yields reproducible and quantitative peaks of separated alleles. Spurious peaks derived from the fragments carrying template-independent nucleotides attached to the 3’-ends of the PCR products during the amplification reaction are avoided in this technique because these extra nucleotides are removed during the postlabeling reaction. PLACE-SSCP analysis has been demonstrated to be suitable for precise estimation of SNP-allele frequencies when applied to pooled DNA (6). The principle of allele frequency estimation by SSCP analysis is depicted in Fig. 12.1. The method has been further improved to increase the throughput of the analysis using a capillary array system (7–10). It is therefore suitable for studies that require allele frequency analyses of many SNPs, such as in the secondary screening after the first genome-wide screen using the array-hybridization method.
allele A
Hetero HA
Pool
PA
allele B HB
PB
RH = HA/HB ….. Peak height ra tio of heterozygote RP = PA/PB ….. Peak height ratio of pool FA: Frequency of allele A in pool FB: Frequency of allele B in pool
FA = (PA/HA)/[(PA/HA)+(PB/HB)] = RP/(RH + RP) FB = 1 – FA = RH/(RH + RP)
Fig. 12.1. Principle of allele frequency estimation by single strand conformation polymorphism (SSCP). Electropherograms from postlabeling method with separation by automated capillary electrophoresis under SSCP conditions analyses of a heterozygous individual and a pooled sample are shown on the left. Allele frequencies are calculated by the equation indicated, where the peak height ratio of heterozygotes (RH) is used as a correction factor for the peak height ratio of pools (Rp) as described in the text.
Estimation of SNP Allele Frequencies by SSCP Analysis of Pooled DNA
195
We present here a suite of detailed protocols for bench works in pooled DNA-SSCP analysis and an introduction to our newly developed data management/interpretation software, QSNPlite, which facilitates systematic PLACE-SSCP analysis of individual and pooled DNA (11).
2. Materials 2.1. Equipment
1. 96-well PCR plates and sealing sheets. 2. Thermal cycler: e.g., ABI 9700 (Applied Biosystems), T1 thermocycler (Biometra). 3. Automated capillary electrophoresis system: ABI Prism 3100 genetic analyzer data collection software (version 1.01 or later). GeneScan or GeneMapper is helpful for quick viewing of the electropherogram to check the runs. 4. 3100/3130xl 36-cm capillary array (Applied Biosystems), 47 cm 50 mm.
2.2. Reagents
1. KOD -Plus- DNA polymerase, 200 U, 1 U/mL and 10x buffer (Toyobo). 2. Thermo Sequenase (GE Healthcare), 1,000 U, 32 U/mL. 3. Klenow fragment of DNA polymerase I (NEB), 1,000 U, 5 U/mL. 4. AmpliTaq and 10x buffer II/MgCl2 (Applied Biosystems), 250 U, 5 U/mL. 5. TaqStart antibody (Clontech), 200 mL, 1.1 mg/mL. 6. dATP, dCTP, dGTP, and dTTP (Applied Biosystems), 1 mL, 10 mM, GeneAmp1 dNTP blend. 7. R6G-ddCTP (PerkinElmer), 2.5 nmol, 0.1 mM. 8. R110-ddUTP (PerkinElmer), 2.5 nmol, 0.1 mM. 9. Long-chain linear poly(N,N-dimethylacrylamide) (LLPD MAA): see Section 3.7. Applied Biosystems supplies conformation analysis polymer. This polymer may be usable, but we do not have any experience with it. 10. 20x TME buffer (pH 6.8): 0.6 M Trizma base (Sigma), 0.7 M 2-morpholinoethanesulfonic acid (Sigma), and 20 mM Na2EDTA (Dojindo). 11. MultiScreen-HV plate (Millipore), 50/Pk.12. Sephadex G-50 Fine (GE Healthcare), 100 g. 12. 1x TE buffer: 10 mM tris(hydroxymethyl)aminomethane (Tris)–HCl, pH 7.5, 1 mM EDTA.
196
Tahira et al.
13. 1x Te buffer: 10 mM Tris–HCl, pH 7.5, 0.1 mM EDTA. 14. MapMarker1 1,000 TMR (BioVentures, Murfreesboro, TN, USA). 15. Microcon YM-30 filters (Millipore) and 0.22-mm Milex-GP filters (Millipore). 16. N,N-Dimethylacrylamide (Aldrich Chemical, Milwaukee, WI, USA). 17. Ammonium persulfate (Bio-Rad, Hercules, CA, USA). Prepare a 10% solution in distilled H2O. Use fresh. 18. 2-Propanol (analytical-reagent grade, e.g., Wako Pure Chemical Industries). 19. N,N,N 0 ,N 0 -Tetramethylethylenediamine (TEMED) (Nacalai Tesque, Kyoto, Japan). 20. Dialysis tubing: Spectra/Pore, molecular weight cutoff 12,000– 14,000 (Spectrum Laboratories, Rancho Dominguez, CA, USA). 21. 0.5 M EDTA (pH 8.0). 2.3. DNA
1. Individual DNA: DNAs are extracted from peripheral blood leukocytes via standard procedures, such as using the QIAamp DNA blood midi kit (Qiagen). They are dissolved at high concentrations, such as 50 ng/mL, in TE buffer and are stored at 4C. They are further diluted with Te buffer to the concentrations indicated (see below). 2. Pooled DNA: This must be a strictly equal part mixture of individual DNAs. The concentrations of individual DNAs are adjusted in multiple steps, involving dilution with TE buffer followed by concentration determination using a spectrofluorometer, such as a SPECTRAmax1 GEMINI XS (Molecular Devices Corporation) after staining with PicoGreen doublestranded DNA quantification reagent (Molecular Probes). These procedures include dispensing liquid at volumes that differ from well to well, or repeated transferring of a fixed volume of liquid from plate to plate. Therefore, the use of a robotic liquid handling system such as a Tecan Genesis RSP150 (Tecan Group, Mannedorf, Switzerland) and a 96channel microdispenser such as a Thermo Scientific Matrix Hydra II (Thermo Scientific, Hudson, New Hampshire) is recommended. The final concentration of each sample should be within 10% of the average value. Pools are made by combining an equal volume of each sample, and then the concentration is adjusted to 8.3 ng/mL in Te buffer.
2.4. Primers
The amplicon length suitable for PLACE-SSCP is less than 200 bp. Primer sequences [L- and R-primers, corresponding to left (forward) and right (reverse) primers] are designed using
Estimation of SNP Allele Frequencies by SSCP Analysis of Pooled DNA
197
various Web-site services such as PRIMER 3 (http://frodo.wi. mit.edu/), and Inazuka tags (ATT and GTT at the 50 -ends of L- and R-primers, respectively) are attached. These tags are required for fluorescent postlabeling of PCR products (see Section 3.3). Primers purchased in desalted form are dissolved in TE buffer at high concentrations, such as 50 mM, and stored frozen at –20C. They are diluted with Te buffer, and the primer pairs (L- and R-primers for each amplicon) are combined to make the primer mix, which has a final concentration of 0.833 mM for each of the two primers. 2.5. Software
We have developed QSNPlite (11), which runs on Windows XP, and facilitates systematic PLACE-SSCP analysis of individual and pooled DNAs. Sequencing analysis modules can be incorporated (optional) so that SNP/indel alleles of individuals detected by SSCP are unambiguously assigned at the nucleotide level, and their frequencies in the pools are determined. The major part of QSNPlite is written in Visual Basic 6.0 (Microsoft). Some parts of the SSCP analysis module are in C on Windows. The sequencing analysis module requires installation of a Linux emulator and external software, Phred/Phrap and PolyPhred (12, 13), which must be obtained directly from the original source by users. QSNPlite can be run without the sequencing analysis module. A detailed description of the installation and operation of QSNPlite is given in the manuals that are supplied with the software program. QSNPlite is available to researchers for nonprofit use from our Web site (http:// qsnp.gen.kyushu-u.ac.jp/placeSSCP/).
3. Methods 3.1. Plate Design
In many cases, it is necessary to examine many SNPs in many DNAs (of individuals or pools), and tracking down the identity of SNPs and DNAs for each trace data of electrophoresis can be a laborious and error-prone task. One solution to this problem is to include the combination of SNP name and DNA name in the sample name (well name) for electrophoresis, which is then included in the file name of the trace data via an input file in the operation of the capillary array machine. In the case of ABI 3100, this input file is called a ‘‘plate file.’’ Although this can be done manually, using the computer program to assign SNPs and DNAs to the wells and to make the plate file can avoid human error at this stage. QSNPlite (described in Section 3.6) has the function ‘‘plate design,’’ which supports the allocation of primers and DNA samples to the well positions of a 96-well plate, and makes a plate file.
198
Tahira et al.
The program also makes a ‘‘sample sheet’’ file, in which the combination of SNP names and DNA names assigned to the wells is described. The operator can print out the sample sheet in an 8 12 table, and carry out bench work following the information described in the table. 3.2. Overall Scheme of PLACE-SSCP Analysis
1. PCR-amplify the target sequences in 96-well plates (see Sections 3.1 and 3.3). 2. Take an aliquot of diluted PCR products, postlabel them, and desalt them by gel filtration (see Section 3.4). 3. Take an aliquot of the filtrate, dispense it into a plate, and add low-salt loading buffer, heat-denature the products and cool them to room temperature (see Section 3.5). 4. Add desalted and diluted TAMRA-labeled double-stranded DNA markers (dsN markers) (see Section 3.7). 5. Start electrophoresis through LLPDMAA in low-pH medium (see Sections 3.5 and 3.8). 6. Interpret the output traces using QSNPlite (see Section 3.6).
3.3. Polymerase Chain Reaction
1. Dispense 4.0 mL of PCR premix solution (described in Table 12.1) into each well. 2. Add 3.0 mL of template DNA (8.3 ng/mL) according to the plate design (see Section 3.2) to make a final amount of 25 ng DNA per well. 3. Add 3.0 mL of primer mix (0.833 mM for each of the L- and R-primers) according to the plate design (see Section 3.2). The final concentration is 0.25 mM for each primer.
Table 12.1 Composition of PCR premix solution
a
Volume per well (mL)
Volume per plate (mL)
Final concentration
4dNTPa
1.6
160
0.2 mM each
10 PCR bufferb
1.0
100
1x
25 mM MgCl2
0.8
80
2 mM
Taq/anti-Taqc
0.1
10
0.25 U per well as Taq
Dimethyl sulfoxide
0.5
50
5%
PCR premix solution
4.0
400
4dNTP is a mixture of the four dNTPs, each at 1.25 mM in H2O. 10x PCR buffer is 0.5 M KCl and 0.1 M tris(hydroxymethyl)aminomethane (Tris)–HCl, pH 8.3. c Taq/anti-Taq is an equal volume mixture of AmpliTaq at 5 U/mL and TaqStart antibody. b
Estimation of SNP Allele Frequencies by SSCP Analysis of Pooled DNA
199
4. Amplify the target sequences by PCR. The thermal profile is 1 min at 94C, followed by 40 cycles of 30 s at 94C, 0.5 min at 60C, and 1 min at 72C. 5. Add 10 mL per well of H2O to dilute the PCR products twofold. 3.4. Two-Color Postlabeling of PCR Products with Fluorescent ddNTPs
The two strands of PCR products are labeled with rhodamine 6G (R6G) and rhodamine 110 (R110), respectively, by the exchange reaction of DNA polymerases. 1. Transfer 2.0 mL per well of diluted PCR product described in Section 3.3 to a new 96-well plate. 2. Add 4.0 mL per well of postlabeling premix solution (described in Table 12.2).
Table 12.2 Composition of postlabeling premix solution Volume per well (mL)
Volume per plate (mL)
H2O
2.71
271
0.1 M MgCl2
0.6
60
10 mM
0.1 M Tris–HCl, pH 8.3
0.6
60
10 mM
100 mM R6G-ddCTP
0.012
1.2
0.2 mM
100 mM R110-ddUTP
0.012
1.2
0.2 mM
0.06
6
0.3 U per well
Thermo Sequenaseb
0.0094
0.94
0.3 U per well
Postlabeling premix solution
4.0
Klenow
a
a
Final concentration
400
Klenow fragment of DNA polymerase I at 5 U/mL. Thermo Sequenase DNA polymerase at 32 U/mL.
b
3. Incubate the mixture for 5 min at 37C, 15 min at 57C, and 10 min at 75C. 4. Add 18 mL per well of 0.5 mM EDTA. 5. Desalt and remove primers/nucleotides by gel filtration through Sephadex G50 equilibrated with 0.5 mM EDTA in the MultiScreen plate, following the instructions for the MultiScreen as provided by Millipore. Incomplete gel filtration may result in the appearance of peaks of unincorporated fluorescent nucleotides in the electropherogram. Do not let the solution run down the wall of the well, but rather make sure it gently lands directly at the center of the gel-bed surface.
200
Tahira et al.
3.5. SSCP Electrophoresis
1. Transfer 2 mL per well of the gel-filtered products described in Section 3.3 to a loading plate. 2. Add 18 mL per well of 0.5 mM EDTA (Table 12.3).
Table 12.3 Composition of the sample solution for single strand conformation polymorphism (SSCP) electrophoresis (ABI 3100) Volume per well (mL) Sample described in Section 3.4
Volume per plate (mL)
2.0
0.5 mM EDTA
18.0
Sample solution
20.0
1800
3. Heat at 90C for 3 min, then cool to 20C. 4. Add 10 mL per well of TAMRA marker, which is either the dsN marker as described in Section 3.6, or 0.5 mL per well of MapMarker1 1,000 TMR. 5. Run capillary array electrophoresis under the following conditions (see Notes 1 and 2). l Sieving matrix: 10% LLPDMAA in 2x TME buffer.
3.6. Data Analysis Using QSNPlite
l
Temperature for equilibration, fill and run: 27C.
l
Capillary fill (36 cm): 100 s.
l
Sample injection: 2 kV for 10 s.
l
Separation: 15 kV for 20 min, data delay 1 s.
Step-by-step instructions for the operation of QSNPlite are given in the manuals supplied with the software program (see Note 3). Figure 12.2 illustrates the flow of operations in QSNPlite, which is functionally divided into three parts: ‘‘set-up,’’ ‘‘analysis,’’ and ‘‘report.’’ The entry points to QSNPlite are indicated by bold rectangles. All entry points except for the one at the top of ‘‘analysis’’ accept input files that are in text format (shadowed in Fig. 12.2). Thus, the upstream operations can be replaced with operations other than the ones indicated as long as the input files contain the necessary information in the appropriate text format. For example, alternative sequencing interpretation software can be used instead of Phred, Phrap, and Polyphred if the output files are converted to a compatible format. Examples of input files are supplied with the program.
Estimation of SNP Allele Frequencies by SSCP Analysis of Pooled DNA DNA list (pools and individuals)
Primer - pair list
Plate design
201
Define samples (primer DNA combinaƟons) for PCR or seq Make plate file for electrophoresis and sample sheet for bench work Plate file
Sample sheet Carry out bench work
Run SSCP/seq electrophoresis Select type of analysis and import/convert trace files to text
Analysis SSCP
Trace data or or seq) Trace data data(SSCP (SSCP seq) Trace data (SSCP or seq) seq) Trace (SSCP or
Detect peaks and align by QUISCA
Base-call, align and trim by P/P/P/C
Resolve fused peaks by wave deconvoluƟon
SSCP.qsl
Resolve indel-heteros by half-homo subtracƟon
Genotype/quanƟfy SSCP-alleles
seq.qsl
Genotype SNP/indel-alleles
SSCP Report
Report
Sequencing
Seq Report
Confirm consistency between SSCP alleles and SNP/indel alleles Assign SNP/indel alleles to individuals and allele frequencies to pools QSNPlite Report
Fig. 12.2. Flow chart of QSNPlite. P/P/P/C stands for Phred, Phrap, Polyphred, and ClustalW. Entry points to QSNPlite are indicated by bold rectangles. Files in text are shadowed. See Section 3.6 for further explanations.
Plate design (see Section 3.2) is made in ‘‘set-up,’’ where a sample name (i.e., a combination of names of the primer pair and DNA) is assigned to each well. The plate design is exported in two forms, ‘‘plate file’’ and ‘‘sample sheet,’’ both in text format. ‘‘Plate file’’ is used as an input file for the operation of the capillary array electrophoresis machine (currently supports only ABI 3100). A ‘‘sample sheet’’ is printed out for use in the bench work, such as PCR. The output trace files from the capillary array electrophoresis machine (*.fsa files for SSCP and *.ab1 files for sequencing in the case of ABI 3100) carry sample names in their file names; thus, the identities of samples and traces are maintained. The trace files are converted to text and further processed by the analysis modules ‘‘SSCP’’ or ‘‘sequencing’’ in the ‘‘analysis’’ section. The ‘‘SSCP’’ analysis module produces a text file, ‘‘SSCP report,’’ in which the genotype of each individual is assigned with SSCP alleles, and the allele frequency of each pool is determined. The ‘‘sequencing’’ analysis module produces a text file, ‘‘Seq report,’’ in which the genotype of each individual is assigned at nucleotide sequence level.
202
Tahira et al.
The operator can stop the analysis at any step and exit QSNPlite. The data at the point of exit are saved as a freeze file that has a .qsl extension. Analysis can then be resumed from the step at the exit by double-clicking the freeze file. The individual genotypes obtained by SSCP and sequencing are compared and consistency is checked in the ‘‘Report’’ section. Then, the genotypes of individuals at the nucleotide level and allele frequencies of pools are exported to the ‘‘QSNPlite report’’ file, which can be opened by spreadsheet software programs such as Excel. Double-stranded PCR fragments serve as internal markers (14), added after heat denaturation of the samples before subjecting them to electrophoresis (see Section 3.4). They provide reference positions during lane-to-lane alignment of peaks in trace data interpretation (see Section 3.6). The procedures described here are for the preparation of 30 PCR products (dsN markers) ranging from 40 to 600 bp in length from a linearized plasmid DNA template (pBluscript SK+/BamHI digest). Primer sequences and product sizes are given in Table 12.4. Amplifications are in 10 mL for each marker, except for 40- and 55-bp markers, which are amplified in 30 mL because of relatively low yield.
3.7. Preparation of Markers
Table 12.4 Primers for marker preparation Name
Product size
Sequence 50 ->30
Primer length
Tam-start*
17**
TAMRA-ATCTCAGCGATCTGTCT
17
A1
40
AGGCAACTATGGATGAAC
18
A2-2
55
ACACGACGGGGAGTC
15
A4-2
95
CTGGGGCCAGATGGT
15
A5-2
115
CTCGCGGTATCATTGC
16
A5b
125
GAGCGTGGGTCTCGC
15
A6
140
AAATCTGGAGCCGGTG
16
A7
160
CTGGCTGGTTTATTGCT
17
A8
180
TCTGCGCTCGGCCC
14
A9
200
GATAAAGTTGCAGGACC
17
B0
225
ACAATTAATAGACTGGATG
19
B1
250
AACTACTTACTCTAGCTTC
19 (continued)
Estimation of SNP Allele Frequencies by SSCP Analysis of Pooled DNA
203
Table 12.4 (continued) Name
Product size
Sequence 50 ->30
Primer length
B2
275
ACGTTGCGCAAACTATTA
18
B3
300
CACGATGCCTGTAGCA
16
B4
325
CCATACCAAACGACGAG
17
B4b
340
CGGAGCTGAATGAAGC
16
B5
350
CGTTGGGAACCGGAG
15
B6
375
GGGGGATCATGTAACTC
17
B7
400
AGCTAACCGCTTTTTTGC
18
B8
425
CTGACAACGATCGGAG
16
B9
450
TGATAACACTGCGGCC
16
B10-2
470
AGTGCTGCCATAACCAT
17
B11-2
495
TGGCATGACAGTAAGAG
17
B11
500
ACGGATGGCATGACAGT
17
B11b
510
AAAGCATCTTACGGATGG
18
B11c
530
GAGTACTCACCAGTCAC
17
B11d
550
ATTCTCAGAATGACTTGG
18
B13
570
ACTCGGTCGCCGCAT
15
B14
580
GGCAAGAGCAACTCGG
16
B15
590
ATTGACGCCGGGCAAG
16
B16
600
ATTATCCCGTATTGACGC
18
*
Tam-start (see Table 12.5) is used in combination with other primers to amplify the products with indicated sizes. **Size of the primer.
1. Distribute 9.5 mL of marker-PCR premix solution (described in Table 12.5) to 34 appropriately spaced wells of a 96-well plate. 2. Add 0.5 mL per well of 5.0 mM primers (Table 12.4, A1–B16) so that each marker is amplified in one well, except for the 40bp marker (A1 in Table 12.4) and the 55-bp marker (A2-2 in Table 12.4), each of which is amplified in three wells. 3. Amplify each target sequence by PCR. The thermal profile is 1 min at 94C, 40 cycles of 15 s at 94C, 15 s at 55C, and 30 s at 68C, followed by 5 min at 68C. 4. Combine all PCR products (340 mL), desalt them, and remove primers/nucleotides by filtration through Microcon YM-30 filters.
204
Tahira et al.
Table 12.5 Composition of premix solution for PCR to prepare marker fragments Volume per well (mL)
Volume per 35 wells (mL)
Distilled H2O
6.58
230.3
4 dNTPs each at 2 mM
1.0
35
0.2 mM each
10 KOD- Plus- buffera
1.0
35
1x
MgSO4a
0.6
21
1.5 mM
25 mM
25 mM Tam-startb
Final concentration
0.1
3.5
0.25 mM
0.02
0.7
0.012 ng/mL
KOD -Plus-a
0.2
7
0.02 U/mL
Marker-PCR premix solution
9.5
332.5
6.0 ng/mL template
c
a
Supplied in KOD -Plus- PCR kit. PCR primer labeled with TAMRA at its 5’-end. See Table 12.4 for the sequences. c pBluscript SK+/BamHI digest. b
5. Dilute purified PCR products with 0.5 mM EDTA to make a final volume of 1,100 mL. This is a 30x dsN marker. Store at 4 C. 6. Dilute the 30x dsN marker with 30 vol of 0.5 mM EDTA to make 1x dsN marker before use. 3.8. Long-Chain Linear Poly(N,Ndimethylacrylamide)
LLPDMAA is used as the sieving matrix in SSCP electrophoresis. This matrix is currently not available commercially, but can be supplied by us on request. Alternatively, the polymer can be synthesized using the following procedure. LLPDMAA is synthesized essentially following the method of Ren et al. (15), as described by Kukita et al. (9). In brief, 24 mL of N,N-dimethylacrylamide, 5.7 mL of 2-propanol, and 222 mL of H2O are mixed, degassed by bubbling with argon gas, and heated at 50C with agitation. Polymerization is started by adding 1.25 mL of 10% ammonium persulfate, and then 1.25 mL of 10% TEMED. The mixture is incubated for 1.5 h at 50C with constant agitation under argon gas. The product is extensively dialyzed (Spectra/Pore, molecular weight cutoff 12,000–14,000), lyophilized, and stored at –20C. The weight-average molecular mass of the product (LLPDMAA) is approximately 863,000, as determined by gel permeation chromatography. To prepare the sieving polymer, dissolve 2 g of LLPDMAA in 15 mL of H2O and 2 mL of 20x TME buffer with slow agitation at 37C for several hours, adjust the volume to 20 mL with H2O,
Estimation of SNP Allele Frequencies by SSCP Analysis of Pooled DNA
205
filter (pore size 0.22 mm, Millex-GP) the solution, and degas it under a vacuum for 5 min before use. The solution should be stored shielded from light.
4. Notes 1. PLACE-SSCP analysis requires a proper color matrix that matches the dye set and electrophoretic conditions (media and temperature) employed, to deconvolute spectral overlaps of fluorescent dyes. The matrix also depends on the optoelectrical characteristics of the individual machine and how it has been tuned. Follow the instructions in the section ‘‘Spatial and spectral calibration’’ of the manual of the ABI Prism 3100 genetic analyzer to make the matrix. Four DNA fragments between 100 and 150 bp in size, each labeled with a different dye (R110, R6G, TAMRA, and ROX) and yielding wellseparated peaks, are required for the calibration. 2. A new run module must be created for the SSCP. Start the 3100 data collection software program version 1.1, go to ‘‘Tools,’’ ‘‘Module Editor’’ and use the module of GeneScan as a template. Change the parameters as described in Section 3.5. Also edit ‘‘Run Current’’ in ‘‘Module Parameter’’ from ‘‘100 mAmps’’ to ‘‘200 mAmps’’ to allow a higher upper limit of the current during the electrophoresis, because of the high conductivity of the buffer system employed. 3. The peak heights of two alleles of a heterozygote in SSCP analysis may not be equal, although the two alleles are always expected to be present at the same amount in the heterozygous DNA (RH of the same amplicon can deviate from 1 and can be variable, see Fig. 12.1). This deviation comes from PCR bias (different efficiency of amplification between alleles), and must be considered in the estimation of allele frequency in pooled DNA samples as shown in Fig. 12.1. We empirically found that the extent of PCR bias can significantly differ if the PCR is carried out in different batches (or in different plates). For this reason, the heterozygous sample to be used in the calculation of RH must be amplified together with the pooled DNA samples in the same plate for reliable allele frequency estimation. A small variation from PCR bias still exists even within the same plate. Therefore, for each amplicon, we recommend devoting at least three wells to the heterozygote sample and three wells to each of the pooled DNA samples, and to use mean values of RH and RP to accurately estimate allele frequencies.
206
Tahira et al.
We also found that the estimation of allele frequency becomes inaccurate for amplicons with large PCR bias. This may imply the presence of hidden SNPs within or near the positions of primers and may seriously influence the efficiency of the amplification reaction. As a rule of thumb, one should redesign the PCR primers when extreme RH values, such as less than 0.25 or greater than 4, are observed.
Acknowledgments This work was supported by KAKENHI (Grant-in-Aid for Scientific Research) on Priority Areas ‘‘Applied Genomics’’ and KAKENHI 18310131 from the Ministry of Education, Culture, Sports, Science and Technology of Japan. References 1. Sham, P., Bader, J. S., Craig I., O’Donovan, M. and Owen, M. (2002) DNA pooling: A tool for large-scale association studies. Nat. Rev. Genet. 3, 862–871. 2. Orita, M., Suzuki, Y., Sekiya, T. and Hayashi, K. (1989) Rapid and sensitive detection of point mutations and DNA polymorphisms using the polymerase chain reaction. Genomics 5, 874–879. 3. Kukita, Y., Tahira, T., Sommer, S. S. and Hayashi, K. (1997) SSCP analysis of long DNA fragments in low pH gel. Hum. Mutat. 10, 400–407. 4. Inazuka, M., Tahira, T. and Hayashi, K. (1996) One-tube post-PCR fluorescent labeling of DNA fragments. Genome Res. 6, 551–557. 5. Inazuka, M., Wenz, H. M., Sakabe, M., Tahira, T. and Hayashi, K. (1997) A streamlined mutation detection system: Multicolor post-PCR fluorescence labeling and singlestrand conformational polymorphism analysis by capillary electrophoresis. Genome Res. 7, 1094–1103. 6. Sasaki, T., Tahira, T., Suzuki, A., Higasa, K., Kukita, Y., Baba, S. and Hayashi, K. (2001) Precise estimation of allele frequencies of single-nucleotide polymorphisms by a quantitative SSCP analysis of pooled DNA. Am. J. Hum. Genet. 68, 214–218. 7. Higasa, K., Kukita, Y., Baba, S. and Hayashi K. (2002) Software for machine-independent quantitative interpretation of SSCP in capillary
8.
9.
10.
11.
12.
13.
array electrophoresis (QUISCA). BioTechniques 33, 1342–1348. Kukita, Y. and Hayashi, K. (2002) Multicolor post-PCR labeling of DNA fragments with fluorescent ddNTPs. BioTechniques 33, 502, 504, 506. Kukita, Y., Higasa, K., Baba, S., Nakamura, M., Manago, S., Suzuki, A., Tahira, T. and Hayashi K. (2002) A single-strand conformation polymorphism method for the largescale analysis of mutations/polymorphisms using capillary array electrophoresis. Electrophoresis 23, 2259–2266. Baba, S., Kukita, Y., Higasa, K., Tahira, T. and Hayashi, K. (2003) Single-stranded conformational polymorphism analysis using automated capillary array electrophoresis apparatuses. BioTechniques 34, 746–750. Tahira, T., Okazaki, Y., Miura, K., Yoshinaga, A.,Masumoto, K., Higasa, K., Kukita, Y. and Hayashi, K. (2006) QSNPlite, a software system for quantitative analysis of SNPs based on capillary array SSCP analysis. Electrophoresis 27, 3869–3878. Ewing, B., Hillier, L., Wendl, M. C. and Green, P. (1998) Base-calling of automated sequencer traces using phred. I. accuracy assessment. Genome Res. 8, 175–185. Nickerson, D. A., Tobe V. O. and Taylor S. L. (1997) PolyPhred: Automating the detection and genotyping of single nucleotide substitutions using fluorescence-based resequencing. Nucleic Acids Res. 25, 2745–2751.
Estimation of SNP Allele Frequencies by SSCP Analysis of Pooled DNA 14. Kuhn, D. N., Borrone, J., Meerow, A. W., Motamayor, J. C., Brown, J. S. and Schnell, R. J. (2005) Single-strand conformation polymorphism analysis of candidate genes for reliable identification of alleles by capillary array electrophoresis. Electrophoresis 26, 112–125.
207
15. Ren, J., Ulvik, A., Refsum, H. and Ueland, P. M. (1999) Applications of short-chain polydimethylacrylamide as sieving medium for the electrophoretic separation of DNA fragments and mutation analysis in uncoated capillaries. Anal. Biochem. 276, 188–194.
Chapter 13 Phenylethynylpyrene Excimer Forming Hybridization Probes for Fluorescence SNP Detection Igor A. Prokhorenko, Irina V. Astakhova, Kuvat T. Momynaliev, Timofei S. Zatsepin, and Vladimir A. Korshun Abstract Excimer formation is a unique feature of some fluorescent dyes (e.g., pyrene) which can be used for probing the proximity of biomolecules. Pyrene excimer fluorescence has previously been used for homogeneous detection of single nucleotide polymorphism (SNP) on DNA. 1-Phenylethynylpyrene (1-1PEPy), a photostable pyrene derivative with redshifted fluorescence, is able to form excimers (emission maximum about 500–510 nm) and is well suitable for nucleic acid labeling. We have shown the utility of 11-PEPy in the excimer-forming DNA probes for detection of 2144A/G and 2143A/G transitions, and 2143A/C substitution in the 23S ribosomal RNA gene of Helicobacter pylori strains resistant to clarithromycin. The phenylethynylpyrene pair can be generated either from 1-1-PEPy pseudonucleoside 4-[4(pyren-1-ylethynyl)phenyl]-1,3-butanediol or from 20 -O-(1-PEPy) modified nucleosides – 20 -O-[3(pyren-1-ylethynyl)benzyl]uridine and 20 -O-[4-(pyren-1-ylethynyl)benzyl]uridine. Key words: DNA probes, single nucleotide polymorphism detection, excimer fluorescence, pyrene, 1-phenylethynylpyrene, Helicobacter pylori.
1. Introduction The development of homogeneous fluorescent PCR-based single nucleotide polymorphism (SNP) analysis techniques is a rapidly growing area (1, 2). Fluorescent tetracyclic aromatic hydrocarbon pyrene and its derivatives have a long excited state lifetime, so their emission is sensitive to the microenvironment. The excimer emission from two pyrenes placed in close proximity (lmax around 470 nm) can be easily distinguished from the emission of a single pyrene label (lmax 370, 400 nm) (3). Pyrene excimer fluorescence was used in various formats of homogeneous SNP detection A.A. Komar (ed.), Single Nucleotide Polymorphisms, Methods in Molecular Biology 578, DOI 10.1007/978-1-60327-411-1_13, ª Humana Press, a part of Springer Science+Business Media, LLC 2003, 2009
209
210
Prokhorenko et al.
Fig. 13.1. Chemical structures of 1-phenylethynylpyrene (1-1-PEPy), 4-[4-(pyren-1ylethynyl)phenyl]-1,3-butanediol (X), Um1-PEPy and Up1-PEPy.
(4–14). 1-Phenylethynylpyrene (1-1-PEPy; Fig. 13.1), a derivative with extended p-conjugation compared with pyrene, is also able to form excimers (15). The 1-1-PEPy fluorophore (see Note 1) has brighter and redshifted emission (compared with pyrene) and is well suitable for nucleic acid labeling (16–23). Conformational shifts during duplex formation cause dramatic changes in the emission spectrum of the probe. Here we describe the detection systems containing an excimerforming 1-1-PEPy pair for homogeneous analysis of 2144A/G and 2143A/G transitions, and 2143A/C substitution in the 23S ribosomal RNA (rRNA) gene of Helicobacter pylori strains resistant to clarithromycin. These mutations determine clarithromycin resistance of H. pylori strains due to inhibition of clarithromycin binding to the 23S rRNA (24–26). The phenylethynylpyrene pair can be assembled using either 1-1-PEPy pseudonucleoside 4-[4-(pyren-1-ylethynyl)phenyl]-1,3-butanediol (X) or 20 -O-(1-1-PEPy) -modified nucleosides – 20 -O-[3-(pyren-1-ylethynyl)benzyl]uridine (Um1-PEPy) and 20 -O-[4-(pyren-1-ylethynyl)benzyl]uridine (Up1-PEPy). The structures of the 1-1-PEPy compounds are shown in Fig. 13.1. Initially, we used pseudonucleoside X inside oligonucleotide probes for SNP analysis (21); however, using this approach, we were unable to detect all mutations, probably because X is a mixture of two enantiomers, and its introduction into oligonucleotide produces a pair of diastereomers. Thus, oligonucleotide probes containing an XX sequence are in fact mixtures of four diastereomers. It is known that oligonucleotides containing adjacent 20 -O(pyren-1-ylmethyl)uridines show hybridization-sensitive pyrene excimer fluorescence (27–29). Therefore, we applied the similar nucleoside derivatives Um1-PEPy and Up1-PEPy, containing 1-1PEPy residue instead of pyrene, to construct probes containing 1-1-PEPy pairs. Um1-PEPy and Up1-PEPy differ in their position of dye attachment to the 20 oxygen atom of uridine – meta or para (Fig. 13.1). There are four possible adjacent pairs that could be formed using these two nucleosides: Um1-PEPyUm1-PEPy, Um1PEPy p1-PEPy U , Up1-PEPyUm1-PEPy, and Up1-PEPyUp1-PEPy sequences (see Note 2). 1-PEPy-labeled nucleosides have two advantages over X: (1) 20 -O-(pyren-1-ylmethyl)uridine is able to form complementary pair with an opposite deoxyadenosine; (2) each oligonucleotide
Phenylethynylpyrene Excimer Forming Hybridization Probes
211
probe containing the adjacent pairs mentioned above is an individual stereoisomer. However, Um1-PEPy and Up1-PEPy are suitable for constructing excimer-forming pairs only when a TT (AA) sequence is present opposite a SNP site. Here we describe Um1-PEPy- and Up1PEPy -containing probes for the detection of 2144A/G, 2143A/G, and 2143A/C mutations.
2. Materials 2.1. Oligonucleotide Probes, Primers, and Targets
2.2. Hybridization of Probes and Model Targets or SingleStranded PCR Products 2.3. Asymmetric PCR Amplification
Nonmodified oligonucleotides are available from many companies, e.g., Sigma. Oligonucleotides containing 1-1-PEPy derivatives X, Um1-PEPy, and Up1-PEPy are available from Primetech (Minsk, Belarus) (30). Modified oligonucleotides should be ordered as double-purified (isolated by polyacrylamide gel electrophoresis followed by high-performance liquid chromatography) and desalted species. 1. Hybridization buffer (5 ): 500 mM NaCl, 50 mM KH2PO4, 0.5 mM Na2EDTA pH 7.0. 2. Fluorescent dye SYBR Green I for DNA staining, 10000x in dimethyl sulfoxide (Invitrogen/Molecular Probes). 1. Thermophilic Taq polymerase (Sigma, D 1806). 2. PCR amplification buffer (10x): 100 mM tris(hydroxymethyl) aminomethane (Tris)–HCl, pH 8.0, 500 mM KCl, 15 mM MgCl2. 3. Deoxynucleoside triphosphates (dNTPs) mixture: solution of dNTP 10 mM in each dNTP (Sigma). 4. DNA samples with a known point mutation in 23S rRNA genes H. pylori for this study was provided by F. Megraud (Bordeaux, France). 5. Forward primer C1 (50 -GCGTTGAATTGAAGCCCGAGTAAAC-30 ) and reverse primer Ap23r8 (50 -AGTAAAGGTCCACGGGGTCT-30 ) were used for the PCR amplification of the 200-bp region of the 23S rRNA gene, which includes SNPs 2143A/G, 2143A/C and 2144A/G.
3. Methods 3.1. Design of Excimeric 1-PEPy Probes
While designing the probes, one substitutes two neighboring nucleotides of the selected probe by 1-PEPy residues in various positions close to the mutation sites mentioned earlier. This allows
212
Prokhorenko et al.
one to select the optimal position of 1-PEPy in the oligonucleotide sequence that gives maximum spectral changes during probe– target hybridization. We made a screening of fluorescent probes by use of 40-bp targets as a model of single-stranded DNA. The sequences of these targets and 1-PEPy excimeric probes are shown in Table 13.1. 3.2. Thermal Denaturation Studies
To verify efficient duplex formation by excimer probes with their targets, we have investigated duplex melting points using SYBR Green I dye fluorescence. Examples of fluorescence melting curves are shown in Fig. 13.2. In all cases, the influence of modifications on duplex stability was minor; thus, one can omit this step (see Note 3). 1. To prepare the model duplexes, mix complementary oligonucleotides in equimolar ratio in the hybridization buffer to reach a final concentration of 1 10–6 L. Anneal oligonucleotides by heating the mixtures at 95C for 5 min and slowly cool the mixture to room temperature within 3 h. 2. For registration of melting curves, add to the duplex solutions SYBR Green I to a final concentration of 1 . Register the temperature dependence of the sample fluorescence in the interval from 30 to 70C with a heating rate of 1C/min. We used an ABI PRISM 8000 sequence detection system spectrofluorometer (Applied Biosystems) for these measurements (see Note 4).
3.3. Fluorescent Analysis of the Model Probe–Target System
All 1-PEPy bifluorophore oligonucleotide probes based on the XX pair show characteristic fluorescence spectra, maximum emission wavelength 402 and 425 nm for the monomer band, and 510 nm for the excimer band (Fig. 13.3). This confirms the close proximity of two 1-PEPy residues. The model duplexes with fluorescent probes should show efficient spectral alterations with the native target as compared with the mutant one. 1. Register the fluorescence spectra of model duplexes, prepared as described in the previous section (excitation wavelength 380 nm). We used a Varian Eclipse fluorescence spectrophotometer and 120-mL cells (see Note 5). 2. Analyze spectral changes if probe duplexes relative to variable base positions. Find the more spectrally sensitive probes and compare their basal responses. Try to reveal a pattern of spectral profile shape relative to a nucleobase in the target sequence (see Note 6). For example, the excimer emission intensity of duplexes notably decreased especially in the case of X9X10 and X10X11 probes, whereas the monomer emission increased, leading to complete predominance of this band (Fig. 13.4). Surprisingly,
WTTs A43Gs A43Cs A44Gs UmUm1 UpUm1 UmUp1 UpUp1 UmUm2 UpUm2 UmUp2 UpUp2
Model 20-mer targets – fragments of the coding chain (sense)
Nucleoside 1-PEPy U1-PEPyU1-PEPy 20-mer excimer probes complementary to the coding sequence
The structures of Um1-PEPy and Up1-PEPy are shown on Fig.13.1
50 -GCAAGACGGAAAGACCCCGT-30 50 -GCAAGACGGGAAGACCCCGT-30 50 -GCAAGACGGCAAGACCCCGT-30 50 -GCAAGACGGAGAGACCCCGT-30
X8X9 X9X10 X10X11 X11X12 NNX
Non-nucleoside 1-PEPy XX 20-mer excimer probes complementary to the noncoding sequence
30 -CGTTCTGCCTUm1-PEPyUm1-PEPyCTGGGGCA-50 30 -CGTTCTGCCTUm1-PEPyUp1-PEPyCTGGGGCA-50 30 -CGTTCTGCCTUp1-PEPyUm1-PEPyCTGGGGCA-50 30 -CGTTCTGCCTUp1-PEPyUp1-PEPyCTGGGGCA-50 30 -CGTTCTGCCUm1-PEPyUm1-PEPyTCTGGGGCA-50 30 -CGTTCTGCCUm1-PEPyUp1-PEPyTCTGGGGCA-50 30 -CGTTCTGCCUp1-PEPyUm1-PEPyTCTGGGGCA-50 30 -CGTTCTGCCUp1-PEPyUp1-PEPyTCTGGGGCA-50
50 -GCAAGACXXAAAGACCCCGT-30 50 -GCAAGACGXXAAGACCCCGT-30 50 -GCAAGACGGXXAGACCCCGT-30 50 -GCAAGACGGAXXGACCCCGT-30 50 -GCAAGACGGAAAGACCCCGT-30
WTTas A43Gas A43Cas A44Gas
30 -TAAGGAGGATGGGCGCCGTTCTGCCTTTCTGGGGCACCTG-50 30 -TAAGGAGGATGGGCGCCGTTCTGCCCTTCTGGGGCACCTG-50 30 -TAAGGAGGATGGGCGCCGTTCTGCCGTTCTGGGGCACCTG-50 30 -TAAGGAGGATGGGCGCCGTTCTGCCTCTCTGGGGCACCTG-50
50 -GAAAATTCCTCCTACCCGCGGCAAGACGGAAAGACCC-CGTGGACCTTTAC-30
2143 2143
Model 40-mer targets – fragments of the non-codingchain (antisense)
The gene fragment of Helicobacter pylori 23S rRNA (the coding sequence; wild type)
Table 13.1 Wild-type and mutant sequences of 23S ribosomal RNA (rRNA) gene and hybridization probes
Phenylethynylpyrene Excimer Forming Hybridization Probes 213
214
a
Prokhorenko et al.
b
Fig. 13.2. Direct (a) and differential (b) melting curves of duplexes X8X9 WTT (a), X10X11 WTT (b), and NNX WTT (c) recorded using SYBR Green I fluorescence.
Fig. 13.3. Emission spectra of 1-1-PEPy excimer-forming oligonucleotide probes in hybridization buffer: X8X9 (a), X10X11 (b), X9X10 (c), and X11X12 (d).
the most effective probes were those in which the dye insertions were located just opposite the variable nucleotide sites, particularly, the 30 -1-PEPy residue. Thus, fluorescent properties of the probes X9X10 and X10X11 appeared to be suitable for the detection of the single nucleotide alterations, 2144A/G or, 2143A/G and/or 2143A/G, in the target sequence. The presence of the excimer emission in the hybridized probe X9X10 indicated a mutation at position 2143 (substitutions 2143A/G or 2143A/C), and, similarly, the excimer emission of the probe X10X11 indicated the presence of the 2144A/G
Phenylethynylpyrene Excimer Forming Hybridization Probes
a
b
c
d
215
Fig. 13.4. Emission spectra of duplexes formed from 1-PEPy probes. (A) X8X9, (B) X9X10, (C) X10X11, and (D) X11X12 with model targets WTT (a), A44G (b), A43G (c), and A43C (d).
substitution. In general, probes X9X10 and X10X11 show excimer emission when the 30 -1-PEPy residue has been placed opposite the nucleotide substitution. The selected probes act as fluorescent molecular switches with three different positions: single stranded, natural duplex, and duplex with a single nucleotide alteration. These two probes were used in hybridization with amplified DNA samples. To distinguish all mutations (i.e., for epidemiological studies) nucleoside 1-PEPy derivatives should be used for design of excimer-forming probes. The wild-type 23S RNA gene contains a TTT sequence opposite positions 2143, 2144, and 2145 of the coding chain. Therefore, there are two possibilities for the location of the adjacent U1-PEPyU1-PEPy pair, and four possibilities of its composition (Um1-PEPyUm1-PEPy, Um1-PEPyUp1-PEPy, Up1-PEPyUm1-PEPy, and Up1PEPy p1-PEPy U ). Combining 1-PEPy pair location and composition gives eight DNA probes (Table 13.1). DNA probes labeled at positions 9 and 10 exhibit fluorescence twice as intense as 10,11-labeled analogues, and in both cases the spectra contain monomer together with excimer emission bands. The excimer wavelength maximum
Prokhorenko et al.
for nucleoside 1-PEPy pairs is about 500 nm. Upon hybridization of U1-PEPyU1-PEPy probes with the complementary or singly mismatched DNA (see Note 7), the I500/I405 ratio changed remarkably (Fig. 13.5). Probes UpUm1, UpUp1, and UmUm2 display I500/I405 of duplexes which allows detection of 2144A/G mutation in the DNA target, while UpUm2 and UpUp2 detect 2143C/A mutation in the 23S RNA gene. However, one can distinguish all the mutations by subsequent application of all the DNA probes followed by calculation of the fluorescence ratio I500/I405. Of course, this is possible only when one type of mutation is present.
a
1.6 ss probe
1.4
probe-WTs probe-A43Gs
1.2
probe-A43Cs probe-A44Gs
I500/I405
1 0.8 0.6 0.4 0.2 0 UmUm1
b
UpUm1
UmUp1
3
UpUp1
ss probe probe-WTs probe-A43Gs probe-A43Cs probe-A44Gs
2.5 2
I500/I405
216
1.5 1 0.5 0
UmUm2
UpUm2
UmUp2
UpUp2
Fig. 13.5. Fluorescent properties (excimer to monomer fluorescence intensity ratio) of nucleoside 1-PEPy excimer-forming probes and their duplexes with complementary or singly mismatched DNA: (A) probes of type 1 (modifications in positions 9, 10); (B) probes of type 2 (modifications in positions 10, 11).
Phenylethynylpyrene Excimer Forming Hybridization Probes
3.4. Asymmetric PCR Amplification
217
1. Prepare 50-mL reactions in 1 PCR amplification buffer containing 10 ng of DNA, 200 nM of each primer, 5 U Taq polymerase, and 2 mM dNTPs. After denaturation at 94C for 3 min, perform 30 cycles of amplification (94C for 1 min, 55C for 20 s, 72C for 30 s) in a thermocycler; we used a Multi Gene II cycler (Labnet International, USA). Use the PCR products obtained in this reaction as templates in the asymmetric PCR. 2. The asymmetric PCR must contain in 50 mL 1 PCR amplification buffer, 5 mL of the reaction mixture from the previous PCR step, 5 U Taq polymerase, 2 mM dNTPs, and a 2 mM solution of the reverse primer Ap23r8. The amplification program included denaturation at 94C (3 min) followed by 25 cycles of 94C for 20 s, 55C for 20 s, and 72C for 20 s.
Table 13.2 Fluorescence of probes X9X10 and X10X11 hybridized with single-stranded PCR products
Sample
Mutation (determined by sequencing)
Excimer to monomer fluoresecnce intensity ratio (I510 nm/I405 nm) after hybridization with the probe X9X10 X10X11
1
2143A/G, wild type
0.54
–a
2
2144A/G, wild type
–a
1.50
3
2143A/G
0.83
–a
4
2144A/G
–a
1.34
5
2144A/G
–a
1.34
6
2143A/G, 2144A/G
0.30
1.47
7
2144A/G
–a
1.18
8
2143A/G, 2143A/C
0.62
–a
9
2143A/G
0.82
–a
10
2143A/G, 2144A/G, wild type
0.43
0.97
11
2143A/G
0.82
–a
12
2143A/G
0.72
–a
13
2144A/G
–a
1.52
14 a
Wild type
No characteristic excimer band was observed.
a
–
–a
218
Prokhorenko et al.
Reaction products were analyzed by electrophoresis (150 V, 80 mA; Tris–acetate–EDTA buffer) in a 2.5% agarose gel containing 0.01% ethidium bromide (see Note 8). 3.5. Fluorescent SNP Analysis
1. Add 2 pmol of the probe to each reaction mixture containing single-stranded PCR product (see Note 9). Anneal the mixtures by heating at 95C for 5 min and cooling to room temperature within 3 h. 2. Analyze changes in the shape of the fluorescence spectra corresponding to the mutations present in the targets, allowing SNP identification. 3. Record the fluorescence spectra with a Varian Eclipse fluorescence spectrophotometer in a 120-mL cell using an excitation wavelength of 380 nm (see Note 10). 4. Calculate the excimer-to-monomer emission intensity ratio (I510 nm/I405 nm). Compare the data with those in Table 13.2 to distinguish the mutation (see Note 11). An example for DNAs with substitution 2144A/G and a wild-type DNA tested with the X10X11 probe is shown in Fig. 13.6. The results for several samples are summarized in Table 13.2. Duplexes with wild-type DNA show only monomer emission. Excimer emission indicates the presence of SNP. The 1-PEPy excimer emission from the X9X10 probe is indicative of
Fig. 13.6. Emission spectra of duplexes of the X10X11 probe (2 pmol) and asymmetric PCR: DNA 14 (wild type) (a); DNA 2 (wild type + 2144A/G mutation) (b), DNA 7 (2144A/G mutation) (c). Examples are from Table 13.2.
Phenylethynylpyrene Excimer Forming Hybridization Probes
219
the mutation at position 2143, and that from the X10X11 probe identifies a mutation at position 2144. Mixed samples cannot be resolved.
4. Notes 1. There are three possible isomers of phenylethynylpyrene dye: 1-1-PEPy, 2-1-PEPy, and 4-1-PEPy. 2-1-PEPy and 4-1PEPy have reduced fluorescence quantum yield as compared with 1-1-PEPy (31, 32), and therefore are less suitable for nucleic acid labeling. In this protocol only 1-1-PEPy is used. 2. Oligonucleotide probes containing Um1-PEPyUp1-PEPy and Up1-PEPyUm1-PEPy pairs are different compounds because of the 50 –30 directionality of the oligonucleotide sequence. 3. 1-PEPy pseudonucleoside X only slightly decreases the stability of the DNA duplex when it is placed in the middle part, and increases the stability of the duplex in dangling positions (17). This is probably because the flat pyrene residue is able to serve as a base pair substitute and stack with other base pairs. 4. Fluorescence of SYBR Green I dye bound to double-stranded DNA is considerably increased as compared with that of single-stranded DNA. The change in the sigmoidal curve of the fluorescence intensity corresponds to the duplex melting temperature. 5. Excitation and emission slits should be usually 1.5–2.5 nm. Broader slits allow good spectra to be obtained from dilute samples of 1-PEPy-labeled oligonucleotides – up to 1 10–7 L. 6. There are no known patterns/rules for excimeric fluorescence of 1-PEPy oligonucleotide probes. Therefore, various probes should be designed for particular mutations, and their fluorescence should be studied on model duplexes. 7. Model oligonucleotides complementary to the U1-PEPyU1PEPy probes, as well as singly mismatched targets are not shown in Table 13.1. These are fragments of wild-type and mutant coding sequence 23S RNA H. pylori of the same length as U1-PEPyU1-PEPy probes (20 nucleotides). 8. Gel staining shows single-stranded and double-stranded PCR products. Note that the sensitivity of double-stranded DNA staining in gel is much higher than that for single-stranded DNA. Thus, if the fluorescence intensities of single-stranded and double-stranded DNA are similar, this is a good result for asymmetric PCR.
220
Prokhorenko et al.
9. It is important for the quantity of the probe to be less than the quantity of the single-stranded PCR product, because nonhybridized probe displays excimer fluorescence and would cause false-positive results. 10. Increased excitation and emission slits (up to 5 nm) may be used to register spectra from probes hybridized with singlestranded PCR products. 11. Data are suitable for the particular case of SNPs in positions 2143 and 2144 of the 23S rRNA gene of H. pylori strains resistant to clarithromycin. The development of the 1-PEPy excimer detection system for particular SNPs requires registration of fluorescence spectra of model duplexes.
Acknowledgment This work was supported by grant 06-03-32426 from the Russian Foundation for Basic Research (RFBR). References 1. Ranasinghe, R. T. and Brown, T. (2005) Fluorescence based strategies for genetic analysis. Chem. Commun., 5487–5502. 2. Kim, S. and Misra, A. (2007) SNP genotyping: technologies and biomedical applications. Annu. Rev. Biomed. Eng. 9, 289–320. 3. Winnik, F. M. (1993) Photophysics of preassociated pyrenes in aqueous polymer solutions and in other organized media. Chem. Rev. 93, 587–614. 4. Paris, P. L., Langenhan, J. M. and Kool, E. T. (1998) Probing DNA sequences in solution with a monomer–excimer fluorescence color change. Nucleic Acids Res. 26, 3789–3793. 5. Masuko, M., Ohtani, H., Ebata, K. and Shimadzu, A. (1998) Optimization of excimerforming two-probe nucleic acid hybridization method with pyrene as a fluorophore. Nucleic Acids Res. 26, 5409–5416. 6. Yamana, K., Iwai, T., Ohtani, Y., Sato, S., Nakamura, M. and Nakano, H. (2002) Bispyrene-labeled oligonucleotides: sequence specificity of excimer and monomer fluorescence changes upon hybridization with DNA. Nucleic Acids Res. 26, 3789–3793. 7. Christensen, U. B. and Pedersen, E.B. (2003) Intercalating nucleic acids with pyrene nucleotide analogues as next-nearest
8.
9.
10.
11.
12.
neighbors for excimer fluorescence detection of single-point mutantions under nonstringent hybridization conditions. Helv. Chim. Acta 86, 2090–2097. Dioubankova, N. N., Malakhov, A. D., Stetsenko, D. A. and Korshun, V. A. (2004) Detection of point mutations using pyrenelabeled DNA probes. Russ. Chem. Bull. Int. Ed. 53, 463–470. Okamoto, A., Ichiba, T. and Saito, I. (2004) Pyrene-labeled oligodeoxynucleotide probe for detecting base insertion by excimer fluorescence emission. J. Am. Chem. Soc. 126, 8364–8365. Hrdlicka, P. J., Babu, B. R., Sørensen, M. D. and Wengel, J. (2004) Interstrand communication between 20 -N-(pyren-1-yl)methyl20 -amino-LNA monomers in nucleic acid duplexes: directional control and signalling of full complementarity. Chem. Commun., 1478–1479. Yamana, K., Fukunaga, Y., Ohtani, Y., Sato, S., Nakamura, M., Kim, W. J., Akaike, T. and Maruyama, A. (2005) DNA mismatch detection using a pyrene–excimer-forming probe. Chem. Commun., 2509–2511. Kashida, H., Asanuma, H. and Komiyama, M. (2006) Insertion of two pyrene moieties into oligodeoxyribonucleotides for the
Phenylethynylpyrene Excimer Forming Hybridization Probes
13.
14.
15.
16.
17.
18.
19.
20.
efficient detection of deletion polymorphisms. Chem. Commun., 2768–2770. Kumar, T. S., Wengel, J. and Hrdlicka, P. (2007) 20 -N-(Pyren-1-yl)acetyl-20 -amino-L-LNA: synthesis and detection of single nucleotide mismatches in DNA and RNA targets. Chembiochem. 8, 1122–1125. Umemoto, T., Hrdlicka, P., Babu, B. R. and Wengel, J. (2007) Sensitive SNP dual-probe assays based on pyrene-functionalized 20 amino-LNA: lessons to be learned. Chembiochem. 8, 2240–2248. Malakhov, A. D., Malakhova, E. V., Kuznitsova, S. V., Grechishnikova, I. V., Prokhorenko, I. A., Skorobogatyi, M. V., Korshun, V. A. and Berlin, Y. A. (2000) Synthesis and fluorescent properties of 5-(1-pyrenylethynyl)-20 -deoxyuridine-containing oligodeoxynucleotides. Russ. J. Bioorg. Chem. 26, 34–44. Korshun, V. A., Prokhorenko, I. A., Gontarev, S. V., Skorobogatyi, M. V., Balakin, K. V., Manasova, E. V., Malakhov, A. D. and Berlin, Y. A. (1997) New pyrene derivatives for fluorescent labeling of oligonucleotides. Nucleosides Nucleotides 16, 1461–1464. Malakhov, A. D., Skorobogatyi, M. V., Prokhorenko, I. A., Gontarev, S. V., Kozhich, D. T., Stetsenko, D. A., Stepanova, I. A., Shenkarev, Z. O., Berlin, Y. A. and Korshun, V. A. (2004) 1-(Phenylethynyl)pyrene and 9,10-bis(phenylethynyl)anthracene, useful fluorescent dyes for DNA labeling: excimer formation and energy transfer. Eur. J. Org. Chem., 1298–1307. Dioubankova, N. N., Malakhov, A. D., Shenkarev, Z. O. and Korshun, V.A. (2004) Oligonucleotides containing new fluorescent 1-phenylethynylpyrene and 9,10-bis(phenylethynyl)anthracene uri0 dine-2 -carbamates: synthesis and properties. Tetrahedron 60, 4617–4626. Filichev, V. V. and Pedersen, E. B. (2005) Stable and selective formation of Hoogsteen-type triplexes and duplexes using twisted intercalating nucleic acids (TINA) prepared via postsynthetic Sonogashira solid-phase coupling reactions. J. Am. Chem. Soc. 127, 14849–14858. Maeda, H., Maeda, T., Mizuno, K., Fujimoto, K., Shimizu, H. and Inouye, M. (2006) Alkynylpyrenes as improved pyrenebased biomolecular probes with the advantages of high fluorescence quantum yields and long absorption/emission wavelengths. Chem. Eur. J. 12, 824–831.
221
21 Prokhorenko, I. A., Malakhov, A. D., Kozlova, A. A., Momynaliev, K., Govorun, V. M. and Korshun, V. A. (2006) Phenylethynylpyrene-labeled oligonucleotide probes for excimer fluorescence SNP analysis of 23S rRNA gene in clarithromycin resistant Helicobacter pylori strains. Mutat. Res. 599, 144–151. 22 Filichev, V. V., Nielsen, M. C., Bomholt, N., Jessen, C. H., and Pedersen, E. B. (2006) High thermal stability of 50 -50 linked alternate Hoogsteen triplexes at physiological pH. Angew. Chem. Int. Ed. 45, 5311–5315. 23. Astakhova, I. A., Malakhov, A. D., Stepanova, I. A., Ustinov, A. V., Bondarev, S. L., Paramonov, A. S. and Korshun, V. A. (2007) 1-Phenylethynylpyrene (1-1-PEPy) as refined excimer forming alternative to pyrene: case of DNA major groove excimer. Bioconjugate Chem. 18, 1972–1980. 24. Versalovic, J., Shortridge, D., Kibler, K., Griffy, M. V., Beyer, J., Flamm, R. K., Tanaka, S. K., Graham, D. Y. and Go, M. F. (1996) Mutations in 23S rRNA are associated with clarithromycin resistance in Helicobacter pylori. Antimicrob. Agents Chemother. 40, 477–480. 25. Stone, G. G., Shortridge, D., Flamm, R. K., Versalovic, J., Beyer, J., Idler, K., Zulawinski, L. and Tanaka, S. K. (1996) Identification of a 23S rRNA gene mutation in clarithromycin-resistant Helicobacter pylori. Helicobacter 1, 227–228. 26. Hulte´n, K., Gibreel, A., Sk¨old, O. and Engstrand, L. (1997) Macrolide resistance in Helicobacter pylori: mechanism and stability in strains from clarithromycin-treated patients. Antimicrob. Agents Chemother. 41, 2550–2553. 27. Mahara, K., Iwase, R., Sakamoto, T., Yamana, K., Yamaoka, T. and Murakami, A. (2002) Bispyrene-conjugated 20 -Omethyloligonucleotide as a highly specific RNA-recognition probe. Angew. Chem. Int. Ed. 41, 3648–3650. 28. Nakamura, M., Shimomura, Y., Ohtoshi, Y., Sasa, K., Hayashi, H., Nakano, H. and Yamana, K. (2007) Pyrene aromatic arrays on RNA duplexes as helical templates. Org. Biomol. Chem. 5, 1945–1951. 29. Nakamura, M., Murakami, Y., Sasa, K., Hayashi, H. and Yamana, K. (2008) Pyrene-zipper array assembled via RNA duplex formation. J. Am. Chem. Soc. 130, 6904–6905. 30. http://www.primetech.by
222
Prokhorenko et al.
31. Astakhova, I. A. and Korshun, V. A. (2008) 2- and 4-Phenylethynylpyrenes, new fluorescent labels for DNA. Russ. J. Bioorg. Chem. 34, 510–512. 32. Filichev, V. V., Astakhova, I. A., Malakhov, A. D., Korshun, V. A. and Pedersen, E. B.
(2008) 1-, 2-, and 4-Ethynylpyrenes in the structure of twisted intercalating nucleic acids: structure, thermal stability, and fluorescence relationship. Chem. Eur. J. 14, 9968–9980.
Chapter 14 The Chemical Cleavage of Mismatch for the Detection of Mutations in Long DNA Fragments Tania Tabone, Georgina Sallmann, and Richard G. H. Cotton Abstract Methods to rapidly scan large regions of DNA that are not dependent on highly specific melting temperatures or suitable only for large-scale discovery are important to reduce the amount of sequencing required for DNA samples that appear to contain a mutation. This protocol describes the chemical cleavage of mismatch method to assess if a region of DNA contains a mutation and accurately localize the position of the mutation in the same reaction. To detect mutations, PCR heteroduplexes are incubated with two mismatch-specific reagents. Hydroxylamine modifies mismatched cytosine residues and potassium permanganate modifies mismatched thymine residues. The samples are then incubated with piperidine, which cleaves the DNA backbone at the site of the modified mismatched base. Cleavage products are separated by electrophoresis, revealing the identity and location of the mutation. The chemical cleavage of mismatch method can efficiently detect point mutations as well as insertions and deletions. Key word: Heteroduplex analysis, mismatch, chemical cleavage, single nucleotide polymorphism, mutation detection.
1. Introduction The chemical cleavage of mismatch (CCM), first described in 1988 (1), remains the only mutation detection method that is able to simultaneously detect and accurately locate the position of a sequence variant in PCR fragments greater than 1 kb in length. The method is suitable for low- to medium-throughput mutation detection applications and is particularly useful for diagnostic applications where large numbers of genes in a small number of DNA samples need to be screened as the same assay conditions can be used to analyze the different PCR products. The CCM A.A. Komar (ed.), Single Nucleotide Polymorphisms, Methods in Molecular Biology 578, DOI 10.1007/978-1-60327-411-1_14, ª Humana Press, a part of Springer Science+Business Media, LLC 2003, 2009
223
224
Tabone, Sallmann, and Cotton
approach is suitable for the detection of all types of DNA mutations, including single nucleotide polymorphisms, insertions, and deletions. The principle of mutation detection by CCM relies on the formation of heteroduplex DNA, which is produced when DNA (amplified by PCR) from a wild-type allele and DNA from a mutant allele are annealed. The two alleles can be derived either from a heterozygous individual or by combining a homozygous patient sample with a wild-type reference sample. CCM analysis utilizes two commonly available reagents to initially detect and modify mismatched pyrimidine bases and a third reagent to cleave the DNA at the site of modification, thus allowing the presence of a mutation to be revealed and the mutation to be located. Hydroxylamine (NH2OH) modifies mismatched C residues by forming adducts across the carbon 5,6 double bonds that are exposed in unpaired or mispaired C residues. Potassium permanganate (KMnO4) modifies mismatched T residues by oxidizing the carbon 5,6 double bonds that are exposed in unpaired or mispaired T residues. These two modification reactions open the pyrimidine ring structure, further exposing the mismatched DNA bases to piperidine-induced b-elimination, which cleaves the DNA sugar phosphate backbone at the modified site. Separation of the cleavage products by electrophoresis confirms both the presence and the location of the mutation. Furthermore, owing to the mismatch specificity of the two chemical reactions, the sequence of the mismatch can be inferred. This provides an additional mechanism to distinguish between a known common sequence variant and a potential functional mutation or rare polymorphism, thereby reducing the amount of sequencing to those samples that appear to have a novel sequence variant (Fig. 14.1). Higher-throughput semiautomated mutation-detection methods, such as high-resolution melt curve analysis (hrMCA) (2) are available; however, CCM is more robust and reliable as it is not dependent on the highly specific melting temperatures required by hrMCA and other commonly used techniques, including denaturing gradient gel electrophoresis (3) (see Chapter 8 in this volume) and single strand conformation polymorphism analysis (4) (see Chapter 12 in this volume), in which a difference in just 1 C can determine whether a mutation is detected. In addition, the limited sensitivity of mutation detection of these methods restricts the analysis of PCR products to those less than 500 bp in length. CCM is also not dependent on enzymatic activities, which can be expensive and often require lengthy optimization to obtain the ideal incubation time or temperature. The CCM method is accurate, reliable, and robust for the detection and genotyping of genetic mutations. In addition, the diagnosis or genotyping of known mutations requires only the application of the chemical modifying the mismatch of interest to be used, further
Chemical Cleavage of Mismatch for the Detection of Mutations in Long DNA Fragments
225
Fig. 14.1. The chemical cleavage of mismatch (CCM) method, demonstrating the detection of a T.G/C.A mismatch. Treatment with potassium permanganate modifies mismatched T bases and treatment with hydroxylamine modifies mismatched C bases. Incubation with piperidine cleaves the modified DNA at the site of the mismatch, which is readily detected after electrophoretic separation.
226
Tabone, Sallmann, and Cotton
reducing reagent costs. For example, genotyping a G.T mutation would only require the DNA sample to be incubated with KMnO4 to detect the mismatched T residue. Alternatively, the reaction can be performed with NH2OH to detect the C.A mismatch on the antisense DNA strand. The use of primers with different 50 -labeled fluorophores, such as 6-HEX for the forward primer and FAM for the reverse primer, allows the detection of the mismatched allele and its assignment to the sense or antisense strand, permitting the genotype to be inferred (Fig. 14.1) as well as enabling multiplexing to improve assay throughput. CCM has proven especially suited to the detection of mutations in diluted samples, such as tumor samples, in which the presence of a mutation can be obscured by an excess of wild-type alleles (5, 6). CCM has been successfully used for the detection and identification of causative mutations in genes implicated in diseases including hemophilia A and B, hereditary angioedema, phenylketonuria, Alzheimer’s disease, colorectal cancer, and congenital nystagmus (7–14). The CCM protocols for detecting mutations using liquidphase and solid-support phase analysis are described here, as both approaches are commonly used and have different advantages according to individual requirements. Liquid-phase analysis is more time-consuming than solid-support phase analysis because it requires an ethanol-precipitation step. However, liquid-phase analysis is cheaper to perform routinely and has a greater sensitivity for detecting mutations in fragments of 2 kb or less in length. Solidsupport phase analysis eliminates the need for an ethanol-precipitation step, allowing faster processing time or adaptation to robotics. However, it requires the addition of silica beads, which might not be desirable owing to the increased expense. Care must be taken to completely remove the beads by transferring the solution to a new reaction vessel, to prevent clogging of the capillary or wells during electrophoretic analysis. Solid-support phase analysis is sensitive for the detection of mutations in PCR fragments 500 bp or less in length, and is ideal for detecting mutations close to the primer site as primer–dimers are removed during the washing steps.
2. Materials 2.1. Reagents
1. Hydroxylamine hydrochloride (NH2OH-HCl) (Sigma-Aldrich, St. Louis, MO, USA). Prepare a 4.2 M solution by dissolving 6.95 g NH2OH-HCl in 7.5 mL distilled H2O (dH2O). Adjust the solution to pH 6 with diethylamine. Adjust the final volume to 20 mL with dH2O. Store the solution at –70C in 1.5-mL aliquots. The solution can be stored for 3 months. Equilibrate the solution to room temperature (15–25C) before use.
Chemical Cleavage of Mismatch for the Detection of Mutations in Long DNA Fragments
227
2. 100 mM KMnO4. Dissolve 0.0165 g KMnO4 (SigmaAldrich) in 1 mL dH2O. Vortex the solution vigorously for approximately 1 min to completely dissolve the KMnO4 crystals. Store the solution at –20C in 100-mL aliquots in 1.2-mL microcentrifuge tubes wrapped in foil. The solution can be stored for up to 6 months (see Note 1). 3. 3 M tetraethylammonium chloride (TEACl), pH 8.0. Dissolve 100 g TEACl (Sigma-Aldrich) in dH2O to 200-mL final volume (see Note 2). Filter the solution through a 0.45-mm membrane and store the filtrate at room temperature (15– 25C). The solution can be stored for up to 6 months. 4. Reconstitute lyophilized fluorescently labeled primers in dH2O to 100 mM and store the solution at –20C. Prepare a 10 mM working dilution in dH2O and store it at –20C in small aliquots, wrapped in foil to minimize freeze/thawing activity and fluorescent degradation. 5. 5 U/mL Taq DNA polymerase (e.g., Invitrogen, Carlsbad, CA, USA). 6. 100 mM dNTP (Bioline, Alexandria, NSW, Australia). Dilute the dNTP to 2 mM working dilution in dH2O and store the solution at –20C in small aliquots to minimize freeze/thawing activity. 7. 2X annealing buffer: 12 mM tris(hydroxymethyl)aminomethane– HCl, pH 7.5, 1.2 M NaCl, 14 mM MgCl2. Autoclave the solution and store it at room temperature (15–25C). 8. 3 M NaAc, pH 5.2. 9. 100% ethanol and 70% (vol/vol) ethanol. 10. Poly(ethylene glycol) (PEG)/NaCl solution: 25% (wt/vol) PEG 8000 and 2.5 M NaCl. Dissolve 250 g PEG 8000 and 146 g NaCl in dH2O to a 1-L final volume. Store the solution at room temperature (15–25C). 11. Stop buffer: 0.3 M NaAc pH 5.2, 0.1 mM EDTA, 25 mg/mL transfer RNA. Store at -20C. 12. Piperidine (Sigma-Aldrich). Prepare a 10% (vol/vol) piperidine solution by diluting piperidine with dH2O immediately before use (see Note 3). 13. Formamide dye: 200 mL blue dextran (50 mg/mL) and 800 mL formamide. Store at –20C in small aliquots to minimize freeze/thawing activity. 14. Piperidine/dye: Prepare a 2 mL piperidine and 8 mL formamide dye solution per sample for acrylamide-based electrophoretic separation. Prepare it fresh on the day of the experiment.
228
Tabone, Sallmann, and Cotton
15. Piperidine/formamide: Prepare a 2 mL piperidine and 8 mL formamide solution per sample for capillary electrophoretic separation. Prepare it fresh on the day of the experiment. 2.2. Equipment
1. Silica beads (e.g., Ultra Bind beads, MO BIO Laboratories, CA, USA). 2. Polyethylene or polypropylene PCR microplates or tubes (see Note 4). 3. Capillary or acrylamide-gel-based DNA sequencer (e.g., ABI 3730, Applied Biosystems, Foster City, CA, USA).
3. Methods 3.1. PCR Amplification
1. In a thin-walled 0.2-mL PCR tube or 96-well microplate, add the PCR components on ice in the following order:
Component
X1 reaction
dH2O (to 25 mL final volume)
17.0 mL
10X PCR buffer
2.5 mL
1X
2 mM dNTP
2.5 mL
0.2 mM
50 mM MgCl2
0.75 mL
1.5 mM
10 mM 5’-6-FAM-labeled forward primer
0.5 mL
0.2 mM
10 mM 5’-HEX-labeled reverse primer
0.5 mL
0.2 mM
20–100 ng/mL test or wild-type template DNA
1.0 mL
20–100 ng
5 U/mL Taq DNA polymerase
0.25 mL
1.25 U
Final
2. Amplify the DNA in a thermal cycler using the following conditions: an initial DNA denaturation at 95C for 5 min; followed by 35 cycles consisting of denaturation at 95C for 20 s, annealing at the optimal primer melting temperature for 20 s, and extension at 72C for 20 s; and a final extension at 72C for 10 min (see Note 5). 3.2. Duplex Formation (Performed as Two Separate Tests)
1. Combine the PCR products in 0.2-mL PCR tubes or 96-well microplates and pipette the solution gently up and down to mix it. Centrifuge the solution briefly to collect liquid at the bottom of the well. Reaction 1: To detect mutations in heterozygote DNA samples
Chemical Cleavage of Mismatch for the Detection of Mutations in Long DNA Fragments
229
Test samples: 12.5 mL test PCR product and 12.5 mL 2X annealing buffer. Reference sample: 25 mL control (known wild-type) PCR product. Reaction 2: To detect mutations in homozygote DNA samples Test samples: 12.5 mL unknown PCR product with 12.5 mL control (wild-type) PCR product. Reference sample: 25 mL control (known wild-type) PCR product. 2. As a negative control, set up a homoduplex annealing reaction with 12.5 mL reference (known wild-type) PCR product and 12.5 mL 2X annealing buffer. This homoduplex DNA sample will act as a control to detect nonspecific cleavage by the DNA-modifying reagents and allow the identification of common polymorphisms present in the wild-type and mutant DNA samples. 3. As a positive control, set up a heteroduplex annealing reaction with 6.25 mL PCR product containing a T allele, 6.25 mL PCR product containing a C allele, and 12.5 mL 2X annealing buffer. This heteroduplex DNA sample will act as a control to confirm that the KMnO4, NH2OH, and piperidine incubation times were sufficient for modification and cleavage and that the reagents are functioning efficiently. 4. Denature DNA samples at 95C for 10 min, then cool them slowly from 80 to 25C, decreasing the temperature by 1C every minute to promote heteroduplex formation in a thermal cycler. 3.3. Chemical Cleavage of Mismatch Assay
1. Chemical cleavage of the DNA strands at the sites of base mismatches can be performed either in a homogeneous liquid phase (option described in Section 3.3.1) or on solid-supportbound substrates (option described in Section 3.3.2).
3.3.1. Chemical Cleavage of Mismatches in the Liquid Phase
1. To modify mismatched C and G residues, in a 96-well microplate add 6 mL (about 300 ng) of duplex DNA and 20 mL of 4.2 M NH2OH. Incubate the mixture at 37C for 40 min (or 50 min for fragments larger than 1 kb) (see Note 6). 2. To modify mismatched T and A residues, in a separate 96-well microplate desiccate 6 mL (about 300 ng) of duplex DNA by vacuum centrifugation in a SpeedVac (Thermo Savant) at 37C for 15 min or perform a standard ethanol-precipitation procedure (15).
230
Tabone, Sallmann, and Cotton
3. To the dried DNA pellet, add 20 mL of 100 mM KMnO4/ 3 M TEACl, vortex the mixture briefly to disrupt the pellet, and incubate the mixture at 25C for 5 min (10 min for fragments larger than 1 kb) (see Note 7). 4. To stop the two DNA modification reactions, add 50 mL of stop buffer and 150 mL of ice-cold (4C) 100% (vol/vol) ethanol to each reaction well, vortex the solution briefly and incubate it at –20C for 30 min (see Note 8). 5. Precipitate the DNA by centrifugation at 13,000 rpm for 20 min and discard the supernatant. 6. Wash the pellet with 200 mL of 70% (vol/vol) ethanol, centrifuge the solution at 13,000 rpm for 10 min, discard the supernatant, and allow the pellet to air-dry for 30 min. 7. To cleave the DNA backbone at the site of the modified mismatched DNA, resuspend the DNA pellet in 10 mL of piperidine/dye, vortex the solution vigorously, incubate it at 90C for 30 min, and immediately plunge it onto ice (see Notes 9 and 10). 8. Load 2–3 mL of the NH2OH and KMnO4 reaction mixtures (samples as well as controls) into wells of a 5% denaturing polyacrylamide gel for electrophoretic separation of the DNA fragments. An example of the results produced is shown in Fig. 14.2. 3.4. Chemical Cleavage of Mismatches in the Solid-Support Phase
1. To bind duplex DNA to a solid support, add 10 mL (about 500 ng) of duplex DNA and 5 mL of ULTRA BIND silica beads to a 96-well microplate and incubate the solution at room temperature (15–25C) for 60 min, with occasional vortexing to keep the beads in suspension and allow the DNA to find an available binding site on the silica matrix. 2. Centrifuge the samples at 13,000 rpm for 5 min, discard the supernatant, and wash the DNA-bound silica beads with 250 mL ULTRA WASH solution. Centrifuge the solution at 13,000 rpm for 5 min and remove supernatant. 3. Add 200 mL of ULTRA WASH solution to a microplate and transfer two 100-mL aliquots of the resulting heterogeneous mixture of DNA-bound silica beads into two separate microplates. This will allow for separate base-mismatch modification reactions. 4. Centrifuge the two samples at 13,000 rpm for 5 min and discard the supernatant.
Chemical Cleavage of Mismatch for the Detection of Mutations in Long DNA Fragments
231
Fig. 14.2. CCM analysis of a 550-bp PCR product with a T.G/C.A mismatch. Potassium permanganate modification followed by piperidine cleavage of (a) the 6-FAM-labeled antisense (reverse) strand and (b) the HEX-labeled sense (forward) strand. Hydroxylamine modification followed by piperidine cleavage of (c) the antisense strand and (d) the sense strand. (a) The presence of a peak after treatment with potassium permanganate suggests a mismatched T base in the heteroduplex DNA sample compared with (b) no cleavage peaks present, only full-length DNA is detected in the homoduplex DNA sample. (c) The presence of a peak after treatment with hydroxylamine suggests the presence of a mismatched C base in the heteroduplex DNA sample compared with (d) no cleavage peaks present, only full-length DNA is detected in the homoduplex DNA sample. The nature of the mutation can therefore be inferred without sequencing. The peak height (amount of DNA fragment) is shown on the vertical scale. The horizontal scale represents the molecular weight of the CCM fragment.
232
Tabone, Sallmann, and Cotton
5. To modify mismatched C and reciprocal G residues, add 30 mL of 4.2 M NH2OH to the first aliquot of DNA-bound silica beads (about 250 ng DNA). Incubate the mixture at 37C for 40 min (see Note 6). 6. To modify mismatched T and reciprocal A residues, add 30 mL of 100 mM KMnO4/3 M TEACl to the second aliquot of DNA-bound silica beads (about 250 ng DNA). Incubate the mixture at 25C for 5 min (see Note 7). 7. (Optional) Following modification by KMnO4 and NH2OH, an extra step may be implemented to rebind the DNA to the silica beads that may have been eluted from the support during the modification reactions. For this purpose, add 2 vol (60 mL) of 25% (wt/vol) PEG/2.5 M NaCl to the reactions and incubate the solution at room temperature (15–25C) for 5 min. Centrifuge the DNA briefly and discard the supernatant. Wash the DNA pellets twice with 200 mL of ULTRA WASH, discard the supernatant, and allow the DNA to airdry for 30 min. 8. Centrifuge the two reaction mixtures at 13,000 rpm for 5 min, discard the supernatants, and allow the pellets to airdry for approximately 30 min. 9. To cleave the DNA backbone at the modified residues, resuspend the two reaction mixtures in 10 mL of piperidine/dye and incubate the solution at 90 C for 30 min to release the DNA from the silica beads (see Notes 9 and 10). 10. Centrifuge the solution at 3,000 rpm for 5 min to pellet the beads and transfer the liquid to a new microplate with care to not aspirate any silica beads. 11. Load 2–3 mL of the NH2OH and KMnO4 reaction mixtures (samples as well as controls) into wells of a 5% denaturing polyacrylamide gel for electrophoretic separation of the DNA fragments. An example of the results produced is shown in Fig. 14.2.
4. Notes 1. KMnO4 oxidizes rapidly when exposed to air and is lightsensitive. Minimize freeze/thawing by storing it in small aliquots at –20C and cover the tubes to protect it from light. 2. TEACl is extremely hygroscopic. The entire TEACl stock should be reconstituted to avoid it absorbing moisture from the atmosphere.
Chemical Cleavage of Mismatch for the Detection of Mutations in Long DNA Fragments
233
3. Piperidine is toxic. All manipulations should be performed in a chemical fume hood. 4. Polycarbonate is not resistant to piperidine and is not suitable for use in post-PCR processes. 5. PCR products can be stored at 4C overnight or at –20C for 6 months. 6. Do not incubate for longer than 50 min. Excessive incubation with NH2OH may subject the DNA to nonspecific modification, resulting in cleavage after treatment with piperidine that will not correspond to the presence of a mutation. 7. Do not incubate for longer than 10 min. Excessive incubation with KMnO4 may subject the DNA to nonspecific modification, resulting in cleavage after treatment with piperidine that will not correspond to the presence of a mutation. 8. DNA can be incubated at –20C overnight, protected from light. 9. If the reaction mixtures are heated in PCR tubes, ensure the lids are pierced to prevent popping of the lids and centrifuge them briefly after incubation to collect liquid at the bottom to prevent splashing of the piperidine. Alternatively, if a microplate is used, ensure the plate is covered with plastic film to prevent vapors. 10. Do not incubate for longer than 30 min as nonspecific cleavage of DNA may occur.
Acknowledgements The authors wish to a thank Vincent Sesto for technical assistance with preparation of this manuscript and Hugo Arnott for critically reviewing the manuscript. This work was supported by an NH&MRC grant (to R.G.H.C.). References 1. Cotton, R. G., Rodrigues, N.R. and Campbell, R. D. (1988) Reactivity of cytosine and thymine in single-base-pair mismatches with hydroxylamine and osmium tetroxide and its application to the study of mutations. Proc. Natl. Acad. Sci. U.S.A. 85, 4397–4401. 2. Wittwer, C. T., Reed, G. H., Gundry, C. N., Vandersteen, J. G. and Pryor, R. J. (2003) High-resolution genotyping by amplicon melting analysis using LCGreen. Clin. Chem. 49, 853–860.
3. Myers, R. M., Maniatis, T. and Lerman, L. S. (1987) Detection and localization of single base changes by denaturing gradient gel electrophoresis. Methods Enzymol. 155, 501–527. 4. Orita, M., Iwahana, H., Kanazawa, H., Hayashi, K. and Sekiya, T. (1989) Detection of polymorphisms of human DNA by gel electrophoresis as singlestrand conformation polymorphisms. Proc. Natl. Acad. Sci. U.S.A. 86, 2766–2770.
234
Tabone, Sallmann, and Cotton
5. Lambrinakos, A., Yakubovskaya, M., Babon, J. J., Neschastnova, A. A., Vishnevskaya, Y. V., Belitsky, G. A. et al. (2004) Novel TP53 gene mutations in tumors of Russian patients with breast cancer detected using a new solid phase chemical cleavage of mismatch method and identified by sequencing. Hum. Mut. 23, 186–192. 6. Tessitore, A., Di Rocco, Z. C., Cannita, K., Ricevuto, E., Toniato, E., Ficorella, C. et al. (2002) High sensitivity of detection of TP53 somatic mutations by fluorescenceassisted mismatch analysis. Genes Chromosomes Cancer 35, 86–91. 7. Bidichandani, S. I., Lanyon, W. G., Shiach, C. R., Lowe, G. D. and Connor, J. M. (1995) Detection of mutations in ectopic factor VIII transcripts from nine haemophilia A patients and the correlation with phenotype. Hum. Genet. 95, 531–538. 8. Rudzki, Z., Duncan, E. M., Casey, G. J. Neumann, M., Favaloro, E. J. and Lloyd, J. V. (1996) Mutations in a subgroup of patients with mild haemophilia A and a familial discrepancy between the one-stage and two-stage factor VIII:C methods. Br. J. Haematol. 94, 400–406. 9. Montandon, A. J., Green, P. M., Giannelli, F. and Bentley, D. R. (1989) Direct detection of point mutations by mismatch analysis: application to haemophilia B. Nucleic Acids Res. 11, 3347–3358.
10. Verpy, E., Biasotto, M., Brai, M., Misiano, G., Meo, T. and Tosi, M. (1996) Exhaustive mutation scanning by fluorescence-assisted mismatch analysis discloses new genotypephenotype correlations in angiodema. Am. J. Hum. Genet. 59, 308–319. 11. Forrest, S. M., Dahl, H. H., Howells, D. W., Dianzani, I. and Cotton, R. G. (1991) Mutation detection in phenylketonuria by using chemical cleavage of mismatch: importance of using probes from both normal and patient samples. Am. J. Hum. Genet. 49, 175–183. 12. Liddell, M. B., Bayer, A. J. and Owen, M. J. (1995) No evidence that common allelic variation in the Amyloid Precursor Protein (APP) gene confers susceptibility to Alzheimer’s disease. Hum. Mol. Genet. 4, 853–858. 13. De Galitiis, F., Cannita, K., Tessitore, A., Martella, F., Di Rocco, Z. C., Russo, A. et al. (2006) Novel P53 mutations detected by FAMA in colorectal cancers. Ann. Oncol. 17, vii78–vii83. 14. Sallmann, G.B., Bray, P. J., Rogers, S., Quince, A., Cotton, R. G. and Carden, S. M. (2006) Scanning the ocular albinism 1 (OA1) gene for polymorphisms in congenital nystagmus by DHPLC. Ophthalmic Genet. 27, 43–49. 15. Sambrook, J., Fritsch, E. F. and Maniatis, T. (1989) Molecular cloning: a laboratory manual. 2nd ed. New York: Cold Spring Harbor Laboratory Press.
Chapter 15 Mismatch Oxidation Assay: Detection of DNA Mutations Using a Standard UV/Vis Microplate Reader Tania Tabone, Georgina Sallmann, and Richard G. H. Cotton Abstract Simple, low-cost mutation detection assays that are suitable for low-throughput analysis are essential for diagnostic applications where the causative mutation may be different in every family. The mismatch oxidation assay is a simple optical absorbance assay to detect nucleotide substitutions, insertions, and deletions in heteroduplex DNA. The method relies on detecting the oxidative modification products of mismatched thymine and cytosine bases by potassium permanganate as it is reduced to manganese dioxide. This approach, unlike other methods commonly used to detect sequence variants, does not require costly labeled probes or primers, toxic chemicals, or a time-consuming electrophoretic separation step. The oxidation rate, and hence the presence of a sequence variant, is detected by measuring the formation of the potassium permanganate reduction product (hypomanganate diester), which absorbs at the 420-nm visible wavelength, using a standard UV/vis microplate reader. Key words: Heteroduplex analysis, mismatch oxidation, single nucleotide polymorphism, mutation detection, spectroscopy, colorimetric assay.
1. Introduction Simple, low-cost methods for the identification of DNA mutations and polymorphisms are essential for the detection and diagnosis of inherited human diseases and basic molecular biology research. While Sanger sequencing represents the ‘‘gold standard’’ for mutation detection and is useful in confirming common mutations, it is expensive for routine use and may not readily identify previously unreported heterozygous changes. Thus, alternative screening methods are often used to examine large genes to reduce the sequencing burden and increase the overall sensitivity for mutation detection. A.A. Komar (ed.), Single Nucleotide Polymorphisms, Methods in Molecular Biology 578, DOI 10.1007/978-1-60327-411-1_15, ª Humana Press, a part of Springer Science+Business Media, LLC 2003, 2009
235
236
Tabone, Sallmann, and Cotton
Established scanning methods differ widely in their accuracy, speed, and cost for mutation detection and each has advantages and disadvantages depending on the resources of the laboratory and requirements of an individual application. Mutation scanning methods that exploit the use of heteroduplex DNA rely on detecting differences in the chromatographic behavior or electrophoretic migration of homoduplex (wild-type) DNA compared with heteroduplex (mutant) DNA samples. These methods include denaturing high performance liquid chromatography (1), denaturing gradient gel electrophoresis (2) (see Chapter 8 in this volume), conformation-sensitive gel electrophoresis (3), and constantdenaturing capillary electrophoresis (4). Limitations of these methods include cost, complexity and reliability or poor sensitivity, and time-consuming methodology. In addition, these methods as a whole commonly require the use of expensive DNA labels and toxic reagents. Here we describe a mutation detection method that is based on the principles of the chemical cleavage of mismatch (CCM) (5), known as the mismatch oxidation assay. CCM utilizes potassium permanganate (KMnO4) and hydroxylamine to modify mismatched DNA and an additional toxic chemical, piperidine, to cleave the modified products. However, as the mismatch oxidation assay is not based on a cleavage reaction to detect mismatched DNA, this approach eliminates the need for toxic chemicals and a time-consuming separation procedure. Instead, this approach relies on directly detecting the oxidative modification products of mismatched thymine (T) and cytosine (C) bases by KMnO4 (6, 7) using a simple microplate method to screen for mutations. Specifically, the technology exploits the selective oxidation of mismatched DNA by KMnO4 to detect mutations in PCR amplicons when present in heteroduplex (mutant) form compared with homoduplex (wild-type) controls. As mismatched T and C bases are oxidized, KMnO4, which is pink in color, is reduced to form a yellow hypomanganate diester intermediate (8). The rate of oxidation of the mismatched bases by KMnO4 and hence the presence of a sequence variant is readily detected by measuring the formation of the yellow hypomanganate diester intermediate, which absorbs at the 420-nm visible wavelength (9), using a standard UV/vis microplate reader (Fig. 15.1). Here we describe the protocol for detecting DNA mutations using KMnO4, a classic reagent for the oxidation of double bonds in pyrimidine bases that allows the development of a simple spectroscopic-based mutation detection assay. The method is not labor-intensive or time-consuming to perform and does not require any complicated preassay set-up or extensive optimization for mutation detection. Following completion of the assay, traces are visually inspected and compared with a wild-type reference
Mismatch Oxidation Assay
237
Fig. 15.1. The mismatch oxidation assay, demonstrating the detection of a T.C/A.G mismatch. The mismatched T and C residues are oxidized by potassium permanganate, which is readily detected by measuring the formation of the hypomanganate diester at 420-nm wavelength using a standard UV/vis microplate reader.
238
Tabone, Sallmann, and Cotton
sample to detect a sequence variant. The universal assay conditions allow a mixture of DNA samples to be analyzed in the same microplate, providing greater throughput for mutation detection.
2. Materials 2.1. Reagents
1. KMnO4 (Sigma-Aldrich, St. Louis, MO, USA). Prepare a 100 mM solution by dissolving 0.0165 g KMnO4 in 1 mL distilled H2O (dH2O). Vortex the solution vigorously for approximately 1 min to completely dissolve the KMnO4 crystals. Store the solution at –20C in 100-mL aliquots protected from light (see Note 1). The solution can be stored for up to 6 months. Prepare a fresh 10 mM working dilution by adding 900 mL dH2O to a 100-mL aliquot of 100 mM KMnO4 on the day of the experiment. 2. Tetraethylammonium chloride (TEACl) (Sigma-Aldrich). Prepare a 3 M solution by dissolving 100 g TEACl in dH2O to a final volume of 200 mL (see Note 2). Filter the solution through a 0.45-mm membrane and store the filtrate at room temperature (15–25C). The solution can be stored for up to 6 months. 3. Reconstitute lyophilized primers in dH2O to 100 mM and store the solution at –20C. Prepare a 10 mM working dilution in dH2O and store it at –20C in small aliquots to minimize freeze/thawing activity (see Note 3). 4. 5 U/mL Taq DNA polymerase (e.g., Invitrogen, Carlsbad, CA, USA). 5. 100 mM dNTP (Bioline, Alexandria, NSW, Australia). Dilute the dNTP to 2 mM working dilution in dH2O and store the solution at –20C in small aliquots to minimize freeze/thawing activity.
2.2. Equipment
1. PCR purification kit (e.g., QIAquick, Qiagen, Valencia, CA, USA). 2. Flat-bottom polystyrene 96-well clear microplates, nonsterile (e.g., Greiner Bio-One, Frickenhausen, Germany). 3. UV/vis spectrophotometer or microplate reader with temperature control (e.g., Varian Cary 50 scanning microplate reader system, Varian, Walnut Creek, CA, USA).
Mismatch Oxidation Assay
239
3. Methods 3.1. PCR Amplification
1. In thin-walled 0.2-mL PCR tubes or 96-well microplates, add the PCR components on ice in the following order: Component
X1 reaction
Final
dH2O (to 25 mL final volume)
17.5 mL
10X PCR buffer
2.5 mL
1X
2 mM dNTP
2.5 mL
0.2 mM
50 mM MgCl2
0.75 mL
1.5 mM
10 mM forward primer
0.5 mL
0.2 mM
10 mM reverse primer
0.5 mL
0.2 mM
100 ng/mL template DNA
0.5 mL
100 ng
5 U/mL Taq DNA polymerase
0.25 mL
2.5 U
2. Amplify the DNA in a thermal cycler using the following conditions: an initial DNA denaturation at 95C for 5 min, followed by 35 cycles consisting of denaturation at 95C for 20 s, annealing at the optimal primer melting temperature for 20 s, and extension at 72C for 20 s; and a final elongation at 72C for 10 min.
3.2. Duplex Formation (Performed as Two Separate Tests)
1. Combine the PCR products in 0.2-mL PCR tubes or 96-well microplates and pipette the solution gently up and down to mix it. Centrifuge the solution briefly to collect liquid at the bottom of the well. Reaction 1: To detect mutations in heterozygote DNA samples Test samples: 25 mL unknown PCR product Reference sample: 25 mL control (known wild-type) PCR product Reaction 2: To detect mutations in homozygote DNA samples Test samples: 12.5 mL unknown PCR product with 12.5 mL control (wild-type) PCR product Reference sample: 25 mL control (known wild-type) PCR product 2. Denature DNA samples at 95C for 10 min, then cool slowly them from 80 to 25C, decreasing the temperature by 1C every minute to promote heteroduplex formation in a thermal cycler (see Note 4).
240
Tabone, Sallmann, and Cotton
3.3. DNA Purification
1. Purify the PCR products to remove unincorporated primers and dNTPs using a standard ethanol-precipitation procedure (10) or using a commercial purification kit, such as the QIAquick PCR purification kit, according to the manufacturer’s instructions. Briefly, add 25 mL PCR product and 125 mL buffer PB (as supplied by the manufacturer) to a 1.2-mL microcentrifuge tube, vortex the solution, and transfer it to a Qiagen column. Centrifuge the column at 13,000 rpm for 2 min and discard the flow-through. To wash the DNA, add 750 mL of buffer PE (as supplied by the manufacturer) to the column and centrifuge the column at 13,000 rpm for 2 min. Discard the flow-through and centrifuge the column for an additional minute to remove all traces of buffer. 2. Place the QIAquick column into a clean 1.2-mL microcentrifuge tube. Add 30 mL dH2O to the membrane and the incubate the contents of the tube at room temperature (15– 25C) for 1 min. Elute the PCR product from the membrane by centrifuging the column at 13,000 rpm for 1 min (see Note 5). 3. Determine the DNA concentration by measuring the optical density at 260 nm. 4. Adjust the DNA concentration to 50 ng/mL with dH2O (see Note 3).
3.4. Mismatch Oxidation Assay
1. Set the microplate reader instrument to 25C constant temperature with measurements recorded at 420-nm wavelength at 1-min intervals for 180 min (see Note 6). 2. In a 96-well flat-bottom polystyrene microplate, add reagents in the following order; dH2O to a 200-mL final reaction volume, 67 mL of 3 M TEACl, 10 mL of 50 ng/mL purified PCR product, and 20 mL of 10 mM KMnO4 (see Note 7). 3. Pipette the solution up and down several times to mix it. 4. Insert the plate into the microplate reader and set the baseline (zero the instrument) and commence scanning immediately. 5. Determine selective rates of oxidation by KMnO4 of the heteroduplex and homoduplex DNA samples by measuring the rate of change in absorbance at 420 nm. The difference between the kinetics curves for each heteroduplex versus its corresponding homoduplex control is recorded by visual identification. Differences can be quantified by determining the total area under each curve using the microplate reader software application. An example of the results produced is shown in Fig. 15.2.
Mismatch Oxidation Assay
241
Fig. 15.2. Spectroscopic scan showing the oxidation rates of PCR products representing all four classes of mismatches after treatment with potassium permanganate. (a) G.T/A.C mismatch, (b) C.T/A.G mismatch, (c) C.C/G.G mismatch, (d) A.A/T.T mismatch. In each scan, the top trace represents the oxidation rate of the mismatched duplex and the bottom trace represents the oxidation rate of the matched duplex.
4. Notes
1. KMnO4 oxidizes rapidly when exposed to air and is lightsensitive. Minimize freeze/thawing by storing it in small aliquots at –20C and cover the tubes to protect from it light. 2. TEACl is extremely hygroscopic. The entire TEACl stock should be reconstituted to avoid it absorbing moisture from the atmosphere. 3. Primers and DNA should always be reconstituted and diluted in dH2O. Avoid buffers containing tris(hydroxymethyl)aminomethane and EDTA as these reagents react strongly with KMnO4 and interfere with the DNA oxidation signal. 4. Heteroduplex formation should be performed immediately before analysis. 5. When DNA is being eluted or resuspended after a PCR purification reaction, ensure all traces of ethanol are completely removed as ethanol reacts strongly with KMnO4 and may interfere with the DNA oxidation signal.
242
Tabone, Sallmann, and Cotton
6. The temperature of the plate reader should be set to room temperature or the plate reader should placed in a room with a constant temperature to ensure reproducible results. The oxidation rate of KMnO4 is increased at temperatures above 30C and decreased at temperatures below 15C. Fluctuations in temperature may affect the reproducibility of some results. 7. Commence scanning immediately after addition of KMnO4 to obtain an accurate background reading for blanking the plate reader as the oxidation of mismatched DNA begins.
Acknowledgements The authors wish to a thank Vincent Sesto for technical assistance with preparation of this manuscript and Sean Stockwell for critically reviewing the manuscript. This work was supported by an NH&MRC grant (to R.G.H.C.). References 1. Oefner, P. J. and Underhill, P. A. (1995) Comparative DNA sequencing by denaturing high-performance liquid chromatography (DHPLC). Am. J. Hum. Genet. 57, A266. 2. Myers, R. M., Maniatis, T. and Lerman, L. S. (1987) Detection and localization of single base changes by denaturing gradient gel electrophoresis. Meth. Enzymol. 155, 501–527. 3. Ganguly, A., Rock, M. J. and Prockop, D. J. (1993) Conformation-sensitive gel electrophoresis for rapid detection of single-base differences in double-stranded PCR products and DNA fragments: evidence for solvent-induced bends in DNA heteroduplexes. Meth. Enzymol. 155, 501–527. 4. Khrapko, K., Hanekamp, J. S., Thilly, W. G., Belenkii, A., Foret, F. and Karger, B. L. (1994) Constant denaturing capillary electrophoresis (CDCE): a high-resolution approach to mutational analysis. Nucleic Acids Res. 22, 364–369. 5. Cotton, R. G., Rodrigues, N. R. and Campbell, R. D. (1988) Reactivity of cytosine and thymine in single-base-pair mismatches with hydroxylamine and osmium tetroxide and its application to the study of mutations.
6.
7.
8.
9.
10.
Proc. Natl. Acad. Sci. U.S.A. 85, 4397–4401. Hayatsu, H., Atsumi, G., Nawamura, T., Kanamitsu, S., Negishi, K. and Maeda, M. (1991) Permanganate oxidation of nucleic acid components: a reinvestigation. Nucleic Acids Symp. Ser. 25, 77–78. Hayatsu, H. (1996) The 5,6-double bond of pyrimidine nucleosides, a fragile site in nucleic acids. J. Biochem. 119, 391–395. Freeman, F., Fuselier, C. O. and Karchefski, E. M. (1975) Permanganate ion oxidation of thymine: spectrophotometric detection of a stable organomanganese intermediate. Tet. Letters. 25, 2133–2136. Tabone, T., Sallmann, G., Webb, E. and Cotton, R. G. H. (2006) Detection of 100% of mutations in 124 individuals using a standard UV/Vis microplate reader: a novel concept for mutation scanning. Nucleic Acids Res. 34, e45. Sambrook, J., Fritsch, E. F. and Maniatis, T. (1989) Molecular cloning: a laboratory manual. 2nd ed. New York: Cold Spring Harbor Laboratory Press.
Chapter 16 High-Throughput Methods for SNP Genotyping Chunming Ding and Shengnan Jin Abstract Single nucleotide polymorphisms (SNPs) are ideal markers for identifying genes associated with complex diseases for two main reasons. Firstly, SNPs are densely located on the human genome at about one SNP per approximately 500–1,000 base pairs. Secondly, a large number of commercial platforms are available for semiautomated or fully automated SNP genotyping. These SNP genotyping platforms serve different purposes since they differ in SNP selection, reaction chemistry, signal detection, throughput, cost, and assay flexibility. This chapter aims to give an overview of some of these platforms by explaining the technologies behind each platform and identifying the best application scenarios for each platform through cross-comparison. The readers may delve into more technical details in the following chapters. Key words: Whole genome association, fine mapping, single nucleotide polymorphism, copy number variation, haplotyping.
1. Introduction Single nucleotide polymorphisms (SNPs) are best known as genetic markers in disease-association studies to identify genes associated with complex diseases (1,2). However, SNPs are also used in many other clinically and biologically important applications (3). A large variety of commercial platforms are available for semiautomated or fully automated SNP genotyping analysis. On the basis of the purposes of the study, SNP genotyping can be divided into two domains: whole genome association (WGA) and fine mapping (Fig. 16.1). Most of the genotyping platforms can be classified accordingly. This chapter aims to briefly explain the principles behind various platforms which lead to a comparison of these platforms so that the readers will get a quick overview before delving into the technical details of some of these methods in the following chapters. A.A. Komar (ed.), Single Nucleotide Polymorphisms, Methods in Molecular Biology 578, DOI 10.1007/978-1-60327-411-1_16, ª Humana Press, a part of Springer Science+Business Media, LLC 2003, 2009
245
246
Ding and Jin SNP Number
Cost Increase
1M
Affymetrix SNP Array 6.0 Illumina 1M-DUO
100K
WGA
Illumina iSelect BeadArray Affymetrix Custom Array
10K 1000
MassArray SNPlex SNPstream
100
Fine Mapping
Taqman OpenArray
10
1
10
100
1000
10000
Sample Size
Fig. 16.1. An overview of platforms with regard to throughput of single nucleotide polymorphisms and sample size. Platforms are selected on the basis of reasonable running costs.
2. Chemistries and Detection Methods for SNP Genotyping
Over the years, a number of chemistries were developed for distinguishing two alleles of a SNP. The key for their adoption in high-throughput studies is dependent on the suitability for automation. An ideal chemistry has to be universally applicable to any SNP (or to a substantial proportion of all human SNPs). Additionally, high automation demands minimum steps in genotyping. It may be fair to say that no single SNP genotyping platform is good enough to serve all purposes. Generally, the chemistries for SNP genotyping can be roughly divided into two types based on the key reaction allowing for the SNP detection: (1) nonenzymatic differential hybridization (see Chapters 18 and 19 in this volume); (2) enzymatic reactions (see Chapter 23 in this volume). Differential hybridization relies on different melting temperatures for matched and mismatched probes binding to the target DNA sequences. The Affymetrix SNP microarray employs this principle. For each SNP, four to six probes (25mers each) are used. Affymetrix arrays can achieve very high density to accommodate millions of probes on a single chip. The newest Affymetrix Human SNP Array 6.0 contains probes for 906,600 SNPs and an additional 946,000 probes for assessing copy number variations (CNVs). All the few million probes will be hybridized to their target sequences under the same temperature and buffer condition for the same amount of time, which is ideal for automated high-throughput SNP genotyping. However, the probes have to be effective in
High-Throughput Methods for SNP Genotyping
247
differentiating matched and mismatched targets. The probe sequences are determined by the local SNP sequences. Consequently, certain SNPs with ‘‘odd’’ local sequences cannot be selected, even if they are crucial tagging SNPs, SNPs in regulatory regions, or SNPs that can change protein coding sequences (see Note 1). Another example of differential hybridization is the TaqMan SNP assay (see Chapters 18 and 19 for details). For each SNP, two TaqMan probes specific for each allele are used. These two probes carry different fluorescent dyes. The presence of an allele (or both alleles for heterozygotes) is detected by the corresponding fluorescence signal(s) generated via 5’-exonuclease cleavage of the probe(s). The main drawback for the TaqMan SNP assay is its incapability to achieve even a very modest multiplex level. However, collaboration between Applied Biosystems and BioTrove (with their OpenArray platform) has enabled 3,072 TaqMan reactions (each reaction has a volume of only 33 nL) on a single slide. This platform may be particularly powerful when an extremely high number of samples is tested. Biomark (Fluidigm) is another system capable of miniaturized TaqMan assays to enable high throughput genotyping. For SNP genotyping based on enzymatic selectivity, there are mainly two types of assays. The first one is the primer extension (or single base extension, or minisequencing; see Chapter 23 in this volume). An extension primer annealing to the 50 end of a SNP site is extended by one or just a few bases. SNP calling is based on either the incorporated fluorescent nucleotide (SNPstream) or the extension product molecular weight (MassArray iPlex Gold assay). These assays provide a low background noise since the enzymatic fidelity in incorporating the right nucleotide is extremely high. The second one is based on DNA ligation. Molecular inversion probe technology (4) developed by ParAllele Biosciences (now part of Affymetrix, and used in the Affymetrix GeneChip custom SNP kits) is one example. Another example is SNPlex (Applied Biosystems). SNPlex achieves up to 48-plex by including a series of unique ZipCodeTM sequences in the allele-specific probes. The corresponding ZipChuteTM probes of different lengths hybridize to the ZipCodeTM sequences, and are subsequently separated and detected by capillary electrophoresis. In general, differential hybridization based platforms rely entirely on hybridization thermodynamic difference between matched and mismatched pairing of probes and targets. The selection of analyzable SNPs is highly dependent on the local SNP sequence. Enzymatic selectivity based platforms are less dependent on SNP local sequences and are likely to be applicable to more SNPs. However, there are often more steps involved in SNP analysis, making full automation more complicated.
248
Ding and Jin
3. Genotyping Platforms 3.1. Genotyping Platforms for WGA Studies
In earlier WGA studies, it was quite common that fewer than 100,000 SNPs were analyzed, since the cost was too high to include more SNPs. However, the paradigm has shifted significantly, thanks to (1) detailed HapMap data guiding the selection of tagging SNPs, and (2) vastly improved ultrathroughput (in terms of SNP number, see Note 2) genotyping platforms. At the moment, the Illumina BeadArray (newest version, High Density Human 1 M-Duo) and the Affymetrix SNP microarray (newest version, Human SNP Array 6.0) are the most widely used platforms in WGA studies. Although both are named as ‘‘array’’ and have similar throughput, these two platforms differ substantially in many aspects. First of all, they use different methods for discriminating the two alleles of a SNP. The Affymetrix microarray technology uses differential hybridization between a set of 25-mer probes matching to one of the two SNP alleles. As discussed earlier, this may limit the selection of SNPs. However, since the human genome contains over five million SNPs, the Affymetrix SNP array can still include close to one million SNPs. The Illumina BeadArray technology uses primer extension to distinguish the two SNP alleles. Theoretically, the enzymatic fidelity in primer extension to distinguish the two SNP alleles is extremely high, regardless of local SNP sequences. Thus, BeadArray may be less limited in SNP selection. However, extra steps of primer extension and staining must be carried out before signals can be detected. Another important difference between the two platforms is the selection of SNPs. The Illumina system places more emphasis on tagging SNPs than the Affymetrix system. This may be due to the two constrains imposed on the Affymetrix system: (1) SNP local sequence content suitable for the universal hybridization condition; (2) a complexity-reduction step through selectively amplifying 200–1,100-bp fragments generated by restriction enzyme digestion. However, whether a strictly tagging SNP based selection approach is superior to a hybrid selection approach (half tagging, half random SNPs) is still being debated. Rigorous comparison is not likely to be carried out given the prohibitive cost. Additionally, it is still not entirely clear how important are the SNPs that are not in the typical haplotype blocks for identifying genes associated with complex diseases. At any rate, with more SNPs detectable on a single chip, we may be able to analyze a sufficient number of tagging and random SNPs simultaneously. There are other technical differences that may not be relevant to the end users. For example, the Illumina BeadArray layout is unique for each chip. A decoding step is needed to determine
High-Throughput Methods for SNP Genotyping
249
geometrically how the beads specific for the SNPs are arranged on the chip. The Affymetrix SNP array uses 25-mers for SNP calling via differential hybridization, while the Illumina BeadArray uses 50-mers for target capture and primer extension via hybridization. 3.2. Genotyping Platforms for Fine Mapping
Fine mapping here is defined as SNP genotyping analysis at a high density for selective genomic regions. Fine mapping often follows large-scale WGA studies to zoom into potential genes associated with the disease of interest. Fine mapping studies differ from WGA studies dramatically in many aspects, notably: 1. Many fewer SNPs (e.g., fewer than 1,000) are genotyped. 2. Such SNPs will be highly dependent on a particular disease of interest. Although one SNP array (Illumina, Affymetrix, or others) can be used for WGA studies of any disease, SNPs selected for fine mapping of one disease are likely to be mostly different from those selected for fine mapping of another disease 3. Fine mapping may involve a larger sample size. In summary, fine mapping will require the genotyping of fewer (fewer than 1,000) SNPs highly specific for each disease for a larger sample size. Once a WGA study has been done and potential targets have been identified, fine mapping should be performed immediately. Additionally, since potentially any SNP can be directly disease causing, it is essential to achieve a high call rate (call rate is defined as the success rate for correctly genotyping the entire SNP panel). Additionally, cost is also an issue to consider (see Note 3). For these reasons, a good genotyping platform for fine mapping should achieve a high call rate for all selected SNPs, without time-consuming assay optimization processes, and at a relatively high multiplex level (e.g., more than 24 SNPs for each individual reaction). SNP calling based entirely on differential hybridization is unlikely to be highly successful in fine mapping. It may be very difficult if one needs to design discriminating probes for all 1,000 selected SNPs as the local sequences of these SNPs may have very different thermodynamic profiles (see Note 4). Possibly for this reason, Affymetrix acquired ParAllele Biosciences for its molecular inversion probe technology for custom SNP genotyping arrays. The custom SNP genotyping arrays do not rely on differential hybridization for SNP calling. Primer extension and allele-specific ligation-based platforms are more suitable for fine mapping applications. A number of commercial platforms are available (Table 16.1). Since systematic and direct comparison of these platforms is not available, we will have to rely on company application notes and publications reporting use of each technology for a rough comparison.
Illumina
Affymetrix (ParAllele)
Sequenom
Applied Biosystems
Beckman Coulter (Orchid)
Applied Biosystems and BioTrove
iSelect BeadArray
GeneChip custom SNP kits
MassArray
SNPlex
SNPstream
TaqMan OpenArray TaqMan
Single base extension
Allele-specific ligation
Single base extension
Molecular inversion probe (primer extension and ligation)
Single base extension or allele-specific primer extension
Chemistry
Fluorescence
Fluorescence
Probe size (capillary electrophoresis)
Molecular weight (Mass Spec)
Fluorescence
Fluorescence
Detection
384
48
64a
a
96 or 384
384
1
12
Number of samples
12 or 48
Up to 48
Up to 40
3,000, 5,000, or 10,000
Up to 60,800
Number of SNPs
SNP single nucleotide polymorphism Not true multiplexing, 64 uniplex TaqMan SNP assays in 64 different nano holes.
Provider
Platform
Table 16.1 Comparison of fine-mapping genotyping platforms
Call rate at higher plex level may be low
Design and delivery take at least 3 months
Design and delivery take at least 3 months
Note
http://www.biotrove.com/products/ open_array/snp/index.asp
http://www.beckmancoulter.com/ products/instrument/geneticanalysis/ ceq/genomelab_snpstream_dcr.asp
https://products.appliedbiosystems.com/ ab/en/US/adirect/ ab?cmd=catNavigate2&catID=600763
http://www.sequenom.com/GeneticAnalysis/Applications/iPLEXGenotyping/iPLEX-Overview.aspx
http://www.affymetrix.com/products /reagents/specific/custom_snp.affx
http://www.illumina.com/pages.ilmn? ID=158
URL
250 Ding and Jin
High-Throughput Methods for SNP Genotyping
251
Two platforms actually significantly surpass the arbitrary 1,000 SNPs cutoff mentioned earlier. The Illumina iSelect BeadArray uses single base extension, the same underlying chemistry and detection as the High Density Human 1 M-Duo array, for genotyping up to 60,800 user-selected SNPs from 12 samples on a single chip. The Affymetrix GeneChip custom SNP kits use the molecular inversion probe technology acquired from ParAllele Biosciences. These custom arrays can analyze 3,000, 5,000, or 10,000 user-selected SNPs for a single sample. One drawback for these two platforms is the turnaround time, since at least 3 months is required for assay designs and array delivery. For a typical fine mapping project following a WGA study, it might not be necessary to analyze tens of thousands of SNPs. Thus, a higher sample number throughput at a reasonable SNP number throughput (fewer than 1,000 SNPs) may be preferred. To this end, a few platforms are great choices for fine mapping, including the MassArray system (Sequenom) (see Chapter 20 in this volume), SNPlex (Applied Biosystems), and SNPstream (Beckman Coulter, in collaboration with Orchid Cellmark). These platforms can all achieve multiplex genotyping at 20-plex or more routinely for 96 or 384 different reactions on a single plate. They are highly flexible in several ways. Firstly, the throughput of SNP number and sample size can be balanced at the users’ discretion. Secondly, the turnaround time for assay design and delivery of reagents is much faster than the custom arrays from Illumina and Affymetrix (Table 16.1). Failed SNP assays can be redesigned and reordered quickly. Unless the SNP number to be analyzed is well above 1,000, these platforms may be the first choices.
4. New Advances and Other Outstanding Issues
There are at least two exciting features about genomic research. One is the constant development of better and more affordable technologies (just like personal computers). The other feature is the acquisition of new insights into gene structure and function. One such example is the CNVs. CNVs are much less frequently found in the human genome than SNPs, with probably around a few thousand to tens of thousand CNVs in the entire human genome. However, these variations involve much larger DNA segments, ranging from a few kilobases to a few megabases (5). Their importance in human health is manifested by a number of diseases, such as CHARGE syndrome (6) and Parkinson’s disease (7).
252
Ding and Jin
The platform suppliers have taken notice of the importance of CNVs. Both the Affymetrix Human SNP Array 6.0 and the Illumina High Density Human 1 M-Duo offer good coverage for CNV analysis. For example, the Human SNP Array 6.0 targets 3,182 distinct, nonoverlapping segments with on average 61 probe sets per region. Earlier versions of these platforms have been used for CNV analysis (8– 11). It is foreseeable that CNV analysis will be part of most, if not all, WGA studies. Other platforms are likely to follow the trend. Given the limited number of CNVs in the human genome, fine mapping genotyping platforms may also be useful for validation studies. For example, the MassArray iPlex platform will launch the CNV genotyping application by 2008. Serious limitations in SNP genotyping are still present though. At least two of them are worth mentioning. The first one is SNP coverage for different ethnic groups. The statistics provided by the best WGA platforms are based on a very limited number of ethnic groups. For example, CHB (Han Chinese in Beijing) is not likely to represent all people in China, given that there are 56 distinct ethnic groups in China. It may be necessary to include more SNPs for better coverage of other ethnic groups. Another limitation is on haplotype analysis. All the platforms mentioned in this chapter, when used in their standard format, cannot achieve direct molecular haplotyping. Instead, statistical methods are used to infer haplotype information. Ultimately, the best solution to all the issues mentioned above, especially related to better and robust identification of the genes associated with complex diseases, may come from the fourthgeneration (see Note 5 and Chapter 5 in this volume), probably single molecule based, capable of sequencing the human genome for less than US $1,000.
5. Summary Scientists and engineers have come a long way developing a wide selection of SNP genotyping platforms. It is now prime time to carry out WGA studies to identify genes associated with complex diseases, potentially yielding biomarkers for disease diagnosis and prognosis, and targets for drug development. Both a WGA platform and a fine mapping platform may be needed for a comprehensive study. The technology will continue to be improved to include more SNPs. New technology (e.g., for whole genome sequencing at low cost; see also Chapters 5 and 6 in this volume) will likely appear in the next 5–10 years and a paradigm shift in WGA studies may happen then.
High-Throughput Methods for SNP Genotyping
253
6. Notes
1. At a fixed hybridization temperature, robust differential hybridization may not be achieved for matched and mismatched targets if a local SNP sequence has very high or very low GC content. 2. Throughput is often defined by the number of SNPs that can be genotyped in one run, but this might not be entirely accurate as in many situations (particularly when SNPs are served as biomarkers for molecular diagnosis) DNA sample throughput may be more important. 3. Genotyping 1,000 SNPs for 2,000 samples (a total of two million SNP genotyping assays) is a lot more costly than genotyping one million SNPs for two samples. In addition, since the 1,000 SNPs are highly dependent on the disease of interest, custom designs and even assay optimization are needed, which further adds to the cost and time. 4. To design a hybridization-based SNP microarray for the selected 1,000 SNPs is a lot more difficult than for a panel of any 1,000 SNPs. For the latter, the designer can choose any 1,000 SNPs from more than five million SNPs available by selecting those SNPs located in sequences with similar thermodynamic profiles. 5. We consider the slab gel sequencing the first generation, capillary sequencing the second generation, and the Roche 454, Illumina Genome Analyzer, and Applied Biosystems SOLiD platforms as the third generation.
Acknowledgements C.D. is supported by the Stanley Ho Centre for Emerging Infectious Diseases and the Li Ka Shing Institute of Health Sciences. References 1. Glazier, A. M., Nadeau, J. H. and Aitman, T. J. (2002) Finding genes that underlie complex traits. Science 298, 2345–2349. 2. Becker, K. G., Barnes, K. C., Bright, T. J. and Wang, S. A. (2004) The genetic association database. Nat. Genet. 36, 431–432. 3. Ding, C. (2007) ’Other’ applications of single nucleotide polymorphisms. Trends Biotechnol. 25, 279–283.
4. 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. and Davis, R. W. (2003) Multiplexed genotyping with sequence-tagged molecular inversion probes. Nat. Biotechnol. 21, 673–678. 5. Iafrate, A. J., Feuk, L., Rivera, M. N., Listewnik, M. L., Donahoe, P. K., Qi, Y.,
254
Ding and Jin
Scherer, S. W. and Lee, C. (2004) Detection of large-scale variation in the human genome. Nat. Genet. 36, 949–951. 6. Jongmans, M. C., Admiraal, R. J., van der Donk, K. P., Vissers, L. E., Baas, A. F., Kapusta, L., van Hagen, J. M., Donnai, D., de Ravel, T. J., Veltman, J. A., Geurts van Kessel, A., De Vries, B. B., Brunner, H. G., Hoefsloot, L. H. and van Ravenswaaij, C. M. (2006) CHARGE syndrome: the phenotypic spectrum of mutations in the CHD7 gene. J. Med. Genet. 43, 306–314. 7. Singleton, A. B., Farrer, M., Johnson, J., Singleton, A., Hague, S., Kachergus, J., Hulihan, M., Peuralinna, T., Dutra, A., Nussbaum, R., Lincoln, S., Crawley, A., Hanson, M., Maraganore, D., Adler, C., Cookson, M. R., Muenter, M., Baptista, M., Miller, D., Blancato, J., Hardy, J. and Gwinn-Hardy, K. (2003) alpha-Synuclein locus triplication causes Parkinson’s disease. Science 302, 841. 8. Redon, R., Ishikawa, S., Fitch, K. R., Feuk, L., Perry, G. H., Andrews, T. D., Fiegler, H., Shapero, M. H., Carson, A. R., Chen, W., Cho, E. K., Dallaire, S., Freeman, J. L., Gonzalez, J. R., Gratacos, M., Huang, J., Kalaitzopoulos, D., Komura, D., MacDonald, J. R., Marshall, C. R., Mei, R., Montgomery, L., Nishimura, K., Okamura, K., Shen, F., Somerville, M. J., Tchinda, J., Valsesia, A., Woodwark, C., Yang, F., Zhang, J., Zerjal, T., Armengol, L., Conrad, D. F., Estivill, X., Tyler-Smith, C., Carter, N. P., Aburatani, H.,
Lee, C., Jones, K. W., Scherer, S. W. and Hurles, M. E. (2006) Global variation in copy number in the human genome. Nature 444, 444–454. 9. Bae, J. S., Cheong, H. S., Kim, J. O., Lee, S. O., Kim, E. M., Lee, H. W., Kim, S., Kim, J. W., Cui, T., Inoue I., and Shin, H. D. (2008) Identification of SNP markers for common CNV regions and association analysis of risk of subarachnoid aneurysmal hemorrhage in Japanese population. Biochem. Biophys. Res. Commun. 373, 593–596. 10. Blauw, H. M., Veldink, J. H., van Es, M. A., van Vught, P. W., Saris, C. G., van der Zwaag, B., Franke, L., Burbach, J. P., Wokke, J. H., Ophoff, R. A. and van den Berg, L. H. (2008) Copy-number variation in sporadic amyotrophic lateral sclerosis: a genome-wide screen. Lancet Neurol. 7, 319–326. 11. Gunnarsson, R., Staaf, J., Jansson, M., Ottesen, A. M., Goransson, H., Liljedahl, U., Ralfkiaer, U., Mansouri, M., Buhl, A. M., Smedby, K. E., Hjalgrim, H., Syvanen, A. C., Borg, A., Isaksson, A., Jurlander, J., Juliusson, G. and Rosenquist, R. (2008) Screening for copy-number alterations and loss of heterozygosity in chronic lymphocytic leukemia – a comparative study of four differently designed, high resolution microarray platforms. Genes Chromosomes Cancer 47, 697–711.
Chapter 17 High-Throughput SNP Genotyping: Combining Tag SNPs and Molecular Beacons Luis B. Barreiro, Ricardo Henriques, and Musa M. Mhlanga Abstract In the last decade, molecular beacons have emerged to become a widely used tool in the multiplex typing of single nucleotide polymorphisms (SNPs). Improvements in detection technologies in instrumentation and chemistries to label these probes have made it possible to use up to six spectrally distinguishable probes per reaction well. With the remarkable advances made in the characterization of human genome diversity, it has been possible to describe empirical patterns of SNPs and haplotype variation in the genome of diverse human populations. These patterns have revealed that the human genome is structured in blocks of strong linkage disequilibrium (LD). Because SNPs tend to be in LD with each other, common haplotypes share common SNPs and thus the majority of the diversity in a region can be characterized by typing a very small number of SNPs; so-called tag SNPs. Herein lies the advantage of the multiplexing ability of molecular beacons, since it becomes possible to use as few as 30 probes to interrogate several haplotypes in a highthroughput approach. Thus, through the combined use of tag SNPs and molecular beacons it becomes possible to type individuals for clinically relevant haplotypes in a high-throughput manner at a cost that is orders of magnitude less than that for high throughput sequencing methods. Key words: Linkage disequilibrium, single nucleotide polymorphism, tagging single nucleotide polymorphisms, DC-SIGN, Mycobacterium tuberculosis, molecular beacons, real-time PCR.
1. Introduction Sanjay Tyagi the inventor of molecular beacons (1) once wrote: Imagine that you have a magic reagent to which you add a droplet of a body fluid from a patient; you wait for a moment and a glow appears in the tube holding the mixture; the glow not only tells you which pathogen is responsible for the patient’s illness, but also indicates which drugs to use to A.A. Komar (ed.), Single Nucleotide Polymorphisms, Methods in Molecular Biology 578, DOI 10.1007/978-1-60327-411-1_17, ª Humana Press, a part of Springer Science+Business Media, LLC 2003, 2009
255
256
Barreiro, Henriques, and Mhlanga
treat the disease. Also imagine that you can perform this diagnosis before any symptoms of the disease appear, improving the chances of success with the treatment, and you can perform this test on a large population with ease. The creation and development of such reagents are the promise of nucleic acid-based detection and are the aspiration of a diverse community of researchers (2). The promise of the technologies evoked by Sanjay Tyagi is borne out in the above quotation. The sequencing of the human genome (3) furnished an unprecedented understanding of its structure and organization, but could not in itself account for human biological variation. To address the latter, a number of international consortiums or private corporations, such as the International SNP Map Working Group, SeattleSNPs PGA, and the Perlegen consortium, have multiplied efforts to resequence genes or genomic regions to characterize single nucleotide polymorphism (SNP) variations in the human genome (4–6). To date, more than 11 million SNPs have been recorded in dbSNP, the public repository for DNA variation data (http://www.ncbi.nlm. nih.gov/SNP/index.html) (see Chapter 3 for details). Decorating the human genome at a frequency of one in every 500–1,000 bp, they are the most common form of human variation and can serve as high-resolution genetic markers. This variation, which represents a legacy of our evolutionary past and in the future may be a treasure trove of information paving the way to personalized medicine, may at least partially explain the wide range of phenotypic differences observed among individuals and populations (7–9). These catalogues of sequence variation therefore provide scientists and clinicians with the precious raw material to be exploited in both human evolutionary studies and medically related research. Here the major challenges have been in devising and implementing cost-effective, easily accessible, and rapid molecular diagnostic methods that can interrogate anywhere from a few dozen to hundreds of thousands of polymorphisms. The comparison of these SNPs among large numbers of individuals can be used in therapy and drug design and even in devising new, more powerful approaches in cell-based screening approaches for drug discovery. It is these diverse and complicated needs that have driven the creation of high-throughput methods of SNP typing. Once genome sequence diversity has been catalogued, the next step is to determine how this diversity is organized within the human genome. Eleven million SNPs discovered to date appear to be not entirely random. When a new mutation arises, it is associated with neighboring variants present on the same chromosome or haploid DNA molecule, forming what is commonly known as a ‘‘haplotype.’’ When two alleles lying on the same chromosome are always observed together, or at least more often
High-Throughput SNP Genotyping
257
than expected by chance, these two variants are said to be in linkage disequilibrium (LD). The HapMap project, a natural extension of the Human Genome Project, was a pioneer in describing empirically the patterns of SNP and haplotype variation in the human genome and in obtaining a general LD map in populations of different ethnic origins (10). HapMap data clearly demonstrate that the human genome is organized in a LD blocklike structure and that these LD blocks are often disrupted by recombination hotspots (11, 12). When SNPs are in LD with each other, redundant information is contained within the haplotype (i.e., by knowing the marker at one locus, we can predict the marker that will occur at the linked loci nearby). Thus, when one infers haplotypes within a region of reasonable LD, the diversity of haplotypes is accounted for by a few common haplotypes and lots of rare ones. The common haplotypes will share a number of SNPs in common with each other, whereas the rarer haplotypes will be characterized by carrying the rarer alleles at certain loci. Thus, one can capture the majority of the diversity within a region by typing those SNPs which allow one to cover the most diversity; so-called tag SNPs. Currently, HapMap phase II provides the most complete available resource for selecting tag SNPs genomewide (12). Importantly, tag SNPs defined on the basis of the HapMap populations have been shown to adequately capture patterns of variation in other human groups; tag SNPs are therefore highly ‘‘portable’’ (13–15). In the practical sense, the HapMap data have already proven to be useful, as attested by the increasing number of successful genomewide association studies on diseases as diverse as type 1 (16, 17) and type 2 (16, 18, 19) diabetes, coronary artery disease (20), obesity-related traits (21, 22), rheumatoid arthritis (16, 23), and human immunodeficiency virus (HIV) disease progression (24). The portability and utility of tag SNPs opens up the possibility of their usage in ‘‘lower’’ highthroughput methods that are cheaper to implement and broadly accessible. Indeed, with a wide range of relatively cheap and robust instruments (see Table 17.1) and multiplexing probes such as molecular beacons, cost-effective high-throughput SNP typing becomes a reality (see Fig. 17.1). Two principal obstacles must be overcome in the detection and analysis of SNPs. The first is the small amounts of nucleic acid present in clinical specimens. This can be overcome by use of differing nucleic acid amplification strategies, most notably polymerase chain reaction (PCR). This and other methods such as nucleic acid sequence based amplification allow the selective amplification and enrichment of a locus of interest by severalthousand-fold over other nucleic acid sequences present (25). The second obstacle is unambiguous detection of the SNP. Herein lies an intrinsic property of nucleic acid chemistry that can be
Model
7300 realtime PCR system
7500 realtime PCR system
PRISM 7700 and 7900HT
StepOne realtime PCR system
MiniOpticon
Chromo 4
ICycler IQ5
SmartCycler II
Rotor-Gene 6000
Company
Applied Biosystems
Applied Biosystems
Applied Biosystems
Applied Biosystems
Bio-Rad
Bio-Rad
Bio-Rad
Cepheid
Corbett Research
FAM, TET, HEX, TMR, ROX, and Texas red
ABLL
LED
LED
THL
LED
LED
CPM, FAM, TET, Texas red, Cy5, and LightCycler Red 705
FAM, Cy3, Texas red, and Cy5
FAM, HEX, TMR, Texas red, and Cy5
FAM, TMR, Texas red, and Cy5
FAM and HEX
FAM, HEX, and ROX
FAM, TET, TMR, Texas red, and Cy5
THL
LED
FAM, TET, TMR, and Texas red
Fluorophore choicea
THL
Excitation source
Table 17.1 Specifications of spectrofluorometric thermal cyclers
6 targets
4 targets
5 targets
4 targets
2 targets
3 targets
6 targets
5 targets
4 targets
Multiplex capability
b,c
72 wellsc
16 units
96 wells
96 wells
All types
All types, less suited for adjacent probes
All types
All types, less suited for adjacent probes
All types, less suited for adjacent probes
All types
48 wellsC
48 wells
All types
All types, less suited for adjacent probes
All types, less suited for adjacent probes
Hybridization probe compatibility
96 wells and 384 wells
96 wells
96 wells
Sample capacity
258 Barreiro, Henriques, and Mhlanga
Mastercycler realplex 4
R.A.P.I.D.
LightCycler 1.5
LightCycler 2.0
LightCycler 480
Mx3000P
Mx3005P
Eppendorf
Idaho Technologies
Roche Applied Science
Roche Applied Science
Roche Applied Science
Stratagene
Stratagene
THL
THL
XL
LED
LED
LED
LED
LED
Best suited for adjacent probes and WS-MB probes
32 wellsc
6 targets
6 targets
4 targets 5 targets
FAM, TMR, Texas red, and Cy5d FAM, HEX, TMR, Texas red, and Cy5d
96 wells
96 wells
All types
All types
All types
Best suited for adjacent probes and WS-MB probes
32 wellsc
3 targets
96 wells and 384 wellsc
All types
3 targets
32 wellsc
All types, less suited for adjacent probes All types, less suited for adjacent probes
96 wells 96 wells
4 targets
2 targets
CPM, FAM, HEX, LightCycler Red 610, LightCycler Red 640, and Cy5
FAM, HEX, LightCycler Red 610, LightCycler Red 640, LightCycler Red 670, and LightCycler red 705
FAM, LightCycler Red 640, and LightCycler Red 705
FAM, LightCycler Red 640, and LightCycler Red 705
FAM, TET, TMR, and Texas red
FAM and HEX
Modified from (39) THL tungsten-halogen lamp, ABLL argon blue-light laser, LED light-emitting diode, XL xenon lamp, WS-MB wavelength-shifting molecular beacon. a Refer to Table 17.2 for alternative fluorophores. b Each unit is independently programmable. c Rapid cycle capabilities d Alternative preinstalled excitation and emission filter sets are available.
Mastercycler realplex 2
Eppendorf
High-Throughput SNP Genotyping 259
260
Barreiro, Henriques, and Mhlanga
Fig. 17.1. Comparative cost between TaqMan assays and molecular beacons. Regardless of the number of individuals or the number of single nucleotide polymorphisms (SNPs) to be genotyped, the cost of molecular beacons is significantly reduced with respect to TaqMan assays owing to the multiplexing power of molecular beacons in a single tube. The cost for TaqMan assays is based on the prices provided by Applied Biosystem when using 96-well plates and 25-mL PCRs. The cost of TaqMan assays can be reduced by approximately 5–10% by performing the assays in 384-well plates and 5-mL reactions.
exploited. A unique property of nucleic acid hybridization is its extremely high fidelity. Such molecular interactions are the most specific and stable known in nature. It becomes possible to monitor and detect hybridization of nucleic acids if it is accompanied by an assayable change in conformation. Two principal methods have emerged in detecting such assayable changes in conformation. The first, TaqMan (26), depends upon the monitoring of enzymatic nucleic acid probe cleavage, resulting in fluorescence (see Chapters 18 and 19 for details). The second, molecular beacons (1), detects a conformational change in the probe, which fluoresces upon hybridization. We will focus principally on the use of molecular beacons. Molecular beacons are single-stranded oligonucleotide probes with a stem-and-loop structure (see Fig. 17.2). The loop is complementary to a known sequence in a target nucleic acid sequence, whereas the stem forms by the hybridization of the arm sequences on either side of the loop sequence. A fluorescent moiety is covalently linked to the extremity of one arm sequence and a quencher is covalently linked to the extremity of another arm. Thus, the fluorophore and quencher are directly juxtaposed when the stem is formed and are in extremely close proximity to each other. This association prevents fluorescence from being emitted from the fluorophore. When the loop portion of the molecule encounters a perfectly complementary target, the entire molecule undergoes a
High-Throughput SNP Genotyping
261
A.
+ Target
Molecular Beacon
Mutant-specific molecular beacon
Wild-type-specific molecular beacon 1.00
TET Fluorescence
FAM Fluorescence
B.
0.75 +
0.50 0.25 0 30
40
50 60
70 80
Temperature (ºC)
1.00 0.75 0.50
+
0.25 0 30
40
50 60
70 80
Temperature (ºC)
C. Forward Primer CGTATGTAGTGGGATGCGCTC GTATAAAATCGCACGCGATTCCAG
Reverse Primer Fig. 17.2. Principle of how molecular beacons function. (a) When the probe sequence (loop portion) encounters a target that is perfectly complementary to it, a conformational reorganization of the molecule occurs, resulting in a separation of the stem and the generation of a fluorescence signal. (b) Thermal denaturation profiles of molecular beacons when they are with wild-type or mutant targets. The wild-type target is represented by solid lines and the mutant target is represented by dashed lines. The absence of target is indicated by a dotted line. The conformational state of the molecular beacon is shown directly above the line. By careful design of molecular beacons, mismatched targets can be easily discriminated from perfectly matched targets with ‘‘windows of discrimination’’ as high as 10C. The optimal temperature for the annealing step from this thermal denaturation profile is found to be 50C and therefore is used in real-time PCR. (c) An example of how each molecular beacon, the ‘‘red’’-labeled or the ‘‘green’’-labeled, competes to hybridize to the same region depending on whether it is perfectly complementary to the region.
conformational change that results in the separation of the arms of the stem. This causes a restoration of fluorescence to the fluorophore as it is moved away from the quencher. Alterations to the length of the probe region strongly influence the stability and
262
Barreiro, Henriques, and Mhlanga
specificity of the probe–target hybrid, contributing to the extreme specificity of molecular beacons. A wide variety of differently colored fluorophores are possible with molecular beacons (27), thus enabling the simultaneous detection of multiple targets in the same solution by using molecular beacons designed to detect differing targets each labeled with a spectrally distinguishable fluorophore. The above-mentioned properties of molecular beacons enable their use in monitoring the progress of nucleic acid amplification reactions (28–32), self-reporting oligonucleotide arrays, and the detection of messenger RNA in living cells (33–36). Molecular beacons are especially adept at the detection of SNPs since they recognize their targets with exquisite specificity unlike conventional linear probes, owing to their hairpin structure (37). Thermodynamic studies where linear and stem–loop probes were compared have revealed that this enhanced specificity is a general feature of conformationally constrained probes such as molecular beacons. Thus, specificity can be ‘‘tuned’’ by altering the degree to which the probes are conformationally constrained. Practically this involves altering the length of the stem structure in relation to the length of the loop. In applications such as SNP detection, molecular beacons can be designed to bind over a wide range of temperatures such that only perfectly complementary probe–target hybrids are formed. This keeps mismatched probes which vary by even as much as one base unbound and dark, whereas only perfectly complementary probe–target hybrids elicit fluorescence. Owing to these unique properties, the use of molecular beacons for SNP detection has proliferated broadly as has its expansion into a cost-effective high-throughput SNP diagnostic tool.
2. Materials 2.1. Reagents and Equipment
1. Molecular beacon probes (see Section 3.4) designed to hybridize to a target sequence carrying SNP of interest (see Note 2) (Biosearch Technologies, http://www.biosearchtech.com). 2. Fluorescent dyes for manual linking to molecular beacons (Glen Research or Molecular Probes/Invitrogen). 3. Black Hole quenchers (Biosearch Technologies, http:// www.biosearchtech.com). 4. Buffer I: 0.1 M sodium bicarbonate, pH 8.5. 5. Buffer II: 10 mM tris(hydroxymethyl)aminomethane (Tris)– HCl, pH 8.0, 4 mM MgCl2, 50 mM KCl. 6. Buffer A: 0.1 M triethylammonium acetate, pH 6.5.
High-Throughput SNP Genotyping
263
7. Buffer B: 0.1 M triethylamonium acetate in 75% acetonitrile, pH 6.5. 8. Ammonium sulfate (3 M). 9. Silver nitrate (0.15 M). 10. Dithiothreitol (0.15 M). 11. Sodium bicarbonate (0.2 M), pH 9.0. 12. 1X TE buffer: 10 mM Tris–HCl, pH 7.5, 1 mM EDTA. 13. Sephadex G-25 column NAP-5 (GE/Amersham-Pharmacia). 14. Filter: 0.2-mm Centrex MF-0.4 filter (Schleicher & Schuell). 15. High-pressure liquid chromatography (HPLC) system Gold (Beckman Coulter) 16. C-18 reverse-phase column (Waters). 17. Molecular beacon buffer: 10 mM Tris–HCl, pH 8.0, 3.5 mM MgCl2. 18. Thermocycler, PRISM 7700 PCR system (Applied Biosystems). 19. AmpliTaq Gold DNA polymerase (Applied Biosystems). Store at –20C. 20. dNTP set, 100 mM solutions (Applied Biosystems). Store at –20C. 21. Spectrofluorometer, QuantaMaster (Photon Technology International). 22. Haploview software program (HapMap project, http:// www.hapmap.org). 23. Zuker/mfold fold software program (http://www.bioinfo. rpi.edu/applications/mfold/). 2.2. Synthesis of Molecular Beacons
Significant advances have been made in solid-phase chemistry enabling the routine synthesis of nucleic acids coupled to fluorophore and quencher moieties (38). Almost all organic dyes that are routinely used in the visible and infrared light range are available as phosphoramidites, which can be coupled to nucleic acid oligomers during routine syntheses. This is also true for quenchers. For complex syntheses and nonstandard molecular beacons, it is also possible to use manual coupling approaches. This is done by using oligonucleotides which contain either amino or sulfahydryl functional groups at either their 50 -ends or their 30 -ends. By using succinimidyl ester, iodoacetamide derivatives, or maleimide derivatives of the fluorophores and quenchers, one can couple most commercially available dyes and quenchers to oligonucleotides possessing either amino or sulfahydryl functional groups. In Section 3.1 and 3.2 we describe a protocol for manual synthesis of modified oligonucleotides.
264
Barreiro, Henriques, and Mhlanga
2.3. Matching the Fluorophore to the Instrument
With the emergence of real-time PCR as a standard instrument in most laboratories, a number of instruments with differing capabilities have become available. For high-throughput applications such as SNP typing, the principal considerations should be multiplexing abilities, throughput (number of wells), and to a certain extent cycling speed. Spectral overlap is minimized with molecular beacons since they are quenched when unbound. In addition, several instruments (Table 17.1) are able to detect up to six spectrally distinguishable dyes (Table 17.2), routinely enabling extremely powerful multiplexing capabilities.
Table 17.2 Fluorophore labels for fluorescent hybridization probes Fluorophore
Alernative fluorophore
Excitation (nm)
Emission (nm)
Coumarin
Biosearch Bluea, LightCycler Cyan 500b
430
475
495
515
525
540
FAM TET
CAL Fluor Gold 540a c
a
HEX
ATTO 532 , CAL Fluor Orange 560 , JOE, VICd
535
555
Cy3
NEDd, Oyster 556f, Quasar 570a
550
570
555
575
g
a
TMR
Alexa 546 , CAL Fluor Red 590
ROX
Alexa 568g, CAL Fluor Red 610a, LightCycler Red 610b
575
605
Texas red
Alexa 594g, CAL Fluor Red 610a, LightCycler Red 610b
585
605
LightCycler Red 640
CAL Fluor Red 635a
625
640
Cy5
ATTO 647 Nc, LC Red 670b, Oyster 645f, Quasar 670a
650
670
LightCycler Red 705
Cy5.5e, Quasar 705a
680
710
Modified from (39) a Biosearch Blue, CAL, and Quasar fluorophores are available from Biosearch Technologies. b LightCycler fluorophores are available from Roche Applied Science. c ATTA dyes are available from ATTO-TEC. d VIC and NED fluorophores are available from Applied Biosystems. e Cyanine dyes are available from Amersham Biosciences. f Oyster fluorophores are available from Integrated DNA Technologies. g Alexa fluorophores are available from Invitrogen.
High-Throughput SNP Genotyping
265
To run this application one would need to have one of the instruments described in Table 17.1. The choice of the instrument depends on the task and the dyes to be used.
3. Methods 3.1. Coupling of Quencher
1. Dissolve 50–250 nmol of dry (commercially obtained or custom-made) oligonucleotide in 500 mL of buffer I. In DMSO dissolve approximately 20 mg succinimidyl ester coupled quencher and add it to a stirring solution of the oligonucleotide in 10-mL aliquots at 20-min intervals. Continue stirring for at least 12 h. Perform this reaction in the dark (see Note 1). We recommend the Black Hole family of quenchers that are available in three variants dependent on the desired wavelength for quenching (see Section 2.2). 2. Remove particulate material by spinning the mixture in a microcentrifuge for 1 min at 16,000 g. To remove unreacted quencher, pass the supernatant through a gel-exclusion column. Equilibrate a Sephadex G-25 column with buffer A, load the supernatant, and elute the contents of the column with 1 mL of buffer A. Filter the eluate through a 0.2-mm Centrex MF-0.4 filter. 3. Purify the oligonucleotides by HPLC on a C-18 reverse-phase column, utilizing a linear elution gradient of 20–70% buffer B in buffer A and run the elution for 25 min at a flow rate of 1 mL/min. Monitor the absorption of the elution stream at 260 nm and the specific quencher absorption maximum. Collect the eluate that absorbs at both wavelengths, and that therefore contains oligonucleotides with a protected sulfhydryl group at their 50 -ends and the quencher at their 30 -ends. 4. Precipitate the collected material with ethanol and 3 M ammonium sulfate, and spin the precipitate in a centrifuge for 10 min at 16,000g, discard the supernatant, dry the pellet, and dissolve it in 250 mL of buffer A.
3.2. Coupling of Fluorophore
1. To remove the trityl moiety, add 10 mL of 0.15 M silver nitrate and incubate the solution for 30 min. Add 15 mL of 0.15 M dye to this mixture and shake the mixture for 5 min. Spin the mixture for 2 min at 16,000g and transfer the supernatant to a new tube. Dissolve about 40 mg og 5-iodoactamido-reactive fluorophore in 250 mL of 0.2 M sodium bicarbonate, pH 9.0, and add it to the supernatant. Incubate the mixture for 90 min. Each of these solutions should be prepared just before use.
266
Barreiro, Henriques, and Mhlanga
2. Remove excess uncoupled fluorophore from the reaction mixture by gel-exclusion chromatography and purify the oligonucleotides coupled to the fluorophore by HPLC, following the instructions in steps 2 and 3 in Section 3.1. Collect the fractions that absorb with a peak at 260 nm and at the specific fluorophore absorption maximum. This eluate should be fluorescent when observed with an ultraviolet lamp in a dark room. 3. Precipitate the collected material and dissolve the pellet in 100 mL 1X TE buffer. Determine the absorbance at 260 nm and estimate the yield (1 OD260 = 33 ng/mL). Store the purified molecular beacon for long-term storage in lyophilized form at –80C (see Notes 1 and 2).
2. Add 10 mL of 1 mM molecular beacon to this solution and record the new level of fluorescence (Fclosed). 3. Add a twofold molar excess of a complementary oligonucleotide target and monitor the rise in fluorescence until it reaches a stable level (Fopen). 4. Calculate the signal-to-background ratio as (Fopen-Fbuffer)/ (Fclosed-Fbuffer).
80000
60000 50000 40000 30000 20000 10000 0
0
add target [Fopen]
add molecular beacon [Fclosed]
70000
buffer [Fbuffer]
3.3.1. Signal-toBackground Ratio
1. Determine the fluorescence of 200 mL of molecular beacon buffer solution (Fbuffer), using 491 nm as the excitation wavelength and the emission wavelength of the fluorophore used (Fig. 17.3).
Fluorescence (CPS)
3.3. Characterization of Molecular Beacons
50
100
150
200
Time (sec)
Fig. 17.3. Spectrofluorometric characterization of molecular beacons. The molecular beacons are functionally characterized in the presence of perfectly complementary oligonucleotide. Here a 30-fold increase is observed.
High-Throughput SNP Genotyping 3.3.2. Thermal Denaturation Profiles
267
1. Prepare two tubes containing 50 mL of 200 nM molecular beacon dissolved in molecular beacon buffer solution and add the oligonucleotide target to one of the tubes at a final concentration of 400 nM (see Fig. 17.2). 2. Determine the fluorescence of each solution as a function of temperature using a spectrofluorometric thermal cycler (see Table 17.1). Decrease the temperature of these tubes from 80 to 10C in 1C steps, with each hold lasting 1 min, while monitoring the fluorescence during each hold (see Fig. 17.2).
3.4. Design of Primers and Molecular Beacons for SNP Detection
The design of molecular beacons for SNP detection is at times challenging since the flexibility in the targeting region to be detected is virtually nil. The region where the SNP of interest occurs must be targeted and molecular beacons with as little as one base variant from this region must not bind under amplification conditions. To satisfy these constraints, the loop portion of the probe is made to be not more than 25 nucleotides in length. As a rule of thumb, the shorter the length of the loop, the more highly discriminating the probe will be. Care must be taken to ensure that the melting temperature of the probe–target hybrid is compatible with the annealing temperature of primers during PCR. With this part of the design complete, stem/arm sequences can be designed that allow the stem to dissociate at about 7–10C above the annealing temperature of the primers during PCR. This design process is made more complex in certain examples where multiple primers are used in a single tube (as in the example given later in this chapter). The challenge when doing multiplex PCR is to optimize all the primers for all the PCRs first. This ensures that all primers make good amplicons at the same temperature. Molecular beacons can then be designed to be SNP-discriminating at the annealing temperature of the primers by alterations in loop size. It is always useful to verify the secondary structure of the designed molecular beacon to ensure that it does not contain secondary structures that restrict the loop from binding to a PCR target. The preferred program for nucleic acid secondary structure prediction is Zuker/mfold fold (http://www.bioinfo. rpi.edu/applications/mfold/). For extremely difficult situations where design for AT- or GC-rich regions makes the stability of annealing variable, this can be circumvented by a number of strategies such as sliding the loop region so the SNP is no longer at its center. A second strategy is to include the stem/arm sequences in the binding sequence so as to create an even more stable hybrid (this could be useful in AT-rich regions). Lastly, if these strategies prove unsuccessful, an additional annealing step for the purposes of detection can be programmed into the thermal cycling profile. This step can be designed to occur at a temperature where it is easier to meet SNP discrimination constraints with the molecular
268
Barreiro, Henriques, and Mhlanga
A. –939G –871A –336A
–139A
2392G 3220T
–939A –871G –336G
–139G
2392A 3220C
DC SIGN
DC SIGN
Tube 1
Tube 2
3838A 4235G 3933G FAM TET Atto 532
3838C 4235C 3933A
Alexa 546 DC SIGN
Alexa 568 Alexa 594
Tube 3 B.
Allele 1 Allele 2
Homozygous Heterozygous
Homozygous
(major) -939G>A
-871A>G
-336A>G
-139A>G
2392G>A
3220T>C
3838A>C
3933G>A
4235G>C
C.
(minor)
G
A
A
A
G
T
A
G
G
Freq. (%) 8.93
H1
A
G
H2
A
G
10.71
H3
A
G
8.93
G
C
Protective Haplotype
28.57
H4 H5
G
G
H6
G
G
H7
G
G
H8
G
G
H9
G
G
5.36 5.36
A
A C
C
5.36
C
5.36
C
7.14
Fig. 17.4. High-throughput SNP scoring of the DC-SIGN locus. (a) Eighteen molecular beacons and corresponding primers were designed to score the major and minor alleles of nine ‘‘tag’’ SNPs of the DC-SIGN locus. Each major and minor SNP allele had a molecular beacon labeled in a spectrally distinct color. This means that in instruments where up to six colors are spectrally distinguishable, it is possible to simultaneously detect up to six major and/or minor alleles. To score
High-Throughput SNP Genotyping
269
beacons designed. It can also potentially result in false priming so it is not a preferred approach. For detailed instructions on the general design of molecular beacons for SNP detection, see (29,32). PCR primers were designed that consistently amplified regions no greater than 250 base pairs. Those design rules were followed to make the probes and primers shown in (see Fig. 17.4). The dedicated software package Beacon Builder (Premier Biosoft International) can be used for the design of similar molecular beacons. The window of discrimination outlined in Fig. 17.4 should be carefully studied and respected in designing molecular beacons to detect SNPs. 3.5. Real-Time PCR
1. Prepare a 50-mL (or as little as 5-mL) reaction that contains 100 nM major allele specific molecular beacon, 100 nM minor allele specific molecular beacon, 500 nM concentration of each primer, at least 1 unit of AmpliTaq Gold DNA polymerase, and 250 mM concentration of each type of dNTP, dissolved in buffer II. 2. Run the PCR. The thermal cycle for most of the machines described in Table 17.1 should be 10 min at 95C followed by 35–40 cycles at 30 s at 95C, 45 s at 50C (or a temperature which is compatible with the window of discrimination), and 30 s at 72C. The fluorescence should be monitored at the appropriate channel during the 50C annealing step (see Notes 3, 4 and 5).
3.6. Data Analysis in a Case Study Using Tag SNPs (HighThroughput SNP Scoring of the DC-SIGN Locus)
In human genetics, association studies aim to identify loci that contribute to disease susceptibility by comparing patterns of genetic variation between people with a disease (cases) and those without (controls). As mentioned earlier, several studies have revealed an interesting feature present in the structure of human genetic variation that can be utilized to dramatically reduce the
Fig. 17.4. (continued) each of the alleles in a given individual, three PCR amplifications were set up with the appropriate primers (not shown) that all annealed at a similar temperature. At each annealing step, depending on the presence or absence of a particular allele, a given molecular beacon would fluoresce. By ‘‘scoring’’ the data for each tube, one can determine, for each individual the specific genotype for each of the nine tag SNPs. (b) The three possibilities for a given SNP locus, either a single major or a single minor allele is present, in which case a homozygous result is obtained and only a single color is observed. Alternatively, both alleles are observed, indicating that the locus is a heterozygote. (c) Haplotypes observed for the combination of these nine tag SNPs in the Cape Town population. The frequencies reported correspond to the frequencies observed for each of these haplotypes in the Cape Town population independent of their disease status. An association was observed between two DC-SIGN promoter variants (-871G and -336A) and decreased risk of developing tuberculosis. Haplotype 3 turned out to be the best predictor of an increased resistance to tuberculosis, at least in the South African population. This haplotype, which contains both -871G and -336A, was found to be more frequently observed in the control group than in people who developed tuberculosis (8.9% vs. 14.2% p = 1.6 103; odds ratio 1.7; 95% confidence interval 1.22–2.38.
270
Barreiro, Henriques, and Mhlanga
cost of association studies (11, 40–43). Specifically, alleles at nearby loci often show strong statistical association (i.e., LD). This can be exploited to design a powerful and cost-effective way to perform association studies by using tag SNPs for a region of interest, i.e., by determining which loci within that region capture the majority of the diversity. In this section we outline a study of the DC-SIGN gene. By using the unique multiplexing power of molecular beacons in a high-throughput assay, we are able to genotype nine tag SNPs thereby obtaining information from 54 SNPs. Thus, with three tubes per individual and with three pairs of molecular beacons per tube, we are able to score all the information of 54 SNPs. DC-SIGN is an innate immunity gene that belongs to the C-type lectin family. C-type lectins are calcium-dependent carbohydrate-binding proteins with a wide range of biological functions, many of which are related to immunity (44). DC-SIGN as well as its homolog L-SIGN are particularly interesting, since they can act as both cell-adhesion receptors and pathogenrecognition receptors (45). DC-SIGN was originally cloned for its ability to bind and internalize the heavily glycosylated HIV gp120 protein (46). DC-SIGN strongly binds all HIV and simian immunodeficiency virus strains examined to date and plays an important role in virus adhesion to dendritic cells (47, 48). These studies have paved the way for further investigations into interactions between DC-SIGN and other pathogens and it has now become clear that this lectin recognizes a vast range of microbes, some of which are of major public health importance (48). Indeed, DC-SIGN captures bacteria such as Mycobacterium tuberculosis, Mycobacterium leprae, Helicobacter pylori, and certain Klebsiela pneumonia strains; viruses such as HIV-1, Ebola virus, cytomegalovirus, hepatitis C virus, Dengue virus, and SARS coronavirus; and parasites such as Leishmania pifanoi and Schistosoma mansoni (47, 49–59). In light of the ability of DC-SIGN to interact with a large plethora of pathogens, it is plausible that variation in its gene may influence the pathogenesis of a number of infectious diseases. Indeed, multiple association studies have shown a relationship between genetic variants in the promoter region of DC-SIGN and susceptibility to several infectious diseases. Specifically, it has been shown that two promoter variants, -871G and -336A, confer protection against tuberculosis. Similarly, the -336A variant has been reported to protect against parental HIV infection and to influence the severity of dengue pathogenesis (60, 61). More recently, two other promoter variants, –139A/G and –939G/A, showed a significant association with an increased risk of developing human cytomegalovirus reactivation and disease (60).
High-Throughput SNP Genotyping
271
How can one efficiently test for an association between DCSIGN variation and susceptibility to disease? Imagine that you want to explore the relationship between DC-SIGN polymorphisms and susceptibility to tuberculosis (62). The best way to do so is to follow the strategy described below: 1. Collect a cohort, from the same population (see Note 6), that includes a group of individuals that developed tuberculosis (i.e., cases) and a group of matched individuals that did not develop the disease (i.e., controls). Ideally, one would need/ like to fully resequence DC-SIGN in the entire cohort to obtain the full extent of diversity present in cases and controls. Nevertheless, full resequencing approaches are unacceptably expensive and time consuming and, therefore, the most powerful and cost-effective way to perform association studies is by defining tag SNPs for a region of interest (see Section 17.1 for details). To do so, you have two alternatives: (a) Begin by fully resequencing the region under study in a subset of your cohort. Typically 20–30 individuals should be enough to capture the most common haplotypes in the population. After haplotype reconstruction (see Note 7) and on the basis of the LD patterns observed, you can then identify the set of SNPs best able to characterize the diversity observed (i.e., tag SNPs) (see Note 8). (b) Use publicly available datasets to identify tag SNPs. The best available resource to choose tag SNPs is the HapMap data. Go to the HapMap Web site (http://www.hapma p.org) and using the genome browser retrieve genotypic data for all the SNPs that have been typed for the region you are interested in; in this case DC-SIGN. Then, upload the data in Haploview (a free software program provided by the HapMap consortium) and run Tagger to identify tag SNPs for your region (see Note 7). The current limitation of using HapMap is that the data are restricted to three human populations – the samples came from an African population from Nigeria (Yoruba; N ¼ 90), a mostly Utah (USA) population of European ancestry (N ¼ 90), and a sample drawn from Japanese (N ¼ 45) and Han Chinese (N ¼ 45) populations. If your population is genetically distinct from these HapMap populations, you will have to follow the resequencing strategy; as the tag SNPs identified using HapMap populations might differ from those characterizing the diversity of your study-population. 2. Once you have identified the set of SNPs best able to characterize the full diversity observed in your population, the next step is to genotype these tag SNPs in the entire cohort.
272
Barreiro, Henriques, and Mhlanga
In Fig. 17.4 we present an example of a haplotyping approach scoring tag SNPs in a high-throughput assay using molecular beacons to easily test for an association between DC-SIGN variation and susceptibility to infectious diseases. This example is based on a previous study that explored the relationship between DC-SIGN polymorphisms and susceptibility to tuberculosis (63). The authors showed that nucleotide variation in the DC-SIGN promoter region is associated with susceptibility to tuberculosis. Specifically they identified a specific haplotype (Fig. 17.4) associated with decreased risk of developing tuberculosis (63).
4. Notes 1. Molecular beacons deteriorate as they are exposed to light. Therefore, avoid exposure to light whenever possible. Molecular beacons should be stored in aluminum-foil-wrapped test tubes at –20C and preferably at –80C in lyophilized form. When preparing them for use, one can resuspended them in TE buffer. 2. Since most oligonucleotide manufacturers worldwide can provide molecular beacons with all these functionalities, obtaining molecular beacons with diverse fluorophore and quencher combinations has become routine. These suppliers can be found at http://www.molecular-beacons.org. 3. At times, false amplicons may appear during PCR and may appear if the sensitivity of the PCR is reduced. Two approaches can be used to circumvent this. Firstly, DNA polymerases that are active only after activation at 95C can be used. Secondly, paying careful attention to the design of primers that function well within the ‘‘window of discrimination’’ is recommended. 4. The real-time PCR machines and fluorescent dyes proposed in Table 17.1 and 17.2 are fairly good at discriminating between the proposed dyes. Thus, if poor discrimination is observed between major and minor alleles, tweaks to the primers and annealing temperatures can be made that permit more stringent discrimination. If these are unsuccessful, modifications to the molecular beacons themselves can be made. One modification is to increase the length of the molecular beacon stem to promote stability and increase stringency. A second modification is to use 20 -O-methyl molecular beacons, which intrinsically have a higher melting temperature than DNA-based molecular beacons. However 20 -O-methyl
High-Throughput SNP Genotyping
273
molecular beacons are more expensive to synthesize. Third, the stem sequence of the molecular beacon can be designed to also bind to the amplicon. 5. Amplicon size has a very important influence on the fluorescence signal obtained with molecular beacons. Thus, it is important to design PCRs where amplicons do not exceed 250 bp. 6. It is important that the groups of cases and controls are genetically matched, as population stratification between cases and controls can be a confounding factor leading to a spurious positive association. This will be particularly harmful if cases and controls are from different populations, but also in admixed populations (e.g. CAP population from South Africa). Indeed, the use of admixed populations in association-mapping studies can be very useful for identification of disease-causing genetic variants that differ in frequency across parental populations. However, when the admixture event is too recent, allelic frequencies can differ coincidentally among cases and controls, reflecting a nonuniform genetic contribution from the parental populations to each subpopulation (i.e., cases and controls), rather than a genuine association between a given genetic variant and the phenotype under study. In this case, the study cohort is said to present population stratification. 7. To reconstruct haplotypes we recommend the Bayesian statistical method implemented in Phase version 2.1.162 (64). Alternatively, you can use the accelerated expectation maximization algorithm implemented in Haploview version 3.163 (65). At least for regions with high levels of LD, both algorithms should give similar results. 8. Tag SNPsfor each population can be selected using Haploview’s Tagger in pairwise tagging mode (r2 0.80, minor allele frequency cutoff 5%, and other settings at default value). References 1. Tyagi, S. and Kramer, F. R. (1996) Molecular beacons: probes that fluoresce upon hybridization. Nat. Biotechnol. 14, 303–308. 2. Tyagi, S. (2000) DNA Probes, In Encyclopedia of Analytical Chemistry: Applications, Theory and Instrumentation (Meyers, R. A., Ed.) John Wiley & Sons Ltd. Chichester, UK, Vol. 6, pp. 4911. 3. Lander, E. S., Linton, L. M., Birren, B. et al. (2001) Initial sequencing and analysis of the human genome. Nature 409, 860–921. 4. Sachidanandam, R., Weissman, D., Schmidt, S. C. et al. (2001) A map of human genome sequence variation containing 1.42 million
single nucleotide polymorphisms. Nature 409, 928–933. 5. Hinds, D. A., Stuve, L. L., Nilsen, G. B. et al. (2005) Whole-genome patterns of common DNA variation in three human populations. Science 307, 1072–1079. 6. Miller, R. D., Phillips, M. S., Jo, I. et al. (2005) High-density single-nucleotide polymorphism maps of the human genome. Genomics 86, 117–126. 7. Kruglyak, L. and Nickerson, D. A. (2001) Variation is the spice of life. Nat. Genet. 27, 234–236.
274
Barreiro, Henriques, and Mhlanga
8. Miller, R. D. and Kwok, P. Y. (2001) The birth and death of human single-nucleotide polymorphisms: new experimental evidence and implications for human history and medicine. Hum. Mol. Genet. 10, 2195–2198. 9. Crawford, D. C., Akey, D. T. and Nickerson, D. A. (2005) The patterns of natural variation in human genes. Annu. Rev. Genomics Hum. Genet. 6, 287–312. 10. Consortium TIH. (2003) The International HapMap Project. Nature 426, 789–796. 11. Consortium TIH. (2005) A haplotype map of the human genome. Nature 437, 1299–1320. 12. Frazer, K. A., Ballinger, D. G., Cox, D. R. et al. (2007) A second generation human haplotype map of over 3.1 million SNPs. Nature 449, 851–861. 13. Conrad, D. F., Jakobsson, M., Coop, G. et al. (2006) A worldwide survey of haplotype variation and linkage disequilibrium in the human genome. Nat. Genet. 38, 1251–1260. 14. Gonzalez-Neira, A., Ke, X., Lao, O. et al. (2006) The portability of tagSNPs across populations: a worldwide survey. Genome Res. 16, 323–330. 15. Eberle, M. A., Ng, P. C., Kuhn, K. et al. (2007) Power to detect risk alleles using genome-wide tag SNP panels. PLoS Genet. 3, 1827–1837. 16. Consortium TWTCC. (2007) Genomewide association study of 14,000 cases of seven common diseases and 3,000 shared controls. Nature 447, 661–678. 17. Todd, J. A., Walker, N. M., Cooper, J. D. et al. (2007) Robust associations of four new chromosome regions from genome-wide analyses of type 1 diabetes. Nat. Genet. 39, 857–864. 18. Saxena, R., Voight, B. F., Lyssenko, V. et al. (2007) Genome-wide association analysis identifies loci for type 2 diabetes and triglyceride levels. Science 316, 1331–1336. 19. Zeggini, E., Weedon, M. N., Lindgren, C. M. et al. (2007) Replication of genomewide association signals in UK samples reveals risk loci for type 2 diabetes. Science 316, 1336–1341. 20. Samani, N. J., Erdmann, J., Hall, A. S. et al. (2007) Genomewide association analysis of coronary artery disease. N. Engl. J. Med. 357, 443–453. 21. Frayling, T. M., Timpson, N. J., Weedon, M. N. et al. (2007) A common variant in the FTO gene is associated with body mass index and predisposes to childhood and adult obesity. Science 316, 889–894.
22. Scuteri, A., Sanna, S., Chen, W. M. et al. (2007) Genome-wide association scan shows genetic variants in the FTO gene are associated with obesity-related traits. PLoS Genet. 3, e115. 23. Thomson, W., Barton, A., Ke, X. et al. (2007) Rheumatoid arthritis association at 6q23. Nat Genet 39, 1431–1433. 24. Fellay, J., Shianna, K. V., Ge, D. et al. (2007) A whole-genome association study of major determinants for host control of HIV-1. Science 317, 944–947. 25. Leone, G., van Schijndel, H., van Gemen, B., Kramer, F. R. and Schoen, C. D. (1998) Molecular beacon probes combined with amplification by NASBA enable homogeneous, real-time detection of RNA. Nucleic Acids Res. 26, 2150–2155. 26. Livak, K. J. (1999) Allelic discrimination using fluorogenic probes and the 50 nuclease assay. Genet. Anal. 14, 143–149. 27. Tyagi, S., Bratu, D. P. and Kramer, F. R. (1998) Multicolor molecular beacons for allele discrimination. Nat. Biotechnol. 16, 49–53. 28. El-Hajj, H. H., Marras, S. A., Tyagi, S., Kramer, F. R. and Alland, D. (2001) Detection of rifampin resistance in Mycobacterium tuberculosis in a single tube with molecular beacons. J. Clin. Microbiol. 39, 4131–4137. 29. Marras, S. A., Kramer, F. R. and Tyagi, S. (2003) Genotyping SNPs with molecular beacons. Methods Mol. Biol. 212, 111–128. 30. Vet, J. A., Majithia, A. R., Marras, S. A. et al. (1999) Multiplex detection of four pathogenic retroviruses using molecular beacons. Proc. Natl. Acad. Sci. U.S.A. 96, 6394–6399. 31. Kostrikis, L. G., Tyagi, S., Mhlanga, M. M., Ho, D. D. and Kramer, F. R. (1998) Spectral genotyping of human alleles. Science 279, 1228–1229. 32. Mhlanga, M. M. and Malmberg, L. (2001) Using molecular beacons to detect singlenucleotide polymorphisms with real-time PCR. Methods 25, 463–471. 33. Bratu, D. P., Cha, B. J., Mhlanga, M. M., Kramer, F.R. and Tyagi, S. (2003) Visualizing the distribution and transport of mRNAs in living cells. Proc. Natl. Acad. Sci. U.S.A. 100, 13308–13313. 34. Mhlanga, M. M., Vargas, D. Y., Fung, C. W., Kramer, F. R. and Tyagi, S. (2005) tRNA-linked molecular beacons for imaging mRNAs in the cytoplasm of living cells. Nucleic Acids Res. 33, 1902–1912.
High-Throughput SNP Genotyping 35. Tyagi, S. and Alsmadi, O. (2004) Imaging native beta-actin mRNA in motile fibroblasts. Biophys. J. 87, 4153–4162. 36. Vargas, D. Y., Raj, A., Marras, S. A., Kramer, F. R. and Tyagi, S. (2005) Mechanism of mRNA transport in the nucleus. Proc. Natl. Acad. Sci. U.S.A. 102, 17008–17013. 37. Bonnet, G., Tyagi, S., Libchaber, A. and Kramer, F. R. (1999) Thermodynamic basis of the enhanced specificity of structured DNA probes. Proc. Natl. Acad. Sci. U.S.A. 96, 6171–6176. 38. Lee, L. G., Livak, K. J., Mullah, B., Graham, R. J., Vinayak, R.S. and Woudenberg T. M. (1999) Seven-color, homogeneous detection of six PCR products. Biotechniques 27, 342–349. 39. Marras, S. A. (2008) Interactive fluorophore and quencher pairs for labeling fluorescent nucleic acid hybridization probes. Mol. Biotechnol. 38, 247–255. 40. Daly, M. J., Rioux, J. D., Schaffner, S. F., Hudson, T. J. and Lander, E. S. (2001) High-resolution haplotype structure in the human genome. Nat. Genet. 29, 229–232. 41. Dawson, E., Abecasis, G. R., Bumpstead, S. et al. (2002) A first-generation linkage disequilibrium map of human chromosome 22. Nature 418, 544–548. 42. Gabriel, S. B., Schaffner, S. F., Nguyen, H. et al. (2002) The structure of haplotype blocks in the human genome. Science 296, 2225–2229. 43. Reich, D. E., Cargill, M., Bolk, S. et al. (2001) Linkage disequilibrium in the human genome. Nature 411, 199–204. 44. Zelensky, A. N. and Gready, J. E. (2005) The C-type lectin-like domain superfamily. FEBS J. 272, 6179–6217. 45. Soilleux, E. J. (2003) DC-SIGN (dendritic cell-specific ICAM-grabbing non-integrin) and DC-SIGN-related (DC-SIGNR): friend or foe? Clin. Sci. (Lond) 104, 437–446. 46. Curtis, B. M., Scharnowske, S. and Watson, A. J. (1992) Sequence and expression of a membrane-associated C-type lectin that exhibits CD4-independent binding of human immunodeficiency virus envelope glycoprotein gp120. Proc. Natl. Acad. Sci. U. S. A. 89, 8356–8360. 47. Geijtenbeek, T. B., Kwon, D. S., Torensma, R. et al. (2000) DC-SIGN, a dendritic cellspecific HIV-1-binding protein that enhances trans-infection of T cells. Cell 100, 587–597. 48. Geijtenbeek, T. B., van Vliet, S. J., Engering, A., Hart, B. A. and van Kooyk, Y. (2004) Self- and nonself-recognition by
49.
50.
51.
52.
53.
54.
55.
56.
57.
58.
275
C-type lectins on dendritic cells. Annu. Rev. Immunol. 22, 33–54. Alvarez, C. P., Lasala, F., Carrillo, J., Muniz, O., Corbi, A. L. and Delgado, R. (2002) C-type lectins DC-SIGN and L-SIGN mediate cellular entry by Ebola virus in cis and in trans. J. Virol. 76, 6841–6844. Appelmelk, B. J., van Die, I., van Vliet, S. J., Vandenbroucke-Grauls, C. M., Geijtenbeek, T. B. and van Kooyk, Y. (2003) Cutting edge: carbohydrate profiling identifies new pathogens that interact with dendritic cell-specific ICAM-3-grabbing nonintegrin on dendritic cells. J. Immunol. 170, 1635–1639. Barreiro, L. B., Quach, H., Krahenbuhl, J. et al. (2006) DC-SIGN interacts with Mycobacterium leprae but sequence variation in this lectin is not associated with leprosy in the Pakistani population. Hum. Immunol. 67, 102–107. Bergman, M. P., Engering, A., Smits, H. H. et al. (2004) Helicobacter pylori modulates the T helper cell 1/T helper cell 2 balance through phase-variable interaction between lipopolysaccharide and DC-SIGN. J. Exp. Med. 200, 979–990. Colmenares, M., Puig-Kroger, A., Pello, O. M., Corbi, A. L. and Rivas L. (2002) Dendritic cell (DC)-specific intercellular adhesion molecule 3 (ICAM-3)-grabbing nonintegrin (DC-SIGN, CD209), a C-type surface lectin in human DCs, is a receptor for Leishmania amastigotes. J. Biol. Chem. 277, 36766–36769. Geijtenbeek, T. B., Van Vliet, S. J., Koppel, E. A. et al. (2003) Mycobacteria target DCSIGN to suppress dendritic cell function. J. Exp. Med. 197, 7–17. Halary, F., Amara, A., Lortat-Jacob, H. et al. (2002) Human cytomegalovirus binding to DC-SIGN is required for dendritic cell infection and target cell trans-infection. Immunity 17, 653–664. Lozach, P. Y., Lortat-Jacob, H., de Lacroix de Lavalette, A. et al. (2003) DC-SIGN and L-SIGN are high affinity binding receptors for hepatitis C virus glycoprotein E2. J. Biol. Chem. 278, 20358–20366. Marzi, A., Gramberg, T., Simmons, G. et al. (2004) DC-SIGN and DC-SIGNR interact with the glycoprotein of Marburg virus and the S protein of severe acute respiratory syndrome coronavirus. J. Virol. 78, 12090–12095. Tailleux, L., Schwartz, O., Herrmann, J. L. et al. (2003) DC-SIGN is the major
276
59.
60.
61.
62.
Barreiro, Henriques, and Mhlanga Mycobacterium tuberculosis receptor on human dendritic cells. J. Exp. Med. 197, 121–127. Tassaneetrithep, B., Burgess, T. H., GranelliPiperno, A. et al. (2003) DC-SIGN (CD209) mediates dengue virus infection of human dendritic cells. J. Exp. Med. 197, 823–829. Martin, M. P., Lederman, M. M., Hutcheson, H. B. et al. (2004) Association of DC-SIGN promoter polymorphism with increased risk for parenteral, but not mucosal, acquisition of human immunodeficiency virus type 1 infection. J. Virol. 78, 14053–14056. Sakuntabhai, A., Turbpaiboon, C., Casademont, I. et al. (2005) A variant in the CD209 promoter is associated with severity of dengue disease. Nat. Genet. 37, 507–513. Mezger, M., Steffens, M., Semmler, C. et al. (2008) Investigation of promoter variations
in dendritic cell-specific ICAM3-grabbing non-integrin (DC-SIGN) (CD209) and their relevance for human cytomegalovirus reactivation and disease after allogeneic stem-cell transplantation. Clin. Microbiol. Infect. 14, 228–234. 63. Barreiro, L. B., Neyrolles, O., Babb, C. L. et al. (2006) Promoter variation in the DCSIGN-encoding gene CD209 is associated with tuberculosis. PLoS Med. 3, e20. 64. Stephens, M. and Donnelly, P. A. (2003) comparison of bayesian methods for haplotype reconstruction from population genotype data. Am. J. Hum. Genet. 73, 1162–1169. 65. Barrett, J. C., Fry, B., Maller, J. and Daly, M. J. (2005) Haploview: analysis and visualization of LD and haplotype maps. Bioinformatics 21, 263–265.
Chapter 18 SNP Genotyping by the 50 -Nuclease Reaction: Advances in High-Throughput Genotyping with Nonmodel Organisms James E. Seeb, Carita E. Pascal, Ramesh Ramakrishnan, and Lisa W. Seeb Abstract Population genetics studies play an increasingly important role in the management and conservation of nonmodel organisms. Unlike studies with model organisms, a typical population genetics study of a nonmodel organism may be conducted by analyzing thousands or hundreds of thousands of individuals for several dozen single nucleotide polymorphisms (SNPs). The use of robust, robotically mediated TaqMan reactions provides substantial advantages in these types of studies. We describe the methods and laboratory setup for analyzing a sustained high throughput of SNP assays in routine university or natural resource agency laboratories with a handful of thermal cyclers. Agencies sustain rates of nearly 150,000 assays per week using uniplex reactions with the Applied Biosystems 7900HT Fast Real-Time PCR System (AB 7900HT). We further describe the medium-density array run on the BioMark from Fluidigm, which increases this rate to over 500,000 assays per week by multiplexing 96 samples for 96 SNPs. Key words: TaqMan, medium-density array, error rate, TaqMan low-density array, genotyping.
1. Introduction Since its first description (1, 2), the 50 -nuclease reaction has become an increasingly important tool in human genetics and genomics. The literature is rich with reports of single nucleotide polymorphism (SNP) screens for candidate genes, genome-association studies, and disease-detection studies employing the 50 nuclease reaction (TaqMan). However, only recently has the use of SNPs and gene detection by TaqMan been broadly applied to population genetics. At the same time, population genetics has played an increasingly important role in the management and conservation of nonmodel organisms. Genetic studies describe A.A. Komar (ed.), Single Nucleotide Polymorphisms, Methods in Molecular Biology 578, DOI 10.1007/978-1-60327-411-1_18, ª Humana Press, a part of Springer Science+Business Media, LLC 2003, 2009
277
278
Seeb et al.
the range of endangered populations or species, units of conservation for healthy species, and are now used to identify components of admixtures of migrating species (3). The use of SNPs in population genetics applications of nonmodel organisms has increased markedly in recent years as DNA sequence data accumulate and replace fragment analyses (4–7). An important difference between human genomics and population genetics of nonmodel organisms is the number of individuals and the number of assays. A typical study, of genome association in humans, for example, may be conducted by assaying a few dozen individuals for hundreds of thousands of SNPs. A typical population genetics study of Pacific salmon may be conducted by analyzing thousands or hundreds of thousands of individuals for several dozen SNPs. The use of robust, robotically mediated TaqMan reactions provides substantial advantages in the latter. The properties and methods of SNP detection using TaqMan have been described elsewhere (8). Among these properties, and important here, are that the reaction is robust, tolerating variability in DNA quality and quantity (9), genotyping error rates are exceedingly low (10), and the reaction is very amenable to automation. The assay is run in a closed-tube format with no post-PCR processing steps. Results are obtained simply and directly by measuring the fluorescence of the reaction. During the 50 -nuclease reaction, fluorogenic hybridization probes are used to detect specific PCR products as they accumulate during PCR cycles. Probes consist of an oligonucleotide labeled with a reporter dye linked to the 50 -end of the probe and a nonfluorescent quencher at the 30 -end of the probe. When the probe is intact, the proximity of the reporter dye to the quencher suppresses the reporter fluorescence. During PCR, the probe anneals specifically to a complementary sequence. Cleavage liberates the reporter dye, resulting in increased fluorescence at each PCR cycle. This increase in fluorescence occurs only if the target sequence is complementary to the probe and is amplified during PCR (see Fig. 18.1). For allelic discrimination of biallelic systems, probes specific for each allele are included in the PCR. Each probe is labeled with one of two fluorescent reporter dyes (FAMTM or VICTM). Mismatches between probe and target result in low-efficiency probe hybridization and cleavage and low fluorescence. A substantial increase in fluorescence is realized from a match. Heterozygotes are identified by an increase in fluorescence of both signals (see Fig. 18.2). 1.1. SNPs in Nonmodel Organisms
As more work is done with nonmodel organisms, especially fish, we find that SNPs are expressed at rates similar to those observed in humans. During the sequencing of 89 kb of Pacific salmon expressed
SNP Genotyping by the 50 -Nuclease Reaction
279
Polymerization Forward Primer
Q
R Probe
3’
5’ 3’ 5’
Reverse Primer
Strand displacement
5’ 3’ 5’
R Q 3’
5’ 3’ 5’
5’ 3’ 5’ Fluorescence occurs
Cleavage
R Q 3’
5’ 3’ 5’
Polymerization completed
5’ 3’ 5’
R
Q 3’
5’ 3’ 5’
5’ 3’ 5’
Fig. 18.1. PCR amplification and allele detection with fluorogenic probes in the TaqMan assay. The single nucleotide polymorphism (SNP) site is interrogated with two probes. Each probe is complementary to one of the two alleles in a biallelic SNP. Each probe has two dyes, a fluorescent reporter (R) that is specific to the allele, and a fluorescent quencher (Q). The reporter dye is quenched when both dyes are attached to the probe. The main steps in the reaction sequence are polymerization, strand displacement, and cleavage. The probe that matches the target SNP will bind, and DNA polymerase will cleave the reporter dye during each extension cycle. The probe that does not match the target SNP will not bind well, and no cleavage occurs. After cleavage the reporter will fluoresce, and the color will discriminate the SNP allele. (After (8)).
sequence tags, one SNP was found every 232 bp (11). On the basis of results from humans (12), one could easily expect to find greater than 10,000,000 SNPs in each nonmodel species with greater than 2,000,000 expressed in any individual. Coupling this density of genetic variation with modern techniques of SNP detection, one
Seeb et al. ADFG|Ots_P450 28000
Amount of FAM probe cleaved
280
23000
AA AT
18000
13000
TT
8000
3000
–2000 –2000
0
2000
4000
6000
8000
10000
12000
Amount of VIC probe cleaved
Fig. 18.2. Scatter plot of 380 individuals assayed for an A/T SNP using an AB 7900 HT. Units on the y-axis are arbitrary fluorescence of FAM probes designed to detect the A allele; units on the x-axis are arbitrary fluorescence of the VIC probes designed to detect the T allele. Four no-template controls cluster near the origin at the bottom left. AA homozygotes cluster on the y-axis and TT homozygotes cluster on the x-axis. Heterozygotes (green) are intermediate. One individual falls between the AT and TT clusters and cannot be scored (no call).
can forecast that SNPs will soon obsolete other marker types in studies of nonmodel organisms for genome screens, gene mapping studies, or basic studies of population genetics. Many high-throughput techniques are available for SNP detection (13). Of these, the TaqMan assay attained the most traction in studies of nonmodel organisms (5, 14–16). Endpoint reads can be done on a variety of instruments, including plate readers or real-time PCR machines. Real-time PCR machines offer the advantages of quantitative PCR for troubleshooting and advanced software for allele calling. Throughput rates will improve with increased plate density (e.g., 1,536-well plates released by Roche Applied Science in lieu of the 384-well plate commonly used today) or improvements in multiplex or array platforms. Further, TaqMan assays are robust to a wide variety of starting DNA concentrations. This advantage is particularly critical with the large number of samples often screened in nonmodel organisms where quantitation and normalization of individual samples are not practical. Fisheries management agencies currently genotype thousands of fish, sustaining rates of nearly 150,000 assays per week, using offline PCR for uniplex reactions to be read on the Applied Biosystems 7900HT Fast Real-Time PCR System (AB 7900HT). In this chapter, we describe the methods and laboratory setup for analyzing a sustained high throughput of 70,000 SNP assays in university or agency laboratories with a handful of thermal cyclers.
SNP Genotyping by the 50 -Nuclease Reaction
281
We further describe the medium-density array run on the BioMark from Fluidigm, which increases this rate to over 500,000 assays per week by multiplexing 96 samples for 96 SNPs. 1.2. Platforms for HighThroughput Analyses
It should be noted that several platforms support uniplex TaqMan chemistry. The Applied Biosystems 7x00 series offers options with 96- or 384-well blocks. The Roche LC 480 comes standard with a 384-well block and was upgraded in 2008 with an optional 1536well block. Each of these can be configured with liquid handling and plate handling robots to provide high-throughput analyses. Because multiple TaqMan reactions do not perform well in single wells (17), research focused on increasing throughput by using parallel reactions in low- or medium-density arrays (18, 19). The TaqMan low-density array (TLDA) from Applied Biosystems (http://tlda.appliedbiosystems.com) can be used to analyze eight samples for each of 48 TaqMan assays in a 384-well microfluidic card. This platform offers rapid, reproducible, and quantitative results for gene expression study of low numbers of individuals (20, 21). However, analysis requires a specialized block on the AB 7900HT, and the configuration of the card is not amenable to automation. For these reasons, the TLDA card is suitable for efficiently genotyping small numbers of individuals for up to 48 SNPs, but will not work well for population genetics studies requiring high-throughput processing. Medium-density arrays for TaqMan are now offered by BioTrove (OpenArray; http:// www.biotrove.com/) and Fluidigm (BioMark arrays; http:// www.fluidigm.com/) (22). The BioTrove OpenArray system is a fixed array that was originally designed to test up to 48 samples against up to 64 TaqMan assays, each assay in a subarray of 64 reaction cells. One disadvantage is that assays must be loaded into the 3,072 cells by the manufacturer. Investigators must identify the panel of SNPs for a particular project and then order the proper number of preloaded arrays. Developments are under way at Applied Biosystems (https://products.appliedbiosystems.com/ab/en/US/adi rect/ab?cmd=catNavigate2&catID=605780&tab=DetailInfo) to introduce prespotted OpenArray formats that vary the sampleto-assay ratios by capitalizing on the subarray format of 48 64. Fluidigm has three medium-density configurations that can be run on the BioMark and EP1 instruments: two for high-throughput genotyping (Dynamic Array) and one for copy number determination using digital PCR (Digital Array) (23). The highthroughput Dynamic Array, of interest in this chapter, may be used to screen either up to 48 samples for 48 TaqMan assays or up to 96 samples for 96 TaqMan assays. These arrays are loaded by the investigator, either manually or robotically, allowing flexibility to alter the panel of SNPs during periods of discovery or assay optimization.
282
Seeb et al.
We describe here high-throughput genotyping using TaqMan assays with two different approaches. First, we focus upon uniplex reactions. Uniplex reactions, using 384-well plates, can be especially effective for population genetics study during the SNP discovery phase for nonmodel organisms where few SNPs are available. Data can be added one SNP at a time to population data sets. Second, we describe an array approach using the Fluidigm 96.96 Dynamic Array. Arrays rapidly decrease cost and increase throughput once enough SNPs have been developed to fill at least half the array’s capacity (Table 18.1).
Table 18.1 Reaction volume and relative cost per genotype of various platforms Format
Reaction volume
Cost breakdown ($)
Total cost/genotype ($)a
96-well plate
20 mL
0.003 plastics 0.453 TaqMan 0.220 assay
0.676
384-well plate
5 mL
0.009 plastics 0.113 TaqMan 0.055 assay
0.177
48 48 array
10 nL
0.102 array 0.002 TaqMan 0.012 assay
0.116
96 96 array
10 nL
0.050 array 0.002 TaqMan 0.005 assay
0.057
a
Total cost is for comparison among platforms only; actual cost will vary depending upon marketing factors.
2. Materials 2.1. Equipment 2.1.1. Uniplex Reaction
1. Twelve 384-well thermal cycling blocks. 2. Automated liquid handler with 96-well pipetting head [e.g., JANUS automated workstation (PerkinElmer), Biomek 1 1 FX laboratory automation workstation (Beckman-Coulter)]. 3. An AB 7900HT with automation accessory or a Roche LightCycler 480 plate-based real-time PCR system with the PlateServer or equivalent.
2.1.2. Array Approach
1. Fluidigm Biomark or EP1 instrument. 2. Fluidigm integrated fluidic circuit (IFC) controller HX.
SNP Genotyping by the 50 -Nuclease Reaction
283
3. Four Fluidigm IFC thermal cyclers. 4. Automated liquid handler with 96-well pipetting head and a four- or eight-tip pipetting arm [e.g., JANUS1 automated workstation (PerkinElmer), Biomek1 FX laboratory automation workstation (Beckman-Coulter)]. 2.2. DNA and Reagents
1. Genomic DNA. DNA can be extracted by any commercially available kits (QIAGEN, Invitrogen, etc.). We recommend that extractions be done in 96-well plates; if the DNA is extracted and stored in tubes, then the first step is to transfer it into 96-well plates. For very low concentrations of DNA, a preamplification step may be necessary (see Note 1). 2. 1x TE buffer for DNA dilution/storage: 10 mM tris(hydroxymethyl)aminomethane–HCl, pH 8.0, 1 mM EDTA. 3. 2x TaqMan universal PCR master mix (Applied Biosystems). 4. 80x primer/probe mix: 16 mM concentration of each probe, 72 mM concentration of each primer, in 1x TE buffer. Alternatively, the 80x primer/probe mix can be purchased from Applied Biosystems. 5. DNA/DNAse-free H2O.
2.2.1. Uniplex Reaction
1. Optical adhesive cover (Applied Biosystems).
2.2.2. Array Approach
1. AmpliTaq Gold DNA polymerase (Applied Biosystems). 2. 2x Dynamic Array assay loading reagent (Fluidigm). 3. 50x ROX (Invitrogen). 4. 20x GT sample loading reagent (Fluidigm).
3. Methods 3.1. Experimental Design for HighThroughput Uniplex Using Offline PCR
First we establish a simple experimental design using end-point reads on a 384-well quantitative PCR instrument with additional offline thermal cycling. Detailed instructions for use of the AB 7900HT are given in Chapter 19. It is important to note that some authors rate the maximum potential throughput of the AB 7900HT at 250,000 genotypes per day (8). This theoretical maximum can only be achieved with the thermal cycling of more than 650 384-well plates per day, a rate that is not achievable in a typical laboratory. A typical laboratory conducting population genetics studies may need to acquire data from thousands of samples with only a handful of PCR machines. Our design for a start-up laboratory will employ 12 thermal cycling blocks in a study of 48 SNPs.
284
Seeb et al.
One or two researchers can easily process twelve 384-well plates for three TaqMan assays each day (throughput of 13,824 assays per day, or nearly 70,000 per week). In this fashion, 4,608 samples are assayed for 48 SNPs every 16 days. 3.1.1. Preparation of 384Well Plates
1. Load 2 mL of genomic DNA with a concentration of 1–10 ng/ mL into each well using an automated liquid handler with a 96well pipetting head; for very low concentrations of DNA, a preamplification step would be necessary (see Note 1). l DNA plates 1–4 are loaded into 384-well plate 1. DNA plates 5–8 are loaded into 384-well plate 2, and continue loading up to 384-well plate 12. 2. Prepare one set of 384-well plates per assay plus two to four extra sets, i.e., 50–52 sets of twelve 384-well plates. Preparing additional plates is advantageous for possible reanalysis and for future testing of newly developed assays. l
3. All 384-well plates are prepared ahead of time; DNA is loaded and allowed to dry overnight. We have successfully genotyped dried DNA plates stored away from light for up to 9 months. 3.1.2. Primer/Probe Mix and Reagent Loading
To maximize efficiency and minimize reagent waste, all DNA plates for each assay are cycled together. To achieve high efficiency, at least the same number of thermal cycling blocks as sets of 384well plates are needed for each day’s work. 1. For each SNP assay, mix the primers and probes as an 80x primer/probe mix (described in Section 2.2) or use the mix purchased from Applied Biosystems. 2. Prepare the following reaction mix for twelve 384-well plates by mixing: 2x TaqMan universal PCR master mix (12,560 mL) (see Note 2 for potential cost savings by reducing the final concentration). 80x primer/probe mix (314 mL). H2O (12,246 mL). 3. Add 5 mL of the above mixture to each well using a repeating 16-channel pipettor. 4. Seal each plate using an optical adhesive cover (see Note 3).
3.1.3. Offline Thermal Cycling
3.1.4. End-Point Read
1. Initial denaturation: 10 min at 95C. 2. Fifty cycles of: 95C for 1 s and 60C for 1 min at a ramp speed of 1C per second. After thermal cycling, load the plates into a 384-well real-time instrument using the automation accessory. The final fluorescence is measured, and the results are analyzed using the instrument software (see Note 4 for a discussion of the quality control protocol).
SNP Genotyping by the 50 -Nuclease Reaction
3.2. Experimental Design for MediumDensity Array (Fluidigm 96.96)
285
Here we establish a simple experimental design for multiplex genotyping using the Fluidigm 96.96 Dynamic Array (http://www.flui digm.com/biomark_genotyping.htm). The Dynamic Array contains a matrix of integrated channels and valves housed in an input frame (Fig. 18.3). On one side of the frame are 96 inlets to accept the sample DNA from 96 individuals, and on the other are 96 inlets to accept the assays for up to 96 SNP markers. Once the components are in the inlets, they are pressurized into the array using the Fluidigm IFC controller HX. The 96 samples and 96 assays are then systematically combined into 9,216 parallel reactions.
Fig. 18.3. The Fluidigm 96.96 Dynamic Array for multiplexing TaqMan assays. The array consists of a matrix of channels and valves housed in an input frame with the footprint of a standard 384-well plate. Up to 96 samples are loaded into the inlets on one side of the array and up to 96 assays are loaded into the other inlets. The samples and assays are pressure-loaded into 9,216 reaction chambers for thermal cycling.
A typical laboratory using a Fluidigm instrument has four offline IFC thermal cyclers that are specifically designed to accept the Dynamic Arrays. Running these three times per day will yield 110,592 genotypes. All of the genotypes for each sample are acquired daily, so the workflow design is simpler than that for uniplex laboratories. 3.2.1. Preparation of Sample Mixture
1. Combine the following for each 96-well plate of DNA in a 1.5-mL tube (sample mixture): 2x TaqMan universal PCR master mix (275 mL).
286
Seeb et al.
AmpliTaq Gold DNA polymerase (5.5 mL). 20x GT sample loading reagent (27.5 mL). DNA/DNAse-free H2O (11 mL). 2. Mix and add 2.9 mL of sample mixture to a 96-well plate. 3. Add 2.1 mL of each DNA sample to the 96-well plate with the sample mixture. For very low concentrations of DNA, a preamplification step may be necessary (see Note 1). 4. Mix and load the solution into the sample inlets on the array (see Notes 5 and 6). 3.2.2. Preparation of Assay Mixture
1. Combine the following for each 80x primer/probe mix (described in Section 2.2). For ease of pipetting, this mixture is prepared for four arrays: 2X DA assay loading reagent (1,100 mL). 50X ROX (110 mL). DNA/DNAse-free H2O (715 mL). 2. Mix and add 17.5 mL into a 96-well plate. 3. Add 2.5 mL of each 80X primer/probe mix to the 96-well plate. 4. Mix and load 5 mL into the assay inlets on the array (see Notes 6 and 7).
3.2.3. Thermal Cycling on a Fluidigm IFC Cycler
1. Load the sample mixtures into one set of inlets and the assay mixture into the opposite set of inlets using the automated liquid handler. 2. Thermal mixing: 50C for 2 min, 70C for 30 min, 25C for 10 min. 3. Initial denaturation: 95C for 10 min. 4. Thermal cycling: 50 cycles of 92C for 15 s and 60 for 1 min.
3.2.4. End-Point Read
The Dynamic Arrays are read on a Fluidigm BioMark or EP1 system after amplification and scored using BioMark or EP1 genotyping analysis software available from Fluidigm.
3.3. Minimizing TaqMan Genotyping Error
Genotyping error can occur either as a result of sample handling or as a result of TaqMan error. Robotic loading helps to reduce handling error; still, we strongly recommend use of quality control protocols to understand and control this (see Note 4). The TaqMan error rate, especially in microliter reaction volumes with quantitated DNA, is invariably very low. Interestingly, both Ranade et al. (10) and Tranah et al. (24) reported error rates of
SNP Genotyping by the 50 -Nuclease Reaction
287
0.05%. Errors have been attributed to factors including buffer conditions or contaminants in the DNA (10). Because the rates are low, the error rate is often ignored and seldom reported. However, we observed elevated error rates when we initially evaluated TaqMan arrays using nanovolume reactions. One of the advantages of TaqMan is the huge range of DNA concentrations suitable for assay, at least 5 orders of magnitude (9). For this reason, laboratories analyzing thousands of samples do not bother with the time-consuming steps necessary to normalize DNA concentrations. Our concern when testing arrays was that low DNA concentrations, suitable for microliter reactions in 384well plates, might not perform well when distributed across 48 or more nanoliter chambers. At low copy numbers, stochastic fluctuation in the number of copies of each allele might result in preferential amplification (25). This might increase the error rate in samples of low concentration, with some heterozygotes appearing to be homozygotes. To test for this phenomenon, we conducted digital PCR experiments (23, 26) to examine copy number effects using the Fluidigm Digital Array. We often observed that the heterozygote scatter became more diffuse, and some known heterozygotes were scored as homozygotes at less than or equal to ten copies per reaction chamber (Table 18.2 Fig. 18.4). Improvements to the original Dynamic Arrays that we initially tested include enhanced mixing of the liquid assay and liquid DNA sample, and, subsequently in 38,000 paired observations, we observed an error rate of 0.07% in nonquantitated DNA from
Table 18.2 Genotype calls in a dilution experiment of 56 DNA samples to test the relationship between copy number per reaction and accuracy of a call Copy number/ reaction
XY
YY
40
5
24
27
0
0
10
5
22
29
0
2
1
16
4
21
15
9
8
0
4
44
5
0.2
Failed calls
Wrong callsa
XX
a All 56 are called correctly at 40 copies per reaction. Some heterozygotes are wrongly called homozygotes at fewer than ten copies per reaction. Amplification drops off sharply at 0.2 copies per reaction.
288
Seeb et al.
A. 40 Copies
B. 10 Copies
C. 1 Copy
D. 0.2 Copy
Fig. 18.4. Scatter plots of a dilution experiment of 56 DNA samples to test the relationship between copy number per reaction and accuracy of the call. No template controls and failed amplifications spot at the origin in the lower left of each plot. (a) At 40 copies per reaction, genotypes cluster discretely, and all calls are correct. (b) At ten copies per reaction, the heterozygote cluster noticeably becomes more diffuse and two heterozygotes are wrongly called YY homozygotes. (c) At one copy per reaction, some samples are not amplified and many heterozygotes are scored as either XX or YY homozygotes. (d) At 0.2 copies per reaction, amplification is poor, and no heterozygotes are called.
routine exactions (Table 18.3). It is still uncertain if this source of error can be ameliorated when it occurs in arrays that use dried assay. An additional advantage of the quality control protocol described in Note 4 is that large data sets, similar to the data set in Table 18.3, will accumulate. Such data sets will help us better understand TaqMan performance. For example, it easy to understand how rare miscalls may occur between adjacent clusters (10, 24) (Fig. 18.4b). It is more difficult to understand XX–YY homozygote–homozygote miscalls. Two of these mismatched scores were observed in the 38,000 paired calls in Table 18.3.
SNP Genotyping by the 50 -Nuclease Reaction
289
Table 18.3 TaqMan genotyping errors observed as mismatched scores in over 38,000 paired calls. This is a compilation of errors observed in paired single nucleotide polymorphism (SNP ) scores from the following platforms: (1) AB 7900HT, (2) Fluidigm 48.48 Dynamic Array, and (3) Fluidigm 96.96 Dynamic Array. The pair of methods is listed in the first column followed by total error. Scores are further broken down into homozygote/heterozygote and homozygote/homozygote miscalls
a
(XX or YY) $ XY
XX $ YY
Methods
Total no. of SNPs read
Total error (%)
n
Error (%)
n
Error (%)
11
16965
0.02
3
0.02
1
< 0.01
1 (2,3)
11970
0.06
7
0.06
0
0.00
22
2103
0.14
3
0.14
0
0.00
2 3a
3363
0.42
13
0.39
1
0.03
33
4288
0.02
1
0.02
0
0.00
Total
38689
0.07
27
0.07
2
<0.01
Ten of 14 failures from one individual (one of 95)
4. Notes 1. Approximately 2–20 ng of genomic DNA per 5 mL TaqMan reaction are recommended for the AB 7900HT (8). Fluidigm recommends 126 ng of genomic DNA per inlet for the Fluidigm 96.96 Dynamic Array. Several situations exist where the recommended amount of genomic DNA is not available. Examples include limited tissue size/number of cells, such as fish scales and hair follicles, or highly degraded samples, such as historical samples (27). To address this limitation, a preamplification step is included as a multiplex PCR where all genotyping primers are included at a low concentration (0.2 mM vs 0.9 mM for the TaqMan reaction). The samples are amplified between ten and 14 cycles and then diluted 1:20 before the TaqMan genotyping reactions are run. 2. Dilution tests can help to reduce costs of uniplex reactions (28). The relative cost per genotype among platforms discussed in this chapter is shown in Table 18.1. A majority of the total cost of a uniplex reaction is the TaqMan PCR master mix. In an attempt to lower per reaction cost, we titrated the master mix between 0.5x and 1x. We found that most assays
290
Seeb et al.
performed reliably at 0.7x. For a small set of low-performing assays, lowering the master mix concentrations was detrimental, so they were kept at the 1x concentration. This is a simple experiment that can lead to substantial cost savings. 3. The 384-well plates can be loaded with reagent mix, sealed and stored at 4C for up to 72 h before cycling (8). 4. High-throughput genotyping necessitates a quality control protocol. A typical project run in our laboratory has 48 96well plates run for 48 assays for a total of 221,184 genotypes. Forty-eight plates (96 48=4,608 samples or 95 48= 4,560 samples with no template controls) need to be entered into the laboratory database, labeled, extracted, and loaded for genotyping. After the genotyping, every plate is scored by two researchers, and the genotypes are entered into the laboratory database. Our quality control protocol consists of a reanalysis of eight samples per 96-well plate (approximately 8%) for all assays. Samples from well positions A3–A10 in each project DNA plate are transferred to a quality control plate: l A3–A10 from project DNA plate 1 are placed in column 1 in the quality control plate. A3–A10 from project DNA plate 2 are placed in column 2 in the quality control plate, and this is continued for all 48 project plates, making four 96-well quality control plates or one 384-well quality control plate. The quality control procedure will measure and identify errors caused by incorrect sample identification, incorrect orientation of the 96-well plate during processing (extraction and loading), and incorrect assay identification. The quality control procedure and the original genotyping are done by two different researchers. This procedure does not check for or measure errors created during sample extraction.
l
5. One inlet is used as a no-template control where DNA is replaced with TE buffer. 6. It may be useful to have larger batches of sample mix and assay mix. Sample mix for approximately 50 arrays for the Fluidigm 96.96 Dynamic Array will include: 2x TaqMan universal PCR master mix (14,400 mL). AmpliTaq Gold DNA polymerase (288 mL). 20x GT sample loading reagent (1,440 mL). DNA/DNAse-free H2O (576 mL). Mix the solution thoroughly and dispense the mixture into 16 wells in a deep-well 96-well plate. Mix 2.9 mL of the sample mix with 2.1 mL DNA and load it into the sample inlets.
SNP Genotyping by the 50 -Nuclease Reaction
291
Assay mix for approximately 40 arrays will contain: 2x DA loading buffer (10,000 mL). 50x ROX (1,000 mL). DNA/DNAse-free H2O (6,500 mL). Mix the solution thoroughly and dispense 175 mL of the mixture into a 96-well plate using a repeating eight- or 12channel pipettor. Add 25 mL of each 80X primer/probe mix. Replace the primer/probe mix with TE buffer if there are fewer than 96 SNP Assays available. Load 5 mL of assay mix into the assay inlets. 7. When fewer than 96 assays are available, replace 80x primer/ probe mix with TE buffer. Every inlet must contain reagent.
Acknowledgement This work was supported by a grant from the Gordon and Betty Moore Foundation to J.E.S. and L.W.S. References 1. Holland, P. M., Abramson, R. D., Watson, R. and Gelfand, D. H. (1991) Detection of specific polymerase chain reaction product by utilizing the 50 30 exonuclease activity of Thermus aquaticus DNA polymerase. Proc. Natl. Acad. Sci. U.S.A. 88, 7276–7280. 2. Holland, P. M., Abramson, R. D., Watson, R., Will, S., Saiki, R. K., and Gelfand, D. H. (1992) Detection of specific polymerase chain reaction product by utilizing the 50 30 exonuclease activity of Thermus aquaticus DNA polymerase. Clin. Chem. 38, 462–463. 3. Hauser, L. and Seeb, J. E. (2008) Advances in molecular technology and their impact on fisheries genetics. Fish Fish. 9, 473–486. 4. Moen, T., Hayes, B., Nilsen, F., Delghandi, M., Fjalestad, K. T., Fevolden, S. E. et al. (2008) Identification and characterisation of novel SNP markers in Atlantic cod: Evidence for directional selection. BMC Genetics 9, 18. 5. Narum, S. R., Banks, M., Beacham, T. D., Bellinger, M. R., Campbell, M. R., Dekoning, J. et al. (2008) Differentiating salmon populations at broad and fine geographical scales with microsatellites and single
6.
7.
8.
9.
10.
11.
nucleotide polymorphisms. Mol. Ecol. 17, 3464–3477. Schlotterer, C. (2004) The evolution of molecular markers – just a matter of fashion? Nat. Rev. Genet. 5, 63–69. Seddon, J. M., Parker, H. G., Ostrander, E. A. and Ellegren, H. (2005) SNPs in ecological and conservation studies: a test in the Scandinavian wolf population. Mol. Ecol. 14, 503–511. Livak, K. J. (2003) SNP genotyping by the 50 -nuclease reaction. In Kwok, P.-Y. (Ed.) Single Nucleotide Polymorphisms: Methods and Protocols. Methods in Molecular Biology, Totowa, Humana Press, pp. 129–147. Heid, C. A., Stevens, J., Livak, K. J. and Williams, P. M. (1996) Real time quantitative PCR. Genome Res. 6, 986–994. Ranade, K., Chang, M. S., Ting, C. T., Pei, D., Hsiao, C. F., Olivier, M. et al. (2001) High-throughput genotyping with single nucleotide polymorphisms. Genome Res. 11, 1262–1268. Smith, C. T., Elfstrom, C. M., Seeb, L. W. and Seeb, J. E. (2005) Use of sequence data from rainbow trout and Atlantic salmon for
292
12.
13.
14.
15.
16.
17.
18.
19.
Seeb et al. SNP detection in Pacific salmon. Mol. Ecol. 14, 4193–4203. The International HapMap Consortium. (2007) A second generation human haplotype map of over 3.1 million SNPs. Nature 449, 851-U3. Kwok, P.-Y. (2003) Single nucleotide polymorphisms; Methods and Protocols, Methods in Molecular Biology, vol. 212, Totowa, Humana Press. Smith, C. T. and Seeb, L. W. (2008) Number of alleles as a predictor of the relative assignment accuracy of short tandem repeat (STR) and single-nucleotide-polymorphism (SNP) baselines for chum salmon. Trans. Am. Fish. Soc. 137, 751–762. Fox, C. J., Taylor, M. I., Pereyra, R., Villasana, M. I. and Rico, C. (2005) TaqMan DNA technology confirms likely overestimation of cod (Gadus morhua L.) egg abundance in the Irish Sea: implications for the assessment of the cod stock and mapping of spawning areas using egg-based methods. Mol. Ecol. 14, 879–884. Sprowles, A. E., Stephens, M. R., Clipperton, N. W. and May, B. P. (2006) Fishing for SNPs: A targeted locus approach for single nucleotide polymorphism discovery in rainbow trout. Trans. Am. Fish. Soc. 135, 1698–1721. Weidmann, M., Armbruster, K. and Hufert, F. T. (2008) Challenges in designing a Taqman-based multiplex assay for the simultaneous detection of herpes simplex virus types 1 and 2 and Varicella-zoster virus. J. Clin. Virol. 42, 326–334. Dahl, A., Sultan, M., Jung, A., Schwartz, R., Lange, M., Steinwand, M. et al. (2007) Quantitative PCR based expression analysis on a nanoliter scale using polymer nano-well chips. Biomed. Microdevices 9, 307–314. Spurgeon, S. L., Jones, R. C. and Ramakrishnan, R. (2008) High throughput gene
20.
21.
22.
23.
24.
25.
26.
27.
28.
expression measurement with real time PCR in a microfluidic dynamic array. PLoS ONE 3, e1662. Osman, F., Leutenegger, C., Golino, D. and Rowhani, A. (2008) Comparison of lowdensity arrays, RT-PCR and real-time TaqMan1 RT-PCR in detection of grapevine viruses. J. Virol. Methods 149, 292–299. Tadros, S. F., D’souza, M., Zhu, X. X. and Frisina, R. D. (2008) Apoptosis-related genes change their expression with age and hearing loss in the mouse cochlea. Apoptosis 13, 1303–1321. Perkel, J. (2008) SNP genotyping: six technologies that keyed a revolution. Nat. Methods 5, 447–453. Qin, J., Jones, R. C. and Ramakrishnan, R. (2008) Studying copy number variations using a nanofluidic platform. Nucleic Acids Res. 36, e16. Tranah, G. J., Lescault, P. J., Hunter, D. J. and De Vivo, I. (2003) Multiple displacement amplification prior to single nucleotide polymorphism genotyping in epidemiologic studies. Biotechnol. Lett. 25, 1031–1036. Walsh, P. S., Erlich, H. A. and Higuchi, R. (1992) Preferential PCR amplification of alleles: mechanisms and solutions. Genome Res. 1, 241–250. Dube, S., Qin, J. and Ramakrishnan, R. (2008) Mathematical analysis of copy number variation in a DNA sample using digital PCR on a nanofluidic device. PLoS ONE 3, e2876. Morin, P. A. and McCarthy, M. (2007) Highly accurate SNP genotyping from historical and low-quality samples. Mol. Ecol. Notes 7, 937–946. Morin, P. A., Saiz, R. and Monjazeb, A. (1999) High-throughput single nucleotide polymorphism genotyping by fluorescent 50 exonuclease assay. Biotechniques 27, 538–552.
Chapter 19 The TaqMan Method for SNP Genotyping Gong-Qing Shen, Kalil G. Abdullah, and Qing Kenneth Wang Abstract Single nucleotide polymorphisms (SNPs) are common DNA sequence variations that occur at single bases within the genome. SNPs have been instrumental in elucidating the genetic basis of common, complex diseases using genome-wide association studies, candidate gene case-control association studies, and genome-wide linkage analyses. A key to these studies is genotyping of SNPs. Various methods for SNP genotyping have been developed. For a particular genotyping project, the choice of method is dependent on the number of SNPs (n) and the number of DNA samples (m) to be genotyped. For a genome-wide or large-scale project with very high n and small m, the Affymetrix SNP GeneChip and Illumina GoldenGate BeadChips assays are the ideal methods. For a project involving a small number of SNPs (small n) and a large population (high m), the TaqMan assay is the preferred technology as it has high throughput and is highly accurate, precise, time-efficient, and cost-effective. Here, we describe the detailed procedures for TaqMan SNP genotyping assay, including preparation of high-quality DNA samples, the operating protocol, clarification of technical issues, and discussion of several cautionary notes. Key words: Single nucleotide polymorphisms, TaqMan, genotyping, case-control association study, genome-wide association study, genetic variation, susceptibility gene.
1. Introduction The most common diseases in the developed world, such as coronary artery disease, hypertension, diabetes, and obesity, are caused by various genetic and environmental factors as well as their interactions. Genetic factors have been defined as an important risk contributor for the pathogenesis of these complex diseases, but the genetic determinants responsible remain largely unidentified (1–4). The most common human genetic variations in the genome are single nucleotide polymorphisms (SNPs). A significant number of SNPs may be associated with complex A.A. Komar (ed.), Single Nucleotide Polymorphisms, Methods in Molecular Biology 578, DOI 10.1007/978-1-60327-411-1_19, ª Humana Press, a part of Springer Science+Business Media, LLC 2003, 2009
293
294
Shen, Abdullah, and Wang
human traits and some of them may be direct, causative factors (5–7). Identification of SNPs that contribute to the risk of acquiring complex diseases has and will continue to provide valuable information important for early diagnosis, prevention, and treatment of common diseases (8–10). Because of their high frequencies in the human populations and the ease of their analysis, SNPs have been used in case-control association studies to identify susceptibility genes for many diseases (11–19). Recently, SNPs have been successfully used as genetic markers to identify genetic loci for various diseases using both genome-wide association studies and genome-wide linkage analysis. Analysis of high-density whole genome SNPs in a large number of samples has become a valuable technique that can identify genetic variants associated with common, complex diseases (20–27). Although SNPs serve as a powerful tool for genetic research, it has become apparent that it is difficult to genotype thousands of SNP for large-scale case-control association studies. Highthroughput, highly accurate and cost-effective SNP genotyping methods have become increasingly popular in the last several years (28–33). Here, we focus on the TaqMan SNP genotyping assay because it is an advanced, mature, validated, and widely used technology (34–36). TaqMan SNP genotyping assay can be performed using an ABI Prism 7900HT sequence detection system (Applied Biosystems, Foster City, CA, USA) at high throughput and with low running costs. The TaqMan SNP genotyping assay is also called the ‘‘50 -nuclease allelic discrimination assay’’ and requires forward and reverse PCR primers, and two differently labeled TaqMan minor groove binder (MGB) probes. The essence of the TaqMan assay was described in detail in Chapter 18. In brief, the biallelic SNP is located in the middle third of the probe. Each allele-specific MGB probe is labeled with a fluorescent reporter dye (either a FAM or a VIC reporter molecule) and is attached with a fluorescence quencher. For the intact MGB probe, the reporter dye is quenched. During PCR, the 50 -nuclease activity of Taq DNA polymerase cleaves the reporter dye (FAM or VIC) from an MGB probe that is completely hybridized to the DNA strand. When separated from the quencher, the reporter dye fluoresces. However, if a single point mismatch is present between the probe and the target DNA strand due to a SNP, the binding of the probe to DNA is destabilized during strand displacement in PCR, which will reduce the efficiency of probe cleavage and quenching of the fluorescent reporter dye. Therefore, an increase in either FAM or VIC dye fluorescence indicates homozygosity for FAM- or VICspecific alleles (X:X or Y:Y), and an increase in the fluorescence of both dyes indicates heterozygosity (X:Y). The ABI Prism 7900HT system can distinguish and quantify the two colors (FAM or VIC),
The TaqMan Method for SNP Genotyping
295
and genotypes can be generated for many samples in a few minutes. A 384-well plate makes the ABI Prism 7900HT system ideally suited to meet the high-throughput requirements. Here we focus our attention on the detailed description of the procedure, which would allow any new researcher to easily operate the ABI Prism 7900HT system and analyze the data using the SDS 2.1 software program.
2. Materials 2.1. DNA
1. DNA isolation kit for mammalian blood (Roche Applied Science, Indianapolis, IN, USA) (see Note 1). 2. 1 TE buffer for DNA dilution: 10 mM tris(hydroxymethyl)aminomethane (Tris)–HCl, 1 mM EDTA, pH 8.0. Prepare it fresh from stock solutions of 1 M Tris–HCl pH 8.0 and 0.5 M EDTA, pH 8.0. Store it at room temperature. 3. 96-well and 384-well plates can be ordered from Denville Scientific (Metuchen, NJ, USA). 4. NanoDrop ND-1000 spectrophotometer from NanoDrop Technologies. (Wilmington, DE, USA). It is a full-spectrum spectrophotometer that measures DNA concentrations using 1 mL of DNA samples with high accuracy and reproducibility. 5. Hydra II, a 96-channel microdispenser with the X/Y plate stage system and syringe-based liquid handling (Thermo Fisher Scientific, Hudson, NH, USA).
2.2. Polymerase Chain Reaction
1. TaqMan genotyping master mix purchased from Applied Biosystems (Foster City, CA, USA). 2. Adhesive PCR sealing foil sheets ordered from Thermo Scientific (ABgene, Epsom, UK). 3. GeneAmp PCR system 9700 with 384 wells (Applied Biosystems, Foster City, CA, USA).
2.3. SNP Genotyping
1. TaqMan SNP genotyping assays (including both Assays-onDemand and Assays-by-Design) from Applied Biosystems (Foster City, CA, USA). 2. MicroAmpTM optical adhesive covers from Applied Biosystems (Foster City, CA, USA). 3. ABI Prism 7900HT sequence detection system (Applied Biosystems, Foster City, CA, USA). 4. Data analysis: SDS 2.1 software program (Applied Biosystems, Foster City, CA, USA).
296
Shen, Abdullah, and Wang
3. Methods 3.1. Preparation of High-Quality DNA Samples
1. Transfer 10 mL of mammalian whole blood stored in EDTA or heparin to a sterile 50-mL centrifuge tube and add 30 mL of red blood cell lysis buffer. 2. Gently shake the solution for 10 min, and centrifuge the tube at 900g for 10 min at 4C. 3. Remove the supernatant, add 5 mL white blood cell lysis buffer, vortex the solution thoroughly, and incubate the solution for 30 min at 65C. 4. Add 2.6 mL protein precipitation solution, vortex the solution thoroughly, and centrifuge the sample at 12,000g for 25 min at 4C (see Note 2). 5. Transfer the supernatant into a new sterile 50-mL centrifuge tube, add 2 vol of room temperature 100% ethanol, and gently mix the solution by inversion until DNA strands precipitate out of solution. 6. Carefully remove the DNA strands from the 100% ethanol using a pipette tip or toothpick and transfer them to a new 1.5-mL micro centrifuge tube containing 1 mL cold 70% ethanol, gently mix the sample two or three times, and centrifuge the tube at 12,000g for 10 min at 4C. 7. Discard the supernatant, and air-dry the DNA pellet for 3–4 h. 8. Add 1 mL of TE buffer, vortex the solution, and incubate it at 65C for 2 h. 9. Measure the concentration of DNA using a spectrophotometer (e.g., NanoDrop ND-1000 spectrophotometer) or a fluorometer; Store the DNA samples in a 4C refrigerator or a –80C freezer until use.
3.2. Addition of DNA Samples to 384-Well PCR Plates
Owing to the requirement of high-throughput SNP genotyping, 384-well reaction plates not only facilitate the screening of more samples simultaneously but also save costs for reagents and other materials. Because of the narrow and dense wells of the plate, loading of DNA samples is a time-consuming process if done manually. The Hydra II 96-channel microdispenser with an X/Y plate stage system (Thermo Fisher Scientific, Hudson, NH, USA) is an instrument with syringe-based liquid handling, which provides high-speed, repetitive aspirating and dispensing of microliter and nanoliter volumes into microplates and other receptacles. The machine can handle up to 96 channels of sample dispensing at the same time, and performs aspiration, dispensing, washing, and emptying functions. The machine also has two manually operated plate positioners for dispensing into 384-well plates.
The TaqMan Method for SNP Genotyping
297
1. Dilute DNA to a concentration of 12.5 ng/mL. 2. Add diluted DNA of concentration 12.5 ng/mL into a 96well plate. 3. Press the power switch on the Hydra II microdispenser to ON. 4. Use the buttons on the front of the microdispenser to edit files (the file number from 1 to 16), including the settings for the operations of dispense, aspirate, empty, and wash. 5. Press Wash to start the washing operation. The syringe needles should be washed five times before aspiration. 6. Press Aspr to start the aspirate operation. 7. Press Disp to start the dispense operation. 8. Set the tray table/stage height. (a) Click Page so the Dispense screen appears on the display. (b) From the Dispense screen, press Edit until the Table is on the top line of the screen. (c) Press Up (slowly) to move the table/stage up until the needle tips are just slightly in contact with the bottom of the wells in the plate (see Note 3). If the plate lifts easily, the tips are not sufficiently near the bottom of the wells. Continue pressing Up (slowly) until the plate cannot be lifted easily. 9. Set the file parameters. (a) From the Ready Mode screen, press Up until File 1 appears on the screen. (b) Click Page, and the Dispense screen should then be on the display. (c) From the Dispense screen, press Edit until Vol is on the top line of the screen. Set Vol to 10.0/40.0/70.0/120.0 by pressing Up or Down. (d) Press Edit until Table is on the top line of the screen. Set Table to 3500 by pressing Up or Down. (e) Press Edit until Speed is on the top line of the screen. Set Speed to 5 by pressing Up or Down. (f) Click Page, and the Aspirate screen should then be on the display. (g) From the Aspirate screen, press Edit until Prime is on the top line of the screen. Set Prime to ON by pressing Up or Down. (h) Press Edit until Vol is on the top line of the screen. Set Vol to 50.0/170.0/290.0/500.0 by pressing Up or Down (i) Press Edit until Air is on the top line of the screen. Set Air to 0 by pressing Up or Down.
298
Shen, Abdullah, and Wang
(j) Press Edit until Table is on the top line of the screen. Set Table to 2700 by pressing Up or Down. (k) Press Edit until Speed is on the top line of the screen. Set Speed to 5 by pressing Up or Down. (l) Click Page, and the Empty screen should then be on the display. (m) From the Empty screen, press Edit until Table is on the top line of the screen. Set Table to 2500 by pressing Up or Down. (n) Press Edit until Speed is on the top line of the screen. Set Speed to 5 by pressing Up or Down. (o) Click Page, and the Wash screen should then be on the display. (p) From the Wash screen, press Edit until Wash is on the top line of the screen. Set Wash to 3 by pressing Up or Down. (q) Press Edit until Vol is on the top line of the screen. Set Vol to 60.0/190.0/340.0/560.0 by pressing Up or Down. (r) Press Edit until Table is on the top line of the screen. Set Table to 6500 by pressing Up or Down. (s) Press Edit until Speed is on the top line of the screen. Set Speed to 5 by pressing Up or Down. (t) Click Page, and the display should then return to the Ready Mode screen. 10. Running a procedure. (a) Place a 96-well DNA sample plate on the tray table/stage, then press Aspr. The syringes will then aspirate the total amount of DNA in each well. (b) Remove the plate from the table/stage. (c) Place a positioner on the Hydra II 96 tray table. Load the first 384-well receiving plate onto the positioner. (d) Turn the thumbwheel on the front of the positioner to move it to one of four positions. (e) Press Disp on the right side of the microdispenser. The DNA is dispensed into the 384-well plate. (f) Remove the plate from the table/stage. Continue to place receiving plates until the last receiving plate is full. (g) Press Empty. (h) Press Wash to wash the syringes. (i) Place another 96-well DNA sample plate on the tray table/stage, then press Aspr. The syringes will then aspirate the total amount of DNA. (j) Remove the plate from the table/stage.
The TaqMan Method for SNP Genotyping
299
(k) Place a positioner on the Hydra II 96 tray table. Load the first 384-well receiving plate onto the positioner. (l) Turn the thumbwheel on the front of the positioner to move it to one of four positions. (m) Press Disp on the right side of the microdispenser. The DNA is dispensed into the 384-well plate. (n) Repeat it four times by dispensing four 96-well DNA plates to each of the four sets into one 384-well plate. 3.3. Design of TaqMan SNP Genotyping Assays
The TaqMan method for high-throughput SNP genotyping using the ABI Prism 7900HT sequence detection system employs the 50 nuclease allelic discrimination assay ((37–42); see also Chapter 18 in this volume). The assay includes the forward target-specific PCR primer, the reverse primer, and the TaqMan MGB probes labeled with two special dyes – FAM and VIC (see Note 4). Assays-onDemand and Assays-by-Design services are offered by Applied Biosystems. Applied Biosystems makes available 4.5 million TaqMan SNP assays (including 3.5 million HapMap SNPs assays) and offers over 10,000 Assays-on-Demand SNP genotyping assays. If your SNPs of interest are not included in the Assays-on-Demand database, you may also request SNP probes by Assays-by-Design, a service of custom TaqMan SNP design based on your DNA sequences. 1. Assays-on-Demand (see Note 5). (a) Go to http://www.appliedbiosystems.com (b) Register and log in. (c) Enter TaqMan SNP genotyping assays and assay search. (d) Search by gene symbol, gene name, public accession number, and multiple assay ID. (e) Choose SNP assay ID and add the product to ‘‘My Baskets/Orders.’’ (f) Proceed to checkout. 2. Assays-by-Design (see Note 6). (a) Choose the DNA target sequence. The sequence length is usually between 300 and 600 bases in the 50 –30 orientation. (b) Designate the target SNP site with square brackets, e.g., ACCA[G/A]CTAG. (c) Provide a suitable name for the sequence record. (d) Save the DNA sequence and your sequence name in a new folder. (e) Go to http://www.appliedbiosystems.com
300
Shen, Abdullah, and Wang
(f) Register and log in. (g) Transfer your folder to ‘‘My Baskets/Orders’’. (h) Submit the order. 3. Guidelines for the TaqMan primer and probe design. (a) Design a TaqMan primer. l Keep the G–C content in the 30–80% range. l
Avoid runs of an identical nucleotide. This is especially true for G, where runs of four or more Gs should be avoided.
l
The melting temperature should be 58–60C.
l
l
The five nucleotides at the 30 end should have no more than two G and/or C bases. Place the forward and reverse primers as close as possible to the probe without overlapping the probe.
(b) Design a TaqMan probe. l Keep the G–C content in the 20–80% range.
3.4. PCR
l
Select the strand that gives the probe more Cs than Gs.
l
Do not put G on the 50 end.
l
Avoidrunsofanidenticalnucleotide.Thisisespeciallytrue for G, where runs of four or more Gs should be avoided.
l
The melting temperature should be 65–67C.
l
Position the polymorphic site approximately in the middle third of the sequence.
l
Adjust the probe lengths so that both probes have the same temperature.
PCR is carried out in the GeneAmp PCR system 9700 with 384 wells (Applied Biosystems, Foster City, CA, USA), and the PCR mixture is assembled as described below: 1. Assays-on-Demand (a) TaqMan genotyping master mix: 2.5 mL (b) TaqMan SNP genotyping assay mix (40X): 0.125 mL (c) DNA samples (12.5 ng/mL): 2 mL (d) Water: 0.375 mL (e) Total: 5 mL 2. Assays-by-Design (a) TaqMan genotyping master mix: 2.5 mL (b) SNP genotyping assay mix (40X): 0.125 mL (c) DNA samples (12.5 ng/mL): 2 mL (d) Water: 0.375 mL (e) Total: 5 mL
The TaqMan Method for SNP Genotyping
301
3. PCR amplification (a) Press the Power Switch on the front of machine to the ON position. (b) From the Main menu screen, press User to display the Select User Name screen. Press New to add a new name to the list. (c) From the User Name screen, enter a unique name of up to six characters in length. (d) Press Accept to accept a name. (e) From the Main menu screen, press User to display the User Name screen. (f) From the User Name screen, select your name, and then click Accept. (g) From the Main menu screen, press Edit to display the View screen. (h) Enter the time and temperature, and then select Start. The time and temperature of PCR for TaqMan SNP genotyping assay: one cycle of denaturation at 95C for 10 min, 50 cycles of denaturation at 92C for 15 s and annealing at 60C for 60 s, and soak at 4C until use. (i) From the Select Method Options screen, enter a reaction volume based on the total amount of material in each well (standard volume of 5 mL), and then select Start again. 3.5. Genotyping
Genotyping or allele calling is carried out with the ABI Prism 7900HT genetic detection system: 1. Turn on the monitor and computer first, and then turn on the ABI Prism 7900HT instrument. If the instrument is active normally, the solid green light on the front of the instrument is on. 2. Select Start > Programs > Applied Biosystems > SDS 2.1 > SDS 2.1 or double-click the SDS 2.1 icon. 3. Select File > New. 4. Configure the New Document dialog box with the following settings: (a) In the Assay drop-down list, select Allelic Discrimination. (b) In the Container drop-down list, select 384 Wells Clear Plate. (c) In the Template drop-down list, select Blank Template. 5. Click OK. 6. From the SDS 2.1 software, select Tools > Detector Manager. 7. From the Detector Manager dialog box, select New.
302
Shen, Abdullah, and Wang
8. From the Add Detector dialog box, click the Name field, and enter a unique name for the detector. In the Reporter drop-down list, select FAM, then click OK to save the detector and return to the Detector Manager dialog box. 9. From the Detector Manager dialog box, select New again. 10. From the Add Detector dialog box, click the Name field, and enter a different name for the detector. In the Reporter dropdown list, select VIC, then click OK and return to the Detector Manager dialog box. 11. Press the Ctrl key on the keyboard, select the detectors you want, and highlight them, then click Create Marker. 12. From Marker Editor, click the Name field, type a new name, and then click OK. 13. From the Detector Manager dialog box, click Done. 14. From the SDS 2.1 software, select Tools > Marker Manager. 15. From the Marker Manager dialog box, search for the name you have just designated and highlight it. 16. Click Copy To Plate Document. 17. From the Copy Markers To Plate dialog box, make sure the name you selected is correct, then click OK. 18. From the Marker Manager dialog box, click Done. 19. From the Allelic Discrimination dialog box, use the Ctrl and Shift keys, and select the wells of the plate grid containing the wells you wish to analyze. 20. Click the Use check box of the marker you want to add to the selected wells. 21. From the Allelic Discrimination dialog box, click the Instrument. 22. From the Plate Read dialog box, first click Connect, and then click Open/Close. The instrument tray rotates to the out position. Transfer the PCR plate from the GeneAmp PCR system 9700 to the tray. Click Open/Close again. The instrument tray rotates to the in position. 23. Click Post Read. 24. When running finishes, click OK. 3.6. Data Analysis
1. Automatic allele calling. (a) Click the large green arrow on the left side. (b) Click the Results tab. (c) Click the small green arrow on the right side. (d) Click the Auto Caller Enabled.
The TaqMan Method for SNP Genotyping
303
(e) Select the Quality Value to suit the parameters of your study design. (f) Click OK (see Note 7). 2. Allele calling by yourself. (a) Click the Results tab. (b) Click the lasso tool. (c) Select the sample cluster. (d) From the Call drop-down list, select Allele X call. (e) Select the other two sample clusters by identifying them as Allele Y or Both based on their grid location. (f) From the File menu, select Print Report (Fig. 19.1).
4. Notes 1. Preparation of high-quality DNA is an important step for the TaqMan SNP genotyping assay. The ABI Prism 7900HT sequence detection system will be unable to perform data analysis after SNP genotyping if the DNA quality is poor. 2. Vortex continuously for 25 s. This is necessary for effective removal of proteins from the sample. 3. It is crucial to know how to set the tray table/stage heights, so that neither operator nor instrument is at risk of injury or damage when the microdispenser is operating. In general, do not allow the needles to contact the bottom of any plate, tube, reservoir, tray, or wash basin. Use the fingertips to gently tilt up the front edge of the plate, off the table/stage and toward the needle tips. 4. The primers are for amplifying the SNP of interest, and the probes are for detecting the two different alleles. 5. Assays-on-Demand is a service of Applied Biosystems that provides ready-made SNP genotyping assays. The customer must submit sequence information and an allele frequency. Assays are immediately available for over 100,000 SNPs in the Applied Biosystems database. 6. If the SNP of interest is not in the database, Applied Biosystems provides the Assays-by-Design service as an alternative. Assays-by-Design is a service that designs, synthesizes, formulates, and delivers analytically quality controlled probe sets for SNP genotyping assays based on specific information submitted by the customer.
304
Shen, Abdullah, and Wang
Fig. 19.1 Representative printout of results from a TaqMan single nucleotide polymorphism (SNP) genotyping assay. The DNA samples were genotyped by the TaqMan assay. The samples were divided into four different clusters after automatic allele calling. The allele calling was processed by the SDS 2.1 software program. The cluster at position X-3.8/Y-2.0 along the horizontal axis represents the samples homozygous for allele X (XX genotype). The cluster at position X-0.8/Y-8.8 along the vertical axis represents the samples homozygous for allele Y (YY genotype). The cluster at position X-3.7/Y-7.6 along the diagonal line represents the samples heterozygous for both X and Y alleles (XY genotype). The cluster at position X-0.6/Y-1.4 has negative controls without any DNA samples (note that DNA samples for which genotyping does not produce any result will fall into this cluster too).
7. A graph on the computer will show the results of the allelic discrimination plot by automatic allele calling. The horizontal axis indicates an allele XX homozygote, the vertical axis indicates allele YY homozygote, and the diagonal is allele X/Y heterozygote.
Acknowledgments This work was supported by NIH grants R01 HL66251, P50 HL77107, and P50 HL81011, and an American Heart Association Established Investigator award (to Q.K.W.). K.G.A. was
The TaqMan Method for SNP Genotyping
305
supported by a seed grant award from the American Medical Association Foundation and funding from the Cleveland Clinic Lerner College of Medicine. References 1. Wang, Q. (2005) Molecular genetics of coronary artery disease. Curr. Opin. Cardiol. 20, 182–188. 2. Topol, E. J., Smith, J., Plow, E. F. and Wang, Q. K. (2006) Genetic susceptibility to myocardial infarction and coronary artery disease. Hum. Mol. Genet. 15, R117–R123. 3. Shore, S. A. (2008) Obesity and asthma: possible mechanisms. J. Allergy Clin. Immunol. 121, 1087–1093. 4. Joy, T., Lahiry, P., Pollex, R. L. and Hegele, R. A. (2008) Genetics of metabolic syndrome. Curr. Diab. Rep. 8, 141–148. 5. Brookes, A. J. (1999) The essence of SNPs. Gene. 234, 177–186. 6. Hemminki, K. and Bermejo, J. L. (2005) Relationships between familial risks of cancer and the effects of heritable genes and their SNP variants. Mutat. Res. 592, 6–17. 7. Yamada, R. (2008) Primer: SNP-associated studies and what they can teach us. Nat. Clin. Pract. Rheumatol. 4, 210–217. 8. Suh, Y. and Vijg, J. (2005) SNP discovery in associating genetic variation with human disease phenotypes. Mutat Res. 573, 41–53. 9. Anderson, J. L., Carlguist, J. F., Horne, B. D. and Hopkins, P. N. (2007) Progress in unraveling the genetics of coronary artery disease and myocardial infarction. Curr. Atheroscler. Rep. 9, 179–186. 10. Wang, L., Luhm, R. and Lei, M. (2007) SNP and mutation analysis. Adv. Exp. Med. Biol. 593, 105–116. 11. Topol, E. J., McCarthy, J., Gabriel, S., Moliterno, D. J., Rogers, W., Newby, L. K. et al. (2001) Single nucleotide polymorphisms in multiple novel thrombospondin genes may be associated with familial premature myocardial infarction. Circulation 104, 2641–2644. 12. Ozaki, K., Ohnishi, Y., Iida, A., Sekine, A., Yamada, R., Tsunoda, T. et al. (2002) Functional SNPs in the lymphotoxin- gene that are associated with susceptibility to myocardial infarction. Nat. Genet. 32, 650–654. 13. Helgadottir, A., Manolescu, A., Thorleifsson, G., Gretarsdottir, S., Jonsdottir, H., Thorsteinsdottir, U. et al. (2004) The gene encoding 5-lipoxygenase activating protein
14.
15.
16.
17.
18.
19.
20.
21.
22.
confers risk of myocardial infarction and stroke. Nat. Genet. 36, 233–239. Connelly, J. J., Wang, T., Cox, J. E., Haynes, C., Wang, L., Shah, S. H. et al. (2006) GATA2 is associated with familial early-onset coronary artery disease. PloS Genet. 2, e139. Wang, L., Hauser, E. R., Shah, S. H., Pericak-Vance, M. A., Haynes, C., Crosslin, D. et al. (2007) Peakwide mapping on chromosome 3q13 identifies the kalirin gene as a novel candidate gene for coronary artery disease. Am. J. Hum. Genet. 80, 650–663. Shen, G. Q., Li, L., Girelli, D., Seidelmann, S. B., Rao, S., Fan, C. et al. (2007) An LRP8 variant is associated with familial and premature coronary artery disease and myocardial infarction. Am. J. Hum. Genet. 81, 780–791. Bouatia-Naji, N., Rocheleau, G., Van Lommel, L., Lemaire, K., Schuit, F., CavalcantiProenca, C. et al. (2008) A polymorphism within the G6PC2 gene is associated with fasting plasma glucose levels. Science 320, 1085–1088. Sun, T., Gao, Y., Tan, W., Ma, S., Shi, Y., Yao, J. et al. (2007) A six-nucleotide insertion-deletion polymorphism in the CASP8 promoter is associated with susceptibility to multiple cancers. Nat. Genet. 39, 605–613. Zhu, G., Vestbo, J., Lenney, W., Silverman, M., Whyte, M., Helms, P. et al. (2007) Association of PTGDR gene polymorphisms with asthma in two Caucasian populations. Genes Immun. 8, 398–403. McPherson, R., Pertsemlidis, A., Kavaslar, N., Stewart, A., Roberts, R., Cox, D. R. et al. (2007) A common allele on chromosome 9 associated with coronary heart disease. Science 316, 1488–1491. Helgadottir, A., Thorleifsson, G., Manolescu, A., Gretarsdottir, S., Blondal, T., Jonasdottir, A. et al. (2007) A common variant on chromosome 9p21 affects the risk of myocardial infarction. Science 316, 1491–1493. Saxena, R., Voight, B. F., Lyssenko, V., Burtt, N. P., de Bakker, P. I., Chen, H. et al. (2007) Genome-wide association
306
23.
24.
25.
26.
27.
28.
29.
30.
31.
Shen, Abdullah, and Wang analysis identifies loci for type 2 diabetes and triglyceride levels. Science 316, 1331–1336. Shojaee, S., Sina, F., Banihosseini, S. S., Kazemi, M. H., Kalhor, R., Shahidi, G.-A. et al. (2008) Genome-wide linkage analysis of a parkinsonian-pyramidal syndrome pedigree by 500 K SNP arrays. Am. J. Hum. Genet. 82, 1375–1384. Shen, G. Q., Li, L., Rao, S., Abdullah, K. G., Ban, J. M., Lee, B. S. et al. (2008) Four SNPs on chromosome 9p21 in a South Korean population implicate a genetic locus that confers high cross-race risk for development of coronary artery disease. Arterioscler. Thromb. Vasc. Biol. 28, 360–365. Shen, G. Q., Rao, S., Martinelli, N., Li, L., Olivieri, O., Corrocher, R. et al. (2008) Association between four SNPs on chromosome 9p21 and myocardial infarction is replicated in an Italian population. J. Hum. Genet. 53, 144–150. Abdullah, K. G., Li, L., Shen, G. Q., Hu, Y., Yang, Y., Mackinlay, K. G. et al. (2008) Four SNPS on chromosome 9p21 confer risk to premature, familial CAD and MI in an American Caucasian population (GeneQuest). Ann. Hum. Genet. 72, 654–657. Assimes, T. L., Knowles, J. W., Basu, A., Iribarren, C., Southwick, A., Tang, H. et al. (2008) Susceptibility locus for clinical and subclinical coronary artery disease at chromosome 9p21 in the multi-ethnic ADVANCE study. Hum. Mol. Genet. 17, 2320–2328. Jenkins, S. and Gibson, N. (2002) Highthroughput SNP genotyping. Comp. Funct. Genom. 3, 57–66. Tsuchihashi, Z. and Dracopoli, N. C. (2002) Progress in high throughput SNP genotyping methods. Pharmacogenomics J. 2, 103–110. Dearlove, A. M. (2002) High throughput genotyping technologies. Brief Funct. Genomic Proteomic. 1, 139–150. Ohnishi, Y. (2002) A high-throughput SNP typingsystemforgenome-wideassociationstudies. Gan To Kagaku Ryoho. 29, 2031–2036.
32. Shen, G. Q., Luo, A. and Wang, Q. K. (2006) High-throughput single-nucleotide polymorphisms genotyping: TaqMan assay and pyrosequencing assay. Methods Mol. Med. 128, 209–224. 33. Lee, J. E. (2007) High-throughput genotyping. Forum Nutr. 60, 97–101. 34. Hampe, J., Wollstein, A., Lu, T., Frevel, H. J., Will, M., Manaster, C. et al. (2001) An integrated system for high throughput TaqMan based SNP genotyping. Bioinformatics 17, 654–655. 35. Giles, J., Hardick, J., Yuenger, J., Dan, M., Reich, K. and Zenilman, J. (2004) Use of applied biosystems 7900HT sequence detection system and Taqman assay for detection of quinolone-resistant Neisseria gonorrhoeae. J. Clin. Microbiol. 42, 3281–3283. 36. Borodina, T. A., Lehrach, H. and Soldatov, A. V. (2004) Ligation detection reactionTaqMan procedure for single nucleotide polymorphism detection on genomic DNA. Anal. Biochem. 333, 309–319. 37. Holland, P. M., Abramson, R. D., Watson, R. and Gelfand, D. H. (1991) Detection of specific polymerase chain reaction product by utilizing the 5’–3’ exonuclease activity of Thermus aquaticus DNA polymerase. Proc. Natl. Acad. Sci. U.S.A. 88, 7276–7280. 38. Livak, K. J. (2003) SNP genotyping by the 50 -nuclease reaction. Methods Mol. Biol. 212, 129–147. 39. McGuigan, F. E. and Ralston, S. H. (2002) Single nucleotide polymorphism detection: allelic discrimination using TaqMan. Psychiatr. Genet. 12, 133–136. 40. Ranade, K., Chang, M. S., Ting, C. T., Pei, D., Hsiao, C. F., Olivier, M. et al. (2001) High-throughput genotyping with single nucleotide polymorphisms. Genome Res. 11, 1262–1268. 41. Livak, K. J. (1999) Allelic discrimination using fluorogenic probes and the 5’ nuclease assay. Genet. Anal. 14, 143–149. 42. Hui, L., DelMonte, T. and Ranade, K. (2008) Genotyping using the TaqMan assay. Curr. Protoc. Hum. Genet. Chapter 2: Unit 2.10.
Chapter 20 Qualitative and Quantitative Genotyping Using Single Base Primer Extension Coupled with Matrix-Assisted Laser Desorption/Ionization Time-of-Flight Mass Spectrometry (MassARRAY) Paul Oeth, Guy del Mistro, George Marnellos, Tao Shi, and Dirk van den Boom Abstract Matrix-assisted laser desorption/ionization (MALDI) time-of-flight (TOF) mass spectrometry (MS) has developed over the past decade into a versatile tool for the analysis of nucleic acids and especially as a reliable genotyping platform. This chapter summarizes its use in the context of the most widely used MALDI-TOF MS genomics platform, the Sequenom MassARRAY system. MassARRAY genotyping is based upon region-specific PCR followed by allele-specific single base primer extension reactions which are then desalted, dispensed onto a silica array preloaded with matrix, and the genotyping products are resolved on the basis of mass using MALDI-TOF MS. The platform is versatile in that it can resolve multiplexed reactions (40þ separate loci per reaction), acquires and interprets data quickly, gives a quantitative output which reflects the amount of product generated for each allele within an assay for multiplexed reactions, and is highly sensitive. These characteristics coupled with integrated software for sequence annotation, assay design, data interpretation, and data storage have lead to its wide adoption and use for multiple nucleic acid analysis applications in both the realm of genomics research and molecular diagnostics. Key words: Genomics, genotyping, single nucleotide polymorphism, insertion/deletion, matrixassisted laser desorption/ionization time-of-flight mass spectrometry, multiplex PCR, primer extension, MassEXTEND, iPLEX, MassARRAY.
1. Introduction 1.1. Overview
This chapter describes, in detail, the use of matrix-assisted laser desorption/ionization (MALDI) time-of-flight (TOF) mass spectrometry (MS) for genotyping genetic variations such as single nucleotide polymorphisms (SNPs), insertion/deletions, and
A.A. Komar (ed.), Single Nucleotide Polymorphisms, Methods in Molecular Biology 578, DOI 10.1007/978-1-60327-411-1_20, ª Humana Press, a part of Springer Science+Business Media, LLC 2003, 2009
307
308
Oeth et al.
mutations using the MassARRAY platform, which is currently the most widely adopted MALDI-TOF MS system for such purposes. Multiple biochemistries have been developed over the last 15 years to generate genotyping products which can be resolved using MALDI-TOF MS (1–3). However, most of them have not been integrated into an easy-to-use workflow which is associated with supporting software, automated instrumentation, and large data storage via a relational database on a commercial scale. This chapter deals with the use of single base primer extension (SBE) reactions in the context of a homogeneous reaction (meaning that reagents are only added, never removed) termed ‘‘MassEXTEND’’ and in particular the iPLEX genotyping biochemistry which have been integrated into the MassARRAY platform, making it a viable option for the average genomics/genetics laboratory, core facility, or large genomics centers. The general principal of this platform is to resolve, via MALDI-TOF MS, differences in primer masses due to changes in sequence such as the incorporation of different terminator nucleotides at the 30 -end of a primer bound adjacent to a variant site. No indirect detection techniques such as mass tags, fluorescent groups, or radioactive labels are required for accurate resolution of single base differences such as an A–G SNP (16-Da difference in mass). This is because the analytical accuracy of MALDI-TOF MS is quite high, 0.1–0.01% of the determined mass. Given that all of the primers and extension products are between 15 and 30 bp in length (approximately 4,500– 9,000 Da) this puts the smallest separation of 9 Da (A to T) within the resolving capability of the instrument. (see Figure 20.1) illustrates a standard mass spectrum generated using iPLEX and MassARRAY. 1.2. Principal of MALDI-TOF MS for Nucleic Acid Analysis
The general principal of MS is to produce, separate, and detect gasphase ions. However, most large biomolecules such as nucleic acids undergo significant decomposition and fragmentation under the conditions used by standard MS to desorb and ionize molecules. The invention of soft ionization techniques (4) alleviated this problem by the use of a buffering matrix which absorbs much of the energy applied to the sample and prevents degradation. In MALDI-TOF MS of nucleic acids, the sample is embedded in the crystalline structure of small organic compounds (a matrix such as 3-hydroxypicolinic acid is used for MassARRAY) and deposited on a conductive sample support (metal plate). The cocrystals are irradiated with a nanosecond ultraviolet laser at a wavelength of 337 nm. The energies introduced are in the range from 1 107 to 5 107 W/cm2. The laser energy causes structural decomposition of the irradiated crystal and generates a particle cloud from which ions are extracted by an electric field and results in the disintegration of the crystal
Qualitative and Quantitative Genotyping
309
Fig. 20.1. Example mass spectrum showing primers and extension products from iPLEX reactions in the mass region of 5200 and 6800 Da (17–23-mers). The figure illustrates that the G to C allele changes (mass difference of 40 Da) can easily be resolved via matrix-assisted laser desorption/ionization time of flight mass spectrometry (MALDI-TOF MS). It also shows the quantitative nature of this process. The case shown is for when both forward and reverse extend primers were used for each of three assays and loci mixtures of 3:1 were used as a template. The peak area ratios track the allele ratios as expected for each assay.
molecules. Following their acceleration through the electric field, the ions drift through a field-free path and finally reach the detector in the form of a secondary electron multiplier. Ion masses [mass-tocharge (m/z) ratios] are calculated by measuring their TOF, which is longer for larger molecules and shorter for smaller molecules (assuming their initial energies are identical). Since predominantly singly charged, nonfragmented ions are generated, parent ion masses can easily be determined from the resulting spectrum without significant data processing. The product masses are available as numerical data at this point. TOF measurements during a standard MALDI run are in the range of several microseconds (see Fig. 20.2). 1.3. Primer-ExtensionBased Qualitative and Quantitative Genotyping Using MassARRAY
(see Figure 20.3) diagrammatically depicts the MassARRAY process for biallelic genotyping of genetic variations such SNPs. Following collective selection of assay loci for a study and sequence screening/annotation, multiplexed reactions ranging from 2-plex to 40-plex are designed using MassARRAY Assay Design software program (see Section 1.4). In general, 24–36 assays are usually multiplexed per reaction. PCR primers for each assay within the multiplex are mixed together with a hot start Taq master mix containing buffers, MgCl2, and dNTPs, which in turn is mixed with the nucleic acid template, generally containing 5–10 ng of DNA. Each PCR primer contains a 10-bp, nontemplated tail which adds mass to the PCR primer so that unextended
310
Oeth et al.
Fig. 20.2. General schematic of a linear MALDI-TOF MS system such as that used with the MassARRAY system (see Section 1.2 for a detailed description).
primer has a mass larger than that of the smallest primer extension product and does not, therefore, interfere with automated genotype interpretation. Following saturation PCR cycling to 45 cycles, dNTPs are dephosphorylated using heat-labile shrimp alkaline phosphatase (SAP) to render the majority of unincorporated dNTPs during PCR unavailable for further primer extension. A SBE cocktail is then added consisting of primers which bind directly 50 to the variant loci so that allele-specific SBE will be carried out, a detergent-free buffer (iPLEX buffer plus), four standard terminator nucleotides (iPLEX termination mix), and a mutant DNA polymerase which preferentially incorporates terminator nucleotides over standard dNTPs (iPLEX enzyme). Linear-amplification SBE is conducted over 200 short cycles. Water and SpectroCLEAN resin are then added to the reaction to conduct cationic exchange of salts such as sodium with ammonium for better MALDI-TOF MS resolution. Reaction product (analyte) is then nanodispensed (about 15 nL) onto a 384-pad SpectroCHIP silica array preloaded with 3-hydroxypicolinic acid matrix using a MassARRAY Nanodispenser. The analyte and matrix cocrystallize, are loaded into the linear, positive mode, MALDI-TOF mass spectrometer (MassARRAY compact analyzer), and each pad is interrogated with 20 laser pulses over five unique raster positions per pad. The spectrum is processed and smoothed with a digitizer and then interpreted genotypes or quantitative measures such as peak areas/allele ratios are assigned on the basis of predetermined values uploaded into a relational database defining unique sample/assay/mass information (Typer software program and Typer database), and the results are
Qualitative and Quantitative Genotyping
311
Fig. 20.3. The iPLEX genotyping reaction. Each single nucleotide polymorphism (SNP) locus is amplified via PCR in a multiplexed reaction using site-specific forward and reverse primers containing a 10-bp mass tag which increases the mass of each PCR primer to prevent their interference with the post-PCR primer extension product resolution of the MALDI-TOF MS system. Following PCR, the reactions are incubated with shrimp alkaline phosphatase to dephosphorylate any unincorporated dNTPs prior to primer extension. Single base primer extension reactions are then carried out in an allele-specific manner using terminator nucleotides so that all four bases can be resolved on the basis of their mass relative to their primer and all other products generated in the reaction. Primer extension products are then desalted via cation exchange, dispensed onto a silicon array preloaded with matrix and genotype products masses (in daltons), and their allele ratios resolved with a MALDI-TOF MS system.
displayed graphically (see Fig. 20.4) The output can be either qualitative, such a genotype AG, or quantitative, such as an allele ratio of 2:1 of A to G based on peak areas. A detailed discussion on all aspects of quantitative applications for nucleic acid analysis using MALDI-TOF MS is beyond the scope of this chapter; however, Section 1.5 describes the core differences between qualitative and quantitative data analysis with MassARRAY. However, applications such as quantitative gene expression analysis (5–7), copy number variant analysis, including trisomy 21 detection (8– 10), oncogene mutation screening (11), loss of heterozygosity (12), splice variant analysis (13), and quantitative methylation
312
Oeth et al.
Fig. 20.4. Illustration of Typer 4.0, which has automated cluster-based genotype assignment functionality, shown at the upper left. Other graphical user interface features include color-coded plate representation, interactive spectra for each well, and a histogram to access assay performance.
determination (14) have all been enabled on MassARRAY; for several good reviews of the platform see (15–17) as well as (18–21) for several of these applications. 1.4. Assay Design 1.4.1. Importance and Methodology for Screening Genomic Sequences Prior to Assay Design and Wet Laboratory Work 1.4.1.1. SNP Selection
High-throughput SNP analysis technologies are increasingly being viewedaspowerfultoolsforgeneticsresearchingeneralandforthestudy ofdiseaseinparticular.Sequenom’splatformhasbeenusedextensivelyto genotypeSNPs,primarilyinhumans,butalsoinotherorganisms. Unless a study starts with previously chosen and validated sets of SNPs, researchers need to select appropriate SNPs from available sources. Publicly available SNPs, notably in NCBI’s dbSNP database (see (22) and http://www.ncbi.nlm.nih.gov/projects/ SNP, as well as Chapter 3 in this volume) are a good source of SNPs in many organisms and are very useful for research, especially if they are judiciously selected on the basis of available annotation. Appropriate frequency: Existing public SNPs have been mainly discovered by in silico data-mining algorithms (see Chapter 4 in this volume) and most of them have not been confirmed
Qualitative and Quantitative Genotyping
313
biologically. It is estimated that only a fraction of publicly available SNPs are polymorphic with appreciable minor allele frequency in a given population and that even fewer are common in that population. For human SNPs there is frequency information for a large number of public SNPs in several populations, currently more than four million in some populations, that have been validated by the HapMap Consortium (see (23) and http://hapmap.org). Such frequency information can help SNP selection for the purposes of any given study. If frequency information in populations of interest is not available, SNPs may be validated with other methods prior to a study. At Sequenom, for instance, we first assessed SNP status by allelotyping (i.e., by testing assays on pooled DNA) on 92 CEPH samples of Caucasian individuals, and then selected appropriate sets of polymorphic SNPs for genome-wide association studies (24, 25). Predictive features: When there is no SNP validation information, public SNP annotation may still contain predictive features to help guide SNP selection. We have found that many of the available dbSNP annotations were predictive of polymorphic status of human SNPs and could be used to improve the selection of SNPs for genetic research (25). One of the strongest predictors of polymorphic status was the number of independent groups reporting the SNP to NCBI. Polymorphic SNPs were also more likely to have longer sequences for their submission, be drawn from more recent submissions, be derived from genomic DNA, be mapped within introns, and map exactly one time to the genome. Other selection criteria: Other annotations that may be important in SNP selection include proximity to genes, location in coding versus noncoding regions, and unique mapping to the genome. Also, the number of SNPs required for a given study may also depend on linkage disequilibrium across the genome in the populations of interest. The HapMap project is characterizing patterns of linkage disequilibrium in the human genome, so its data could help identify more informative sets of SNP markers (23). 1.4.1.2. Assay Screening
After the SNP sequences (or other genomic sequences) have been selected for a study, the sequences and the assays for them need to be screened before use in the laboratory, to achieve optimal results. The screening method we use at Sequenom centers around the eXTEND algorithm, and is our best effort to account for known aspects of assay design quality control, from validating raw SNP sequences to post-design sign-off validation of multiplexes. The main purpose of eXTEND is to check for known effects that might result in assays failing or producing misleading results (false positives) when actually executed in an experiment. eXTEND identifies potential amplicons for given PCR primer pairs, allowing for a
314
Oeth et al.
small number of mismatches and favoring 30 hybridization in a manner similar to that of the ePCR program (26). ePCR identifies potential amplicons for given PCR primer pairs; eXTEND, on the other hand, is tailored for MassEXTEND assay design and so identifies potential MassEXTEND sites on the genome considering both the PCR primer pair and the corresponding probe MassEXTEND primer of each assay. The eXTEND algorithm runs in parallel on multiple processors, each analyzing a 50Mb stretch of the genome, to analyze many alternative primer triplets and select the best PCR primers for assay design. Each primer is annealed in silico to the genome, allowing up to a very small number of mismatches over its length. PCR primers are mapped in all intended and unintended combinations of order, orientation, and spacing that could produce an amplicon of up to 1,000 bp and possibly contain a MassEXTEND primer target within the amplicon; see Fig. 20.5 for assay mapping configurations that count as designed mappings (true hits) or unintended targets (false hits and null hits). Generally users will want to work only with assays that have a single true hit.
Fig. 20.5. Assay mapping configurations that count as true mappings (true hits) or false positives (false hits and null hits).
The eXTEND SNP sequence screening and assay design process consists of a suite of four applications that may be used independently, by uploading the right input files (SNP sequences or assay designs), or in series, passing the results of one application on to the next as shown in Fig. 20.6; users can start at any module and continue in the direction of the flow. The applications of the eXTEND suite, in the order they are intended to be used, are:
Qualitative and Quantitative Genotyping
315
Fig. 20.6. Assay screening/assay design process flow.
ProxSNP: Used to map SNP sequences to a genome and reformat the sequences to demark the presence of known SNPs that are within a given proximity of the specified (assay) SNP. Ensuring that proximal SNPs are denoted in SNP sequences prevents primers from being designed over regions where these polymorphisms may occur. PreXTEND: Used to (pre-)design amplification (PCR) primers with respect to a genome and reformat the sequences so that subsequent assay design will employ these primers. Designing with respect to a genome ensures that only the intended region of the genome (around the assay SNP) is amplified by a given pair of PCR primers.
316
Oeth et al.
PreXTEND accomplishes: – Validation of SNP sequences for primer design – Genome-wide PCR primer pair validation, accounting for macro and micro subsequence repeats and the presence of suitable probe primer target(s) – Demarcation of the best PCR primers in the SNP sequence for assay design Assay Design: Used to design (multiplexed) assays for a given set of SNP sequences using the latest version of Sequenom MassARRAY Assay Design. The default settings for assay design are set to expect sequences formatted by PreXTEND. See more details about Assay Design in the following section. PleXTEND: Used to validate multiplexed assay designs with respect to a genome. Assuming assays were designed through the preceding applications in the eXTEND suite, the purpose of this application is to check for unintended amplification and MassEXTEND products resulting from random couplings of all oligos present in a multiplex of assays. PleXTEND achieves: – Detection of potential false homogeneous MassEXTEND results due to unintended PCR products resulting from random association of primers in multiplexed assays (rare) – Detection of potentially nonfunctional or multiply targeted homogeneous MassEXTEND assays: for old assay designs or revalidation against new genome sequence releases – Retrieval of amplified sequences with respect to known genome sequence 1.4.2. Overview of the Automated Assay Design Software for MassARRAY
The creation of assay designs is performed using SEQUENOM’s proprietary MassARRAY Assay Design software program. The current version of this software program (3.1) is capable of designing a high throughput of SNP sequences to high multiplex levels with numerous design options and chemistry variations, although is more typically used to generate tens to hundreds of iPLEX assays at the 10-plex to 30-plex multiplexing levels. To use the assay design software, SNP sequences must be prepared in the SEQUENOM SNP Group file format. This format is a tab-delimited text (table) file containing just named columns (fields) that identify the name (SNP_ID) and sequence (SEQUENCE) of each SNP sequence to be analyzed. The sequences themselves identify the assay SNP using brace and forward slash characters to denote the sequence alternatives, e.g., a C/T SNP (IUPAC code ‘‘Y’’) is represented as ‘‘[C/T]’’ (or ‘‘[T/C]’’). This format also allows for more general mutations other than SNPs such as insertion/deletion and multiple nucleotide polymorphisms, e.g., ‘‘[CAT/-]’’ and ‘‘[A/TG/G].’’ The
Qualitative and Quantitative Genotyping
317
assay SNP must be flanked by at least 100 bases of sequence to allow for primer design. This sequence should contain only uppercased nucleotide codes (e.g., no space characters), but may also denote proximal SNPs using IUPAC codes, e.g., ‘‘Y’’ or ‘‘N.’’ Mixed uppercased and lowercased characters may be employed to annotate the sequence for amplicon design, as described below. To create an assay design, a SNP Group file is specified (opened) using the software. Design settings are then specified, e.g., for chemistry and multiplexing level, either manually or using one of the preset configurations. A design run will take anywhere from a few seconds to a few hours to complete, depending on the number of input SNP sequences, design options, and multiplexing level specified. A design run will produce several output files. The Design Summary file will contain statistical details of the run, such as the number of wells created, and all of the design settings used. Together with the original input SNP group, this file may be reloaded into the design software to examine, modify, or reproduce an assay design run. This file also contains a textual description of each assay and well designed, along with a depiction of the expected mass spectrum peaks and a list of any warnings, e.g., where the primers from two assays in the same well might have a potential for extending off one another. An Assay Group file will also be produced to describe each well designed. This file has a more condensed tab-delimited column/row format and contains the essential definition of the assays. It is the primary file necessary to upload assay designs into the Typer database and for ordering oligos. A third output file that may be created is the Failed Strands file. This file contains all the input SNP sequences that failed assay design, along with a description of the most significant reason why they were rejected. The failed strands file format is an extension of the SNP group format and these files may be used as an input file for further assay design, i.e., with adjusted design settings. The process of initially designing MassEXTEND assays from input SNP sequences is referred to as ‘‘de novo assay design.’’ The key considerations for de novo assay design are described in Sections 1.4.2.1, 1.4.2.2, and 1.4.2.3. It is often necessary to use the design software multiple times with modified settings to optimize the number of sequences and/or wells designed. Additionally it may be necessary to redesign assays add to existing designs, or even change the type of assay, after wetlab testing. The most useful secondary design options and variations are described Sections 1.4.2.3 and 1.4.2.4. 1.4.2.1. Primer Design
A single MassEXTEND assay analyzes a single-assay SNP (or sequence variation) and is defined by three oligo primer sequences. More generally, it also requires a definition of the MassEXTEND stop mix ddNTP and dNTP composition, but for iPLEX chemistry this is a constant. Two oligos are PCR primers that are
318
Oeth et al.
designed to amplify a small region (about 100 bases) of DNA containing the targeted SNP site. The third oligo is the MassEXTEND probe primer that is designed to extend into the SNP site in a subsequent and distinct polymerase reaction to produce an allele-specific analyte sequence, which may be later identified in a mass spectrum. Although PCR and probe primers are used for quite different purposes, their design is mostly restricted by the same criteria. Both PCR and probe primers must hybridize their 30 sequence to the targeted DNA for subsequent polymerization, although both may have 50 tag sequence (or chemical group) that does not hybridize. The templated part of the primer is variable with respect to the position it targets along the DNA sequence and its length. The predicted temperature of primer hybridization (or dissociation), Tm, is used as an indication of the potential of an oligo to polymerase at the targeted site and this value must be greater than a minimum value or within a range specified for design. The primer should also specifically bind to its target site and therefore must not be designed over a region that contains another sequence variation, e.g., a proximal SNP, nor have significant potential for false priming, i.e., extending at an equivalent or marginally less favorable site. Primer sequences must also be screened for potential extension reactions amongst themselves as a result of hairpin substructures, homodimers, and heterodimers. The inclusion of Gn subsequences (where n 4) is also considered, since this can be problematic for oligo synthesis. Some criteria will debar primer designs immediately, such as not meeting the required predicted Tm within the specified length bounds or nonspecificity of 30 hybridization of a sequence with repeated subunits, e.g., CGACATCATCATCATCAT. Otherwise each particular feature of primer design is given a normalized score and then rejected if that score falls below a fixed threshold (0.5). Design criteria for PCR primers verses probe primers differ in the limits and optimum values for these features, and in their relative weightings. In particular, false priming (extension) is not considered as important for PCR primers as probe primers. In the former case amplification efficiency may be affected, whereas in the latter case a false extension would lead to a false-positive result for genotyping. Where a particular primer design is not rejected for any single feature, a combined score is generated as a product of individual feature scores and the design is rejected if this combined score is less than another cutoff value. (Optimal values, boundaries, score weightings, and combined score cutoffs relating to design features may be changed from their default values via the expert settings dialogs of the software.) For sequences where primers may be designed within all the scoring criteria, the combined primer design score is used to select between alternative assay designs.
Qualitative and Quantitative Genotyping 1.4.2.2. Amplicon Design
319
The term ‘‘amplicon’’ is used in relation to MassEXTEND chemistry to refer to the short double-stranded sequence of DNA resulting from PCR amplification using a particular pair of PCR primers and sample DNA. Typically there are many choices for primer pairs about a given SNP site. The assay design software will select from a list of potential primers from each side of the SNP site such that individual primer scores and the resulting amplicon length are optimized. An overall amplicon score is generated for each potential design. It is used to select the best design and in evaluation of multiplexed assay designs. This amplicon score is also weighted by an extra term that accounts for the PCR primers forming a heterodimer such that the 30 -end of one primer might extend off the other. The selection of potential PCR primers for amplicon design may be restricted owing to annotation of the individual SNP sequence and how this annotation is interpreted. In constricted design mode, a PCR primer may not be chosen to overlap any lowercased base in the sequence and must use any whole sequence that is between the specified limits for PCR primer length (typically 18–24 bases). This option is used where the input SNP sequences have been annotated to indicate exact PCR primers chosen by the user or another application, such as Sequenom’s eXTEND suite described earlier. In scan and restrict design mode, a PCR primer may be chosen such that only a maximum of 50% of its length may overlap a lowercased region of the sequence and its score is penalized by the amount of overlap such that primers are most likely chosen from all uppercased regions. This option is useful when external applications produce annotated sequences after searching for common sequence motifs and larger repeat regions. Additionally, in this mode each SNP sequence is internally scanned and annotated for upstream (50 to 30 ) 10-mer sequence repeats, checking in both senses of the DNA strand. Scanning for such repeats ensures that the primers target a specific site and produce the smallest amplicon containing the SNP site. This is particularly useful where primer selection cannot be prevalidated against a genome and such repeats are often missed by applications looking for specific or longer repeated regions. Other features of standard amplicon design for MassEXTEND assays include the restriction that the PCR primers may not be targeted to within six bases of the SNP site and the addition of a specific universal 10-mer tag to the 50 -ends of each templated PCR primer design. The first of these features is to avoid creating a false target for probe primer extension should the PCR primers amplify an unintended region of the DNA sample. The addition of the 10-mer tags increases the typical length of the primers to 30 bases, which is useful should any undigested PCR primers be present after the probe extension reaction as their singly and doubly charged masses would fall outside the typical analyte
320
Oeth et al.
mass window (4,500–8,500 Da). More importantly, the addition of universal 50 tags assists in the efficiency and balance of PCR amplification for multiplexed assays. These tags become part of the amplicon design and are considered in the calculation of its length when a PCR primer pair is chosen. In the case of insertion/deletion polymorphisms, the length of the amplicon target is calculated assuming insertion of the longest sequence alternative. The best-scoring amplicon design is used to define the PCR primers regardless of the multiplexing level. This greatly reduces the amount of false priming analysis necessary while designing multiplexed primers relative to allowing amplicon design to be flexible. Amplicon design typically has no effect on the available choices for probe primer and it is rare that PCR primers for one assay have significant false priming potential against the amplicon of another assay. Where this does occur it is typically because SNPs are in very close proximity on the genome and amplicon designs overlap. If it is necessary to design such SNP assays for the same well, then use of a multiSNP sequence may be required, as discussed in Section 1.4.2.4. 1.4.2.3. Extend Primer and Multiplexed Assay Design
The choice of probe primers (aka extend primers) is limited to only two sites directly upstream (50 ) of the SNP site on either the forward or the reverse sense strand of the double-stranded amplicon. For SBE assays, such as iPLEX, the probe reads the SNP by polymerization of a single ddNTP base that complements the first 30 allele base encountered. For some less typical SNP definitions, e.g., for [A/AC/G] and A[A/-]A, this might invalidate MassEXTEND design in one or both sense directions because there is not a distinct extension product for all alleles. Where possible, the software may correct for this by effectively adjusting the position of the given SNP, e.g., [A/AC] is equivalent to A[-/ C] in the forward direction. For multiple base extension assays, as formally employed for homogeneous MassEXTEND, the choice of probe primers is more complicated and allows for more alternative designs requiring different stop mixes. Probe primer design may also be restricted in one or both directions owing to the general primer design features described earlier, e.g., the predicted Tm requirement may not be satisfied within the length bounds for A.T-rich regions or the primer may have a strong potential for autoextension via a hairpin structure. Typically, where a probe primer can be designed for a certain length in one direction, several alternative designs can be created by adding templated bases to the 50 -end. Since probe primers target short amplicons rather than DNA, their length can be as low as 15 bases. The upper limit to extended primer length is typically restricted to 30 since primers of this length are at the high mass limit where multiplexed analyte signals are clearly detected and resolved in the mass spectrum. Longer sequences
Qualitative and Quantitative Genotyping
321
might also produce doubly charged mass signals that overlap with low mass analyte signals. Probe primers have an additional scoring component that penalizes their design if the mass difference between their analyte masses is close to that of a sodium adduct at þ22 Da. Hence an A/C SNP is disfavored for design in the forward direction relative to a reverse probe that produces T/G extensions. Evaluating as many alternative probe primer designs as possible is necessary to assist with multiplexed assay design. If only a single assay is designed for a well (a uniplex), then selecting probes of shorter length is preferred. For designing multiplexed assays, there are two primary considerations. The probe primer sequence for one assay must not have a significant potential for extending against any amplicons or probe primers of other assays in the same multiplex, since this is likely to result in false-positive analyte signals. Each of the analyte mass signals, one for each allele extension, must be clearly resolved from all other expected peaks in the mass spectrum. These expected peaks include those of other possible analyte signals and signals from unextended primer peaks, in addition to potential by-product peaks for each species. By-product peak mass offsets and fixed mass contaminants may be specified explicitly for a design run, but the most commonly observed salt adduct peaks, Naþ at þ22 Da and Kþ at þ38 Da, are typically avoided by leaving the minimum analyte peak separation design setting at its default value of 30 Da. Additionally, a minimum separation value may be specified to prevent unextended probe primer peaks being designed with similar masses to ensure that the ratio of unextended to extended primers can later be accurately estimated from their peak areas. If the number of designable SNP sequences N is greater than the targeted multiplex level M, as is usually the case, many alternative well designs may be available. This leads to two secondary considerations for multiplexed assay design. A multiplex of assays should have a high score relative to alternative choices of combinations of SNP assays and the individual probe primers chosen for those assays. For a given set of SNP sequences, it is also desirable to produce the optimum multiplex design efficiency, i.e., the greatest number of largest multiplexes or the minimum number of total wells. The maximum number of possible SNP assay combinations to evaluate to find the ‘‘best’’ M-plex is calculated as N!/M!(N-M)!, which becomes a significantly large number as N increases at the typical levels of multiplexing employed. For a SBE design, a typical SNP sequence has two design directions and, considering Tm restrictions, an average of ten alterative primer lengths. Therefore, for each subset of SNP sequences there could be M20 possible combinations of probe primer combinations to evaluate for analyte peak mass overlaps and false priming potentials. The assay design software optionally permits probe primers to be designed with an
322
Oeth et al.
additional nontemplated 50 nucleotide sequence and/or the addition of optional 50 mass tags. This allows for much greater flexibility for designing the masses of longer probe primers but also significantly increases the number of alternative sequences to evaluate. Hence, it is generally not possible to evaluate all potential multiplexed designs in a reasonable computation time, although the best scoring solution may be evaluated for small values of M and N. The algorithmic method the assay design software employs to evaluate and select multiplexed designs is termed subplexing. The first step of this method is to evaluate all valid subplexes of size m (m M) for a subset of n SNP amplicon designs (n N). To facilitate this process and to enhance performance, a logical biplexing matrix is created for the n SNP amplicons. Each bit (logical 1 or 0) in the matrix indicates if a particular probe primer sequence for one SNP assay design can form a 2-plex, or biplex, with a particular probe primer sequence for a second SNP assay, as uniquely identified by the row/column address of the bit. In building up the matrix, one makes shortcuts by testing design features in order. For example, if the PCR primers for two SNP amplicon designs have a significant potential for false priming, then all biplexing bits for the probes are set to false (0) without the need to evaluate any probe primer designs. Noting that no multiplex can contain any pair of assays that do not separately biplex, one can rapidly scan this matrix using bitwise AND operations to detect all available multiplexes up to an n-plex. The algorithm actually scans for m-plexes from a subset of n SNP amplicons (m n 2m) as each new SNP amplicon is added to the logical biplexing matrix. If there are already n SNP amplicons represented in the subset, a new SNP amplicon may replace any other in the subset that is not a member of an m-plex and the search proceeds until either each of the n SNP amplicons is a member of at least one m-plex or there are no more SNP amplicons to consider. Only the best scoring set of probe primer designs is recorded for each set of valid m-plexes found. This score is given by the product of the individual probe primers scores and factors that depend on the average mass of the probes and the number of nontemplated 50 bases added to the probes. This is to favor designs that have higher individual probe primer scores, at the lower end of the mass spectrum and with a minimum number of nontemplated bases. Once the multiplexing buffer is at capacity, all potential m-plex designs recorded are given a multiplex confidence score, MPCONF. At this point all multiplexed probe designs are assumed to be optimal for each m-plex of SNP amplicons and attention switches back to multiplexed PCR design. The weighted factors that make up the multiplex confidence score are the geometric average of all the individual amplicon design scores; the mean square variation of all predicted PCR primer hybridization Tm values; the root-mean-square variation of the amplicon lengths;
Qualitative and Quantitative Genotyping
323
and the largest false priming potential for cross-hybridization of a PCR primer to an amplicon in the multiplex. Hence, a high MPCONF score should indicate that a multiplexed assay should not have cross-hybridization issues and should be balanced with respect to amplification rates of individual assays. At this point some multiplexed designs may be rejected if the MPCONF score is less than a cutoff value, which may be changed from its default value, along with other score component weightings, via an expert settings dialog of the assay design software. If the subplexing level m equals the targeted multiplex level M then the best m-plex design may become a completed well design. However, the best scoring design may not be the best choice considering the overall design efficiency for the whole set of N input SNP sequences. Instead, the assay design software considers all m-plex designs that have scores within 40% of the best scoring design and selects from these the m-plex that has a set of probe designs involved in the least number of alternative m-plex designs. This strategy is understood in the most general sense as selecting well designs for the most difficult SNP assays to multiplex first, thereby leaving the greatest potential to multiplex the remaining SNP assays. If the subplexing level m is less than the targeted multiplex level M, then the best m-subplex is converted into a subplex constraint for creating a new subplex of level m (or M-m if 2m < M). A subplex constraint is essentially a fixed set of probe primer designs that is initially employed to screen out the set of available probe designs for new SNP amplicons before they are evaluated to create a separate m-plex. Since dividing up an M-plex search into separate m-subplex searches amounts to a rational tree pruning strategy, it dramatically decreases the amount of search space at the cost of potentially missing some of the best M-plex solutions. It is analogous to deciding on a particular sequence of moves in chess before looking further ahead. Using a subplex constraint also has other performance advantages. The collection of expected primer masses may be mapped so that other primer masses are tested against them with a single array access rather than sorting through a list of peaks. Searching for probe primer designs with additional nontemplated 50 modifications becomes practical. For each probe primer design tested for a SNP amplicon that does not multiplex with the subplex constraint owing to a mass conflict, a quick search may be performed for a 50 modification that increases the masses of the primer and analytes to avoid any conflict. In the case of nontemplated nucleotides, the shortest available sequence is used that does not result in any new false extension potential. The principle of subplexing is easily extended to create multiplexes of any size. The assay design software initially attempts to design assays for a subset of ten SNP amplicons such that each SNP amplicon forms at least one 5-plex with four others in the subset.
324
Oeth et al.
Because analyzing the full set of unrestrained probe primer lengths takes a significant amount of time and is unnecessary, creation of initial 5-plexes is limited to scanning only the first four available lengths for each extend direction. The first 5-plex chosen becomes the constraint in choosing the next 5-plex. This is then combined to create a subplex constraint of ten assays that is used to constrain the creation of the next 5-plex and so on. At some point it will not be possible to find the next 5-plex. In the trivial case where this is because there are too few SNP sequences left to multiplex, the multiplex level is reduced to proceed. However, at higher levels, multiplexing is more difficult and, since only a fraction of the total multiplexing combinations can be evaluated, it is necessary to repeat the search sequence from a different starting point to find more solutions. The foremost issue with searching for suitable multiplexed assay designs at high multiplexing levels is predicting when there are no more solutions to be found. The assay design software has several techniques for performing multiple search-pass iterations, including adjusting the subplexing level, and design settings that may be adjusted to trade search depth for run time. A discussion of this subject requires describing programmatic details that are beyond the scope of this document. Because it is not possible to look far ahead for multiplex design efficiency, it is quite common for a design run to result in a few low-plex wells for the remaining SNP sequences at the end of a design run. These are often a result of false extension potentials for probe primers between a few assays. The 3.1 version of the assay design software has an exchange replexing option that allows an extra search procedure to be performed at the end of a design run. This procedure attempts to remove these small wells by exchanging assays between well designs and adding them to other wells where space exists. 1.4.2.4. Further Assay Design Options
As mentioned already, it may be necessary to run the assay design software multiple times to create a final set of assay designs, either to optimize the initial designs or after assays have been experimentally tested against real DNA samples. The assay design software has a number of Replex design modes to assist with modifying existing assay designs. For replex design runs, both an existing assay group and a SNP group, containing the original SNP sequences, are required. The most versatile of these options is the Superplex with new SNPs design mode. This process, also referred to as ‘‘superplexing,’’ involves designing assays for SNP sequences that will be added to existing multiplexed assay designs or used to create new wells where this not possible. The most direct use of superplexing is with the output of a de novo design with modifications to particular design settings to allow previously
Qualitative and Quantitative Genotyping
325
rejected SNP sequences to be designed. The advantage of using this method to increase design throughput over performing a new de novo design run is that the original assay designs remain designed at the more stringent design settings. A superplex design run is also typically much faster, allowing the effect of various design setting modifications to be tested in a short time over multiple runs. Another important use of superplexing is to add completely new assays to existing designs. In this case new SNP sequences are added to the original SNP group file. In combination with modifications to the assay group file, this allows for high-level design strategies. For example, a set of validated control assay designs (at low multiplexing levels) may be multiplexed with sets of test SNP sequences. Another useful replex design option is Re-multiplex assays, which is a procedure also referred to as ‘‘replexing.’’ When this option is employed, the assay design software attempts to redesign the multiplexing combinations of assays to create new wells without modifying any details of the individual assay designs. This option may be used to lower, or slightly raise, the average multiplexing levels, but is primarily intended to regroup assays to a smaller number of wells after particular assay designs have been discarded from the set, e.g., after the assays are found to be ineffective in practice against real DNA samples. In this case the original assay group may be modified to remove the unwanted assays or created as a result of exporting a new assay group from the Typer database. Other options for more directed assay design may be facilitated using options available to the SNP sequence or SNP group file format. In the situation where there are multiple SNP sites in close proximity, a useful option is to specify these using a multiSNP sequence. This is simply a SNP sequence that has more than one SNP site identified, e.g., . . .[A/G]. . .[C/A]. . .. For such sequences the assay design software will attempt to design probe primers to read each SNP but targeting a single common amplicon. This will allow such SNPs to be analyzed in the same well design, whereas the SNPs defined by separate SNP sequences, e.g., . . .[A/G]. . . M. . . and . . .R . . .[C/A]. . ., may not be multiplexed in the same well should the designed amplicon sequences overlap. In this case it would appear that primers from one assay false-prime against the amplicon of the other. When using multiSNP sequences, one usually needs to modify the settings that constrain amplicon length to allow PCR primers to be designed that flank all of the SNP sites. Other SNP group file options pertaining to SNP sequence sets and the Quantitative allele discrimination of single nucleotide polymorphisms (qSNP) design options are not described here as they are intended to assist designs for Quantitative gene expression (QGE) and copy number variation (CNV) studies rather than genotyping.
326
Oeth et al.
1.5. MassARRAY Data Analysis 1.5.1. Data Acquisition, Signal Processing, and Real-Time Calling
In MALDI-TOF MS, data acquisition happens at the end of the flying tube, where a detector in the form of a secondary electron multiplier detects the arrivals of ions. The number of ions hitting the detector in a given time interval (usually in nanosecond range) is recorded as ‘‘intensity.’’ Therefore, the raw data produced by MALDI TOF MS from each laser shot can be thought of as a numeric matrix consisting of two columns: intensity and TOF. The control of the mass spectrometer and the acquisition of the spectra data are handled by the SEQUENOM proprietary software program SpectroACQUIRE. The signal processing of mass spectra for the DNA sample is similar to that being used in the proteomics field: they both go through steps such as denoising, calibration, baseline estimation, and peak quantification. However, the key difference between the two is that for genotyping we already know the masses of the peaks of interest, including those of all the primers, SBE products, and even their adducts, so no peak identification is required. Usually, several spectra corresponding to the different locations of the crystals are acquired from a single sample. The raw spectra are first calibrated (i.e. transforming the TOF to the m/z coordinate) using the external calibrants with known masses, which are located in a separated location on the SpectroCHIP. The relationship between TOF and m/z is usually quadratic. The coefficients of this quadratic function can be easily estimated using least-squares fitting from the resulting spectra of the calibration well. The calibrated spectra are then averaged. This averaged spectrum then goes through a wavelet-based digital filter to remove the high-frequency noises. The baseline intensities of this processed spectrum are estimated using the data where the mass ranges of prespecified window size surrounding the expected peaks are masked. The baseline noises are calculated as the root mean square of the difference between the local intensity and the baseline value. For each peak of interest, we determine the best-fitting Gaussian curve using the data points surrounding each peak. The resulting parameters from the curve-fitting procedure are peak height, width, area, mass offset, and signal-tonoise ratio. For genotype calling, we first convert each of the aforementioned parameters derived from curve fitting to a probability measurement based on a corresponding preparameterized probability function. The product of these probabilities is then used as a quality measure for each peak of interest. The final genotype assignment primarily depends on the number of allele peaks identified by their probabilities, the relative skew of allele peaks, and a set of predefined parameter cutoffs. For example, for a biallelic SNP assay, if both allele peaks have P < ‘‘no-peak cutoff’’, a ‘‘No Alleles’’ call is made; if only one
Qualitative and Quantitative Genotyping
327
allele peak has P > ‘‘aggressive cutoff’’, a homozygous call may be made; if both allele peaks are present and the skewness is low, a heterozygote is made; etc. In allelotyping, instead of the assignment of genotypes, SNP allele frequencies are calculated from the relative ratios of the areas of the two allele peaks. The peak areas are estimated by integrating the fitted Gaussian curve. The variation in the peak area is estimated using the area for a Gaussian with amplitude equal to the local noise in the same region of the spectrum. In addition, rather than using the average spectrum created from shooting a number of spectra (rasters) for a pad as done for genotyping, one records areas for each raster separately for allelotyping. So, the frequency uncertainty for an allele can be calculated on the basis of the variation in the individual area measurements between rasters and on the basis of the area variations. The signal processing and real-time calling are done on the fly immediately after acquisition of the spectra of each sample. This task is handled by MassARRAY Caller in the background while SpectroACQUIRE is running in the foreground. 1.5.2. Post-Real-Time Analysis
The MassARRAY system combined with the iPLEX Gold chemistry allows us to design assays with high multiplex level (about 36-plex) routinely. With 384 pads per SpectroCHIP, one single run could yield over 13,000 genotypes. This large amount of genotyping data allows us to examine the behavior of a single genotype assay over a large number of samples and helps us gain more insight into assay performance, improve real-time calls by applying a clustering algorithm, and identify potential CNVs.
1.5.2.1. Genotype Quality Control
Genotype quality control is a crucial step before carrying out any downstream analyses, e.g., GWAS study. Besides the genotype call scores and flags provided by MassARRAY Caller, one can assess the genotyping quality by examining the behavior of a SNP assay across many genotyping samples. The quality of genotype calls can be assessed by the following criteria: l Genotype consistency between the replicated samples. l
Genotyping completeness (call rate): low call rate indicates poor assay performance.
l
Hardy–Weinberg equilibrium check: deviation from Hardy– Weinberg equilibrium may indicate problematic genotyping.
l
Concordance of genotype or allele frequency with those in the public database (e.g., HapMap database).
l
Check Mendelian inheritance given the family structures of the samples are known: Mendelian inheritance error may indicate genotyping errors.
l
Clustering performance (see the next section for details): poor clustering performance may indicate poor assay performance.
328
Oeth et al.
1.5.2.2. Clustering Analysis of Genotyping Samples
Quantitative genotyping data like those produced by the MassARRAY system can be presented in a scatter plot where each point represents one sample and the two axes are the heights of the two allele peaks (hH and hL) from the mass spectra (Fig. 20.7a and b). If an assay performs perfectly, the homozygote samples should lie along the two axes, while the heterozygote samples should lie along the 45 line. However, in reality, assays are often skewed owing to preferential incorporation of one of the two alleles in the PCR step. Sometimes, they are so skewed to the point that the real-time genotyping calling algorithm is unable to decide whether they are homozygote or heterozygote (e.g., those ‘‘no calls’’ samples on the edges of the two homozygote clusters, as shown in ( see Fig. 20.7a). However, on the basis of the clustering pattern, we can tell that they should be assigned to the corresponding homozygous genotypes. One of the major goals of applying a clustering algorithm to genotyping data is to rescue these ‘‘no call’’ samples and, therefore, improve the call rate (see Fig. 20.7b). The k-means algorithm is one of the earlier methods used for genotyping data; however, it is not effective for handling different within-variations and for finding outliers. One also needs to supply the number of clusters upfront. This clustering algorithm is based on the work reported in (27). The nature of this algorithm makes handling within-variations and finding outliers easy, and using a penalized likelihood approach in this model also eliminates the need to supply the number of clusters a priori. The model works well even when one or two clusters are missing.
Fig. 20.7. The genotyping data of the 92 HapMap samples. (a) The genotyping data of the 92 HapMap samples (in quadruplicates) on SNP rs2953438 are shown in a scatter plot based on their allele peak heights. The sample points are shaped according to the real-time genotype calls. (b) The same scatter plot as in (a), but the sample points are shaped according to the clustering calls.
Qualitative and Quantitative Genotyping
329
In (see Fig. 20.8) we illustrate how the clustering algorithm works. For a given genotyping assay, the angle values [defined by tan-1(hH/hL)] of all the genotyped samples are modeled as random samples drawn from a mixture of three Gaussian distributions, each of which corresponds to one of the possible genotype clusters. The task of the algorithm is to determine the mean and the standard deviation of component and its weight (the contribution to the mixture model). One classical approach to this kind of problem is to use the expectation maximization (EM) algorithm, which iteratively maximizes the likelihood of the model for a given data set. Once the parameters have been determined, the posterior probabilities for each sample given that sample belongs to one of the three genotype clusters can be calculated on the basis of the Bayes rule. The final assignment of clustering membership is governed by the following three sets of user-defined parameters: 1. Specificity: Each sample is assigned to the cluster that gives the highest posterior probability and this probability must be higher than the specificity cutoff. 2. Sensitivity: This is used to define a confidence interval around each cluster mean (shown as the regions defined by the solid vertical lines in Fig. 20.8) and only the samples that lie within the intervals are considered for clustering into a given group.
Fig. 20.8. The genotyping data of the HapMap samples. The data from Fig. 20.7 are replotted on the basis of their magnitudes (defined as the Euclidean norm of the two allele peak heights) and angles and the sample points are shaped according to the clustering calls. The fitted Gaussian density is plotted for each genotype cluster in light color, while the overall Gaussian mixture density is shown in darker color. The confidence intervals determined by the sensitivity cutoffs are defined by the solid vertical lines. The static angular limits are shown as the dashed vertical lines. Finally, the red horizontal dashed line indicates the signal to noise ratio based signal cutoff. All samples with a signal-to-noise ratio below this cutoff are assigned to ‘‘no calls’’.
330
Oeth et al.
3. Static angular limits: They are used to define the core regions of each genotype cluster (shown as the regions defined by the dashed vertical lines in Fig. 20.8). All the samples that lie within the core regions are automatically assigned to the corresponding genotypes. As a by-product of the clustering results, one can calculate each sample’s clustering strength, which is a measurement of how close the sample is to other samples in the same cluster compared with the samples in other clusters. The clustering performance of each genotype cluster is defined as the average of the clustering strengths of the samples in that cluster. Both values range from –1 to 1, with 1 being the best. One can use clustering strength to identify outliers and clustering performance to flag poor performing assays.
1.5.2.3. Using Genotyping Data for CNV Analysis (a SNP Allele Ratio Approach)
CNVs are segments of DNA ranging in size from thousands to millions of DNA bases in which the copy number differs among individuals. They have attracted tremendous attention in the field of genomics owing to their possible implications in disease association. The SNP allele ratio approach to CNV identification using quantitative genotyping data is based on the fact that depending on its actual allelic status and copy number change, the allele ratio (i.e., hH/hL) of a heterozygous SNP within a duplication or multiallelic CNV region will be deviated from the core heterozygote cluster. In a scatter plot based on the heights of the two allele peaks, the CNV samples are shown as distinct clusters lying between the heterozygote and homozygote clusters, e.g., the circled samples in (see Fig. 20.9). Where exactly they are located will depend on the actual allelic status and copy number change. For example, a sample with genotype ‘‘CC/T’’ should lie along the 26.6 line (tan26.6 ¼ 0.5) or the 63.4 line (tan63.4 ¼ 2), depending on which allele is plotted on the x-axis. In reality, however, the clusters are often skewed and spread out, which makes the identification of the CNV samples difficult. We have implemented a function (findCNV) in the R.SQNM package to do SNP allele ratio analysis. This function is an extension of the Gaussian mixture model based clustering algorithm we described in the previous section. The outliers identified by the clustering algorithm based on the preset sensitivity cutoff can be flagged as potential CNV samples. However, this does not distinguish CNV samples from true experimental outliers. To improve the CNV detection, three additional criteria were added. The final set of criteria used for CNV detection is summarized below.
Qualitative and Quantitative Genotyping
331
Fig. 20.9. The genotyping data on a SNP from a copy number variation region. The data are shown in a scatter plot based on allele peak heights. The sample points are shaped according to the clustering calls. The potential copy number variation samples identified by the function findCNV in the R.SQNM package are circled.
l
The samples are flagged as potential CNVs if: They were identified as outliers by the clustering algorithm.
l
Their clustering strength is below a certain cutoff (an indication of the sample being on the edge of the cluster).
l
At least n of the technical replicates meet the above criteria (the default is 2 assuming four replicates per sample);
The samples are located within the angular region (the default is between 5 and 85). A SNP is flagged as a CNV SNP if at least one sample is flagged as a CNV sample. One can think of many drawbacks of this approach. For example, only a heterozygous SNP is informative and the analysis does not yield the actual copy number. However, easy experimental implementation and quick interpretation of the results make it a good starting point for CNV validation after the large chip studies. l
332
Oeth et al.
2. Materials 2.1. PCR
1. Assay Design version 3.1 (Sequenom, San Diego, CA, USA). 2. 96-channel liquid handler with corresponding software (PlateMate2x2 with Matrix ControlMate software program, Sequenom, San Diego, CA, USA). 3. Sterile and filtered tips for the liquid handler (D.A.R.T., matrix, Thermo Fisher Scientific, Hudson, NH, USA). 4. Deionized water for the liquid handler tip wash. 5. Centrifuge for 96-well/384-well PCR plates (Eppendorf, Hamburg, Germany). 6. Standard thermocycler for cycling of 384-well plates (GeneAMP PCR system 9700, Applied Biosystems, Foster City, CA, USA). 7. 384-well reaction plates (Abgene, Epsom, UK). 8. Clear adhesive PCR sealing film (Abgene, Epsom, UK). 9. Single pipettors with corresponding sterile-filtered tips: 10, 20, 200, and 1,000 mL (Eppendorf, Hamburg, Germany). 10. Multichannel pipettors with corresponding sterile-filtered tips: 10, 20, 200, and 1,000 mL (Eppendorf, Hamburg, Germany). 11. Repeater Plus with Combitips Plus (0.1 mL) (Eppendorf, Hamburg, Germany). 12. Microtubes: 1.5 and 2 mL (Eppendorf, Hamburg, Germany). 13. Reagent reservoirs (Fisher Scientific, Pittsburgh, PA, USA). 14. High performance liquid chromatography grade water (J.T. Baker, Phillipsburg, NJ, USA). 15. 10x PCR buffer (contains 20 mM MgCl2) (Sequenom, San Diego, CA, USA). 16. MgCl2 (25 mM) (Sequenom, San Diego, CA, USA). 17. PCR dNTP mix (25 mM) (Sequenom, San Diego, CA, USA). 18. 1 mM F/R PCR primer mix (forward and reverse PCR oligo per assay, desalted, resuspended, and diluted in water) /Integrated DNA Technologies, Coralville, IA, USA). 19. 5 U/mL hot start Taq DNA polymerase (Sequenom, San Diego, CA, USA). 20. 5 ng/mL DNA.
2.2. SAP Reaction
1. 96-channel liquid handler with corresponding software (PlateMate2x2 with Matrix ControlMate software program, Sequenom, San Diego, CA, USA).
Qualitative and Quantitative Genotyping
333
2. Tips for the liquid handler (D.A.R.T., matrix, Thermo Fisher Scientific, Hudson, NH, USA). 3. Deionized water for the liquid handler tip wash. 4. Centrifuge for 96-well/384-well PCR plates (Eppendorf, Hamburg, Germany). 5. Standard thermocycler for cycling of 384-well plates (GeneAMP PCR system 9700, Applied Biosystems, Foster City, CA, USA). 6. 96-well, V-bottom plate (SARSTEDT, Nu ¨ mbrecht, Germany). 7. 384-well reaction plate (‘‘sample plate’’ from the PCR). 8. Clear adhesive PCR sealing film (Abgene, Epsom, UK). 9. Single pipettors with corresponding tips: 10, 20, 200, and 1,000 mL (Eppendorf, Hamburg, Germany). 10. Multichannel pipettors with corresponding tips: 10, 20, 200, and 1,000 mL (Eppendorf, Hamburg, Germany). 11. Repeater Plus with Combitips Plus: 0.1 mL (Eppendorf, Hamburg, Germany). 12. Microtubes: 1.5 and 2 mL (Eppendorf, Hamburg, Germany). 13. Reagent reservoirs (Fisher Scientific, Pittsburgh, PA, USA). 14. Nanopure water, autoclaved. 15. 10x SAP buffer (Sequenom, San Diego, CA, USA). 16. 1.7 U/mL SAP (Sequenom, San Diego, CA, USA). 2.3. iPLEX MassEXTEND Reaction
1. 96-channel liquid handler with corresponding software (PlateMate2x2 with Matrix ControlMate software program, Sequenom, San Diego, CA, USA). 2. Tips for the liquid handler (D.A.R.T., matrix, Thermo Fisher Scientific, Hudson, NH, USA). 3. Deionized water for the liquid handler tip wash. 4. Centrifuge for 96-well/384-well PCR plates (Eppendorf, Hamburg, Germany). 5. Standard thermocycler for cycling of 384-well plates (GeneAMP PCR system 9700, Applied Biosystems, Foster City, CA, USA). 6. 96-well, V-bottom plate (SARSTEDT, Nu ¨ mbrecht, Germany). 7. 384-well reaction plates (sample plate). 8. Clear adhesive PCR sealing film (Abgene, Epsom, UK). 9. Single pipettors with corresponding tips: 10, 20, 200, and 1,000 mL (Eppendorf, Hamburg, Germany). 10. Multichannel pipettors with corresponding tips: 10, 20, 200, and 1,000 mL (Eppendorf, Hamburg, Germany).
334
Oeth et al.
11. Repeater Plus with Combitips Plus: 0.1 mL (Eppendorf, Hamburg, Germany). 12. Microtubes: 1.5 and 2 mL (Eppendorf, Hamburg, Germany). 13. Reagent reservoirs (Fisher Scientific, Pittsburgh, PA, USA). 14. Nanopure water, autoclaved. 15. iPLEX buffer plus (Sequenom, San Diego, CA, USA). 16. iPLEX termination mix (Sequenom, San Diego, CA, USA). 17. iPLEX enzyme (Sequenom, San Diego, CA, USA). 18. 5 mM/10 mM/15 mM iPLEX extend primer mix (extend PCR oligo per assay, desalted, resuspended, and diluted in water, Integrated DNA Technologies, Coralville, IA, USA). 2.4. Conditioning
1. 96-channel liquid handler with corresponding software (PlateMate2x2 with Matrix ControlMate software program, Sequenom, San Diego, CA, USA). 2. Tips for the liquid handler (D.A.R.T., Matrix, Thermo Fisher Scientific, Hudson, NH, USA). 3. Deionized water for the liquid handler tip wash. 4. Centrifuge for 96-well/384-well PCR plates (Eppendorf, Hamburg, Germany). 5. Rotator capable of holding microplates (Thermo Fisher Scientific, Hudson, NH, USA). 6. 384-well reaction plates (sample plate). 7. Clear adhesive PCR sealing film (Abgene, Epsom, UK). 8. CleanRESIN (Sequenom, San Diego, CA, USA). 9. CleanRESIN 6 mg dimple plate (Sequenom, San Diego, CA, USA). 10. CleanRESIN scraper (Sequenom, San Diego, CA, USA). 11. CleanRESIN spoon (Sequenom, San Diego, CA, USA). 12. Nanopure water, autoclaved.
2.5. Nanodispensing
1. 24-pin nanodispenser with corresponding software (Samsung with MassARRAY software, Sequenom, San Diego, CA, USA). 2. Three-point calibrant (Sequenom, San Diego, CA, USA). 3. 384-well reaction plates (sample plate). 4. SpectroCHIP (Sequenom, San Diego, CA, USA). 5. 100% ethanol (ACS quality) for the nanodispenser pin conditioning (EMD Chemicals, Gibbstown, NJ, USA). 6. 50% ethanol (diluted from above solution with Nanopure water) for the nanodispenser pin wash. 7. Deionized water for the nanodispenser pin wash.
Qualitative and Quantitative Genotyping
2.6. Matrix-Assisted Laser Desorption/ Ionization Time-of Flight Mass Spectrometry
335
1. MALDI-TOF mass spectrometer (Bruker Compact, Sequenom, San Diego, CA, USA). with corresponding software (Genoflex and RT Workstation 3.4 with Chip Linker, Caller, and Acquire, Sequenom, San Diego, CA, USA). 2. SpectroCHIP loaded with samples from above (Sequenom, San Diego, CA, USA).
3. Methods 3.1. Assay Design
1. Assay designs for iPLEX reactions require MassARRAY Assay Design version 3.1 or later to allow for SBE designs. In this new version of Assay Design, the primer design algorithms are the same, but the multiplexing efficiency and flexibility are enhanced since all SBE products use the same termination mix, making multiplexing of any set of assays easier. 2. The smaller mass separation compared with homogeneous MassEXTEND also contributes to the greater efficiency, as does the new nontemplated base addition functionality.
3.2. DNA Isolation
3.3. PCR Amplification
Any preferred isolation kit can be used. However, DNA is resuspended in autoclaved, nanopure water instead of tris(hydroxymethyl)aminomethane–EDTA (TE) buffer (see Note 1) to a final concentration of 50 ng/mL and stored at 4C prior to use (in general). 1. Each PCR has a final volume of 5 mL. 2. The reactions are set up and performed in 384-well plates. 3. The PCR master mix is prepared without DNA; the omitted reagent is added to each reaction individually (see Note 2). The volumes needed for each reaction are shown in Table 20.1. 4. The PCR cycling program for iPLEX reactions is shown below: 95C 10 min
g
95C 20 s 56C 30 s
68C 1 min 72C 3 min 68C hold
45 cycles
336
Oeth et al.
Table 20.1 iPLEX PCR cocktail master mix Reagent
Concentration (1 reaction)
Volume (1 reaction) (mL)
Volume (384 reactions plus 11.5% overhang) (mL)
Water, HPLC grade
None
1.3
556.4
10 PCR buffer with 20 mM MgCl2
1 (2 mM MgCl2)
0.5
214
25 mM MgCl2
2 mM
0.4
171.2
25 mM dNTP mix
0.5 mM each
0.1
42.8
1 mM PCR primer mix -F, -R
0.1 mM each
0.5
214
5 U/mL hot start Taq DNA polymerase
1U
0.2
85.6
5 ng/mL DNA
10 ng
2
None
Total volume (mL)
None
5
1284 (without DNA)
HPLC high performance liquid chromatography.
1. After PCR amplification, most unincorporated dNTPs are dephosphorylated with SAP. The volumes needed for each reaction are shown in Table 20.2.
3.4. SAP Reaction
2. The SAP incubation program for iPLEX reactions is shown below. 37C 40 min 85C 5 min 4C hold
Table 20.2 iPLEX shrimp alkaline phosphatase (SAP) cocktail master mix Concentration (1 reaction)
Volume (1 reaction) (mL)
Volume (384 reactions plus 38% overhang) (mL)
Nanopure water, autoclaved
None
1.53
810.9
10 SAP buffer
0.24
0.17
90.1
1.7 U/mL SAP enzyme
0.5 U
0.3
159
Total volume (mL)
None
2 (þ5 mL PCR)
1060
Reagent
Qualitative and Quantitative Genotyping
3.5. Adjusting Extension Primers
337
In MS the signal-to-noise ratios of peaks tend to decrease as masses increase. In extreme cases, signals become indistinguishable from noise, resulting in calling errors. Therefore, when multiplexing experiments are conducted it is highly recommended to adjust the concentration of oligos to equilibrate signal-to-noise ratios. A general method to adjust extension primers is to divide the primers into a low-mass group and a high-mass group (two tier): 1. All primers in the high-mass group are doubled in concentration with respect to the low-mass group. For example, in a 24-plex, the 12 lowest mass primers would be at a concentration of 0.625 mM and the 12 highest mass primers would be at 1.25 mM in the final 9-mL reaction. 2. For a plex level above 24-plex, the primers can be divided into three groups (three tier), which is shown in the following iPLEX reaction set up. In the case of a 36-plex, the 12 lowest mass primers would be at a concentration of 0.52 mM, the 12 medium mass primers would be at a concentration of 1.04 mM, and the 12 high mass primers would be at a concentration of 1.57 mM in the 9-mL reaction.
3.6. iPLEX MassEXTEND Reaction
In the iPLEX extend reaction, mass-modified nucleotides are used that make it impossible for the iPLEX enzyme to extend beyond one extension, which helps increase the plex level. A 200-shortcycling program with two cycling loops was chosen for optimum results. The program consists of one loop of five cycles sitting inside a loop of 40 cycles. These two loops result in a 200-cycle program. 1. The volumes needed for each reaction are shown in Table 20.3.
Table 20.3 iPLEX extension cocktail master mix Concentration (1 reaction)
Volume (1 reaction) (mL)
Volume (384 reactions plus 38% overhang) (mL)
Nanopure water, autoclaved
None
0.619
328.07
10 iPLEX buffer plus
0.222X
0.2
106
iPLEX termination mix
0.2 mM
0.2
106
5/10/15 mM iPLEX extend primer mix
0.52/1.04/ 1.57 mM
0.940
498.2
iPLEX enzyme, 33 U/mL
1.35 U
0.041
21.73
Total volume (mL)
None
2 (þ7 mL PCR/ SAP)
1060
Reagent
338
Oeth et al.
2. The PCR cycling program for iPLEX MassEXTEND reactions is shown below: 94C 30 sec
3.7. Sample Conditioning
94C
5 sec
52C
5 sec
80C
5 sec
72C
3 min
4C
hold
g
g
5 cycles
40 cycles
Prior to MS the samples need to be conditioned with CleanRESIN to exchange alkaline metals from reaction buffers with ammonia. This desalting is needed to decrease the noise and increase the signal-to-noise ratios of the product peaks in the MS spectrum. 1. Add 16 mL of Nanopure autoclaved water to each sample, followed by 6 mg CleanRESIN. 2. Rotate the sample plate for at least 10 min.
3.8. Dispensing
1. Prior to dispensing, the sample plate needs to be centrifuged (5 min at maximum speed) to settle down all CleanRESIN particles (see Note 3). 2. The samples and three-point calibrant (ten extra pads) are dispensed to the SpectroCHIP with a nanodispenser. The optimum dispensing speed needs to be defined with a volume check immediately before the dispensing of the actual samples. The results of dispensing samples to a SpectroCHIP with a nanodispenser depend upon a number of parameters (see Notes 4–8).
3.9. MALDI-TOF MS Analysis
A linear TOF mass spectrometer with delayed extraction is used for the analysis. All spectra are acquired in positive mode. 1. Under high-vacuum conditions, the matrix crystals are irradiated with a 337-nm laser, leading to formation of a plume of volatilized matrix and analyte as well as charge transfer from matrix ions to analyte molecules. After electric-fieldinduced acceleration in the mass spectrometer source region, the gas-phase ions travel through an approximately 1 m field-free region at a velocity inversely proportional to their m/z ratio. 2. The resulting time-resolved spectrum is translated into a mass spectrum upon calibration. MS can show more peaks than expected, which could be explained by a number of factors (see Notes 9–17).
Qualitative and Quantitative Genotyping
339
3. The mass spectra are further processed and analyzed by proprietary software (MassARRAY Caller and MassARRAY Typer) for baseline correction and peak identification. 4. The genotype determination occurs during data acquisition and takes about 7 s in total for each sample, including the acquisition and the transit time from element to element.
4. Notes 1. As the PCR is performed in only 5 mL, it is important that the TE concentration in the DNA does not inhibit the multiplexed PCR. The DNA solution should not contain more than 0.25x TE buffer (final concentration). 2. PCR enhancers have been tested and have not shown any improvement of the reaction, but have shown a disruption of the matrix/sample crystallization, which leads to a decrease of the spectral quality. PCR enhancers should therefore not be used. 3. It is very important to spin-down the sample plate before the dispensing of the samples to make sure all CleanRESIN particles settle down and are not transferred from the samples to the SpectroCHIP. CleanRESIN particles imbedded in the matrix can lead to bad or no results in the MALDI-TOF analysis. 4. Quality of the chip. Before dispensing, the chip needs to be clean and intact; it should have shiny surface without any artifacts such as dust, hair or matrix satellites, and no fingerprints (wear gloves); no scratches or bits broken off. The chip needs to have pads that are fully covered with matrix; pads with only matrix dispensed are bright white squares, which turn gray after sample dispensing. After dispensing, the user needs to wait until the matrix/sample mix has completely dried before removing the chip from the nanodispenser. The chip still needs to be clean and intact, with the same criteria as before dispensing, but with the color of the pads changed to slightly grsy. The pads still need to be fully covered with matrix. Pads with bright white squares after dispensing indicate that no or too little sample was dispensed to these pads. 5. Composition of the sample/reaction (amount of detergents, glycerol, etc.). ‘‘Watery’’ solutions such as oligos in water (calibrant) might require a higher dispensing speed than solutions with enzyme/detergents.
340
Oeth et al.
6. Maintenance of the nanodispenser. Weekly and daily maintenance needs to be performed followed by very intense rinsing of the pins (water). Use deionized water. The maintenance prevents the buildup of residue. Dirty pins can lead to sample contamination and bad crystallization. The sonicator needs to be properly filled with 50% ethanol (use Nanopure water for dilution) before the start of dispensing (check before each run). If drops of ethanol still stick to the pins after drying them in a vacuum, the drying time should be increased. Ethanol changes the surface tension of the sample, which changes the dispensing behavior of it. 7. Dispensing humidity (and speed). It is believed that proper results from dispensing depend on the humidity (which has not been thoroughly validated). Low humidity: higher dispensing speed required. High humidity: lower dispensing speed required. Humidity can change over the day/week and dispensing speeds might need to be adjusted accordingly. Lower dispensing speed: less volume dispensed Higher dispensing speed: more volume dispensed A volume check before each run on the nanodispenser is recommended. 8. Dispensing speed (solely). The optimum dispensing speed has to be used to apply samples to a SpectroCHIP. A volume check can find the optimum dispensing speed. For noncomparative and nonquantitative analysis (e.g., genotyping, allelotyping) a volume check can be performed on the same chip on which the final dispensing is performed after the matrix/sample mix has dried. Multiple dispensing does not interfere with this type of analysis. For comparative and quantitative analysis (e.g., research, QGE) it is recommended to do the volume check on a separate chip. If the volume check results in volumes that are too small, the next volume check can be performed on the same pads (with increased dispensing speed). If the volume check results in volumes that are too large, the next volume check should be performed on different pads (with decreased dispensing speed). Samsung nanodispenser: The volume calculation for the Samsung-based nanodispenser calculates volume for a complete sphere. However, what is dispensed is semispherical. Volume values resulting from the volume check need to be divided by 2. For example, if the required volume to be dispensed is 10 nL, the value to be achieved in the volume check is 20 nL. 9. Alkaline metals are introduced to the reaction through reaction and enzyme storage buffers. The presence of alkali metals can lead to an adduct formation, with the reduction of the amount of expected product (especially in G- or T-rich DNA
Qualitative and Quantitative Genotyping
341
molecules). The amount of CleanRESIN is optimized to the amount of buffers used. The ratio between the two should not be changed unless the amount of CleanRESIN is increased. If the reaction is not properly conditioned, alkaline adducts can occur. The CleanRESIN replaces alkaline metals on the DNA backbone with ammonium. DNA is prone to attract adducts that are difficult to remove. The addition of ammonium can reduce such adduct formations because it forces the DNA molecule to form ammonium adducts rather than alkali adducts, which then decay into ammonia and the molecular ion without alkali metal adducts. 10. Although ammonia is used to reduce the alkali metal adducts, in rare cases it can become an adduct itself; mostly in combination with the potassium ion. The source or reason for this is unknown at the moment. 11. Multiplex reactions can lead to unexpected primer extension products due to primer/primer interactions or primer binding to unintended templates. 12. Incomplete dephosphorylation by SAP can lead to the incorporation of additional nucleotides in addition to termination nucleotides. An insufficient SAP reaction leaves nucleotides (dNTPs) from the PCR available for the extend reaction. These nucleotides (dNTPs) will not stop the reaction, which will continue to extend until a ‘‘stop nucleotide’’ is incorporated. 13. MALDI-TOF analysis requires the use of a matrix (main component 3-hydroxypicolinic acid) that can introduce a polymeric structure (161 Da) in the very low mass area (mostly not seen in mass area for iPLEX reactions) and specific adducts: þ 63 Da: carbonic acid þ 94 Da: decarboxylated 3-hydroxypicolinic acid þ 138/139 Da: pure 3-hydroxypicolinic acid þ 188 Da: twice decarboxylated 3-hydroxypicolinic acid 14. Nucleotides can lose parts of their molecular structure through physical impact such as freeze–thaw cycles or laser impact (depurination and depyrimidation): Depurination: A-134 Da and G-150 Da (most common) Depyrimidation: C-110 Da and T-125 Da 15. Molecules can get multiple charges (protons), which allows the molecule to fly faster (refer to the examples below): Two charges per molecule (doubly charged): half the mass of the expected peak Three charges per molecule (triply charged): one third of the mass of the expected peak Four charges per molecule (quadruply charged): one quarter of the mass of expected peak
342
Oeth et al.
16. Molecules can build dimers or multimers (of the same or different species), which can share a proton (charge). Sharing a proton slows down the molecule(s) (refer to the examples below): Two molecules (dimer) of the same species per charge: twice the mass of the expected peak Three molecules of the same species per charge: three times the mass of the expected peak Two molecules (dimer) of different species per charge (e.g., molecule 1 with 5,000 Da + molecule 2 with 8,000 Da): peak at 13,000 Da 17. Multiple molecules can also share multiple charges. For example, in the example in Note 16 the peak at 13,000 Da could have resulted from two charges which lead to a peak at 6,500 Da. References 1. Kim, S., Edwards, J. R., Deng, L., Chung, W. and Ju, J. (2002) Solid phase capturable dideoxynucleotides for multiplex genotyping using mass spectrometry. Nucleic Acids Res. 30, e85. 2. Mengel-Jørgensen, J., Sanchez, J. J., Børsting, C., Kirpekar, F. and Morling, N. (2005) Typing of multiple single-nucleotide polymorphisms using ribonuclease cleavage of DNA/RNA chimeric single-base extension primers and detection by MALDI-TOF mass spectrometry. Anal. Chem. 77, 5229–5235. 3. Sauer, S., Reinhardt, R., Lehrach, H. and Gut, I. G. (2006) Single-nucleotide polymorphisms: analysis by mass spectrometry. Nat. Protoc. 1, 1761–1771. 4. Nordhoff, E., Ingendoh, A., Cramer, R., Overberg, A., Stahl, B., Karas, M., Hillenkamp, F. and Crain, P. F. (1992) Matrixassisted laser desorption/ionization mass spectrometry of nucleic acids with wavelengths in the ultraviolet and infrared. Rapid. Commun. Mass Spectrom. 6, 771–776. 5. Ding, C. and Cantor, C. R. (2003) A highthroughput gene expression analysis technique using competitive PCR and matrixassisted laser desorption ionization time-offlight MS. Proc. Natl. Acad. Sci. U.S.A. 100, 3059–3064. 6. Ding, C., Maier. E., Roscher, A. A., Braun, A. and Cantor, C. R. (2004) Simultaneous quantitative and allele-specific expression analysis with real competitive PCR. BMC Genet. 5, 8.
7. Elvidge, G. P., Price, T. S., Glenny, L. and Ragoussis, J. (2005) Development and evaluation of real competitive PCR for highthroughput quantitative applications. Anal. Biochem. 339, 231–241. 8. Huang, D. J., Nelson, M. R., Zimmermann, B., Dudarewicz, L., Wenzel, F., Spiegel, R., Nagy, B. Holzgreve, W. and Hahn, S. (2006) Reliable detection of trisomy 21 using MALDI-TOF mass spectrometry. Genet. Med. 8, 728–734. 9. Lo, Y. M., Tsui, N. B., Chiu, R. W., Lau, T. K., Leung, T. N., Heung, M. M., Gerovassili, A., Jin, Y., Nicolaides, K. H., Cantor, C. R. and Ding, C. (2007) Plasma placental RNA allelic ratio permits noninvasive prenatal chromosomal aneuploidy detection. Nat. Med. 13, 218–223. 10. Williams, N. M., Williams, H., Majounie, E., Norton, N., Glaser, B., Morris, H.R., Owen, M. J. and O’Donovan, M. C. (2008) Analysis of copy number variation using quantitative interspecies competitive PCR. Nucleic Acids Res. 36, e112. 11. Thomas, R. K., Baker, A. C., Debiasi, R. M., Winckler, W., Laframboise, T., Lin, W. M., Wang, M., Feng, W., Zander et al. (2007) High-throughput oncogene mutation profiling in human cancer. Nat. Genet. 39, 347–351. 12. van Puijenbroek, M., Dierssen, J. W., Stanssens, P., van Eijk, R., Cleton-Jansen, A. M., van Wezel, T. and Morreau, H. (2005) Mass spectrometry-based loss of heterozygosity analysis of single-nucleotide polymorphism
Qualitative and Quantitative Genotyping
13.
14.
15.
16.
17.
18.
19.
20.
loci in paraffin embedded tumors using the MassEXTEND assay: single-nucleotide polymorphism loss of heterozygosity analysis of the protein tyrosine phosphatase receptor type J in familial colorectal cancer. J. Mol. Diagn. 7, 623–630. McCullough, R. M., Cantor, C. R. and Ding, C. (2005) High-throughput alternative splicing quantification by primer extension and matrix-assisted laser desorption/ ionization time-of-flight mass spectrometry. Nucleic Acids Res. 33, e99. Tong, Y. K., Ding, C., Chiu, R. W., Gerovassili, A., Chim, S. S., Leung, T. Y., Leung, T. N., Lau, T. K., Nicolaides, K. H. and Lo, Y. M. (2006) Noninvasive prenatal detection of fetal trisomy 18 by epigenetic allelic ratio analysis in maternal plasma. Clin. Chem. 52, 2194–2202. Jurinke, C., Oeth, P. and van den Boom, D. (2004) MALDI-TOF mass spectrometry: a versatile tool for high-performance DNA analysis, Mol. Biotechnol. 26, 147–164. Jurinke, C., Denissenko, M. F., Oeth, P., Ehrich, M., van den Boom, D. and Cantor, C. R. (2005) A single nucleotide polymorphism based approach for the identification and characterization of gene expression modulation using MassARRAY. Mutat. Res. 573, 83–95. Ragoussis, J., Elvidge, G. P., Kaur, K. and Colella, S. (2006) Matrix-assisted laser desorption/ionisation, time-of-flight mass spectrometry in genomics research. PLoS Genet. 2, e100. Wjst, M. and van den Boom, D. (2005) Matrix-assisted laser desorption/ionization time-of-flight mass spectrometry. Methods Mol. Biol. 311, 125–137. Ding, C. (2006) Qualitative and quantitative DNA and RNA analysis by matrix-assisted laser desorption/ionization time-of-flight mass spectrometry. Methods Mol. Biol. 336, 59–71. Huang, D. J., Nelson, M. R. and Holzgreve, W. (2008) Maldi-TOF mass spectrometry
21.
22.
23.
24.
25.
26. 27.
28.
343
for trisomy detection. Methods Mol. Biol. 444, 123–132. van den Boom, D. and Ehrich, M. (2009) Mass spectrometric analysis of Cytosine methylation by base-specific cleavage and primer extension methods. Methods Mol. Biol. 507, 207–227. Sherry, S. T., Ward, M. H., Kholodov, M., Baker, J., Phan, L., Smigielski, E. M. and Sirotkin K. (2001) dbSNP: the NCBI database of genetic variation. Nucleic Acids Res. 29, 308–311. The International HapMap Consortium. (2007) A second generation human haplotype map of over 3.1 million SNPs. Nature 449, 851–861. Buetow, K. H., Edmonson, M., MacDonald, R., Clifford, R., Yip, P., Kelley, J., Little, D. P., Strausberg, R., Koester, H., Cantor, C. R. and Braun, A. (2001) Highthroughput development and characterization of a genomewide collection of genebased single nucleotide polymorphism markers by chip-based matrix assisted laser desorption/ionization time-of-flight mass spectrometry. Proc. Natl. Acad. Sci. U.S.A. 98, 581–584. Nelson, M. R., Marnellos, G., Kammerer, S., Hoyal, C. R., Shi, M. M, Cantor, C. R., and Braun, A. (2004) Large-scale validation of single nucleotide polymorphisms in gene regions. Genome Res. 14, 1664–1668. Schuler, G. D. (1997) Sequence mapping by electronic PCR. Genome Res.7, 541–550. Fujisawa, H., Eguchi, S., Ushijima, M., Miyata, S., Miki, Y., Muto, T. and Matsuura, M. (2004) Genotyping of single nucleotide polymorphism using modelbased clustering. Bioinformatics 20, 718–726. R Development Core Team (2008). R: a language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, http://www.Rproject.org
Chapter 21 SNP Detection Using Trityl Mass Tags Klara R. Birikh, Pablo L. Bernad, Vadim V. Shmanai, Andrei D. Malakhov, Mikhail S. Shchepinov, and Vladimir A. Korshun Abstract A new method suitable for single nucleotide polymorphism (SNP) detection using differential oligonucleotide probe extension has been developed. Sulfur-linked laser-cleavable trityl labels are implemented in this protocol. The method is based on mass spectrometry and utilizes a single surface for affinity purification of extended probes and matrix-independent desorption–ionization of the cleavable labels. The usefulness of this method for SNP genotyping is demonstrated. Key words: Mass tags, single nucleotide polymorphism detection, trityl cation, modified oligonucleotides.
1. Introduction Mass-spectrometry-based methods attract substantial interest in molecular biology research and represent a very fast developing field. Admittedly, so far biological applications of mass spectrometry have been mostly limited to proteomic studies (1–3), although examples of the method spread into the molecular biology of DNA also exist (4–6). Two main different principles are used for single nucleotide polymorphism (SNP) detection by mass spectrometry. One approach is based on the detection of the oligonucleotide probes themselves (7–9) (see Chapter 20 in this volume for details), including short photocleaved probes (10) or probes equipped with a charge label (11–13). The other approach relies on the detection of cleavable labels (14, 15). The disadvantages of analyses of entire oligonucleotides are that they have a much broader spectrum owing to a large number of isotopic variants, and a lower A.A. Komar (ed.), Single Nucleotide Polymorphisms, Methods in Molecular Biology 578, DOI 10.1007/978-1-60327-411-1_21, ª Humana Press, a part of Springer Science+Business Media, LLC 2003, 2009
345
346
Birikh et al.
sensitivity, because oligonucleotides do not desorb very well. On the other hand, this method does not require any labels (see Chapter 20) and relies on a direct SNP detection. Mass labels attracted the attention of molecular biologists owing to their extremely sharp spectra, which allows one to simultaneously detect labels that are only a few daltons apart in mass. This opens up the possibility of a large degree of multiplexing, as opposed to fluorescent labels, whose spectra are rather broad. This becomes especially important in the field of SNP genotyping (16, 17). Trityl derivatives easily ionize, desorb, and can be efficiently detected by mass-spectrometry techniques (in other words, they ‘‘fly’’ well) owing to the increased stability of the corresponding carbocations. Importantly, ionization is a result of laser irradiation and does not require the assistance of a matrix, although the presence of the latter does not adversely affect the flying. This has been put to practical use by employing trityl mass tags as labels in combinatorial chemistry (18) in combination with (matrixassisted) laser desorption–ionization (LDI) time-of-flight (TOF) detection. The latter study used mass labels comprising a standard dimethoxytrityl protecting group with variable additional masses attached. An improved version of the trityl mass tags for DNA labeling has since been developed (19). Here we describe a SNP genotyping method based on trityl labels. As opposed to the previously described method for SNP genotyping with cleavable mass labels (14), the present protocol omits minicolumn purification of the reaction mixtures and an additional step of the label cleavage by irradiation. The superior matrix-independent flying properties of the trityl-based labels made it possible to use the surface of the mass-spectrometry target as the affinity substrate in the purification protocol followed by label cleavage by a laser upon desorption–ionization, altogether largely reducing sample handling. A schematic representation of the SNP detection protocol is given in Fig. 21.1. The first step of the procedure, which is common in most SNP detection methods, includes the PCR amplification of the fragment, containing the SNP site. The second step involves a well-developed and widely used principle of allele-specific primer extension in which alleles are discriminated by the inhibited extension of a 30 -mismatched probe. For each allele variant, a probe labeled with a certain trityl is introduced into the reaction. Thereby, each nucleotide in the variable position is assigned a mass tag (the trityl, the mass of which is unique in the reaction). As a result of the extension reaction, probes which match the SNP variant are extended and the corresponding trityls are incorporated into a relatively long doublestranded DNA; whereas all mismatched probes along with an excess of matching probes remain in the form of single-stranded oligonucleotides.
SNP Detection Using Trityl Mass Tags
347
Fig. 21.1. The single nucleotide polymorphism (SNP) detection method. Mass tags are depicted as circles; the disulfide group is indicated as SS.
The last and crucial step is to detect only the extended probes. To this end the nonextended probes have to be removed from the mixture, because as they bear cleavable labels they will produce the same kind of signal in the mass spectrometer as the extended probes and bury the specific signal. To provide the means for such separation, a disulfide (–SS–) group was incorporated at the 50 -end of the reverse primer in the extension reaction. To be able to use a universal SS-modified primer for any set of primers and probes we introduced a dangling segment (tail) at the 50 -end of the reverse PCR primer, which provided an annealing site for a universal primer during the extension reaction (the universal primer is
348
Birikh et al.
depicted in Fig. 21.1 as a gray line with SS at one terminus). As a result, double-stranded DNA molecules bearing trityl labels also appeared to bear a disulfide group at the opposite terminus. Therefore, these molecules are able to bind to gold, whereas nonextended probes do not possess this property. In a subsequent experiment, crude reaction mixtures (with some extra MgCl2 added) were spotted onto a disposable gold-coated mass-spectrometry target plate (Applied Biosystems), dried, then washed from unbound material, and analyzed in a (matrix-assisted) LDI mass spectrometer without any matrix added. In the mass spectrum, each allele present in the analyzed DNA sample should express itself as a peak of the corresponding mass tag. The utility of this method has been demonstrated using a model system – a polymorphic site within the MHCIII gene [locus DJ201G24; M(19761) is C or A]. Primers for amplification of the SNP-containing fragment and allele-specific probes labeled with mass tags are presented in Fig. 21.2.
Fig. 21.2. Primers and probes for SNP 19761 of the MHCIII gene; gray bars indicate the position of the primers on the genomic DNA, corresponding sequences are given beside them, and the names of the oligonucleotides are in italic. SS a disulfide-containing group attached at the 50 -end of the oligonucleotide, Tag1 and Tag2 mass tags (flying ion masses are 558 and 572 Da).
2. Materials 2.1. Synthesis of Trityl Phosphoramidite
1. Reagents: 5-Hexynoic acid, tert-butyl trichloroacetimidate, boron trifluoride diethyl etherate, 4,40 -dimethoxybenzophenone, Pd(PPh3)4, Et3N, (NH4)2EDTA, 4-methoxyphenylmagnesium bromide, trifluoroacetic acid, N,N-disuccinimidyl carbonate, N,N-diisopropylethylamine (DIEA), 4-hydroxypenthanethiol (20), diisopropylchlorophosphoramidite. The reagents can be obtained from common commercial suppliers and are used as indicated.
SNP Detection Using Trityl Mass Tags
349
2. Inorganic reagents: Iodine, 58% HNO3, CuI, 10% Pd/C, hydrogen gas, Na2SO4, NaHCO3 (available from common commercial suppliers). 3. Solvents: Dichloromethane (DCM), diethyl ether, petroleum ether, ethyl acetate, dioxane, ethanol, dimethylformamide (DMF), toluene (PhMe), tetrahydrofuran (THF), hexane, acetone, acetic acid (available from common commercial suppliers). The solvents are high performance liquid chromatography (HPLC) grade and are used without further purification unless otherwise noted. DCM is always used freshly distilled over CaH2. THF is distilled over powdered ˚ molecular sieves under nitrogen. LiAlH4 and stored over 4-A DMF is freshly distilled under reduced pressure. 4. Chromatography: Analytical thin-layer chromatography is performed on Kieselgel 60 F254 precoated aluminum plates (Merck); spots are visualized under UV light (254 nm). Column chromatography is performed on silica gel (Merck Kieselgel 60 0.040–0.063 mm) or aluminum oxide (Fluka, activity I, 0.050–0.150 mm). 2.2. Oligonucleotide Synthesis
1. Standard reagents for oligonucleotide synthesis: Nucleoside phosphoramidites, acidic trityl deprotection solution, oxidizing solution, capping solution, acetonitrile wash, activating solution, and nucleoside solid supports from any commercial supplier. 2. Modifying phosphoramidites: Trebler phosphoramidite (Glen Research, catalog no. 10-1922-xx), spacer phosphoramidite 18 (Glen Research, catalog no. 10-1918-xx), dithiol phosphoramidite (Glen Research, catalog no. 10-1937-xx). 3. Amines for tagging: 1-Octylamine (C8H17NH2) and 1-nonylamine (C9H19NH2) (Aldrich) are used. Other primary amines from other suppliers are also suitable. 4. Buffers: 20 mM triethylammonium acetate (TEAA) buffer, pH 8.0, and HPLC-grade acetonitrile for HPLC purification of conjugates. NAP-10 columns and ‘‘saltless’’ buffer: 0.1 mM tris(hydroxymethyl)aminomethane (Tris)–HCl, pH 8.0, for desalting.
2.3. PCR and Probe Extension
1. PCRs: TopoTaq polymerase (Fidelity Systems, USA), genomic DNA, standard PCR mixture. Thermocycling in our case was carried out using a PCR amplificator Mastercycler (Eppendorf).
2.4. LDI-TOF detection of Trityl Mass Tags
1. Loading on mass-spectrometry target plate: MgCl2 (1 M); washing buffer 10 mM Tris–HCl pH 9.0, 3 M guanidine thiocyanate, 0.1 mM HSCH2CH2SO3Na, 50% THF; 50% EtOH.
350
Birikh et al.
3. Methods The methods described below outline (1) synthesis of trityl phosphoramidite (2) preparation of trityl-labeled oligonucleotides (3) PCR and probe extension using mass-tag-labeled oligonucleotides, and (4) LDI-TOF of trityl mass tags. 3.1. Synthesis of Trityl Phosphoramidite
The reaction sequence for the synthesis of trityl-bearing phosphoamidite reagent is shown in Fig. 21.3. 5-Hexynoic acid 1 is converted to tert-butyl ester 2. 4,40 -Dimethoxybenzophenone 3 is halogenated to 3-iodo derivative 4. Sonogashira coupling of compounds 2 and 4 gives substituted benzophenone 5. The next step is the hydrogenation of the triple bond. The reduced compound 6 is treated with Grignard reagent to give tritanol 7. The tert-butyl group is removed in acidic conditions, and the resulting acid 8 is converted into oxysuccinimide ester 9. This is treated with
Fig. 21.3. Synthesis of trityl phosphoramidite. Reagents and conditions: (i) tert-butyl trichloroacetimidate, BF3 Et2O, dichloromethane (DCM), petroleum ether, room temperature, 2 h; (ii) I2, HNO3, dioxane, water, 60C, 9 h; (iii) Pd(PPh3)4, CuI, Et3N, room temperature, 16 h; (iv) H2 (150 Torr), Pd/C, EtOAc, room temperature, 24 h; (v) 4methoxyphenylmagnesium bromide, tetrahydrofuran, room temperature, 16 h; (vi) CF3CO2H, DCM, room temperature, 3 h; (vii) N,N-disuccinimidyl carbonate, Et3N, DCM, room temperature, 16 h; (viii) 4-hydroxypenthanethiol, AcOH, room temperature, 2 h; (ix) i-Pr2NP(Cl)OCH2CH2CN, N,N-diisopropylethylamine, DCM, room temperature, 2 h.
SNP Detection Using Trityl Mass Tags
351
oxythiol to yiels S-trityl compound 10. The secondary hydroxyl of 10 is phosphitylated in standard conditions to afford phosphoramidite 11. 3.1.1. tert-Butyl 5-Hexynoate (Compound 2)
1. Dissolve 5-hexynoic acid (10.88 g, 97 mM) in DCM (100 mL). Dissolve tert-butyl trichloroacetimidate (34.8 mL, 194 mM) in petroleum ether (200 mL). Mix the two solutions in one flask. 2. Add boron trifluoride diethyl etherate (450 mL, 3.6 mM) and magnetically stir the mixture for 2 h. 3. Filter the solution into an evaporating flask and half-evaporate. Dilute the residue with Et2O (250 mL), wash with 5% NaHCO3 (3 100 mL), then dry the solution over anhydrous sodium sulfate (overnight). 4. Filter the solution, evaporate it, and subject the residue to chromatography on silica gel in petroleum ether containing 4% ethyl acetate. A colorless oil is obtained. Yield 10.4 g (64%). NMR [dimethyl-d6 sulfoxide (DMSO-d6)]: 2.77 (t, 1H, 4J ¼ 2.6 Hz, ”CH); 2.28 (t, 2H, J ¼ 7.5 Hz, COCH2); 2.17 (dt, 2H, 4J ¼ 2.6 Hz, J ¼ 7.1 Hz, ”CCH2); 1.65 (m, 2H, CH2CH2CH2); 1.40 (s, 9H, CH3).
3.1.2. 3-Iodo-4,40 dimethoxybenzophenone (Compound 4)
1. Dissolve 4,40 -dimethoxybenzophenone (18.1 g, 75 mM) in dioxane (70 mL) at 60C, add iodine (9.9 g, 39 mM), and mechanically stir the mixture for 15 min. 2. With stirring, add water (20 mL), and then add 58% HNO3 (41 mL) dropwise within 2 h. Continue stirring at 60C until the iodine coloring disappears (about 6–7 h). 3. Evacuate the flask (0.3 Torr, 30 min) to remove nitrogen oxides, then dilute the mixture with water (100 mL), and allow to cool to rt. A white precipitate is formed. 4. Filter off the precipitate, and wash it successively with 5% NaHCO3 and water. Suspend the slightly wet solid in EtOH (200 mL), reflux for 15 min, filter the mixture hot, and cool the filtrate to room temperature and then in a freezer to 4C. Filter the solid (16 g) off and dry it in air. The solid contains (gas–liquid chromatography) 75% 3-iodo-4,40 dimethoxybenzophenone (the desired compound), 24% starting 4,40 -dimethoxybenzophenone, and 0.7% 3,30 diiodo-4,40 -dimethoxybenzophenone. 5. Purify the product by column chromatography on silica gel in PhMe. Yield 11.2 g (40%), off-white solid. Electrospray ionization (ESI) TOF high-resolution mass spectrometry (HRMS9: m/z ¼ 383.0139 [MþH]þ, calc. for [C16H16IO3]þ 383.0139. 1 H NMR (DMSO-d6): 8.09 (d, 4J2,6 ¼ 2.0 Hz, 1H; H-2), 7.73 (dd, J5,6 ¼ 8.7 Hz, 4J2,6 ¼ 2.0 Hz, 1H; H-6), 7.70
352
Birikh et al.
(d, J20 ,30 ¼ J50 ,60 ¼ 8.7 Hz, 2H; H-20 ,60 ), 7.12 (d, J5,6 ¼ 8.7 Hz, 1H; H-5), 7.07 (d, J20 ,30 ¼ J50 ,60 ¼ 8.7 Hz, 2H; H-30 ,50 ), 3.93 (s, 3H, 4-OCH3), 3.86 (s, 3H, 40 -OCH3). 13C NMR (DMSOd6): 191.95 (CO), 162.80 (C-40 ), 160.88 (C-4), 140.25 (C-2), 132.03 (C-6), 131.91 (2C, C-20 ,60 ), 131.85 (C-1), 129.55 (C10 ), 113.90 (2C, C-30 ,50 ), 110.90 (C-5), 86.15 (C-3), 56.87 (4-OCH3), 55.57 (40 -OCH3). 3.1.3. 3-[5-(tertButyloxycarbonyl)pent-1ynyl]-4,4’dimethoxybenzophenone (Compound 5)
1. Dissolve 3-iodo-4,40 -dimethoxybenzophenone 4 (4.50 g, 12.2 mM) and tert-butyl 5-hexynoate 2 (2.10 g, 12.5 mM) in DMF (50 mL) with magnetic stirring. Add successively under argon Pd(PPh3)4 (1.40 g, 1.22 mM), CuI (465 mg, 2.44 mM), and Et3N (2.55 mL, 18.3 mM). Stir the reaction mixture overnight. 2. Dilute the mixture with water (200 mL), and extract the product with EtOAc (200 mL) in a separatory funnel. Wash the organic layer with water (4 200 mL), 0.1 M (NH4)2EDTA (4 200 mL), dry the solution over Na2SO4 overnight, and evaporate off the liquid. 3. Subject the residue to chromatography on a silica gel column using a gradient of EtOAc in PhMe (0 to 5%). Yield 3.43 g (69%), viscous yellowish oil. Rf 0.36 [10% EtOAc in PhMe (v/ v)]. ESI-TOF HRMS: m/z ¼ 409.2027 [M+H]+, calc. for [C25H29O5]þ 409.2010. 1H NMR (DMSO-d6): 7.72–7.67 (m, 3H, H-6,20 ,60 ); 7.64 (d, 1H, 4J ¼ 2.1 Hz, H-2); 7.17 (d, 1H, J ¼ 8.7 Hz, H-5); 7.08 (d, 2H, J ¼ 8.7 Hz, H-30 ,50 ); 3.91 (s, 3H, OCH3); 3.86 (s, 3H, OCH3); 2.47 (t, 2H, J ¼ 7.0 Hz, ”CCH2); 2.37 (t, 2H, J ¼ 7.4 Hz, COCH2); 1.75 (m, 2H, CH2CH2CH2); 1.39 (s, 9H, CCH3). 13C NMR (DMSO-d6): 192.64 (ArCOAr), 171.86 (OCO), 162.74 (C40 ), 162.60 (C4), 134.36 (C2), 131.86 (2C, C20 ,60 ), 131.48 (C6), 129.93 (C1), 129.74 (C1´), 113.87 (2C, C30 ,50 ), 112.23 (C3), 110.98 (C5), 94.42 (ArC”C), 79.65 (C(CH3)3), 76.65 (ArC”), 56.12 (4-OCH3), 55.56 (40 -OCH3), 33.67 (COCH2), 27.79 (3C, C(CH3)3), 23.77 (CH2CH2CH2), 18.30 (”CCH2).
3.1.4. 3-[5-(tertButyloxycarbonyl) pentyl]-4,40 dimethoxybenzophenone (Compound 6)
1. Dissolve 3-[5-(tert-butyloxycarbonyl) pent-1-ynyl]-4, 40 dimethoxybenzophenone 5 (2.03 g, 4.96 mmol) in EOAc (40 mL), add 10% Pd/C (100 mg), and hydrogenate the mixture at 150 Torr for 24 h. 2. Filter the solution, dilute the filtrate with EtOAc (150 mL), wash it with 5% NaHCO3 (100 mL) followed by 0.1 M (NH4)2EDTA (100 mL). Dry the solution over Na2SO4 overnight and evaporate off the liquid. 3. Purify the residue by column chromatography (0 to 10% of EtOAc in PhMe). Yield 1.50 g (73%), colorless oil. Rf 0.45 [10% EtOAc in PhMe (v/v)]. ESI-TOF HRMS: m/z ¼ 413.2304
SNP Detection Using Trityl Mass Tags
353
[M+H]+, calc. for [C25H33O5]+ 413.2323. 1H NMR (DMSO-d6): 7.69 (d, 2H, J ¼ 8.7 Hz, H-20 ,60 ); 7.58 (dd, 1H, J ¼ 8.4 Hz, 4J ¼ 2.2 Hz, H-6); 7.53 (d, 1H, 4 J ¼ 2.2 Hz, H-2); 7.07 (m, 3H, H-5,30 ,50 ); 3.88 (s, 3H, OCH3); 3.86 (s, 3H, OCH3); 2.59 (t, 2H, J ¼ 7.4 Hz, ArCH2); 2.16 (t, 2H, J ¼ 7.3 Hz, COCH2); 1.57–1.47 (m, 4H, COCH2CH2CH2CH2); 1.36 (s, 9H, CCH3); 1.28 (m, 2H, COCH2CH2CH2). 13C NMR (DMSO-d6): 193.38 (ArCOAr), 172.28 (OCO), 162.51 (C40 ), 160.51 (C4), 131.81 (2C, C20 ,60 ), 131.08 (C2), 130.23 (C3), 130.09 (C6), 129.93 (C1), 129.67 (C10 ), 113.77 (2C, C30 ,50 ), 110.17 (C5), 79.33 (C(CH3)3), 55.65 (4OCH3), 55.33 (40 -OCH3), 34.73 (COCH2), 29.31 (ArCH2), 28.75 (ArCH2CH2), 28.17 (ArCH2CH2CH2), 27.79 (3C, C(CH3)3), 24.48 (COCH2CH2). 3.1.5. 3-[5-(tertButyloxycarbonyl) pentyl]-4,40 ,400 trimethoxytritanol (Compound 7)
1. Dissolve 3-[5-(tert-butyloxycarbonyl)pentyl]-4,40 -dimethoxybenzophenone 6 (412 mg; 1.0 mmol) in dry THF (10 mL). To the magnetically stirred solution add 1 M 4-methoxyphenylmagnesium bromide (1.2 mL, 1.2 mM) via a syringe under argon. Keep the reaction mixture at ambient temperature overnight. 2. Dilute the reaction mixture with water (50 mL) and saturated aqueous NH4Cl (20 mL), and extract with EtOAc (2 100 mL). Dry the organic phase over Na2SO4, and evaporate off the liquid. 3. Purify the residue by chromatography on aluminum oxide using a gradient of 0 to 10% EtOAc in PhMe with 0.5% of Et3N. The yield of compound 7 is 274 mg (53%), colorless oil. Rf 0.43 [10% EtOAc in PhMe (v/v)]. ESI-TOF HRMS: m/z ¼ 503.2811 [M–OH]+, calc. for [C32H39O5]+ 503.2792. 1H NMR (DMSO-d6): 7.06 (m, 4H, H20 ,60 ,200 ,600 ); 6.97 (d, 1H, 4J ¼ 1.9 Hz, H-2); 6.88–6.79 (m, 6H, H-5,6, 30 ,50 ,300 ,500 ); 6.02 (s, 1H, OH); 3.74 (s, 3H, OCH3); 3.72 (s, 6H, OCH3); 2.45 (t, 2H, J ¼ 7.5 Hz, ArCH2); 2.12 (t, 2H, J ¼ 7.3 Hz, COCH2); 1.44 (m, 4H, COCH2CH2CH2CH2); 1.37 (s, 9H, CCH3); 1.20 (m, 2H, COCH2CH2CH2). 13C NMR (DMSO-d6): 172.27 (CO), 157.76 (2C, C40 ,400 ), 155.61 (C4), 140.68 (2C, C10 ,100 ), 140.08 (C1), 129.25 (C3), 128.89 (4C, C20 ,60 ,200 ,600 ), 128.64 (C2), 126.36 (C6), 112.68 (4C, C30 ,50 ,300 ,500 ), 109.35 (C5), 79.67 (Ar3COH), 79.34 (C(CH3)3), 55.31 (4-OCH3), 55.02 (2C, 40 -OCH3, 400 -OCH3), 34.74 (COCH2), 29.62 (ArCH2), 28.99 (ArCH2CH2), 28.14 (ArCH2CH2CH2), 27.79 (3C, C(CH3)3), 24.45 (COCH2CH2).
354
Birikh et al.
3.1.6. 3-[5-(Succinimid-1yloxycarbonyl)pentyl]4,40 ,400 -trimethoxytritanol (Compound 9)
1. Dissolve 3-[5-(tert-butyloxycarbonyl)pentyl]-4,40 ,400 -trimethoxytritanol 7 (274 mg; 0.53 mM) in dry DCM (2 mL), add trifluoroacetic acid (2 mL) and stir the mixture magnetically at ambient temperature for 3 h. 2. Evaporate the mixture, and coevaporate it with DCM (4 50 mL). The free acid 8 obtained is suitable for the next step. An analytical sample of the acid 8 can be purified by column chromatography on silica gel (20 to 30% acetone in PhMe). Rf 0.30 [30% Me2CO in PhMe (v/v)]. 1H NMR (DMSOd6): 11.92 (br, s, 1H, CO2H); 7.07 (d, 4H, J ¼ 8.7 Hz, H20 ,60 ,200 ,600 ); 6.98 (d, 1H, 4J ¼ 1.9 Hz, H-2); 6.88 (dd, 1H, J ¼ 8.4 Hz, 4J ¼ 1.9 Hz, H-6); 6.85–6.79 (m, 5H, H-5, 30 ,50 ,300 ,500 ); 6.02 (br, s, 1H, OH); 3.75 (s, 3H, OCH3); 3.72 (s, 6H, OCH3); 2.45 (t, 2H, J ¼ 7.8 Hz, ArCH2); 2.15 (t, 2H, J ¼ 7.3 Hz, COCH2); 1.45 (m, 4H, COCH2CH2CH2CH2); 1.22 (m, 2H, COCH2CH2CH2). 13 C NMR (DMSO-d6): 174.47 (CO2H), 157.76 (2C, C40 ,400 ), 155.63 (C4), 140.68 (2C, C10 ,100 ), 140.10 (C1), 129.29 (C3), 128.89 (4C, C20 ,60 ,200 ,600 ), 128.68 (C2), 126.37 (C6), 112.70 (4C, C30 ,50 ,300 ,500 ), 109.38 (C5), 79.68 (Ar3COH), 55.32 (4-OCH3), 55.03 (2C, 40 -OCH3, 400 -OCH3), 33.69 (CH2CO2H), 29.66 (ArCH2), 29.11, 28.36, 24.37 (ArCH2CH2CH2CH2). 3. Dissolve the product in DCM (15 mL), add triethylamine (0.60 mL, 4.3 mM) and N,N-disuccinimidyl carbonate (556 mg, 2.17 mM) to a magnetically stirred solution, then leave the mixture for 16 h. 4. Dissolve the residue in EtOAc (50 mL), wash with 5% NaHCO3 (50 mL) and water (50 mL), and dry the solution over Na2SO4 overnight and evaporate off the liquid. 5. Purify the residue by column chromatography on silica gel using a 15 to 30% gradient of EtOAc in PhMe as the solvent. Yield 277 mg (93%), pink amorphous solid. Rf 0.64 [30% Me2CO in PhMe (v/v)]. ESI-TOF HRMS: m/z ¼ 544.2388 [M–OH]+, calc. for [C32H34NO7]+ 544.2330. 1H NMR (DMSO-d6): 7.71 (d, 1H, J ¼ 8.9 Hz, H-5); 7.07 (d, 4H, J ¼ 8.9 Hz, H-20 ,60 ,200 ,600 ); 6.99 (d, 1H, 4J ¼ 2.1 Hz, H-2); 6.89–6.78 (m, 5H, H-6,30 ,50 ,300 ,500 ); 6.02 (s, 1H, OH); 3.89 (s, 3H), 3.86 (s, 6H) (OCH3); 2.80 (s, 4H, COCH2CH2CO); 2.61 (t, 2H, J ¼ 7.3 Hz), 2.46 (t, 2H, J ¼ 7.6 Hz) (ArCH2CH2CH2CH2CH2); 1.54 (m, 2H), 1.44 (m, 2H), 1.32 (m, 2H) (ArCH2CH2CH2). 13C NMR (DMSO-d6): 171.22 (CO2N), 170.37 (2C, COCH2CH2CO), 157.75 (2C, C40 ,400 ), 155.66 (C4), 140.64 (2C, C10 ,100 ), 140.12 (C1), 129.28 (C3), 128.87 (4C, C20 ,60 ,200 ,600 ), 128.66 (C2), 126.35 (C6), 112.68 (4C, C30 ,50 ,300 ,500 ), 109.36 (C5), 79.65
SNP Detection Using Trityl Mass Tags
355
(Ar3COH), 55.30 (4-OCH3), 55.01 (2C, 40 -OCH3, 400 OCH3), 33.67 (CH2CO2H), 29.64 (ArCH2), 29.10, 28.35, 24.39 (ArCH2CH2CH2CH2), 22.48 (2C, COCH2CH2CO). 3.1.7. 1-S-{4,40 ,400 Trimethoxy-3-[5-(Nsuccinimidyloxycarbonyl) pentyl]trityl}-4-O(diisopropylamino-2cyanethoxyphosphinyl)-4hydroxypentanethiol (Compound 11)
1. Dissolve 3-[5-(succinimid-1-yloxycarbonyl)pentyl]-4,40 ,400 trimethoxytritanol 9 (1.50 g, 2.68 mM) in AcOH (15 mL) and add 4-hydroxypenthanethiol (355 mg, 2.95 mM), then stir the mixture magnetically for 2 h. 2. Pour the mixture into water (300 mL) and extract the mixture with EtOAc (150 mL). Wash the organic layer with water (2 100 mL), dry the solution over Na2SO4, and evaporate it to dryness. 3. Purify the residue by column chromatography on silica gel (30% EtOAc in PhMe) to obtain the intermediate 1-S{4,40 ,400 -trimethoxy-3-[5-(succinimid-1-yloxycarbonyl)pentyl]trityl}-4-hydroxypentanethiol 10. Yield 1.52 g (85%), colorless oil. 1H NMR (DMSO-d6): 7.19 (d, 4H, J ¼ 8.8 Hz, H-20 ,60 ,200 ,600 ); 7.07 (d, 1H, 4J ¼ 2.3 Hz, H-2); 7.02 (dd, 1H, J ¼ 8.5 Hz, 4J ¼ 2.6 Hz, H-6); 6.85 (m, 5H, H-5,30 ,50 ,300 ,500 ); 4.25 (d, 1H, J ¼ 4.8 Hz, OH); 3.76 (s, 3H, OCH3); 3.73 (s, 6H, OCH3); 3.43 (m, 1H, CHOH); 2.80 (s, 4H, COCH2CH2CO); 2.60 (t, 2H, J ¼ 7.3 Hz, COCH2); 2.47 (t, 2H, J ¼ 7.5 Hz, ArCH2); 2.08 (m, 2H, SCH2); 1.60 (m, 2H), 1.47 (m, 2H), 1.38 (m, 1H), 1.30 (m, 3H), 1.23 (m, 2H) (COCH2CH2CH2CH2, SCH2CH2CH2); 0.94 (d, 3H, J ¼ 6.2 Hz, CHCH3). 4. Dissolve 1-S-{4,40 ,400 -trimethoxy-3-[5-(succinimid-1-yloxycarbonyl)pentyl]trityl}-4-hydroxypentanethiol 10 (1.52 g, 2.30 mM) in dry DCM (20 mL), add with stirring DIEA (0.52 mL, 3.0 mM), and then add dropwise 2-cyanoethyl diisopropylchlorophosphoramidite (0.27 mL, 2.5 mM) under argon. Continue stirring for 2 h. 5. Dilute the reaction mixture with CHCl3 (50 mL), wash it with water (2 50 mL), it dry over Na2SO4, and evaporate off the liquid. 6. Dissolve the residue in DCM (4 mL) and induce precipitation by adding the solution dropwise to magnetically stirred hexane (300 mL). Remove the solvent, dissolve the residue in dry DCM (5 mL), evaporate off the liquid, then dry the flask in vacuo. The procedure gives phosphoramidite 11 as a colorless oil (1.71 g, 86%). 1H NMR (DMSO-d6): 7.29 (m, 4H, H20 ,60 ,200 ,600 ); 7.16–7.11 (m, 2H, H-2,6); 6.87–6.81 (m, 5H, H-5,30 ,50 ,300 ,500 ); 3.87–3.53 (m, 14H, OCH3, POCH, POCH2, NCH); 2.78 (s, 4H, COCH2CH2CO); 2.64–2.52 (m, 6H, ArCH2, COCH2, CH2CN); 2.16 (m, 2H, SCH2);
356
Birikh et al.
1.69 (m, 2H), 1.56–1.32 (m, 8H) (COCH2CH2CH2CH2, SCH2CH2CH2); 1.19–1.10 (m, 15H, CHCH3). 31P NMR (MeCN-d3): 147.95, 147.08. 3.2. Oligonucleotide Synthesis
Oligonucleotides can be made using any standard solid-phase DNA synthesizer (in our case the Applied Biosystems 392 DNA/RNA synthesizer) following standard phosphoramidite chemistry. Nonmodified oligonucleotides: primers US (5 0 GCTCGCTGGGCGGTGCCGATTTCTG) and DS-tail (5 0 AGAAGGTCGGAGTCAACGGATCTCCGGGGCATTGTC TAAGCGGGAC). Modified oligonucleotides: Tail-SS (5 0 [dithiol]-AGAAGGTCGGAGTCAACGGAT), probe A (5 0 [C 9 H 19 NH-tritylS-spacer18] 3 -trebler-spacer18-CCGCCA GTCTGGATGTAATGGGCCA), probe C (5 0 -[C 8 H 17 NHtritylS-spacer18]3-trebler-spacer18-CCGCCAGTCTGGATGT AATGGGCCC).
3.2.1. Synthesis of 50 Disulfide-Containing Oligonucleotides
The disulfide group at the 50 -end of the reverse primers is introduced using dithiol phosphoramidite reagent in the last step of the automated oligonucleotide synthesis. Incorporation of the phosphoramidite is carried out under conditions recommended by the manufacturer.
3.2.2. Synthesis of TritylLabeled Probes
The steps in the synthesis of trityl-labeled probes (probe A and probe C) are shown in Fig. 21.4. The procedure includes solidphase oligonucleotide synthesis, coupling of several modifying phosphoramidites, tagging with particular amines, and isolation of the mass-tag-labeled probe. 1. Synthesize the DNA part of the probe using standard phosphoramidite chemistry in a DNA synthesizer. 2. After the last step, introduce PEG-hexamer linker phosphoramidite reagent (spacer phosphoramidite 18). Incorporation of this phosphoramidite is carried out under conditions recommended by the manufacturer. 3. Introduce the branching reagent trebler phosphoramidite (21). Incorporation of this phosphoramidite is carried out under conditions recommended by the manufacturer. 4. Repeat step 2: introduce PEG-hexamer linker phosphoramidite reagent (spacer phosphoramidite 18). Incorporation of this phosphoramidite is carried out under conditions recommended by the manufacturer (see Note 1). 5. Introduce trityl phosphoroamodite reagent 11 using the standard synthetic cycle of automated oligonucleotide synthesis, apart from the oxidation step, in which the concentration of iodine is fivefold lower as compared with the standard conditions (see Note 2).
SNP Detection Using Trityl Mass Tags
357
Fig. 21.4. Synthesis of trityl mass-tag-labeled oligonucleotides and structures of trityl mass-tag cations.
6. Take the DNA synthesis column out of the synthesizer. Using a syringe, treat the solid support manually with 0.05 M solution of aliphatic amine in THF (10 min). In this study 1octylamine (C8H17NH2) and 1-nonylamine (C9H19NH2) were used. Wash the support with THF, then with MeCN and dry it (see Note 3). 7. Carry out ammonolysis deprotection under standard conditions. 8. Purify oligonucleotide conjugates by reverse-phase HPLC on a C8 column. Conditions: acetonitrile gradient from 20 to 80% in 20 mM TEAA buffer, pH 8.0. Collect fractions containing the most hydrophobic compound (containing three trityl labels). 9. Evaporate the solution and desalt the conjugate on NAP-10 column. 3.3. PCR and Probe Extension
PCR and probe extension reactions are carried out using TopoTaq polymerase (see Note 4) under the conditions recommended by the manufacturer. 1. Synthesize the double-stranded template for probe extension by PCR in a 20-mL reaction volume with 10 ng of genomic DNA, primers US and DS-tail (0.1 mM in each), buffer, and NTPs as recommended by the manufacturer. The thermal cycle is 45 s at 94C, 1 min at 60C, and 30 s at 72C (30–40 cycles).
358
Birikh et al.
2. For the following extension reaction, prepare a reaction mixture (20 mL) containing 2 mL of the PCR mixture, 0.5 mM probes A and C and primer Tail-SS, buffer, and NTPs as recommended by the manufacturer. The thermal cycle for the extension reaction is 30 s at 90C, 1 min at 55C, and 30 s at 72C (six cycles). 3.4. LDI-TOF Detection of Trityl Mass Tags 3.4.1. Gold Affinity Purification
Care should be taken while performing this step, which should help avoid subsequent false-positive signals (see Note 5). 1. Adjust the magnesium concentration in the extension mixture to 0.1 M using 1 M MgCl2. 2. Spot 4 mL of the extension mixture on a circular spot of a disposable gold-plated mass-spectrometry target plate (Applied Biosystems, USA), and leave it at room temperature for approximately 30 min to dry (see Note 6). 3. Rinse the plate with water, place it in a container with 20–30 mL of washing buffer (10 mM Tris–HCl pH 9.0, 3 M guanidine thiocyanate, 0.1 mM HSCH2CH2SO3Na, 50% THF), and incubate the container for 10 min at room temperature with gentle shaking. 4. Rinse the plate thoroughly with 50% EtOH and take it for mass spectrometry (see Note 7).
Fig. 21.5. SNP genotyping. Representative spectra of the samples obtained from homozygous DNA (C at position 19761) (top spectrum) and heterozygous DNA (both alleles present) (bottom spectrum).
SNP Detection Using Trityl Mass Tags 3.4.2. LDI-TOF Analysis
359
LDI-TOF mass spectra in our case were obtained using a Voyager Elite Biospectrometry Research Station (PerSeptive Biosystems, Vestec Mass Spectrometry Products) in the positive ion mode. Set the laser power to the maximum. Signals from three to five shots are accumulated for each spectrum (see Note 8). A representative example of the typical result is shown in Fig. 21.5. Homozygous DNA (top spectrum) produces only one major peak, corresponding to probe C, whereas the heterozygous DNA (bottom spectrum) shows peaks corresponding to both probe A and probe C. The structures of the trityl cations detected are shown in Fig. 21.5.
4. Notes 1. Branching reagent trebler was used to attach three trityl labels to each DNA molecule during automated synthesis to further increase sensitivity. PEG linkers were introduced between the oligonucleotide and the trebler as well as between the trebler and trityl labels. These linkers were employed to prevent interference of bulky hydrophobic trityl residues with oligonucleotide hybridization and extension. 2. Standard (nondiluted) iodine oxidizing reagent causes deprotection of the S-trityl compound, thus dramatically lowering the yield of the desired conjugate. 3. Other aliphatic amines (C2H5NH2, C3H7NH2, C4H9NH2, etc.) are suitable for generation of a variety of mass tags. More prolonged amine treatment can cause cleavage of an oligonucleotide from a solid support and loss of the material. 4. This PCR system was chosen because it utilizes a relatively high pH (9.0), which is important for trityl stability during extension reaction; besides, the TopoTaq buffer turned out to be favorable for immobilization of the labeled DNA on gold (see also Note 5). 5. False-positive signals might appear either as a result of extension through a 3´-mismatch or owing to incomplete washout of the nonextended probes from the gold surface. To distinguish between these two backgrounds and check that the washing works well, a preliminary test experiment needs to be carried out. In this experiment only one probe is used for the extension reaction, while the second probe (at the same concentration) is added after the reaction; hence, the DNA remains totally single stranded. Then loading on gold and washing is performed according to the instructions in Section 3.4, followed by
360
Birikh et al.
mass spectrometry. Nonspecific signal in this set up in our hands was within 15 % of the main peak (of the extended probe) intensity. 6. Usually, salt-containing solutions dry into uneven white patches; however, our samples dried into a transparent film, evenly covering the surface of the sample circle of the plate. We attribute this effect to the presence of some detergent or polymer in the reaction buffer for the TopoTaq polymerase used in these experiments. 7. The quality of the reagents used for washing the plate may dramatically affect the quality of the spectra; therefore, all solvents used for washing should be HPLC grade; salts should be molecular biology grade. 8. Sometimes the spectra appeared cleaner if the signal was accumulated starting from the second shot.
Acknowledgments We thank Susan Wheeler and Kaajal Patel from Oxford Gene Technology Ltd for their helpful advice regarding mass spectrometry, and Irina Udalova from the Wellcome Trust Centre for Human Genetics, Oxford University, for her help in selecting the genetic model. References 1. Lane, C. S. (2005) Mass spectrometrybased proteomics in the life sciences. Cell. Mol. Life Sci. 62, 848–869. 2. Domon, B. and Aebersold, R. (2006) Mass spectrometry and protein analysis. Science 312, 212–217. 3. Cravatt, B. F., Simon, G. M. and Yates J. R. 3rd. (2007) The biological impact of massspectrometry-based proteomics. Nature 450, 991–1000. 4. Hofstadler, S. A., Sannes-Lowery, K. A. and Hannis, J. C. (2005) Analysis of nucleic acids by FTICR MS. Mass Spectrom. Rev. 24, 265–285. 5. Sauer, S. (2006) Typing of single nucleotide polymorphisms by MALDI mass spectrometry: principles and diagnostic applications. Clin. Chim. Acta 363, 95–105. 6. Tost, J. and Gut, I. G. (2006) DNA analysis by mass spectrometry-past, present and future. J. Mass Spectrom. 41, 981–995.
7. Haff, L. A., Belden, A. C., Hall, L. R., Ross, P. L. and Smirnov, I. P. (2001) SNP genotyping by MALDI-TOF mass spectrometry. In: J. N. Housby (Ed.) Mass Spectrometry and Genomic Analysis, Kluwer, New York, pp. 16–32. 8. Griffin, T. J. and Smith, L. M. (2001) Single-nucleotide polymorphism analysis by MALDI-TOF mass spectrometry. In: J. N. Housby (Ed.) Mass Spectrometry and Genomic Analysis, Kluwer, New York, pp. 1–15. 9. Storm, N., Darnhofer-Patel, B., van den Boom, D. and Rodi, C. P. (2003) MALDI-TOF mass spectrometry-based SNP genotyping. Methods Mol. Biol. 212, 241–262. 10. Wenzel, T., Elssner, T., Fahr, K., Bimmler, J., Richter, S., Thomas, I. and Kostrzewa, M. (2003) Genosnip: SNP genotyping by MALDI-TOF MS using photocleavable oligonucleotides. Nucleosides Nucleotides Nucleic Acids 22, 1579–1581.
SNP Detection Using Trityl Mass Tags 11. Sauer, S., Lechner, D., Berlin, K., Lehrach, H., Escary, J.-L., Fox, N. and Gut, I. G. (2000) A novel procedure for efficient genotyping of single nucleotide polymorphisms. Nucleic Acids Res. 28, e13. 12. Sauer, S., Lechner, D., Berlin, K., Plancon, C., Heuermann, A., Lehrach, H. and Gut, I. G. (2000) Full flexibility genotyping of single nucleotide polymorphisms by the GOOD assay. Nucleic Acids Res. 28, e100. 13. Sauer, S., Lechner, D. and Gut, I. G. (2001) The GOOD assay. In: J. N. Housby (Ed.) Mass Spectrometry and Genomic Analysis, Kluwer, New York, pp. 50–65. 14. Kokoris, M., Dix, K., Moynihan, K., Mathis, J., Erwin, B., Grass, P., Hines, B. and Duesterhoeft, A. (2000) High-throughput SNP genotyping with the Masscode system. Mol. Diagn. 5, 329–340. 15. Hammond, N., Koumi, P., Langley, G. J., Lowe, A. and Brown, T. (2007) Rapid mass spectrometric identification of human genomic polymorphisms using multiplexed photocleavable mass-tagged probes and solid phase capture. Org. Biomol. Chem. 5, 1878–1885.
361
16. Marnellos, G. (2003) High-throughput SNP analysis for genetic association studies. Curr. Opin. Drug Discov. Dev. 6, 317–321. 17. Shi, M. M. (2002) Technologies for individual genotyping: detection of genetic polymorphisms in drug targets and disease genes. Am. J. Pharmacogenomics 2, 197–205. 18. Shchepinov, M. S., Chalk, R. and Southern, E. M. (2000) Trityl mass-tags for encoding in combinatorial oligonucleotide synthesis. Tetrahedron 56, 2713–2724. 19. Birikh, K. R., Korshun, V. A., Bernad, P. L., Malakhov, A. D., Milner, N., Khan, S., Southern, E. M. and Shchepinov M. S. (2008) Novel mass tags for single nucleotide polymorphism detection. Anal. Chem. 80, 2342–2350. 20. Filippi, J.-J., Fernandez, X., Lizzani-Cuvelier, L. and Loieseau, A.-M. (2002) Convenient enantioselective synthesis of new 1,4sulfanylalcohols from g-lactones. Tetrahedron Lett. 43, 6267–6270. 21. Shchepinov, M. S., Udalova, I. A., Bridgman, A. J. and Southern, E. M. (1997) Oligonucleotide dendrimers: synthesis and use as polylabelled DNA probes. Nucleic Acids Res. 25, 4447–4454.
Chapter 22 Putting the Invader Assay to Work: Laboratory Application and Data Management Yi Zhang, Edward Smith, and Michael Olivier Abstract Choosing a single nucleotide polymorphism genotyping method that suits specific research needs is not much less of a challenge than determining the genetic components underlying the disease and/or trait being investigated. This is especially true with a long list of tempting methodologies available, as summarized in this book. Here, from an end-user point of view, we discuss how a commercially available genotyping platform, the Invader assay, can be utilized to meet the needs and demands of human genomic research in a laboratory. Key words: Invader assay, single nucleotide polymorphism, genotyping, fluorescence resonance energy transfer, serial invasive signal amplification reaction, sequence management pipeline.
1. Introduction An individual’s DNA sequence is unique. Although the differing loci constitute only a fraction of the entire genome, they encode the variabilities ranging from appearance to governing susceptibility to disease. As the most common form of genetic variation in humans, single nucleotide polymorphisms (SNPs) influence a number of conditions and have served as successful targets within genes responsible for a variety of diseases. When it comes to choosing a SNP genotyping method that is most suitable for a particular research project, investigators can easily feel overwhelmed by the list of available A.A. Komar (ed.), Single Nucleotide Polymorphisms, Methods in Molecular Biology 578, DOI 10.1007/978-1-60327-411-1_22, ª Humana Press, a part of Springer Science+Business Media, LLC 2003, 2009
363
364
Zhang, Smith, and Olivier
technologies and platforms. On many occasions, modifications of the original commercialized version of a chosen technology are necessary to develop a truly cost and time efficient platform. Genotyping projects of small to medium size have been carried out using a method based on the Invader technology in our laboratory (1–7). As has been described (8, 9), a structure-specific flap endonuclease can cleave and therefore distinguish an oligonucleotide triplex formed by hybridization of probes to target DNA with a particular SNP allele (8, 9). To generate detectable genotype results, a fluorescence resonance energy transfer (FRET) cassette has been engineered to anneal with a fragment of the cleaved product from the first step to elicit a second flap endonuclease cleavage reaction. However, when it is cleaved this time, a fluorescent dye will emit signals upon its release from the proximity to a quencher on the same strand of the FRET cassette (Fig. 22.1).
Primary reaction
Secondary reaction
Cleavage site
ap
Fl
5’
Cleavage site
F1
1
Primary probe 1 Tgca 3’ Invader probe 5’ g t c N cagAcgt 3’ 5’ Allele 1 target DNA
5’ 5’
T
F1
3’ FRET cassette 1
ap
Fl
5’
her
enc
Qu
2
C
Primary probe 2 gca 3’ Invader probe 5’ g t c N cagAcgt 3’ 5’ Allele 1 target DNA
No cleavage, thus no secondary reaction
Fig. 22.1. The serial invasive signal amplification reaction process. Two oligonucleotides (an allele-specific and an invader probe) anneal to the complementary region of the target molecule to form a three-dimensional structure. The allele-specific probe has a 50 noncomplementary flap which extends past the single nucleotide polymorphism (SNP) locus and does not anneal to the target. This flap, including the base complementary to the SNP allele, is cleaved in a primary reaction, and then acts as an invader probe in a secondary reaction with a fluorescence resonance energy transfer cassette. Once again the required Invader three-dimensional structure is formed and 0 a fluorophore-labeled 5 nucleotide is cleaved from a quencher molecule, generating a measurable signal. As depicted, a piece of target DNA with the SNP A allele can bind with primary probe 1, which will trigger the secondary reaction to emit fluorescence. In contrast, this target DNA cannot form the three-dimensional structure with primary probe 2, so there is no subsequent emission of fluorescence.
Putting the Invader Assay to Work
365
2. Materials 2.1. Amplification of Regions of Interest for Invader Reaction
1. Genomic DNA can be isolated using many commercial reagents (e.g., Puregene, Gentra Systems, Minneapolis, MN, USA). Genomic DNA (1 ng/mL) is stored in 96-well deep-well plates at –20C.
2.1.1. Preparation of DNA Samples
2. 96-well deep-well plates (Thermo Scientific, Rockford, IL, USA), plus adhesive plate seals (Eppendorf North America, Westbury, NY, USA). 3. 384-well plates (Eppendorf North America, Westbury, NY, USA), plus adhesive plate seals. 4. Biomek FXP automation station (Beckman, Fullerton, CA, USA), plus compatible BioRobotix tips (Molecular BioProducts, Petaluma, CA, USA).
2.1.2. PCR Amplification of Regions of Interest
1. DNA samples in 384-well plates (5 mL per well) (see Section 2.1.1). 2. Forward and reverse PCR primers, 20 mM ineach. 3. 10 PCR buffer (Invitrogen, Carlsbad, CA, USA). 4. 50 mM MgCl2 (Invitrogen, Carlsbad, CA, USA). 5. 20 mM (each) dNTPs (Invitrogen, Carlsbad, CA, USA). 6. Platinum Taq polymerase (Invitrogen, Carlsbad, CA, USA). 7. Autoclaved purified water. 8. 50-mL reagent reservoirs (Corning, Corning, NY, USA). 9. 12-channel pipette (capable of pipetting 5 mL; Rainin, Woodburn, MA, USA). 10. 15-mL conical tubes (BD Biosciences, Franklin Lakes, NJ, USA). 11. Sterilized pipette tips (four boxes for each 384-well plate). 12. Ice. 13. Plate seals. 14. A roller with a rubber head. 15. Eppendorf 5810R centrifuge with plate adaptors (Eppendorf North America, Westbury, NY, USA). 16. GeneAmp PCR system 9700 thermocyclers with 384-well plate holders (Applied Biosystems, Foster City, CA, USA).
2.2. Genotyping with the Invader Assay 2.2.1. Preparation of the ‘‘Replicator Pins’’
1. ‘‘Replicator pins’’ (V & P Scientific, San Diego, CA, USA). 2. Five clean empty pipette tip boxes (washed but no need to be sterilized). 3. Blotting paper, precut to the size of a pipette tip box (Whatman, Maidstone, UK).
366
Zhang, Smith, and Olivier
4. A glass container (13.97 cm wide, 19.05 cm long, 4.45 cm high). 5. A mushroom brush. 6. VP110 surfactant (V & P Scientific, San Diego, CA, USA). Add 160 mL sterilized ultrapure water to one bottle (30 mL) of surfactant, and store (covered) the mixture in the glass container. This solution can be stored at room temperature for up to 2 months. 7. 10% Clorox bleach solution. 8. 95% ethanol. 9. A hair dryer. 10. Autoclaved purified water. 2.2.2. Wet-Format ‘‘Replicator Pin’’ Genotyping Using Invader Technology
1. 10 mL PCR products in 384-well plates (see Section 2.2). 2. Invader and primary probe oligonucleotides designed at the Third Wave Web site (weblink can be requested from the corresponding author), used at a working concentration of 200 mM. 3. 2.6 M betaine (Sigma, St. Louis, MO, USA). 4. TE buffer: 10 mM tris(hydroxymethyl)aminomethane–HCl pH 7.5, 1 mM EDTA. 5. CleavaseXI/MgCl2 mix solution (Third Wave Technologies, Madison, WI, USA). 6. FRET mix (Third Wave Technologies, Madison, WI, USA). 7. Prepared replicator pins (see Section 2.3), plus washing solutions in tip boxes. 8. 15-mL conical tubes (BD Biosciences, Franklin Lakes, NJ, USA). 9. Centrifuge with plate adaptors. 10. Aluminum foil plate seals and a rubber roller. 11. Thermocyclers with 384-well plate holders (Applied Biosystems, Foster City, CA, USA).
2.3. Genotype Data Collection and Management 2.3.1. Raw Data Collection
1. An LJL Analyst HT (LJL BioSystems, Sunnyvale, CA, USA). 2. 384-well plates with Invader assayed PCR products (see Section 2.2). 3. Seals and a roller. 4. A directory with six subdirectories: Fluorescein; Red dye; Sample lists; Raw data; Scores; Plots
2.3.2. Cluster Analysis of Genotype Data
1. Output files from the fluorescent plate reader, one for each fluorophore. 2. Four sample lists, one for each 96-well plate used initially. 3. Access to MCW Sequence Management Pipeline (weblink can be requested from the corresponding author).
Putting the Invader Assay to Work
367
3. Methods 3.1. Amplification of Regions of Interest for Invader 3.1.1. Preparation of DNA Samples for the Invader Genotyping
3.1.2. Amplification of Regions of Interest Using PCR
Genomic DNA. Five microliters of DNA (1 ng/mL) is required per sample. Large volumes (e.g., 600 mL) of DNA from stock solutions are diluted and aliquoted into 96-well deep-well plates and stored at – 20C. DNA samples are thawed at 4C overnight, and 5 mL is transferred into 384-well plates using a Biomek FXP automation station (Fig. 22.2). Therefore, samples originally on four 96-well plates are combined into one 384-well plate (see Note 1). Included on each 384well plate are eight CEPH (Centre d’Etude du Polymorphisme Humain) DNA positive controls and 16 negative (water) controls. 1. Standard PCR primers are designed for the genomic regions containing the target SNPs using, e.g., Primer3 (weblink can be requested from the corresponding author). Default settings are used, with the following exceptions: (1) change the ‘‘Product Size Ranges’’ to ‘‘80-150 150-250 250-400 400-500’’; (2) change the ‘‘Max 3’ Stability’’ to ‘‘8.0’’; (3) change the ‘‘Primer Size’’ to ‘‘Min:21,’’ ‘‘Opt: 23,’’ ‘‘Max: 26’’; (4) change the ‘‘Primer Tm’’ to ‘‘Min: 59.0,’’ ‘‘Opt:62.0,’’ ‘‘Max: 65.0’’; (5) change the ‘‘Max’’ of the ‘‘Primer GC%’’ to ‘‘50.0.’’ 2. Invader probes are designed at the Third Wave Web site (weblink can be requested from the corresponding author). To distinguish the allele-specific primary probes, sequences with 5’ ‘‘CGC’’ will be named ‘‘FAM’’ probes, whereas those with ‘‘ACG’’ will be named ‘‘RED’’ (see Note 2).
a
b Sample_1 (A1) Sample_2 (A2) Sample_3 (A3) Sample_4 (A4) Sample_5 (A5) Sample_6 (A6) Sample_7 (A7) Sample_8 (A8) Sample_9 (A9) NTC (A10) NTC (A11) NTC (A12) NTC (B1)
Fig. 22.2. An example of converting DNA samples in 96-well format into 384-well format. (a) The layout of four 96-well plates when combined into a single 384-well plate. Each 96-well plate is distinguished by a different color code and all positions in the 384-well plate are labeled according to their original positions on the 96-well plate. (b) Example sample list file. Do not include the data in parentheses; these are shown to clarify the order of the list.
368
Zhang, Smith, and Olivier
3. Once the allele-specific and the Invader probes have been prepared, DNA regions containing the SNPs of interest can be amplified by PCR from genomic DNA. The PCR master mix recipe for a single reaction and a 384-well plate and the thermocycler program are given in Tables 22.1, 22.2 and 22.3 (see Notes 3 and 4).
Table 22.1 Master mixes used for PCR amplification of genomic loci Reagent
Starting concentration
Final concentration
Volume per well (mL)
Volume per 96well plate (mL)
Volume per 384well plate (mL)
Primer F
20 mM
2 mM
0.5
55
207.5
Primer R
20 mM
2 mM
0.5
55
207.5
PCR buffer
10
2
1.0
110
415.0
MgCl2
50 mM
5 mM
0.5
55
207.5
dNTPs
2.5 mM
3.2 mM
0.8
88
332.0
Platinum Taq
5 U/mL
0.07 U/mL
0.07
7.7
29.05
Water
NA
NA
1.63
179.3
676.45
Total
NA
NA
5.00
550
2075.00
NA
Table 22.2 Master mixes used for Invader genotyping Reagent
Concentration
Volume for 1 SNP per 384well plate (mL)
Volume for more than 1 SNP per platea (mL)
Primary probe 1
200 mM
9.1
2.30
Primary probe 2
200 mM
9.1
2.30
Invader probe
200 mM
1.0
0.25
Betaine
2.6 M
500.0
130.40
TE buffer
NA
1230.0
330.00
Cleavase/MgCl2
NA
158.0
43.30
FRET mix
NA
553.0
152.00
Total
NA
2460.2
660.55
SNP single nucleotide polymorphism, TE tris(hydroxymethyl)aminomethane–EDTA, FRET fluorescence resonance energy transfer a Volumes used for 96 samples, i.e., up to four SNPs assayed per plate
Putting the Invader Assay to Work
369
Table 22.3 Thermocycler programs used for PCR amplification of genomic loci (A) and Invader genotyping (B) A Stage
Temperature (C)
Duration
No. of cycles
Denature
94
3 min
1
Denature
94
30 s
Anneal
60
30 s
Extend
72
30 s
Extend
72
10 min
Store
4
1
40
1
B Stage Denature
Invader reaction
3.2. Wet-Format ‘‘Replicator Pin’’ Genotyping Using Invader 3.2.1. Preparation of the ‘‘Replicator Pin’’
Temperature (C)
Duration (min)
95
5
63
15
1. Set up five reservoirs (use empty tip boxes) containing (in order) (1) 10% bleach (2) Milli-Q water (3) Milli-Q water (4) Milli-Q water, and (5) 95% ethanol. Fill each successive reservoir with increasing volumes of the respective washing solution. Lay one piece of precut filter paper in front of each reservoir for drying pins. 2. Soak the replicator pin in cold soapy water for 5 min. 3. Brush the pins with a mushroom brush. Wash the grooves thoroughly. 4. Rinse the pins with distilled water. 5. Dip and swirl the pins in the three Milli-Q water reservoirs, beginning with the lowest volume. Blot the pins briefly but vigorously on filter paper between each wash. 6. Wash the pins in 95% ethanol and blot them on paper. 7. Dry the pins with the hairdryer on a cool setting. 8. Dip the pins into the reservoir containing the surfactant VP110 to half pin length. Avoid air bubbles on the grooves. 9. Blot the pins on paper. Repeat from step 8. 10. Dry the pins with the hairdryer on a cool setting. 11. Repeat steps 5 and 6 twice.
370
Zhang, Smith, and Olivier
12. Repeat step 7. The replicator pins are now ready for use. 13. To clean the pins after each transfer of PCR product, dip and swirl the pins in all five reservoirs and blot them after each dipping and swirling. Dry the pins with the hairdryer and store them at room temperature. The pins will need to be treated with surfactant after six to eight PCR product transfers, and before the first use on a given day. 3.2.2. The Invader Assay
1. Make a genotyping master mix on ice for each SNP in a 15-mL conical tube (Table 22.2). 2. Pipette 5.5 mL of the master mix solution into the wells of a fresh 384-well plate. 3. Using the replicator pin, transfer about 0.5 mL of a PCR product from the 384-well plate described in Section 3.1 into the fresh one described in Section 3.2.2. This is done by dipping the replicator pins into the wells of the PCR plate (see Notes 5 and 6), followed by dipping them into the new plate with the Invader master mix. When the pins are touching bottoms of the wells, rock them back and forth to mix the two solutions (PCR product and genotyping master mix). 4. Seal the plates with foil seals and the roller. Briefly centrifuge the plates (1,000 rpm for 1 min) to collect the samples at the bottom of the wells. 5. Run the Invader assay on a thermocycler as described in Table 22.3. Fifteen minutes is the recommended incubation time for the initial genotyping of a SNP. The incubation period can be shortened or lengthened on the basis of the resulting signal strength (see Note 7).
3.3. Genotype Data Collection and Management 3.3.1. Raw Data Collection
3.3.2. Cluster Analysis of Genotype Data
1. Create a directory with six subdirectories: Fluorescein; Red dye; Sample lists; Raw data; Scores; Plots. 2. Once the Invader reaction is complete, read the plates in the LJL plate reader. Save the results of genotyping with the ‘‘FAM’’ probe in the ‘‘Fluorescein’’ folder and the ‘‘RED’’ probe in the ‘‘Red dye’’ folder. The excitation wavelengths for FAM and RED are 485 and 580 nm, respectively, whereas the emission wavelengths are 530 and 630 nm, respectively. 1. A genotype clustering algorithm designed for use with the Invader system (weblink can be requested from the corresponding author) (see Note 8).
Putting the Invader Assay to Work
371
2. To begin the analysis, click on ‘‘Raw data muncher.’’ This will then open a page which will prompt you to enter an e-mail address, the raw probe readings, matched sample lists, and the designated names for the results. 3. The e-mail address is needed to function as a data storage folder for SNPs under investigation. This could be a real or fake e-mail address, but it is recommended to adopt a format like name_of_the_investigator@name_of_the_project for the sake of easy recognition. Second, provide FAM probe readings and RED dye probe readings as prompted, by browsing to their respective folders. Next, sample IDs are required; our system uses a 4 96well plate to 1 384-well plate format, i.e. the order of the samples on the original 96-well plate is maintained throughout (Fig. 22.2a). Thus, sample IDs are entered as quarter sectors (one for each 96-well plate); these files should be saved in the ‘‘sample lists’’ folder in plain text format (one sample ID per line; for an example, see Fig. 22.2b). The negative (water) controls should be named ‘‘NTC’’ (no target control), otherwise the analysis software will generate an error message and the analysis will fail. Finally, each of the four sectors needs to be named. Once the form is completed, click on ‘‘submit.’’ 4. On a new page, raw clustering data will be presented. These will be displayed in four separate charts (one for each original 96well plate). These raw data provide real fluorescence levels that are important for evaluating whether an assay has been successful. Look for a clustering of dots (one for each sample) with approximate values of 100,000 on the x-axis (FAM) and 10,000 on the y-axis (RED). Below each plot is a link to the results in text format. If the raw data are satisfactory (see Note 9), click on ‘‘concatenate data.’’ 5. On a new page, the user is prompted to enter the ‘‘e-mail’’ he or she used to store the data. 6. A page with all archived data in the same project will be shown. The user needs to provide a name for the data set to be concatenated on the top and also select the four sectors for that genotyped SNP (see Note 10). 7. On a new page, concatenated data will be presented. The algorithm will have scaled raw fluorescence values from 0 to 1. Next, click on ‘‘download_result file’’ and a new window will open. Save the data in plain text format in the ‘‘Raw data’’ folder. Close the new window. 8. On the main page, select ‘‘Cluster Analysis.’’ On the ‘‘upload data’’ interface, browse to the raw data folder, highlight the file you have just saved, and click on ‘‘Submit.’’
372
Zhang, Smith, and Olivier
9. The concatenated data are presented in a single chart. Each cluster will have been assigned a color-coded center point. The coordinates of the point can be altered manually by selecting the cluster (in the box at the top) followed by clicking on the position where the new center will be located. Continue by clicking on ‘‘Start Clustering.’’ 10. On a new page, change the number for the confidence level to 99.5, to get a 99.5% confidence call of clustered data points. Also on this page, data points with indeterminate cluster assignments can be removed manually by selecting the data point and clicking on ‘‘Remove this point.’’ Once this has been completed, click on ‘‘Perform Cluster.’’ This will assign clusters to all data points with a confidence of 99.5%. 11. Click on ‘‘Save Data.’’ In a new window, a sheet with the signal scores that will be used to assign genotypes is shown. This file should then be saved in the ‘‘Scores’’ folder in text format. Close the new window. 12. Lastly, click on ‘‘Save Plot.’’ Save the cluster plot in .jpg format in the ‘‘Plots’’ folder for future reference (see Note 11). 3.3.3. Preparation of Genotyping Data for Association Analysis
1. Clustered genotype data for analysis can be processed using Microsoft Excel. To begin, open the ‘‘score’’ file of the SNP of interest in Excel. A window will pop up with warnings; choose ‘‘Finish.’’ Now a spreadsheet will appear with sample ID in the first column and the cluster ID (1, 2, 3, 4, or 8) in the fourth column. These two columns are all that are required for data processing. So, delete all other information in the sheet and save the file with a different filename. 2. If multiple SNPs are to be analyzed for the same project, clean them up as described above and combine them in the same spreadsheet. Use the first column for sample IDs, the second for SNP1, the third for SNP2, and so on. 3. The cluster ID numbers can be translated to genotype groups as follows: 1 – water controls (or failed reactions); 2 – homozygotes for red-dye-inferred allele 1; 3 – homozygotes for fluorescein-inferred allele 2; 4 – heterozygotes; 8 – failed or removed calls. 4. It is important to accurately assign the nucleotide allele to its representing dye. This is can be inferred by comparing the designed oligo sequences between the two probes.
Putting the Invader Assay to Work
373
Fig. 22.3. An example of the sample tracking worksheet.
5. Once the interrogated SNP alleles have been determined, cluster IDs can be converted to the genotypes. This can be processed on the same clustering data spreadsheet. A sample layout for ease of data handling is given in Fig. 22.3. Once the genotypes have been assigned, this file can be used for computational analyses.
4. Notes 1. In addition to processing samples as we have described, it is also possible to genotype up to four SNPs on 96 samples in one 384-well plate. The same set of (96) samples can be aliquoted into each quadrant of the plate (Fig. 22.2b). At the PCR stage, make up four separate master mixes (add this to the master mix table) and ensure that each is pipetted into one quadrant. Run the PCR as normal. Equally, when making the genotyping master mix (again add to the table), aliquot the genotyping mix to the corresponding PCR quadrant. Again, run the reaction as normal. For the analysis, skip the ‘‘concatenate data’’ step and instead cluster each of the four separate files individually, providing of course that the genotyping has been successful (ensure that there are sufficient, i.e., four, NTC samples per plate for the analysis to run successfully).
374
Zhang, Smith, and Olivier
a
b
c
d
e
Fig. 22.4. SNP genotyping results. (a) An example of a successfully clustered genotype result. x-axis, FAM fluorescence resulting from excitation at 485 nm and measured at 530 nm; y-axis, Texas red fluorescence
Putting the Invader Assay to Work
375
2. Sometimes very low design scores are generated owing to sequence features in the SNP region. We aim to use assays with a score above 50, but sometimes this is not possible, and we have had success with assays with much lower scores than this. 3. The PCR setup and cycling programs as shown in Tables 22.1 to 22.3 are considered the ‘‘standard’’ setup that can be used for the first-time amplification of the region of interest. However, for certain amplicons, adjustments for some of the parameters are required. The most frequently adjusted variables are the magnesium concentration, the annealing temperature, and the number of cycles in the amplification program. 4. It is extremely important for people not familiar with the 384well system to know that evaporation can be a big problem. Therefore, always remember to seal the plates tightly with the self-adhesive aluminum foil seals by pressing the rubber roller as firmly as possible. It is also recommended to seal the four edges of the plate with a thumb/fingernail wrapped in a Kimwipes wiper (Kimberley-Clark Global Sales, Roswell, GA, USA). Samples in the wells on the edge (particularly at the corners) are the most prone to evaporation. 5. The same PCR product can be used for multiple SNPs; it is not necessary to perform a separate PCR for each of them. Up to three SNPs have been successfully genotyped using the same PCR product. 6. PCR plates can be stored at 4C for at least 2 weeks without noticeable changes in quality. Genotyping plates can be wrapped in foil and stored at 4C for 1 week. For longer-term storage, PCR plates should be wrapped tightly in Parafilm (Pechiney Plastic Packaging, Menasha, WI, USA) and stored at –20C. It is of no value to
Fig. 22.4. (continued) resulting from excitation at 580 nm and measured at 630 nm. Each dot on the plot represents an individual sample on one of the 384-well plates. A successful cluster plot of a confirmed SNP genotyping result should show four clusters each containing (1) water controls and all failed samples (the one closest to the x–y intersect), (2) samples that are homozygous for one allele of the SNP (the one in gray with high x-axis scores), (3) samples that are homozygous for the other allele of the SNP (the one in black with high y-axis scores), and (4) heterozygous samples [the cluster between (2) and (3)] (b) A successful clustering result with a few data points with uncertain genotype assignment. Four such data points have been removed from the final score list (they are depicted in light gray ). (c) An Invader assay that worked but needed longer incubation time. (d) An assay in which only one allele can be typed. Note the low values on the y-axis. This may not be a SNP, or may be a very rare minor allele that cannot be detected in the samples analyzed. The assay only worked on one of the alleles could be another possibility. (e) Neither allele was successfully typed. A failed PCR may be the reason, or both assays may have failed to work.
376
Zhang, Smith, and Olivier
store the Invader reaction plates long term as the dye fluorescence will fade; rather we recommend repeating the genotyping if necessary. 7. The recommended reaction time for the Invader assay is 15 min. Depending on the quality of the DNA samples and the intensity of the perceived fluorescence signals, this standard protocol can be modified. The incubation time can be shortened to 5 min or lengthened up to 60 min. 8. The clustering algorithm developed can also be used with data from other genotyping platforms to permit highthroughput analyses (10). 9. No matter how perfectly the experiments have been conducted, failed Invader assays do occur. Several examples, including a typical successful assay and assays with identified or unknown causes, are presented in Fig. 22.4. 10. It is possible to concatenate more than the four plates at the concatenate data stage; however, we do not recommend this as inherent experimental variation (PCR/Invader efficiency, use of different master mixes) can lead to poor results at the clustering stage. 11. To facilitate tracking the progress of genotyping more than one SNP, use of a work progress spreadsheet is helpful. The format that is used in our laboratory is shown in Fig. 22.3. Check the cell once each step has been completed. This is most helpful and important when more than one person is conducting the project.
Acknowledgements The authors are thankful for the technical support from Regina Cole, Milena Zelembaba, and Brian Gau. This work was supported by NIH/NHLBI grant HL74168 to M.O. References 1. Pennacchio, L. A.,Olivier, M., Hubacek, J. A., Cohen, J. C., Cox, D. R., Fruchart, J. C., Krauss, R. M. and Rubin, E. M. (2001) An apolipoprotein influencing triglycerides in humans and mice revealed by comparative sequencing. Science 294, 169–173. 2. Olivier, M., Chuang, L. M., Chang, M. S., Chen, Y. T., Pei, D., Ranade, K., de Witte, A., Allen, J., Tran, N., Curb, D., Pratt, R.,
Neefs, H., de Arruda Indig, M., Law, S., Neri, B., Wang, L. and Cox, D. R. (2002). High-throughput genotyping of single nucleotide polymorphisms using new Invader technology. Nucleic Acids Res. 30, e53. 3. Pennacchio, L. A., Olivier, M., Hubacek, J. A., Krauss, R. M., Rubin, E. M. and Cohen, J. C. (2002) Two independent apolipoprotein A5 haplotypes influence human plasma
Putting the Invader Assay to Work triglyceride levels. Hum. Mol. Genet. 11, 3031–3038. 4. Eichenbaum-Voline, S., Olivier, M., Jones, E. L., Naoumova, R. P., Jones, B., Gau, B., Patel, H. N., Seed, M., Betteridge, D. J., Galton, D. J., Rubin, E. M., Scott, J., Shoulders, C. C. and Pennacchio, L. A. (2004) Linkage and association between distinct variants of the APOA1/C3/A4/A5 gene cluster and familial combined hyperlipidemia. Arterioscl. Thromb. Vasc. Biol. 24, 167–174. 5. Olivier, M., Wang, X., Cole, R., Gau, B., Kim, J., Rubin, E. M. and Pennacchio, L. A. (2004) Haplotype analysis of the apolipoprotein gene cluster on human chromosome 11. Genomics 83, 912–923. 6. Olivier, M., Hsiung, C. A., Chuang, L. M., Ho, L. T., Ting, C. T., Bustos, V. I., Lee, T. M., de Witte, A., Chen, I., Rodriguez, B., Wen, C. C. and Cox, D. R. (2004) Single nucleotide polymorphisms in protein tyrosine phosphatase 1B (PTPN1) are associated with essential hypertension and obesity in Japanese and Chinese. Hum. Mol. Genet. 13, 1885–1892.
377
7. Baessler, A., Hasinoff, M., Fischer, M., Reinhard, W., Sonnenberg, G., Olivier, M., Erdmann, J., Schunkert, H., Doering, A., Jacob, H. J., Comuzzie, A. G., Kissebah, A. H. and Kwitek, A. E. (2005) Genetic linkage and association of the growth hormone sectretagogue receptor (GHSR, Ghrelin receptor) gene in human obesity. Diabetes 54, 259–267. 8. Olivier, M. (2005) The Invader assay for SNP genotyping. Mut. Res. 573, 103–110. 9. Hall, J. G., Eis, P. S., Law, S. M., Reynaldo, L. P., Prudent, J. R., Marshall, D. J., Allawi, H. T., Mast, A. L., Dahlberg, J. E., Kwiatkowski, R. W., de Arruda, M., Neri, B.P. and Lyamichev, V. I. (2000) Sensitive detection of DNA polymorphism by the serial invasive signal amplification reaction. Proc. Natl. Acad. Sci. U.S.A. 97, 8272–8277. 10. Smith, E. M., Littrell, J. and Olivier, M. (2007) Automated SNP genotype clustering algorithm to improve data completeness in high-throughput SNP genotyping datasets from custom arrays. Genomics Proteomics Bioinformatics 5, 256–259.
Chapter 23 SNP Genotyping Using Multiplex Single Base Primer Extension Assays Daniele Podini and Peter M. Vallone Abstract Single nucleotide polymorphisms (SNPs) are the most common form of polymorphisms present in the human genome. The single base primer extension (SBE) method is an effective and sensitive tool that can type over 30 known loci scattered throughout an organism’s genome in a single reaction. It allows the typing of tetra-allelic SNPs and has been adapted to a broad range of analytical necessities: single-cell analysis, molecular diagnosis of monogenic diseases, forensic mitochondrial DNA analysis on highly degraded human remains, and high-throughput SNP screening for population studies. Every SBE-based assay will need customized optimization efforts that are generally proportional to the number of desired SNPs typed in a single reaction. This chapter offers a detailed outline on which to base the design and optimization of any multiplex SBE assay that can then be tailored to the analytical conditions that characterize each specific application. Key words: Single nucleotide polymorphism, multiplex PCR, capillary electrophoresis, SNaPshot1, single base extension.
1. Introduction The single base primer extension (SBE) technique, also known as minisequencing (1–3), allows for the simultaneous typing from one to over 30 single nucleotide polymorphisms (SNPs) scattered throughout the organism’s genome (4). The advantages of this method include the possibility of typing tetra-allelic SNPs, sensitivity, specificity, robustness, and amenability to automation (5). Once the assay has been optimized, it generally allows one to obtain robust results with a broad range of both quantity and quality of genomic DNA template. SBE has been applied in several A.A. Komar (ed.), Single Nucleotide Polymorphisms, Methods in Molecular Biology 578, DOI 10.1007/978-1-60327-411-1_23, ª Humana Press, a part of Springer Science+Business Media, LLC 2003, 2009
379
380
Podini and Vallone
different applications: single-cell analysis for preimplantation genetic diagnosis, prenatal and postnatal molecular diagnosis of monogenic diseases (6), forensic mitochondrial DNA analysis on highly degraded human remains, and high-throughput SNP screening for population studies (7–9). The SBE method is based on an initial multiplex PCR amplification of fragments that can be small (about 50 base pairs) as long as the targeted SNP is included in the amplicon (i.e., in-between the primer binding sites but not included in the primer sequence). Generally, the smaller the amplified fragment, the greater the amplification efficiency; this is particularly relevant in the situation where the starting template is available at very low copy numbers/ concentrations and/or the template is highly degraded (7). After the PCR multiplex amplification has been performed, the reaction product is purified to eliminate unincorporated PCR primers and dNTPs. The SBE reaction then uses the purified PCR product as a template. SBE primers are designed similarly to a standard sequencing primer. The SBE primer binds in a 50 !30 orientation to the PCR amplicon with the 30 end of the primer adjacent to the SNP of interest. The second sequence-specific annealing step adds further specificity to the assay. The SNaPshot1 reagent kit contains buffer, polymerase, and fluorescently labeled dideoxynucleotides (ddNTP) (one dye for each nucleotide). During thermal cycling the SBE primer binds to the PCR amplicon and the appropriate ddNTP is incorporated at the SNP site (Fig. 23.1). Following the SBE reaction, samples are further purified to eliminate unincorporated labeled ddNTPs that would interfere with data analysis, and are loaded onto a capillary electrophoresis (CE) instrument. Electropherograms can then be analyzed with commercially available programs. Dedicated macros can be created to facilitate data interpretation, processing, and management. A flowchart of the SBE SNP typing method is shown in Fig. 23.2. Multiplexing of a SBE assay is accomplished by adding a nonbinding tail sequence to the 50 end of the SBE primer. The tail is typically a poly-T or a repeating AGCT sequence. The total length of SBE primers can range from 20 to 80 nucleotides (the length is somewhat defined by limitations in the automated synthesis of DNA oligomers). Each SBE primer is usually separated in size by three to four nucleotides to ensure resolution on a gel or capillary detection platform. Every SBE-based assay will need customized optimization owing to a number of factors, including differences in template, flanking sequences of the target sites, number of SNPs, and the target (i.e., haploid or diploid). Generally, the greater the number of desired SNPs typed in a single reaction, the greater the necessary optimization efforts. It should be noted that developers of SBE assays are responsible for the design of PCR and SBE primers that must be compatible in a multiplex format.
SNP Genotyping Using Multiplex Single Base Primer Extension Assays
381
Multiplex PCR T TT
(1) EXO-SAP Purification
TT
TT TT T TT TT TT TT TT TT TT
T TT
T TT
Non-binding tail
5’ 3’
TTT
3’
5’
TT
SBE Primer
TT
TT TT
TT
TT
T-
T TT
SBE multiplex reaction
ddGTP ddCTP ddTTP ddATP
5’
-
TT
5’
T TT
3’
---
CGGAATCAGGTCAATCAC T TGCCTTAGTCCAGTTAGTGATCAGT
3’
T TT
3’
----
5’
-
3’
5’ PCR template
Target SNP
-T
Smallest
T TT TT
TT
-
TT TT
T TT
TT TT
TT
Homozygote
Heterozygote
TT
TT
-
TT
-
-T
TT TT TT TT TT TT TT TT TT TT TT TT
(1) SAP Purification (2) Capillary Electrophoresis
Largest
TT
AA
CC
RFU
GA
Length in nucleotides
Fig. 23.1. The single base primer extension (SBE) assay. Initial multiplex PCR amplification is performed targeting the flanking regions of the single nucleotide polymorphisms (SNPs). Following amplification, samples are purified to eliminate unincorporated PCR primers and dNTPs. The SBE reaction then uses the purified PCR product as a template. SBE primers bind in a 50 !30 orientation to the corresponding PCR amplicon with the 30 end of the primer adjacent to the SNP of interest and the appropriate ddNTP is incorporated at the SNP site. Following the SBE reaction, samples are purified to eliminate unincorporated labeled ddNTPs and then loaded on a capillary electrophoresis apparatus. In this example, the targets are four diploid loci, of which the first three (left to right) are homozygous and the forth one (the largest) is a heterozygote (G/A). Note that the migration of the SBE primers is affected by the specific dye attached by the incorporated nucleotide. The two alleles, although having the same number of bases, exhibit different electrophoretic mobility and appear as two separate peaks. Single Base Primer Extension Assay Workflow Amplification
Primer Extension
Analysis
Genomic DNA sample
Add SBE primers and SNaPshot® mix
Sample prep for CE analysis
Fig. 23.2. Workflow diagram of the SBE assay.
Multiplex PCR
ExoI-SAP Digestion
Multiplex SBE (cycle sequencing)
SAP treatment
Run on CE Platform
Data Analysis Genotyping Macros
382
Podini and Vallone
2. Materials 2.1. Primers and Primer Design
1. There are several free software tools for multiplex PCR primer design, including, but not limited to, those at http://frodo. wi.mit.edu/ (Primer 3) and http://www.idtdna.com. SBE primers can be selected with sequencing primer design tools and then tested for unwanted primer–dimer interactions to evaluate potential performance in a multiplex assay. A useful multiplex primer screening tool (AutoDimer) for this purpose can be found at http://yellow.nist.gov:8444/dnaAnalysis/index.do. 2. Standard desalting purification is sufficient for primers that are shorter than 50 nucleotides, but polyacrylamide gel electrophoresis purification should be performed for longerlength oligos. Most vendors provide lyophilized products. Prepare 100 mM stock solutions by adding the appropriate volumes of deionized water. Owing to potential discrepancies between the amount of oligo reported by the manufacturer and the actual amount measured, it is recommended that the final concentration of each extension primer be determined by UV spectroscopy. This is particularly important when reporting concentrations for ‘‘balanced’’ multiplex PCR assays, SBE multiplexes, or otherwise. The quality of the oligos contributes greatly to the success of the assay and facilitates optimization; thus, it is recommended to choose a reputable and referenced manufacturer for primer synthesis.
2.2. PCR and Purification of the Products
Several Taq polymerases, buffers, and dNTPs are commercially available for developing a multiplex PCR. The objective at this stage is to efficiently amplify all the fragments containing the targeted SNPs. Here we present a generalized method based on previous experiences in assay design for developing a multiplex amplification assay with the understanding that alternative strategies might be just as efficient. Examples of successful multiplex SBE assays have been detailed in the literature (4, 6–13) 1. AmpliTaq Gold1 DNA polymerase (5 U/mL) (Applied Biosystems). 2. GeneAmp1 PCR buffer II 10X (Applied Biosystems). 3. 25 mM MgCl2 (Applied Biosystems). 4. 100 mM dNTPs (Roche Diagnostic). 5. PCR-grade distilled H2O (dH2O). 6. Agarose gel electrophoresis capability. 7. Exonuclease I (10 U/mL) (USB). 8. Shrimp alkaline phosphatase (SAP) (1 U/mL) (Roche Diagnostic).
SNP Genotyping Using Multiplex Single Base Primer Extension Assays
383
2.3. Extension Reaction Reagents
1. SNaPshot1 (Applied Biosystems). Sold as 2x master mix.
2.4. Capillary Electrophoresis
1. A 3130xl genetic analyzer (Applied Biosystems). Alternative CE platforms capable of detecting the dyes present in the SNaPshot1 kit can be utilized (Applied Biosystems 3100, 310, 3130, 3730, etc.). 2. Genescan-120 LIZ size standard (Applied Biosystems). 3. Hi-Di formamide (Applied Biosystems).
2.5. Data Analysis
1. Fragment analysis software: Genemapper1 (Applied Biosystems) or GeneMarkerTM (Softgenetics1).
3. Methods As previously mentioned, each multiplex assay requires customized optimization. The first step is to design PCR and extension primers, then the performance of PCR and extension reactions for each SNP is individually tested to determine the reaction efficiency and electrophoretic mobility of each SBE primer. Multiplex optimization can begin once these aspects have been determined and potential issues have been addressed. 3.1. Preliminary Testing 3.1.1. Primer Design
1. Multiplex PCR primers should be designed to perform efficiently in the amplification reaction. Design parameters to consider include avoiding primer–dimers and aiming for similar annealing temperatures (55–60C) and percentage of GC content amongst primers and amplicons of a similar size. Generally, smaller fragments have higher amplification efficiencies (see Note 1). PCR primers can be designed to anneal very close to the target SNP, even one base short, but must not include it (see Note 2). 2. SBE primers target the PCR template and lie adjacent (50 ! 30 ) to the SNP site. Again, the initial design should avoid primer– dimers and intramolecular hairpins, and should aim for similar annealing temperatures and percentage of GC content. Although a forward and reverse SBE can be designed, only one SBE primer is required. It is often useful to design both candidates and then select the one that works best in the multiplex assay (based on avoiding unwanted primer–primer interactions). Once the extension primers have been designed, poly-T tails or repeating AGTC tails are added to the 50 end to alter the electrophoretic mobility of the SBE primers. This allows for the spatial differentiation on the electrophoretic detection platform. Generally, at least three or four bases separating each SBE primer are sufficient, although the electrophoretic mobility of such small
384
Podini and Vallone
fragments is not only determined by primer length but also by the sequence and specific dye attached to the incorporated base (see Note 3). Adjustments (i.e., addition or removal of bases from the 50 tail end of extension primers or redesigning a primer targeting the opposite strand) may be necessary during optimization of the assay for better separation and to facilitate data interpretation. Unfortunately, this adjustment is determined empirically and requires the synthesis of a new primer. 3.1.2. PCR
1. Singleplex conditions: 10 pmol of forward and reverse primer (0.2 mM), 5 mL 10X buffer II, 3 mL 25 mM MgCl2 (final concentration 1.5 mM), 0.1 mL of each dNTP (0.2 mM), 0.5 mL AmpliTaq Gold (2.5 U), 0.5–2 ng (see Note 4) template DNA in a final volume of 50 mL. Singleplex PCR thermal cycling profile: 95C for 10 min, 35 cycles of 94C for 30 s, 55C for 1 min, 72C for 30 s, 72C for 10 min, 4C storage (see Note 5). 2. Load the PCR amplification products on a 2% agarose gel (or comparable detection system) to verify successful amplification and the correct size of the amplicons (see Note 6). All samples should yield strong bands of the expected size. Primers that yield no or very weak PCR products for specific SNPs will, most likely, not perform well in multiplex; thus, new primers should be designed for these target sites. 3. Initial multiplex conditions: 10 pmol of each forward and reverse primer (0.2 mM), 5 mL 10X buffer II, 3 mL 25 mM MgCl2 (final concentration 1.5 mM), 0.1 mL of each dNTP (0.2 mM), 0.5 mL AmpliTaq Gold (2.5 U), 0.5–2 ng (see Note 4) of template DNA in a final volume of 50 mL. The multiplex PCR thermal cycling profile is as follows: 94C for 11 min, three cycles of 95C for 30 s, 50C for 55 s, and 72C for 30 s, then 19 cycles of 95C for 30 s, 50C for 55 s with an increase of 0.2C per cycle, and 72C for 30 s, and finally 11 cycles of 95C for 30 s, 55C for 55 s, and 72C for 30 s. A final extension at 72C for 7 min is followed by a storage step at 4C (see Note 7). 4. Purification: Add 1 mL of Exo I (10 U) and 1 mL of SAP (1U) to 5 mL of PCR product (see Note 8). Vortex and incubate the mixture for 70 min at 37C and 20 min at 72C for enzyme deactivation (see Note 9).
3.1.3. Single Base Primer Extension Reaction
Before performing a multiplex extension reaction, verify that all extension primers are working efficiently by executing a singleplex extension reaction with the corresponding template. A SBE primer can self-extend (by folding upon itself during the annealing step); thus, SBE primers need to be tested also in a singleplex blank
SNP Genotyping Using Multiplex Single Base Primer Extension Assays
385
reaction (replacing purified PCR product with dH2O). Once determined that all primers perform efficiently proceed with the multiplex reaction and its optimization. 1. Singleplex extension reaction: using individual tubes for each SNP, add 1.5 pmol (0.15 mM) of the single extension primer, 1 L SNaPshot1 reaction mix, 1 L of the purified PCR product encompassing the corresponding SNP, and dH2O to a 10 L final volume (see Note 10). The SBE cycling profile is as follows: 25 cycles of 95C for10 s, 50C for 5 s, and 60C for 30 s, and final storage at 4C. 2. Initial multiplex extension conditions: 1.5 pmol (0.15 mM) of each extension primer, 2.5 mL SNaPshot1 reaction mix, 1 mL of the purified multiplex PCR product, and dH2O to bring the final volume to 10 mL; same cycling profile as in step 1 (see Note 11). 3. Extension reaction purification: add 1 mL of SAP (1 U) to each reaction tube and incubate the mixture for 70 min at 37C, and 20 min at 72C. 3.1.4. Capillary Electrophoresis
1. Single sample preparation: add 1 mL of each purified extension product to 11.75 mL of Hi-Di formamide (Applied Biosystems) and 0.25 mL of Genescan-120 LIZ size standard (Applied Biosystems) in individual wells or follow the manufacturer’s instructions. 2. The run parameters depend on the CE platform available. Follow the manufacturer’s instructions taking into account instrumentation, type of polymer, and capillary length (see Note 12).
3.1.5. Analysis
1. Analyze the electropherograms using fragment analysis software such as Genemapper1 (Applied Biosystems) or GeneMarkerTM (Softgenetics1), following the manufacturer’s instructions. 2. Each extension reaction should yield at least a single peak (homozygote or haploid marker) or two peaks (heterozygote) (see Note 13). Extension primers that yield no peaks (except for blank reactions which should show none) or peaks with a relative fluorescence unit (RFU) value below 500 in singleplex may not perform well in multiplex. Try increasing the SBE primer concentration and if no significant improvement in signal is observed, new extension primers should be designed for these target sites. Selfextending SBE primers resulting in a peak in the blank reaction may have to be redesigned. If all SBE reactions are successful proceed with steps 2 and 3 in Section 3.1.3 (see Note 14).
386
Podini and Vallone
3.2. Optimization
Optimization is performed at four levels: multiplex PCR, multiplex SBE reaction, confirming the electrophoretic mobility of extension primers, and data analysis.
3.2.1. Multiplex PCR Optimization
To evaluate efficiency of the multiplex PCR amplification, perform individual SBE reactions using purified multiplex PCR product as the template: 1. Mix 1.5 pmol (1.5 mM) of the extension primer(s) (test each SNP individually), 1 mL SNaPshot1 reaction mix, 1 mL of the purified multiplex PCR product, and add dH2O to bring the total volume to 10 mL. Perform the same SBE thermal cycling profile as in step 1 in Section 3.1.3. 2. Purify the SBE products and perform CE as previously described. The absence of one or more peaks will be caused by the absence of the corresponding PCR template. If this is the case, the multiplex PCR conditions can be varied. This includes increasing MgCl2 concentration, increasing PCR primer concentration(s) of the fragments that failed to be amplified, and lowering the annealing temperatures. Vary only one experimental parameter at a time and evaluate the results. The final optimization may require the modification of more than one parameter. The final remedy will be the redesign of the failing PCR primer pairs. Once all SBE fragments have individually yielded peaks higher than 500 RFU using purified multiplex PCR product as the template, proceed to optimizing the multiplex extension reaction.
3.2.2. Multiplex Extension Reaction Optimization
1. The concentration of extension primers yielding low signal intensities can be increased, while, conversely, the concentration of high-signal-producing SBEs can be reduced. The goal is to improve the overall signal balance within the SBE multiplex. If necessary, the concentration of the PCR primers amplifying the corresponding SNP can also be adjusted. To increase the overall sensitivity of the assay, increase the amount of SNaPshot1 reaction mix and/or the number of cycles of the reaction (can be increased up to 40 cycles). 2. It is expected that different alleles exhibit different peak heights owing to variations in the intensities of the fluorescent dyes attached to the different ddNTPs. When possible, SNPs should be selected with a C/T polymorphism owing to the even balance that corresponding dyes (red and black) exhibit when using the SNaPshot1 kit with CE detection. 3. Occasionally, artifacts can be present in the CE electropherogram. These may be caused by several factors, including poor SBE primer synthesis, interactions between different SBE primers, or SBE primer interaction with a nonspecific PCR template. These can be ignored if they do not interfere with
SNP Genotyping Using Multiplex Single Base Primer Extension Assays
387
the correct allele calling of the SNPs; otherwise serially remove each SBE primer from the multiplex to determine the cause of the artifact. A simple and efficient way to determine whether there are interactions between the SBE primers is to perform a multiplex SBE reaction without including PCR template (blank test). If any of the SBE primers are extended, this would result from another SBE primer acting as a template. In theory this should not happen if SBE primers were screened for primer–dimers, yet it cannot be excluded until a blank test has been executed (see Table 23.1 for troubleshooting summary).
Table 23.1 Troubleshooting Problem
Possible cause
Solution
Unexpected/ interfering peaks around 18–25
PCR primers were not fully degraded in the Exo I– SAP step
Increase Exo I in purification
Mobility of the fragment is off by a few nucleotides
Dye varying the ‘‘true’’ mobility
Add/remove nucleotides from the tail of the SBE primer to adjust
Nonspecific artifacts
Poor SBE synthesis SBE primer–primer interactions SBE primer interaction with nonspecific PCR template
Serially remove each SBE primer from the multiplex to determine which one is causing the artifact Possible to accept an artifact that does not interfere with allele call Perform blank reaction (see Section 3.2.2)
Weak overall signal
Inefficient and/or insufficient SNaPshot1 reagent
Check SNaPshot1 reagent Increase SNaPshot1 reagent
Exo I exonuclease I, SAP shrimp alkaline phosphatase, SBE single base primer extension
3.2.3. Optimization of the Electrophoretic Mobility of Extension Primers
1. Migration of the SBE primers should be optimized to avoid overlap between extension primers. This is done to facilitate data interpretation and SNP genotyping. Optimization involves the reordering and resynthesis of SBE primers with longer or shorter non-template-binding 50 tails (depending on what is required). 2. It is also recommended to type a representative sample cohort to test samples that exhibit all alleles and verify their electrophoretic mobility when optimizing the assay. The overlapping of two SBE fragments that can only incorporate different bases will not obscure interpretation (see Note 15).
388
Podini and Vallone
3.2.4. Analysis Optimization
1
Once the assay performs consistently, create custom panels in the fragment analysis software that will enable automated SNP genotyping. This will facilitate data interpretation and management and increase the throughput of the assay (Fig. 23.3).
3 2
Fig. 23.3. Example of an electropherogram of a 16-plex human mitochondrial DNA coding region assay for haplogroup screening. A customized panel was developed to facilitate data analysis. The mitochondrial genome is haploid so only one allele per SNP is expected. Note that (1) primers targeting SNPs 13708 (C) and 12372 (A) have similar mobility and both target a G/A polymorphisms yet correct genotyping is not compromised because the 13708 primer was designed in the reverse reading frame and will thus incorporate either a C or a T, whereas the 12372 will incorporate either an A or a G; (2) the presence of minor artifacts that do not interfere with the correct allele call; (3) imbalance between peaks (further optimization of the assay is possible).
4. Notes 1. If the assay is being developed to target SNPs on degraded DNA extracts, it is paramount to minimize PCR amplicon size to maximize amplification yield. Typically 50–75 base pair PCR amplicons are beneficial for degraded SNP typing. 2. If the PCR primer includes the probed SNP site, this will result in the downstream extension assay targeting the primer sequence and not the actual sample. This will compromise accurate SNP genotyping.
SNP Genotyping Using Multiplex Single Base Primer Extension Assays
389
3. Once the SBE primers have been designed, the fragment migration is optimized by adding a 50 tail to the binding region of the primer. This tail can be a poly-T or a repeating AGTC sequence on the 50 end of the primer. Sort the SBE primers on the basis of their length, do not add a 50 tail to the shortest, add to the second shortest the number of nucleotides necessary to make it four bases longer than the shortest one, then to the third shortest add the number of nucleotides to make it four bases longer than the previous one, and so on. 4. This is a sensitive and robust method and can be used for diagnostic purposes on clinical samples (blood, tissue, etc.), for singlecell analysis (preimplantation genetic diagnosis), and for forensic applications on highly degraded human remains. However, if large amounts of pristine starting template are available (more than 3 ng), the downstream optimization of the assay is less challenging. 5. The annealing temperature (Ta) for the PCR should be selected on the basis of the average calculated melting temperature (Tm) of all PCR primers. 6. An advantage of this system is that owing to the added specificity of the SBE primer, nonspecific PCR products can be tolerated since the SBE primer only incorporates a ddNTP after binding to the correct template sequence. 7. This PCR thermal cycling profile is based on the incremental increase of temperature between the annealing and extension conditions. This allows the specific binding of the different sets of primers in the multiplex, as both primers in each set have nearly identical Tm but do not necessarily have the same Tm as the other sets of primers in the multiplex. The final temperature indicated here is 55C, but should be based on the average annealing temperature of all PCR primers. 8. A mixture of these two enzymes can be prepared and added to tubes prior to addition to the PCR product. If necessary, volumes can be scaled up, maintaining the same volume to enzyme units ratio. Another commercially available product commonly used in place of separately adding the two enzymes is ExoSap-IT (USB). 9. It is essential that all PCR primers are degraded after PCR as they can act as SBE primers, visualized as peaks migrating in the range of 18–30 nucleotides (depending on their size), potentially obscuring peaks from true SBE primers. Larger multiplexes may require extra exonuclease I and/or longer incubation times to ensure the degradation of all PCR primers. 10. The manufacturer recommends 5 mL of SNaPshot1 reaction mix, but 1 mL for singleplex reactions and 2.5 mL for multiplex reactions are generally sufficient. 10X PCR buffer II can be added to compensate for the reduction of the recommended salt concentration in the reaction.
390
Podini and Vallone
11. This reaction will be subject to the modifications addressed in Section 3.2. 12. Examples: Applied Biosystems 3130 genetic analyzer with POP 7 as a sieving polymer and a 36-cm array uses SNaPshot_pop7 default run module and E5 dye set; ABI Prism1 310 genetic analyzer with POP 4 polymer and a 47-cm capillary uses GS STR POP4 (1 mL) E5 run module, 5-s injection time, 15-kV injection and run voltage, 60C run temperature, and a 15-min run time. POP 6 can also be used by varying parameters accordingly. 13. SBE extension primers that have the same length and sequence and differ only by the incorporated base usually do not produce peaks that overlap, but usually appear as two separate peaks. This is due to the electrophoretic mobility of the dye attached to the incorporated nucleotide. 14. Note the mobility for each of the SBE primers and verify the peak separation that would be obtained when the run is in multiplex. Optimization of the electrophoretic mobility of SBE primers can begin at this stage (see Section 3.2.3). If the CE apparatus is an ABI Prism1 310 genetic analyzer (a singlecapillary instrument) pooling of the singleplex extension products can help speed up the optimization. In this case add 0.5 mL of each purified extension product to 11.75 mL of Hi-Di formamide (Applied Biosystems) in a single tube/well and 0.25 mL of Genescan-120 LIZ size standard (Applied Biosystems). 15. For example, a SBE primer designed in the forward strand orientation targeting a A/G SNP (incorporation of green/ blue dyes) can overlap with an extension primer designed on the forward strand targeting a T/C SNP (incorporation of red/yellow dyes). Or an extension primer designed on the forward strand targeting an A/G SNP (incorporation of green/blue dyes) can overlap with one designed on the reverse strand for an A/G SNP (incorporation of red/yellow dyes) because they will incorporate different bases, thereby generating distinguishable peaks.
Acknowledgements The authors would like to thank Melinda Jean Hung for her valuable work on the design and optimization of the human mtDNA haplogroup SBE typing assay shown in Fig. 23.3 and Becky Hill and Amy Decker for their valuable comments. Disclaimer: Certain commercial equipment, instruments, and materials are identified to specify experimental procedures as completely as possible. In no case does such identification imply a
SNP Genotyping Using Multiplex Single Base Primer Extension Assays
391
recommendation or endorsement by the National Institute of Standards and Technology nor does it imply that any of the materials, instruments, or equipment identified are necessarily the best available for the purpose. References 1. Sokolov, B. P. (1990) Primer extension technique for the detection of single nucleotide in genomic DNA. Nucleic Acids Res. 18, 3671. 2. Pastinen, T., Kurg, A., Metspalu, A., Peltonen, L. and Syvaenen, A. C. (1997) Minisequencing: a specific tool for DNA analysis and diagnosis on oligonucleotide arrays. Genome Res. 7, 606–614. 3. Syvanen, A. C. (1999) From gels to chips: ‘‘minisequencing’’ primer extension for analysis of point mutations and single nucleotide polymorphism. Hum. Mutat. 13, 1–10. 4. Phillips, C., Salas, A., Sanchez, J. J., Fondevila, M., Gomez-Tato, A., Alvarez-Dios, J., Calaza, M., Casares de Cal, M., Ballard, D., Lareu, M. V. and Carracedo, A. (2007) Inferring ancestral origin using a single multiplex assay of ancestry-informative marker SNPs. Forensic Sci. Int.: Genetics 1, 273–280. 5. Fiorentino, F., Magli, M. C., Podini, D., Ferraretti, A. P., Nuccitelli, A., Vitale, N., Baldi, M. and Gianaroli, L. (2003) The minisequencing method: an alternative strategy for preimplantation genetic diagnosis for single gene disorders. Mol. Hum. Reprod. 9, 399–410. 6. Fiorentino, F., Biricik, A., Karadayi, H., Berkil, H., Karlkaya, G., Sertyel, S., Nuccitelli, A., Podini, D., Baldi, M., Magli, M. C., Gianaroli, L. and Kahraman, S. (2004) Development and clinical application of a strategy for PGD of single gene disorders combined with HLA matching. Mol. Hum. Reprod. 10, 445–460. 7. Vallone, P. M., Just, R. S., Coble, M. D., Butler, J. M. and Parsons, T. J. (2004) A multiplex allele-specific primer extension assay for forensically informative SNPs distributed throughout the mitochondrial genome. Int. J. Leg. Med. 118, 147–157.
8. Nelson, T. M., Just, R. S., Loreille, O., Schanfield, M. S. and Podini, D. (2007) Development of a multiplex single base extension assay for mtDNA haplogroup typing. Croat. Med. J. 48 460–472. 9. Vallone, P. M. and Butler, J. M. (2004) YSNP typing of U.S. African American and Caucasian samples using allele-specific hybridization and primer extension. J. Forensic. Sci. 49 723–732. 10. Sanchez, J. J., Borsting, C., Balogh, K., Berger, B., Bogus, M., Butler, J. M., Carracedo, A., Syndercombe Court, D., Dixon, L. A., Filipovic, B., Fondevila, M., Gill, P., Harrison, C. D., Hohoff, C., Huel, R., Ludes, B., Parson, W., Parsons, T. J., Petkovski, E., Phillips, C., Schmitter, H., Schneider, P. M., Vallone, P. M. and Morling, N. (2008) Forensic typing of autosomal SNPs with a 29 SNP-multiplex— results of a collaborative EDNAP exercise. Forensic Sci. Int.: Genetics 2, 176–183. 11. Sanchez, J. J., Phillips, C., Borsting, C., Balogh, K., Bogus, M., Fondevila, M., Harrison, C. D., Musgrave-Brown, E., Salas, A., Syndercombe-Court, D., Schneider, P. M., Carracedo, A. and Morling, N. (2006) A multiplex assay with 52 single nucleotide polymorphisms for human identification. Electrophoresis 27, 1713–1724. 12. Sanchez, J. J., Borsting, C., Hallenberg, C., Buchard, A., Hernandez, A. and Morling, N. (2003) Multiplex PCR and minisequencing of SNPs: a model with 35 Y chromosome SNPs. Forensic Sci. Int. 137, 74–84. 13. Brandstatter, A., Parsons, T. J. and Parson, W. (2003) Rapid screening of mtDNA coding region SNPs for the identification of west European Caucasian haplogroups Int. J. Legal. Med. 117, 291–298.
Chapter 24 High-Throughput SNP Detection Based on PCR Amplification on Magnetic Nanoparticles Using Dual-Color Hybridization Nongyue He, Song Li, and Hongna Liu Abstract A microarray-based method for detecting single nucleotide polymorphisms (SNPs) using solid-phase polymerase chain reaction (PCR) on magnetic nanoparticles (MNPs) was developed. In this method one primer with a biotin label is captured by streptavidin-coated MNPs (SA-MNPs), and the PCR products are directly amplified on the surface of SA-MNPs in a 96-well plate. The samples are further probed by hybridization with a pair of dual-color probes to determine SNP. The genotype of each sample can be simultaneously identified by scanning the microarray printed with the denatured fluorescent probes. All the reactions can be performed in the same reaction volume without the necessity of purification of intermediates. This approach represents a novel, simple, and labor-saving method for SNP genotyping and can be applied in automation system(s) to achieve high-throughput SNP detection. Key words: Single nucleotide polymorphism, solid-phase PCR, magnetic nanoparticles, dual-color hybridization.
1. Introduction With the completion of the Human Genome Project, attention is now rapidly shifting towards the study of individual genetic variations. The most abundant source of genetic variation in the human genome is represented by single nucleotide polymorphisms (SNPs), which can account for heritable interindividual differences in complex phenotypes. SNPs promise great medical utility, since they give rise to individual gene variants that can alter susceptibility to common diseases and
A.A. Komar (ed.), Single Nucleotide Polymorphisms, Methods in Molecular Biology 578, DOI 10.1007/978-1-60327-411-1_24, ª Humana Press, a part of Springer Science+Business Media, LLC 2003, 2009
393
394
He, Li, and Liu
might be helpful in prediction disease progression as well as individual responses to drugs (1, 2). The identification of SNPs, their association with disease and other phenotypes, and a careful assessment of the mechanistic basis of their functional impact on these phenotypes will fulfill this promise, providing highly accurate diagnostic information that will facilitate early diagnosis, prevention, and treatment of human diseases. Over the past several years, the advancement of increasingly high throughput and cost effective methods to detect and discover SNPs has begun to open the door towards this endeavor. Microarray platforms have been widely used for highly parallel genomic analyses, owing to their high multiplex capabilities, low cost, and ability to provide highly parallel readout for large-scale samples. Over the past several years, multiplexing high-throughput methods based on microarrays to discover and measure SNPs have been developed and commercialized (3, 4). However, these technologies are usually time-consuming and not suitable for automatic operation, since they involve additional procedures for purification and concentration of targets in sample preparation. Therefore, these methods have limited utility in high-throughput polymorphism detection to meet the challenges of the new genomics era. Furthermore, a rapid, simple, high-throughput parallel screening protocol for SNP detection over thousands of samples is still required. Magnetic nanoparticles (MNPs) have already been successfully used in various fields of biology and medicine, such as magnetic targeting (of drugs, genes), magnetic resonance imaging, immunoassays, cell separation, and RNA and DNA purification (5–7). MNPs are recognized for their unique higher dispersion capability in aqueous solution, higher separation efficiency in a magnetic field, and easy operation in autoworkstations (5–7). Approaches and methods using MNPs as platforms for SNPs genotyping have been developed recently (8, 9). Herein, we present a method for multiplex SNP profiling in conjunction with microarrays, which involves solid-phase PCR amplification directly on MNPs and hybridization with allele-specific probes labeled with dual-color fluorescence (Cy3, Cy5) (8, 9). In this method, all steps of the preparation can be performed in the same vessel by simple additions of solution and incubation, and then genotypes are discriminated by scanning the microarray printed with the denatured fluorescent probes on an unmodified glass slide. Therefore, this method is particularly suitable for automation and represents a simple, labor-saving, and highly sensitive approach to SNP genotyping.
High-Throughput SNP Detection Based on PCR Amplification
395
2. Materials 1. Genomic DNA was extracted from 100 mL blood samples from patients using a QIAamp DNA blood mini kit (Qiagen). Written informed consent was obtained from all individuals.
2.1. DNA
2. 100 ng template DNA is required for a reaction. 1. SiO2/[poly(methyl methacrylate) (PMMA)/Fe3O4] NMPs about 100 nm in diameter are prepared according to a previously published method (10).
2.2. Reagents for Streptavidin-Coated MNP Preparation
2. Ethanol, molecular biology grade (Sigma-Aldrich). 3. The MNPs are dispersed in ethanol at a concentration of 4 mg/mL (see Note 1). 4. 3-Aminopropyl triethoxysilane (APTES) (Sigma-Aldrich). 5. 1x, 0.1 M PB buffer, pH 7.4. Prepare the stock of 0.1 M PB by mixing 77.4 mL of 1 M Na2HPO4 and 22.6 mL of 1 M NaH2PO4 and add ddH2O up to 1 L. 6. 50% glutaraldehyde (Sigma-Aldrich). 7. Prepare a 5% glutaraldehyde solution by diluting the 50% glutaraldehyde with 0.01 M PB buffer, pH 7.4 (see Note 2). 8. Streptavidin (Amresco) dissolved in 0.1 M PB buffer at 2 mg/mL; store at 20C. 9. 1x phosphate-buffered saline (PBS): 10 mM phosphate, 137 mM NaCl, 2.7 mM KCl, pH 7.4. 10. Bath sonicator (from any commercial source). 1. 50 -end biotin labeled reverse primer (Table 24.1), containing the 10–15 nt polyT or polyA spacer preceding the main sequence can be obtained from any commercial source.
2.3. Solid-Phase PCR
Table 24.1 Oligonucleotides used in this protocol Name
Type
Sequence 5’–3’
C677T FP
Forward primer
TGAAGGAGAAGGTGTCTGCGGGA
C677T RP
Biotin-labeled reverse
Biotin-(T)15-AGGACGGTGCGGTGAGAGTG
677CC probe
Wild-type probes
Cy3-CGGGAGCCGATTT
677TT probe
Mutant-type probes
Cy5-CGGGAGTCGATTT
The italicized base represents the discriminating/recognition position.
396
He, Li, and Liu
2. TE buffer: 10 mM tris(hydroxymethyl)aminomethane (Tris)– HCl, pH 8.0, 1 mM EDTA. 3. Biotin-labeled primer is dissolved in TE buffer at 10 mM; stored at –20C. 4. Reagents for PCR: 10x PCR buffer: 100 mM Tris–HCl pH 8.3, 500 mM KCl; MgCl2; deoxynucleotide triphosphates (dNTPs); Taq DNA polymerase (TaKaRa Bio, Japan). 5. 96-well PCR plates. 6. Thermal cycler: MJ PTC-220 (MJ Research). 2.4. Dual-Color Hybridization Probes
1. A pair of allele-specific dual-color SNP detection probes can be obtained from any commercial source (Table 24.1). Wildtype probe is labeled with Cy3. Mutant-type probe is labeled with Cy5. Probes are dissolved in TE buffer at 10 mM and are stored at –20C. 2. MicroHyb hybridization buffer (Invitrogen). 3. 20 SSC: 3 M NaCl, 0.3 M sodium citrate, pH 7.0. 4. Washing buffer 1:2 SSC, 1% (w/v) sodium dodecyl sulfate, stored at room temperature. 5. Washing buffer 2: 0.1 SSC, 1% (w/v) sodium dodecyl sulfate, stored at room temperature. 6. Resuspension buffer: 3x SSC. 7. Neodymium–boron (Nd–B) magnet (TDK Tokyo, Japan) for particle separation.
2.5. Slide Printing and Microarray Analysis
1. Glass slides (available from any commercial source). 2. Sulfuric acid available from any commercial source (see Note 3). 3. Hot plate, from any commercial source (keep at 100C). 4. GeSiM Nanoplotter (Gesim, Dresden, Germany). Probe solutions are printed directly onto a cleaned glass slide to fabricate a microarray. 5. A 4,100 A microarray analysis system (Axon, USA). 6. GenePix Pro 6 (microarray analysis software): http://www. moleculardevices.com/pages/software/gn_genepix_pro.html
3. Methods The PCR products are directly amplified on the surface of streptavidin-coated MNPs (SA-MNPs). After amplification, the reaction complexes are denatured and magnetically separated. Further, the single-stranded DNA (ssDNA) is bound to SA-MNPs and
High-Throughput SNP Detection Based on PCR Amplification
397
hybridized with a pair of dual-color probes. Finally, the probes hybridized with ssDNA-MNPs are denatured and printed on an unmodified glass slide. Numerous samples can be assessed simultaneously by scanning the dual-color microarray. A schematic outline of the solid-phase PCR-based genotyping method using dual-color hybridization is shown in Fig. 24.1. As an example, the methylenetetrahydrofolate reductase (MTHFR) gene C677T polymorphism was selected as a target.
Fig. 24.1. The solid-phase PCR-based genotyping method. The PCR products are directly amplified on the surface of streptavidin-coated magnetic nanoparticles (SA-MNPs). After amplification, the reaction complexes are denatured and magnetically separated, then the single-stranded DNA (ssDNA) molecules bound on SA-MNPs are hybridized with a pair of dual-color probes. Finally, the probes hybridized with ssDNA-MNPs are denatured and printed on an unmodified glass slide. Green, yellow, and red spots indicate homozygous wild-type (HoW, CC), heterozygote (He, CT), and homozygous mutant (HoM, TT) genotypes, respectively. FP forward primer. RP reverse primer.
398
He, Li, and Liu
3.1. Preparation of SA-MNPs
1. Incubate 4 mg of MNPs with 2% APTES in 95% ethanol (1 mL) at room temperature for 20 min. 2. Magnetically separate the MNPs using a Nd–B magnet, and discard the supernatant. 3. Wash the amido-modified MNPs (NH2-MNPs) with 95% ethanol and 0.1 M PB buffer (three times), respectively. 4. Incubate the NH2-MNPs with streptavidin (2 mg/mL) in 1x PBS buffer in a total volume of 200 mL for 30 min at room temperature with pulsed sonication. 5. Wash the SA-MNPs magnetically with 0.1 M PB buffer to remove the excess of unconjugated streptavidin, and resuspend the SA-MNPs in 1x PBS at a final concentration of 4 mg/mL. Store them at 4C (see Note 4).
3.2. Solid-Phase PCR on MNPs
1. Prepare a 10 mM solution of streptavidin-labeled primer in TE buffer. 2. Incubate 1.2 mM biotin-labeled primers with 4 mg SA-MNPs for 1 h at room temperature. 3. Wash the SA-MNPs with 0.1 M PB buffer three times, and resuspend the SA-MNPs in 0.1 M PB buffer at a final concentration of 4 mg/mL. Store them at 4C. 4. Prepare the PCR mixture. The 30-mL reaction mixture contains 10 mM Tris–HCl pH 8.3, 50 mM KCl, 2.0 mM MgCl2, 200 mM in each of the dNTPs, 100 ng template DNA, 1.25 U of Taq DNA polymerase, 0.5 mL 10 mM forward primer, and 50 mg MNP-bound reverse primer. 5. Run the PCR. PCR amplification is carried out in a 96-well plate in the MJ PTC-220 PCR system. After an initial denaturation at 95C for 5 min, amplification is carried out for 35 cycles: denaturation at 95C for15 s, annealing at 62C for 30 s, and extension at 72C for 30 s; and extension is additionally performed at 72C for 7 min. 6. Wash the reaction complexes twice with double-distilled H2O (ddH2O) after PCR amplification is over and then denature the probes at 95C for 5 min and keep them on ice for 1 min. (see Note 5). 7. Remove the ssDNA (supernatant) after magnetic separation from each well and save it for further analysis using 1% agarose gel electrophoresis. A representative result is shown Fig. 24.2. The concentration of ssDNA is determined by measuring the absorbance at 260 nm using a UV spectrophotometer. 8. Wash the ssDNA-MNP complexes twice with 0.1 M PB buffer, and store them in the microwell plate at 4C before hybridization.
High-Throughput SNP Detection Based on PCR Amplification
399
Fig. 24.2. Agarose gel (1%) analysis of supernatant ssDNA after PCR amplification on SA-MNPs consisted of a denaturation procedure. The target fragment amplified by solidphase PCR is 213 bp in length. A dispersed smear is also observed between 213 and 2000 bp. Compared with the MNP-bound reverse primer, the forward primer suffers a smaller effect of the steric resistanceand is at a higher concentration, so an asymmetric PCR is performed, but the smear would not interfere with genotyping results because of suffers the washing steps before hybridization. MK DNA marker.
3.3. Dual-Color Hybridization
1. Separate the ssDNA-MNP complexes using a Nd–B magnet, and discard the supernatant. 2. Prepare the hybridization mixture (final volume 20 mL), by adding 50 mg ssDNA-MNP complexes, 7 mL hybridization solution, 20 pmol wild-type and mutant probes, respectively, and add ddH2O to 20 mL. 3. Hybridization is conducted in a PCR tube using a PTC-220 thermocycler under the following conditions: lead cover temperature 102C, tube hybridization mixture temperature 38C, 40 min (see Note 6). 4. Separate hybrid-MNP complexes after hybridization using a Nd–B magnet, wash them subsequently with washing buffer 1 and washing buffer 2 (5 min for each wash, followed by a second wash with 3 SSC buffer for 3 min), and resuspend them in 20 mL of the resuspension buffer (3 SSC). 5. Denature the DNA-MNP complexes at 95C for 5 min and keep them on ice for 1 min. 6. Separate the probes using the Nd–B magnet. After magnetic separation, the MNPs are collected at the bottom of the wells, and the denatured probes are in the supernatant. They are further used for the spotting process.
3.4. Fabrication of Microarray and SNP Genotyping
1. Rerinse the glass slides thoroughly using consecutively sulfuric acid, ddH2O, and 95% ethanol. Dry them in air until use. 2. Perform the spotting. The process is carried out using a GeSiM Nanoplotter by spotting 250-pL droplets of sample solutions (in 3 SSC buffer; spot size 400 mm in diameter).
400
He, Li, and Liu
3. After spotting, snap-dry the slide for 2 s on a hot plate (100C). The microarray slide is now ready for scanning. 4. Scan the microarray using a 4,100 A microarray analysis system, fitted with filters for Cy3 and Cy5. A representative result is shown in Fig. 24.3a. 5. The images acquired by the scanner are further analyzed with the Genepix Pro 6.0 software program. For each fluorescence image, the average pixel intensity within each circle is determined and a local background using the mean pixel intensity is computed for each spot. The net signal is determined by subtraction of this local background from the mean average intensity for each spot. 6. Herein, we define two factors to identify the genotype of each sample, the signal intensities factor I – I=log [signal(Cy3+Cy5)] – and genotype factor G – G=signal ratio(Cy3/ Cy3+Cy5). When the signal intensities factors I are over 3.0, this would indicate that the summation of Cy3 and Cy5 signal intensities was over 1,000, which is sufficiently above background signals. The boundaries for clusters are defined as mutant, heterozygote, and wild type when G<0.2, 0.350.8, respectively. If samples fall outside these boundaries they would be classified as ambiguous. An example of the result is shown in Fig. 24.3b.
Fig. 24.3. The genotyping results of 126 different samples and four negative controls (NTC) assayed for the C677T locus of the MTHFR gene. (a) Each sample is spotted as four replicates in a column. (b) Cluster analysis of tag array hybridization results for 126 samples using factors I and G.
High-Throughput SNP Detection Based on PCR Amplification
401
4. Notes 1. Except for SiO2/(PMMA/Fe3O4) MNPs, the Fe3O4 nanoparticles produced by the chemical coprecipitation can also be used in this method (11). 2. Glutaraldehyde (5%) can be used up to three times within 1 week at room temperature. 3. Extreme care should be taken when using sulfuric acid. Wear gloves and eye goggles, and perform all operations under the hood. The acid is extremely corrosive. Skin contact leads to burns. Exposure to aerosols can cause damage of the respiratory tract. 4. Streptavidin should be modified on MNPs covalently. In this protocol, coating of MNPs by streptavidin is performed by using glutaraldehyde for linkage between –NH2 of APTES on the surface of MNPs and streptavidin. 5. In this protocol, the double-stranded DNA also can be denatured by adding 25 mL of 0.1 M NaOH to the well and letting it stand for 1–2 min. Compared with thermal denaturation, this variant is more suitable for automated operation. 6. This protocol can be adapted for identification of different SNP loci. In the hybridization-based genotyping method, the hybridization temperature is an important factor for accurate discrimination. When dealing with different loci, the optimum temperature of each locus can be obtained according to the previously described method (12).
Acknowledgements This work was supported by the National Science Foundation of China (60801007 and 90606027), the National Key Program for Developing Basic Research (2007CB936104), the Scientific Research Fund of Hunan Provincial Education Department (08B018), the Hunan Natural Science Foundation (09JJ4004 and 09JJ3017), the Chinese 863 Program (2006A A02Z133 and 2007AA022007), the China Postdoctoral Science Foundation (20080430160), and Jiangsu Planned Projects for Postdoctoral Research Funds (0801004B).
402
He, Li, and Liu
References 1. Marshall, E. (1997) The hunting of the SNP. Science 278, 2046–2048. 2. Pinnisi, E. (1998) A closer look at SNPs suggests difficulties. Science 281, 1787–1789. 3. Flavel, A. J., Bolshakov, V. N., Booth, A., Jing, R., Russell, J., Ellis, T. H. and Isaac, P. (2003) A microarray-based high throughput molecular marker genotyping method: the tagged microarray marker (TAM) approach. Nucleic. Acids. Res. 31, e115. 4. Hultin, E., Ka¨ller, M., Ahmadian, A. and Lundeberg, J. (2005) Competitive enzymatic reaction to control allele-specific extensions. Nucleic. Acids. Res. 35, e48. 5. Niemeyer, C. M. (2001) Nanoparticles, proteins, and nucleic acids: biotechnology meets materials science. Angew. Chem. Int. Ed. 40, 4128–4158. 6. Zhao, X. J., Tapec-Dytioco, R., Wang, K. M. and Tan, W. H. (2003) Collection of trace amounts of DNA/mRNA molecules using genomagnetic nanocapturers. Anal. Chem. 75, 3476–3483. 7. Ingram, R., Tagoh, H., Riggs, A. D. and Bonifer, C. (2005) Rapid, solid-phase based automated analysis of chromatin structure and transcription factor occupancy in living eukaryotic cells. Nucleic. Acids. Res. 33, e1.
8. Li, S., Liu, H. N., Wang, Z. F., Hou, P., Guo, Y. F., He, Q. G. and He, N. Y. (2006) Magnetic-particles-based high-throughput genotyping method with dual-color fluorescence hybridization. Anal. Biochem. 359, 277–279. 9. Liu, H. N., Li, S., Ji, M. J., Nie, L. B. Chen, J. R., Miao Y. Q., and He, N. Y. (2008) Fabrication and application of single nucleotide polymorphisms library on magnetic nanoparticles using adaptor PCR. J. Nanosci. Nanotechnol. 8, 405–409. 10. Wang, Z. F., Guo Y. F., Li, S., Sun, Y. M. and He N. Y. (2008) Synthesis and characterization of SiO2/(PMMA/Fe3O4) magnetic nanocomposites. J. Nanosci. Nanotechnol. 8, 1797–1802. 11. Philipse, A. P., van Bruggen, M. P. B. and Pathmamanoharan, C. (1994) Magnetic silica dispersions: preparation and stability of surface-modified silica particles with a magnetic core. Langmuir 10, 92–99. 12. Liu, H. N., Li, S., Wang, Z. F., Ji, M. J., Nie, L. B. and He, N. Y. (2007) Highthroughput SNP genotyping based on solid-phase PCR on magnetic nanoparticles with dual-color hybridization. J. Biotech. 131, 217–222.
Chapter 25 Restriction Enzyme Analysis of PCR Products Masao Ota, Hideki Asamura, Takahito Oki, and Masaharu Sada Abstract Progress in single nucleotide polymorphism (SNP) detection technologies has provided information for SNP-based studies, such as identification of candidate genes for the complex genetic diseases, pharmacogenetic analysis, drug development, population genetics, evolutionary studies, and forensic investigations. SNP detection is performed by many methods, including hybridization, allele-specific polymerase chain reaction (PCR), primer extension, oligonucleotide ligation, direct DNA sequencing, and endonuclease cleavage. Each of these methods has its specific advantages and disadvantages. Here we introduce the PCR–restriction fragment length polymorphism (RFLP) method, which has certain advantages over many other techniques used for analysis of SNPs. The PCRRFLP method allows very rapid, simple, and inexpensive detection of point mutations within the sequences of PCR products. The mutation is discriminated by the specific restriction endonuclease and is identified by gel electrophoresis followed by staining with ethidium bromide. This convenient and simple method is useful in a small basic research study. Key words: PCR–restriction fragment length polymorphism, restriction endonuclease, single nucleotide polymorphism, electrophoresis, polyacrylamide gel.
1. Introduction Molecular genetics research has dramatically progressed since the invention of the polymerase chain reaction (PCR) (1, 2). Detection of genetic polymorphisms using the sensitive PCR-based technique has been applied to finding the biological meaning of nucleic acid divergences in living organisms. The most common form of genetic polymorphism is accounted for by SNPs (3). SNP genotyping assays play an important role in the study of molecular genetics and genomics, such as the investigation of the genetic causes of complex human diseases, pharmacogenetics (response to therapeutic drugs), population genetics, and evolution. A.A. Komar (ed.), Single Nucleotide Polymorphisms, Methods in Molecular Biology 578, DOI 10.1007/978-1-60327-411-1_25, ª Humana Press, a part of Springer Science+Business Media, LLC 2003, 2009
405
406
Ota et al.
Over the past few years, many different methods have been developed for SNP genotyping, based on a molecular mechanism: allele-specific oligonucletide hybridization (4), fluorescent probes (5), melting-temperature-based separation including single-strand conformational polymorphism (6), heteroduplex analysis (7), allelespecific amplification (8), direct DNA sequencing (9) or primer extension (10), and PCR–restriction fragment length polymorphism (RFLP) (11). Most of these applications are described in this volume. Recent technologies, e.g., the TaqMan method (Chapters 18 and 19), the Invader method (Chapter 22), the matrix-assisted laser desoprtion/ionization time off light method (Chapter 20), and GeneChips posses a capacity for large-scale SNPs genotyping; however, they are costly and usually require expensive equipment. Admittedly, the level of throughput in SNP detection depends mostly on the aim of the study. One of the inexpensive, simple, and convenient methods for SNP genotyping is PCR-RFLP, which is based on endonuclease digestion of the PCR-amplified DNA. The presence of a SNP usually creates new restriction site(s) or abolishes the existing one(s), thus allowing the detection. We describe here a general protocol for the PCR-RFLP method.
2. Materials 2.1. SNP sample
1. Template DNA in tris(hydroxymethyl)aminomethane (Tris)– EDTA (TE) buffer (see below) or sterile water (see Note 1). Generally prepare 10–50 ng of good-quality DNA.
2.2. Polymerase Chain Reaction
1. 10x PCR buffer (supplied by the manufacturer with Taq polymerase). 2. Taq polymerase (general PCR or hot-start PCR depending on the PCR specificity). 3. 10 mM solution of each dNTP (dATP, dCTP, dGTP, dTTP; supplied by the manufacturer with the Taq polymerase). 4. PCR primers; 20 mM solution of each in TE buffer (see below) or sterilized water. 5. DNase-free distilled water (usually autoclaved). 6. TE buffer: 10 mM Tris–HCl, pH 8.0, 1 mM EDTA (store at room temperature). 7. PCR chubs, 96-well mutliwell plates (when using multiple samples). 8. PCR thermal cycler. 9. Pipettors and autoclaved tips (use filter tips, if available).
Restriction Enzyme Analysis of PCR Products
2.3. Analysis of the PCR Product(s)
407
1. DNA size markers (100- or 123-bp ladder, can be obtained by restriction digestion of pBR322 with HaeIII, ’x174 DNA with HaeIII, ’x174 DNA with HinfI, etc., or can be obtained from commercial sources). 2. Agarose (low electroendosmosis and/or NuSieve1 3:1 agarose; used to prepare 2–4% agarose gels). 3. Running TBE buffer (1X): 89 mM Tris–borate, pH 8.0, 2 mM EDTA (see Note 2). 4. Gel-loading dye: 0.25% bromophenol blue, 0.25% xylene cyanol, 50% glycerol, 1 mM EDTA, pH 8.0. 5. Horizontal minigel electrophoresis apparatus. 6. Ethidium bromide stock solution: 10 mg/mL (store at 4C in a brown bottle, see Note 3). 7. UV transilluminator. 8. Gel documentation apparatus (an instant camera or a CCD camera).
2.4. RFLP Analysis
1. Appropriate restriction enzyme(s) for detection of SNP(s) of interest (see Section 3.2 and Note 4). 2. Gel-loading dye (stops restriction digest): 50 mM EDTA, pH 8.0, 50% glycerol, and 0.05% bromophenol blue. 3. 10% (w/v) ammonium persulfate (see Note 5). 4. Acrylamide/bisacrylamide stock solution (29:1) (see Note 6). Prepare 30% stock solution by adding 29 g acrylamide and 1 g N0 ,N0 -methylene bisacrylamide, and bring the volume to 100 mL with water. 5. N,N,N0 ,N0 -Tetramethylethylenediamine (TEMED) (store at 4C). 6. Vertical or horizontal mini gel electrophoresis apparatus.
3. Methods 3.1. Primer Design
Before the start of PCR-RFLP analysis, one should design the optimum primer pair(s) and search for restriction enzymes for discriminating target SNPs within the alignment of the product amplified by PCR. Primer design is essential to obtain high-quality results, i.e., amplify a given DNA sequence and not to amplify orthologous nonspecific sequences (see Note 7). 1. Design the optimal primer pair taking into consideration the GC composition, melting temperature, length, self-complementarity, and complementarity to each other, particularly in the 30 region. To facilitate primer design with complicate and
408
Ota et al.
cumbersome steps, software is available (12, 13). Primers should generally satisfy a simple set of rules: they should be about 17–25 bases in length; (G+C) base composition should be 50–60%; primers should preferably end (30 ) in a G or C, or CG or GC; Tms should be between 55 and 62C; self-complementarity should be avoided and 30 ends of the primers should not be complementary to each other. 3.2. Selection of Restriction Enzymes for Discriminating SNPs
1. To search for appropriate restriction enzymes for discriminating target SNPs within the sequences of PCR-amplified product, several comprehensive tools have been described and are supported by either commercial software (Genetyx, Genetyx; Gene Construction Kit, Textco BioSoftware) or free software (e.g., http://re.neb.com/rebase/rebase.ftp.html (14) or http://bioinfo.bsd.uchicago.edu/SNP_cutter.htm (15)). 2. If it is impossible to find restriction enzymes discriminating the target SNP of interest, then try to use the mismatch PCRRFLP method. Mismatch PCR-RFLP is performed by providing an artificial restriction enzyme site proximate to the SNP site. The artificial restriction site is constructed by adding mismatch bases in PCR primer sequences.
3.3. Template DNA Preparation
1. Isolate the target DNA. DNA can be isolated from a variety of sample sources (e.g., fresh or frozen tissues and cells, blood, body fluids, paraffin-embedded tissue, formalin-fixed tissue, insects, yeasts, or bacteria). High-quality and high-quantity template DNA can be extracted by many commercially available preparation kits as well as them commonly used phenol/ chloroform extraction method (16). 2. Assess the purity of the template DNA for PCR, by measuring the optical density at 260 and 280 nm (the DNA sample should be dissolved in TE buffer or water). An A260/A280 ratio of 1.7–1.9 suggests good quality DNA suitable for further PCR amplification.
3.4. Polymerase Chain Reaction
PCR amplification (17) is performed in the automated thermal cycler. The amplification parameters are generally designed according to the manufacturer’s instructions for Taq polymerase and the primers used in PCR. Temperatures and cycling times are optimized for templates or primer pairs (see Note 7). 1. Set up a PCR mixture following the recipe presented in Table 25.1. 2. Include one negative control (no DNA) in a different tube during the PCR setup. 3. Run the PCR, for example, using the following standard scheme: one cycle 94C for 2 min; 25–30 cycles 94C for 30 s, 55C for 30 s, 72C for 45 s; one cycle 72C for 5 min.
Restriction Enzyme Analysis of PCR Products
409
Table 25.1 PCR preparation to amplify DNA samples for digestion with a single restriction enzyme Volume (mL) n=1 n = 20
Component
Final concentration
10 x PCR buffer
2.0
44.0
1x PCR
dNTP mix (2.5 mM in each dNTP)
1.6
35.2
0.2 mM: each dNTP
Forward primer (100 pmol/mL)
0.2
4.4
1 mM
Reverse primer (100 pmol/mL)
0.2
4.4
1 mM
Taq polymerase (5 U/mL)
1.0
22.0
Sterile water
14.8
325.6
Total
20.0
440.0
1U
When setting up more than one sample, add an extra 10% of each component. 3.4.1. Checking the PCR Products
Confirm that PCR ran successfully; check the PCR yield, expected size, and nonspecific background on a 2% agarose gel (see Note 7). 1. Place 6 mL (5 mL of each reaction product plus 1 mL of gelloading dye) on a horizontal agarose gel. Load on a separate lane an optimal DNA size marker for comparison. 2. Electrophorese the samples for appropriate times and stain thehm with ethidium bromide (0.5 mg/mL) for 10 min. Alternatively, use the gel already containing ethidium bromide (0.1 mg/mL) (see Note 3).
3.5. RFLP Analysis of PCR Products 3.5.1. Digestion of PCR Products
3. Visualize the PCR products on a UV transilluminator and document the results (see Note 8). The most convenient method for digestion of PCR-amplified DNA is the addition of restriction enzyme(s) (5–10 U) directly to the mixture containing aliquots (5–8 mL) of the PCR products. One should not forget to add the appropriate restriction buffer. Digestion is usually performed by adding 1 mL of enzymes (10 U) to the PCR product in a final volume of 10 mL, containing the appropriate 1X buffer, and followed by an incubation time of more than 1 h at an optimal temperature for the restriction enzyme chosen. Fewer units of enzyme might be available; in that case the incubation time should be extended. 1. Set up the reaction(s) as suggested in Table 25.2, in the case of using a single restriction enzyme for digestion of multiple DNA samples. 2. When using multiple restriction enzymes for digestion of multiple DNA samples, set up the reaction as shown in Table 25.3. Generally, 10 U of restriction enzymes is sufficient to overcome variability in DNA source, quantity, and purity.
410
Ota et al.
3. Stop the reactions by adding the DNA loading buffer. If further manipulations of digested DNA are required, heatinactivation of the digestion reaction (65 or 80C for 20 min) is recommended.
Table 25.2 Preparation of a master-mix to digest PCR products with a single restriction enzyme Component
Volume (mL) n=1
n = 20
10x enzyme buffer
1.0
22.0
Restriction enzyme (10 U/mL)
0.5
11.0
Sterile water
2.5
55.0
Total
4.0
88.0
When setting up more than one sample, add an extra 10% of each component. Each sample is subjected to a reaction with a total of 10 mL of mixture, containing 4 mL of the digestion master mix and 6 mL of the PCR products.
Table 25.3 Preparation of a master-mix to digest PCR products with multiple restriction enzymes Component
Volume (mL) n=1
n = 10
10x enzyme buffer
1.0
11.0
PCR product
6.0
66.0
Sterile water
2.0
22.0
Total
9.0
99.0
When setting up more than one enzyme, add an extra 10% of each component. When using nine restriction enzymes, prepare a master mix as for ten. One microliter of each restriction enzyme is added to 9 mL of a master mix.
3.5.2. Gel Electrophoresis of SNP Samples Digested with Restriction Enzymes
1. Load samples [5 mL of each reaction mixture product (amplified DNAs digested by restriction enzymes) plus 1 mL of gel-loading dye] onto 6–12% vertical polyacrylamide gel, or, alternatively, onto 4% horizontal agarose (NuSieve1 3:1) gel (see Note 9). 2. Load on a separate lane the appropriate DNA size markers (e.g., 100-bp ladder, pBR322-HaeIII, ’x174-HaeIII, or ’x174-HinfI). Each gel should also include the appropriate controls for enzyme activity (see Note 10).
Restriction Enzyme Analysis of PCR Products
411
3. Subject the gels to electrophoresis (see Note 11). 4. Stain the gel (after the run is over) with ethidium bromide at 0.5 mg/mL for 10 min. 5. Visualize the results on a UV transilluminator. 6. Document the results. 7. Compare the cleaved fragments and undigested DNA (see Notes 10 and 12). The PCR-RFLP method is a simple, sensitive, and reliable method that requires minimal investment in instrumentation. This method is available for genotyping not only of SNPs but also for insertion/deletion polymorphisms (18, 19) and multiple mutations such as human leukocyte antigen alleles (Fig. 25.1) (20, 21).
1
2
3
267 bp 174 bp 124 bp
HLA-DRB1*1501/1302 1 2 3 4 5
4
6 7 8
9 10 11
267 bp 120 bp
174 bp 124 bp
DRB1*1501 1: marker,pBR322/HaeIII digestion
109 bp
DRB1*1302 2: AvaII, uncut, 265 bp 3: Fok I, cut, 170 bp, 95 bp
2: Fok I, cut, 166 bp, 95 bp 3: Cfr13I, cut, 197 bp, 64 bp 4: HphI, cut, 145bp, 120bp
4: Kpn I, uncut, 265 bp 5: Hae II, uncut, 265 bp 6: Cfr13I, cut, 201 bp, 63 bp, 1 bp 7: SfaNI, cut, 145 bp, 120 bp 8: SacII, cut, 207 bp, 58 bp 9: BsaJI, cut, 204 bp, 61 bp 10: ApaI, uncut, 265 bp 11: HphI, cut, 145bp, 109bp, 11bp
Fig. 25.1. Human leukocyte antigen (HLA)–DRB1 genotyping from DNA of a normal individual (DRB1*1501/1302 heterozygote) (20). HLA alleles are determined with the modified polymerase chain reaction (PCR)–restriction fragment length polymorphism method (20, 21) incorporating informative enzymes, which have a single recognition site in some alleles but none in other alleles in the amplified region. Samples of amplified DNAs cleaved by restriction enzymes were run on a 12% polyacrylamide horizontal gel. Assignment of alleles is based on the combination of band patterns produced by digestion with multiple restriction enzymes. The gel pictures show the restriction patterns for PCR products amplified by DRB1*2 group specific primers (left) and DRB1*3,5,6,8 group specific primers (right). The left-end lane is a size marker of HaeIII-digested pBR322 DNA.
412
Ota et al.
4. Notes 1. For long-term storage of DNA samples, dissolving them in TE buffer and storing them at –20C is recommended, since DNA stored in water is subject to acid hydrolysis. 2. TBE buffer is frequently used in DNA gel electrophoresis. Make a (10X) stock solution by adding 108 g Tris, 55 g boric acid, and 40 mL 0.5 M EDTA, pH 8.0, and bring the volume to 1 L with sterile water. Store the buffer at room temperature. 3. Care should be taken in handling ethidium bromide because it is a mutagen and environmental hazard. Wear gloves and dispose of it properly. 4. Enzyme activity decreases rapidly if restriction enzymes are not kept at –20C. Always keep stocks of restriction enzymes on ice during handling and place them immediately at –20C after handling. 5. Prepare a fresh solution before use. 6. Both acrylamide and N0 ,N0 -methylene bisacrylamide are highly neurotoxic and can be readily absorbed through the skin. Wear a protective mask and gloves when them handling. 7. A number of potential problems can be encountered in the PCR-RFLP method. Conceivable suggestions may help to resolve them. In the case of absence of the PCR product, check whether the primers have been carefully designed, the template DNA concentration was determined correctly, or verify that the template DNA was indeed added to the reaction mixture. Also check whether the thermal cycler is in fact working properly. Using fresh Taq polymerase or using Taq polymerase from a different company might sometimes help to resolve the problem. In the case of appearance of extra PCR products, repeat the amplification with new preparations of all reagents [especially if products appeared in a negative control (no DNA) sample reaction]. Likewise, changing the annealing temperature, extension temperature, or cycling numbers and/or changing the primer and DNA concentration, and/ or alternatively changing the Mg2+ concentration (generally a decrease in Mg2+ concentration increases the stringency of the reaction) might help to resolve the problem. 8. Care should be taken when using UV light sources. Wear safety glasses or a face mask. 9. Polyacrylamide gels are prepared with appropriate acrylamide concentrations depending on the restriction fragment lengths used for analysis. For example, to make 100 mL of 12%
Restriction Enzyme Analysis of PCR Products
413
polyacrylamide gel, mix 40 mL of 30% acrylamide/bisacrylamide stock (29:1) solution, 10 mL of 10x TBE buffer, and then add water to 99 mL. Subsequently, add 940 mL of 10% ammonium persulfate and 60 mL of TEMED for polymerization. A 12% polyacrylamide gel is usually used for separation of DNA fragments between 30 and 500 bp in length. A 4% agarose gel allows reliable separation of fragments between 50 and 600 bp in length. 10. When DNA samples are incompletely digested or not digested at all, purification of the PCR product(s) is sometimes useful for improving or recovering the efficiency of the restriction enzyme activity. Also, storage buffers for most restriction enzymes contain 25–50% glycerol to prevent freezing during storage at –20C. During restriction enzyme digestion, high glycerol concentrations inhibit the activity of restriction enzymes or affect the specificity (resulting in unexpected cleavages; the so-called star activity). It is, therefore, important to keep the glycerol concentration at less than 5% and for it not to exceed 10% of the total reaction volume. 11. Running conditions are different depending on the gel composition and density. Run the gel for a long enough time to obtain the desired resolution of the fragments after digestion with restriction enzymes. Run the gel at a constant voltage of 2–10 V/cm. The bromophenol blue dye front migrates approximately with 50-bp DNA fragments, whereas the xylene cyanol dye front runs with 400-bp fragments in a 5% polyacrylamide gel. 12. If unexpected fragments appear after RFLP analysis, check whether the reaction conditions allowed for complete digestion of the PCR fragment(s). In the control reaction use the DNA fragment which has been proven to have cleavage site(s) for the same restriction enzyme(s). Also, one may use a fresh batch of the enzyme or an enzyme from a different company. Increase the volume of the enzyme and/or extend the incubation time. Check if an appropriate digestion buffer has been used. Also see Note 10. References 1. Saiki, R. K., Scharf, S., Faloona, F., Mullis, K. B., Horn, G. T., Erlich, H. A. and Arnheim N. (1985) Enzymatic amplification of beta-globin genomic sequences and restriction site analysis for diagnosis of sickle cell anemia. Science 230, 1350–1354. 2. Saiki, R. K., Bugawan, T. L., Horn, G. T., Mullis, K. B. and Erlich, H. A. (1986) Analysis of enzymatically amplified beta-globin
and HLA-DQ alpha DNA with allele-specific oligonucleotide probes. Nature 324, 163–166. 3. Collins, F. S., Brooks, L. D. and Chakravarti, A. (1998) A DNA polymorphism discovery resource for research on human genetic variation. Genome Res. 8, 1229–1231. 4. Saiki, R. K., Chang, C. A., Levenson, C. H., Warren, T. C., Boehm, C. D., Kazazian, Jr.,
414
5.
6.
7.
8.
9.
10.
11.
12. 13.
Ota et al. H. H. et al. (1988). Diagnosis of sickle cell anemia and b-thalassemia with enzymatically amplified DNA and nonradioactive allelespecific oligonucleotide probes. N. Engl. J. Med. 319, 537–541. Tyagi, S. and Kramer, F. R. (1996) Molecular beacons: probes that fluoresce upon hybridization. Nat. Biotechnol. 14, 303–308. Orita, M., Iwahana, H., Kanazawa, H., Hayashi, K. and Sekiya, T. (1989) Detection of polymorphisms of human DNA by gel electrophoresis as single-strand conformation polymorphisms. Proc. Natl. Acad. Sci. U.S.A. 86, 2766–2770. Prior, T. W., Papp, A. C., Snyder, P. J., Burghes, A. H., Sedra, M. S., Western, L. M. et al. (1993) Identification of two point mutations and a one base deletion in exon 19 of the dystrophin gene by heteroduplex formation. Hum. Mol. Genet. 2, 311–313. Newton, C. R., Graham, A., Heptinstall, L. E., Powell, S. J., Summers, C., Kalsheker, N. et al. (1989) Analysis of any point mutation in DNA. The amplification refractory mutation system (ARMS). Nucleic Acids Res. 17, 2503–2516. Sanger, F., Nicklen, S. and Coulson, A. R. (1977) DNA sequencing with chain terminating inhibitors. Proc. Natl. Acad. Sci. U.S.A. 74, 5463–5467. Syvanen, A. C., Aalto-Setala, K., Kontula, K. and Soderlund, H. (1990) A primer guided nucleotide incorporation assay in the genotyping of apolipoprotein E. Genomics 8, 684–692. Ota, M., Fukushima, H., Kulski, J. K. and Inoko, H. (2007) Single nucleotide polymorphism detection by polymerase chain reaction-restriction fragment length polymorphism. Nat. Prot. 2, 2857–2864. van Baren, M.J. and Heutink, P. (2004) The PCR suite. Bioinformatics 20, 591–593. Gardner, S. N. and Wagner, M. C. (2005) Software for optimization of SNP and PCR-
14.
15.
16.
17.
18.
19.
20.
21.
RFLP genotyping to discriminate many genomes with the fewest assays. BMC Genomics 6, 73. Roberts, R. J., Vincze, T., Posfai, J. and Macelis, D. (2005) REBASE-restriction enzymes and DNA methyltransferases. Nucleic Acids Res. 35, D230–D232. Zhang, R., Zhu, Z., Zhu, H., Nguyen, T., Yao, F. and Xia, K. (2005) SNP cutter: a comprehensive tool for SNP PCR-RFLP assay design. Nucleic Acids Res. 33, W489–W492. Moore, D. D. (1994) Purification and concentration of DNA from aqueous solutions. In: Current Protocols in Molecular Biology (eds. Ausubel, F. M., Brent, R., Kignston, R. E., Moore, D. D., Seidman, J. G., Smith, J. A., and Struhl, K.), John Wiley and Sons Inc., Hoboken, NJ, USA, 2.1.1–2.1.9. Innis, M. A., Gelfand, D. H., Sninsky, J. J. and White, T. J. (ed.) (1990) PCR Protocols. Academic Press Inc., San Diego, CA, USA. Simsekm, M., Al-Wardy, N., Al-Khayat, A. and Al-Khabory, M. (2002) A PCR-RFLP test for simultaneous detection of two single-nucleotide insertions in the Connexin-26 gene promoter. Genet. Test. 6, 225–228. Bu, H., Rosdahl, I., Sun, Z. -F. and Zhang, H. (2007) Importance of polymorphisms in } BI and NF-BI genes for melaNF-U noma risk, clinicopathological features and tumor progression in Swedish melanoma patients. J. Cancer Res. Clin. Oncol. 183, 856–866. Ota, M., Seki, T., Fukushima, H., Tsuji, K. and Inoko, H. (1992) HLA-DRB1 genotyping by modified PCR-RFLP method combined with group-specific primers. Tissue Antigens 39, 187–202. Inoko, H. and Ota, M. (1993) PCR-RFLP. In: Handbook of HLA typing Techniques (eds. Hui, K. M. and Bidwell, J. L.), CRC press, Boca Raton, Ann Arbor, London, Tokyo, pp. 9–70.
Chapter 26 Allele-Specific PCR in SNP Genotyping Muriel Gaudet, Anna-Giulia Fara, Isacco Beritognolo, and Maurizio Sabatti Abstract The increasing need for large-scale genotyping applications of single nucleotide polymorphisms (SNPs) in model and nonmodel organisms requires the development of low-cost technologies accessible to minimally equipped laboratories. The method presented here allows efficient discrimination of SNPs by allelespecific PCR in a single reaction with standard PCR conditions. A common reverse primer and two forward allele-specific primers with different tails amplify two allele-specific PCR products of different lengths, which are further separated by agarose gel electrophoresis. PCR specificity is improved by the introduction of a destabilizing mismatch within the 30 end of the allele-specific primers. This is a simple and inexpensive method for SNP detection that does not require PCR optimization. Key words: Single nucleotide polymorphism genotyping, allele-specific PCR, destabilizing mismatch, tailed primers, primer design.
1. Introduction Single nucleotide polymorphisms (SNPs) are the most abundant genetic markers in the genome (1). Many SNPs are widely used in this regard in animal genomics, and are increasingly being adopted in plant genetic studies. A wide variety of SNP detection methods have been developed and many of them use automated highthroughput systems, which as a rule require expensive equipment and/or have development costs (2, 3). Simple and cost-effective methods for SNP genotyping would increase the accessibility of an average research laboratory to SNP markers (4). Among the simple SNP genotyping methods, the allelespecific (AS) PCR (AS-PCR), also called ‘‘PCR allele-specific amplification’’ (PASA) or ‘‘amplification refractory mutation,’’ A.A. Komar (ed.), Single Nucleotide Polymorphisms, Methods in Molecular Biology 578, DOI 10.1007/978-1-60327-411-1_26, ª Humana Press, a part of Springer Science+Business Media, LLC 2003, 2009
415
416
Gaudet et al.
allows SNP to be detected in minimally equipped laboratories (4). This method relies on obtaining a PCR product specific to the SNP polymorphism using AS primers that have the 30 -end base complementary to the SNP site. The advantage of the basic AS-PCR is the low cost and fast detection of the amplification products on agarose gel. Nevertheless, two reactions are needed, one for each SNP allele (5). Additionally, the single base-pair change at the 30 end is not always sufficient to ensure reliable discrimination between the two SNP alleles (6). Improvement was further achieved by the detection of all the SNP alleles in a single-tube reaction with the use of at least two specific primers together (7– 9). This method is also called ‘‘PCR amplification of multiple specific alleles’’ (7). However, the apparent simplicity of a single-tube reaction has its own drawbacks. Such a type of technology often requires more expensive equipment and reagents, e.g., associated with the need to use (and detect) fluorescent probes (9–13) to allow discrimination of the products in a single test tube. Other methods based on the use of AS primers with different lengths were developed to allow the discrimination of alleles by gel electrophoresis. Two previous protocols described an AS-PCR with one AS primer either 31 nucleotides (7) or five and two nucleotides (8) longer than the other AS primer. In the first case (7), the technique required rigorous optimization of the PCR conditions, because of the substantial difference in the lengths between the primers. In the second case (8), separation of the products on acrylamide gel and optimization of the molar ratio of the AS primers were needed. The bidirectional PASA is a variant of the AS-PCR, and also allows the allele separation on agarose gel. Four primers are needed: two outer primers and two AS inner primers (14). One of the alleles is amplified by an AS-PCR in one direction with an AS inner primer and an outer primer, whereas the second allele is amplified in the opposite direction with the other inner and the outer pair of primers. The two outer primers also amplify a constant fragment, which serves as a positive control for PCR. Once again, this method requires PCR optimization because of the competition between multiple primer sets in the PCR (14–17), and the efficient amplification of all the expected fragments cannot always be achieved and cannot be guaranteed (16). An important improvement of the AS-PCR method and its variants involves the introduction of a destabilizing mismatched base pair within four bases of the 30 primer end, in addition to the 30 -end base complementary to the SNP site. The mismatch increases the primer specificity and was successfully used in different AS-PCRs (4, 5, 8, 18). The new AS-PCR method that we describe here allows efficient discrimination of the SNP alleles on agarose gel in a single reaction, without the need for PCR optimization. Three
Allele-Specific PCR in SNP Genotyping
417
unlabeled primers are required: two AS primers and a common primer. In the AS primers, a destabilizing mismatch is introduced within the five bases of the 30 end, and a 50 tail is added for the amplification of different-length PCR products. Additional mismatches and tails confer high allele specificity, and allow megapriming to be avoided. The length of the tail is adapted to allow allele discrimination on agarose gel and avoid PCR optimization.
2. Materials 2.1. Primer Design
2.2. PCR Amplification
This method requires knowledge of the sequence around the SNP in order for the PCR primers to be designed (see Section 2.1 for details). General rules for the PCR primer design apply. Software that allows one to test the formation of hairpin loops, dimers, and duplexes of the primers is useful. We performed the design of the primers using Vector NTI 10.3.0 (Invitrogen). 1. Taq DNA polymerase (5 U/mL). Stored at –20C. 2. Taq buffer (10X): 50 mM KCl, 10 mM tris(hydroxymethyl) aminomethane (Tris)–HCl pH 9, 25 mM MgCl2 (GE Healthcare). Stored at –20C. 3. dNTPs (GE Healthcare) diluted to 2 mM (in each) in water (see Note 1) and stored at –20C. A solution containing all four dNTPs should be prepared, in which each dNTP would have a concentration of 2 mM. 4. Primers should be diluted to 10 mM in water and stored at –20C. 5. Thermocycler.
2.3. Gel Electrophoresis
1. MetaPhor1 agarose (Cambrex), stored at room temperature (see Note 2). 2. Buffer for gel preparation and electrophoresis: Tris–borate– EDTA (TBE; 5X stock solution): 45 mM Tris–borate, pH 8.0, 1 mM EDTA in distilled water (see Note 3). Also have 1X and 0.5X working solutions in distilled water. Store at room temperature. 3. Gel-loading buffer 6X: 0.5% bromophenol blue, 30% glycerol in water. Store at 4C. 4. DNA size standards, e.g., pUC19 DNA/MspI (HpaII) marker 23, prepared and stored as indicated by the manufacturer (M-Medical).
418
Gaudet et al.
5. Ethidium bromide staining solution (Sigma-Aldrich) at 1.25 mg/mL in distilled water. Store at room temperature in the dark (see Note 4). 6. Standard horizontal gel electrophoresis apparatus. 7. UV transilluminator.
3. Methods This method is based on the choice/design of two AS primers that would include (end on) a SNP and a one common (reverse) primer to amplify the PCR fragments (Fig. 26.1). A tail is incorporated in the AS primers, thus allowing differentiation of the alleles through the length of the PCR products on agarose gel. The different
Fig. 26.1. Principle of the allele-specific (AS) PCR (AS-PCR). (a) Template DNA containing the single nucleotide polymorphism (SNP) site is amplified by two forward AS primers and a common reverse primer. The AS primers end at the SNP position and a destabilizing mismatch is added within the five bases of the 30 end to enable allele specificity. Allele 1 is amplified by the AS primer with a short tail, while allele 2 is amplified by the AS primer with a long tail. Therefore, the two alleles are discriminated by the length of the amplification products. (b) Amplification products obtained by AS PCR for the case presented in a. The amplification products were separated on 3% MetaPhor1 agarose gel and stained with ethidium bromide. The last lane is the DNA size standard and the fragment size is given in base pairs on the right. The SNP genotypes are depicted at the top of each lane.
Allele-Specific PCR in SNP Genotyping
419
sequences of the tails allow discrimination of the two AS PCR products; the 50 end of the primer is not considered essential for the specificity. The key to efficient discrimination of the two alleles is the incorporation of a destabilizing mismatch within the five bases of the 30 primer end, which is different between the two AS primers. In this way, cross-amplification as a result of complementarity between one AS primer and the product of the others is avoided. Effectively, certain mismatches, such as G–T or C–A, are poorly discriminated by the Taq DNA polymerase and yield an extension product (6, 19). The addition of the second mismatch abolishes the nonspecific amplification. The amplification efficiency of the two (expected) PCR fragments could be different, but the pattern remains clearly readable for the genotyping. 3.1. Primer Design
1. Firstly, we design the AS forward primers without mismatch and with a tail. They are positioned with the 30 -terminal end at the SNP base. The length is adjusted so that the melting temperature (Tm) is between 50 and 65C and with no more than 5C difference between the three primers. The starting primer length is 20 nucleotides and, if necessary, we add or remove one or two nucleotides at the 50 end to adjust Tm. 2. The program Vector NTI 10.3.0 can be used to design the reverse (common) primer. One should take into account the following considerations. The forward primer is one of the AS primers defined in step 1. The length of the PCR product should be within the 100–200 base pair (bp) range (see Note 5). The primer Tm should be between 50 and 65C. The range of the primer lengths should be within 18–22 bp; the maximum Tm difference between the forward and the reverse primer is set at 5C. The software helps to find a set of reverse primers corresponding to these criteria. 3. The best reverse primers found by the software are then tested for the formation of the stable hairpin loops and dimers against themselves and against the two AS primers. The reverse primer that would reveal the less stable secondary structure formation, especially in the 30 region, should then be chosen. 4. For each AS primer, a mismatch is added within the five bases of the 30 end. The added mismatch must differ between the two AS primers (Fig. 26.2, see Note 6) either in its base composition or in the position. These mismatched primers are then tested for hairpin loops and dimer formation. The position and the nature of the base of the mismatch can be further adjusted to reduce the hairpin and dimer formation. 5. Tails of different lengths are added to obtain a difference of 10 bp between the amplification products (5 and 15 bp when the primer is 20 bp long). In this way the tails are not too long
420
Gaudet et al.
Fig. 26.2. General scheme of the AS-PCR for a SNP with identical or different destabilizing mismatches between the two AS primers. The same samples were genotyped for a specific SNP with two different couples of AS primers. The amplification products were separated on a 3% MetaPhor1 agarose gel and stained with ethidium bromide. The expected genotypes are indicated under the gel pictures. The amplification product for the allele A is shorter than the PCR product for the allele G. (a) AS-PCR was performed with two AS primers having the same destabilizing mismatch (G) at the same position. In this case the two SNP alleles are not discriminated. (b) AS-PCR carried out with two AS primers having different destabilizing mismatches (G and C) at the same position. In the latter case, the SNP alleles are efficiently discriminated.
and the difference of 10 bp between the PCR products can be easily detected on a MetaPhor1 agarose gel. The base composition of the tails is chosen such that it should limit the formation of hairpin loops and dimers in the primer. 6. The three primers (the two forward AS primers, ASF1 and ASF2, and the common reverse primer, R) are finally tested pair by pair (R/ASF1, R/ASF2, ASF1/ASF2) to check the formation of stable dimers, and their composition is further adjusted, if necessary (see Note 7). 3.2. PCR Amplification
PCRs are performed in a total volume of 12.5 mL. Genomic DNA (obtained using any commercially available kit) is diluted in water at 10 ng/mL. The reaction mixtures are prepared on ice.
Allele-Specific PCR in SNP Genotyping
421
1. Prepare a PCR mix with 1.4 mL of 10X PCR buffer (1X in the PCR), 0.4 mL of 25 mM MgCl2 (final concentration 2.25 mM), 1.4 mL of 2 mM dNTP (final concentration 0.2 mM in each), 0.3 mL of 10 mM concentration of each primer (final concentration 0.25 mM), 0.1 mL of 5 U/mL Taq DNA polymerase (0.02 U/mL), and 7.8 mL of water (see Note 8). 2. To 1.5 mL of DNA (15 ng) add 11 mL of the PCR mix. Mix well and spin down. The samples are ready for the PCR. 3. Perform the PCR. The amplification is carried out using standard PCR conditions. The annealing temperature depends on the primers. There are several formulas to calculate Tm (20, 21). The formula chosen has to be the same for all primers (see Note 9). We conducted the PCR with 3 min of denaturation at 94C, 35 cycles of 94C for 20 s, 50C (annealing) for 30 s, 72C for 15 s, and a final extension at 72C for 4 min. 3.3. Gel Electrophoresis 3.3.1. Gel Preparation
The amplification products are separated on a 3% MetaPhor1 agarose gel prepared in 1X TBE buffer and run in 0.5X TBE buffer. The gel and the buffer should be cooled to 4C (see Note 10). 1. In a beaker that is 2–4 times the volume of the agarose solution (required to prepare the gel), add the electrophoresis buffer (1X TBE buffer) and a stir bar. Rapidly stir the solution. 2. Slowly sprinkle the agarose powder on the TBE solution, and soak the agarose in the buffer for 10 min. This reduces the tendency of the agarose solution to foam during heating. 3. Cover the beaker with a plastic wrap and pierce it to form a small hole for ventilation. Heat the solution at 150C while stirring. Hold the solution at the boiling point until all of the particles have dissolved. 4. Cool the solution to 50–60C and cast the gel at room temperature. When the agarose has solidified, place the gel at 4C and keep it for at least 20 min (see Note 10).
3.3.2. Sample Preparation and Electrophoresis
1. Add 2 mL of the gel-loading buffer (6X) to the 12.5 mL of each PCR product. 2. Load 4 mL of each sample, and 2 mL of the DNA size standard on a gel to check the size of the PCR fragments. 3. Run the gel at 6 V/cm constant voltage for about 2 h.
3.3.3. Gel Staining
1. Immerse the gel in ethidium bromide solution for 15 min at room temperature in the dark (see Note 4). 2. Rinse the gel with water for 15 min at room temperature in the dark. 3. Visualize the products by illumination with 300-nm UV light (see Note 11).
422
Gaudet et al.
4. Notes 1. Water used for the preparation of dNTP mix is sterile distilled Milli-Q water, 0.2-mm filtered. Milli-Q water 0.2-mm filtered should have a resistivity of about 18.2 M cm and is free of nucleic acid, RNase, and DNase. In the text, the term ‘‘water’’ refers to this type of water. 2. MetaPhor1 agarose is a high-resolution gel agarose with twice the resolution capabilities of the finest-sieving agarose products. It challenges polyacrylamide gel with the ease of preparation of standard agarose gel. 3. TBE 5X stock solution is more stable than 10X stock because the solutes do not precipitate during storage. Passing the concentrated buffer stocks through a 0.45-mm filter and storing the buffer in a plastic bottle can prevent or delay formation of precipitates (22). 4. Ethidium bromide is a powerful mutagen and is toxic. Wear gloves, a laboratory coat, and safety glasses. Handle with care. Contaminated waste should be treated separately. 5. MetaPhor1 agarose gel has the best resolution (up to two bases) for fragments in a range of 100–200-bp length. 6. If the mismatches are at the same position in the two AS primer, the base must be different. If the two AS primers have a mismatch with the same nucleotide in the same position, they could lose their discrimination efficiency. In fact, in the first PCR cycle, the two AS primers generate the expected AS-PCR fragments, which incorporate the mismatches. If the mismatches added in the AS primers are the same, the two PCR fragments are identical near the SNP. Therefore, in the successive PCR cycles, the two amplified alleles differ only by the base of the SNP, which allows cross-amplification and indifferent generation of the two alternative PCR fragments (Fig. 26.2). 7. On one hand, the appropriate choice of the additional mismatch and tail could help to limit stable hairpin loops and dimer formation, especially in the 30 -end region. On the other hand, the SNP position forces the choice of the sequence of the AS primers and it is sometimes difficult to completely avoid the formation of secondary structures. Nevertheless, these primers are used for qualitative genotyping. Therefore, the presence of dimers in the PCR is acceptable if the main fragments are amplified and readable. The objective is to obtain the primers which work.
Allele-Specific PCR in SNP Genotyping
423
8. The volumes indicated in the protocol are for one sample and include the pipetting error (estimated to be about 10%). Therefore, the PCR mix includes an extra volume and could be prepared for the exact number of samples. 9. With more than 20 PCR cycles, the primers work efficiently over a large range of temperatures (23). We have calculated Tm using the following simple formula: Tm (in C) ¼ 2(AþT) þ 4(GþC); the lowest Tm of the three primers is chosen as the annealing temperature in the PCR. Tm of the AS primers is calculated without the tails. 10. The agarose gel and the electrophoresis buffer (TBE 0.5X) are cooled to 4C to obtain the optimal gel resolution. 11. UV radiation is dangerous for the eyes, so wear appropriate safety glasses. References 1. Soleimani, V. D., Baum, B. R. and Johnson, D. A. (2003) Efficient validation of single nucleotide polymorphisms in plants by allele-specific PCR, with an example from barley. Plant Mol. Biol. Rep. 21, 281–288. 2. Gut, I. G. (2001) Automation in genotyping of single nucleotide polymorphisms. Hum. Mutat. 17, 475–492. 3. Gupta, P. K., Roy, J. K. and Prasad, M. (2001) Single nucleotide polymorphisms: A new paradigm for molecular marker technology and DNA polymorphism detection with emphasis on their use in plants. Curr. Sci. India 80, 524–535. 4. Bundock, P. C., Cross, M. J., Shapter, F. M. and Henry, R. J. (2005) Robust allele-specific polymerase chain reaction markers developed for single nucleotide polymorphisms in expressed barley sequences. Theor. Appl. Genet. 112, 358–365. 5. Kim, M. Y., Van, K., Lestari, P., Moon, J. K. and Lee, S.H. (2005) SNP identification and SNAP marker development for a GmNARK gene controlling supernodulation in soybean. Theor. Appl. Genet. 110, 1003–1010. 6. Ahmadian, A., Gharizadeh, B., O’Meara, D., Odeberg, J. and Lundeberg, J. (2001) Genotyping by apyrase-mediated allele-specific extension. Nucleic Acids Res. 29, e121. 7. Dutton, C. and Sommer, S. S. (1991) Simultaneous detection of multiple singlebase alleles at a polymorphic site. BioTechniques 11, 700–7002. 8. Okimoto, R. and Dodgson, J.B. (1996) Improved PCR amplification of multiple specific alleles (PAMSA) using internally
9.
10.
11.
12.
13.
14.
mismatched primers. BioTechniques 21, 20–26. Germer, S. and Higuchi, R. (1999) SingleTube Genotyping without Oligonucleotide Probes. Genome Res. 9, 72–78. Hansson, B. and Kawabe, A. (2005) A simple method to score single nucleotide polymorphisms based on allele-specific PCR and primer-induced fragment-length variation. Mol. Ecol. Notes 5, 692–696. Hinten, G. N., Hale, M. C., Gratten, J., Mossman, J. A., Lowder, B. V., Mann, M. K. and Slate, J. (2007) SNP-SCALE: SNP scoring by colour and length exclusion. Mol. Ecol. Notes 7, 377–388. Wu, W. M., Tsai, H. J., Pang, J. H. S., Wang, H. S., Hong, H. S. and Lee, Y. S. (2005) Touchdown thermocycling program enables a robust single nucleotide polymorphism typing method based on allele-specific real-time polymerase chain reaction. Anal. Biochem. 339, 290–296. Ishiguro, A., Kubota, T., Soya, Y., Sasaki, H., Yagyu, O., Takarada, Y. and Iga, T. (2005) High-throughput detection of multiple genetic polymorphisms influencing drug metabolism with mismatch primers in allelespecific polymerase chain reaction. Anal. Biochem. 337, 256–261. Liu, Q., Thorland, E. C., Heit, J. A. and Sommer, S. S. (1997) Overlapping PCR for bidirectional PCR amplification of specific alleles: A rapid one-tube method for simultaneously differentiating homozygotes and heterozygotes. Genome Res. 7, 389–398.
424
Gaudet et al.
15. Sasvari-Szekely, M., Gerstner, A., Ronai, Z., Staub, M. and Guttman, A. (2000) Rapid genotyping of factor V Leiden mutation using single-tube bidirectional allele-specific amplification and automated ultrathin-layer agarose gel electrophoresis. Electrophoresis 21, 816–821. 16. Waterfall, C. M. and Cobb, B. D. (2001) Single tube genotyping of sickle cell anaemia using PCR-based SNP analysis. Nucleic Acids Res. 29, e119. 17. Waterfall, C. M. and Cobb, B. D. (2002) SNP genotyping using single-tube fluorescent bidirectional PCR. BioTechniques 33, 80–90. 18. Gaudet, M., Fara, A. G., Sabatti, M., Kuzminsky, E. and Mugnozza, G. (2007) Single-reaction for SNP genotyping on agarose gel by allele-specific PCR in black poplar (Populus nigra L.). Plant Mol. Biol. Rep. 25, 1–9. 19. Day, J. P., Bergstrom, D., Hammer, R. P. and Barany F. (1999) Nucleotide analogs
20.
21.
22.
23.
facilitate base conversion with 30 mismatch primers. Nucleic Acids Res. 27, 1810–1818. Sambrook, J. and Russell, D. W. (2001) Working with synthetic oligonucleotide probes. In Molecular cloning: a laboratory manual. 3 ed. New York: Cold Spring Harbor Laboratory; pp. 10.1–10.52. SantaLucia, J. J. (1998) A unified view of polymer, dumbbell, and oligonucleotide DNA nearest-neighbor thermodynamics. Proc. Natl. Acad. Sci. U.S.A. 95, 1460–1465. Sambrook, J. and Russell, D. W. (2001) Gel electrophoresis of DNA and pulsed-field agarose gel electrophoresis. In Molecular cloning: a laboratory manual. 3 ed. New York: Cold Spring Harbor Laboratory; pp. 5.1–5.9. Rychlik, W., Spencer, W. J. and Rhoads R.E. (1990) Optimization of the annealing temperature for DNA amplification in vitro. Nucleic Acids Res. 18, 6409–6412.
Chapter 27 Modified Multiple Primer Extension Method Toshihide Yanagawa and Hisashi Koga Abstract A multiple primer extension (MPEX) was originally developed for the hybridization, extension, and amplification of a DNA template on a planar substrate by Kinoshita et al. in 2006. Herein we present a modified MPEX method refined by our group for single nucleotide polymorphism (SNP) detection. In this method, hybridization and extension reactions are performed on a plastic S-BIO1 PrimeSurface1 substrate, with a biocompatible polymer. Its surface chemistry offers extraordinarily stable thermal properties, as well as chemical properties advantageous for enzymatic reactions on the surface. To visualize allele-specific PCR products on the surface, biotin–dUTP was incorporated into newly synthesized complementary strands during the extension reaction. The products were ultimately detected by carrying out a colorimetric reaction with a substrate solution containing 5-bromo-4-chloro-3-indolyl phosphate/nitro blue tetrazolium. We have further successfully combined this method with multiplex PCR. We demonstrate the advantages of this combined method by analyzing representative SNPs on different linkage disequilibrium blocks of the m opioid receptor gene (OPRM1), which is a marker gene for pain threshold. Key words: Single nucleotide polymorphisms, multiple primer extension, biotin, colorimetric, focused analysis, human m opioid receptor gene.
1. Introduction Recent advances in technology have enabled the establishment of many single nucleotide polymorphism (SNP) detection methods; however, many of these methods require specialized infrastructure and expensive equipment (e.g., for fluorescence signal detection) (1– 4). For use in institutions with limited financial resources, we have adopted and modified a multiple primer extension (MPEX) method which does not require such infrastructure. For detection
A.A. Komar (ed.), Single Nucleotide Polymorphisms, Methods in Molecular Biology 578, DOI 10.1007/978-1-60327-411-1_27, ª Humana Press, a part of Springer Science+Business Media, LLC 2003, 2009
425
426
Yanagawa and Koga
of SNPs using this method, visualized allele-specific PCR products could be recorded using a simple desktop personal imaging scanner (5). It ought to be emphasized that this method was developed as a result of a recent innovation in material science with special applicability to polymer materials. The new plastic S-BIO1 PrimeSurface1 is exploited in a solid planar substrate in the MPEX method because it has significant hybridization properties in terms of the surface chemistry applied to a cyclic olefin copolymer surface with the random copolymerization of 2-methacryloyloxyethyl phosphorylcholine, n-butyl methacrylate, and p-nitrophenyloxycarbonyl poly(ethylene glycol) methacrylate (6, 7). Although the protocol described in this chapter involves 20 oligonucleotides on a 7 mm 7 mm area, at least 50 oligonucleotides for different SNPs can be, in principle, spotted onto the same area. Therefore, this method seems to be suitable for highthroughput SNP analysis focused on a particular disease or physiological status.
2. Materials 2.1. DNA Extraction
1. Cotton swabs. 2. QIAamp1 DNA mini kit (QIAGEN, Valencia, CA, USA): Store at room temperature (15–25C).
2.2. Oligonucleotide Module Fabrication
1. S-Bio1 PrimeSurface1 (BS-11608) consists of cyclic olefin copolymer grafted with an original biocompatible phospholipid polymer, poly[2-methacryloyloxyethyl phosphorylcholine-co-n-butyl methacrylate-co-p-nitrophenyloxycarbonyl polyethyleneglycol methacrylate] hydrophilic polymer (Sumitomo Bakelite, Tokyo, Japan). 2. The oligonucleotide probes (see Note 1 and Table 27.1) designed to hybridize to allele-specific PCR products of, e.g., the m opioid receptor gene (OPRM1). 3. Spotting solution: 250 mM sodium carbonate buffer, pH 9.0. Store at room temperature. 4. BioChip Arrayer1 spotting robot (Filgen, Nagoya, Japan). 5. The oligonucleotide modules [gasket-type hybridization cassettes: one module consists of 16 (8 2 lanes) hybridization wells] (Sumitomo Bakelite). 6. Blocking buffer solution: 0.5 N NaOH. Store at room temperature.
Modified Multiple Primer Extension Method
427
Table 27.1 Immobilizedoligonucleotide probes for modified multiple primer extension Length (bp)
GC (%)
Tm (C)
TCA ACT TGT CCC ACT TAG ATG GCA
24
46
68.5
A118G_G_allele_forward
TCA ACT TGT CCC ACT TAG ATG GCG
24
50
69.9
A118G
A118G_A_allele_reverse
CGC ATG GGT CGG ACA GGT T
19
63
70.6
A118G
A118G_G_allele_reverse
CGC ATG GGT CGG ACA GGT C
19
68
71.5
C691G
C691G_C_allele_forward
TTA GCT CTG GTC AAG GCT AAA AAT C
25
40
64.4
C691G
C691G_G_allele_forward
TTA GCT CTG GTC AAG GCT AAA AAT G
25
40
65.1
C691G
C691G_C_allele_reverse
TGT TAA TAC TGC CAT TTT GCT CAT TG
26
35
66.3
C691G
C691G_G_allele_reverse
TGT TAA TAC TGC CAT TTT GCT CAT TC
26
35
65.5
G5953A
G5953A_A_allele_forward
AGT TTT TTT TAC AAG ACA TCT GTG GAA
27
30
63.6
G5953A
G5953A_G_allele_forward
AGT TTT TTT TAC AAG ACA TCT GTG GAG
27
33
63.3
G5953A
G5953A_A_allele_reverse
GAA ACA ATT ACT TTC CCA AAT TAA CTT
27
26
61.3
G5953A
G5953A_G_allele_reverse
GAA ACA ATT ACT TTC CCA AAT TAA CTC
27
30
61.8
A2109G
A2109G_A_allele_forward
GTT GAG AGC AAC TTT GTC TTC AAG TA
26
38
63.6
A2109G
A2109G_G_allele_forward
GTT GAG AGC AAC TTT GTC TTC AAG TG
26
42
66
A2109G
A2109G_A_allele_reverse
TTG TGG AAA AAG ATA GAT CAG GTC CT
26
38
65.6
A2109G
A2109G_G_allele_reverse
TTG TGG AAA AAG ATA GAT CAG GTC CC
26
42
67.7
Target SNP
Probe name
Sequence (50 !30 )
A118G
A118G_A_allele_forward
A118G
Polymorphisms are indicated in bold. SNP single nucleotide polymorphism
428
Yanagawa and Koga
2.3. Preparation of Template Multiplex PCR Products and Their Validation
1. Multiplex PCR Mix1 (TaKaRa Bio, Japan). Store at –30C. 2. The primer pairs (see Note 2 and Table 27.2) designed to amplify allele-specific PCR products of OPRM1.
Table 27.2 The primer pairs for amplification of allele-specific PCR products Forward primer (50 !30 )
Reverse primer (50 !30 )
Product size (bp)
A118G
AAT GTC AGA TGC TCA GCT CG
ACA TGA CCA GGA AGT TTC CG
464
C691G
CCT GAA CGA AAG CTT AAT GTG
TCT GGG ATT GTA ACC AAG GC
538
G5953A
CTG AAA TCT TAT TTA AAC TGT ACC
ATG TTA TTA TGA ATA GTC TAA AAG CC
862
A2109G
GAC ATA GTA TTG CTT TTG CT
AAA ATA ACT CCC TTT ATT GTG T
217
Target SNP
3. Thermal cycler: TaKaRa PCR thermal cycler Dice1 model TP600 (TaKaRa Bio, Japan). 4. Wizard1 SV 96 PCR cleanup system (Promega, Madison, WI, USA). Store at room temperature (22–25C). 5. EtOH (80%). Store at room temperature. 6. Agarose S (Nippon Gene, Tokyo, Japan). Store at room temperature. 7. 50X TAE buffer: 2 M tris(hydroxymethyl)aminomethane (Tris)–acetate, pH 8.3, 50 mM EDTA. To prepare the buffer, mix 242 g of Tris, 57.1 mL of glacial acetic acid, and 100 mL of 0.5 M EDTA (pH 8.0), and adjust the volume to 1,000 mL with water. Store at room temperature. 8. 100-bp DNA ladder (New England Biolabs, Ipswich, MA, USA). Store at –20C. 9. Ethidium bromide (Nippon Gene). Store at 4C. 2.4. Modified MPEX Reaction
1. dNTP set, 100 mM solutions (GE Healthcare), working solution 1 mM in each dNTP. Store at –30C. 2. HotStar TaqTM DNA polymerase (QIAGEN). Store at –30C. 3. Biotin-11-dUTP (PerkinElmer, Wellesley, MA, USA), working solution 1 mM. Store at –30C, and protect from extended exposure to light and minimize the freeze–thaw cycles.
Modified Multiple Primer Extension Method
429
4. 10X MPEX buffer A: 1% TritonX-100. Store at room temperature. 5. 2X MPEX buffer B: 0.1 M phosphate buffer, pH 7.0. Store at room temperature. 6. Hybridization oven: Hybaid Midi Dual-14 (Hybaid, Middlesex, UK). 2.5. Visualization by Colorimetric Reaction
1. Washing buffer A is TBS-T: 10 mM Tris–HCl, pH 7.6, 150 mM NaCl, 0.1% Tween 20. Store at room temperature. 2. Washing buffer B: 0.1% Tween 20. Store at room temperature. 3. Streptavidin–alkaline phosphatase (AP) (PerkinElmer). Store at –20C, store at 4C after thawing, do not refreeze this product. 4. 5-Bromo-4-chloro-3-indolyl phosphate/nitro blue tetrazolium (BCIP/NBT) substrate solution (PerkinElmer). Store at 2–8C. 5. Sigma 4K15 laboratory centrifuge (QIAGEN). 6. Rotating shaker: NR-3 double shaker (Taitec, Saitama, Japan). 7. GT-9700F personal image scanner (EPSON, Tokyo, Japan).
3. Methods MPEX was originally developed for the hybridization, extension, and amplification of a DNA template on a planar substrate (6). In our case, amplification was first carried out according to our SNP detection method, and hybridization and extension reactions were performed only using the substrate; we therefore referred to our method as a modified MPEX approach (Fig. 27.1). To determine the best experimental conditions sufficient for the visualization of multiple SNPs, we selected representative SNPs on four linkage disequilibrium blocks of the m opioid receptor gene (OPRM1) as targets (Fig. 27.2). We first adjusted the conditions for amplification of the template PCR products, and then determined the optimal one. Four DNA fragments spanning each SNP were successfully amplified to a similar extent under the PCR condition chosen (Fig. 27.3). We next adjusted the conditions for the subsequent MPEX reaction using these samples, and we finally determined the MPEX condition that allowed the genotyping (Fig. 27.4). We therefore applied the modified MPEX approach to 88 voluntarily accumulated human samples. Overall, the accuracy of the modified MPEX approach was validated at a maximum
430
Yanagawa and Koga
biotin
oligonucleotide probes
allelespecific PCR product
biotin
Extension & incorporation of biotin-dUTP
MEONP MPC BMA S-BIO® PrimeSurface®
Fig. 27.1. The modified multiple primer extension (MPEX) reaction on an S-BIO1 PrimeSurface1. The 50 -ends of C6-aminated DNA primers were covalently attached to the 2-methacryloyloxyethyl phosphorylcholine (MPC) surface of S-BIO1 via the active ester moiety(p-nitrophenyloxycarbonyl polyethyleneglycol methacrylate). The MPEX reaction proceeded in the presence of allele-specific PCR product, dNTP, and HotStar TaqTM DNA polymerase in a buffer on the S-BIO1 PrimeSurface1. BMA butyl methacrylate.
Fig. 27.2. Polymorphisms and estimated linkage disequilibrium (LD) blocks in OPRM1. Polymorphisms located in all exons, parts of the 50 and 30 flanking regions, and parts of the introns that have been previously found are indicated (8–13). The polymorphisms in OPRM1 are well linked, and four LD blocks are estimated to exist (lower horizontal bars). Representative single nucleotide polymorphisms (SNPs) on these four LD blocks used in this study are framed. del deletion, ins insertion, IVS intervening sequence.
431
2H
2G
2F
2E
2D
2C
2B
2A
1H
1G
1F
1E
1D
1C
1B
M ar ke r 1A
Modified Multiple Primer Extension Method
500 bp 400 bp 300 bp 200 bp 100 bp
Fig. 27.3. Representative multiplex PCR products used for the subsequent modified MPEX reaction. We extracted genomic DNA from 16 oral mucosa samples and amplified the template PCR products for the subsequent modified MPEX reaction. The multiplex PCR products were then purified using the Wizard1 SV 96 PCR cleanup system. Two microliters of purified PCR products was used for the agarose gel electrophoresis. Arrowheads indicate the PCR products for four different SNPs (the sizes of the PCR products were 862 bp for G5953A, 538 bp for C691G, 464 bp for A118G, and 217 bp for A2109G).
1
B
A Allele 1
A
Allele 2 600 µm
A118G
2
B
A2109G C691G
C
G5953A G8449A
D forward
E reverse A118G_A_F
A118G_A_R
A118G_G_F
A118G_G_R
A2109G_A_F
A2109G_A_R
A2109G_G_F
A2109G_G_R
C691G_C_F
C691G_C_R
C691G_G_F
C691G_G_R
G5953A_A_F
G5953A_A_R
G5953A_G_F
G5953A_G_R
G8449A_A_F
G8449A_A_R
G8449A_G_F
G8449A_G_R
F
G
H
Fig. 27.4. Oligonucleotide module layout and representative modified MPEX results. (a) Forward and reverse oligonucleotide pairs were spotted at a volume of 12.5 nL per spot using a BioChip Arrayer1 spotting robot. The names of the SNPs corresponding to the spotted oligonucleotides are given in the bottom column. (b) Overview of the oligonucleotide module. One module consists of 16 (8 2 lanes) gasket-type hybridization cassettes. Twenty oligonucleotide spots are located in a 7 mm 7 mm grid area, and the cassette forms a flow cell with 100 mL of MPEX reaction buffer. Hybridized and extended allele-specific DNA fragments were visualized with 5-bromo-4-chloro-3-indolyl phosphate/nitro blue tetrazolium substrate, and the subsequently appearing dark-purple spots were scanned on a GT-9700F personal image scanner. A combination of the indicated numbers (1–2 ) and letters (A–H ) was used to designate anonymous samples. The same numerical and alphabetical combination was used as in Fig. 27.3. We did not succeed in carrying out the reaction when we used an allele-specific primer for G8449A, owing to the AT-rich sequence encompassing the SNP. We thus excluded these results (dealing with G8449A) from the text and tables.
432
Yanagawa and Koga
of 99.4% (345/347), and thus was not inferior to the accuracy of the other SNP detection methods (sequence-specific primer cycle elongation fluorescence correlation spectroscopy and direct sequencing) (5). 3.1. DNA Extraction
1. Cotton swabs are used to obtain oral mucosa samples. 2. The oral mucosa samples adhering to cotton swabs were stored at –80C until use. 3. Genomic DNA was extracted from the oral mucosa samples using the QIAamp1 DNA mini kit according to the manufacturer’s instructions (http://bio.calpoly.edu/ubl/proto cols_files/qiaalu.pdf) (see Note 3). 4. The extracted DNAs were stored at –80C until use (see Note 4).
3.2. Oligonucleotide Module Fabrication
1. The oligonucleotide probes (see Note 1) were designed to hybridize to allele-specific PCR products of the OPRM1 gene and were dissolved in spotting solution at a final concentration of 0.2 mM. 2. The oligonucleotides were spotted (about 600 mm in diameter, about 12.5 nl per spot) on the surface of S-Bio1 PrimeSurface1 (BS-11608) using a BioChip Arrayer1 spotting robot. 3. The modules (gasket-type hybridization cassettes) were incubated overnight in a humid chamber with 250 mM sodium phosphate buffer at room temperature. 4. The excess amine-reactive group (MEONP) was inactivated for 5 min at room temperature in blocking buffer solution. 5. After the modules had been washed in boiling water for 2 min, they were washed in water at room temperature for 2 min and then dried by centrifugation for 2 min at 200g. 6. The oligonucleotide modules were stored in a desiccated state at 4C until use.
3.3. Preparation of Template Multiplex PCR Products and Their Validation
1. PCRs were carried out using Multiplex PCR Mix1 in a 20-mL total reaction volume, containing 10-mL Multiplex PCR Mix 1, 0.1 mL Multiplex PCR Mix 2, 1 mL template genomic DNA (0.5–19.75 ng), and an appropriate amount of each primer (5.0 mM G5953A, 5.0 mM A2109G, 1.0 mM C691G, 1.0 mM A118G:). The PCR was performed for 40 cycles which consisted of denaturation at 94C for 30 s, annealing at 54C for 90 s, and extension at 72C for 90 s. 2. After the PCR, the multiplex PCR products were purified with the Wizard1 SV 96 PCR cleanup system according to the manufacturer’s instructions (http://www.promega. com/tbs/tb311/tb311.pdf). An additional washing step, using 80% EtOH was also performed.
Modified Multiple Primer Extension Method
433
3. The amplified products were electrophoresed on 2.0% agarose gels and then visualized by ethidium bromide staining. 4. An 8/10 volume of purified PCR product was used further for the modified MPEX reaction. 3.4. Modified MPEX Reaction
1. After the template PCR products had been denatured at 95C for 20 min on a thermal cycler, the PCR products were subjected to an annealing reaction with immobilized oligonucleotides on the modules in a 100-mL reaction volume containing 1X PCR buffer supplied by QIAGEN (information regarding the components is not available), 0.1% TritonX-100, 0.04 mM in each dNTP, and 2.5 U HotStar TaqTM DNA polymerase. 2. The modules were rinsed with MPEX buffer A three times. 3. The samples were further incubated at 66C for 3 h, and biotin–dUTP was incorporated during the extension of the complementary strand. This reaction was performed in a hybridization oven, which should be prewarmed for at least 1 h before the reaction. To avoid a decrease in temperature, the modules were covered with aluminum foil. 4. The residual reaction mixture was removed by decantation, and then modules were agitated with 100 mL of MPEX buffer A for 1 min on an NR-3 double shaker (65 rpm; Taitec, Saitama, Japan). 5. After the removal of MPEX buffer A, the modules were further agitated with 100 mL of MPEX buffer B for 1 min. 6. MPEX buffer B was completely removed by decantation, followed by centrifugation at 2000 rpm for 2 min on a Sigma 4K15 laboratory centrifuge.
3.5. Visualization by Colorimetric Reaction
1. A working solution of streptavidin–AP should be prepared immediately before the reaction (mix 0.15 mL of streptavidin– AP, 15 mL of 10 X MPEX buffer A, 75 mL of 2 X MPEX buffer B, and add distilled water to 150 mL). 2. The working solution was added to the modules, and the samples were incubated at 37C for 10 min. During this reaction, the modules were covered with aluminum foil. 3. The residual reaction mixture was removed by decantation, and modules were then agitated with 100 mL of MPEX buffer A for 1 min on an NR-3 double shaker (65 rpm; Taitec, Saitama, Japan). 4. After removal of MPEX buffer A, the modules were further agitated with 100 mL of MPEX buffer B for 1 min. 5. MPEX buffer B was completely removed by decantation and 2,000 rpm centrifugation for 2 min on a Sigma 4K15 laboratory centrifuge.
434
Yanagawa and Koga
6. The colorimetric detection of the AP-labeled complementary strand was performed in BCIP/NBT substrate solution at 37C for 30 min. 7. The BCIP/NBT substrate solution was removed, and then the modules were agitated with 100 mL of distilled water for 1 min on an NR-3 double shaker (65 rpm). 8. For the purpose of recording the results, the modules were immediately dried by centrifugation for 2 min at 2,000 rpm on a Sigma 4K15 laboratory centrifuge. 9. The dark-purple stains were recorded (scanned) on a GT9700F personal image scanner. 10. The data were collected using the free software program EPSON TWAIN 5, available from the EPSON Web site (http:// www.epson.jp/dl_soft/list/1379.htm), and were then adjusted if necessary using Adobe1 Photoshop Elements version 4.0. The recommended resolution of the images is at least 600 dpi. 11. The presence of SNPs was assessed by visual inspection of the signal intensities.
4. Notes 1. The oligonucleotides used here were single-stranded 19–27mer 50 -C6-amino-oligonucleotides. The oligonucleotides were designed and synthesized by NovusGene (Tokyo, Japan) (see Table 27.1). 2. The primer pairs designed to amplify allele-specific PCR products of OPRM1. The primer pairs were designed and synthesized by NovusGene to span the SNPs (Table 27.2). 3. Minor modification of the manufacturer’s protocol. The manufacturer suggests the use of 400 or 600 mL of buffer; we always used 500 mL. 4. Using this method, we generally obtained 3–23 ng genomic DNA in a total volume of 150 mL.
Acknowledgments The authors gratefully acknowledge the technical assistance of Nobutake Suzuki and Junko Hasegawa, and the kind supply of the data for Fig. 27.2 by Kazutaka Ikeda and Soichiro Ide
Modified Multiple Primer Extension Method
435
(Director, Molecular Psychiatry Research, Tokyo Institute of Psychiatry). This study was supported by a research grant from the Ministry of Health, Labour, and Welfare of Japan (MHLW; H17pharmaco-001) and by a grant from the Kazusa DNA Research Institute to H.K. References 1. Romkes, M. and Buch, S. C. (2005) Genotyping technologies: application to biotransformation enzyme genetic polymorphism screening. Methods Mol. Biol. 291, 399–414. 2. Brennan, M. D. (2001) High throughput genotyping technologies for pharmacogenomics. Am. J. Pharmacogenomics 1, 295–302. 3. Bannai, M., Higuchi, K., Akesaka, T., Furukawa, M., Yamaoka, M., Sato, K. and Tokunaga, K. (2004) Single-nucleotide-polymorphism genotyping for whole-genomeamplified samples using automated fluorescence correlation spectroscopy. Anal. Biochem. 327, 215–221. 4. Hori, K., Shin, W. S., Hemmi, C., Toyooka, T. and Makino, T. (2003) High fidelity SNP genotyping using sequence-specific primer elongation and fluorescence correlation spectroscopy. Curr. Pharm. Biotechnol. 4, 477–484. 5. Imai, K., Ogai, Y., Nishizawa, D., Kasai, S., Ikeda, K. and Koga, H. (2007) A novel SNP detection technique utilizing a multiple primer extension (MPEX) on a phospholipid polymer-coated surface. Mol. Biosyst. 3, 547–553. 6. Kinoshita, K., Fujimoto, K., Yakabe, T., Saito, S., Hamaguchi, Y., Kikuchi, T., Nonaka, K., Murata, S., Masuda, D., Takada, W., Funaoka, S., Arai, S., Nakanishi, H., Yokoyama, K., Fujiwara, K. and Matsubara, K. (2007) Multiple primer extension by DNA polymerase on a novel plastic DNA array coated with a biocompatible polymer. Nucleic Acids Res. 35, e3. 7. Michikawa, Y., Fujimoto, K., Kinoshita, K., Kawai, S., Sugahara, K., Suga, T., Otsuka, Y., Fujiwara, K., Iwakawa, M. and Imai, T. (2006) Reliable and fast allele-specific extension of 30 -LNA modified oligonucleotides covalently immobilized on a plastic
8.
9.
10.
11.
12.
13.
base, combined with biotin-dUTP mediated optical detection. Anal. Sci. 22, 1537–1545. Shi, J., Hui, L., Xu, Y., Wang, F., Huang, W. and Hu, G. (2002) Sequence variations in the mu-opioid receptor gene (OPRM1) associated with human addiction to heroin. Hum. Mutat. 19, 459–460. Smolka, M., Sander, T., Schmidt, L. G., Samochowiec, J., Rommelspacher, H., Gscheidel, N., Wendel, B. and Hoehe, M. R. (1999) Mu-opioid receptor variants and dopaminergic sensitivity in alcohol withdrawal. Psychoneuroendocrinology 24, 629–638. Gelernter, J., Kranzler, H. and Cubells, J. (1999) Genetics of two mu opioid receptor gene (OPRM1) exon I polymorphisms: population studies, and allele frequencies in alcohol- and drug-dependent subjects. Mol Psychiatry 4, 476–483. Bergen, A. W., Kokoszka, J., Peterson, R., Long, J. C., Virkkunen, M., Linnoila, M. and Goldman, D. (1997) Mu opioid receptor gene variants: lack of association with alcohol dependence. Mol. Psychiatry 2, 490–494. Bond, C., LaForge, K. S., Tian, M., Melia, D., Zhang, S., Borg, L., Gong, J., Schluger, J., Strong, J. A., Leal, S. M., Tischfield, J. A., Kreek, M. J., and Yu, L. (1998) Singlenucleotide polymorphism in the human mu opioid receptor gene alters beta-endorphin binding and activity: possible implications for opiate addiction. Proc. Natl. Acad. Sci. U.S.A. 95, 9608–9613. Hoehe, M. R., Kopke, K., Wendel, B., Rohde, K., Flachmeier, C., Kidd, K. K., Berrettini, W. H. and Church, G. M. (2000) Sequence variability and candidate gene analysis in complex disease: association of mu opioid receptor gene variation with substance dependence. Hum. Mol. Genet. 9, 2895–2908.
Chapter 28 Detection of SNP by the Isothermal Smart Amplification Method Alexander Lezhava and Yoshihide Hayashizaki Abstract The complexity of molecular diagnostic assays is a significant barrier to employment at the point of care, despite the growing need for such technologies. We have developed a sensitive, accurate, rapid, and simple DNA amplification scheme that shows potential for translational medicine from pharmacogenomics-based drug discovery through to point-of-care diagnostics. Called the ‘‘Smart Amplification process 2’’ (SmartAmp 2), the method is isothermal, and employs a unique primer design and background suppression technology that can amplify target sequences from crude cell lysates. The specificity of the SmartAmp 2 assay enables detection of single-nucleotide differences such as somatic mutations in tumors and single nucleotide polymorphism (SNP) variants. Because mismatch amplification can be completely suppressed in SmartAmp 2, a reliable diagnostic result can be achieved based exclusively on amplification alone. Furthermore, mutation detection and SNP genotypes can be obtained in as little as 30–40 min from sample preparation of raw blood or tissue specimens. Key word: Single-nucleotide polymorphisms, Smart Amplification process 2, Aac DNA polymerase, DNA-mismatch-binding protein, competitive probe.
1. Introduction The standard protocol of medical care is shifting towards genomic medicine/pharmacogenomics/etc., supported by the human genome sequence (1,2) and genome diversity databases (3–5) (see also Chapter 3 in this volume). This new type of medicine is based on the accumulating knowledge of gene polymorphisms (singlenucleotide polymorphisms, SNPs) and their relationship to specific phenotypes, such as disease predisposition, drug metabolism, and disease development (see Chapters 1 and 2 in this volume). A.A. Komar (ed.), Single Nucleotide Polymorphisms, Methods in Molecular Biology 578, DOI 10.1007/978-1-60327-411-1_28, ª Humana Press, a part of Springer Science+Business Media, LLC 2003, 2009
437
438
Lezhava and Hayashizaki
Point-of-care molecular diagnostics is a rapidly growing field which is gradually becoming more user-friendly with the introduction of portable devices and shortening of the time of detection of nucleic acids and their variations (6). Successful point-of-care diagnostics require rapidness of the reaction, low cost, low energy consumption and simple analysis with no offinstrument processing or reagent preparation, the inclusion of reliable controls, and minimal technical training. SmartAmp 2, the Smart Amplification process 2, is a novel isothermal DNA amplification method for rapid detection of single-nucleotide differences in DNA (7). The method requires little to no DNA purification and displays extreme fidelity, sensitivity, and single-nucleotide precision, enabling rapid analysis of SNPs, as well as other mutations in DNA targets. Unlike other amplification methods, SmartAmp 2, when properly optimized, achieves complete suppression of the background noise (mismatch/background amplification). To accomplish this, we use a uniquely original primer design and the key enzymes DNA polymerase and mismatch-binding protein (see below). Since the background is eliminated, amplification becomes equivalent to detection, even if DNA targets differ by only a single nucleotide. Therefore, amplification as a tool for SNP detection constitutes the basic principle of SmartAmp. Compared with PCR, the yield of the amplified product by SmartAmp 2 is typically 100 times greater. Lysed blood and tissue samples can be used directly in the assay with no inhibition of amplification, yield, or specificity. From crude sample to diagnostic result, the process usually takes 20–60 min. The SmartAmp 2 reaction is typically monitored using real-time PCR instrumentation. Each assay is designed to contain a unique primer mix for a specific target. With an amplification process of absolute specificity, the detection can be observed with a nonspecific double-stranded DNA (dsDNA) intercalating fluorescent dye. The sensitivity and precision of this method for determination of SNPs and mutations is at least equivalent to, and usually exceeds, those of other diagnostic technologies. The unique design of the primers contributes to the specificity of SmartAmp (Figs. 28.1, 28.2 and 28.3). Although only two primers are critical for the amplification process, four or five primers are commonly employed to accelerate the process and enhance specificity. The primer design employs a strategy whereby single-stranded intermediate amplified molecules can self-hybridize and self-prime the synthesis of dsDNA. On the other hand, only the primer sequence can be used to optimize the reaction condition because the reaction is isothermal, unlike PCR, and utilizes a unique Aac DNA polymerase from Alicyclobacillus acidocaldarius. This is a bivalent feature of SmartAmp 2; its primer design allows much more flexibility than
Detection of SNP by the Isothermal Smart Amplification Method
439
A Primer mis-match
Primer exact match
Amplification
Mis-amplification
MutS MutS
Pol
Amplification
ttMutS
MutSbinds to mismatch
Pol
No amplification
B
Fig. 28.1. Suppression of mismatch amplification in SmartAmp. (a) The mismatch binding protein Taq MutS does not bind to fully matched double-stranded DNA, and hence through the action of Aac polymerase, amplification proceeds. All types of mismatches, including the loop-out of single-stranded DNA (e.g., microdeletions), are recognized by MutS. The MutS-mismatch DNA complex is highly stable, and cannot be dissociated by the displacement activity of Aac polymerase. Hence the amplification of mismatched DNA molecules is inhibited in SmartAmp. (b) Typical suppression of mismatch amplification signals.
traditional isothermal amplification methods (8–10). The DNA excision and repair pathway enzyme Taq MutS from Thermus aquaticus allows the detection of a mismatch (SNP) (Fig. 28.1). Another point of development here is the rapid
440
Lezhava and Hayashizaki
Fig. 28.2. SmartAmp primer design. The general design and critical features of primers for amplification by the SmartAmp process are illustrated. The 50 -ends of the turn-back primer (TP) and the folding primer (FP) define the unit amplicon length, which is usually 70–200 bp.
Fig. 28.3. SmartAmp process. The mechanism of DNA amplification in the SmartAmp reaction is shown. The center outlines the events leading to the generation of intermediate species IM1 and IM2. The outer edges outline the selfpriming events of pathways A and B, and the free primer hybridization and priming event that drives pathway C. The final products are concatenated DNA molecules of multimeric amplicons whose unit lengths are bounded by the ends of the TP and the FP.
pretreatment protocol for the samples, such as clinical samples, veterinary materials, and biological samples from industrial sources, that significantly decreases the time for analysis. Many different future applications can be expected for this technology. A rapid SNP test of each patient can be achieved during a consultation and the correct prescription can be given
Detection of SNP by the Isothermal Smart Amplification Method
441
for a patient before she or he goes home. Owing to these advantageous features of SmartAmp 2, various new fields of use will likely be developed in the future. This technology is cost- and energy-efficient partly because of the small reaction volume and isothermal reaction. The device employing this technology can therefore be miniaturized for use in the field or at bedside. This technology also requires a very small amount of reagent to cut the cost to a level where it will become universally available. Here we present a protocol for the use of the Smart Amplification process version 2 (SmartAmp 2).
2. Materials 2.1. DNA Samples
2.2. Enzymes
1. The minimum blood sample is 10 mL. This amount can be obtained with a finger prick. The blood sample is pretreated (see Section 3.1) and is directly used in the SmartAmp reaction. The enzymes used in the SmartAmp reaction are commercially available from DNAFORM (Japan). 1. Aac polymerase is a DNA polymerase from the thermophilic bacteria Alicyclobacillus acidocaldarius. The optimal in vitro polymerization temperature for this polymerase is 60–65C and the optimal pH is 8.0–8.2. The enzyme has high polymerization rates typical for other displacement polymerases from thermophiles and requires a magnesium ion concentration of 8 mM for optimal activity. The enzyme used in Smart Amp is a cloned large fragment of 610 amino acids (69.5 kDa) that also possesses displacement activity. 2. Thermus aquaticus MutS (Taq MutS) is a nuclear protein that belongs to a class of molecules involved in the DNA excision and repair pathway. This protein is an early factor for recognizing base-pair mismatches during DNA replication. It consists of 819 amino acids (molecular mass 91.4 kDa) and has been well characterized (11). It has three structural domains and four different functional activities that include (1) dsDNA binding activity, (2) ATPase activity, (3) MutS– MutS dimerization activity, and (4) base-mismatch recognition (see Notes 1 and 2).
2.3. SmartAmp Primer Set
The primer set and the process of amplification during the Smart Amp assay are illustrated in Figs. 28.2 and 28.3, respectively. Primers for this assay can be obtained from any commercial source. We used the ones supplied by Invitrogen. A typical
442
Lezhava and Hayashizaki
assay employs five primers: turn-back primer (TP), folding primer (FP), boost primer (BP), outer primer 1 (OP1), and outer primer 2 (OP2). 1. The TP consists of the 30 -end sequence (Ac’), complementary to the target sequence (A), and the 50 -end sequence (B’), which is complementary to the sequence (Bc) that is present on the same strand of DNA, but is located approximately 15–40 bp further downstream. Bc is complementary to sequence B, which is a genomic DNA target sequence on the strand to which the TP hybridizes. 2. The folding primer (FP) consists of the 3’-end sequence (Cc’), complementary to the target genome sequence (C), and the 5’-end sequence (Dc’), which is self-annealing and complementary to the intermediate region (D). In this nomenclature rule, (’) represents the sequence of the primer and the target genome sequences, shown as A, B, and C, indicate the genomic region where each primer binds. 3. Additional primers are employed in SmartAmp: a set of the so-called outer primers (OP1 and OP2), which strand-displace the DNA synthesized from FP and TP extension reaction, and a BP, which increases the amplification speed. The efficiency of the reaction has minimum dependence on the BP position and changes very little when BP is positioned within the loop or in other regions of the target sequence. 4. The genomic sequences between and including the TP and the FP define the target region to be amplified by the SmartAmp reaction. 2.4. Other Reagents and Equipment for SmartAmp
1. dNTPs (available from any commercial source). 2. Dimethyl sulfoxide (DMSO) (available from any commercial source). 3. 10x SmartAmp buffer: 200 mM tris(hydroxymethyl)aminomethane (Tris)–HCl pH 8.0, 100 mM KCl, 100 mM (NH4)2SO4, 80 mM MgSO4, 0.1% Tween20. 4. SYBR1 Green I (Molecular Probes, Invitrogen). 5. Real-time PCR Mx3000P system (Agilent Technologies), Lightcycler 480 (Roche), and other compatible machines.
3. Methods 3.1. SmartAmp DNA Samples
1. For direct SmartAmp reactions, blood samples are diluted 4-fold with 40 mM NaOH and heated at 98C for 3 min. They are ready now to be used directly in the SmartAmp assay.
Detection of SNP by the Isothermal Smart Amplification Method
3.2. Designing Primer Sets for the SmartAmp Assay
443
Establishing primer sets for the SmartAmp assays requires a minimum of 120–150 bases of sequence information (see Note 3). Software is available that assists the design process (see Note 4). The DNA sequence targeted in primer selection depends on the design objectives of the assay (see Note 4). Primers should be designed to avoid the formation of intermolecular and intramolecular secondary structures as followed in the design of PCR primers. We have been successful in designing primers with a G+C content from 40 to 60%. However, the amplicon lengths should be less than 200 bp to achieve the highest efficiency/fastest rates of amplification (see Note 5). 1. To design assays for SNP detection, we determine the sequence (60–75 bases) immediately upstream and downstream of the SNP of interest. However, there are more options if the diagnostic assays is designed to detect foreign sequences in the background of the host genomic DNA (see Notes 3 and 4). 2. Amplicon size in a SmartAmp reaction is the distance between the 30 -ends of the TP and the FP, as well as the size of the TP and the FP. For SNP detection, an amplicon size of about 100 bp is ideal. However, in other cases, depending on sequence specificity, an amplicon size of 100–200 bp can be used (see Note 5). 3. The recommended melting temperature and distance between primer regions for each primer are illustrated in Fig. 28.2.
3.2.1. Discrimination Primer and Discrimination Nucleotide Positioning
For the primer design software (see Note 4), it is necessary to choose where to position the discrimination nucleotide and how to input the sequence information. 1. This sequence may be positioned on the FP, the TP, or the BP, and hence that primer is referred to as the discrimination primer (DP). On any DP, there are several possibilities for where to incorporate the discrimination nucleotide relative to the ends. 2. Table 28.1 summarizes the primers that can be used as the DP and the optional positions for the discrimination nucleotide. The SmartAmp primer design software will examine and design possible primer sets that target the discrimination nucleotide on both strands of the target DNA.
3.3. SmartAmp Assay Reaction
1. Twenty five microliters of the SmartAmp reaction mixture typically contains 1.6 mM FP, 1.6 mM TP, 0.2 mM OP1, 0.2 mM OP2, 0.8 mM BP, 1,400 mM dNTPs, 6% DMSO,
444
Lezhava and Hayashizaki
Table 28.1 Optional positions for the discrimination nucleotide Position
TP
FP
BP
30 -end (n)
X
X
X
30 -end (n-1)
X
X
X
30 -end (n-2)
X
X
X
30 -end (n-3)
X
X
X
50 -end (n)
X
50 -end (n-1)
X
50 -end (n-2)
X
50 -end (n-3)
X
TP turn-back primer, FP folding primer, BP boost primer
20 mM Tris–HCl pH 8.0, 10 mM KCl, 10 mM (NH4)2SO4, 8 mM MgSO4, 0.1% Tween20, 1/100,000 diluted original SYBR1 Green I, 20 units of Aac DNA polymerase, 3 mg of Taq MutS, and 1 mL of heat denatured blood sample (prepared as in Section 3.1). 2. SmartAmp reactions are performed at 60C. 3. In the first step, the FP and the TP hybridize to the template genomic DNA (Fig. 28.3). Both products primed and synthesized from the FP and the TP are then dissociated from the template genomic DNA by the strand-displacement activity of the DNA polymerase primed from the flanking outer primer regions adjacent to the 50 -end of the target sequence. 4. Additional single-stranded displaced DNA products designated IM1 and IM2 are then generated by FP-primed synthesis on the TP-synthesized template and vice versa. These intermediate products (IM1 and IM2) are milestone products for the subsequent amplification steps. IM1 carries the TP sequence at the 50 -end and the FP complementary sequence at the 30 -end. IM2 has the TP complementary sequence at the 30 -end and the FP sequence at the 50 -end. The initial selfpriming elongation site on IM1 is the 30 -end of the FP sequence of IM1; this priming event also occurs later on other more mature products. 5. Concatenated products of IM1 are synthesized by an elongation process named ‘‘pathway A’’ (Fig. 28.3). For the products in pathway A, both 50 - and 30 -ends carry the TP and its
Detection of SNP by the Isothermal Smart Amplification Method
445
complementary sequence, forming long, double-stranded hairpin DNA. The initial self-priming elongation site on IM2 is located at the 30 -end of the TP sequence of IM2; this priming also occurs later on other more mature products. 6. Long concatenated DNA products are synthesized in the same way as in pathway A, but the synthesis process is denoted as ‘‘pathway B,’’ because the end product is different (Fig. 28.3). These long DNA products carry the FP and its complementary sequence at both the 50 -end and the 30 -end. 7. In the SmartAmp assay there is one other elongation event, defined as ‘‘pathway C’’ (Fig. 28.3). The elongation of pathway C starts from the 30 -end of a free TP that hybridizes to the looping structure of the TP complementary sequence, which is located at the intermediate region of the long products from pathway A. 8. Ladder patterns typical of SmartAmp-amplified DNA are shown in Fig. 28.4. In the example shown, the DNA band indicated by the asterisk is approximately 60 bp in length, likely corresponding to the monomeric self-
Fig. 28.4. Gel electrophoresis separation of DNA products after SmartAmp amplification. Agarose gel (5%) stained with ethidium bromide. Lane M, a 20-bp DNA ladder; lane 1, an aliquot of a SmartAmp reaction; lane 2, no primer control. The asterisk indicates a DNA band about 60 bp in size, which corresponds to the expected amplicon length of a monomer in this reaction. The DNA banding pattern supports the general scheme of selfprimed amplification that has been described (7).
446
Lezhava and Hayashizaki
primed amplification product bounded by the 50 -ends of the FP and the TP. Other larger bands are alternative forms (derived from IM1 or IM2) and self-hybridized hairpin species generating large DNA polymers. 9. The course of the reaction is monitored following the increase in fluorescence of the intercalating dye. 3.3.1. SmartAmp Assay Optimization
Trial and evaluation is the only way to select the best performing SmartAmp primer set. The general process is illustrated in Fig. 28.5. The steps for SmartAmp assay optimization are described below: 1. Employ a primer design software tool and generate candidate primer sets (see Note 4). 2. Synthesize (see Note 6) and test candidate sets without Taq MutS. (a) Start with FP and TP only (include BP if it is a DP). (b) Test with all primers (include OP1, OP2, BP).
Fig. 28.5. SmartAmp assay primer optimization strategy. Beginning with the software design tool, primer sets are selected and optimized for amplification speed and efficiency on full-match templates. Optimization of specificity and mismatch suppression is the next stage and this requires the use of mismatch templates and Taq MutS. These cycles are repeated as necessary, until the ideal primer set is generated. Finally, sensitivity is checked and sample preparation options are explored to minimize complexity but maximize sensitivity.
Detection of SNP by the Isothermal Smart Amplification Method
447
3. Select the best performers on the basis of the speed of the reaction and yield. 4. Optimize the design for speed and yield (if necessary) – return to step 2. 5. Test the top performers with Taq MutS for specificity (mismatch suppression) against mismatch templates. 6. Return to step 4, and optimize for specificity if necessary. 7. Test the top performers for sensitivity on full-match (target) templates. 8. Examine and optimize various sample preparations using clinical samples.
3.4. SNP Typing of ALDH2, Background Suppression with Taq MutS
To demonstrate the amplification efficiency and specificity of SmartAmp, we used the SNP (G1543A) of aldehyde dehydrogenase 2 (ALDH2) (the gene involved in alcohol sensitivity (13)), as a test. On a blood sample from an individual known to be homozygous wild type to ALDH2 as determined by PCR sequencing (data not shown), amplification occurred when we used a wildtype primer, BP(W), both in the presence of Taq MutS and in its absence (Fig. 28.6). When using the SNP-variant primer, BP(M), on the wild-type blood sample, we observed misamplification of the wild-type target after 25 min in the absence of Taq MutS, but the amplification was completely suppressed by inclusion of Taq MutS in the reaction, with no signal observed even after 255 min of the reaction (Fig. 28.6). These data indicated that Taq MutS was capable of completely suppressing the mismatch background amplification. 1. Reactions were performed in 25-mL total reaction volume as described above. 2. Enzymatic components (Aac DNA polymerase and Taq MutS) for the SmartAmp assays were supplied by DNAFORM in a prototype SmartAmp core kit. 3. In the case of SNP (G1543A) in ALDH2, each reaction contained 3.2 mM FP, 3.2 mM TP, 0.4 mM OP1, 0.4 mM OP2, 1.6 mM BP, 1.4 mM dNTPs, 5% DMSO, 20 mM Tris–HCl, pH 8.8, 10 mM KCl, 10 mM (NH4)2SO4, 8 mM MgSO4, 0.1% Tween20, 1/100,000 diluted original SYBR Green I, 40 units of Aac DNA polymerase, 1.5–2.4 mg of Taq MutS, and 1 ml of prepared blood sample. 4. The reaction was performed at 60C. We believe this method will be become a valuable tool for rapid and cost-effective SNP detection.
448
Lezhava and Hayashizaki
A 5’ACAAGATGTCGGGGAGTGGCCGGGAGTTGGGCGAGTACGGGCTGCAGGCATACACTGAAGTGAAA ACT GTGAGTGTGGGACCTGCTGGGGGCTCAGG -3‘
B ALDH2 TP CGAGTACGGGCCCACACTCACAGTTTTCAC ALDH2 FP TTTATATATATATAAACCGGGAGTTGGGCGAG ALDH2 BP(w)
GCAGGCATACACTGA
ALDH2 BP(m)
GCAGGCATACACTAA
ALDH2 OP1
ACAAGATGTCGGGGAGTG
ALDH2 OP2
CCTGAGCCCCCAGCAGGT
C
D
Fig. 28.6. Single nucleotide polymorphism (SNP) typing of the ALDH2 allele using the SmartAmp assay. (a) Sequence of the ALDH2 gene. The arrowhead indicates the position of the SNP nucleotides of the ALDH2 allele (G1543A). DNA sequences used for primer design are marked. (b) Primer set for SNP typing of ALDH2. BP(W) and BP(M) were designed to be wild-type allele-specific and mutant allele-specific, respectively. The arrowheads indicate the nucleotides corresponding to the SNP site. (c) The effect of Taq MutS on SNP detection of the ALDH2 gene. The intensity of fluorescence of SYBR Green I dye monitored during the SmartAmp reaction with or without Taq MutS. The blood sample of a homozygous wildtype DNA was used as a template for all reactions. (d) Amplification time course of the SmartAmp reaction with ALDH2 allele-specific primers using human blood specimens. Three possible diploid genotypes of ALDH2 alleles are shown. The left graph, the center graph, and the right graph show time courses for a homozygous wild type, a heterozygote, and a homozygous mutant, respectively. dR baseline-subtracted fluorescence reading.
Detection of SNP by the Isothermal Smart Amplification Method
449
4. Notes 1. A strand-displacement DNA polymerase and DNA-mismatchbinding protein (in addition to a unique primer design/set) contribute to the remarkable specificity of the SmartAmp 2 assay. In the amplification process, primers first hybridize to a denatured target template. The optimal temperature for SmartAmp is 60C, at which hybridization is fairly stringent. However, similarly to PCR, mismatch hybridization and primer extension can occur even under stringent conditions. An allele-specific primer (e.g., SNP) hybridizing to a wild-type template is an example of a mismatch hybridization event that can lead to amplification of background products. However, in SmartAmp, when mismatch hybridization and extension occurs, the mismatched base pairs are recognized by and subsequently form a stable complex with the mismatchbinding protein. The strand-displacement activity of the Aac DNA polymerase used in SmartAmp is not sufficient to dissociate the complex, and consequently amplification of the mismatched molecules is inhibited (Fig. 28.1a). 2. Taq MutS usually requires six or seven bases of dsDNA on either side of a mismatch to efficiently recognize and bind single-base mismatched DNA (Table 28.2). MutS from other thermophiles has also been cloned, isolated, and characterized (12,13) and can be employed in the assay. 3. Primer extension-based SNP detection systems usually require the SNP detection nucleotide(s) to be engineered precisely at the 30 -end of a specific primer; however, SmartAmp does not possess this limitation and thus there is a far greater versatility in its ability to detect almost any target sequence. Some recommendations are outlined in this protocol regarding the primer design process; however, the design options are numerous and the primer design flexibility is unrivaled. 4. It is most convenient to use the SMAP (SmartAmp) primer design software program available on the Web at http:// www.smapDNA.com (or a Windows compatible desktop version). Initial candidate primer sets can be generated with this program. Depending on how much sequence information is loaded and the sequence itself, we suggest you use anywhere from zero to five candidate primer sets. SMAP version 1.1 (Web) interfaces with a modification of Primer3, which is a commonly used freeware PCR primer design software program on the MIT Web site (http://frodo.wi.mit.edu/cgi-bin/ primer3/primer3_www.cgi). SMAP version 1.1 cannot rank-order the candidate primer sets according to expected
450
Lezhava and Hayashizaki
Table 28.2 Rank order of nucleotide mismatch recognition affinity by Taq MutS Rank
Very higha
Mismatch
(del)
High G:T
C:T
Medium A:G
T:T
A:C
A:A
Low C:C
G:G
a The affinity of Taq MutS for various mismatched DNAs is generally determined by gel-mobility shift assays on hybridized synthetic oligonucleotides to which various kinds of mismatches were specifically engineered. The order provided is a general guideline for the observed performance in the SmartAmp assay from highest affinity to lowest affinity. represents any microdeletion or insertion (one to about 50 bases) resulting in a loop structure.
performance in the SmartAmp assay. Therefore, performance must be determined empirically through trial and evaluation. 5. Amplicon sizes exceeding 200 bp begin to slow the amplification process considerably and should be avoided. 6. Recommended concentrations of primers used in the SmartAmp assay are shown in Table 28.3.
Table 28.3 Recommended concentrations of primers used in the SmartAmp assay Primer
Recommended concentration (mM)
Working range (mM)
TP
2
1–3
BP
1
0.5–3
FP
2
1–3
OP
0.25
0.1–0.5
OP outer primer
Acknowledgments This study was supported by research grants from the RIKEN Genome Exploration Research Project from the Ministry of Education, Culture, Sports, Science and Technology of the Japan (MEXT) to Y.H. References 1. Venter, J. C., Adams, M. D., Myers, E. W., Li, P. W. et al. (2001) The sequence of the human genome. Science 291, 1304–1351.
2. Lander, E. S., Linton, L. M., Birren, B. et al. (2001) Initial sequencing and analysis of the human genome. Nature 409, 860–921.
Detection of SNP by the Isothermal Smart Amplification Method 3. Miller, R. D., Phillips, M. S., Jo, I. et al. (2005) High-density single-nucleotide polymorphism maps of the human genome. Genomics 86, 117–126. 4. Human Genome Sequencing Consortium. (2004) Finishing the euchromatic sequence of the human genome. Nature 431, 931–945. 5. Sachidanandam, R., Weissman, D., Schmidt, S. C. et al. (2001) A map of human genome sequence variation containing 1.42 million single nucleotide polymorphisms. Nature 409, 928–933. 6. Holland, C. A. and Kiechle, F. L. (2005) Point-of-care molecular diagnostic systems – past, present and future. Curr. Opin. Microbiol. 8, 504–509. 7. Mitani, Y., Lezhava, A., Kawai, Y., Kikuchi, T., Oguchi-Katayama, A., Kogo, Y. et al. (2007) Rapid SNP diagnostics using asymmetric isothermal amplification and a novel mismatch suppression technology. Nat. Methods 4, 257–262. 8. Berard, C., Cazalis, M. A., Leissner, P. and Mougin, B. (2004) DNA nucleic acid sequence based amplification-based genotyping for polymorphism analysis. Biotechniques 37, 680–686.
451
9. Pickering, J., Bamford, A., Godbole, V., Briggs, J., Scozzafava, G., Roe, P., Wheeler, C., Ghouze, F. and Cuss, S. (2002) Integration of DNA ligation and rolling circle amplification for the homogeneous, endpoint detection of single nucleotide polymorphisms. Nucleic Acids Res. 30, e60. 10. Wang, S. S., Thornton, K., Kuhn, A. M., Nadeau, J. G. and Hellyer, T. J. (2003) Homogeneous real-time detection of single-nucleotide polymorphisms by strand displacement amplification on the BD ProbeTec ET system. Clin. Chem. 49, 1599–1607. 11. Takamatsu, S., Kato, R., and Kuramitsu, S. (1996) Mismatch DNA recognition protein from an extremely thermophilic bacterium, Thermus thermophilus HB8. Nucleic Acids Res. 24, 640–647. 12. Biswas, I. and Hsieh, P. (1996) Identification and characterization of a thermostable MutS homolog from Thermus aquaticus. J. Biol. Chem. 271, 5040–5048. 13. Whitehouse, A., Deeble, J., Parmar, R., Taylor, G. R., Markham, A. F. and Meredith, D. M. (1997) Analysis of the mismatch and insertion/deletion binding properties of Thermus thermophilus, HB8, MutS. Biochem. Biophys. Res. Commun. 233, 834–837.
SUBJECT INDEX A ABCC2 (gene).................................................................. 27 ABC transporters........................................................ 27, 30 ABI 7900 HT (instrument).......294, 295, 299, 301, 303 n1 ABI Prism 3100 Genetic Analyzer ...................195, 205 n1 AB SNPlex (genotyping system) ............54, 247, 250 t, 251 AB TaqMan (genotyping system) ..................54, 247, 250 t Acquisition bias .....................................................51, 69 n9 Acyl-CoA dehydrogenase (MCAD)................................ 26 Adducts ............... 224, 321, 326, 340 n9, 341 n10, 341 n13 Affymetrix genotyping arrays .............................................. 105, 106 human SNP array .....................................0, 6, 246, 252 SNP microarray ................................................ 246, 248 AFLP (Amplified Fragment Length Polymorphisms) ... 76, 85, 86 AlanyltRNA synthetase .................................................... 28 Alcohol sensitivity........................................................... 447 Aldehyde dehydrogenase 2 (ALDH2) (gene) ..........................................447–448, 448 f Alicyclobacillus acidocaldarius .................................... 438, 441 Aligned sequence fragments ............................................. 74 Alignment multiple ........................................................... 78, 83, 86 views............................................................................ 53 Alistair Brown................................................................... 30 Allele frequency..........51, 56, 63, 66, 86, 128, 194, 194 f, 201, 205, 206, 273 n8, 303 n5, 313, 327 polymorphic .............................................................. 193 -specific amplification ..................................187, 190 n1, 415 PCR, see also Allele-specific (AS) PCR (AS-PCR) Allele-specific (AS) PCR (AS-PCR) MetaPhor1 agarose gel .................. 418 f, 420 f, 422 n5 primer(s) inner .................................................................... 416 outer ..................................................416, 442, 450 t tailed.................................................................... 415 reaction......................................185, 186, 187, 415–423 A-allele-specific (primer).............................................. 188 f Allelotyping ...............................................313, 327, 340 n8 American Association for Clinical Chemistry ...176 t, 179 f Aminoacid-tRNAs ........................................................... 24
Amplification refractory mutation (ARMS), see Allele-specific (AS) PCR (AS-PCR) Analytical thin-layer chromatography............................ 349 Anfinsen’s principle .................................................... 29, 35 Annealing temperature (Ta) ........180 n6, 187, 267, 272 n4, 375, 383, 386, 389 n5, 389 n7, 412 n7, 421, 423 n9 Anton Komar.................................................................... 30 Archon Genomics X Prize................................................ 96 Arrays digital ................................................................ 281, 287 dynamic...... 281, 282, 283, 285, 285 f, 286, 287, 289 t, 289 n1, 290 n6 Aspirate .................................129 n3, 129 n9, 232, 297, 298 Assays-on-Demand ............. 44 t, 46, 49, 50, 51, 295, 299, 300, 303 n5 Assays-by-Design ..............................295, 299, 300, 303 n6 Association studies candidate gene case-control ...................................... 293 case-control ....................................................... 293, 294 Asymmetric PCR Amplification ............................ 211, 217 ATM (Ataxia Telangiectasia Mutated) gene ................... 65 Atomically flat gold .................117, 118, 119, 120 f, 121 n9 ATP-binding cassette (ABC) transporter .................. 27, 30 ATP-dependent efflux...................................................... 27 ATP-driven efflux pump .................................................. 30 Atrophy of the globe........................................................... 5 Auditory function ............................................................... 9 Autism............................................................................... 26 AutoDimer (tool)............................................................ 382 Automatic allele calling ........................................302, 304 f AutoSNP (software) .....................................82, 84–85, 87 t Autosomal...............................................10, 11, 12, 14 f, 17 Average pain responsiveness haplotype (APS)........ 28, 140, 144, 148 n6, 155, 159, 162 n6
B BAC (bacterial artificial chromosome)..................... 76, 100 Backbone validated SNPs ........................................... 53–54 Base calling quality........................................................ 74, 76–77, 84 scores ..................................................................... 76, 77 software ................................................................. 76, 97 Bayesian (statistical method) .................................... 82, 273
A.A. Komar (ed.), Single Nucleotide Polymorphisms, Methods in Molecular Biology 578, 10.1007/978-1-60327-411-1, Ó 2009 Humana Press, a part of Springer Science+Business Media, LLC 2003, 2009
453
SINGLE NUCLEOTIDE POLYMORPHISMS
454 Subject Index
Bidirectional-PASA (bi-PASA), see Allele-specific (AS) PCR (AS-PCR) Bifurcation plots ............................................................... 56 BioMark (company)................................ 247, 281, 282, 286 Biotin-dUTP ........................................................430 f, 433 Biotrove (company) ......................................247, 250 t, 281 Biplexing ......................................................................... 322 BLAST ......................................................53, 68 n5, 78, 83 result(s)........................................................................ 53 BLAT ............................................................................... 78 Blindness.........................................................5, 8, 9, 11, 12 bilateral................................................................ 5, 9, 11 congenital.......................................................... 5, 9, 226 Blood peripheral .................................................. 172, 185, 196 sample(s) .........................395, 441, 442, 444, 447, 448 f transfusion................................................................... 15 type.............................................................................. 15 Bone (mineral density)...................................................... 12 Boolean (in SNP mining software/tools) ........57–62, 67 n2 Brain................................................................................ 5, 6 Brugada patients.................................................................... 179 f syndrome.........................................................172, 179 f
C Cambridge Reference Sequence descriptions..............67 n2 Cancer ......................................................23, 30, 67 n2, 226 colorectal ................................................................... 226 CAPG (gene).................................................................... 65 CAP3 (software)................................................... 78, 84, 85 Capillary electrophoresis..... 194, 194 f, 195, 236, 247, 250 t, 380, 381 f, 383, 385 sequencer(s)........................................................... 95, 97 Capillary maturation......................................................... 15 Carbocations ................................................................... 346 Cardiac sodium channel gene (SCN5A)....... 171, 172–176, 177, 179 f, 185, 186, 189 f Cataract............................................................................... 5 Cathechol-O-methyltransferase (COMT) ................ 28, 29 Causative factors ....................................................... 12, 294 CC chemokine receptor 2 (CCR2) gene........................ 139 CCM, see Chemical Cleavage of Mismatch (CCM) method Celera Discovery System (CDS)............................................ 50 SNP database .............................................................. 50 Cerebellar granular layer..................................................... 6 CHARGE (syndrome)................................................... 251 Chemical Cleavage of Mismatch (CCM) method...223–233 in liquid-phase .......................................................... 226 mismatched cytosine........................................................ 223, 236 thymine ....................................................... 223, 236
mismatch-specific reagents ....................................... 223 in solid-support phase............................... 226, 230–232 Chicken............................................................................. 85 ChIP-seq................................................................... 96, 109 Chloramphenicol acetyltransferase ................................... 30 Chronic skin disorder psoriasis (PSORS1–9) loci ........... 27 Cis factors.................................................................... 26, 27 Clarithromycin.................................................210, 220 n11 CleanRESIN1 (particles) ............334, 338, 339 n3, 341 n9 Cluster Report(s) .......... 46, 47, 48, 49, 60 f, 61–63, 69 n10 Cochlea ............................................................................... 6 Coding region(s) ...... 24, 26, 32 t, 33, 36, 43, 48, 49, 54, 63–64, 172, 388 f sequence(s) ........ 13, 24, 27, 36, 60 f, 213 t, 219 n7, 247 Codon bias .............................................................................. 31 frequency..................................................................... 33 nonsynonymous .......................................................... 36 synonymous............................................... 33, 34, 35, 36 Colorimetric.................................................... 429, 433, 434 reaction...................................................... 429, 433–434 Conformational change ........................8, 31, 171, 260, 261 Conformation-sensitive monoclonal antibodies UIC2 .... 30 Congenital nystagmus colorectal .................................... 226 Constant denaturing capillary electrophoresis (CDCE)........................................................ 236 Contig ......................................................... 80, 82, 100, 105 Control region (CR)...................................154, 158 f, 161 f Copy number variations (CNVs) .....48, 96, 107, 246, 251, 252, 325, 327, 330, 331, 331 f Coriell panels .................................................................... 63 Corneodesmosin (CDSN) gene ....................................... 27 Cost breakdown ............................................................282 t CRoPS (Complexity Reduction of Polymorphic Sequences).................... 76, 81 f, 82, 85–86, 88 t Crystal(s)...... 227, 238, 308, 310, 326, 338, 339 n2, 340 n6 CT dinucleotide repeats ..................................................... 5 Curly hair ...................................................................... 4, 15 Cyclen (1, 4, 7, 10-tetraazacyclododecane) ............ 169–181 Cyclic olefin copolymer (COC) ..................................... 426 Cyclic vomiting syndrome .............................................. 154 Cyclosporine A ................................................................. 30 Cysteine rich domain (CRD) .................................5, 6 f, 11 Cystic fibrosis.................................................................... 37 Cystine knot motif...................................................... 5, 8, 9 Cytomegalovirus ............................................................. 270
D D2cluster (method)..................................................... 80, 84 DbSNP-announce ............................................................ 52 DbSNP (database) ........ 43, 44 t, 45, 46, 47, 48, 49, 50, 51, 52, 55, 58, 60 f, 61–63, 66, 68 n4, 68 n5, 68 n6, 69 n8, 69 n10, 85, 96, 110 n1, 256, 312, 313
SINGLE NUCLEOTIDE POLYMORPHISMS
Subject Index 455
DC-SIGN (locus).........................................268 f, 269–272 DDJB (DNA Data Bank of Japan) .................................. 73 Deafness bilateral sensorineural ................................................... 8 progressive sensorineural .............................................. 5 Denaturing gradient gel electrophoresis (DGGE) ....................... 137–150, 154, 224, 236 Bio-Rad DCode System.............143, 149 n17, 163 n13 chemical gradient ...................................................... 154 direction parallel...............................................141, 144, 146 f perpendicular...... 141, 142, 144, 149 n16, 150 n18, 150 n19 gradient delivery system............................ 140, 143, 144 visualization .............................................. 140, 145–147 Denaturing high performance liquid chromatography (DHPLC) ..................................................... 236 Dengue (virus) ................................................................ 270 De novo sequence data ............................................ 76, 77, 80–81 SNP mining .......................................................... 75, 82 Deoxythymidine (dT) ...........................................169, 170 f Desiccator .....................................................180 n3, 190 n6 DGGE, see Denaturing gradient gel electrophoresis (DGGE) Diabetes ............................................................ 23, 257, 293 type I ......................................................................... 257 type II........................................................................ 257 Dialysis tubing ................................................................ 196 Diastereomers ................................................................. 210 Dimethoxytrityl protecting group .................................. 346 Disease Alzheimer’s ............................................................... 226 cardiovascular ........................................................ 23, 51 causing allele(s) .................................................................. 24 gene(s).................................................16 n1, 273 n6 clinical expression ..................................................... 154 Coats ........................................................................... 13 common ..............................4, 15, 23, 34, 293, 294, 393 complex .....................................245, 248, 252, 293, 294 coronary artery .................................................. 257, 293 development.............................................................. 437 human .........................15, 25, 26, 35, 36, 235, 394, 405 inherited................................10, 13, 171, 172, 177, 235 Mendelian .................................................24, 34, 36, 37 monogenic................................................................. 380 multifaceted ................................................................ 26 nonfamilial .................................................................. 13 Norrie..........................................................4–9, 13, 14 f ocular........................................................................... 15 Parkinson’s ................................................................ 251 pathogenesis.......................................... 12–13, 270, 293 predisposition................................................ 13, 25, 437
susceptibility....................................15, 24, 26, 193, 269 traits ............................................................................ 24 treatment........................................................... 294, 394 Disorder allelic ........................................................................... 10 mitochondrial............................................................ 154 neurological................................................................. 25 Disulfide bonds........................................................... 10, 11 D-loop............................................................................. 154 DNA doublings ................................................................ 98 DNA linkage analysis See also Linkage disequilibrium (LD) DNA Stretching on a Conductive Substrate See also Sequencing, scanning probe DNA sugar-phosphate backbone ................................... 224 Dominant, see Trait(s) Dopamine receptor D2 (DRD2) gene ............................. 27 Double-stranded DNA (dsDNA) ..............120 f, 171, 198, 219 n4, 219 n8, 348, 401 n5, 438, 439 f, 441, 449 n2 Drug..............4, 15, 24, 27, 30, 31, 34, 252, 255, 256, 394, 405, 437 response...................................................15, 24, 34, 394 Drug-binding pocket.......................................................................... 31 site(s)........................................................................... 31 DTNBP1 (gene) ............................................................... 65 Dual-color fluorescence............................................................... 394 hybridization ..................................................... 393–401 probes .................................................................. 396 Dubin-Johnson syndrome ................................................ 27
E Ebola (virus).................................................................... 270 Electron tunneling .................................................. 115, 116 Electropherogram(s) ...................194 f, 195, 199, 380, 385, 386, 388 f Enantiomers.................................................................... 210 ENCODE (project).............................................. 45, 48, 51 Endoplasmic reticulum ..................................................... 11 Ensembl (genome browser) ..............44 t, 46, 49, 50, 54, 64 Entrez-Gene (genome browser) ....................................... 46 EntrezProtein (genome browser) ..................................... 46 EntrezSNP (genome browser)..............46, 57, 58 t, 59, 60, 60 f, 61, 69 n8 Environmental cues.............................................................................. 23 factors............................................................ 13, 15, 293 inputs........................................................................... 35 Enzymatic activity......................................................................... 28 fidelity ............................................................... 247, 248 reaction...................................................................... 246
SINGLE NUCLEOTIDE POLYMORPHISMS
456 Subject Index
Epidemiological (studies) ............................................... 215 Epidermal growth factor (EGF) ...................................... 12 Epidermis.......................................................................... 27 Epilepsy............................................................................... 9 ESEfinder ................................................................... 54, 55 ESE’s, see Exonic splicing enhancer(s) (ESEs) ESS’s, see Exonic splicing silencer(s) (ESSs) Excel (software) .....................................63, 67 n1, 202, 372 Excimer emission ....................................209, 212, 214, 215, 218 fluorescence..........................................209, 210, 220 n9 formation .................................................................. 209 Excimeric PEPy-Probes Design............................. 209–220 Exonic splicing enhancer(s) (ESEs) ..................... 26, 54, 55 Exonic splicing silencer(s) (ESSs) .................................... 26 5’-exonuclease cleavage ................................................... 247 ExoSAP-IT (reagent) ........................147, 156, 161, 389 n8 Expectation Maximization (EM) algorithm .....273 n7, 329 Expressed Sequence Tags (ESTs) ..74, 75 f, 79 f, 82, 84, 85 Extended haplotype homozygosity (EHH) ......... 55, 56, 66 Exudates intra- ............................................................................. 8 subretinal....................................................................... 8 Eye ......................... 4, 5, 8, 9, 10 f, 11, 14, 401 n3, 423 n11 disorders .................................................................... 4, 9
F Fabrication of microarray........................................ 399–400 Familial exudative vitreoretinopathy (FEVR)...4, 7 t, 9–12, 13, 14 f FASTA ........................................................................68 n4 Fields (in SNP mining software/tools)............................. 57 Fine mapping ................................245, 246 f, 249–251, 252 Fish ............................................................278, 280, 289 n1 Flap endonuclease (FEN) ............................................... 364 Flavivirus West Nile virus (WNV)................................... 26 Fluidigm............. 247, 281, 282, 283, 285–286, 287, 289 t, 289 n1, 290 n6 Fluorescence resonance energy transfer (FRET)........... 364, 366, 368 t Fluorescence SNP analysis .............................................................. 212–216 detection............................................................ 209–220 Fluorescent dye ..... 97, 194, 205 n1, 211, 247, 262, 272 n4, 364, 386, 438 terminators .................................................................. 97 Fluorophore ............. 100, 102, 210, 212, 226, 258 t, 259 t, 260, 261, 262, 263, 264, 264 t, 265–266, 272 n2, 364 f, 366 Focused ion beam (FIB) ....................115, 117, 118, 120 n3 Footprinting (experiments)............................................. 170 Forensic (mitochondrial DNA analysis)......................... 380 Formamide....................138, 140, 149, 227, 228, 349, 383, 385, 390 n14
Fragile Histidine Triad (FHIT) gene ............................ 139 Frizzled-4 (FZD4) receptor .........................................5, 6 f Frizzleds (FZ) ................................................................... 14 Full-term infants............................................................... 12
G Gabor Marth..................................................................... 82 G-allele-specific (primer) .............................185, 187, 188 f Ganglion cell................................................................... 5, 6 GC-rich clamp/sequence and/or GC clamp ......... 108, 139, 140, 141, 142–143, 147 n2, 150, 155, 156, 157, 158 f, 162 n3 GenBank.............. 46, 47, 53, 54, 60 f, 100, 172, 176 t, 185 GeneID................................................................. 48, 64, 68 GeneMapper (software).................................. 195, 383, 385 GeneScan (software)...........195, 205 n2, 383, 385, 390 n14 Genetic.......... 4, 9, 10, 12–13, 15, 16 n1, 16 n3, 24, 25, 26, 28, 34, 36 n1, 36 n3, 48, 61, 64, 73, 74, 177, 195, 205 n1, 224, 245, 250 t, 256, 269, 270, 273 n6, 277, 279, 293, 294, 301, 307, 309, 313, 363, 380, 383, 389 n4, 390 n12, 390 n14, 393, 405, 415 code redundancy............................................................ 24 factors.............................................................. 9, 26, 293 heterogeneity......................................................... 10, 13 mapping studies .....................................................36 n1 predisposition.............................................................. 13 variation density ......................................................... 279–281 GeneView (software/tool) ................................................ 62 Genome coverage..................................................... 104, 105, 108 diversity database(s)........................................................... 437 evolution ................................................................. 4, 15 map(s) ............................................................. 48, 50, 52 mitochondrial..........................................153, 154, 388 f nuclear....................................................................... 154 sequence ....................50, 76, 78, 85, 256, 316, 437, 442 Variation Server (GVS) .............................................. 52 -wide association studies............................. 49, 294, 313 Workbench ................................................................. 52 Genomics ....15, 96, 176, 277, 278, 308, 330, 394, 405, 415 platform..................................................................... 307 Genotype polymorphic .............................................. 73, 76, 85, 86 Genotyping biallelic ...................................................................... 309 high-throughput ............................................... 277–291 low-cost..................................................................... 194 platform(s) Sequenom’s ......................................................... 312 results ............................ 16 n4, 374 f, 375 f, 399 f, 400 f GeSiM Nanoplotter ............................................... 396, 399
SINGLE NUCLEOTIDE POLYMORPHISMS
Subject Index 457
Gibbs free energy .............................................................. 28 GigaBayes (software) ................................................83, 87 t Glaucoma............................................................................ 5 Gliosis ............................................................................... 11 Glutamate neurotransmission ....................................................... 25 synaptic levels.............................................................. 25 GoldenGate BeadChips assay(s) (Illumina)................... 293 Google....................................................................44, 67 n2 Scholar ................................................................... 67 n2 Graphical browser............................................................. 49 Growth factor b2 .................................................................................. 5 BB platelet derived........................................................ 5 nerve.............................................................................. 5 GUI (Graphical User Interface) .............................83, 312 f
H HaploSNPer (software) .................................................... 85 Haplotter (software) ................................. 44, 55, 56, 65–66 Haplotype block(s).................................. 53, 54, 64–65, 248 Haploview (software)........53, 64, 263, 271, 273 n7, 273 n8 HapMap phase II ......................................................... 56, 66, 257 phase III ................................................................ 57, 66 Hardy–Weinberg Equilibrium (HWE) ......................... 327 Hearing impairment .....................................................7 t, 8 Heat-labile shrimp alkaline phosphatase (SAP) ........... 310, 332–333, 336 t, 337 t, 341 n12, 381 f, 382, 384, 385, 387 t -helices............................................................................ 31 Helicobacter pylori ..........................................210, 213 t, 270 Helicos (sequencing technology), See also Sequencing, Helicos (technology), single molecule -hemolysin.................................................................... 114 Hemophilia ..................................................................... 226 Hemorrhage.............................................................. 5, 9, 11 intraretinal and subretinal............................................. 9 Hepatitis C (virus) .......................................................... 270 Hereditary angioedema................................................... 226 Heteroduplex .......170, 171, 172, 180, 224, 229, 231 f, 236, 239, 240, 241 n4, 406 Heteroduplexing (technique)................................171 f, 177 Heteroplasmy........................................................154, 161 f Heterozygosity .......... 58, 58 t, 59, 60 f, 66, 124, 132 f, 170, 172, 177, 179 f, 294, 311 High Density Human 1 M-Duo (Illumina BeadArray)............................................ 248, 251 High-fidelity polymerase .......................................... 98, 108 High pain responsiveness (HPS), see Pain High performance liquid chromatography (HPLC)..... 211, 236, 263, 265, 266, 332, 336 t, 349, 357, 360 n7
High-resolution melt curve analysis (hrMCA).............. 224 High-Throughput analyses...................................................................... 281 platform(s)................................................... 281–282 Hippocampus...................................................................... 6 HIV......................................................................... 257, 270 Homoduplex .................................171, 229, 231 f, 236, 240 Homopolymer errors ......................................................... 100–102, 107 runs.................................................................... 100, 102 stretches............................................. 100, 101, 102, 108 Hospitalized................................................................ 26, 34 Hot start Taq master mix ............................................... 309 HPLC, see High performance liquid chromatography (HPLC) HUGO (Human Genome Organisation) .............48, 68 n7 Human eye disorders, see Eye, disorders Human genome diversity panel (HGDP).................. 57, 66 Human Genome Project (HGP).............. 25, 109, 257, 393 Human leukocyte antigen (HLA) alleles ..........142 f, 146 f, 411, 411 f Humidity................................................................... 59, 340 Hybridization buffer...............................................211, 212, 214 f, 396 cassettes...................................................426, 431 f, 432 probe(s) ............................................................. 209–220 in situ............................................................................. 5 Hydroxylamine............................224, 225 f, 226, 231 f, 236 Hydroxypicolinic acid ..............................308, 310, 341 n13 Hypertension .................................................................. 293 Hypervariable region 1 (HV1) (gene) ............................ 154 Hypoxia............................................................................. 12
I Idiopathic ventricular fibrillation (IVF) ................. 172, 179 IHS (integrated haplotype score) ......................... 56, 65, 66 Illumina BeadArray ............................ 246 f, 248, 249, 250 t, 251 GA (sequencing technology), see Sequencing, Illumina (technology) Immediate gene .................................................................. 5 Inconsequential ........................................................... 24, 36 Indel(s) (insertion-deletion(s)..........74, 78, 84, 86, 96, 102, 107, 108, 109, 124, 197, 201 f Individuality.................................................................. 4, 15 Infrastructure .......................................................... 114, 425 Inherited arrhythmia (syndromes).......................... 171, 172 In silico data-mining........................................................ 312 In situ hybridization, see Hybridization InSNP ................................................................. 84, 87, 406 Intrafamilial variability ....................................................... 5 Invader1 assay................................................................... 363–376 efficiency ................................................................... 376
SINGLE NUCLEOTIDE POLYMORPHISMS
458 Subject Index
Invader1 (continued) master mix...............................................368 t, 370, 376 reaction plates ...................................296, 332, 333, 334, 376 system........................................................................ 370 technology......................................................... 364, 366 Ionic currents .................................................. 114, 115, 121 IPLEX, see Matrix-assisted laser desorption/ionization time-of-flight mass spectrometry Irradiation ................................................................. 27, 346 Isoform.............................................................................. 25 Isothermal (reaction) ...................................................... 441 Isothermal Smart Amplification Method, see Smart Amplification Process (SmartAmp) IUPAC allele codes .................................................................. 58 code base ..................................................................... 62
K Keratinocytes..................................................................... 27 Kim-wipe1 ..................................................................... 375 Klebsiela pneumonia ......................................................... 270 Klenow Fragment ................................................... 195, 199 KOD-plus DNA polymerase.......................................... 177
L Labor-efficiency.............................................................. 194 Laser energy........................................................................ 308 ultraviolet .................................................................. 308 LDI-TOF Detection of Trityl Mass-Tags .... 349–350, 358 Leishmania pifanoi........................................................... 270 Leukocytes .............................................. 172, 185, 196, 411 Light infra-red .................................................................... 263 visible ........................................................................ 263 Likelihood value ............................................................... 77 Linkage disequilibrium (LD) ....24, 36, 44, 49, 64, 65, 257, 270, 313, 429, 430 LD Plot....................................................................... 64 Lipid bilayer.................................................................... 114 LLPDMAA (long-chain linear poly-N, Ndimethylacrylamide) ..... 195, 196, 198, 200, 204 Lobe dysfunction.................................................................. 25 frontal.......................................................................... 25 temporal ...................................................................... 25 Long QT syndrome type 3 (LQT3),....................172, 179 f Loss of heterozygosity (LOH) .............124, 128, 132 f, 311 Low-density lipoprotein receptor protein (LRP5).........6 f, 11, 12, 14, 15 Low pain responsiveness (LPS)........................................ 28 See also Pain Lung................................................................ 5, 44, 51, 230
M MacMelt (software) ................139, 155, 156, 157, 162, 163 Macula .......................................................................... 9, 11 Magnetic nanoparticles (MNPs) streptavidin coated (SA-MNPs)....395, 396, 397, 398 f, 399 f Major histocompatibility complex (MHC).............. 69, 139 MALDI-TOF MS, see Matrix-assisted laser desorption/ionization time-of-flight mass spectrometry Manual coupling (approaches) ....................................... 263 MapViewer (software) .................................... 52, 60, 63, 69 MAQ (software) ............................................................... 78 Marker(s) polymorphic ................................................................ 55 MassARRAY, see Matrix-assisted laser desorption/ ionization time-of-flight mass spectrometry MassEXTEND, see Matrix-assisted laser desorption/ ionization time-of-flight mass spectrometry Mass-spectrometer.................................. 310, 326, 335, 338 Matrix-assisted laser desorption/ionization (MALDI) Time-of-flight (TOF) Mass spectrometry (MS), see Matrix-assisted laser desorption/ionization time-of-flight mass spectrometry Matrix-assisted laser desorption/ionization time-of-flight mass spectrometry amplicon design ................................ 317, 319, 320, 322 assay design (software).....309, 316, 319, 321, 322, 323, 324, 325 eXTEND (algorithm, program) ePCR (program) ................................................. 314 PleXTEND......................................................... 316 PreXTEND ................................................ 315–316 ProxSNP ............................................................. 315 genotype QC............................................................. 327 iPlex (assay) chemistry.....................................308, 316, 317, 327 mass-to-charge ratios................................................ 309 MassEXTEND (assay)....308, 314, 316, 317, 318, 319, 320, 333, 335, 337 mass spectrum ..........308, 309, 317, 318, 320, 321, 322, 338, 348 primer design ............................69, 128, 316, 317, 320, 322, 382, 383, 390, 407, 417, 419, 438, 440, 443, 449 SpectroACQUIRE (software).......................... 326, 327 SpectroCHIP1 silica array ............................... 310, 340 Matrix-independent desorption-ionization.................... 346 MAVIANT (Multipurpose Alignment VIewing and Annotation Tool)................................ 82, 85, 88 MDR, see Multiple Drug Resistance 1 (MDR1) gene Medication........................................................................ 15 Melting curves ........................................................ 212, 214
SINGLE NUCLEOTIDE POLYMORPHISMS
Subject Index 459
Melting temperature (Tm) ....137, 141, 143, 156, 157, 162, 170, 180, 219, 224, 228, 239, 246, 267, 272, 300, 406, 407, 419, 443 Mendeilan disease, see Disease, Mendelian Mental retardation...................................................... 5, 8, 9 MeSH (i.e. Medical SubHeadings) vocabulary ................................................................... 61 Metabotropic glutamate receptors (GRMs)..................... 25 2-methacryloyloxyethyl phosphorylcholine (MPC) ................................................426, 430 f Methylation (determination).......................................... 312 Methylenetetrahydrofolate reductase (MTHFR) gene.. 397 MHCIII (gene) ............................................................348 f MHC, see Major histocompatibility complex (MHC) Mice .......................................................................... 5, 6, 15 Microdispenser................196, 295, 296, 297, 298, 299, 393 Micro-fluidic card........................................................... 281 MicroRNA................................................................ 96, 109 Migraines ........................................................................ 154 Mining SNP ............................................................... 73–89 Mismatch Oxidation Assay hypomanganate diester intermediate........................ 236 UV/Vis Microplate Reader............................... 235–242 Mitochondria ............................67, 154, 158, 161, 380, 388 Mitochondrial DNA (mtDNA) ...................154, 380, 388 f Modified Multiple Primer Extension (MPEX) (method) BioChip Arrayer1 spotting robot ............ 426, 431, 432 S-BIO1 PrimeSurface1 (substrate)........ 426, 430, 432 Molecular beacons fluorophore(s) derivatives iodoacetamide ..................................................... 263 maleimide............................................................ 263 succinimidyl ester........................................ 263, 265 quencher(s) black hole .................................................... 262, 265 signal to background ratio ........................................ 266 stem-and-loop structure ........................................... 260 thermal denaturation profiles ........................... 261, 267 Molecular diagnosis postnatal.................................................................... 380 prenatal...................................................................... 380 Molecular Mass standard........................................ 149, 163 Monozygotic twins ........................................................... 13 MPEX, see Modified Multiple Primer Extension (MPEX) (method) MRNA splicing ...................................................... 24, 25, 26, 29 stability...................................................... 15, 24, 26, 27 structure ....................................................24, 28, 31, 35 MTMR9 (gene)................................................................ 61 Mucins ................................................................................ 5 Multi-exonic (genes)......................................................... 25 Multiple base extension (MBE) ..................................... 320
Multiple Drug Resistance 1 (MDR1) gene ............... 30, 31 Multiple sclerosis ........................................................ 26, 59 Multiplex capabilities................................................................. 394 extension conditions ................................................. 385 PCR .................................................................. 380, 386 reaction.............................................. 341, 381, 385, 389 Single Base Primer Extension (assay)............... 307–342 typing ........................................................................ 255 Muscle............................................................................. 5, 9 Mutant-allele-specific amplification (MASA)............... 190 Mutational analysis ........................................... 6, 10, 11, 16 Mutation(s) causative .................................................................... 226 deletions .......................................................... 11, 12, 26 disease-causing.................................................. 169, 172 frameshift ...................................................................... 6 heterozygous .............................................11, 171, 179 f homozygous ................................................................ 12 insertions..................................................................... 12 length-based.............................................................. 102 missense ..............................................4, 8, 9, 10, 12, 26 nonsense.............................................................. 6, 8, 26 submicroscopic .............................................................. 8 Mycobacterium leprae ....................................................... 270 Mycobacterium tuberculosis ............................................... 270 MyNCBI..................................................................... 52, 61
N Nanopores protein....................................................................... 114 synthetic .................................................... 114, 115, 117 Nanopore Sequencing, see Sequencing, nanopore Nanopure (water)............................333, 334, 335, 336, 340 N-butyl methacrylate (BMA)...............................426, 430 f NCBI Entrez ........................................................ 43, 57, 60 Neighborhood Quality Standard (NQS).......................... 84 Neodymium-boron (Nd-B) magnet............................... 396 Next generation machines ..................................................................... 75 sequencing (NGS) methods......................................................... 95–110 technologies ................................................ 107, 109 Nomenclature faulty ........................................................................... 24 Non-enzymatic differential hybridization ...................... 246 Non-model organism(s).......................................... 277–291 Non-synonymous substitutions (Ka) ................................ 36 Norrie disease, see Disease Norrin .......................................................5, 6, 8, 9, 10, 14 f NovoSNP (software)...........................................83, 84, 87 t 5’-nuclease reaction, see Taqman Nuclease S1....................................................................... 28
SINGLE NUCLEOTIDE POLYMORPHISMS
460 Subject Index
Nucleic Acid Sequence Based Amplification (NASBA)...................................................... 257 Nucleotide mismatches................................................... 450
O Obesity ...................................................... 61, 257, 274, 293 Olfactory bulb ..................................................................... 6 Oligoadenylate synthetase gene(s) (OASs) ...................... 26 2’,5’-oligoadenylate synthetase like gene (OASL) ........... 26 OMIM (Online Mendelian Inheritance in Man database) ...........................46, 47, 48, 60, 65, 68 Oncogene (mutation screening) ..................................... 311 Operator(s) (in SNP mining software/tools)............ 57, 59, 60, 61, 67, 198, 303 n4 m opioid receptor gene (OPRM1) ................ 426, 428, 429, 430 f, 432, 434 n2 Optical absorbance assay, see Mismatch Oxidation Assay Optic nerve atrophy .......................................................................... 5 Osteoporosis-pseudoglioma.............................................. 12 Oversampling............................................................ 97, 108
P Pacific salmon ................................................................. 278 Pain desensitized ................................................................. 28 experimental................................................................ 28 perception.................................................................... 28 responsiveness high (HPS) ........................................................... 28 low (LPS).............................................................. 28 sensitivity..................................................................... 28 Parafilm1 ...................................................................375 n6 Paralogue(s) ......................................80, 82, 85, 86, 87–88 t Pathogen(s) ............................................................. 255, 270 Patient(s)...... 5, 6, 8, 9, 10, 11, 12, 13, 15, 25, 26, 27, 172, 179 f, 186, 224, 255, 395, 440 PCR allele-specific amplification (PASA), see Allele-specific (AS) PCR (AS-PCR) PCR amplification of multiple specific allele (PAMSA), see Allele-specific (AS) PCR (AS-PCR) Perlegen ................................44, 46, 49, 50, 56, 64, 66, 256 Persistent fetal vasculature (PFV)..................................... 13 Personalized medicine ........................................ 24, 36, 256 P-glycoprotein (P-gp)....................................................... 30 Pharmacogenetics ........................................................... 405 Pharmacogenomics ................................................... 24, 437 Phased haplotype display ................................................................... 64, 65 plot .............................................................................. 65 Phenomena(non) psychophysiological..................................................... 25 Phenotype(s) clinical ...............................................................4, 7 t, 35
cognitive.................................................................... 6–7 distinct......................................................................... 25 heterogeneity................................................................. 8 manifestation................................................................. 9 mild ............................................................................... 8 NDP.............................................................................. 8 severe......................................................................... 8, 9 variability..................................................................... 10 variation ......................................................9, 15, 23, 35 Phenylethynylpyrene Excimer Forming Hybridization Probes .......... 209–220 Phenylketonuria .............................................................. 226 Phosphate-Affinity Polyacrylamide Gel Electrophoresis dinuclear metal complex ........................................... 183 phosphate-affinity gel ....................................... 184, 186 phosphomonoester dianion (R-OPO32–) ................. 184 phos-tag .............180 n2, 183, 184, 185, 186, 187, 188, 189 f, 190 n5, 191 n11 Phosphopeptides ............................................................. 184 Phosphoprotein(s)........................................................... 184 Phosphoramidites ...........263, 348, 349, 350, 351, 355, 356 Phosphorothioate oligonucleotides .................................. 28 Phrap (software)..........................78, 84, 85, 197, 200, 201 f PHRED score ................................................................ 76, 77, 97 value(s) .................................................................. 96–97 Phthisis bulbi ...................................................................... 8 Phylogenetic tree views..................................................... 53 PicoGreen dsDNA (reagent).......................................... 196 Picoliter reactors ............................................................... 96 Pipeline tools .................................................................... 85 Piperidine... 225 f, 227, 228, 229, 230, 231 f, 232, 233, 236 Piperidine-induced b-elimination .................................. 224 Placenta............................................................................... 5 PLACE-SSCP, see Post-Labelling method with separation by Automated Capillary Electrophoresis under SSCP conditions (PLACE-SSCP) P-nitrophenyloxycarbonyl polyethyleneglycol methacrylate (MEONP) ............................................ 426, 432 Point-of-care (POC) diagnostics................................................................. 438 PolyBayesShort ................................................................. 83 PolyBayes (software)...........................................82–85, 87 t Polycarbonate.................................................................. 233 Polymorphism(s) diallelic ...................................................................... 124 microsatellite repeat .................................................... 16 Polyphen (software) .................................................... 54–55 PolyPhred (software) ..........................................82–84, 87 t Polyploid samples ............................................................. 79 Poly(vinylidene difluoride), see PVDF membrane Pooled DNA samples ...................................................... 132, 193–206 -SSCP analysis (PD-SSCP)............................. 193–206
SINGLE NUCLEOTIDE POLYMORPHISMS
Subject Index 461
Population ASW( African ancestry in SW USA) ......................45 t CEU (Utah residents with N & W European ancestry) ..................................................45 t, 50 CHB (Han Chinese in Beijing, China)....45 t, 50, 66, 252 CHD (Chinese in Metropolitan Denver, Colorado)......................................................45 t GIH (Gujarati Indians in Houston, Texas) .............45 t JPT (Japanese in Tokyo, Japan)..........................45 t, 66 LWK (Luhya in Webuye, Kenya) ............................45 t MEX (Mexican ancestry in Los Angeles, California).....................................................45 t MKK (Maasai in Kinyawa, Kenya) ..........................45 t panel probing................................................................45 t TSI (Toscans in Italy).........................................45 t, 66 YRI (Yoruba in Ibadan, Nigeria) .................45 t, 50, 56 Post-Labelling method with separation by Automated Capillary Electrophoresis under SSCP conditions (PLACE-SSCP)............... 194, 195, 196, 197, 198, 205 n1 Potassium permanganate ... 224, 225 f, 231 f, 236, 237, 241 f Potato................................................................................ 85 Precordial electrocardiogram .......................................... 172 Pre-implantation genetic diagnosis (PGD).................... 172 Prescription ............................................................. 440–441 Pretreatment ................................................................... 440 PRIMER 3 (software) .................................... 124, 197, 382 Program for Genomic Applications (PGA).....44 t, 51, 256 Proliferation ...................................................................... 11 Promoter .................................4, 13, 60, 69, 269 f, 270, 272 2-propanol............................................................... 196, 204 b-propeller ........................................................................ 12 Protein folding cotranslational ........................................... 24, 29, 31, 35 Proteome........................................................................... 25 Proteostasis ....................................................................... 25 Pseudoglioma (NDP) .............................................5, 6 f, 14 Pseudonucleoside ...............................................210, 219 n3 Psoriasis............................................................................. 27 Psychiatric disorders ........................................................................ 3 issues ........................................................................... 23 PubMed ........................................46, 47, 48, 59, 61, 67, 68 PupaSNP........................................................................... 54 PupaSuite (Putative Phenotype Alterations Suite) ... 43–44, 54, 55, 63, 64, 65 PVDF membrane............................................................ 184 Pyrene ........................................................209–210, 219 n3 Pyrosequencing, see Sequencing, pyrosequencing
Q QIAamp DNA Blood (Kit).................... 172, 185, 196, 395 QIAquick column........................................................... 240
QSNPlite (software) .............195, 197–198, 200, 201 f, 202 Quality control...........30, 31, 128, 284, 286, 288, 290, 303, 313, 327 QualitySNP.........................................................82, 85, 88 t
R Rabbit.............................................................................. 5–6 Real-time PCR ......177, 186, 261, 264, 269, 272, 280, 282, 438, 442 RefSNPs............................................................................ 47 Region(s) coding............................................24, 26, 33, 172, 388 f exonic .................................................................... 48, 54 intronic............................................................ 24, 48, 54 noncoding .................. 4, 12, 13, 24, 63, 154, 213 t, 313 RepeatMasker(software) ...........48, 62, 69, 70, 78, 80, 104 f Repeat sequences, see Sequence(s), repeat Replexing ........................................................................ 325 Replication protein A ....................................................... 28 Replicator pins ................................................ 365–366, 370 Restriction Enzyme Analysis of PCR products (PCR-RFLP)........................................ 405–413 Restriction enzyme(s) ............................... 86, 248, 405–413 Retina peripheral ............................................................ 8, 9, 11 vascularization................................................... 9, 11, 12 Retinal avascularity .................................................................. 11 detachment....................................5, 8, 9, 10, 11, 12–13 development............................................................ 9, 15 dysgenesis.................................................................... 10 dysplasia ........................................................................ 9 traction .......................................................................... 8 Retinopathy of prematurity (ROP) ..........4, 12–13, 14 f, 15 Retrolental mass.................................................................. 5 Rheumatoid arthritis....................................................... 257 Ribonuclease L (RNASEL) ............................................. 26 Ribosome stalling.............................................................. 31 Robotic liquid handling system .............................................. 196 loading....................................................................... 286 Roche GS (sequencing technology), see Sequencing, Roche (technology) Rs-number(s) ........47, 48, 49, 50, 55, 59, 62, 66, 67, 68, 69
S Sanjay Tyagi............................................................ 255, 256 Santa Cruz In Silico PCR ................................................ 53 SARS coronavirus........................................................... 270 Scanning Probe Sequencing, see Sequencing, scanning probe Scanning tunneling microscope (STM) ............... 115, 116, 117, 118 Scatter plot(s)........................... 280 f, 328, 328 f, 330, 331 f Schistosoma mansoni......................................................... 270
SINGLE NUCLEOTIDE POLYMORPHISMS
462 Subject Index
Schizophrenia ................................................................... 25 SeattleSNPs ................................................44 t, 51–52, 256 Self-assembled monolayer (SAM)........................ 116, 118, 119, 120 f Self-closing tweezers See also Sequencing, nanopore Sephadex G–25....................................................... 263, 265 Sequence(s) coding..................................... 13, 60 f, 213 t, 213 t, 219 data.............................................................................. 48 flanking .......................................60, 62, 68, 70, 80, 380 HLA-DRB1 ...................................... 142 f, 146 f, 411 f information .............................75, 76, 80, 303, 443, 449 intergenic .................................................................... 24 M13Forward ..................................................... 148, 162 M13Reverse ...................................................... 148, 162 pattern ......................................................................... 77 repeat........................................................... 78, 103, 105 similarity ......................................................... 53, 62, 75 variants ..........................................74, 75, 77, 78–79, 83 Sequence Tagged Site(s) (STSs) ..............................75 f, 84 Sequencing 454 (technology) ......................................................... 77 accuracy ........................................................... 97, 98, 99 errors ...................................................74, 77, 78, 79, 80 Helicos (technology)....98, 99, 100, 101, 102, 103, 104, 107, 108, 109 high throughput (HT).................................... 74, 82, 85 Illumina (technology) ......................... 96, 101, 102, 103 nanopore self-closing tweezers .................117, 120 n1, 121 n5 silicon nitride membrane .................... 114–115, 116 platforms ............................................................. 77, 116 polony.................................................................. 96, 104 pyrosequencing............................ 99, 100–108, 123–132 CpG island(s)...................................................... 124 Dispensation .............................124, 127, 130, 131 f PSQ TM.................................96, 125, 126, 127, 128 instrument........................................... 126, 127, 128 reagent cartridge ................................. 126, 127, 128 Pyro Gold Reagents.................................... 126, 127 Streptavidin Sepharose HP ................................ 125 Vacuum Prep Tool.............................. 125, 127, 129 redundancy................................................................ 103 Roche (technology).......... 74, 77, 82, 83, 85, 87 t, 88 t, 253, 259 t, 280, 281, 282, 295, 382, 442 Sanger (dideoxy) .......................76, 77, 79, 83, 97–100, 103, 108, 235 scanning probe DNA Stretching on a conductive substrate ........................................ 117, 118–119 single molecule, See also Helicos (sequencing technology) single-pass .............................................................6 f, 97
Solexa (technology)......................................... 74, 77, 83 SOLiD (technology)........................................... 99, 103 ultra-deep.................................................................... 76 Sequencing cycle ............................................................... 95 454 (sequencing technology), see Sequencing 454 (technology) Severe, X-linked recessive neurodevelopmental disorder... 5 b-sheets ............................................................................. 31 Shotgun fragments ............................................................74, 75 f Signaling pathway Wnt...........................................................................14 f Signal to noise ratio (SNR) ..........................329 f, 337, 338 Signal transduction ....................................................... 4, 15 Silica beads.............................................. 226, 228, 230, 232 Silicon nitride membrane See also Sequencing, nanopore Single Base Primer Extension (SBE) cocktail ...................................................................... 310 multiplex, see Multiplex, Single Base Primer Extension (assay) primer concentration................................................. 385 product(s) .................................................. 326, 335, 386 reaction............................308, 380, 381 f, 385, 386, 387 Single gene disorder............................................................ 6 Singleplex (condition)...............384–385, 389 n10, 390 n14 Single-strand conformational polymorphism (SSCP).................................................. 193–206 Single stranded conformation polymorphism (SSCP) (method) ............................................... 193–206 conformation analysis polymer, see LLPDMAA (long-chain linear poly-N, N-dimethylacrylamide) data analysis ...................................................... 200–202 LLPDMAA (long-chain linear poly-N, N-dimethylacrylamide)......... 198, 200, 204–205 overall scheme ........................................................... 198 plate design ....................................................... 197–198 TAMRA-labeled double-stranded DNA markers... 198 Two-Color Post-Labeling of PCR Products ................................................ 199–200 Single-stranded DNA fragments (ssDNA).................. 124, 125, 126, 127, 128, 129, 130, 396, 397, 397 f, 398, 399 f SISAR (serial invasive signal amplification reaction) process.........................................................364 f Slits .............................................................219 n5, 220 n10 Smart Amplification Process (SmartAmp) ............ 437–450 Aac DNA polymerase..................438, 444, 447, 449 n1 assay optimization................................................ 446–447 reaction........................................................ 443–446 Competitive Probe (CP)........................................... 437 DNA mismatch binding protein (MutS) ......... 441, 447
SINGLE NUCLEOTIDE POLYMORPHISMS
Subject Index 463
intermediate products IM1 ...................................................440 f, 444, 446 IM2 ...........................................440 f, 444, 445, 446 pathway ...................................439, 440 f, 441, 444, 445 primer set boost primer (BP) .....................................442, 444 f discrimination primer (DP) ................................ 443 folding primer (FP).........................440 f, 442, 444 f outer Primer(s) OP1............................................. 442, 443, 446, 447 OP2............................................. 442, 443, 446, 447 turn-back primer (TP) ....................440 f, 442, 444 f SMAP primer design software ......................... 449–450 SmartAmp 2 (SMAP2) ............................ 438, 441, 449 SNaPshot1 reaction.............................................................. 381, 390 reagent kit ...............................................................387 t SNP2Prot.......................................................................... 55 SNPBrowser ......................................................... 51, 53–54 SNP cosegregation score................................................... 84 SNPdetector..............................................................84, 87 t SNP discovery............................................. 73–89, 177, 282 SNPeffect.......................................................................... 54 SNP filter criteria............................................................75 f SNP mining, see Mining SNP SNPper......................................................43, 48, 54, 55, 63 SNP-PHAGE ..........................................................84, 87 t SNP redundancy score...................................................... 84 SOAP................................................................................ 78 Solexa (sequencing technology), see Sequencing, Solexa (technology) Solid-phase polymerase chain reaction (PCR)............... 393 SOLiD (sequencing technology), see Sequencing, SOLiD (technology) Spectral overlap............................................................... 264 Spectrofluorometer ......................................... 196, 212, 263 Spectrofluorometric thermal cycler................................. 267 Spidey................................................................................ 78 Spiral ganglion cells................................................................. 6 ligament cells ................................................................ 6 Splicing ...........................12, 25–26, 28, 29, 35, 54, 55, 179 SPSmart ..............................................44, 50, 56–57, 65, 66 Spurious peaks ................................................................ 194 23S RNA ...................................................215, 216, 219 n7 SsahaSNP..................................................................84, 87 t SSAHA (software)....................................................78, 87 t SSCP, see Single stranded conformation polymorphism (SSCP) (method) Ss-number(s)..................................................................... 47 Stria vascularis..................................................................... 6 Subarray .......................................................................... 281 Subplexing....................................................... 322, 323, 324 Sulfahydryl functional groups......................... 322, 323, 324
Sulfur-linked laser-cleavable trityl labels........................ 345 Superplexing ........................................................... 324–325 Surface plasmon resonance ............................................. 184 Susceptibility gene .......................................................... 294 SWEEP (software) ..................................................... 55, 56 Swiss-prot/Uniprot ........................................................... 50 SYBR Gold (stain) ...............140, 145, 146 f, 148, 156, 160 Synonymous substitutions (Ks) ........................................ 36 Syntax (in SNP mining software/tools)............................ 57
T T4 polynucleotide kinase ........................................ 185, 187 Tabular views of data (in SNP mining software/tools).... 53 Tags................. 15, 57, 58 t, 60, 74, 79 t, 86, 197, 279, 308, 319–320, 322, 345–360 Tag SNP Picker ................................................................ 64 TagSNPs (tag-SNPs) ....................................................... 53 Tallness ............................................................................... 4 Taqman assay........... 247, 260 f, 279 f, 280, 281–282, 284, 285 f, 294, 304 f chemistry.................................................................250 t end-point read .................................................. 283, 284 error rate.................................................................... 287 genotyping master mix...................................... 295, 300 offline thermal cycling ...................................... 283, 284 primer........................................................................ 284 probes MGB........................................................... 294, 299 reaction(s).......................................................... 295, 296 TaqStart Antibody.................................................. 195, 198 TATA box .......................................................................... 5 T-cell receptor (Tcf) ......................................................... 14 Tecan Genesis RSP150, see Robotic, liquid handling system Temperature Gradient Gel Electrophoresis (TGGE)........................................................ 154 Temporal Temperature Gradient Electrophoresis (TTGE) ................................................ 153–164 Bio-Rad DCode System........................... 149, 159, 163 casting and running TTGEs .................................... 155 optimal temperature.............................................163 n8 preparation of acrylamide solution(s) ....................... 155 visualization ...................................................... 156, 160 TeraClu (algorithm) ................................................... 80, 85 Teratogen........................................................................ 140 Tetra-allelic SNPs .......................................................... 379 TGGE, see Temperature Gradient Gel Electrophoresis (TGGE) TGICL (software) ...................................................... 80, 86 Thermal Denaturation (studies) ..................................... 212 Thermo Sequenase................................................195, 199 t Thermus aquaticus .................................................... 439, 441 Thymine-dependent electrophoresis .............................. 172 TIGR (The Institute for Genomic Research).................. 80
SINGLE NUCLEOTIDE POLYMORPHISMS
464 Subject Index
Tissue sample(s) ................................................................... 438 Trait(s) autosomal dominant ...............................10, 11, 12, 14 f autosomal recessive .......................................10, 12, 14 f obesity-related........................................................... 257 X-linked recessive .............................................5, 7 t, 10 Transcription factor binding sites .................................................... 24, 54, 64 Transcriptome........................................................... 76, 109 Transition(s) ..................................................................... 33 Translational pauses.................................................... 30, 35 Translation (protein) inefficiency .................................................................... 8 premature termination ................................................ 12 Transmission electron microscope (TEM)............ 115, 117, 118, 121 Transverse electrodes ...................................................... 115 Transversion(s)......................................................102, 179 f Trisomy 21 (detection) ................................................... 311 Trityl cation......................................................................... 345 mass-tags (MT) ................................................ 345–360 TRNA(s)............................................................... 24, 28, 35 Trypsin digestion...................................................................... 30 TTGE, see Temporal Temperature Gradient Electrophoresis (TTGE) Tumor samples ...................................................................... 226 Turnaround time ............................................................ 251 Typer database..................................................... 310, 317, 325 software ..................................................................... 339
U Unigene............................................................................. 76 Uniplex reaction.........................................280, 282, 283, 289 n2 Universal SS-modified primer........................................ 347 Untranslated region(s) ........................5, 13, 15, 24, 27, 32 t Urea.........................137–138, 140, 143, 149, 154, 155, 157 Uridine – meta........................................................................ 210 – para......................................................................... 210
V Variability...........5, 6, 10, 24, 26, 45, 46, 49, 50, 51, 56, 57, 63, 66, 278, 409 Vascular development ............................................... 5, 6, 15 VEGA (Vertebrate Genome Annotation) ....................... 49 Verapamil.......................................................................... 30 Vertebrate (species)......................................................... 154 Vessel growth ......................................................................... 12 Visual Basic (software).................................................... 197 Vitreous hemorrhage ................................................................... 5 opacity ........................................................................... 8 Von Willebrand factor........................................................ 5 VP110 surfactant ............................................................ 366
W Western blotting............................................................. 184 Wet lab work .................................................................. 312 Whole genome association (WGA) mapping................ 245 Wild-card asterix (in SNP mining software/tools) .......... 59 WinMelt (software) ............139, 142, 142 f, 155, 156, 157, 158 f, 161 f, 162 n3, 163 n8 Wnt/b - catenin...................................................... 5, 11, 14 Written informed consent .............................. 172, 185, 395
X X-chromosome skewed........................................................................... 5 Xp11............................................................3 – p4, 5, 11
Y Yin-Yang haplotypes ........................................................ 65
Z ZipChute (software) ....................................................... 247 ZipCode (software)......................................................... 247 Zn(II)–Cyclen Polyacrylamide Gel Electrophoresis (Zn2+–cyclen–PAGE).......................... 169–181 Atto model AE-6500 mini-slab gel system...... 178, 188 dT-rich regions ......................................................... 170 Imide-containing nucleoside .................................... 169 Imide-deprotonated (species) ................................... 170 Zn2+–cyclen (complex)...........................169–170, 170 f Zuker/mfold (software) .......................................... 263, 267