Microarray Methods for Drug Discovery [1 ed.] 1607616629, 9781607616627

Citation preview

METHODS

IN

MOLECULAR BIOLOGY™

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

Microarray Methods for Drug Discovery Edited by

Sridar V. Chittur Department of Biomedical Sciences, School of Public Health, Center for Functional Genomics, University at Albany-SUNY, Rensselaer, NY, USA

Editor Sridar V. Chittur, Ph.D. Department of Biomedical Sciences Schools of Public Health Centre for Functional Genomics University at Albany-SUNY Rensselaer, NY USA [email protected]

ISSN 1064-3745 e-ISSN 1940-6029 ISBN 978-1-60761-662-7 e-ISBN 978-1-60761-663-4 DOI 10.1007/978-1-60761-663-4 Springer New York Dordrecht Heidelberg London Library of Congress Control Number: 2010921137 © Humana Press, a part of Springer Science+Business Media, LLC 2010 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 Science+Business 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. Printed on acid-free paper Humana Press is a part of Springer Science+Business Media (www.springer.com)

Preface The postgenomic era presents a multitude of challenges for scientists in all areas of science. The information overload from new discoveries in genomics and proteomics highlight how little we really know about the functioning of a cell. The advent of Next-Generation Sequencing technologies promises to make our genetic blueprint available to the common man. The availability of the plethora of biological information has lead to the development of new areas of science and the coining of new “omics” terms including transcriptomics, methylomics, toxicogenomics, pharmacogenomics, metabolomics, lipidomics, and so on. Remarkable research is being conducted to understand the various aspects of human health and how processes like histone modifications, promoter usage, alternative splicing, posttranscriptional, and posttranslational modifications contribute to disease. The advent of systems biology has unified chemists and biochemists alike in the struggle to eradicate or treat human disease. Microarrays have blossomed into a fast developing and cutting-edge technology that promises to become a major component of personalized medicine. The 1990s witnessed a boom in many areas including genome sequencing, combinatorial chemistry, and computers, all of which have contributed to the development of microarray technology from its infancy into a mature tool. The growing potential of this tool is evident from the number of publications since 1991 when Fodor et al. of Affymax (now Affymetrix) first described the microarray prototype. The number of publications using microarrays in 1990–1999 was approximately 300, while over 8,200 journal articles have been published in the first half of this year alone. The usage of microarrays in experiments designed to identify differential gene expression is well accepted now. Since the seminal work of Pat Brown’s group at Stanford, microarrays became a technology that could be developed by any individual researcher using simple spotting robots. Currently, few laboratories make their own arrays due to the availability of commercial cost-effective solutions that are less prone to variation. Microarrays have evolved from traditional oligonucleotide arrays for gene expression into tools that have even more fascinating applications. Today, one can find arrays containing not only DNA oligonucleotides but antibodies, carbohydrates, small molecules, and enzymes. The diversity of these applications makes this field exciting and limited only by imagination. This, however, makes it challenging for an inexperienced scientist wishing to enter this arena. I am still surprised by the lack of general information amongst individuals regarding how to design and conduct a microarray experiment. As we get exposed to the concept of “Personalized Medicine”, we find ourselves confounded by the myriad of platforms and applications attributed to microarrays. This book aims at enlightening individuals with all levels of experience about some of the

v

vi

Preface

most common applications of microarrays in drug discovery and development. I hope that this book will serve as a reference for students and scientists alike who would like to enter this exciting field but are a bit intimidated. I am especially grateful to the many friends, colleagues, and family who encouraged me in this effort. Rensselaer, NY

Sridar V. Chittur

Contents Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

v ix

1 Multicenter Clinical Sample Collection for Microarray Analysis . . . . . . . . . . . . . . . Tony S. Mondala, Daniel R. Salomon, and Steven R. Head 2 Isolation of Total RNA from Transgenic Mouse Melanoma Subsets Using Fluorescence-Activated Cell Sorting . . . . . . . . . . . . . . . . . . . . . . . . Scott Tighe and Matthew A. Held 3 Microarray Analysis of Embryonic Stem Cells and Differentiated Embryoid Bodies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Alexander C. Zambon and Christopher S. Barker 4 Determination of Alternate Splicing Events Using the Affymetrix Exon 1.0 ST Arrays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Sita Subbaram, Marcy Kuentzel, David Frank, C. Michael DiPersio, and Sridar V. Chittur 5 Profiling microRNA Expression with the Illumina BeadChip Platform . . . . . . . . . Julissa Tsao, Patrick Yau, and Neil Winegarden 6 TaqMan® Array Cards in Pharmaceutical Research . . . . . . . . . . . . . . . . . . . . . . . . David N. Keys, Janice K. Au-Young, and Richard A. Fekete 7 DMET ™ Microarray Technology for Pharmacogenomics-Based Personalized Medicine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . James K. Burmester, Marina Sedova, Michael H. Shapero, and Elaine Mansfield 8 The Use of Microarray Technology for Cytogenetics. . . . . . . . . . . . . . . . . . . . . . . Bassem A. Bejjani, Lisa G. Shaffer, and Blake C. Ballif 9 PCR/LDR/Universal Array Platforms for the Diagnosis of Infectious Disease . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Maneesh Pingle, Mark Rundell, Sanchita Das, Linnie M. Golightly, and Francis Barany 10 RIP-CHIP in Drug Development . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Ritu Jain, Francis Doyle, Ajish D. George, Marcy Kuentzel, David Frank, Sridar V. Chittur, and Scott A. Tenenbaum 11 ChIPing Away at Global Transcriptional Regulation . . . . . . . . . . . . . . . . . . . . . . . Kelly Jackson, James Paris, and Mark Takahashi 12 HELP (HpaII Tiny Fragment Enrichment by Ligation-Mediated PCR) Assay for DNA Methylation Profiling of Primary Normal and Malignant B Lymphocytes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Rita Shaknovich, Maria E. Figueroa, and Ari Melnick

1

vii

27

45

63

73 87

99

125

141

159

173

191

viii

Contents

13 High-Throughput Screening of Metalloproteases Using Small Molecule Microarrays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Mahesh Uttamchandani 14 Metabolic Enzyme Microarray Coupled with Miniaturized Cell-Culture Array Technology for High-Throughput Toxicity Screening . . . . . . . . . . . . . . . . . Moo-Yeal Lee, Jonathan S. Dordick, and Douglas S. Clark 15 Use of Tissue Microarray to Facilitate Oncology Research. . . . . . . . . . . . . . . . . . . Panagiotis Gouveris, Paul M. Weinberger, and Amanda Psyrri 16 Small Molecule Selectivity and Specificity Profiling Using Functional Protein Microarrays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Peter R. Kraus, Lihao Meng, and Lisa Freeman-Cook 17 Production and Application of Glycan Microarrays . . . . . . . . . . . . . . . . . . . . . . . . Julia Busch, Ryan McBride, and Steven R. Head

203

221 239

251 269

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 283

Contributors JANICE K. AU-YOUNG • Molecular Biology Division, Life Technologies Corporation, Foster City, CA, USA BLAKE C. BALLIF • Signature Genomic Laboratories, Spokane, WA, USA FRANCIS BARANY • Department of Microbiology, Weill Medical College, Cornell University, New York, NY, USA CHRISTOPHER S. BARKER • Gladstone Institute of Cardiovascular Disease, University of California, San Francisco, CA, USA BASSEM A. BEJJANI • Signature Genomic Laboratories, Spokane, WA, USA JAMES K. BURMESTER • Center for Human Genetics, Marshfield Clinic Research Foundation, Marshfield, WI, USA JULIA BUSCH • DNA Array Core Facility, The Scripps Research Institute, La Jolla, CA, USA SRIDAR V. CHITTUR • Department of Biomedical Sciences, Center for Functional Genomics, School of Public Health, University at Albany-SUNY, Rensselaer, NY, USA DOUGLAS S. CLARK • Department of Chemical Engineering, University of California, Berkeley, CA, USA SANCHITA DAS • Department of Microbiology, Weill Medical College, Cornell University, New York, NY, USA C. MICHAEL DIPERSIO • Center for Cell Biology and Cancer Research, Albany Medical College, Albany, NY, USA JONATHAN S. DORDICK • Department of Chemical and Biological Engineering, Rensselaer Polytechnic Institute, Troy, NY, USA FRANCIS DOYLE • NanoBio Constellation, College of Nanoscale Science and Engineering, University at Albany-SUNY, Albany, NY, USA RICHARD A. FEKETE • Molecular Biology Division, Life Technologies Corporation, Foster City, CA, USA MARIA E. FIGUEROA • Division of Hematology/Oncology, Department of Medicine, Weill Medical College, Cornell University, New York, NY, USA DAVID FRANK • Department of Biomedical Sciences, Center for Functional Genomics, School of Public Health, University at Albany-SUNY, Rensselaer, NY, USA LISA FREEMAN-COOK • Life Technologies, Carlsbad, CA, USA AJISH D. GEORGE • Department of Biomedical Sciences, Gen*NY*Sis Center for Excellence in Cancer Genomics, School of Public Health, University at Albany-SUNY, Rensselaer, NY, USA LINNIE M. GOLIGHTLY • Department of Microbiology, Weill Medical College, Cornell University, New York, NY, USA PANAGIOTIS GOUVERIS • Division of Hematology Oncology, Department of Internal Medicine, Yale University, New Haven, CT, USA

ix

x

Contributors

STEVEN R. HEAD • DNA Array Core Facility, The Scripps Research Institute, La Jolla, CA, USA MATTHEW A. HELD • Departments of Dermatology and Pathology, Yale University School of Medicine, New Haven, CT, USA KELLY JACKSON • UHN Microarray Center, Toronto, ON, Canada RITU JAIN • NanoBio Constellation, College of Nanoscale Science and Engineering, University at Albany-SUNY, Albany, NY, USA DAVID N. KEYS • Molecular Biology Division, Life Technologies Corporation, Foster City, CA, USA PETER R. KRAUS • Life Technologies, Carlsbad, CA, USA MARCY KUENTZEL • Department of Biomedical Sciences, Center for Functional Genomics, School of Public Health, University at Albany-SUNY, Rensselaer, NY, USA MOO-YEAL LEE • Solidus Biosciences, Inc., Troy, NY, USA ELAINE MANSFIELD • Application Sciences Department, Affymetrix, Inc., Santa Clara, CA, USA RYAN MCBRIDE • DNA Array Core Facility, The Scripps Research Institute, La Jolla, CA, USA ARI MELNICK • Division of Hematology/Oncology, Department of Medicine, Weill Medical College, Cornell University, New York, NY, USA LIHAO MENG • Life Technologies, Carlsbad, CA, USA TONY S. MONDALA • The Scripps Research Institute, La Jolla, CA, USA JAMES PARIS • UHN Microarray Center, Toronto, ON, Canada MANEESH PINGLE • Department of Microbiology, Weill Medical College, Cornell University, New York, NY, USA AMANDA PSYRRI • Division of Hematology Oncology, Department of Internal Medicine, Yale University, New Haven, CT, USA MARK RUNDELL • Department of Microbiology, Weill Medical College, Cornell University, New York, NY, USA DANIEL R. SALOMON • The Scripps Research Institute, La Jolla, CA, USA MARINA SEDOVA • Assay and Application Product Development, Affymetrix, Inc, Santa Clara, CA, USA LISA G. SHAFFER • Signature Genomic Laboratories, Spokane, WA, USA RITA SHAKNOVICH • Division of Immunopathology, Department of Pathology, Weill Medical College, Cornell University, New York, NY, USA; Division of Hematology/Oncology, Department of Medicine, Weill Medical College, Cornell University, New York, NY, USA MICHAEL H. SHAPERO • Assay and Application Product Development, Affymetrix, Inc, Santa Clara, CA, USA SITA SUBBARAM • Center for Cell Biology and Cancer Research, Albany Medical College, Albany, NY, USA MARK TAKAHASHI • UHN Microarray Center, Toronto, ON, Canada SCOTT A. TENENBAUM • NanoBio Constellation, College of Nanoscale Science and Engineering, University at Albany-SUNY, Albany, NY, USA SCOTT TIGHE • Microarray Core Facility, University of Vermont, College of Medicine, Burlington, VT, USA

Contributors

JULISSA TSAO • UHN Microarray Center, Toronto, ON, Canada MAHESH UTTAMCHANDANI • Defense Medical and Environmental Research Institute (DMERI), DSO National Laboratories, Singapore; Department of Chemistry, National University of Singapore, Singapore PAUL M. WEINBERGER • Department of Otolaryngology, Medical College of Georgia, Augusta, GA, USA NEIL WINEGARDEN • UHN Microarray Center, Toronto, ON, Canada PATRICK YAU • UHN Microarray Center, Toronto, ON, Canada ALEXANDER C. ZAMBON • Department of Pharmacology, University of California, San Diego, La Jolla, CA, USA

xi

Chapter 1 Multicenter Clinical Sample Collection for Microarray Analysis Tony S. Mondala, Daniel R. Salomon, and Steven R. Head Abstract In this chapter, we describe numerous methods to extract RNA, DNA, and protein from tissue, represented by kidney transplant biopsies, and from peripheral blood cells collected at various clinical sites. Gene expression profiling and SNP-based genome-wide association studies are done using various microarray platforms. In addition, protocols that enable simultaneous protein purification from these clinical samples, enable additional strategies for understanding of the molecular processes involved in organ transplantation, immunosuppressive drug regimens, and the elements determining allograft success and failure. Successfully establishing a multicenter clinical study was essential to meet our objectives for subject enrollment and transplant outcomes. This chapter focuses on our experience setting up and coordinating clinical sample collection from multiple transplant centers for the purpose of microarray analysis. Key words: Microarrays, Genomics, Transplantation, Multicenter clinical study, Nucleic acid extraction, Protein extraction

1. Introduction The analysis of clinical samples utilizing microarray technology has advanced the field of clinical research including transplantation medicine (1–6). Our research group, the Transplant Genomics Collaborative Group (TGCG; http://www.genetics. ucla.edu/transplant-genomics/index2.php) is involved in a large study of kidney transplantation outcomes with an emphasis on defining genomic biomarkers that could be used to monitor and individualize the adequacy and efficacy of immunosuppressive drug therapy. Organizing the Transplant Genomics project has provided a better understanding of the challenges facing anyone

Sridar V. Chittur (ed.), Microarray Methods for Drug Discovery, Methods in Molecular Biology, vol. 632, DOI 10.1007/978-1-60761-663-4_1, © Humana Press, a part of Springer Science+Business Media, LLC 2010

1

2

Mondala, Salomon, and Head

planning a large, multicenter clinical project that involves collection of multiple sample types, at multiple predefined time points, with multiple sample handling protocols and where obtaining precise clinical data and outcomes are necessary. The key to success of any study involving multiple clinical centers is the efficient collection, preservation, and transport of clinical material from collection sites to a central processing facility where samples can be prepared for analysis using microarrays and other analytical techniques. Organizing a multicenter research study requires effective training and support of physicians, nurse coordinators, and laboratory personnel in order to guarantee adherence to enrollment (inclusion/exclusion) criteria, proper sample collection, preshipment specimen processing, and documentation of patient/subject data. A key insight is that physicians, nurses, and laboratory personnel all have different jobs, training, and work environments so that strategies to effectively communicate study objectives and monitor sample collection and data integrity must be developed for each. As with any research study involving human subjects, Institutional Review Board approval of a Human Subjects Protocol is required as well as informed consent for study participants. Setting up a central processing center is necessary to create, test, and then provide kits for sample collection and transport. The central processing center is also tasked with the tracking of all collected specimens from the various clinical centers as well as coordinating sample preparation and archiving. Finally, it is critical to have a highly secure clinical database that is readily accessible to all the participating centers and a parallel, but integrated, specimen tracking database in the central processing center. The wealth of scientific information that can be obtained through the establishment of a well-organized system to collect, document, and process clinical research samples provides a foundation for advancing clinical research and translational medicine. The purpose of this chapter is to discuss our experience in a large multicenter clinical study, specifically the Transplant Genomics project. 1.1. Setting up the Clinical Centers

Before recruitment of candidate clinical centers can begin, it is imperative to have reagents and protocols thoroughly established and validated for sample collection, processing at the collection sites, storage and shipment to the central processing center. In order to collect kidney biopsies and blood samples from transplant recipients and donors, we prepared kits containing suitable containers to hold the specimens along with detailed instructions written in plain language as well as illustrations for collecting, processing (when needed), and temporarily storing specimens at the clinical sites. A second set of kits were prepared containing all the necessary components to facilitate shipping samples from the clinical sites to the central processing center. Finally, a specimen labeling system, which includes bar coding, was implemented

Multicenter Clinical Sample Collection for Microarray Analysis

3

to facilitate linking specific specimens with the clinical data for each patient. In our experience, every clinical center is going to have specific needs and support requirements. We have found it helpful to have a single person in the central processing team responsible for any given center to evaluate the capabilities of the center, provide on-site training, phone, and email support. In addition, this person is tasked with monitoring problems in sample collection, enrollment, processing, and shipping as well as ensures patient data associated with each sample are properly entered into a clinical database and timely follow-up and outcome data are also obtained per the study protocols. The role of supporting clinical centers is a key component and should not be under resourced. For our Transplant Genomics project, we ask each clinical center to collect whole blood and core needle kidney biopsies from kidney transplant recipients at specific time points or at the time of specific events such as acute rejection. We also require the clinical centers to process a portion of the whole blood into purified lymphocytes and plasma. This requires centrifugation of whole blood within 2 h of collection followed by separation of the lymphocyte and plasma fractions. Therefore, to be considered for inclusion in our study, we required each candidate clinical center to be capable of performing this procedure. Clinical centers are frequently staffed by nurses with varying levels of laboratory experience and competing demands on their time. In some cases, all on-site sample processing is done by clinical laboratory staff and in other cases, it is done by nursing staff. This adds complexity to the process of establishing clinical center-specific procedures required to efficiently collect, process, and store samples. However, an enthusiastic and highly motivated staff at the clinical site can often find creative solutions to complete the tasks within the required parameters. One key component of collecting samples from clinical centers is the parallel collection of accurate and complete patient data and records. We designed an online database accessible via a secure web portal using a 256-bit Secure Socket Layer (SSL) encryption protocol. On-site training in clinical data entry for the project was a key and final step in launching the study at each clinical center and is repeated whenever local staff changes. The clinical database, maintained by a database administrator at our center, is accessible only by authorized personnel at each clinical center with individual usernames and passwords. To further protect information confidentiality, clinical center staff can only access patient-identified data from their own clinical center. It is critical to continuously monitor the patient data entry to ensure that every patient and sample has completed records for all sample collections as well as any follow-up clinical information relevant to the study. Again, this should not be under resourced as it is often impractical to retroactively fill in required patient data and records.

4

Mondala, Salomon, and Head

1.2. Institutional Review Boards, HIPAA, and Compensation for Participation

Institutional Review Boards (IRBs) are critical to the integrity of clinical research. A comprehensive discussion of the importance and functions of the IRB is beyond the scope of this chapter. For an excellent starting source of information, the reader is referred to the FDA website (http://www.fda.gov/oc/ohrt/irbs/). In the case of a multicenter clinical trial, the IRB of the primary project center takes the lead in the first review and approval of the research protocols. But this approval is based on all the procedures agreed upon by the local IRB of the primary center. Next, the research protocols must be reviewed and approved by each participating center’s IRB and again, this involves all the local procedures for these different IRBs. Thus, the format of the protocol submissions can be very different in every center, the requirements for the informed consents can be different, and a local IRB can raise issues with any element of the research independent of any other IRB’s approval. Our experience is that several candidate clinical centers were never able to complete their IRB reviews. A common challenge was our study’s requirements for DNA collections that raised concerns over unwanted dissemination of genetic information, an issue that many IRBs had never confronted. Another critical element of planning are the provisions of the Health Insurance Portability and Accountability Act (HIPAA). Again, it is beyond the scope here to detail HIPAA’s provisions and the website of U.S. Department of Health and Human Services is an excellent first source (http://www.hhs.gov/ocr/ privacy/index.html). However, as a starting point, it is necessary to create a strategy to protect patient-identified data so that no one can access this information beyond the authorization given by the IRBs to the principal investigators and key personnel of the project as well as relevant regulatory governmental agencies (e.g., FDA). Such strategies include assigning coded patient identification numbers to decouple sample IDs from patient names. Thus all downstream processing of samples and analysis of data by the technicians and scientists is done with no patient-identifiable information, just anonymous alpha-numeric codes. We purposefully created two project databases to facilitate our compliance with HIPAA. The first was the highly secure clinical database that is accessible by password protection at each site via a web portal. This database contains all the sensitive patient identified information. The second database is designed only for sample tracking, disposition, and archiving. It has no patient-identified data and all samples are listed as anonymous alpha-numeric codes assigned to each sample collected and bar-coded. The two databases are linked through the clinical database where the key to the alpha-numeric codes are kept. Finally, we encountered a number of issues involving compensation of research subjects. For one example, different transplant

Multicenter Clinical Sample Collection for Microarray Analysis

5

programs had different standards of practice. In programs like ours that does serial monitoring transplant biopsies in all patients, we could not compensate our research subjects for the biopsy due to issues of fairness. In contrast, in clinical centers where such serial biopsies are not standard of practice, the local IRBs did allow subject compensation for research participation as long as it was reasonable, not coercive, and covered acceptable things such as loss of work time, travel expenses, childcare etc. In general, the subject of compensating subjects for research remains a very sensitive issue and every IRB has different views of what is reasonable. 1.3. Central Processing Laboratory

Setting up a central processing laboratory is essential for coordinating the shipment, tracking, and processing of clinical samples. In addition, the preparation and validation of sample collection and sample shipping kits is most efficiently accomplished by the same laboratory facility that will be receiving and processing the samples. A key point is that all protocols, reagents, and kits must be fully developed, tested, and ready for implementation by every clinical center before on-site training and study initiation. It is a major error to launch a project and then find out that one or more elements of the sample collection protocol require significant changes. Another important role of the central processing facility is to ensure that the clinical centers are always adequately supplied with kits for sample collection and shipping. In kidney transplantation, newly enrolled patients are frequently in need of urgent care and if the clinical center is missing specimen collection kits we miss a valuable opportunity. Frequent communication with the clinical center staff ensures adequate supplies are always available. Since most biological specimens are perishable, specimen shipments by clinical centers should be made early in the week to prevent weekend deliveries. All shipments must be immediately unpacked and safely stored until they can be logged into a specimen tracking database and processed. We created a separate database (linkable to the clinical database by several key variables) for tracking the arrival of samples and following their progress through extraction of RNA, DNA, and protein from blood, cell pellets, and kidney biopsy material. In addition, the sample tracking database records how much and where aliquots of RNA, DNA, and protein are shipped for further analysis. Finally, the archived sample amounts and locations (freezer, drawer, box, and position) are recorded in the specimen tracking database. The use of a bar coding system facilitates tracking the processing and archiving of clinical specimens. We use a single 5 digit barcode number to label all collection tubes within a specific kit. Each kit is used for all specimens collected from a single patient at a single collection time point. The kit barcode is recorded into the clinical database by clinical center staff at the time of sample

6

Mondala, Salomon, and Head

procurement in order to provide a unique identifier that allows the clinical database data to be linked to the data in the specimen tracking database. When specimens are processed into RNA, DNA, and protein at the central processing laboratory, a one letter prefix is added to the barcode (new labels are printed) to identify the sample type and source (i.e., RNA derived from whole blood vs. RNA derived from biopsy core). An additional aliquot number as a suffix is also added (5 digits) to the barcode, which is now one letter and 10 digits long, to identify each specific tube and facilitate tracking the shipping or archiving of each aliquot. We use several protocols for extraction of RNA, DNA, protein, and plasma from whole blood, cell pellets, and biopsies, which are described below in Subheadings 2 and 3.

2. Materials 2.1. Specimen Collection Kits

1. PAXgene Blood RNA tube, 2.5 ml (Qiagen). 2. Vacutainer Cell Preparation tube (CPT) with sodium citrate, 8 ml (Becton-Dickenson). 3. Vacutainer Plasma Preparation tube (PPT) with EDTA, 5 ml (Becton-Dickenson). 4. RNAlater (Ambion). 5. Phosphate buffered saline, pH 7.2 (Invitrogen). 6. 2.0 and 4.0 ml cryovials (USA Scientific). 7. Color cap inserts (USA Scientific). 8. 15 and 50 ml conical tubes (USA Scientific). 9. 3 ml disposable transfer pipet (VWR). 10. Cardboard boxes (Office Depot). 11. Barcode label printer (TLS PC Link, Brady Worldwide). 12. Barcode labels (PTL-76-461, Brady Worldwide) (see Note 1).

2.2. RNA Extraction from PAXgene Blood Samples

1. PAXgene Blood RNA Kit (Qiagen). 2. Ethanol (100%) (Sigma). 3. GLOBINclear Kit – Human (Ambion). 4. Isopropanol (100%) (Sigma). 5. Magnetic Stand (Ambion).

2.3. RNA, DNA, and Protein Extraction from Mononuclear Cells

1. AllPrep DNA/RNA/Protein Mini Kit (Qiagen). 2. 14.3 M b-mercaptoethanol (Sigma). 3. Ethanol (100%) (Sigma). 4. 5 ml syringes and 18G needles (Becton-Dickenson).

Multicenter Clinical Sample Collection for Microarray Analysis

2.4. Plasma Separation from Whole Blood Samples (CPT)

1. 3 ml disposable transfer pipet (VWR).

2.5. RNA, DNA, and Protein Extraction from Biopsies

1. Trizol Reagent (Invitrogen).

7

2. 2 ml Wheaton Plastic Coated Tissue Grinder with Teflon Pestles (VWR). 3. Chloroform (Sigma). 4. Isopropanol (100%) (Sigma). 5. Ethanol (100, 80, and 75%) (Sigma). 6. DEPC-treated water (Ambion). 7. RNeasy Kit (Qiagen). 8. 14.3 M b-mercaptoethanol (Sigma). 9. 0.1 M NaCitrate in 10% ETOH. 10. 8 mM NaOH. 11. Phase Lock Gel tube (Eppendorf). 12. Phenol Chloroform (Ambion). 13. 3 M Sodium Acetate. 14. Glycogen (5 mg/ml) (Ambion). 15. 0.3 M Guanidine HCl.

2.6. DNA Extractions from Whole Blood

1. QIAamp DNA Blood Midi Kit – 100 (Qiagen). 2. Ethanol (100%). 3. 15 ml Centrifuge tubes (USA Scientific).

2.7. DNA Extractions from Mononuclear Cells

1. QIAamp DNA Mini Kit – 50 (Qiagen). 2. Ethanol (100%). 3. Phosphate buffered saline, pH 7.2 (Invitrogen).

3. Methods The proper procurement and handling of human blood and tissues for translational studies is critical to ensure that the quality of the specimen remains intact and suitable for downstream analysis. We discuss below in detail how the blood and biopsies are collected and handled in transport and the extraction of RNA, DNA, and protein. 3.1. Blood and Tissue Procurement

Each clinical center is supplied with multiple sample collection kits for procurement of specimens. Each kit (7 × 4 × 3 in. cardboard box) contains two 2.5 ml PAXgene tubes, one 8.5 ml CPT

8

Mondala, Salomon, and Head

(Cell Preparation Tube) tube, two 4.5 ml PPT (Plasma Preparation Tube) tubes, one 2 ml cryovial containing 1 ml of RNAlater for holding the biopsy core, one 50 ml conical tube containing 35 ml of PBS for washing the cell pellet obtained from the CPT tube, one empty 15 ml conical tube for washing the cell pellet from the CPT tube, one empty 4 ml cryovial tube to hold the plasma from the CPT tube, one 2 ml cryovial tube containing 1.5 ml RNAlater to resuspend the washed cell pellet, one empty 2 ml cryovial to hold the cells resuspended in RNAlater, and one transfer pipet used to transfer plasma from the CPT tube to the 4 ml cryovial. In addition, a detailed set of instructions for specimen collection, processing, and temporary storage at the clinical center are included as well as a laboratory requisition form used to document the date, time, and person collecting and processing the specimens. Finally a set of barcode labels are provided to attach to each collection tube and vial. This barcode information specifies a 5 digit number assigned to the specific collection kit and allows us to uniquely label specimens from a specific patient and collection time point. This barcode is subsequently entered into both the clinical and specimen tracking databases. Standard phlebotomy technique is utilized in collecting the whole blood samples. 2.5 ml of blood is collected into each of two PAXgene tubes and immediately inverted several times to efficiently mix the blood with the RNA stabilizing reagent. This mixing step is critical. Nursing and phlebotomy staff must be made aware of the importance of mixing the tubes immediately after draw. 8.5 ml of blood is then collected into the CPT tube and 4.5 ml of blood into each of two PPT tube. As with the PAXgene tubes, the CPT and PPT tubes must be inverted immediately several times to ensure the blood and anti-coagulant is thoroughly mixed (see Note 2). Core needle kidney biopsies are collected by trained transplant physicians or radiologists. Biopsies are immediately submerged in 1 ml of RNAlater in a 2 ml cryovial and stored at 4°C overnight and then frozen at –20°C the following day (see Note 3). The blood collection tubes are transported to the onsite processing laboratory for further processing and storage until shipment to the Central Processing Laboratory. 3.2. Transport

Each clinical center is provided with sample shipping supplies. This includes detailed shipping instructions, biohazard plastic zip lock bags, absorbent tube sleeves, FedEx forms, dry ice labels, diagnostic specimen labels (UN3373), styrofoam shipping boxes contained within a secondary cardboard box. All frozen samples are shipped to the central processing center with a sufficient amount of dry ice for next day delivery. Packages are shipped only on Mondays, Tuesdays, and Wednesdays to lessen the chance that the shipment would be delayed over a weekend.

Multicenter Clinical Sample Collection for Microarray Analysis

9

When packaging the frozen specimens for shipment, it is very important to make sure that the frozen tubes are properly protected from damage by rough handling during transport. The frozen tubes are very brittle and prone to crack thus they should be packaged carefully cushioned in a smaller box that is then placed in the larger styrofoam box that also contains the dry ice. This also prevents the larger pieces of dry ice from crushing the frozen tubes. The clinical center documents in the clinical database when specimens are shipped to the central processing center. The database then automatically sends an email notification to the central processing center that a shipment is on its way and to expect arrival the following day. 3.3. RNA, DNA, and Protein Extraction

Once the samples are received at the central processing center they are immediately unpacked and placed in –20°C storage for up to 60 days until further processing. When samples are ready for processing the specimens are removed from the freezer and thawed on ice. The following protocols are used to extract RNA, DNA, and protein from whole blood, cell pellet, and biopsy tissue. Once the extraction procedure is completed each sample type (RNA, DNA and protein) is assigned a new barcode consisting of the original 5-digit barcode plus an additional letter prefixed which identifies the sample type as well as assigning a unique 5 digit aliquot identification number. At this point, the samples are archived at –80°C and ready for downstream analysis by microarray and other analytical techniques. At the time this project started, we utilized Trizol reagent to extract RNA, DNA, and protein from both cells and biopsies. Subsequently,thecreationoftheQiagenAllPrepDNA/RNA/Protein Mini kit allowed us to extract all three fractions with improved efficiency. This is a simplified protocol, takes less time, avoids the use of toxic phenol and chloroform, does not require ethanol precipitation, and it allows greater ease when processing multiple samples simultaneously. The DNA extracted using AllPrep also proved to be better suited for genotyping using the Affymetrix SNP microarrays than DNA extracted using Trizol reagent. This was observed in the increased SNP call rates for DNA samples extracted using the AllPrep method. In our experience biopsies processed using the AllPrep did not yield RNA, DNA, and protein in amounts comparable to the Trizol method. Thus for biopsies we continued to use the Trizol method and for mononuclear cells we have adapted the AllPrep method. Table 1 shows the different extraction protocols used for both biopsies and cells and the downstream applications for each fraction. In our Transplant Genomics project, we also collect extracted donor and recipient DNA specimens for analysis on genome-wide Affymetrix SNP microarrays. In cases where only peripheral blood

10

Mondala, Salomon, and Head

Table 1 Extraction methods utilized based on starting tissue type, desired purified fraction, and downstream analysis platform Tissue type Kidney biopsy

Mononuclear cells

Extraction method

Sample type Application

Invitrogen Trizol reagent

RNA

Obtained higher yields Gene expression: (Affymetrix than the Qiagen Human Genome U133 AllPrep Method Plus 2.0 Array) Whole-transcript gene expression and alternative splicing: (Affymetrix Human Exon 1.0 ST Array; Human Gene1.0 ST Array)

DNA

SNP based genome-wide association studies: (Affymetrix Genome-Wide Human SNP Array 6.0)

Affymetrix SNP call rates were not as high compared to DNA derived using Qiagen AllPrep Method

Protein

Proteomic analysis: (MudPIT Tandem Mass Spectrometry, quantification, expression)

Obtained higher yields than the Qiagen AllPrep Method

RNA

Gene expression: (Affymetrix Human Genome U133 Plus 2.0 Array) Whole-transcript gene expression and alternative splicing: (Affymetrix Human Exon 1.0 ST Array; Human Gene1.0 ST Array)

Obtained yields comparable to the Trizol method. All QC metrics similar to that derived using the Trizol method

DNA

SNP based genome-wide association studies: (Affymetrix Genome-Wide Human SNP Array 6.0)

Obtained yields comparable to the Trizol method. Affymetrix SNP call rates were higher than the Trizol method

Protein

Proteomic analysis: (MudPIT Tandem Mass Spectrometry, quantification, expression)

Obtained yields similar to the Trizol method

SNP based genome-wide association studies: (Affymetrix Genome-Wide Human SNP Array 6.0)

Extraction method used when only DNA is required for SNP genotyping

Qiagen Allprep DNA/ RNA/ Protein Kit

DNA Qiagen, Whole blood QIAamp, or DNA Blood mononuclear Midi and cells DNA Mini Kit

Notes

Multicenter Clinical Sample Collection for Microarray Analysis

11

lymphocytes or frozen anti-coagulated whole blood is available such as those found in most tissue typing laboratory archives we extract DNA using the QIAamp DNA kits from Qiagen. From a Vacutainer tube, 4 ml draw with EDTA we use the QIAamp Blood Midi Kit to extract DNA. When we have isolated lymphocytes to extract DNA from we use the QIAamp Mini Kit. The specific protocols for extraction of RNA, DNA, and protein are described in a following section. 3.3.1. RNA Extraction from PAXgene Blood Tubes

The manufacturer (Qiagen) recommended protocol was followed. PAXgene Blood RNA tubes are intended for the collection of whole blood and the stabilization of cellular RNA for up to 3 days at 18–25°C or up to 5 days at 2–8°C. Tubes should be stored at –20 to –80°C for longer periods of time. Draw 2.5 ml of blood directly into PAXgene tube and invert the tube 10 times immediately, do not shake (see Note 4). 1. Buffer BR4 is supplied as a concentrate. Before using for the first time, add 4 volumes of ethanol (100%) as indicated on the bottle to obtain a working solution. Buffer BR2 may form a precipitate upon storage, warm to 37°C to dissolve if necessary. 2. Prepare DNase I stock solution when using the RNase-Free DNase set for the first time. Dissolve the solid DNase I in 550 ml of RNase-free water provided. Take care that no DNase I is lost when opening the vial. Mix gently by inverting the tube. Do not vortex (see Note 5). 3. All centrifugation steps for this protocol are done at room temperature. 4. Centrifuge the PAXgene tube containing 2.5 ml of blood for 10 min at 3,500 × g, brake on using a swing-out rotor with adapters for round-bottom tubes. 5. Remove the supernatant by decanting and discard the supernatant. Dry the rim of the tube with a Kim wipe. Add 4 ml of RNase free water to the pellet and cover the tube using a new, fresh secondary Hemogard closure provided with the kit. 6. Vortex until the pellet is visibly dissolved. Centrifuge for 10 min at 3,500 × g, brake on using a swing-out rotor. Remove the supernatant by decanting and discard the supernatant. Dry the rim of the tube with a Kim wipe. 7. Add 350 ml Buffer BR1 and vortex until the pellet is visibly dissolved. 8. Pipet the sample into a 1.5 ml microcentrifuge tube. Add 300 ml Buffer BR2 and 40 ml proteinase K. Mix by vortexing for 5 s and incubate for 10 min at 55°C using a shakerincubator at 1,400 rpm.

12

Mondala, Salomon, and Head

9. Pipet the lysate directly into a PAXgene Shredder column placed in a 2 ml processing tube and centrifuge for 3 min at maximum speed (not to exceed 20,000 × g). Carefully transfer the entire supernatant of the flow through fraction to a fresh 1.5 ml microcentrifuge tube without disturbing the pellet in the processing tube. 10. Add 350 ml ethanol (100%). Mix by vortexing and quick spin for only 1–2 s to collect the droplets from inside of the tube lid. 11. Add 700 ml of the sample to a PAXgene RNA spin column placed in a 2 ml processing tube and centrifuge for 1 min at maximum speed. Place the spin column in a new 2 ml processing tube and discard the old processing tube containing the flow-through. 12. Add remaining sample to the spin column and centrifuge for 1 min at maximum speed. Place the spin column in a new 2 ml processing tube and discard the old processing tube containing the flow-through. 13. Add 350 ml Buffer BR3 to the spin column. Centrifuge for 1 min at maximum speed. Transfer the spin column to a new processing tube and discard the old processing tube containing the flow-through. 14. Add 73.5 ml Buffer RDD to each thawed 10.5 ml DNase I stock solution aliquot. A single DNase I aliquot per PAXgene tube being processed. Mix by gently flicking the tube, do not vortex. Centrifuge briefly to collect residual liquid from the sides of the tube. 15. Add 80 ml of the DNase I incubation mix directly onto the PAXgene RNA spin column membrane and incubate at room temperature for 15 min. 16. Add 350 ml Buffer BR3 to the spin column. Centrifuge for 1 min at maximum speed. Place the spin column in a new 2 ml processing tube and discard the old processing tube containing the flow-through. 17. Add 500 ml Buffer BR4 to the spin column and centrifuge for 1 min at maximum speed. Place the spin column in a new 2 ml processing tube and discard the old processing tube containing the flow-through. 18. Add another 500 ml BR4 to the spin column and centrifuge for 3 min at maximum speed. Place the spin column in a new 2 ml processing tube and discard the old processing tube containing the flow-through. Centrifuge for 1 min at maximum speed to dry the spin column membrane. 19. Place the spin column in a 1.5 ml microcentrifuge and discard the old processing tube. Add 40 ml Buffer BR5 directly onto

Multicenter Clinical Sample Collection for Microarray Analysis

13

the spin column membrane. Centrifuge for 1 min at maximum speed to elute the RNA. Do not discard the eluate. 20. Add another 40 ml of Buffer BR5 directly onto the spin column membrane. Centrifuge for 1 min at maximum speed. Discard the spin column. 21. Incubate the eluate for 5 min at 65°C in a constant temperature incubator. Do not exceed incubation time or temperature. After incubation immediately chill on ice. 22. Quantify RNA using NanoDrop or other spectrophotometer, blank instrument with Buffer BR5. A 260/280 absorbance ratio between 1.8 and 2.0 is typical of pure RNA. Assess quality of the RNA using the Agilent Bioanalyzer. If the RNA will not be used immediately, store at –80°C. Typical RNA yield from a single PAXgene tube is 4–10 mg. 3.3.2. Globin Reduction of RNA Derived from PAXgene Blood Tubes

It is known that the presence of globin mRNA in total RNA samples derived from whole blood can reduce detection sensitivity when using gene expression arrays as seen as a decrease in present calls and an increase in signal variation (7–10). We followed the manufacturer (Ambion) recommended protocol for reducing the presence of globin mRNA in our total RNA samples derived from PAXgene RNA blood tubes. This protocol utilizes magnetic beads and biotin/streptavidin binding to remove 95% or more of alpha and beta globin mRNA from whole blood derived total RNA samples. 1. Set constant temperature incubators to 50 and 58°C. 2. Prior to starting the procedure, prepare the following reagents. Add 2 ml isopropanol (100%) to the bottle labeled RNA Binding Buffer. Concentrate, mix well, and mark the label to indicate that the isopropanol was added. Add 4 ml ethanol (100%) to the RNA Wash Solution Concentrate bottle, mix well, and indicate on the label that the ethanol was added. 3. Prepare Bead Resuspension Mix prior to starting procedure by combining in a 1.5 ml microcentrifuge tube, 10 ml of RNA Binding Beads (mix thoroughly by vortexing before dispensing) and 4 ml of RNA Bead Buffer for a single reaction, mix briefly, then add 6 ml of isopropanol (100%), mix by vortexing. Scale volumes for multiple reactions include 5% overage for pipetting error. 4. Prepare Streptavidin Magnetic Beads prior to starting procedure by warming the 2× Hybridization Buffer and the Streptavidin Bead Buffer to 50°C for at least 15 min and vortex well before use. Vortex the tube of Streptavidin Magnetic Beads and aliquot in to a 1.5 ml microcentrifuge tube 30 ml for each sample to be processed. Briefly centrifuge for less than 2 s at low speed to collect the mixture at the bottom of the tube.

14

Mondala, Salomon, and Head

Place the tube on a magnetic stand to capture the Streptavidin Magnetic Beads. Leave the tube until the mixture becomes transparent (~5 min). Carefully aspirate the supernatant using a pipet without disturbing the Streptavidin Magnetic Beads. Discard the supernatant and remove the tube from the magnetic stand. Add Streptavidin Bead Buffer to the Streptavidin Magnetic Beads, use a volume equal to the original volume of Streptavidin Magnetic Beads. Vortex vigorously until beads are resuspended and keep at 50°C for at least 15 min before being used later in procedure. 5. Warm Elution Buffer to 58°C prior to using later in procedure. 6. Combine 1–10 mg human whole blood total RNA (in a maximum volume of 14 ml) with 1 ml of Capture Oligo Mix in a 1.5 ml microcentrifuge tube. Add nuclease-free water to the sample mixture as necessary to a final volume of 15 ml. 7. Add 15 ml of 50°C 2× Hybridization Buffer, vortex briefly to mix, and centrifuge briefly for less than 2 s at low speed to collect contents in the bottom of the tube. Incubate at 50°C for 15 min. 8. Remove the prepared Streptavidin Magnetic Beads from the 50°C incubator and resuspend them by gentle vortexing. Briefly centrifuge for less than 2 s at low speed. Add 30 ml of prepared Streptavidin Magnetic Beads to each RNA sample, vortex to mix well, and centrifuge briefly for less than 2 s at low speed. Flick the tube very gently to resuspend the beads, being careful to keep the contents at the bottom of the tube. Incubate at 50°C for 30 min. 9. Remove sample and vortex briefly to mix, centrifuge for less than 2 s at low speed. Capture the Streptavidin Magnetic Beads on a magnetic stand. Leave the tube until the mixture becomes transparent (~5 min). Carefully draw up the supernatant, which contains the globin mRNA depleted RNA, and transfer the RNA to a new 1.5 ml microcentrifuge tube. Place RNA on ice, discard the tube with the Streptavidin Magnetic Beads. 10. Add 100 ml RNA Binding Buffer to each sample. Vortex the Bead Resuspension Mix to resuspend the beads thoroughly and immediately dispense 20 ml to each sample. Vigorously vortex the sample for 10 s, briefly centrifuge for less than 2 s at low speed. 11. Capture the RNA Binding Beads by placing the tube on a magnetic stand. Leave the tube until the mixture becomes transparent (~5 min). Carefully aspirate the supernatant without disturbing the RNA Binding Beads and discard the supernatant (it is important to remove as much of the supernatant as possible). Remove the tube from the magnetic stand.

Multicenter Clinical Sample Collection for Microarray Analysis

15

12. Add 200 ml RNA Wash Solution to each sample and vortex for 10 s. Briefly centrifuge for less than 2 s at low speed. Capture the RNA Binding Beads on a magnetic stands as in the previous magnetic bead capture steps. Carefully aspirate and discard the supernatant and remove the tube from the magnetic stand. Briefly centrifuge the tube as in previous steps and place it back on the magnetic stand. Remove any liquid in the tube with a small-bore pipet tip, remove the tube from the magnetic stand, and allow the beads to air-dry for 5 min with the caps left open (see Note 6). 13. Add 30 ml warm Elution Buffer to each sample and vortex vigorously for 10 s to thoroughly resuspend the RNA Binding Beads. Incubate at 58°C for 5 min. Vortex the sample vigorously for 10 s to thoroughly resuspend the RNA Binding Beads and centrifuge for less than 2 s at low speed. 14. Capture the RNA Binding Beads on a magnetic stand as in the previous magnetic bead capture steps. Be especially careful at this step to avoid disturbing the RNA Binding Beads when collecting the supernatant. The purified RNA will be in the supernatant, and transfer to a new 1.5 ml microcentrifuge tube (frequently some of the RNA Binding Beads are carried over to the eluate, tinting it brownish but this does not affect absorbance or downstream applications). 15. Quantitate RNA using the NanoDrop or any other spectrophotometer, blank the instrument with Elution Buffer. A 260/280 absorbance ratio between 1.8 and 2.0 is typical of pure RNA. Assess quality of the RNA using the Agilent Bioanalyzer. If the RNA will not be used immediately, store at –80°C. Processing whole blood total RNA with the GLOBINclear Human Kit can reduce RNA yield by as much as 30% though we have observed average decrease in yield of about 15% (see Note 7). 3.3.3. Isolation of Mononuclear Cells and Separation of Plasma from CPT Tubes and the Processing of the PPT Tubes

The CPT tube and the 2 PPT tubes are centrifuged (swing-bucket rotor) at room temperature, 20 min at 1,700 × g. It is important to centrifuge the CPT tube within 2 h of draw as cell yields considerably decline and red blood cell contamination of the cell fraction increases thereafter. (http://www.bd.com/vacutainer/ products/molecular/citrate/limitations.asp). After centrifugation the 2 PPT tubes are frozen upright in a –20°C freezer. As for the CPT tube, the stopper is removed and approximately 3 ml of clear plasma from the upper phase is transferred to a 4 ml cryovial and then frozen at –20°C. The CPT tube is then recapped with the stopper and inverted a couple of times to mix the remaining plasma with the mononuclear cells. This cell suspension is then decanted into a 15 ml conical tube and PBS is added to bring the volume to 15 ml and mixed. This tube is then centrifuged

16

Mondala, Salomon, and Head

(swing-bucket rotor) at room temperature, 15 min at 300 × g. The supernatant is then removed and the cell pellet is resuspended with a few microliters of PBS and the volume is brought up to 15 ml with more PBS (see Note 8).This cell washing step is repeated a second time after which the cell pellet is resuspended in 1 ml of RNA later, transferred to a 2 ml cryovial, and then frozen at –20°C. 3.3.4. Extraction of RNA, DNA, and Protein from Mononuclear Cells

The manufacturer (Qiagen) recommended protocol was followed. The AllPrep DNA/RNA/Protein Mini Kit allows the simultaneous purification of genomic DNA, total RNA, and total protein from a single sample. In this protocol the RNA is first extracted to completion then the protein is processed to the point that it is pelleted and finally the DNA is purified to completion (see Note 9). 1. Prior to starting the procedure prepare the following reagents. Add 10 ml b-mercaptoethanol per 1 ml Buffer RLT. This is stable at room temperature for 1 month. Buffer RLT may form a precipitate during storage, redissolve by warming, and then return to room temperature. 2. Buffer RPE, Buffer AW, and Buffer AW2 are supplied as a concentrate, add the appropriate volume of ethanol (100%) as indicated on the bottle to obtain a working solution. 3. Thaw frozen cells resuspended in RNA later on ice. Pellet cells by centrifuging at maximum speed for 2 min. Carefully remove all the supernatant by aspiration. 4. Flick the tube to loosen the pellet and then disrupt the cells by adding 600 ml Buffer RLT, vortex lightly or pipet to mix till no cell clumps remain. 5. Homogenize the lysate by passing it through an 18-gauge needle attached to a 5 ml syringe at least 5 times. 6. Transfer the lysate to an AllPrep DNA spin column placed in a 2 ml collection tube, centrifuge for 30 s at 8,000 × g. 7. Place the DNA spin column in a new 2 ml collection tube and store at 4°C for later DNA purification. Use the flow-through for RNA and protein purification. 8. To the flow-through add 400 ml ethanol (100%), mix well by pipetting. Do not centrifuge. Proceed immediately to next step. 9. Transfer 700 ml of the sample, including any precipitate that may have formed to an RNeasy spin column placed in a 2 ml collection tube. Centrifuge for 15 s at 8,000 × g. Transfer the flow-through to a 2 ml microcentrifuge tube for protein purification. Add the remaining sample to the same RNeasy spin column, centrifuge for 15 s at 8,000 × g. Combine the flowthrough in the 2 ml microcentrifuge tube. 10. Add 700 ml Buffer RW1 to the RNeasy spin column, centrifuge for 15 s at 8,000 × g, discard the flow-through.

Multicenter Clinical Sample Collection for Microarray Analysis

17

11. Add 500 ml Buffer RPE to the RNeasy spin column, centrifuge for 15 s at 8,000 × g, and discard the flow-through. 12. Add 500 ml Buffer RPE to the RNeasy spin column, centrifuge for 2 min at 8,000 × g, and discard the flow-through. 13. Carefully remove the RNeasy spin column and transfer to a new 2 ml collection tube, centrifuge at maximum speed for 1 min. 14. Place the RNeasy spin column in a new 1.5 ml microcentrifuge tube, carefully add 50 ml RNase-free water directly on the spin column membrane, centrifuge for 1 min at 8,000 × g to elute the RNA. 15. Quantitate RNA using the NanoDrop or any other spectrophotometer, blank the instrument with water. A 260/280 absorbance ratio between 1.8 and 2.0 is typical of pure RNA. Assess quality of the RNA using the Agilent Bioanalyzer. If the RNA will not be used immediately, store at –80°C. Typical RNA yield from a single 8 ml draw CPT tube is 3–7 mg. 16. To start the total protein precipitation, add 1 volume (~1,000 ml) of Buffer APP to the flow-through from step 7. Mix vigorously and incubate at room temperature for 10 min to precipitate protein. 17. Centrifuge at full speed for 10 min and carefully decant the supernatant. 18. Add 500 ml 70% ethanol to the pellet, centrifuge at full speed for 1 min, and then remove as much of the supernatant by carefully decanting followed by a pipet. 19. Dry the protein pellet for about 20 min at room temperature. Protein pellet can now be stored at –80°C. This protein fraction can now be quantified and is suitable for tandem mass spectrometry analysis. Typical protein yield from a single 8 ml draw CPT tube is 50–90 mg. 20. To complete the purification of genomic DNA, add 500 ml Buffer AW1 to the DNA spin column from step 5. Centrifuge for 15 s at 8,000 × g to wash the membrane. Discard the flowthrough. 21. Add 500 ml Buffer AW2 to the DNA spin column. Centrifuge for 2 min at full speed to wash the membrane. 22. Carefully remove the spin column avoiding contact with the flow-through. Place the spin column in a new 1.5 ml microcentrifuge tube. Add 100 ml Buffer EB directly to the spin column membrane, incubate at room temperature for 1 min, and then centrifuge for 1 min at 8,000 × g to elute the DNA. 23. Quantitate DNA using the NanoDrop or any other spectrophotometer, blank the instrument with Buffer EB. A 260/280 absorbance ratio between 1.7 and 1.9 is typical of pure DNA.

18

Mondala, Salomon, and Head

Check size distribution (15–30 kb) by running an aliquot on an agarose gel. If the DNA will not be used immediately store at –80°C. Typical DNA yield from a single 8 ml draw CPT tube is 4–12 mg. 3.3.5. Extraction of RNA, DNA, and Protein from Kidney Biopsies 3.3.5.1. RNA Extraction: Trizol

1. Bring Trizol reagent to room temperature. 2. Turn on constant temperature incubator to 60°C. 3. If precipitate has formed in Buffer RLT, redissolve by warming and place at room temperature. 4. Buffer RPE is supplied as a concentrate. Add 4 volumes of ethanol (100%) before using for the first time. 5. In a chemical fume hood/biosafety cabinet, thaw biopsy core submerged in RNAlater on ice. Add 1 ml of Trizol reagent into a properly decontaminated, RNase-free 2 ml Wheaton Plastic Coated Tissue Grinder. 6. Carefully and quickly transfer core tissue and all smaller pieces into the grinder using forceps. Manually homogenize tissue using the Teflon coated pestle until completely homogenized as determined by visual inspection. Wear eye protection during this step. Incubate at room temperature for 5 min. 7. Transfer sample to a 1.5 ml microcentrifuge tube and add 200 ml chloroform, cap securely and vortex lightly for 20 s. Incubate at room temperature for 3 min. Centrifuge at 12,000 × g, 4°C for 15 min. 8. Carefully remove upper aqueous layer down to the interphase using a P200 pipet tip and transfer into a new 1.5 ml microcentrifuge tube (~500 ml volume). Save the tube containing the Trizol/Chloroform mixture for subsequent DNA and protein extraction (freeze the sample if the isolation of DNA and protein will be done another day). 9. Add 500 ml of room temperature isopropanol (100%) and mix by inversion. 10. Incubate at room temperature for 10 min. 11. Spin at 12,000 × g, 4°C for 10 min. 12. Carefully decant the supernatant. 13. Add 500 ml ETOH (70%), do not resuspend pellet. 14. Spin at 7,000 × g, 4°C for 5 min. 15. Carefully decant the supernatant and turn tubes upside down on Kim wipes. Using a P10 pipet tip, aspirate any remaining ETOH and then immediately add 100 ml DEPC water. Do not allow the RNA pellet to dry completely, do not speedvac, do not resuspend the pellet. 16. Incubate the tube at 55°C in a constant temperature incubator for 10 min.

Multicenter Clinical Sample Collection for Microarray Analysis

19

17. Pellet usually dissolves by itself, tap the tube a couple times, and then quick spin. 18. In a fume hood, make up enough Buffer RLT with b-mercaptoethanol (10 ml bME: 1 ml Buffer RLT). 19. Add 350 ml Buffer RLT containing bME to each tube containing sample (total RNA in 100 ml DEPC-treated water). 20. Add 250 ml cold ethanol (100%), mix. 21. Add sample to an RNeasy spin column. 22. Spin at 10,000 × g, room temperature for 1 min. 23. Reapply flow-through to the column. 24. Spin at 10,000 × g, room temperature for 1 min. 25. Place column in a new 2 ml collection tube. 26. Add 500 ml Buffer RPE buffer. 27. Spin at 10,000 × g, room temperature for 1 min. 28. Discard flow-through. 29. Add 500 ml Buffer RPE buffer. 30. Spin at 10,000 × g, room temperature for 1.5 min. 31. Discard flow-through. 32. Spin at 10,000 × g, room temperature for 2 min. 33. Heat an aliquot of water to 70°C. 34. Place column in a new 1.5 ml tube. 35. Add 50 ml of 70°C water to the membrane, incubate at room temperature for 1 min. 36. Spin at 14,000 × g, room temperature for 2 min. 37. Quantitate RNA using the NanoDrop or any other spectrophotometer, blank the instrument with water. A 260/280 absorbance ratio between 1.8 and 2.0 is typical of pure RNA. Assess quality of the RNA using the Agilent Bioanalyzer. If the RNA will not be used immediately, store at –80°C. Typical RNA yield from a single 18G kidney biopsy core is 3–10 mg. Yield varies greatly depending on the cellular make up and overall size of the needle core. 3.3.5.2. DNA Extraction: Trizol

1. Total RNA Isolation must be performed prior to DNA Isolation. 2. Set constant temperature incubator to 37°C. 3. Prepare wash buffer that is, 0.1 M NaCitrate in 10% ethanol, 2 ml is used per sample. 4. Spin the tube containing Trizol reagent and sample from step 8 above, at 12,000 × g, room temperature for 2 min to separate phases.

20

Mondala, Salomon, and Head

5. Carefully remove any remaining upper aqueous phase using a pipet and discard. 6. Add 400 ml of ethanol (100%), mix by inversion. 7. Spin at 2,000 × g, 4°C for 5 min. 8. Aliquot supernatant equally into two separate tubes (~500 ml each) and set aside for the protein extraction later. 9. Add 1 ml 0.1 M NaCitrate in 10% ethanol (wash buffer) to pellet, vortex lightly. 10. Incubate at room temperature for 30 min, mix by inversion periodically. 11. Spin at 2,000 × g, 4°C for 5 min. 12. Remove supernatant and add 1 ml of wash buffer. 13. Incubate at room temperature for 30 min, mix by inversion periodically. 14. Spin at 2,000 × g, 4°C for 5 min. 15. Remove supernatant and add 1 ml ethanol (75%). 16. Incubate at room temperature for 20 min, mix by inversion periodically. 17. Spin at 2,000 × g, 4°C for 5 min. 18. Remove supernatant, dry down in speed-vac on medium heat for ~30 s, do not over dry. 19. Add 300 ml 8 mM NaOH, pass the pellet through a pipet tip a few times, and incubate overnight at 37°C. 20. Pre-spin Phase Lock Gel (PLG) tube at 14,000 rpm for 20 s. Add equal volume of phenol/chloroform as NaOH (~300 ml) to sample, vortex, transfer phenol/sample mix to PLG tube and spin at 14,000 rpm for 2 min. Transfer top, clear aqueous phase to new 1.5 ml microcentrifuge tube. 21. Add 0.1 volumes 3 M Sodium Acetate (~30 ml). Add 1 ml 5 mg/ml Glycogen. Add 2.5 volumes ice-cold 100% ethanol (~830 ml). Incubate 1 h at –80°C. 22. Spin at 14,000 rpm, 4°C for 20 min. 23. Remove and discard supernatant and add 1 ml ice-cold ethanol (80%). 24. Spin at 14,000 rpm, room temperature for 2 min. 25. Remove and discard supernatant and add 1 ml ice-cold ethanol (80%). 26. Spin at 14,000 rpm, room temperature for 2 min. 27. Remove and discard supernatant, let pellet air dry, and resuspend in 22 ml of water. 28. Quantitate on NanoDrop or other spectrophotometer, blank instrument with water. A 260/280 absorbance ratio between

Multicenter Clinical Sample Collection for Microarray Analysis

21

1.7 and 1.9 is typical of pure DNA. Check size distribution by running an aliquot on an agarose gel. Typical DNA yield from a single 18G kidney biopsy core is 1–15 mg. This can vary greatly depending on the cellular make up and overall size of the needle core. 3.3.5.3. Protein Extraction: Trizol

1. Total RNA Isolation and DNA Isolation must be performed prior to Protein Isolation. 2. Add 750 ml isopropanol (100%) into each of the 2 tubes from step 8 above, invert 20 times. 3. Incubate at room temperature for 10 min. 4. Spin at 12,000 × g, 4°C for 7 min. 5. Remove supernatant, wash with 750 ml 0.3 M guanidine HCl in 95% ethanol, vortex lightly. 6. Incubate at room temperature for 20 min. 7. Spin at 7,500 × g, 4°C for 5 min. 8. Wash a total of 3 times. 9. After the last wash, add 1 ml ethanol (100%) to the pellet, vortex lightly. 10. Incubate at room temperature for 20 min. 11. Spin at 7,500 × g, 4°C for 5 min. 12. Remove supernatant. 13. Store protein pellet at –80°C. This protein fraction can now be quantified and is suitable for tandem mass spectrometry analysis. Typical protein yield from a single 18G kidney biopsy core is 50–100 mg. This can vary greatly depending on the overall size of the needle core.

3.3.6. DNA Extraction from Whole Blood

The QIAamp DNA Blood Midi Kit (100 reaction kit) provides a simple, fast method for purifying DNA from blood. The separation of leukocytes is not necessary and no phenol/chloroform or alcohol precipitation is required. DNA purified using this method ranges in size up to 50 kb (see Note 10). 1. Equilibrate samples to room temperature, thoroughly mix by inversion. 2. Set constant temperature incubator to 70°C. 3. Before starting the DNA purification prepare the Protease Working Solution by pipetting 5.5 ml water into the vial of lyophilized Qiagen Protease. Freeze aliquots of the unused protease at –20°C for later use. 4. Add 125 ml of ethanol (100%) to Buffer AW1. 5. Add 150 ml of ethanol (100%) to Buffer AW2.

22

Mondala, Salomon, and Head

6. If a precipitate has formed in Buffer AL, dissolve by incubating at 56°C. 7. Pipet 400 ml of Protease into the bottom of 15 ml centrifuge tube. 8. Add 4 ml of blood and mix briefly (add appropriate amount of PBS if sample is less than 4 ml). 9. Add 4.8 ml Buffer AL to the samples, mix thoroughly by vortexing at least 3 times for 5 s each time. Do not add Protease directly to Buffer AL. 10. Incubate at 70°C for 10 min in a constant temperature incubator. 11. Add 4 ml of ethanol (100%) to the sample and mix again by vortexing. 12. Carefully transfer 3.3 ml of sample onto a QIAamp Midi column placed in a 15 ml centrifugation tube. Close the cap and centrifuge for 3 min at 1,900 × g, room temperature, brake on using a swing-out rotor with adapter for round-bottom tubes (do not over tighten caps, if the caps are tightened until they snap they may loosen during centrifugation and damage the centrifuge). 13. Remove the Midi column, discard the filtrate, and wipe off any spillage from the thread of the 15 ml centrifugation tube before re-inserting the column in to the 15 ml centrifugation tube. Load another 3.3 ml of sample onto the column, close the cap, and centrifuge for 3 min at 1,900 × g, room temperature, brake on. 14. Remove the Midi column, discard the filtrate and wipe off any spillage from the thread of the 15 ml centrifugation tube before re-inserting the column in to the 15 ml centrifugation tube. Load remaining sample onto the column, close the cap, and centrifuge for 3 min at 1,900 × g, room temperature, brake on. 15. Remove the Midi column, discard the filtrate, and wipe off any spillage from the thread of the 15 ml centrifugation tube before re-inserting the column in to the 15 ml centrifugation tube. Carefully without moistening the rim add 4 ml Buffer AW1 to the column, close the cap and centrifuge for 5 min at 3,500 × g, room temperature, brake on. 16. Remove the Midi column, discard the filtrate, and wipe off any spillage from the thread of the 15 ml centrifugation tube before re-inserting the column in to the 15 ml centrifugation tube. Carefully without moistening the rim add 4 ml Buffer AW2 to the column, close the cap, and centrifuge for 5 min at 3,500 × g, room temperature, brake on.

Multicenter Clinical Sample Collection for Microarray Analysis

23

17. Wipe off any spillage off the Midi column and place the column in a clean 15 ml centrifugation tube and discard the tube containing the filtrate. Add 600 ml of water, close the cap, and incubate at room temperature for 5 min. Centrifuge at 3,500 × g for 10 min, room temperature, brake on. 18. Reload the 600 ml eluate containing the DNA onto the membrane of the Midi column. Close the cap and incubate at room temperature for 5 min. Centrifuge at 3,500 × g for 10 min, room temperature, brake on. 19. Quantitate on NanoDrop, blank instrument with water. A 260/280 absorbance ratio between 1.7 and 1.9 is typical of pure DNA. Check size distribution (15–30 kb) by running an aliquot on an agarose gel. If the DNA will not be used immediately store at –80°C. Typical DNA yield from 4 ml blood is 40–60 mg. 3.3.7. DNA Extraction from Mononuclear Cells

The QIAamp DNA Mini Kit (50 reaction kit) provides a simple, fast method for purifying DNA from cells. DNA purified using this method ranges in size up to 50 kb. The manufacturer (Qiagen) recommended protocol was followed. In our Transplant Genomics Project we typically would process 2–5 million cells suspended media. 1. Equilibrate samples to room temperature. 2. Set constant temperature incubator to 56°C. 3. Before starting the DNA purification prepare the Protease Working Solution by pipetting 1.2 ml protease solvent in to the vial of lyophilized Qiagen Protease. Freeze aliquots of unused protease at –20°C for later use. 4. Add 25 ml of ethanol (100%) to Buffer AW1. 5. Add 30 ml of ethanol (100%) to Buffer AW2. 6. If a precipitate has formed in Buffer AL, dissolve by incubating at 56°C. 7. Centrifuge sample containing cells at maximum speed (20,000 × g) to pellet cells, remove supernatant, and resuspend pellet with 200 ml PBS. 8. Pipet 20 ml Qiagen Protease into a 1.5 ml microcentrifuge tube. 9. Add 200 ml sample to the microcentrifuge tube. 10. Add 200 ml Buffer AL to the samples, mix by pulse-vortexing for 15 s. Note: Do not add Protease directly to Buffer AL. 11. Incubate at –56°C for 10 min. 12. Briefly centrifuge the sample to collect all drops from the inside of the lid.

24

Mondala, Salomon, and Head

13. Add 200 ml of ethanol (100%) to the sample and mix by pulse-vortexing for 15 s. After mixing briefly centrifuge the tube to collect all drops from the inside of the lid. 14. Carefully apply sample mixture directly on to a Mini spin column placed in a 2 ml collection tube without wetting the rim. Close cap and centrifuge at maximum speed for 1 min. Place the spin column in a clean 2 ml collection tube and discard the tube containing the filtrate. 15. Carefully open the spin columns and add 500 ml Buffer AW1 without wetting the rim. Close the cap and centrifuge at 6,000 × g for 1 min. Place the spin column in a clean 2 ml collection tube and discard the tube containing the filtrate. 16. Carefully open the spin column and add 500 ml Buffer AW2 without wetting the rim. Close the cap and centrifuge at maximum speed for 3 min. 17. Place the spin column in a new 2 ml collection tube and discard the old tube with the filtrate. Centrifuge at full speed for 1 min. 18. Place the spin column in a clean 1.5 ml microcentrifuge tube and discard the tube containing the filtrate. Carefully open the spin column and add 200 ml water. Incubate at room temperature for 5 min and then centrifuge at 6,000 × g for 5 min. 19. Reload the eluate containing the DNA onto the membrane of the spin column and close the cap and incubate at room temperature for 5 min and then centrifuge at 6,000 × g for 5 min. 20. Quantitate on NanoDrop, blank instrument with water. A 260/280 absorbance ratio between 1.7 and 1.9 is typical of pure DNA. Check size distribution (15–30 kb) by running an aliquot on an agarose gel. If the DNA will not be used immediately store at –80°C. Typical DNA yield from 2 to 5 million mononuclear cells is 4–10 mg.

4. Notes 1. Various labels were tested, these labels passed requirement that they adhere, stay intact and that the printed barcode not smear (remain scanable using a handheld barcode reader) under extreme cold temperature such as dry ice and after numerous freeze-thaw cycles. 2. We instruct all clinical centers to collect blood tubes in a specific order to ensure all collection procedures are done as

Multicenter Clinical Sample Collection for Microarray Analysis

25

uniformly as possible. PAXgene tubes are collected first followed by the CPT tube and then the PPT tubes. 3. It is important that detailed instruction must be given to the physicians or radiologists performing the biopsies to immediately and completely submerge the kidney core into RNAlater to eliminate the impact of RNA degradation. 4. Before starting the PAXgene RNA purification procedure, incubate the tube at room temperature for at least 2 h in order to ensure complete blood cell lysis. If the tube was immediately frozen or stored at 2–8°C after blood collection, then after removal from storage, first thaw to room temperature for at least 2 h, invert 10 times and then incubate at room temperature for an additional 2 h. After incubating, invert the tube another 10 times. 5. Divide DNase I into single-use aliquots of 10.5 ml and store at 2–8°C for up to 6 weeks or at –20°C for up to 6 months. Thaw appropriate number of DNase I stock solution aliquots for on-column DNase digestion. Do not refreeze the aliquots after thawing. 6. Do not air-dry the RNA Binding Beads for more than 5 min, in our experience over-drying at this step resulted in lower yield. 7. Initially we had observed as much as a 30% reduction in total RNA after globin reduction. Subsequently, yield was reduced on average 15% only. This was attributed to an increase in familiarity with the magnetic bead technique decreasing the unintentional loss of binding beads when removing supernatant from the captured beads. 8. The cell pellet can be easily resuspended by flicking the tube a couple of times. 9. Qiagen now has a supplemental protocol for the purification of miRNAusingtheAllPrepDNA/RNA/ProteinMiniKitandRNeasy MinElute Cleanup Kit (http://www1.qiagen.com/products/ RnaStabilizationPurification/AllPrepDNARNAProteinMiniKit. aspx#Tabs=t2). 10. A modified version of the manufacturer (Qiagen) recommended protocol was followed. We increased the starting blood volume from 2 to 4 ml.

Acknowledgments This work is supported by the National Institute of Allergy and Infectious Diseases (NIAID) Program Project Grant U19 AI63603 Genomics for Kidney Transplantation.

26

Mondala, Salomon, and Head

References 1. Copland JA, Davies PJ, Shipley GL, Wood CG, Luxon BA, Urban RJ (2003) The use of DNA microarrays to assess clinical samples: the transition from bedside to bench to bedside. Recent Prog Horm Res 58:25–53 2. Al-Mulla F (2007) Utilization of microarray platforms in clinical practice. Methods Mol Biol 382:115–136 3. Flechner SM, Kurian SM, Head SR, Sharp SM, Whisenant TC, Zhang J, Horvath S, Mondala T, Gilmartin T, Cook DJ, Kay SA, Walker, JR, Salomon DR (2004). Characterizing acute kidney transplant rejection by gene profiling of biopsies and peripheral blood lymphocytes. Am J Transplant 4(9): 1475–1489 4. Flechner SM, Kurian SM, Solez K, Cook DJ, Burke JT, Rollin H, Hammond JA, Whisenant T, Lanigan CM, Head SR, Salomon DR (2004) De novo kidney transplantation without use of calcineurin inhibitors preserves renal structure and function at two years. Am J Transplant 4(11):1776–1785 5. Kurian SM, Flechner SM, Kaouk J, Modlin C, Goldfarb D, Cook DJ, Head S, Salomon DR (2005) Laparoscopic donor nephrectomy gene expression profiling reveals upregulation of

6.

7.

8.

9.

10.

stress and ischemia associated genes compared to control kidneys. Transplantation 80(8): 1067–1071 Kurian S, Grigoryev Y, Head S, Campbell D, Mondala T, Salomon DR (2007) Applying genomics to organ transplantation medicine in both discovery and validation of biomarkers. Int Immunopharmacol 7(14):1948–1960 Burczynski ME et al (2005) Transcriptional profiles in peripheral blood mononuclear cells prognostic of clinical outcomes in patients with advanced renal cell carcinoma. Clin Cancer Res 11(3):1181–1189 Cobb JP et al (2005) Application of genomewide expression analysis to human health and disease. Proc Natl Acad Sci USA 102(13): 4801–4806 Tsuang MT, Nossova N, Yager T, Tsuang MM, Guo SC, Shyu KG, Glatt SJ, Liew CC (2005) Assessing the validity of blood-based gene expression profiles for the classification of schizophrenia and bipolar disorder: a preliminary report. Am J Med Genet B Neuropsychiatr Genet 133(1):1–5 Feezor RJ et al (2004) Whole blood and leukocyte RNA isolation for gene expression analyses. Physiol Genomics 19:247–254

Chapter 2 Isolation of Total RNA from Transgenic Mouse Melanoma Subsets Using Fluorescence-Activated Cell Sorting Scott Tighe and Matthew A. Held Abstract The majority of tumors, including melanoma, are phenotypically heterogeneous in that they contain various cell populations with differential expression of cell surface antigens such as CD133/Prominin-1. We have used fluorescence-activated cell sorting (FACS) technology to purify CD133+ and CD133− cellular subsets from mouse melanoma models for high-quality total RNA practical for downstream applications such as expression profiling. Implementation of this strategy can lead to higher resolution of transcripts that are potentially important for the survival and functionality of one cancer cell population relative to another. Suboptimal extraction of RNA after FACS is common and can ultimately result in misinterpretations that impede the effective design of novel therapies. Here, we describe a number of methods that have been amenable to the successful isolation of high-quality total RNA after FACS of CD133+ and CD133− mouse melanoma cell fractions. Key words: Melanoma, Mouse models of cancer, FACS, Cell surface markers, Cell subsets, RNA isolation, RNA FACS sorting

1. Introduction Methods for genome-wide expression analyses, such as DNA microarrays (1), can be used to delineate global RNA expression differences between cancer cell subsets that show variations in function such as their abilities to resist chemotherapy or propagate tumors. The cell surface antigen CD133 has been demonstrated to identify cancer cells from a variety of solid-tissue cancers such as melanoma that display higher tumorigenicity or treatment resistance (2–5) and can be characterized through, for example, gene expression profiling of the CD133+ and CD133− subset phenotypes. To accomplish such a task requires well-established flow cytometric sorting methods and RNA extraction protocols (6, 7). Sridar V. Chittur (ed.), Microarray Methods for Drug Discovery, Methods in Molecular Biology, vol. 632, DOI 10.1007/978-1-60761-663-4_2, © Humana Press, a part of Springer Science+Business Media, LLC 2010

27

28

Tighe and Held

It is well known that RNA is a sensitive nucleic acid that can easily degrade as a result of erroneous introduction of ribonucleases (RNases) either from instrumentation, immunostaining procedures, end user, or endogenously from the sample itself (8, 9). In addition, improper extraction and storage of RNA can decrease its overall half-life, compromising future utility. Here we establish several workflows for fluorescence-activated cell sorting (FACS) of CD133+ and CD133− melanoma cell subsets for highquality total RNA purification including instrument decontamination, cell surface marker labeling, cell sorting procedures, and RNA handling and extraction methods. In addition, we discuss quantification techniques and integrity analyses used for validating the RNA quality of these cellular subsets after FACS.

2. Materials 2.1. Cell Culture and Antibody Staining

1. Dulbecco’s Modified Eagle’s Medium-F12, 1:1 (Gibco). 2. Fetal-bovine serum (FBS), US-origin, irradiated, heatinactivated (Hyclone). 3. Modified Eagle’s (Cellgro).

Medium

Nonessential

amino

acids

4. Trypsin 0.25%/2.2 mM EDTA (Cellgro). 5. Penicillin streptomycin (Pen/Strep), 1 × 104 U/ml each (Cellgro). 6. Phosphate-buffered saline (PBS), RNase-free (Ambion). 7. Bovine serum albumin (BSA), RNase-free (Equitech-Bio). 8. RNase inhibitor, e.g., RiboLock (Fermentas Corp). 9. Dulbecco’s PBS containing 100 U/ml RiboLock (PBSRIBO). 10. RNase-free 1.5 ml microcentrifuge tubes (Axygen, #MCT175c). 11. RNaseZap (Ambion). 12. Rat anti-CD133 mouse monoclonal antibody (eBioscience). 13. AlexaFluor488 chicken antirat IgG secondary antibody (Invitrogen). 2.2. Flow Cytometry and Sorting

1. BD FACSAria flow cytometer or equivalent. 2. RNase-free water (VWR Scientific). 3. Sterile polystyrene round-bottom tubes for flow 5 ml (BD Falcon). 4. Bleach 10% (0.525% sodium hypochlorite).

Isolation of Total RNA from Transgenic Mouse Melanoma Subsets Using Fluorescence

29

5. Sterile sheath fluid (saline) RNase-free. 6. Propidium iodide (1 mg/ml) solution-ultra-high purity (Enzo, #enz-52403R). 7. Bovine serum albumin, RNase-free (Equitech-Bio). 8. RNaseAlert RNase detection system (Ambion). 2.3. RNA Isolation

1. Trizol or Trizol LS or equivalent. 2. RNeasy Micro Kit (Qiagen). 3. Beta-mercaptoethanol. 4. Chloroform (100% ACS Grade). 5. 100% Ethanol (Electron Microscopy Sciences). 6. MaxyClear RNase-free tubes 1.5, 15, and 50 ml (Axygen). 7. QIAvac-24 Plus Vacuum manifold (Qiagen). 8. Nanodrop ND1000 spectrophotometer. 9. Qubit Spectrofluorometer (Invitrogen). 10. Quant-IT RNA reagents (Invitrogen). 11. Agilent 2100 Bioanalyzer or equivalent.

3. Methods 3.1. Quality Control of the FluorescenceActivated Cell Sorter

Before proceeding with FACS of cell subsets, stringent quality control of the instrumentation is mandatory to ensure the success of good quality RNA isolation from sorted cell populations. This involves thorough decontamination followed by empirical validation of FACS machine sanitation. Decontamination time will depend on the instrument type, age, and degree of contamination. However, procedures for sanitizing any FACS instrument are similar, and so a review of the following steps is warranted. It is urged to perform all steps with RNaseZap-treated gloves in a low contamination environment (see Note 1). Once decontamination is complete, a test sort using noncritical cells with a known viability >80% should be performed to test the instrument (see Note 2). 1. Ensure the dip tube, septa, flow cell, tubing lines, and nozzles have been decontaminated with 10% bleach, 100% ethanol, RNaseZap, autoclaving, or other suitable qualifying technique prior to the sort. 2. Ensure sheath tank and fluid are RNase-free. Quality control sampling of each may be tested with RNase-detecting reagents such as RNaseAlert.

30

Tighe and Held

3. Replace all contaminated fluid lines and filters as outlined by the manufacturer. 4. Prior to sort, run several tubes of 10% bleach through the flow cytometer including the sorting components followed by flushing with sterile RNase-free sheath fluid or PBSRIBO. 5. Perform a sort using a bead solution containing 400 U/ml of an RNase inhibitor just prior to sorting critical samples. 3.2. Preparation of Melanoma Cells and Antibody Staining for FACS

Melanoma cell lines were derived from transgenic conditional mouse melanoma tumors as previously described (10). Tumors were finely minced using aseptic technique and enzymatically dissociated with 0.05% trypsin/0.55mM EDTA for 30 min at room temperature, with thorough mincing every 10 min. Dissociated tumors were then lightly triturated 15–20 times, and the resulting suspensions were transferred to tissue culture treated 10 cm adherent dishes. Melanoma cultures were grown in 1:1 DMEM:F12 media with 5% FCS and 1% Pen/Strep (media complete) in a cell culture incubator at 37°C with 5% CO2 and allowed to grow until approximately 75% confluent. The following protocol was then followed with proper RNA handling in a biosafety cabinet or PCR hood for indirect antibody labeling of cells for the surface marker CD133 followed by FACS of CD133+ and CD133− cellular subsets using a BD FACSAria flow cytometer (see Note 3). 1. Aspirate media from 10 cm adherent melanoma culture dishes and detach cells by briefly incubating (2–3 min) with 1 ml of 0.25% trypsin/2.2 mM EDTA, followed by neutralization of trypsin with 10 ml of media complete. 2. Centrifuge cell suspensions at 800 × g for 5 min, aspirate supernatant, and resuspend cell pellet in 1 ml of 1× PBSRIBO with 2% BSA (PBS-RIBO-BSA) (see Note 4). 3. Perform a viability count using a hemocytometer and Trypan Blue dead-cell discrimination dye (see Note 5). 4. For each sample and control, transfer 5 × 105 cells to a new tube. Controls should include samples with primary antibody only, secondary antibody only (or isotype-control only), unstained cells, and propidium iodide only. These are required for fluorescent compensation and proper gate positioning. 5. All samples and controls are centrifuged at 800 × g, aspirated, and resuspended in 100 ml of PBS-RIBO-BSA followed by staining with 1 mg/ml final concentration anti-CD133 primary antibody for 30 min at 4°C (in fridge, not on ice). 6. Samples are quenched with 900 ml PBS-RIBO-BSA, centrifuged, aspirated, and resuspended in 100 ml PBS-RIBO-BSA.

Isolation of Total RNA from Transgenic Mouse Melanoma Subsets Using Fluorescence

31

7. Each sample is stained with a species-matched, AlexaFluor-488 chicken antirat IgG secondary antibody at 1:1,000 for 20 min in a dark fridge at 4°C. 8. Centrifuge cells, aspirate supernatant, and resuspend cells in 500 ml PBS-RIBO-BSA with a final concentration of 1 µg/ml RNase-free propidium iodide for dead-cell discrimination. Transfer samples to FACS machine-compatible, sterile 5 ml round-bottom tubes and cap. 9. During the flow cytometric procedure, exclude all propidium iodide-positive signals (i.e., dead cells). Whenever possible, use forward scatter (FSC), side scatter (SSC) height, width, and area measurements to exclude any potential doublets or putative apoptotic/dead cells. Live, single cells are then analyzed and sorted by FACS on CD133 signal into precooled, RNase-free 1.5 ml microcentrifuge tubes for subsequent total RNA extraction of purified cell subsets (see Note 6). 3.3. Methods for Sorting Cell Subsets for Total RNA Extraction

There are a variety of procedures for recovering total RNA from sorted cells. The choice of any one protocol depends on two factors: (1) whether the type of FACS machine used for cell purification is mechanical or electrostatic and (2) whether high or low dispensed sort volumes are expected. Mechanical sorters, such as the BD FACSCalibur, use a mechanical sorting device called the “catcher tube” positioned near the flow cell and sort relatively slowly (e.g., 300 events/s) with a relatively high sort volume (e.g., 100 nl–10 µl per event). Therefore, direct sorting of cells using mechanical sorters is not ideal for sorting large numbers of cells directly into RNA extraction buffer as the high dispensed volumes will dilute the buffer substantially and impede RNA recovery. When using mechanical sorters, it is recommended to first centrifuge the sorted cells to form a cell pellet followed by addition of the chosen RNA extraction buffer as outlined below in Subheading 3.3.1. It is important to consider that any additional handling before adding the RNA extraction buffer, such as centrifuging, may lead to consequential gene expression changes (see Note 7). Electrostatic sorters or “stream-in-air” FACS machines can operate at much higher speeds (e.g., 25,000 events/s or more) and involve a vibrating nozzle by which cells exit within single droplets resulting in much smaller dispensed sort volumes (11). Electrostatic sorters are also capable of fitting various sized nozzles in order to accommodate for cell size and maximize cell viability during the procedure. For example, a 70 mm nozzle decreases flow stream width, thereby resulting in droplet volumes of 1 nl drops per event – an approach applicable to sorting small cell types (e.g., T lymphocytes). In contrast, a 100 mm nozzle will relax flow stream width slightly to accommodate larger sized cells

32

Tighe and Held

(e.g., tumor cells) which results in volumes of 5–10 nl drops per event. Although the exact droplet size may vary slightly based on the system settings of each individual flow cytometer, smaller dispensed volumes allow for sorting directly into RNA extraction buffer such as Trizol LS or RLT buffer from the RNeasy system (see Note 8). Although other alternative methods for RNA isolation from sorted cells are available, they will not be described here (see Note 9). 3.3.1. RNeasy System for RNA Isolation After Centrifugation

Isolation of cells from high sorted volumes, such as those from a mechanical sorter, will require a centrifugation step to collect the cell pellet followed by RNA isolation using a silica column approach, such as the RNeasy microcolumn, or a standard Trizol precipitation method as described by the manufacturer (see Note 10) (12). When sorting cells for RNA, it is important to consider adding an RNase inhibitor to the sort recovery tube prior to the sort and adjust to 5–20 U/ml following the sort whenever possible (see Note 11). Sorting directly into a cell preservation reagent for future RNA isolation should be avoided (see Note 12). 1. Immediately following the sort, aseptically centrifuge cells to a pellet at 1,000 × g for 10 min using a refrigerated centrifuge. 2. Using a sterile aspirator, remove all supernatant from the cell pellet. 3. Add 100 ml of RNase-free water and 350 ml of RLT buffer and vortex for 30 s (see manufacturer’s protocol) (13). 4. Add 250 ml of 100% EMS grade ethanol and vortex. 5. Using a micropipettor with aerosol resistant tip, transfer sample to the RNeasy microcolumn and centrifuge at >10,000 × g for 15 s. Replace waste capture tube containing the passthrough liquid. 6. A DNase treatment (steps 8–10) may be required when downstream methods involving random hexamer priming such as in the case of exon microarrays, RT-qPCR, or equivalent are used. If no DNase treatment is required, proceed to step 10 (see Note 13). 7. Apply 200 ml of RW1 buffer to the column and centrifuge at >10,000 × g for 15 s. 8. For each sample, prepare the DNase solution from the Qiagen RNase-free DNase kit by combining 70 ml of RDD buffer with 10 ml of DNase I (27.3 Kunitz units total) and applying 80 µl to the column’s silica membrane. Incubate at room temperature for 20 min. 9. Add 200 ml of RW1 buffer to the column and centrifuge at >10,000 × g for 15 s. Replace the waste capture tube containing the pass-through liquid.

Isolation of Total RNA from Transgenic Mouse Melanoma Subsets Using Fluorescence

33

10. Apply 0.5 ml RPE buffer to the column and centrifuge at >10,000 × g for 15 s. Replace the waste capture tube containing the pass-through liquid. 11. Repeat step 10. 12. Using a 20 ml pipette, remove the remaining liquid that may be caught up on the edge of the column’s inner O-ring. 13. Perform an extended centrifugation for 3–5 min to remove as much liquid from the membrane as possible. Do not centrifuge with column open as described in the manufacturer’s protocol. 14. Replace waste tube with a new standard RNase-free 1.5 ml microcentrifuge tube. 15. Apply 15 ml of 60°C RNase-free water directly to the center of the RNeasy microcolumn membrane and incubate at room temperature for 30 s. 16. Centrifuge at >10,000 × g for 15 s. 17. Carefully remove the 15 ml of sample from the tube and reapply it to the same RNeasy membrane again. Close column and centrifuge at >10,000 × g for 15 s. This reelution is performed with the same 15 ml aliquot to assure complete recovery of RNA from the entire surface of the column’s silica membrane. 18. Remove the RNeasy microcolumn from the microcentrifuge tube containing the 15 ml of sample, and add the equivalent of 20 U of RNase inhibitor and vortex. Store sample at −20°C. 19. Quantify the RNA using a high resolution spectrometer such as the Nanodrop ND-1000 and Qubit fluorometer (see Subheading 3.4.1). 20. Analyze the RNA quality using an Agilent 2100 Bioanalyzer or equivalent (see Subheading 3.4.2). 3.3.2. Direct Sort Method

When low sort volumes are expected, it is advantageous to sort directly into extraction reagent such as Trizol LS or RLT buffer in order to minimize downstream handling and inadvertent gene expression changes. Regardless of the method selected, it is imperative to maintain the exact ratio of aqueous sorted volume to extraction reagent consistent with the manufacturer’s recommendations and to extract RNA promptly. If immediate extraction is not possible, then short-term storage in dilute extraction reagent may be considered (see Note 14). Although direct sorting into extraction buffer is optimal for RNA recovery, secondary analyses such as microscopy or postsort cell purity validation will require additional steps (see Note 15).

34

Tighe and Held

3.3.2.1. Direct RNA Extraction Using a Combined Trizol LS-RNeasy Method

1. Start with 500 ml of Trizol LS in a sterile RNase-free FACS tube if choosing to sort directly into the Trizol LS. Otherwise, sort into 1× PBS-RIBO and spin cells down at 800 × g for 10 min, aspirate supernatant, and then add 500 ml of Trizol LS. 2. Multiple sort tubes may be used to collect cells if dispensed sort volumes exceed the volume capacity of the sort collection tubes. If so, use 500 ml starting Trizol LS volume for the extra sort tubes as well. 3. After the sort, use a pipette equipped with an aerosol resistant tip to measure the final volume in the tube. Subtract the amount of Trizol LS to determine the amount of dispensed liquid. 4. Adjust the amount of Trizol LS required to maintain the sample at a Trizol:dispensed volume ratio of at least 3:1. This may require the solution to be transferred to a larger RNase-free tube (see Note 16 for a mathematical example). 5. Add 200 ml of chloroform for every 750 ml of Trizol LS to the tube and mix. Let the samples sit on bench top for 3 min. Alternative organic phases may be used in place of chloroform but are not preferred by the authors (see Note 17). 6. Centrifuge at >10,000 × g for 10 min at 4°C to separate the top aqueous layer from the bottom layer and interface. If the volume of solution is too large to fit into a microcentrifuge tube, it can be transferred to a 15 or 50 ml centrifuge tube and spun down with a larger centrifuge (see Note 18). 7. Carefully remove samples from the centrifuge and transfer the top aqueous layer to an RNase-free tube. Determine the exact volume of the aqueous layer and add 1.5 times the volume of 100% RNase-free ethanol and mix (see Note 19). 8. Filter the entire volume through an RNeasy microcolumn. For larger volumes (e.g., >5 ml), a vacuum manifold is suggested for faster sample processing (step 9a). Smaller volumes can be processed as individual 700 ml applications to the same RNeasy column and centrifuged (step 9b). 9a. Vacuum manifold technique: Using the QiaVac manifold (see Fig. 1) or equivalent (14), turn on the vacuum pump, and open the selected receiver ports to allow suction. Saturate receiver ports with RNaseZap for 30 s followed by rinsing with 100% ethanol. Turn off pump and aseptically install RNeasy microcolumn to selected receiver port(s). Turn vacuum pump on and repeatedly load 700 ml aliquots of the same sample into the RNeasy column until all of the sample volume has been filtered through. Remove column from vacuum manifold and place in a standard 2 ml capture tube and continue to step 10.

Isolation of Total RNA from Transgenic Mouse Melanoma Subsets Using Fluorescence

35

9b. Centrifugation technique: Apply no more than 700 ml to the RNeasy microcolumn and spin at >10,000 × g for 15 s. If the total volume is greater than 700 ml, multiple loadings to the same column will be required. 10. Perform DNase I treatment if required as per Subheading 3.3.1, step 7. If a DNase treatment is not needed, proceed to step 11 below. 11. Apply 700 ml RPE buffer to each column and centrifuge at >10,000 × g for 15 s. Discard and replace the waste capture tube containing the pass-through liquid. 12. Repeat step 11 a total of four times. This is required to remove any remaining Trizol that may otherwise be bound to the silica membrane when a Trizol-based lysis protocol is performed. Any residual Trizol contamination will lead to inaccurate UV-based RNA quantitation at 260 nm (see Note 19). 13. Using a 20 ml pipette, remove the remaining liquid that may be caught on the inner edge of the column’s O-ring. 14. Perform an extended “dry” centrifugation at >10,000 × g for 2 min to remove as much residual liquid from the RNeasy microcolumn as possible. Do not centrifuge with column cap open. 15. Replace waste tube with a new standard RNase-free 1.5 ml microcentrifuge tube. 16. Apply 15 ml of 60°C RNase-free water directly to the center of the RNeasy microcolumn membrane, and incubate at room temperature for 30 s. 17. Centrifuge at >10,000 × g for 15 s. 18. Carefully remove the 15 ml of sample from the tube and reapply it to the same RNeasy membrane again. Close column

Fig. 1. Standard configuration for a vacuum manifold system fitted with RNeasy microcolumns. This approach allows the processing of large volumes of RNA extraction buffer (e.g., >5 ml) through the silica membrane without the use of a centrifuge.

36

Tighe and Held

and centrifuge at >10,000 × g for 15 s. This reelution is performed with the same 15 ml aliquot to assure complete recovery of RNA from the entire surface of the column. 19. Remove the RNeasy microcolumn from the microcentrifuge tube, and add the equivalent of 20 U of RNase inhibitor and vortex. At this point, samples may be stored at −20°C for short-term use or at −80°C for long-term storage. 20. Using both a UV spectrophotometer and fluorometer, such as the NanoDrop ND1000 and Qubit, determine the concentration of each sample. Make note of possible Trizol contamination as noted by a 270 nm absorbance peak on the UV spectrometer (see Subheading 3.4.1). In most cases, quantitative results for the fluorometer are lower than that of the UV spectrophotometer, but are considered more accurate. 3.3.2.2. Direct RNA Extraction Using RNeasy Microcolumn Method

When sorting directly into RLT buffer (guanidium isothiocyanate), a ratio of 100 ml of sorted sample to 350 ml of RLT should be maintained. In general, the Trizol LS method has a greater RNA recovery on cells with more resistant cell membranes, aggregated cells, or organisms with a cell wall, but is more costly and involves more reagents. RNA recovered by directly sorting into RLT buffer is typically much cleaner than that recovered with Trizol and does not require additional quantitation with a Qubit spectrofluorometer because there is no interfering 270 nm absorbance from trace amounts of Trizol carryover. 1. Start with 500 ml of RLT buffer with 5 ml BME in a sterile RNase-free FACS tube. 2. While sorting, periodically mix to get liquid off sides of the tube. Keep sample cold whenever possible. 3. After the sort, using a pipette with sterile tip, measure the final volume and calculate the exact volume of sample sorted into the RLT. 4. Adjust the amount of RLT required, so that a ratio of 350 ml of RLT buffer to every 100 ml of sorted sample is maintained and then vortex. Samples may need to be transferred to larger RNase-free tubes if final volumes are high. 5. Add 250 ml of 100% ethanol for every 350 ml RLT buffer, and then mix samples. 6. If volumes from step 5 are high (e.g., >5 ml), then use of a vacuum manifold is suggested. Smaller volumes can be processed as individual 700 ml applications to the same RNeasy microcolumn and centrifuged at >10,000 × g for 15 s. 7. Complete protocol by referring to steps 9a–19 in Subheading “Direct RNA Extraction Using a Combined Trizol LS-RNeasy Method”.

Isolation of Total RNA from Transgenic Mouse Melanoma Subsets Using Fluorescence

3.4. Analyzing RNA from Sorted Melanoma Cell Fractions 3.4.1. Quantitation of RNA

3.4.2. Assessing RNA Quality

37

After extracting RNA from CD133+ and CD133– subsets using methods described above, the concentration of RNA was determined using a Nanodrop spectrophotometer and Qubit spectrofluorometer (15). Both methods are necessary because UV absorbance from the Nanodrop or other similar instruments alone cannot effectively discriminate some contaminants from true RNA; therefore, additional quantitation using a fluorescent RNA intercalation dye along with the Qubit spectrofluorometer is required (16). If residual Trizol carryover is present, an absorbance at 270 nm (Fig. 2) may be observed and interface with the absorbance value at 260 nm used for RNA and other nucleic acids resulting in erroneous quantification data. In cases where this carryover is problematic, further purification steps may be necessary. This may include an adjustment to the Trizol procedure to include an additional chloroform wash or a subsequent RNA cleanup step using a standard RNeasy MinElute column provided there is sufficient RNA available (see Note 19). RNA integrity was analyzed using the Agilent 2100 Bioanalyzer by loading 1 ml of sample RNA into the appropriate analysis cassette according to the manufacturer’s protocol. For low RNA recovery samples (e.g., 25 ml), it is not economical to maintain a final RNase inhibitor concentration at 20 U/ml. Regardless of the final concentration selection, it is most important to maintain consistency for samples belonging to the same experiment. 12. During a sort, it is not recommended to sort into RNAlater or other ammonium sulfate solutions as the resulting viscosity will be too high to centrifuge the cells properly and result in poor cellular recovery and compromised RNA quality. This is not unexpected as this reagent is designed for tissue preservation and not for purified cells from FACS (18). 13. A DNase I treatment will be required when downstream methods involving random hexamer priming such as in the case of exon microarrays, RT-qPCR, or equivalent. If no DNase I treatment is required, it should be omitted as results from our laboratory indicate that an expected loss of 30–40% of RNA may be observed when performing an on-column digestion (unpublished data). 14. Freezing directly sorted extracts in Trizol or RLT buffer often results in degraded RNA and is not recommended. However, we have observed that samples that are maintained at 4°C overnight in a dilute (~20%) Trizol LS solution followed by proper RNA extraction the next day have resulted in good quality RNA. Any storage method should be evaluated on each sample type prior to beginning an experiment because some cell types do not tolerate any lengthy Trizol or RLT exposure. 15. Although direct sorting of cells into RNA extraction buffer will negate a postsort cell purity check, this can still be performed by separately sorting a fraction of the cells into another tube containing PBS with 2% BSA, so that purity analysis can be performed after FACS is complete. 16. The example below indicates the amount of each reagent required to process a sample from the method outlined in Subheading 3.3.2. In this example, the cell lysis, nucleic acid

Isolation of Total RNA from Transgenic Mouse Melanoma Subsets Using Fluorescence

43

separation, and ethanol steps must be done in either a 15 or 50 ml RNase-free centrifuge tube and a vacuum manifold will be needed for processing the RNeasy microcolumn: Original Trizol LS in FACS Tube (presort)

0.5 ml

Sorted volume (Total volume postsort minus 0.5 ml Trizol LS above)

2.2 ml

Trizol LS needed to maintain 3:1 ratio (Trizol:sample ratio, 6.6:2.2)

6.1 ml

Amount of chloroform needed (0.2 ml/0.75 ml Trizol LS)

1.8 ml

Total volume for centrifugation

10.6 ml

Recovered aqueous phase (AQP)

4.5 ml

100% ethanol needed (1.5 × AQP v/v)

6.8 ml

Total volume to be applied to column

11.3 ml

17. The use of alternative organic phases in Trizol precipitations, such as 1-bromo-3-chloropropane (BCP) and 4-bromoanisole (4BA), has proven to be less desirable in our facility as their vapor pressures are low and do not benefit by evaporating from the final sample such as in the case of chloroform. 18. When using larger centrifuge tubes to processing larger volumes of the Trizol sample mix, it is not possible to centrifuge at 12,000 × g, and we have found that spinning as low as 1,000 × g results in good quality RNA. 19. Unfortunately, the Nanodrop and Qubit instruments cannot effectively discriminate RNA from DNA and other 260 nm absorbing contaminants. Any resulting DNA contamination must either be characterized or digested before proceeding to downstream reactions. Trizol carryover (absorbance at 270 nm) can sometimes be minimized by adding an additional chloroform cleanup step. This is done by combining the recovered aqueous phase with an equal volume of fresh chloroform at step 7 in Subheading “Direct RNA Extraction Using a Combined Trizol LS-RNeasy Method”. The tube is mixed and incubated at room temperature before centrifuging at full speed. The resulting aqueous phase is then processed exactly as the original aqueous phase at step 7 of Subheading “Direct RNA Extraction Using a Combined Trizol LS-RNeasy Method” by combining with a 1.5× volume of ethanol.

44

Tighe and Held

References 1. Gershon D (2005) DNA microarrays: more than gene expression. Nature 437:1195–1198 2. Rappa G, Fodstad O, Lorico A (2008) The stem cell-associated antigen CD133 (Prominin-1) is a molecular therapeutic target for metastatic melanoma. Stem Cells 26:3008–3017 3. Zabierowski SE, Herlyn M (2008) Melanoma stem cells: the dark seed of melanoma. J Clin Oncol 26:2890–2894 4. Klein WM, Wu BP, Zhao S, Wu H, KleinSzanto AJ, Tahan SR (2007) Increased expression of stem cell markers in malignant melanoma. Mod Pathol 20:102–107 5. Mizrak D, Brittan M, Alison MR (2008) CD133: molecule of the moment. J Pathol 214:3–9 6. Barrett MT, Glogovac J, Prevo LJ, Reid BJ, Porter P, Rabinovitch PS (2002) High-quality RNA and DNA from flow cytometrically sorted human epithelial cells and tissues. Biotechniques 32:888–896 7. Mack E, Neubauer A, Brendel C (2007) Comparison of RNA yield from small cell populations sorted by flow cytometry applying different isolation procedures. Cytometry A 71:404–409 8. D’Alessio G, Riordan JF (1997) Ribonucleases: structures and functions. Academic, San Diego, CA

9. Beintema JJ (1998) Introduction: the ribonuclease A superfamily. Cell Mol Life Sci 54:763–765 10. Bosenberg M, Muthusamy V, Curley DP, Wang Z, Hobbs C, Nelson B, Nogueira C, Horner JW, Depinho R, Chin L (2006) Characterization of melanocyte-inducible Cre recombinase transgenic mice. Genesis 44:262–267 11. Introduction to flow cytometry: a learning guide (2002) Becton, Dickinson and Company. 11-11032-03 rev. A 12. TRIzol reagent and TRIzol LS reagent technical note. Invitrogen Corp. Carlsbad, California 13. RNeasy® Micro Kit handbook (2007) QIAGEN sciences. Germantown, MD 14. QIAvac® 24 Plus handbook (2005) QIAGEN sciences. Germantown, MD 15. Qubit™ fluorometer instruction manual (2007) Invitrogen Corp. Carlsbad, California 16. Quant-iT™ RiboGreen RNA assay kit. Invitrogen Corp. Carlsbad, California 17. Kuschel M, Ausserer W (2000) Characterization of RNA quality using the Agilent 2100 Bioanalyzer. Agilent Technologies Application Notes 18. RNAlater® handbook (2006) QIAGEN sciences. Germantown, MD

Chapter 3 Microarray Analysis of Embryonic Stem Cells and Differentiated Embryoid Bodies Alexander C. Zambon and Christopher S. Barker Abstract By altering the cellular microenvironment and culture media composition, embryonic stem cells (ESCs) can be induced to differentiate in vitro into somatic cell types from the three primitive germ layers. ESC differentiation is regulated by an intricate series of signaling events that result in their transcriptional reprogramming, asymmetric cell division, and differentiation. Using various pharmacological agents and/or genetic manipulations, one can drive and enrich ESC differentiation to specific cell lineages. Identifying the transcriptional fingerprint during ESC differentiation could yield novel targets for genetic or pharmacological regulation to increase the yield of desirable cell types. We discuss here how to culture undifferentiated mouse ESCs (E14 line from 129/Ola) and generate embryoid bodies (EBs). We also discuss in detail how to prepare Affymetrix samples, how to hybridize and scan arrays, and how to quality control data and generate signal values and permutation based P-values. Key words: Embryonic stem cells, Stem cell differentiation, Embryoid bodies, Expression profiling

1. Introduction Culturing mouse ESCs (1) in vitro was a major scientific breakthrough that led to a series of significant biomedical research advances in transgenic (2) and knockout mouse models (3, 4) and provided valuable insight for the subsequent culture of human ESCs (5). Cultured ESCs have been used as a developmental model system to study gene and signaling networks that drive stem cells to differentiate into specific somatic cell types (e.g., cardiac myocytes (6)). Microarrays provide an opportunity to make unbiased genome-wide surveys to identify the transcriptional fingerprints of the gene networks that drive ESC differentiation (7) into somatic cell types. The use of genetically engineered

Sridar V. Chittur (ed.), Microarray Methods for Drug Discovery, Methods in Molecular Biology, vol. 632, DOI 10.1007/978-1-60761-663-4_3, © Humana Press, a part of Springer Science+Business Media, LLC 2010

45

46

Zambon and Barker

ESCs that express selectable transgenes (e.g., neomycin) (8) or fluorescent marker genes with tissue-specific promoters enables one to enrich and purify desired cell types. Such lines are well suited for genome-wide expression profiling. Several considerations should be made when choosing an ESC line, including the strain of mouse from which the line was derived from (considering that the ESC line will be used for genetargeted mutations and the generation of chimeric mice), whether or not the line requires coculture with mitotically inactive embryonic fibroblast feeder layers (i.e., culture of “feeder free” ESCs is less labor-intensive), and the potential for in vitro differentiation into desired cell types. This last consideration is supported by evidence of variability in the cardiogenic potential of various human (9) and mouse (8) ESC lines. It is important to note that variations in the culture conditions and genetic background of ESC lines can have a dramatic effect on gene expression signatures and should be taken into consideration when planning and interpreting expression profiles of ESCs and ESC-derived cells (10). A variety of microarray platforms and sample preparations have been described (for review (11)). The most commonly used array platforms available today are Affymetrix, Agilent, and Illumina microarrays. When selecting an array it is important to keep in mind that while different array suppliers may detect the same RNA transcript, the exact probe sequences used on each array can be quite different and located on different exons within a transcript. As a result, it can be problematic to directly compare data from similar samples that were run on different kinds of arrays. We recommend that array users use the same array across multiple data sets to facilitate future meta-analyses. While we discuss the classical sample preparation for Affymetrix microarrays by in vitro transcription, in many cases, it is not possible to obtain the amounts of RNA required for the described protocol. Numerous commercial sample preparation kits available also work quite well, including Affymetrix GeneChip One-Cycle Target Labeling kit (³1 µg of total RNA needed per sample), Applied Biosystems MessageAmp II – Biotin Enhanced kit (³100 ng of total RNA/sample), NuGEN Technologies WT-Ovation Pico kit (³500 pg of total RNA/sample), Molecular Devices Arcturus RiboAmp HS Plus kit (³100 pg total RNA/sample), and NuGEN Technologies WT-Ovation FFPE kit (³50 pg total RNA/sample). When selecting sample amplification methods, it is best to scale the reactions for the study to the sample with the least amount of RNA, and then pick the sample preparation kit that best meets those needs while ensuring that the sample preparation is compatible with the microarray platform used. In analyzing microarray results, a variety of strategies and techniques can be employed that are beyond the scope of this

Microarray Analysis of Embryonic Stem Cells and Differentiated Embryoid Bodies

47

chapter (for reference (12, 13)). In planning microarray studies, typically, the largest source of variation in a study is due to biological variation. As a result, we highly recommend designing studies with multiple (three or more) experimental, not technical, replicates to allow the use of statistical analysis to reduce experimental noise and allow the researcher to validate these changes with alternate methods (e.g., real-time PCR). We also recommend isolating RNA from extra experimental replicates (when possible) in case of RNA contamination or problems during microarray processing. The protocol below highlights one of these cases, in which there was unusable data generated by an array despite multiple quality control checks of the sample. Once expression signal values are generated, multiple testing procedures or other statistical tests can be conducted to define which transcripts show different expression. For a two sample comparison (i.e., ESCs compared to EBs), we generated permutation based unadjusted and Westfall and Young multiple-testing adjusted P-values (14) and employ a greater than twofold and P < 0.05 cutoff to call a gene differentially expressed. Most of the basic analysis can be done directly in Excel or another spreadsheet with basic search and filtering functions. To generate permutation-based P-values and to quality control array images, we use several R-based (http://www.r-project.org/) statistical packages that are freely available from the Bioconductor (15) website (http://www.bioconductor.org). R is a free software environment for statistical computing and graphics and in conjunction with packages available from Bioconductor can be used to both visualize and analyze a variety of genomic data sets (e.g. SAGE, SNP arrays). For Affymetrix arrays, we use the following R packages. For quality control of scanned microarray images, we use the affyQCReport. To generate log2 expression signal values, we use the gcRMA package. It has greater accuracy and precision than other available algorithms (16). To generate permutation-based unadjusted and Westfall and Young adjusted p-values, we use the multtest package.

2. Materials 2.1. Mouse E14 Feeder-Independent ESC Culture

1. Dulbecco’s phosphate-buffered saline (PBS) without calcium and magnesium. 2. 1000× beta-mercaptoethanol (b-ME) stock solution: Add 70 µl of b-ME to 20 ml of distilled, deionized water. Filter sterilize with a 22 mm Steriflip (Millipore, Billerica, MA), and store at 4°C for up to 2 weeks.

48

Zambon and Barker

3. ESC culture medium: Glasgow MEM/BHK21 medium (Sigma, St. Louis, MO) supplemented with 10% ES cell-characterized FBS (Hyclone, Logan, UT), 1× MEM nonessential amino acids (Invitrogen, Carlsbad, CA), 2 mM L-glutamine (Invitrogen), 1 mM Na-Pyruvate (Invitrogen), 1× b-ME, 1 × 106 units/L of ESGRO (Millipore). Culture medium can be stored at 4°C for up to 4 weeks; after that, resupplement the remaining medium with L-glutamine. 4. ESC trypsin solution: Add 100 mg of EDTA tetrasodium salt to 500 ml of PBS. Filter-sterilize and add 10 ml of 2.5% porcine trypsin solution (Invitrogen) and 5 ml of chicken serum (Invitrogen). Store as 20 ml aliquots at −20°C (avoid multiple freeze-thawing cycles). 5. 0.1% gelatin solution: Add 25 ml of a 2% bovine gelatin solution (Sigma) to 500 ml of PBS. Store at 4°C. 2.2. Embryoid Body (EB) Formation by Hanging Drops

1. EB differentiation medium (EBDM): Glasgow MEM/ BHK21 medium supplemented with 20% ES cell characterized FBS, 1× MEM nonessential amino acids, 2 mM L-glutamine, 1 mM Na-Pyruvate, and 1× b-ME. 2. 96-well sterile conical bottom polypropylene plates (E&K Scientific). 3. Sterile 96-well plate lids (E&K Scientific). 4. Wide orifice tips (Rainin RT-L250WS).

2.3. Total RNA Extraction

1. TRIzol Reagent (Invitrogen). 2. Phase Lock Heavy Gel Tubes (2 ml) (Eppendorf). 3. RNAeasy Mini Kit (Qiagen).

2.4. cDNA Synthesis

1. T7-(dT)24 Primer, HPLC Purified (Operon Technologies). 2. Superscript double-stranded cDNA Synthesis kit (Invitrogen).

2.5. cRNA Synthesis and Labeling

1. BioArray High Yield DNA Transcript kit (Affymetrix). 2. RNAeasy Mini Kit (Qiagen). 3. 5× fragmentation buffer: 200 mM Tris–acetate, pH 8.2, 500 mM potassium acetate, 150 mM magnesium acetate.

2.6. Genechip Hybridization

1. U430 2.0 GeneChip (Affymetrix). 2. GeneChip Eukaryotic Hybridization Control Kit including 20× hybridization controls and control oligonucleotide B2 (Affymetrix). 3. 12× MES stock: Resuspend 70.4 g MES hydrate (Sigma) plus 193.3 g MES sodium (Sigma) in 800 ml of molecular biology grade H2O (Gibco), mix and adjust volume to 1 l. The pH

Microarray Analysis of Embryonic Stem Cells and Differentiated Embryoid Bodies

49

should be 6.5–6.7. Filter through a 0.2 µM filter. Store at 4°C in a foil-covered bottle (protect from light). 4. 2× hybridization buffer: 19.9 ml molecular biology grade H2O + 8.3 ml of 12× MES stock + 17.7 ml of 5 M NaCl (Ambion) + 4.0 ml of 0.5 M EDTA (Sigma) + 0.1 ml of 10% Tween 20 (Pierce). Store at 4°C in a foil-covered bottle (protect from light). 5. Herring sperm DNA 10 mg/ml (Promega). 6. DMSO (Sigma). 2.7. GeneChip Wash, Stain and Scan

1. Prepare antibody and stain solutions immediately before use. 2. Wash buffer A (nonstringent): 300 ml of 20× SSPE (Fisher) + 1 ml of 10% Tween 20 + 699 ml of molecular biology grade H2O to final volume 1 l and filter through a 0.2-µm filter. 3. Wash buffer B (stringent): 83.3 ml 12× MES Buffer + 5.2 ml 5 M NaCl + 1 ml 10% Tween 20 + 910.5 ml Molecular Biology Grade H2O to final volume 1 l and filter through 0.2-µm filter and store at 4°C, protected from light. 4. 2× stain buffer: 41.7 ml 12× MES Buffer + 92.5 ml 5 M NaCl + 2.5 ml 10% Tween 20 + 113.3 ml molecular biology grade H2O to a final volume of 250 ml and store 4°C protected from light. 5. Goat IgG stock 10 mg/ml (Sigma): resuspend 50 mg in 5 ml of 150 mM NaCl and store at 4°C. 6. SAPE stain solution: 600 µl of 2× stain buffer + 48 µl of 50 mg/ml BSA (Invitrogen) + 12 µl of 1 mg/ml streptavidin/phycoerythrin (SAPE) (Invitrogen) + 540 µl of DI H2O. Mix well and divide into two aliquots of 600 µl each. 7. Antibody solution: 300 µl of 2× stain buffer + 24 µl of 50 mg/ml BSA + 6 µl of 10 mg/ml goat IgG stock + 3.6 µl of 0.5 mg/ml biotinylated antibody (Vector Laboratories) + 266.4 µl of DI H2O.

3. Methods We use a feeder-independent ESC line derived from the 129/ Ola strain of mice as shown in Fig. 1 (17). These cells are easy to maintain and significantly reduce the amount of tissue culture required. The parental cell line E14Tg2A (denoted as E14 herein) was established from delayed blastocysts on gelatinized tissue culture dishes in ES cell medium containing leukocyte inhibitory factor (LIF) (17). Sublines were isolated by plating cells at a single-cell density, picking and expanding single colonies,

50

Zambon and Barker

Fig. 1. Phase images of (a) mouse E14 ESCs and (b) 5-day-old embryoid bodies. Images captured at 20×

and testing several clones for germline competence. To differentiate ESCs into embryoid bodies that contain spontaneously beating cardiac myocytes, we have made slight modifications to the method of Boheler et al. (18). We isolated RNA from seven T25-cm flasks of undifferentiated E14 cells and seven 10-cm dishes each containing EBs recovered from one 96-well plate of EBs. Eight days after the initiation of hanging drops, beating cardiomyocytes could be visualized with a microscope. We then prepared five samples of either ESC or EB total RNA for quality control and array hybridization and analyzed the resultant data using the protocols described. Yields at various steps in the protocol are reported in Table 1 for reference. 3.1. Mouse E14 Feeder-Independent ESC Culture

1. Coat a 25-cm2 tissue-culture flask with 0.1% gelatin and aspirate off the excess immediately before use. 2. Thaw ES cells (approximately 2.5 × 106 cells, equivalent to ½ of a confluent 6-well or 1/4 of a confluent 25-cm2 flask) in a 37°C water bath and dilute into 10 ml of prewarmed ES cell medium.

Microarray Analysis of Embryonic Stem Cells and Differentiated Embryoid Bodies

51

3. Collect the cells by spinning for 3 min at 1,100 rpm (130 g) in a bench-top clinical centrifuge. 4. Aspirate off medium and gently resuspend cells in 10 ml of prewarmed medium. 5. Transfer cell suspension to a 25-cm2 flask and grow at 37°C in a humidified 7% CO2 incubator. 6. Change medium the following day to remove dead cells and residual DMSO (see Note 1). 7. ES cells are routinely passaged every 2 days, and the medium is changed on alternate days. Thus, ES cells require daily attention (see Fig. 1a for an example of subconfluent E14 ES cells) (see Note 2). To passage every other day, aspirate the culture media, rinse with 5 ml PBS, aspirate and add 1 ml of ESC trypsin and incubate in the tissue culture incubator for ~3 min. Neutralize with 9 ml of ESC media and passage at 1/10 split. 3.2. Embryoid Body Formation by Hanging Drops

1. Day 0: From a confluent 25-cm2 flask of cells, aspirate off the medium and wash with 5 ml of room temperature PBS, pipetting it away from the cells. Rock flask gently and aspirate medium. 2. Cover cells with 1 ml of 1× trypsin solution, and return to 37°C incubator for 2 min or until cells are uniformly dispersed into small clumps. 3. Add 9 ml of EBDM to inactivate the trypsin and pipette up and down gently to create a single cell suspension. 4. Count cells and dilute to 25,000 cells/ml (approximately 40-fold) in EBDM. Using a multichannel pipet, transfer 20 ml (500 cells) to the center of each well of a 96-well polypropylene plate with conical bottoms. Each plate will require 2 ml of cell suspension. Invert plates gently and incubate at 37°C for 2 days. 5. Day 2: Invert plate right side up and use multichannel pipet to add 200 µl/well fresh 20% EBDM. 6. Day 5: Use multichannel pipet to remove 100 µl of medium, being careful not to disturb developing EB at bottom of well, and replace with 100 µl of fresh EBDM. 7. Day 7: EBs are collected from the 96-well plate by rinsing/ scraping the V-bottom wells with a multichannel pipet set at 150 µl and wide orifice tips. Transfer to a sterile reservoir with ~5 µl of EBDM (see Fig. 1b for an example of a Day 7 EB). After all the EBs have been collected, transfer to a 50 ml conical tube, and allow the EBs to sediment by gravity flow (~5 min). After the EBs have settled to the bottom of the conical tube, aspirate all but ~5 ml of medium, where the settled EBs resided and transfer to a 10-cm tissue culture dish with 10 ml of fresh EBDM (1 dish/96 well plate = sample replicates).

52

Zambon and Barker

Change the medium every 2–3 days and look for beating foci with a microscope (you should see beating areas). 3.3. Total RNA Extraction

1. Add 1 ml of TRIzol Reagent per 10-cm dish of EBs or ES cells. 2. Scrape cells with a cell scraper and pipet up and down several times with a 1 ml pipet. 3. Incubate the homogenized sample for 5 min at R/T. 4. Add 0.2 ml of chloroform per ml of TRIzol reagent. Cap tubes securely. 5. Handshake for 15 s and incubate for 2–3 min at R/T. 6. Transfer aliquots of 500 µl (up to 750 µl) homogenates to prespin (12,000 rpm (15,300 g) for 30 s) heavy phase lock tubes. 7. Centrifuge for 10 min at 12,000 rpm (15,300 g) at 4°C. 8. Remove upper colorless aqueous phase remaining the RNA to a fresh tube. 9. Precipitate RNA with 0.5 ml of isopropanol per ml of TRIzol reagent. 10. Incubate for 10 min at R/T. 11. Centrifuge at 12,000 rpm (15,300 g) for 10 min at 4°C. 12. Remove the supernate carefully. 13. Wash pellet with 1 ml of 75% ethanol per ml TRIzol Reagent. 14. Vortex and centrifuge at 9,000 rpm (8,600 g) for 5 min at 4°C. 15. Remove the supernate and briefly dry the RNA pellet by air-dry or vacuum-dry for 5–10 min. Do not dry RNA by centrifugation under vacuum. It is very important not to let the RNA pellet dry completely. 16. Dissolve RNA in RNase-free water by passing the solution a few times through a pipet tip and incubating for 10 min at 55–60°C. 17. Take 1 µl or an aliquot for quality and quantity measurements. 18. Store sample at −80°C, if necessary, or go on to next step. 19. Adjust the volume of the total RNA sample to 100 µl with RNase-free water. 20. Add 350 µl buffer RLT (with b-ME) (RNAeasy Mini Kit) to the sample and mix thoroughly (see Note 3 and Table 1 for yield). 21. Add 250 µl of 100% ethanol to the lysate and mix by pipetting. 22. Apply sample (700 µl) to an RNAeasy column sitting in a 2-ml collection tube. Spin for 15 s at max speed. Discard the flow-through.

Microarray Analysis of Embryonic Stem Cells and Differentiated Embryoid Bodies

53

Table 1 Yields for microarray sample preps Total RNA Sample # Sample yield (mg)

Post cleanup Adjusted A260/A280 [RNA] mg/ml A260/A280 IVT yield [IVT]

1

EB

137.2

1.47

0.86

1.90

47.90

43.75

2

ES

89.4

1.52

2.21

1.70

45.98

41.83

3

EB

142.4

1.38

0.79

1.90

53.09

48.94

4

ES

103.8

1.53

1.27

1.90

58.50

54.35

5

EB

143.7

1.30

0.76

1.90

51.55

47.40

6

ES

96.9

1.55

1.92

1.90

59.74

55.59

7

EB

52.8

1.83

0.84

1.90

39.49

35.34

8

ES

75.1

1.79

1.37

1.90

47.62

43.47

9

EB

53.5

1.83

0.97

1.90

47.01

42.86

10

ES

57.8

1.80

1.42

1.90

59.49

55.34

23. Add 500 µl RPE buffer (ethanol added) onto the column and spin for 15 s at max speed to wash. Discard the flow-through. 24. Add additional 500 µl RPE buffer and spin at max speed for 2 min to dry RNAeasy membrane. 25. Carefully transfer column to a new 1.5-ml tube and pipet 30–50 µl of RNase-free water directly onto the membrane. Wait for 3–4 min. Spin at max speed for 1 min to elute. 26. Repeat step 7 if more than 30 µg RNA yield is expected and elute into the same collection tube. 27. Take a 1-µl aliquot for quality assessment. We recommend using the Agilent 2100 BioAnalyzer (see Fig. 2a for a representative and acceptable tracing of total RNA). The output file will also generate an RNA integrity number (RIN). The RIN is generated on a scale of 1–10 (poor to excellent quality). We recommend using samples with RIN ³ 7. An approximation of the amount of RNA can also be derived from this tracing, but this measurement is frequently imprecise. 28. If there is no access to an Agilent 2100 BioAnalyzer, check the total RNA quality on 1% agarose (RNase-free) gel by loading 1 µl of the total RNA sample. Treat gel equipment with 3% peroxidase before use. Use RNase-free water when making TAE buffer needed for the agarose gel preparation and electrophoresis buffer. Run at 60 V for 30 min or until RNA bands are well separated. Look for 2 kbp (28S), 0.9 kbp (18S), and 200 bp (5S) ribosomal RNA bands.

Zambon and Barker

a 6

total RNA

Fluorescence

5 4 3 2

19

b

24

29

34

3

1

0

2

1

39 44 49 Time (seconds)

54

59

64

69

2.5

cRNA

Fluorescence

2.0 1.5 1.0 0.5 0.0 19

24

29

34

39

44

49

54

59

64

69

Time (seconds)

c 2.5

Fragmented cRNA

2.0 Fluorescence

54

1.5 1.0 0.5 0.0 19

24

29

34

39 44 49 Time (seconds)

54

59

64

69

Fig. 2. Agilent BioAnalyzer tracings of (a) total RNA (b) cRNA (c) fragmented cRNA

29. The amount of RNA should be determined spectroscopically by measuring the A260 value by standard methods. The A260/A280 OD ratio should be 1.8–2.0 for pure RNA when RNA sample diluted in 10 mM Tris–HCl, pH 7.5.

Microarray Analysis of Embryonic Stem Cells and Differentiated Embryoid Bodies

3.4. cDNA Synthesis

55

1. Add DEPC treated H2O to 16–24 µg of total RNA (no more than 24 µg) to achieve the final volume of 9 µl. 2. Add 1 µl of the T7-(dT)24 primer (100 pmol/µl)/sample. 3. Incubate for 10 min at 70°C in PCR machine with heated cover. 4. Quick spin and then put the samples on ice. 5. Add 4 µl of 5× first-strand cDNA buffer, 2 µl 0.1 M DTT, 1 µl of 10 mM dNTP mix. 6. Mix and incubate at 42°C for 2 min. 7. Add the 3 µl of the SSII RT (final volume = 20 µl). 8. Mix well and incubate at 42°C for 1 h. 9. Set the water bath to 16°C. Spin samples briefly to bring down condensation on side of tube. 10. On ice, add the following reagents, in the order shown, to the first strand reaction tube: 91 µl of DEPC-treated water, 30 µl of 5× second strand buffer, 3 µl of 10 mM dNTP mix, 1 µl of 10 U/µl DNA ligase, 4 µl of 10 U/µl DNA polymerase I, E. coli, 1 µl 2 U/µl RNase H, E. coli (final volume, 150 µl). 11. Tap the tube and mix. Spin briefly and incubate for 2 h at 16°C. 12. Add 2 µl (10 U) of T4 DNA polymerase and continue incubating for 5 min at 16°C. 13. Place the reaction on ice and add 10 µl of 0.5 M EDTA. 14. Store at −20°C or proceed with cleanup steps. 15. Pellet the material in a 1.5-ml green phase lock light tube (PLG) at max speed for 30 s. 16. Add 162 µl phenol:chloroform:isoamyl (25:24:1) alcohol to the final volume (162 µl) of the cDNA reaction (total volume 324 µl). Vortex thoroughly. 17. Transfer the entire volume to the PLG tubes. Do not vortex. 18. Spin at maximum speed for 2 min. 19. Transfer the aqueous upper phase to a new tube. 20. Add 0.5 volume of 7.5 M NH4Ac and 2.5 volume of 100% ethanol (stored at −20°C). 21. Vortex and spin at maximum speed for 20 min at R/T. 22. Remove the supernatant carefully. Wash with 1 ml of 80% ethanol (stored at −20°C). Spin for 5 min. Discard supernatant. Repeat once. 23. Air-dry pellet. Resuspend the pellet in 12 µl of DEPC-treated H2O. 24. Remove resuspended sample and place in fresh 200-µl PCR tube (proceed or store at −20°C).

56

Zambon and Barker

3.5. cRNA Synthesis and Labeling

1. Add the reagents in the following order to new RNase-free tubes (final volume: 40 µl) at R/T: 4 µl of cDNA, 18 µl of DEPC-treated H2O, 4 µl of 10× reaction buffer (BioArray High-Yield DNA Transcript kit), 4 µl of 10× biotin-labeled ribonucleotides, 4 µl of 10× DTT, 4 µl of RNase inhibitor mix, and 2 µl of T7 RNA polymerase enzyme. 2. Carefully mix and then spin briefly. 3. Immediately place the tube at 37°C. Incubate for 5 h (the longer, the higher yield). Gently mix the tube every 45 min during the incubation. 4. Store at −20°C if not purifying cRNA immediately. 5. Bring the volume of the IVT reaction to 100 µl with 60 µl of RNase-free water, then add 350 µl Buffer RLT (with b-ME) to the sample and mix thoroughly. 6. Add 250 µl of 100% ethanol to the lysate and mix well by pipetting. 7. Apply sample (700 µl) to an RNAeasy column with a new 2-ml collection tube (supplied in the kit). Spin for 15 s at max speed. Discard the flow-through. 8. Add 500 µl diluted RPE buffer and centrifuge for 15 s at maximum speed to wash. Discard the follow-through. 9. Add additional 500 µl diluted RPE buffer onto the RNAeasy column, and centrifuge for 2 min at maximum speed to dry RNAeasy membrane. 10. Carefully (without touching ethanol), transfer RNAeasy column into a new 1.5 ml collection tube (supplied) and pipet 30 µl of RNase-free water directly onto the RNeasy membrane. Wait for 3–4 min. Spin for 1 min at maximum speed to elute. 11. Usually repeat step 10 if more than 30 µg of cRNA yield is expected (50–100 µg expected). 12. Lightly vortex tubes before quantification. Save a 1 µl aliquot for quality and quantity measurement (see Note 4). 13. Use the same procedure as described for RNA above. Check for concentration (1 OD at 260 nm equals 40 µg/ml). A260/A280 ratio of 1.8–2.0 is acceptable purity. 14. The cRNA must be at a minimum concentration of 0.6 µg/µl. If it is not, it can be concentrated with ethanol precipitation or SpeedVac Concentrator. 15. If ethanol precipitation is required, add 0.5 volumes of 7.5 M NH4Ac and 2.5 volumes of 100% (−20°C) ethanol. Vortex. 16. Precipitate at −20°C for 1 h to overnight. 17. Spin for 30 min at maximum speed in a microfuge at 4°C.

Microarray Analysis of Embryonic Stem Cells and Differentiated Embryoid Bodies

57

18. Remove the supernatant carefully. Wash with 0.5 ml of 80% ethanol (−20°C). Spin for 5 min and then discard the supernatant. Repeat once. 19. Air-dry pellet. Resuspend the pellet in 10–20 µl DEPC-treated H2O. 20. Save 1 µl for quality control analysis with the Agilent 2100 BioAnalyzer (see Fig. 2b for a representative and acceptable tracing of cRNA). 21. Use 40 µg of adjusted cRNA for fragmentation: Adjusted cRNA yield = RNAm − (total RNAi)*(Y ); RNAm = amount of cRNA measured after IVT (µg), Total RNAi = starting amount of total RNA (µg), Y = fraction of cDNA reaction used in IVT. 22. Add 2 µl of 5× fragmentation buffer for every 8 µl of RNA plus H2O. The final concentration of RNA in the fragmentation mix can range from 0.5 µg/µl to 2 µg/µl. 23. Bring the volume of 40 µg cRNA to 64 µl with RNase-free H2O. 24. Add 16 µl 5× fragmentation buffer, final concentration is 0.5 µg/µl. 25. Incubate for 35 min at 94°C. Putting on ice after the incubation. 26. Save a 1 µl aliquot for analysis on the Agilent BioAnalyzer (see Fig. 2c for a representative and acceptable tracing of fragmented cRNA). Store at −20°C or at −80°C until ready to perform the hybridization. Fragmented cRNA is very stable at −80°C. 3.6. GeneChip Hybridization

1. Heat 20× Eukaryotic Hybridization Controls to 65°C for 5 min to completely resuspend before aliquoting. 2. Mix hybridization cocktail components at room temperature: 15 mg of fragmented cRNA + 5 µl control oligonucleotide B2 + 15 µl of 20× Eukaryotic Hybridization Controls + 3 µl herring sperm DNA + 3 µl BSA + 150 µl 2× hybridization buffer + 30 µl DMSO + molecular biology grade H2O to final volume 300 µl. 3. Warm GeneChip to room temperature immediately before use. Fill GeneChip with 1× hybridization buffer and incubate at 45°C for 10 min while mixing in hybridization oven. 4. Heat hybridization cocktail to 99°C for 5 min and cool to 45°C in heating block for 5 min. 5. Microfuge hybridization cocktail for 5 min at room temperature to remove any insoluble material. 6. Remove hybridization buffer from GeneChip and fill with hybridization cocktail. Cover GeneChip septa with ToughTag spots.

58

Zambon and Barker

7. Place GeneChips in hybridization oven at 45°C and incubate rotating at 60 rpm for 16 h. 3.7. GeneChip Wash, Stain, and Scan

1. Load two 600 µl tubes of SAPE solution and one 600 µl tube of antibody solution into Affymetrix fluidics station. 2. Load GeneChip into fluidics station. 3. Select Fluidics Script EukGE-WS2v4_450 for Affymetrix Fluidics Model 450 Stations in Affymetrix Command Console software. 4. GeneChip wash and stain should take about 2 h. 5. Remove GeneChip from fluidics station and place in Affymetrix Scanner. 6. Start array scan using Command Console Software.

3.8. Quality Control of Hybridized Microarrays

The .cel files for each microarray were quality tested with the Bioconductor packages Affy and affyPLM. Please see the package help file for specifics regarding commands for running the analysis. Several plots in the reports of these analysis indicated that the data for sample 4 did not fit the RMA model well despite acceptable A260/280 ratios (Table 1) and Agilent 2100 BioAnalyzer tracings. RMA (19) is an algorithm that generates signal values from the .cel files from the scanned chip image. The chip pseudoimage (shown here in Fig. 3 in gray scale) function in the affyPLM package, which plots the weights and residuals from RMA signal

01.EB.Cel

02.ES.Cel

03.EB.Cel

04.ES.Cel

05.EB.Cel

06.ES.Cel

07.EB.Cel

08.ES.Cel

09.EB.Cel

10.ES.Cel

Fig. 3. ES and EB microarray data was quality controlled with the R program with Bioconductor plots

Microarray Analysis of Embryonic Stem Cells and Differentiated Embryoid Bodies

59

model fitting procedures. A white area indicates the probe fits the model well, and a dark area indicates the probe does not. Thus, sample 4 contained an even distribution of probes which did not fit the model well indicating a problem with either the microarray or the sample prep/hybridization. We did not detect any abnormalities of the sample at the RNA, IVT or fragmented IVT as indicated by quality control checks using the Agilent BioAnalyzer (see Fig. 2 for representative data). One can note the markings on samples 1 and 3 which are typical. These marks represent local chip or hybridization anomalies. Since the probes for a gene are distributed randomly over the chip, and since these probes are downweighted, local problems do not affect the final model and expression estimates in a significant manner, especially when there are lots of chips used in the model (10 is fairly large). We excluded sample four and proceeded to generate signal values with the gcRMA package from Bioconductor in R. Please see the package help file for specifics regarding commands for running the analysis. The resultant gcRMA signal values are log2 expression values and are converted to geometric folds with standard spreadsheet calculations. Permutation adjusted P-values were then generated with the multtest package in R. Figure 4 indicates the number of probe sets that are either up or down regulated with a permutation based P-value > 0.05 and an absolute fold change greater than 2. We would focus on these genes for downstream pathway analysis or validation and have shown previously a high degree of correlation in gcRMA generated fold changes and real-time PCR validated fold changes (20). 2000 1500

Number of probesets changed (Fold >2 and P 2, P < 0.05)

60

Zambon and Barker

4. Notes 1. ES cells are frozen in medium containing 10% DMSO. Since DMSO may induce the differentiation of ES cells, we advise thawing the cells late in the day and changing the medium the following morning to minimize the effects of residual DMSO. 2. In our experience, feeder-independent ES cells grow rapidly and quickly acidify the medium, turning it yellow. Allowing the cells to acidify the medium (by not changing the medium every day or by passaging the cells at too low a dilution) will cause the cells to undergo crisis, triggering excess differentiation and cell death, after which their pluripotency cannot be guaranteed. Plating cells at too low a density, insufficient dispersion of cells during passage, or uneven plating can cause similar problems, as the cells will form large clumps before reaching confluence, and the cells within these clumps will differentiate or die. Germline transmission is significantly reduced in cells that have been mistreated, even when they appear healthy at the time of injection. 3. Do not exceed 100 µg RNA/spin column. Add 10 µl b-ME per ml of RLT buffer. Make sure four volumes of 100% ethanol were added to the RPE buffer. All centrifugation steps should be performed at 20–25°C. We see a significant increase in RNA purity as noted by the improved A260/A280 ratios (Table 1) after RNA cleanup with RNeasy cleanup kit. 4. It is suggested to purify one half of the IVT product and check yields before purifying the second half.

Acknowledgments The authors would like to thank Drs. Whitmore Tingley and Roland Russnak for their contributions to the protocols for E14 ESC culture and embryoid body formations and Bruce Conklin and the late Karen Vranizan for their contributions to the design and interpretation of the data presented in this manuscript. We would also like to thank the Gladstone editorial department Gary Howard and Stephen Ordway for their contributions. Dedicated to the memory of Karen Vranizan. References 1. Evans MJ, Kaufman MH (1981) Establishment in culture of pluripotential cells from mouse embryos. Nature 292:154–156 2. Robertson E, Bradley A, Kuehn M, Evans M (1986) Germ-line transmission of genes

introduced into cultured pluripotential cells by retroviral vector. Nature 323:445–448 3. Capecchi MR (1989) Altering the genome by homologous recombination. Science 244: 1288–1292

Microarray Analysis of Embryonic Stem Cells and Differentiated Embryoid Bodies 4. Koller BH, Hagemann LJ, Doetschman T, Hagaman JR, Huang S, Williams PJ et al (1989) Germ-line transmission of a planned alteration made in a hypoxanthine phosphoribosyltransferase gene by homologous recombination in embryonic stem cells. Proc Natl Acad Sci USA 86:8927–8931 5. Thomson JA, Itskovitz-Eldor J, Shapiro SS, Waknitz MA, Swiergiel JJ, Marshall VS et al (1998) Embryonic stem cell lines derived from human blastocysts. Science 282: 1145–1147 6. Beqqali A, Kloots J, Ward-van OD, Mummery C, Passier R (2006) Genome-wide transcriptional profiling of human embryonic stem cells differentiating to cardiomyocytes. Stem Cells 24:1956–1967 7. Chang HY, Thomson JA, Chen X (2006) Microarray analysis of stem cells and differentiation. Methods Enzymol 420:225–254 8. Zandstra PW, Bauwens C, Yin T, Liu Q, Schiller H, Zweigerdt R et al (2003) Scalable production of embryonic stem cell-derived cardiomyocytes. Tissue Eng 9:767–778 9. Moore JC, Fu J, Chan YC, Lin D, Tran H, Tse HF et al (2008) Distinct cardiogenic preferences of two human embryonic stem cell (hESC) lines are imprinted in their proteomes in the pluripotent state. Biochem Biophys Res Commun 372:553–558 10. Allegrucci C, Young LE (2007) Differences between human embryonic stem cell lines. Hum Reprod Update 13:103–120 11. Hardiman G (2004) Microarray platforms – comparisons and contrasts. Pharmacogenomics 5:487–502

61

12. Durinck S (2008) Pre-processing of microarray data and analysis of differential expression. Methods Mol Biol 452:89–110 13. Page GP, Zakharkin SO, Kim K, Mehta T, Chen L, Zhang K (2007) Microarray analysis. Methods Mol Biol 404:409–430 14. Westfall PH, Young SS (1993) Resamplingbased multiple testing: Examples and methods for p-value adjustment. Wiley, NY 15. Gentleman RC, Carey VJ, Bates DM, Bolstad B, Dettling M, Dudoit S et al (2004) Bioconductor: open software development for computational biology and bioinformatics. Genome Biol 5:R80 16. Wu Z, Irizarry RA (2004) Preprocessing of oligonucleotide array data. Nat Biotechnol 22:656–658, author reply 8 17. Nichols J, Evans EP, Smith AG (1990) Establishment of germ-line-competent embryonic stem (ES) cells using differentiation inhibiting activity. Development 110:1341–1348 18. Boheler KR, Czyz J, Tweedie D, Yang HT, Anisimov SV, Wobus AM (2002) Differentiation of pluripotent embryonic stem cells into cardiomyocytes. Circ Res 91:189–201 19. Irizarry RA, Hobbs B, Collin F, BeazerBarclay YD, Antonellis KJ, Scherf U et al (2003) Exploration, normalization, and summaries of high density oligonucleotide array probe level data. Biostatistics 4:249–264 20. Zambon AC, Zhang L, Minovitsky S, Kanter JR, Prabhakar S, Salomonis N et al (2005) Gene expression patterns define key transcriptional events in cell-cycle regulation by cAMP and protein kinase A. Proc Natl Acad Sci USA 102:8561–8566

Chapter 4 Determination of Alternate Splicing Events Using the Affymetrix Exon 1.0 ST Arrays Sita Subbaram, Marcy Kuentzel, David Frank, C. Michael DiPersio, and Sridar V. Chittur Abstract Alternative splicing plays an important role in regulation of normal cellular function. Alternative splicing of pre-mRNA leads to the diversity of downstream protein products in the cell. The Affymetrix Exon arrays allow for a high throughput evaluation of the differences in spliced mRNA expressed in a biological system. In this study, we describe a method using this technology to study the generation of alternative mRNA transcripts in breast cancer cells that differ in the levels of a particular integrin, a3b1. Key words: Alternative splicing, Gene regulation, Expression profiling, Microarray, Exon splicing, Integrins

1. Introduction a3b1 integrin belongs to a family of heterodimeric cell surface receptors that mediate cell adhesion to the extracellular matrix. Integrins can mediate both inside-out and outside-in signal transduction, and they have been demonstrated to be involved in many aspects of cellular biology such as adhesion, migration, and survival. Laminin-332 is the primary ligand for a3b1 that is expressed in a variety of epithelial cell types. a3b1 is overexpressed in a variety of human cancers and experiments conducted in breast cancer cells have indicated an important role for this integrin in invasion (1, 2). In addition, we have shown that a3b1 in epithelial cells can induce the expression of EMT and angiogenesis promoting genes such as MMP9 and Mrp3 (3–5). a3b1-dependent induction of MMP9 gene expression was established to occur via enhanced stability of the MMP9 mRNA transcript in mouse keratinocytes, Sridar V. Chittur (ed.), Microarray Methods for Drug Discovery, Methods in Molecular Biology, vol. 632, DOI 10.1007/978-1-60761-663-4_4, © Humana Press, a part of Springer Science+Business Media, LLC 2010

63

64

Subbaram et al.

resulting in increased protein expression (3). To investigate the role of a3b1 in regulating differential gene expression as well as gene splicing events in the breast cancer cell line MDA-MB-231, we performed microarray analysis using the Affymetrix Human Exon 1.0 ST array platform (6). We have identified various candidate genes that are differentially spliced in cells that stably express an shRNA that targets the a3 integrin subunit (a3-knockdown cells), compared to cells that express a control shRNA. One of these genes was identified as POLR2I, which encodes a subunit of RNA Polymerase II. POLR2I mRNA was found to be differentially spliced at the 3¢-end, where part of Exon 6 was excluded from mRNA isolated from control breast cancer cells, but was included in mRNA from the a3-knockdown cells. This difference in Exon 6 processing could be attributed to differential usage of the 3¢-untranslated region of the gene or variations in polyadenylation.

2. Materials 2.1. Equipment

1. Agilent Bioanalyzer 2100 system. 2. Nanodrop ND-1000 UV–Vis spectrophotometer. 3. Affymetrix Genechip® System.

2.2. Materials for Cell Culture

1. MDA-MB-231 breast cancer cell lines were stably infected with lentivirus expressing a control shRNA (control cells; MISSIONTM shRNA, Sigma). 2. MDA-MB-231 breast cancer cell lines stably infected with lentivirus expressing shRNA that targets the human a3 mRNA (a3-knockdown cells; MISSIONTM shRNA, Sigma). 3. Phosphate buffered saline (PBS) 137 mM NaCl, 2.7 mM KCl, 4.3 mM Na2HPO4 7H2O, 1.4 mM KH2PO4, pH 7.4.

2.3. Materials for RNA Isolation

1. All tips, tubes, and reagent bottles must be DNase and RNase free (see Note 1). 2. Tri-reagent (Molecular Research Inc, cat#TR118) or TRIzol (Invitrogen cat#15586-026). 3. 1-Bromo-3-chloropropane(MolecularResearchInc,cat#BP151) or chloroform. 4. Isopropanol. 5. We recommend the use of nuclease-free water (Ambion cat#AM9932) to prepare all buffers and solutions. 6. RNeasy mini RNA isolation kit (Qiagen cat#74104). 7. DNase I (Ambion cat#AM2222). 8. RNase Zap (Ambion cat#AM9780).

Determination of Alternate Splicing Events Using the Affymetrix Exon 1.0 ST Arrays

2.4. Materials for RNA QC and Microarray Experiment

65

1. RNA 6000 Nanokit (Agilent cat#5067-1511). 2. GeneChip WT Sense Target Labeling and Control reagents (Affymetrix cat#900652). This catalog number includes all kits required for this protocol including cDNA synthesis, amplification, labeling, cleanup and hybridization. 3. GeneChip® Human Exon 1.0 ST arrays (Affymetrix cat#900650). 4. RiboMinus™ Transcriptome Isolation Kit (Human/Mouse) (Invitrogen cat#K1550-02). 5. Magna-Sep™ Magnetic cat#K1585-01).

Particle

Separator

(invitrogen

6. Betaine 5 M (Sigma cat#B-0300).

3. Methods 3.1. Cell Culture and Harvesting of Cells for RNA Isolation

1. Indicated cell lines were cultured in DMEM (Gibco) supplemented with 10% fetal bovine serum (Hyclone) 100 U/ml penicillin and 100 mg/ml streptomycin and 2 mM L-glutamine. Cells were maintained in 10-cm2 dishes in a 37°C incubator under 5% CO2. 2. Wash the cells with PBS to remove any residual media prior to harvesting. 3. Add 1 ml Tri-reagent or TRIzol directly to the cells in each 10-cm2 dish. Do not trypsinize the cells prior to treatment with tri-reagent or TRIzol (see Note 2). Move the TRIzol around the flask and gently tap to slough off all attached cells. Pipette into a clean tube and store at −20°C till further use.

3.2. RNA Isolation

3.3. Qiagen RNEasy Mini-Cleanup

The specific RNA isolation method that you choose will depend on your downstream application. Generally either method is acceptable for microarray, RT-PCR, or Northern blotting. The Qiagen spin-column cleanup offers the advantage of performing an optional DNase I digestion while purifying the RNA so further processing is avoided. However, detection of RNA molecules of 200 bp or smaller will be limited if using the Qiagen cleanup procedure and hence not advised if you intend to use the RNA for miRNA analysis. While using arrays such as the Exon ST 1.0, Gene ST 1.0 or Tiling arrays, ensure that the RNA is DNase treated since DNA contaminants will be amplified and labeled in the array protocol. 1. Add nuclease free water to the 150–200 ml RNA from the RNA isolation to adjust the volume to 200 ml (see Note 3). 2. Perform the RNEasy micro-cleanup as per the manufacturer’s protocol.

66

Subbaram et al.

3.4. Assessment of RNA Quality

1. Using a NanoDrop® spectrophotometer, measure the optical absorbance characteristics of the sample (see Note 4). The A260/A280 as well as the A260/A230 ratio will ideally be close to 2.0, signifying the purification of nucleic acids away from protein and other organics, respectively. If either ratio is lower than 1.6, expect problems with downstream applications of the RNA (see Note 5). 2. Performance of a NanoChip assay using Agilent’s BioAnalyzer allows for measurement of the molecular weight profile of the isolated RNA. In this way, you may evaluate the 28S/18S ratio measurements. A total RNA ratio between 1.8 and 2.0 is desirable; however, ratios 1.6–1.8 may be acceptable. A RNA Integrity number (RIN) score should be between 7 and 10 if the samples are to be used in a microarray or QPCR experiment downstream (see Note 6).

3.5. Expression Analysis of mRNA from Cells

1. While we have used many different microarray platforms for standard gene expression analysis, we recommend the use of Affymetrix Exon 1.0 ST arrays for experiments where alternate splicing is of interest. 2. There are two methods recommended by Affymetrix to amplify and label the RNA for hybridization to Exon arrays starting with 100 ng or 1 mg of total RNA. We will demonstrate the use of 1 mg protocol in this example (7) (Note 7). We have had good results with both protocols and also with the Nugen protocol, which enables starting with small amounts of RNA as seen with LCM or flow sorted samples. Please remember that since data generated by each of these protocols are not directly cross-comparable, process all samples of a given study using the same protocol.

3.6. Synthesis of Labeled cDNA and Microarray Hybridization 3.6.1. Ribosomal Reduction of 1 mg Total RNA

1. Make serial dilutions of the GenChip PolyA controls (1:20; 1:50, and finally 1:50) using the polyA dilution buffer supplied with the kit. The final concentration of the polyA controls is 1:50,000 of the original stock. 2. Add 1 part 5 M betaine to 2 parts hybridization buffer supplied in the Invitrogen ribominus kit (162 ml/sample). 3. Aliquot 1 mg total RNA in RNase free tube. The total volume of the sample should not exceed 3.2 ml. Add 2 ml of the diluted poly A controls to the sample. 4. Prepare a master mix composing of 1 ml ribominus probe (100 pmol/ml) and 30 ml of the betaine buffer per reaction. Add this to the tube from step 3. Incubate at 70°C for 5 min and then place on ice. 5. Resuspend the bottle containing the magnetic beads by flicking it until no sediment is seen at the bottom. Aliquot 50 ml

Determination of Alternate Splicing Events Using the Affymetrix Exon 1.0 ST Arrays

67

of the resuspended bead solution per reaction to a fresh tube. Add 50 ml of RNase-free water, briefly spin and place the tube on the magnetic stand for 1 min. Gently aspirate and discard the supernatant. Repeat this wash again with 50 ml water and then resuspend the beads in the hybridization buffer with betaine prepared in step 2. Spin briefly and place on the magnetic stand. Aspirate and discard the supernatant. Resuspend the beads in 20 ml hybridization buffer with betaine and incubate at 37°C for 10 min mixing once during incubation. 6. Transfer the cooled sample mix from step 4 to the bead suspension from step 5. Mix gently and incubate at 37°C for 10 min, mixing once during incubation. Place on the magnetic stand and aspirate the supernatant into a clean labeled tube. Add 50 ml of hybridization buffer with betaine to the beads, mix, place on magnetic stand, and aspirate the supernatant and combine into the previously labeled tube. The total volume of this rRNA reduced sample is approximately 100 ml. 7. Add 350 ml of cRNA binding buffer (containing ethanol) from the GenChip IVT cRNA cleanup kit to each rRNA reduced sample. Vortex and then add 250 ml of 100% ethanol to each reaction. Mix well and apply the sample to the IVT cRNA cleanup column. Centrifuge 15 s at 8,000 × g, transfer column to a fresh tube, add 500 ml cRNA wash buffer and centrifuge again for 15 s at 8,000 × g. Discard the flow through, add 500 ml of 80% ethanol to the column and spin again for 15 s at 8,000 × g. Discard the flow through, open the column cap and centrifuge for 5 min at 20,000 × g with the cap open. Transfer the column to a fresh tube and add 11 ml of RNase-free water directly to the membrane. Spin at 20,000 × g for 1 min to elute the rRNA reduced total RNA/Poly A RNA control mix. 8. Check the sample from step 7 on a bioanalyzer to ensure that ribosomal peaks are reduced in the sample. We typically see greater than 80% reduction after this protocol (see Fig. 1). Samples with less than optimal reduction may be subjected to an additional ribosomal reduction step. 3.6.2. Synthesis of Labeled cDNA

1. Prepare a 1:5 dilution of the supplied T7-(N)6 primers and add 1 ml of the diluted solution to 4 ml of the rRNA reduced total RNA/Poly A RNA control mix from step 8 of Subheading 3.6.1. Flick the tube to mix, spin down, and incubate 5 min at 70°C followed by 2 min at 4°C. Place on ice. 2. Prepare the double-stranded cDNA using the GeneChip WT cDNA synthesis kit as per the manufacturer’s protocol (see Note 8).

68

Subbaram et al.

before

after

Fig. 1. Electropherogram traces of total RNA before and after ribosomal reduction

3. This is then converted to complimentary RNA (cRNA) by in vitro transcription using the GeneChip WT cDNA amplification kit as per the manufacturer’s protocol. This protocol should yield at least 15–30 mg cRNA. 4. This cRNA (10 mg) from the first cycle is then reverse transcribed back to cDNA using random primers and a 10 mM nucleotide mix containing dNTP and dUTP. Typical yields of the sense DNA is in the range of 6–7.5 mg. 5. The uridylated single stranded cDNA (5.5 mg) is then fragmented using Uracil DNA glycolase (UDG) and human apurinic/apyrimidinic endonuclease (APE 1). This procedure fragments the cDNA reproducibly at locations where dUTP is incorporated in the DNA during the second-cycle firststrand reverse transcription step. 6. The fragmented cDNA is end labeled using terminal deoxynucleotidyl transferase (TdT) and the kit supplied DNA labeling reagent that is covalently linked to biotin.

Determination of Alternate Splicing Events Using the Affymetrix Exon 1.0 ST Arrays 3.6.3. Hybridization and Scanning

69

1. The labeled cDNA (5.5 mg) is mixed with 20× eukaryotic hybridization controls, denatured and hybridized to Human Exon 1.0 ST arrays as recommended in the kit (Note 9). 2. After 18 h hybridization, the arrays are subjected to a fluidics protocol that washes and stains the array with streptavidin phycoerithrin. 3. The stained arrays are then scanned in a GeneChip 3000G scanner and the data is exported as CEL files.

3.7. Analysis of Human Exon 1.0 ST Array Data 3.7.1. Gene Level Analysis

1. Traditional microarray analysis methods present a steep learning curve for the average user. The problem resides primarily in the normalization techniques used to distribute the signal intensities on the array. To obtain a robustly confident list of genes associated with a given condition, we use the iterPLIER algorithm as the probe intensity summarization method (6, 8). We have successfully used Agilent GeneSpring GX v10, Biotique X-ray as well as Partek Genomics software to analyze Exon array data. 2. We strongly recommend the use of replicates in the experiments using microarray technology for gene expression profiling. While we realize that these experiments can be cost prohibitive, confidence in that data from microarray experiments requires the use of at least 2–3 biological replicates. While generating preliminary data, one could resort to pooling of multiple samples to neutralize the biological variance; however, this could lead to loss of meaningful important data. 3. After summarization, we routinely conduct a Principle Component Analysis to identify any outliers in the samples. We also evaluate the control spikes and hybridization metrics as described by Affymetrix (9). 4. Next, we filter the data to exclude probesets that fall in the bottom 20th percentile for signal intensity and do not show good signal in all replicates of any given condition. This reduces the noise in the data and makes it manageable. 5. A statistical test (Students t-test or ANOVA) with a p-value 3 standard deviations above the mean signal for all proteins). These criteria may be relaxed to Z-Factor >0.35 (indicating signal:noise of 1.5) or Z-Score >2 if the stringency of the standard thresholds results in no interacting proteins being identified. Relative Signal Used (backgroundcorrected signal) values should typically be greater than 500 for proteins that interact with the small molecule. Replicate spot CV should be less than 0.5 in order to be considered an interacting protein. 7. Alternately, Microsoft Excel can be used to analyze the raw .gpr data (the .gpr file is a text file which can be open directly in Excel), although the signal scatter correction algorithm used in ProtoArray® Prospector will not be incorporated. The background-subtracted signal values (F635medianB635median) may be used to calculate Z-factors using the following formula: Z-Factor = 1–3 × (stdev feature signal + stdev negative signal)/| (avg feature signal−avg negative signal)| Interacting proteins are defined as those yielding a Z-factor > = 0.5 and a replicate spot CV < 0.5. Coefficient of variation (CV) = standard deviation/mean. As an example, the specificity profile for a tritiated small molecule generated on a high content protein microarray is shown in Figs. 3–5.

264

Kraus, Meng, and Freeman-Cook

Fig. 4. Images of ProtoArray® Protein Microarrays from Phase 2. (a) Images corresponding to ProtoArray® Human Protein Microarrays probed with 115 nM (10 nCi/µl) or 23 nM (2 nCi/µl) [3H]-SM1 in the absence of sodium chloride. The negative control array, probed with 40 pCi/µl [3H]-estradiol, is shown for comparison. (b) Images corresponding to ProtoArray® Human Protein Microarrays probed with 115 nM (10 nCi/µl) or 23 nM (2 nCi/µl) [3H]-SM1 in the presence of 150 mM sodium chloride. The negative control array, probed with 40 pCi/µl [3H]-estradiol in the presence of NaCl, is shown for comparison.

4. Notes 1. Spots may smear or merge if arrays are not equilibrated before use due to the formation of condensation on the array surface.

Small Molecule Selectivity and Specificity Profiling Using Functional Protein Microarrays

265

Fig. 5. Examples of [3H]-SM1 candidate protein interactors. The plots show background subtracted signal (Signal Used) values corresponding to each assay included in Phase 2 of the study.

2. For certain small molecule-protein interactions, such as interactions between ligand and G-protein coupled receptors, the favorable incubation temperature might be higher (ex. room temperature). 3. Assay performance with the optimal blocking reagents is small molecule-specific and can be determined through the pilot experiment described in Phase 1 of this procedure. Buffers containing casein should be heated to 50°C until casein is completely dissolved. The casein used should be Hammarsten grade casein. Do not exceed 60°C and do not microwave the solution. Buffer should be cooled to 4°C before use. 4. Use a shaker that keeps the arrays in one plane during rotation. Nutating or rocking shakers are not recommended because of increased risk of cross-well contamination. 5. The recommended activity range for the final concentration of small molecule probe is 50 pCi/ml-50 nCi/µl, with weaker interactions requiring activity of 10–50 nCi/µl. The tritiated small molecule stock activity should be at least 1 mCi/µl with a specific activity of at least 10 Ci/mmol, and a minimum of 60 mCi should be available to perform each small moleculeprotein interaction experiment. If the tritiated small molecule

266

Kraus, Meng, and Freeman-Cook

is dissolved in an organic solvent such as ethanol or DMSO, the final concentration should be less than 5% ethanol by volume or 1% DMSO by volume. To avoid non-specific interactions and/or high background, the final concentration of the small molecule should not be higher than 1 mM. 6. Do not allow any part of the array surface to dry before adding the next solution as this can cause high and/or uneven background. 7. If air bubbles are trapped under the LifterSlip™, tap the slides gently to drive them out or lift one edge of the LifterSlip™ allowing bubbles to move to the fluid front and then gently lower down again. 8. Do not remove the LifterSlip™ with forceps if the LifterSlip™ is not dislodged from the array. Continue to gently move the array in the tube until the LifterSlip™ floats off. 9. The tritium-sensitive phosphor screen will eventually be damaged due to tritium contamination. Directly washing the screen with methanol can remove some contamination, but for critical experiments we recommend the use of a new screen or a screen that has been verified to be contaminantfree by pre-exposure in an empty cassette followed by scanning and imaging. 10. The following phosphorimagers have been tested with ProtoArray® microarrays: Cyclone® Storage Phosphor System (Perkin Elmer, Inc.) and Typhoon™ Imager (GE Healthcare Life Sciences). 11. High salt concentrations can in some cases modulate binding interactions between small molecules and protein targets through ionic interactions and solvation effects at the protein surface. 12. In general, the use of pixel-based segmentation (irregular feature finding) results in more reproducible Signal Used values. References 1. Predki PF (2004) Functional protein microarrays: ripe for discovery. Curr Opin Chem Biol 8:8–13 2. Boyle SN, Michaud GA, Schweitzer B, Predki PF, Koleske AJ (2007) A critical role for cortactin phosphorylation by Abl-family kinases in PDGF-induced dorsal-wave formation. Curr Biol 17:1–7 3. Gupta R, Kus B, Fladd C, Wasmuth J, Tonikian R, Sidhu S, Krogan NJ, Parkinson J, Rotin D (2007) Ubiquitination screen using

protein microarrays for comprehensive identification of Rsp5 substrates in yeast. Mol Systems Biol 3(116):1–12 4. Hudson ME, Pozdnyakova I, Haines K, Mor G, Snyder M (2007) Identification of differentially expressed proteins in ovarian cancer using high-density protein microarrays. Proc Natl Acad Sci USA 104:17494–17499 5. Satoh J, Obayashi S, Misawa T, Sumiyoshi K, Oosumi K, Abunoki H (2008) Protein microarray analysis identifies human cellular

Small Molecule Selectivity and Specificity Profiling Using Functional Protein Microarrays

6.

7.

8.

9.

10.

prion protein interactors. Neuropathol Appl Neurobiol 35:16–35 Schnack C, Hengerer B, Gillardon F (2008) Identification of novel substrates for Cdk5 and new targets for Cdk5 inhibitors using high-density protein microarrays. Proteomics 8:1980–6 MacBeath G, Schreiber SL (2000) Printing proteins as microarrays for high-throughput function determination. Science 289: 1760–3 Ge H (2000) UPA, a universal protein array system for quantitative detection of protein-protein, protein-DNA, protein-RNA, and protein-ligand interactions. Nucleic Acids Res 28:e3 Fang Y, Frutos AG, Lahiri J (2002) Membrane protein microarrays. J Am Chem Soc 124: 2394–5 Schweitzer B, Predki P, Snyder M (2003) Microarrays to characterize protein interactions

267

on a whole-proteome scale. Proteomics 3: 2190–9 11. Huang J, Zhu H, Haggarty SJ, Spring DR, Hwang H, Jin F, Snyder M, Schreiber SL (2004) Finding new components of the target of rapamycin (TOR) signaling network through chemical genetics and proteome chips. Proc Natl Acad Sci USA 101:16594–9 12. Singh J, Salcius M, Liu S-W, Staker BL, Mishra R, Thurmond J, Michaud G, Mattoon DR, Printen J, Christensen J, Bjornsson JM, Pollok BA, Kiledjian M, Stewart L, Jarecki J, Gurney ME (2008) DcpS as a therapeutic target for Spinal Muscular Atrophy. ACS Chem Biol 3:711–22 13. Zhang J-H, Chung TDY, Oldenburg KR (2000) Confirmation of primary active substances from high-throughput screening of chemical and biological populations: A statistical approach and practical considerations. J Com Chem 2:258–265

Chapter 17 Production and Application of Glycan Microarrays Julia Busch, Ryan McBride, and Steven R. Head Abstract Glycans are vital elements of living organisms and are involved in recognition, communication, cell growth and development, motility, and other significant processes. The interactions of glycans with the proteins that bind them provide valuable information about protein interaction and specificity. By printing glycans on microarrays, investigators are able to effectively determine the binding specificity of certain proteins with an extremely efficient and precise result. Such chips are performed by standard robotic microarray printing. Incubating the slides with various GBP-containing substances not only reveals clear receptor preferences of the proteins, but also detects minute differences in structure specificity. Key words: Glycan, Carbohydrate, Lectin, Microarray, Glycan-binding protein, Chip

1. Introduction The study of interactions between GBPs and glycan ligands has progressed greatly through the development and use of glycan microarrays. The technology of these arrays allows investigators to observe and analyze GBP interactions on individual chips, each of which can contain hundreds of different glycan structures. Arrays can be used to study a number of GBPs including antibodies specific to tumors or HIV, plant lectins, and virus GBPs such as hemagglutinin (HA) (2). Glycan arrays can be very easily customized to contain desired structures, which gives the capability to explore pathogen-specific glycan interactions (3).

Sridar V. Chittur (ed.), Microarray Methods for Drug Discovery, Methods in Molecular Biology, vol. 632, DOI 10.1007/978-1-60761-663-4_17, © Humana Press, a part of Springer Science+Business Media, LLC 2010

269

270

Busch, McBride, and Head

2. Materials 2.1. Plate Setup

1. Printing Buffer: 300 mM phosphate buffer, pH 8.5 containing 0.005% Tween-20. Store at 4°C long term, but must be at room temperature before use to avoid crystallization. 2. 96-well plates: Nunc 96-well conical bottom (Nunc 249944). 3. 96-well plate lids: Nunc 96-well cap, natural rubber, nonsterile (Nunc 276002). 4. Matrix standard clear 384 well small volume microplates. 5. Matrix universal polystyrene lid use with 96, 384, and 1,536 well plates. 6. USA Scientific Sealing Film (2921-0000). 7. Glycans (natural and/or synthetic) with amine derivatized spacers. Store at −20°C long term, but must be at room temperature when printing.

2.2. Printing

1. SCHOTT Nexterion® Slide H, polymer layer activated with N-Hydroxysuccinimide (NHS) esters which covalently binds amine groups. Store sealed in bags at −20°C long term. The slides must be removed from freezer for at least 4 h (usually overnight) before opening bags and printing. 2. ArrayIt Stealth Micro Spotting Pins (SMP4B and SMP4 models were used and have the same printing quality and produce the same spot size) or acceptable equivalent. Stored in the arrayer head or in manufacturers’ box with tip protection intact.

2.3. Humidification/ Immobilizing

1. Large Pyrex dish (5 cm × 30 cm × 15 cm). 2. Rack, such as test tube. 3. Plastic Wrap.

2.4. Numbering, Blocking, and Storage

1. VWR black lab marker. 2. Slide staining rack, 50 slide rack, and dish (Wheaton 90040). 3. Blocking buffer: 50 mM ethanolamine in 50 mM borate buffer, pH 9.2. Stored at 4°C long term but for best results, should be at room temperature before contacting slides. 4. Glass Pyrex loaf dish with fitted glass lid ( 3 1 4 × 7 7 8 × 4 1 8 ). 5. Glass slide staining rack (Wheaton 900200). 6. Drierite Anhydrous Calcium Sulfate, Indicating (Item#23005). Stored at room temperature in tray on bottom of desiccator

Production and Application of Glycan Microarrays

271

boxes. Should be replaced when color indicates-time period varies depending on how often the containers are opened and exposed to humidity. 2.5. Incubation

1. Bioworld PBS 20× solution, pH at 25°C: 7.3–7.5 stored at room temperature. 2. Bioworld PBS 1× solution made by diluting 20× solution 1:20 with ddH2O stored in 1 L quantities at 4°C. 3. Tween® 20, 500 mL (Aldrich Chemical Company, Inc.). Stored at room temperature. 4. Seracare Life Sciences Bovine Serum Albumin standard grade powder. Stored at 4°C. 5. Super PAP PEN hydrophobic slide marker or acceptable alternative. Stored at room temperature lying horizontally on its side. 6. Kimwipes. 7. Invitrogen Molecular Probes Strepdavidin Alexa Fluor 488 conjugate 2 mg/mL in PBS, pH 7.2, 5 mM azide, 5 moles dye/mole. Stored at 4°C. Light sensitive. 8. Plastic Wrap. 9. Aluminum Foil.

3. Methods The process of printing, preparing, and analyzing glycan microarrays, if done properly and carefully, will yield extremely high quality data. The slides used in these prints are very sensitive to a variety of factors including humidity, temperature, and physical contact. In order to draw conclusions based on reliable and reproducible results, special attention must be paid to the storage, handling, and application of all the components of an array. It is assumed that the reader is familiar with the general process of designing an array as well as the access to and experience with a robotic microarrayer and slide scanner. A glycan array is composed of a library of structurally defined sugars that have been modified by the addition of a linker, also referred to as a spacer, containing a terminal amine. These terminal groups bind immediately and irreversibly to the NHS esters on the slide surface during printing and are immobilized before storage to preserve the quality of the chips. By then blocking any additional unwanted bonding to the slide, clear and precise results can be obtained. To be able to analyze the glycan arrays,

272

Busch, McBride, and Head

the GBPs must be either directly or indirectly (through the use of a secondary antibody) fluorescently labeled (1). 3.1. Plate Setup

The design and layout of the glycan array should be completed before this process is started. All of the pipette work in the following method can be performed by hand or with the use of a robot, but great attention to detail and correct placement are extremely important in either case. 1. To prepare a stock of printing buffer, first make about 500 mL of a 300 mM stock solution of dibasic sodium phosphate. Also make about a 50 mL stock of 300 mM monobasic sodium phosphate solution. In a flask using a magnetic stir bar and pH meter, slowly titrate the dibasic sodium phosphate solution with the monobasic solution until a pH of 8.5 is reached. Using a 10% stock solution of Tween-20 in ddH2O, add the appropriate amount to the phosphate buffer to reach an end result of 0.005% Tween20 content. If the buffer has been prepared at a previous date, it might be necessary to place the bottle in a 37°C water bath to remove crystals that may have precipitated over time 2. The sugars are kept at a 1 mM stock concentration in printing buffer stored in 2 mL screw cap microtubes at −20°C. Thaw sugars in warm water bath of 37°C, mix by vortexing, then briefly spin down with centrifuge or small tube spinner to avoid droplets on cap 3. Add 90 mL of printing buffer to each well of the 96-well plates. 10 mL of each sugar should then be added to and thoroughly mixed in each well for a 1:10 dilution, bringing the concentration to a desired 100 mM. Our lab includes an additional concentration of 10 mM sugars on the array. To add these samples to the plates, pipette 10 mL of the 100 mM sugars into wells with 90 mL of printing buffer to perform an additional 1:10 dilution 4. Place sealing lid on 96-well plate and gently mix with vortex. Spin plates in centrifuge briefly, allowing them to reach ~2,000 rpm (1666g) for a moment just to remove any liquid that may have splashed onto the inside of the lid during mixing 5. Transfer 10 mL of each glycan from the 96 well plates to the 384 well plates 6. After the 384 well plates are complete, seal with adhesive film and spin them the same as the 96-well plates to remove any bubbles in the bottom of the wells. Keep all of the plates on ice until it is time to print the arrays

Production and Application of Glycan Microarrays

3.2. Printing

273

The authors make the assumption that the reader has access to a microarrayer and the knowledge of its use. Our lab uses a MicroGrid II 600 by Digilab, but any comparable machine would be acceptable. The slides are sensitive to humidity and as soon as the sealed bags are opened, their surface begins to hydrolyze with any moisture over the course of time. This should be taken into account when planning especially long prints at high humidity. 1. Place arrayer pins in the printing head according to the correct configuration of the desired layout 2. Remove slides from bags, open boxes, and remove any dust particles or small pieces of plastic that might be on their surfaces by blowing ultra-high purity Nitrogen gas on each slide. Anything that is on the printing surface of the slide could potentially get picked up by the pins and inhibit the ability to print. Place the desired number of slides, coated side up, on the arrayer stage (see Note 6) 3. Program the arrayer with the appropriate conditions, load the 384 well plates into the printer, and begin. Periodically check that all the pins are printing by shining a flashlight onto the stage, so the spots are visible with the glare. Relative humidity while printing should be kept between 55 and 65%. A standard hygrometer placed inside the lid of the arrayer is a good measure of chamber humidity.

3.3. Humidification/ Immobilizing

1. Once the slides have finished printing, they must undergo a temporary immobilization step, also called humidifying. This includes the slides being placed in a chamber with 100% humidity immediately after printing for a period of 30 min. The chamber can be constructed by simple placing a few very wet paper towels flat in the bottom of a large Pyrex glass dish, placing the slides on some sort of rack, such as a test tube rack, print side up, and sealing with plastic wrap to trap in the moisture 2. After the humidification step, the slides can be immediately numbered and blocked or stored in a desiccation box.

3.4. Numbering, Blocking, and Storage

Marking, blocking, and storing the slides are very important steps to preparing the slides for incubation. When the slides are bordered and numbered, the orientation of the print must be paid attention to. Desiccating conditions can be achieved by placing a tray of Drierite in the bottom of an air tight container which keeps the slides at 0% humidity for long term storage. 1. Blocking buffer can be prepared in generous quantities since a fairly large amount is used per batch of slides printed.

274

Busch, McBride, and Head

First prepare about 700 mL of a 50 mM boric acid solution. While constantly stirring and monitoring the pH, add the appropriate amount of ethanolamine from a 16.54 M stock solution to reach the desired 50 mM buffer solution. Finally, add concentrated sodium hydroxide to bring the pH up to 9.2 2. After the slides have been humidified, the spots are still visible as small crystals left by the printing buffer. It is necessary to inactivate any unbound groups on the slide surface in order to prevent any nonspecific binding. Once the blocking step has been completed, the crystals will wash away and the grids will no longer be visible on the slide surface, so the print area must be checked and marked prior to the slides being placed in the buffer. Before marking a slide, do a visual inspection to make sure that there are no large parts of grids are missing, smears, or any other major imperfections. This is also when slide numbering should occur. If the barcodes are not the preferred form of identification, a series of numbers can simply be assigned to the slides. This step is also vital to designating the directionality of the array. By numbering each of the slides in an identical fashion, the position of the array can be determined even after the print is no longer visible. This is necessary knowledge to have during the analysis step so if numbering is not used, some other marking to designate an orientation should be used. Using a black VWR lab marker on the back side (not the print surface), mark brackets around each of the four corners of the entire print and number if desired. Be sure not to mark directly behind any portion of the array which could obstruct the scanner from reading a signal (see Fig. 1) (see Note 4) 3. After each slide is marked, carefully place up to 50 into each metal rack. Place the rack inside the Pyrex loaf dish, moving the handle down, so the lid can sit properly on top. Pour enough blocking buffer just to cover the tops of the slides. Place the lid on and shake gently for 1 h 4. Once the slides have finished blocking, remove them from the dish, allowing the excess buffer to drain off. In a separate loaf dish filled about halfway with ddH2O, dip the entire rack of slides 10 times – completely removing and submerging them from the water each time. Dispose of the blocking buffer in the appropriate waste container 5. Transfer the slides to smaller glass dishes and spin dry in centrifuge at 20°C for 5 min at 200 rcf (1,024 rpm) 6. Once the slides are dry, they can be incubated immediately or stored under desiccating conditions for up to 6 months

Production and Application of Glycan Microarrays

275

Fig. 1. The view of the print surface of a slide that has been bordered and numbered on the back

3.5. Incubation

Our lab uses primarily three different protocols for incubating the glycan arrays. On every batch of slides printed, it is important to do some form of quality control. This can be performed by preparing a cocktail of biotinylated plant lectins which are incubated on the slides then fluorescently labeled. Serum can also be incubated on slides and the GBPs can then indirectly be labeled through a secondary antibody. Viruses are another sample that can be incubated on the arrays and require antisera as well as secondary antibodies for the labeling to be successful. All incubations using dangerous of infectious substances should be performed in the hood while non-hazardous samples, such as lectins, can be done on the bench. A sample volume of 1 mL should cover the print surface for most configurations, but large prints using 48 or more pins may need approximately 1.2 mL, and the same logic applies for smaller prints and volumes. For all incubations involving fluorescence, cover the humidification chamber with foil during the labeling step to avoid any loss of signal. The first four steps are universal for all glycan array incubations.

3.5.1. For All Array Incubations

1. Draw two lines, one on either side of the print area, with hydrophobic marker on outside of print area and let it dry for several minutes. Be sure to mark the slide on the print surface

276

Busch, McBride, and Head

and to go all the way to the edge of the slide to create a true barrier. The purpose of this step is to limit the area the liquid will cover on the array during incubation and also helps preserve sample volume 2. Place a couple of paper towels in the bottom of a Pyrex dish (size depending on how many slides are being incubated at a time) and soak them generously with water. Use a test tube rack or lid of some sort inside the dish to put the slides on during the incubation to raise the slides up off the wet towels. Place the dish, which is now the humidification chamber, on a rotating shaker 3. Soak the slides in PBS for 2 min to hydrate the print surface. This is an important step to eliminate nonspecific binding, which can cause high background 4. Remove the slides one at a time and dry off the backside with a Kimwipe, be careful not to wipe the print surface. Return each slide in the humidification chamber and pipette 1 mL of the sample in between the hydrophobic marker barrier (see Note 5) 3.5.2. For Plant Lectins

1. Dilute the lectin(s) in incubation buffer composed of 1× PBS and 0.05% Tween-20 to a final concentration of 10 mg/mL 2. Pipette 1 mL of the sample onto the print surface and incubate in the sealed humidification chamber while rotating gently for 1 h. Take care that the speed is not too fast and the liquid on the slide surface does not spill 3. At the end of the first incubation period, discard sample by simply pouring it off the slide into the bottom of the humidification chamber 4. Wash the slides one at a time by holding the edges and dipping them four times in a mixture of PBS and 0.05% Tween, then four times in 1× PBS, and finally four times in ddH2O 5. Dry off the backside of the slide with a Kimwipe and place back in the humidification chamber 6. Add 1 mL of Strepdavidin at 0.4 mg/mL concentration to the print surface and incubate in the sealed humidification chamber covered with foil while rotating gently for 1 h 7. After the labeling step, wash the slides with four dips in PBS/0.05% Tween-20, four dips in 1× PBS, and lastly a 3 × 3 ddH2O wash. This final wash step includes three separate dishes with ddH2O, and the slides are dipped three times in each dish 8. Spin dry in centrifuge at 20°C for 5 min at 200 rcf (same drying protocol as after blocking step) or with a gentle stream of ultra high purity Nitrogen gas

Production and Application of Glycan Microarrays 3.5.3. For Serum

277

1. Dilute the serum in incubation buffer composed of 1× PBS, 3% BSA, and 0.01% Tween-20. Various dilutions can be incubated on the arrays, or even pure serum with no incubation buffer can be used. However, more concentrated serum has a lesser specificity than that which is diluted. This is something that can be tested with a series of dilutions on a group of slides 2. Pipette 1 mL of the sample onto the print surface and incubate in the sealed humidification chamber while rotating gently for 1 h. Take care that the speed is not too fast and the liquid on the slide surface does not spill 3. At the end of the first incubation period, discard sample by simply pouring it off the slide into the bottom of the humidification chamber 4. Wash the slides one at a time by holding the edges and dipping them four times in a mixture of PBS and 0.05% Tween, then four times in 1× PBS, and finally four times in ddH2O 5. Dry off the backside of the slide with a Kimwipe and place back in the humidification chamber 6. Pipette 1 mL of a biotinylated secondary antibody onto the print surface and incubate in the sealed humidification chamber while rotating gently for 1 h. The secondary antibody is usually diluted to 10 mg/mL from its stock solution in PBS/3% BSA/0.05% Tween-20 incubation buffer, but this is also something that can be varied and tested for optimal results. Take care that the speed is not too fast and the liquid on the slide surface does not spill 7. At the end of the second incubation period, discard sample by again simply pouring it off the slide into the bottom of the humidification chamber 8. Add 1 mL of Strepdavidin at a 2 mg/mL concentration in PBS/3% BSA/0.05% Tween-20 incubation buffer to the print surface and incubate in the sealed humidification chamber covered with foil while rotating gently for 1 h 9. After the labeling step, wash the slides with four dips in PBS/0.05% Tween-20, four dips in 1× PBS, and lastly a 3 × 3 ddH2O wash. This final wash step includes three separate dishes with ddH2O, and the slides are dipped three times in each dish 10. Spin dry in centrifuge at 20°C for 5 min at 200 rcf (same drying protocol as after blocking step) or with a gentle stream of ultra high purity Nitrogen gas

3.5.4. For Viruses

1. Dilute the virus in incubation buffer composed of 1× PBS and 3% BSA. Various dilutions resulting in different HA

278

Busch, McBride, and Head

concentrations can be used on the arrays. This is something that can be tested with a series of dilutions on a group of slides (see Note 8) 2. Pipette 1 mL of the sample onto the print surface and incubate in the sealed humidification chamber while rotating gently for 1 h. Take care that the speed is not too fast and the liquid on the slide surface does not spill 3. At the end of the first incubation period, discard sample by simply pouring it off the slide into the bottom of the humidification chamber 4. Wash the slides one at a time by holding the edges and dipping them four times in a mixture of PBS and 0.05% Tween and then four times in 1× PBS 5. Dry off the backside of the slide with a Kimwipe and return it to the humidification chamber 6. Pipette 1 mL of antisera specific to the virus diluted in the PBS/3% BSA incubation buffer onto the print surface and incubate in the sealed humidification chamber while rotating gently for 1 h. Take care that the speed is not too fast and the liquid on the slide surface does not spill. The antisera is typically incubated at a 1:1,000 dilution, however, this can be varied and tested for optimal results. Take care that the speed is not too fast and the liquid on the slide surface does not spill 7. At the end of the second incubation period, discard sample by again simply pouring it off the slide into the bottom of the humidification chamber 8. Wash the slides one at a time by holding the edges and dipping them four times in a mixture of PBS and 0.05% Tween and then four times in 1× PBS 9. Pipette 1 mL of a biotinylated secondary antibody onto the print surface and incubate in the sealed humidification chamber while rotating gently for 30 min. The secondary antibody is usually diluted to 10 mg/mL from its stock solution in PBS/3% BSA incubation buffer, but this is also something that can be varied and tested for optimal results. Take care that the speed is not too fast and the liquid on the slide surface does not spill 10. At the end of the third incubation period, discard sample by again simply pouring it off the slide into the bottom of the humidification chamber 11. Add 1 mL of Streptavidin at a 2 mg/mL concentration in PBS/3% BSA incubation buffer to the print surface and incubate in the sealed humidification chamber covered with foil while rotating gently for 30 min

Production and Application of Glycan Microarrays

279

12. After the labeling step, wash the slides with four dips in PBS/0.05% Tween-20, four dips in 1× PBS, and lastly a 3 × 3 ddH2O wash. This final wash step includes three separate dishes with ddH2O, and the slides are dipped three times in each dish 13. Spin dry in centrifuge at 20°C for 5 min at 200 rcf (same drying protocol as after blocking step) or with a gentle stream of ultra high purity Nitrogen gas. 3.6. Scanning and Data Analysis

The glycan arrays can be scanned at different qualities and powers. By varying the PMT, laser power, and resolution, the instrument can produce an image with as many signals as possible with its dynamic range. Our lab uses a Perkin Elmer slide scanner with a 20-slide auto loader and normally scan at a fixed laser power while varying the PMT. It is assumed that the reader is familiar with creating a GeneID or map file, made by the arrayer or by hand in a tab-delimited spreadsheet. 1. After the arrays have been incubated, they are ready to be immediately scanned. Regardless of what type of scanner being used, take care to make sure the print surface is facing the laser source 2. Using the scanner software, set up the instrument to scan with the desired parameters. The settings can be tested and optimized, but keep in mind that each scan lowers the fluorescence of the signals. The images should be saved to a known location in the form of a TIFF file (see Fig. 2) (see Note 7) 3. Once the slides have finished scanning, transfer the images to a computer with the image processing program installed. ImaGene (Biodiscovery Inc.) is the software most often used in our lab;however, there are a variety of other acceptable programs available. Using a previously constructed GeneID or map file and customized grids, measure the array and save the data results. The placement of a sample marker known to react upon analysis is important in order to accurately place the grid file. Sample markers might include: a fluorophore like GFP; fluorescently labeled antibody; or a detectable tag like His. Our lab prints NHS-biotin, diluted to 100 nM in printing buffer, at the corner of each subarray. This allows the correct placement of the analysis grid by visualization with labeled streptavidin 4. The output file, usually in a text format, can then be opened in Microsoft Excel. With the use of a multistep macro, the data can be quickly sorted, organized, and graphed as desired for all slides of the same print format (see Fig. 3)

280

Busch, McBride, and Head

Fig. 2. An image, as it appears in ImaGene version 6.1 software, of a glycan array printed for the CFG that was incubated for quality control according to the given procedure with a cocktail of plant lectins. The slide was scanned at 60% PMT and 80% Laser Power at 10 mm resolution. The plant lectins used are from Vector and include AAL, ACL, BPL, ConA, GS-I, Jac, LEL, LTL, MAA, PTL-I, RCA-I, SBA, SJA, SNA, STL, WFA, and WGA

Production and Application of Glycan Microarrays

281

Fig. 3. The output file produced by a macro run in Excel for a quality control slide incubated with plant lectin cocktail

4. Notes 1. Only ultra high purity Nitrogen gas must be used on slides, pins, and any other pieces of equipment involved in printing. If a lower quality gas is used, the oils and debris will cause endless problems with pins sticking and printing to be defective 2. Always use gloves when handling slides, pins, and any part of the arrayer. Oils from the skin can have the same negative effect as dirty gas tanks when in contact with printing materials. It is even recommended to wash gloved hands with soap and water before handling or cleaning the pins to get rid of any powder or particles that could clog the tips 3. Even if the pins in the arrayer are clean, be sure to add at least 2 wash cycles before the first pick up of a print. This ensures no dust is in the tips, and it has been found that wetting and drying the pins briefly right before printing improved the quality of the first set of spots printed 4. Do not use Sharpie© or other regular permanent marker when marking and numbering slides. While the product description

282

Busch, McBride, and Head

may claim permanence, the ink will gradually soak off in the blocking buffer resulting in a mess and wasted slides. VWR lab markers have shown to withstand the buffer for such an extended amount of time 5. If during incubation, the wrong side of the slide is accidently wiped, the array is not necessarily destroyed. The stability of slide surface and the covalent bond formed between the slide surface and the glycans is sufficiently robust, and the print will most likely be fine except for a few minor smears 6. Attach a piece of plastic hose to the Nitrogen gas tank and insert a 1,000 mL pipette tip inside the end of tubing. Snip the tip off to create a larger opening, creating a precise and easy to use way of cleaning off slides and drying pins 7. If slides cannot be scanned immediately after incubation, they should be stored in a dark place such as a cabinet. This is not recommended because of the possible loss of fluorescence, however, sometimes it is necessary if the scanner is being used or breaks down 8. Do not use Tween-20 or other detergents in incubation buffers for viruses. The virus will break down and not bind as effectively causing less than desirable results

Acknowledgments The authors would like to thank Ola Blixt, James Paulson, Nahid Razi, and Julia Hoffmann for all of their help and guidance. Andrew Hemingway and all of Schott/Nexterion for outstanding product support. Special thanks to the Consortium for Funtional Glycomics (CFG) for making everything possible. References 1. Blixt O, Head S, Mondala T, Scanlan C, Huflejt ME, Alvarez R, Bryan MC, Fazio F, Calarese D, Stevens J, Razi N, Stevens DJ, Skehel JJ, van Die I, Burton DR, Wilson IA, Cummings R, Bovin N, Wong C-H, Paulson JC (2004) Printed covalent glycan array for ligand profiling of diverse glycan binding proteins. Proc Natl Acad Sci USA 101:17033–17038 2. Stevens J, Blixt O, Paulson JC, Wilson IA (2006) Glycan microarray technologies: tools

to survey host specificity of influenza viruses. Nat Rev Microbiol 4:857–864 3. Stevens J, Blixt O, Glaser L, Taubenberger JK, Palese P, Paulson JC, Wilson IA (2006) Glycan microarray analysis of the hemagglutinins from modern and pandemic influenza viruses reveals different receptor specificities. J Mol Biol 355:1143–1155

INDEX A

F

Absorption, distribution, metabolism, and elimination (ADME)...................100–102, 118 Adverse drug reaction .................................................... 100 Affymetrix GeneChip .............. 46, 48, 58, 64–65, 115, 161 Agilent bioanalyzer................................................... 13, 15, 17, 19, 29, 33, 37–39, 53–54, 57–59, 64, 66, 71, 161, 166 Allele-specific extension (ASE)........................... 75, 77–78 Alternate splicing ...................................................... 63–72 Apurinic/apyrimidinic endonuclease ....................... 68, 167 Automated quantitative protein expression (AQUA) analysis ........................241–242, 245–246

Feeder-independent ESC culture ........................ 47–51, 60 Flourescence activated cell sorting............................. 27–43 Flow cytometer ...............................................28, 30, 32, 39 Fragmentation ......................................... 38, 48, 54, 57, 59, 68, 105, 111–115, 122, 130–131, 167, 175–176, 178–180, 182, 191–200

G

Bacterial artificial chromosomes (BACs) ....................... 125

Gene regulation ..................................................... 102, 174 Genomics .................................... 1–3, 9, 16–17, 23, 38, 47, 69, 87, 102–103, 105–110, 118, 121, 126–132, 137, 161, 174, 192–193, 195, 198, 224, 227, 241 Genotyping ........................ 9, 100, 105–106, 116–121, 123 Globin reduction ................................................. 13–15, 25 Glycan arrays .......................... 269, 271–272, 275, 279–280

C

H

Carbohydrate ................................................................... 71 Cell differentiation .......................................................... 74 Cells-to-CT.......................................................... 91–92, 94 cell surface markers .................................................... 28, 39 Chromatin immunoprecipitation............179, 181, 186–188 Combinatorial synthesis ........................................ 204–211 Comparative genomic hybridization ............................. 126 Copy variation ............................................................... 118 CpG island .............................................142, 174, 188, 192 Cytogenetics .......................................................... 125–139

Haplotype .............................................................. 118–119 Health Insurance Portability and Accountability Act (HIPAA)..................................................... 4–5 High throughput screening ....................203–218, 221–236 HpaII tiny fragment enrichment by ligation-mediated PCR (HELP) assay ................................... 191–200 Hypermethylation ......................................................... 192 Hypomethylation .......................................................... 192

D

Illumina beadchip ...................................................... 73–86 Immunoprecipitation .....................................161, 163–167, 169–170, 176–188 In-situ hybridization...................................................... 240 Integrins .................................................................... 63–64 Isotype-control ................................................................ 30

B

DASL assay ..................................................................... 75 DataChip........................................222–223, 227, 230–233 DNA methyaltion ..................................173–174, 191–200 Drug metabolism....................................100–101, 222, 228 Drug metabolizing enzymes and transporters (DMET) assay ................................................... 119

E Embryoid bodies ....................................................... 45–60 Embryonic stem cells................................................. 45–60 Endogenous control .................................................. 90–91 Epigenomics .................................................................. 192 Exon splicing ............................................................. 70–71 5¢-Exonuclease assay........................................................ 88 Expression profiling.......................... 27, 46, 69, 73–86, 192

I

L Lectin.. ...........................................269, 275–276, 280–281 Leukemia....................................................................... 192 Ligase detection reaction (LDR) ........................... 141–156 LM-PCR ...................................................................... 192

M Melanoma ................................................................. 27–43 b-Mercaptoethanol..................6–7, 16, 19, 29, 47, 186, 194

283

MICROARRAY METHODS FOR DRUG DISCOVERY 284 Index Metabolic enzyme ................................................. 221–236 MetaChip ...............................................222–223, 227–233 Metalloprotease ..................................................... 203–218 Microarrays.................................... 1–25, 27, 32, 42, 45–60, 64–69, 71, 73–74, 83, 87–89, 99–123, 125–139, 142, 144–145, 148, 160–162, 167–168, 174, 177–178, 182, 185–186, 192, 197–198, 203–218, 221–236, 239–249, 253, 255, 262, 266, 269–282 microRNA ................................................................. 73–86 Molecular inversion probe (MIP) ...........105, 108–111, 122 Mononuclear cells .............................. 6–7, 9, 15–18, 23–24 Multicenter clinical study .................................................. 2 Multiplexed detection.................................................... 142 Multiplex polymerase chain reaction (mPCR) ............... 95, 102, 105–109, 121, 142, 155–156

Ribominus ......................................................... 65–66, 170 Ribonomics.................................................................... 161 Ribonucleases (RNases)............................................. 28, 41 RIP-Chip .............................................................. 159–170 RNA-binding protein (RBP) .................160–167, 169–170 RNaseAlert ................................................................ 29, 39 RNaseZap ......................................................28–29, 34, 39 RNA viruses .......................................................... 144–145

S

NanoDrop ................................... 13, 15, 17, 19–20, 23–24, 29, 33, 36–38, 43, 64, 66, 71, 128–129, 161, 166 Nucleic acid extraction .................................................. 149

Selectivity profiling................................................ 251–266 shRNA ............................................................................ 64 Single nucleotide polymorphisms (SNPs) .................. 9, 47, 101, 105, 109–110, 117–120, 193 Slide printing .................................................147, 213–214, 218, 227, 255, 271, 273, 275–276 Small molecule arrays .....................................252, 256, 259 Specificity profiling ............................................... 251–266 Streptavidin magnetic beads ...................................... 13–14 Streptavidin-phycoerythrin ................................49, 69, 113

O

T

Oligonucleotide-based................................................... 126 Oropharyngeal squamous cell cancer............................. 240

TaqMan Array Card .................................................. 87–96 Tissue microarray .................................................. 239–249 Toxicity ...........................................................100, 221–236 Transcription factors .......................173–174, 178, 186–187 Transcriptome ....................................................65, 87, 160 Transfection, Transplantation.................................................1–5, 8–9, 23

N

P PAXgene Blood RNA tube ..........................6–8, 11–15, 25 PCR amplification...........................................90, 102, 105, 142, 144, 146, 149, 155, 182–183, 187 Personalized medicine ............................................. 99–123 Pharmacogenetics ............................................ 99–100, 119 Pharmacogenomics............................................ 91, 99–123 Poly-A polymerase (PAP) ................................. 75–77, 271 Post-transcriptional ....................................................... 159 Protein extraction ................................................ 6–7, 9–24 Protein microarray ................................................. 251–266

U Uracil DNA glycolase (UDG) ......................68, 75, 79, 167

W Whole blood......................... 3, 6–9, 11, 13–15, 21–23, 127 Whole transcript (WT)................................................. 159

Q

Z

QPCR. ..........................32, 42, 66, 87–90, 93–95, 243, 247 immunohistochemistry .................................................. 239

Zip-codes oligonucleotides............. 142, 144, 147, 151–155

R Radiofrequency...................................................... 206, 209 Reverse transcription ..............................68, 71, 92–95, 145