Natural killer cell protocols: cellular and molecular methods (volume 612)

ME T H O D S

IN

MO L E C U L A R BI O L O G Y

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

TM

Natural Killer Cell Protocols Cellular and Molecular Methods Second Edition

Edited by

Kerry S. Campbell Fox Chase Cancer Center, Philadelphia, PA, USA

Editor Kerry S. Campbell Fox Chase Cancer Center Institute for Cancer Research 333 Cottman Avenue Philadelphia PA 19111-2497 USA [email protected]

ISSN 1064-3745 e-ISSN 1940-6029 ISBN 978-1-60761-361-9 e-ISBN 978-1-60761-362-6 DOI 10.1007/978-1-60761-362-6 Library of Congress Control Number: 2009939929 © 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 springer.com

Preface It is becoming increasingly clear that natural killer (NK) cells are critical sentinels of the innate immune response. NK cells play important roles in protecting the body from numerous pathogens and cancer in addition to contributing to normal pregnancy and impacting the outcomes of transplantation. They have the unique capacity to detect and immediately respond to abnormal cells in the body without prior exposure. NK cell responses include the classical tumor cytolytic activity for which they were named and for the production of a number of cytokines that directly contribute to or potentiate the immune response. Although efficient tolerance mechanisms appear to prevent NK cells from causing autoimmune diseases, a great deal of genetic evidence suggests that match or mismatch of certain NK cell regulatory receptors [namely killer cell Ig-like receptors (KIR)] and their MHC class I ligands can influence a wide variety of human pathological conditions, including altering outcomes of viral infections, transplantations, pregnancies, and tumor therapies. The second edition of Natural Killer Cell Protocols provides a broad collection of some of the most important methods currently being used to study NK cells both in vitro and in vivo. The authors are international leaders in the field, who are directly using these methods to advance our understanding of this fascinating subset of lymphocytes. While the first edition provided a valuable collection of classical cellular and in vivo techniques to study NK cell functions, the chapters in the second edition focus on more recently developed methods, more refined techniques, and protocols designed to study NK cells within specialized tissue sites. These include protocols to analyze the various stages of NK cell development/maturation, to assess NK cell interactions with target cells and dendritic cells, to evaluate signal transduction by NK cell receptors, and to define KIR expression profiles by genotyping or flow cytometry. Additional chapters describe methods for the study of unique subsets of NK cells within the uterus during pregnancy and at intestinal mucosal surfaces, as well as techniques to evaluate NK cell responses to viral infections and malaria. The collection also includes specialized techniques to identify ligands for NK cell receptors, to define promoters regulating human KIR expression, to map receptors encoded within the murine NK cell gene locus that are responsible for resistance to pathogens, and to introduce cDNAs and shRNAs into NK cells using recombinant retrovirus or lentivirus. Finally, the book’s appendix provides a rich resource summarizing available reagents to study NK cells, cross-referencing KIR nomenclature, and detailing the many HLA ligands for various KIR family members. NK cells in rodents and humans are regulated by very different repertoires of receptors. Therefore, protocols from both perspectives are provided in this volume, with the species noted in the title or abstract. Furthermore, we have emphasized the “Notes” sections, which provide important details within each protocol, thereby extending the longstanding tradition of the Methods in Molecular Biology series. NK cells play unique roles in the immune response, but despite several decades of study, there is still much to learn about their functions, maturation, and regulation. The goal of Natural Killer Cell Protocols is to provide open access to important techniques

v

vi

Preface

written by key researchers in the field. I am indebted to the authors who have contributed their time and energies to provide high quality protocols. Their commitment to providing detailed descriptions of the methods was essential in making this project a success. I hope that this collection of methods will make significant contributions to your research and thereby further advance our collective understanding of these fascinating cells for many years to come. Kerry S. Campbell

Contents Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

v

Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

xi

1.

2.

Purification of Human NK Cell Developmental Intermediates from Lymph Nodes and Tonsils . . . . . . . . . . . . . . . . . . . . . . . . . . Aharon G. Freud and Michael A. Caligiuri

1

In Vitro Development of Human Killer–Immunoglobulin Receptor-Positive NK Cells . . . . . . . . . . . . . . . . . . . . . . . . . . . . Frank Cichocki and Jeffrey S. Miller

15

3.

Subset Analysis of Human and Mouse Mature NK Cells . . . . . . . . . . . . . Yoshihiro Hayakawa, Daniel M. Andrews, and Mark J. Smyth

27

4.

Assessing Licensing of NK Cells . . . . . . . . . . . . . . . . . . . . . . . . . . A. Helena Jonsson and Wayne M. Yokoyama

39

5.

Use of Stem Cell Radiation Chimeras to Analyze How Domains of Specific Proteins Impact on Murine NK Cell Development In Vivo . . . . . . Rebecca H. Lian and Vinay Kumar

51

6.

Use of Transfected Drosophila S2 Cells to Study NK Cell Activation . . . . . . . Michael E. March, Catharina C. Gross, and Eric O. Long

67

7.

Natural Killer Cell Conjugate Assay Using Two-Color Flow Cytometry . . . . . Deborah N. Burshtyn and Chelsea Davidson

89

8.

Studying NK Cell/Dendritic Cell Interactions . . . . . . . . . . . . . . . . . . Mathias Lucas, Cedric Vonarbourg, Peter Aichele, and Andreas Diefenbach

97

9.

Analysis of the NK Cell Immunological Synapse . . . . . . . . . . . . . . . . . 127 Keri B. Sanborn, Gregory D. Rak, Ashley N. Mentlik, Pinaki P. Banerjee, and Jordan S. Orange

10.

Measuring Intracellular Calcium Signaling in Murine NK Cells by Flow Cytometry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149 Alexander W. MacFarlane IV, James F. Oesterling, and Kerry S. Campbell

11.

Intracellular Staining for Analysis of the Expression and Phosphorylation of Signal Transducers and Activators of Transcription (STATs) in NK Cells . . . . 159 Takuya Miyagi, Seung-Hwan Lee, and Christine A. Biron

12.

A Model System for Studying NK Cell Receptor Signaling . . . . . . . . . . . . 177 Lukasz K. Chlewicki and Vinay Kumar

13.

Expression of cDNAs in Human Natural Killer Cell Lines by Retroviral Transduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 199 S. M. Shahjahan Miah and Kerry S. Campbell

vii

viii

Contents

14.

Lentiviral Gene Transduction in Human and Mouse NK Cell Lines . . . . . . . 209 Ram Savan, Tim Chan, and Howard A. Young

15.

Introduction of shRNAs into Human NK-Like Cell Lines with Retrovirus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 223 Amanda K. Purdy and Kerry S. Campbell

16.

Introduction of shRNAs into Primary NK Cells with Lentivirus . . . . . . . . . 233 Sam K.P. Kung

17.

Methods to Identify and Characterize Different NK Cell Receptors and Their Ligands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 249 Dikla Lankry, Roi Gazit, and Ofer Mandelboim

18.

Generating NK Cell Receptor-Fc Chimera Proteins from 293T Cells and Considerations of Appropriate Glycosylation . . . . . . . . . . . . . . . . . 275 Alon Zilka, Michal Mendelson, Benyamin Rosental, Oren Hershkovitz, and Angel Porgador

19.

Identification of NK Cell Receptor Ligands Using a Signaling Reporter System . 285 Yoshie-Matsubayashi Iizuka, Nikunj V. Somia, and Koho Iizuka

20.

Determining Ligand Specificity of Ly49 Receptors . . . . . . . . . . . . . . . . 299 Kerry J. Lavender and Kevin P. Kane

21.

Probing the Interactions of NK Cell Receptors with Ligand Expressed in trans and cis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 313 Jonathan Back, L´eonardo Scarpellino, and Werner Held

22.

A Simple Method to Measure NK Cell Cytotoxicity In Vivo . . . . . . . . . . . 325 Aurore Saudemont, Shannon Burke, and Francesco Colucci

23.

Functional Analysis of Human NK Cells by Flow Cytometry . . . . . . . . . . . 335 Yenan T. Bryceson, Cyril Fauriat, Jo˜ ao M. Nunes, Stephanie M. Wood, Niklas K. Bj¨orkstr¨om, Eric O. Long, and Hans-Gustaf Ljunggren

24.

Analysis of the KIR Repertoire in Human NK Cells by Flow Cytometry . . . . . 353 Niklas K. Bj¨orkstr¨om, Cyril Fauriat, Yenan T. Bryceson, Johan K. Sandberg, Hans-Gustaf Ljunggren, and Karl-Johan Malmberg

25.

KIR Genotyping by Multiplex PCR-SSP . . . . . . . . . . . . . . . . . . . . . 365 Smita Kulkarni, Maureen P. Martin, and Mary Carrington

26.

Identification and Analysis of Novel Transcripts and Promoters in the Human Killer Cell Immunoglobulin-like Receptor (KIR) Genes Hongchuan Li, Paul W. Wright, and Stephen K. Anderson

. . . . . 377

27.

Use of Inbred Mouse Strains to Map Recognition Receptors of MCMV Infected Cells in the NK Cell Gene Locus . . . . . . . . . . . . . . . . . . . . 393 Nassima Fodil-Cornu, Michal Pyzik, and Silvia M. Vidal

28.

Studying NK Cell Responses to Ectromelia Virus Infections in Mice . . . . . . . 411 Min Fang and Luis Sigal

Contents

ix

29.

Activation of Human NK Cells by Malaria-Infected Red Blood Cells . . . . . . . 429 Amir Horowitz and Eleanor M. Riley

30.

Natural Killer Cells in Human Pregnancy . . . . . . . . . . . . . . . . . . . . . 447 Victoria Male, Anita Trundley, Lucy Gardner, Jacquie Northfield, Chiwen Chang, Richard Apps, and Ashley Moffett

31.

Analysis of Uterine Natural Killer Cells in Mice . . . . . . . . . . . . . . . . . . 465 B. Anne Croy, Jianhong Zhang, Chandrakant Tayade, Francesco Colucci, Hakim Yadi, and Aureo T. Yamada

32.

Isolation of NK Cells and NK-Like Cells from the Intestinal Lamina Propria . . . 505 Stephanie L. Sanos and Andreas Diefenbach

Appendix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 519 Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 545

Contributors PETER AICHELE • Institute of Medical Microbiology & Hygiene, University of Freiburg Medical Center, Freiburg, Germany STEPHEN K. ANDERSON • Cancer and Inflammation Program, Laboratory of Experimental Immunology, SAIC-Frederick Inc., National Cancer Institute-Frederick, Frederick, MD, USA DANIEL M. ANDREWS • Peter MacCallum Cancer Centre, Parkville, Victoria, Australia RICHARD APPS • Department of Pathology, University of Cambridge, Cambridge, UK JONATHAN BACK • Ludwig Institute for Cancer Research Ltd., Lausanne Branch, and University of Lausanne, Epalinges, Switzerland PINAKI P. BANERJEE • Division of Immunology, Children’s Hospital of Philadelphia, Philadelphia, PA, USA CHRISTINE A. BIRON • Department of Molecular Microbiology and Immunology, Division of Biology and Medicine, Brown University, Providence, RI, USA NIKLAS K. BJo¨ RKSTRo¨ M • Center for Infectious Medicine, Department of Medicine, Karolinska Institutet, Karolinska University Hospital Huddinge, Stockholm, Sweden YENAN T. BRYCESON • Center for Infectious Medicine, Department of Medicine, Karolinska Institutet, Karolinska University Hospital Huddinge, Stockholm, Sweden SHANNON BURKE • Lymphocyte Signalling and Development, The Babraham Institute, Cambridge, UK DEBORAH N. BURSHTYN • Department of Medical Microbiology and Immunology, University of Alberta, Edmonton, AB, Canada MICHAEL A. CALIGIURI • The Comprehensive Cancer Center and The James Cancer Hospital & Solove Research Institute, The Ohio State University, Columbus, OH, USA KERRY S. CAMPBELL • Institute for Cancer Research, Fox Chase Cancer Center, Philadelphia, PA, USA MARY CARRINGTON • Cancer and Inflammation Program, Laboratory of Experimental Immunology, SAIC-Frederick Inc., National Cancer Institute-Frederick, Frederick, MD, USA TIM CHAN • Cancer and Inflammation Program, Laboratory of Experimental Immunology, National Cancer Institute-Frederick, Frederick, MD, USA CHIWEN CHANG • Department of Pathology, University of Cambridge, Cambridge, UK LUKASZ K. CHLEWICKI • Department of Pathology, The University of Chicago, Chicago, IL, USA FRANK CICHOCKI • University of Minnesota Cancer Center, Minneapolis, MN, USA FRANCESCO COLUCCI • Lymphocyte Signalling and Development, The Babraham Institute, Cambridge, UK B. ANNE CROY • Department of Anatomy and Cell Biology, Queen’s University, Kingston, ON, Canada CHELSEA DAVIDSON • Department of Medical Microbiology and Immunology, University of Alberta, Edmonton, AB, Canada

xi

xii

Contributors

ANDREAS DIEFENBACH • Institute of Medical Microbiology & Hygiene, University of Freiburg Medical Center, Freiburg, Germany MIN FANG • Institute for Cancer Research, Fox Chase Cancer Center, Philadelphia, PA, USA CYRIL FAURIAT • Center for Infectious Medicine, Department of Medicine, Karolinska Institutet, Karolinska University Hospital Huddinge, Stockholm, Sweden NASSIMA FODIL-CORNU • Department of Human Genetics, McGill University, Montreal, QC, Canada AHARON G. FREUD • Department of Pathology, Stanford University, Stanford, CA, USA LUCY GARDNER • Department of Pathology, University of Cambridge, Cambridge, UK ROI GAZIT • The Hebrew University – Hadassah Medical School, Jerusalem, Israel CATHARINA C. GROSS • Laboratory of Immunogenetics, National Institute of Allergy and Infectious Diseases, Rockville, MD, USA YOSHIHIRO HAYAKAWA • The University of Tokyo, Bunkyo-Ku, Tokyo, Japan WERNER HELD • Ludwig Institute for Cancer Research Ltd., Lausanne Branch, and University of Lausanne, Epalinges, Switzerland OREN HERSHKOVITZ • Department of Microbiology and Immunology, Ben Gurion University, Beer Sheva, Israel AMIR HOROWITZ • Department of Infectious and Tropical Diseases, London School of Hygiene and Tropical Medicine, London, UK KOHO IIZUKA • Department of Medicine and Center for Immunology, Masonic Cancer Center, University of Minnesota, Minneapolis, MN, USA YOSHIE-MATSUBAYASHI IIZUKA • Department of Medicine and Center for Immunology, Masonic Cancer Center, University of Minnesota, Minneapolis, MN, USA A. HELENA JONSSON • Medical Scientist Training Program, Rheumatology Division, Departments of Medicine, Pathology and Immunology, Washington University School of Medicine, St. Louis, MO, USA KEVIN P. KANE • Department of Medical Microbiology and Immunology, University of Alberta, Edmonton, AB, Canada SMITA KULKARNI • Cancer and Inflammation Program, Laboratory of Experimental Immunology, SAIC-Frederick Inc., National Cancer Institute-Frederick, Frederick, MD, USA VINAY KUMAR • Department of Pathology, University of Chicago, Chicago, IL, USA SAM K.P. KUNG • Laboratory of Innate Immunobiology, Department of Immunology, University of Manitoba, Winnipeg, MB, Canada DIKLA LANKRY • The Hebrew University – Hadassah Medical School, Jerusalem, Israel KERRY J. LAVENDER • The Edward Jenner Institute for Vaccine Research, University of Oxford, Oxford, UK SEUNG-HWAN LEE • Department of Molecular Microbiology and Immunology, Division of Biology and Medicine, Brown University, Providence, RI, USA HONGCHUAN LI • Cancer and Inflammation Program, Laboratory of Experimental Immunology, SAIC-Frederick Inc., National Cancer Institute-Frederick, Frederick, MD, USA REBECCA H. LIAN • Department of Pathology, University of Chicago, Chicago, IL, USA HANS-GUSTAF LJUNGGREN • Center for Infectious Medicine, Department of Medicine, Karolinska Institutet, Karolinska University Hospital Huddinge, Stockholm, Sweden ERIC O. LONG • Laboratory of Immunogenetics, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Rockville, MD, USA

Contributors

xiii

MATHIAS LUCAS • Institute of Medical Microbiology & Hygiene, University of Freiburg Medical Center, Freiburg, Germany ALEXANDER W. MACFARLANE IV • Institute for Cancer Research, Fox Chase Cancer Center, Philadelphia, PA, USA VICTORIA MALE • Department of Pathology, University of Cambridge, Cambridge, UK KARL-JOHAN MALMBERG • Center for Infectious Medicine, Department of Medicine, Karolinska Institutet, Karolinska University Hospital Huddinge, Stockholm, Sweden OFER MANDELBOIM • The Hebrew University – Hadassah Medical School, Jerusalem, Israel MICHAEL E. MARCH • Laboratory of Immunogenetics, National Institute of Allergy and Infectious Diseases, Rockville, MD, USA MAUREEN P. MARTIN • Cancer and Inflammation Program, Laboratory of Experimental Immunology, SAIC-Frederick Inc., National Cancer Institute-Frederick, Frederick, MD, USA MICHAL MENDELSON • Department of Microbiology and Immunology, Ben Gurion University, Beer Sheva, Israel ASHLEY N. MENTLIK • Cell Biology and Physiology Graduate Group, Division of Immunology, Children’s Hospital of Philadelphia, University of Pennsylvania School of Medicine, Philadelphia, PA, USA S. M. SHAHJAHAN MIAH • Institute for Cancer Research, Fox Chase Cancer Center, Philadelphia, PA, USA JEFFREY S. MILLER • University of Minnesota Cancer Center, Minneapolis, MN, USA TAKUYA MIYAGI • Department of Molecular Microbiology and Immunology, Division of Biology and Medicine, Brown University, Providence, RI, USA ASHLEY MOFFETT • Department of Pathology, University of Cambridge, Cambridge, UK JACQUIE NORTHFIELD • Department of Pathology, University of Cambridge, Cambridge, UK JO˜aO M. NUNES • Center for Infectious Medicine, Department of Medicine, Karolinska Institutet, Karolinska University Hospital Huddinge, Stockholm, Sweden JAMES F. OESTERLING • Institute for Cancer Research, Fox Chase Cancer Center, Philadelphia, PA, USA JORDAN S. ORANGE • The Joseph Stokes Jr. Research Institute, Division of Immunology, Children’s Hospital of Philadelphia, University of Pennsylvania School of Medicine and School of Veterinary Medicine, Philadelphia, PA, USA ANGEL PORGADOR • Department of Microbiology and Immunology, Ben Gurion University, Beer Sheva, Israel AMANDA K. PURDY • Institute for Cancer Research, Fox Chase Cancer Center, Philadelphia, PA, USA MICHAL PYZIK • Department of Human Genetics, McGill University, Montreal, QC, Canada GREGORY D. RAK • Cell Biology and Physiology Graduate Group, Division of Immunology, Children’s Hospital of Philadelphia, University of Pennsylvania School of Medicine and School of Veterinary Medicine, Philadelphia, PA, USA ELEANOR M. RILEY • Department of Infectious and Tropical Diseases, London School of Hygiene and Tropical Medicine, London, UK BENYAMIN ROSENTAL • Department of Microbiology and Immunology, Ben Gurion University, Beer Sheva, Israel

xiv

Contributors

KERI B. SANBORN • Immunology Graduate Group, Division of Immunology, Children’s Hospital of Philadelphia, University of Pennsylvania School of Medicine, Philadelphia, PA, USA JOHAN K. SANDBERG • Center for Infectious Medicine, Department of Medicine, Karolinska Institutet, Karolinska University Hospital Huddinge, Stockholm, Sweden STEPHANIE L. SANOS • Institute of Medical Microbiology & Hygiene, University of Freiburg Medical Center, Freiburg, Germany AURORE SAUDEMONT • Lymphocyte Signalling and Development, The Babraham Institute, Cambridge, UK RAM SAVAN • Cancer and Inflammation Program, Laboratory of Experimental Immunology, National Cancer Institute-Frederick, Frederick, MD, USA L´eONARDO SCARPELLINO • Ludwig Institute for Cancer Research Ltd., Lausanne Branch, and University of Lausanne, Epalinges, Switzerland LUIS SIGAL • Institute for Cancer Research, Fox Chase Cancer Center, Philadelphia, PA, USA MARK J. SMYTH • Peter MacCallum Cancer Centre, Parkville, Victoria, Australia NIKUNJ V. SOMIA • Department of Genetics, Cell Biology and Development, Masonic Cancer Center, University of Minnesota, Minneapolis, MN, USA CHANDRAKANT TAYADE • Department of Anatomy and Cell Biology, University of Guelph, Guelph, ON, Canada ANITA TRUNDLEY • Department of Pathology, University of Cambridge, Cambridge, UK SILVIA M. VIDAL • Department of Human Genetics, McGill University, Montreal, QC, Canada CEDRIC VONARBOURG • Institute of Medical Microbiology & Hygiene, University of Freiburg Medical Center, Freiburg, Germany STEPHANIE M. WOOD • Center for Infectious Medicine, Department of Medicine, Karolinska Institutet, Karolinska University Hospital Huddinge, Stockholm, Sweden PAUL W. WRIGHT • Cancer and Inflammation Program, Laboratory of Experimental Immunology, SAIC-Frederick Inc., National Cancer Institute-Frederick, Frederick, MD, USA HAKIM YADI • Lymphocyte Signalling and Development, The Babraham Institute, Cambridge, UK AUREO T. YAMADA • Laboratory of Histochemistry and Cytochemistry, Institute of Biology, UNICAMP, Campinas, Brazil WAYNE M. YOKOYAMA • Howard Hughes Medical Institute, Rheumatology Division, Departments of Medicine, Pathology and Immunology, Washington University School of Medicine, St. Louis, MO, USA HOWARD A. YOUNG • Cancer and Inflammation Program, Laboratory of Experimental Immunology, National Cancer Institute-Frederick, Frederick, MD, USA JIANHONG ZHANG • Department of Anatomy and Cell Biology, Queen’s University, Kingston, ON, Canada ALON ZILKA • Department of Microbiology and Immunology, Ben Gurion University, Beer Sheva, Israel

Chapter 1 Purification of Human NK Cell Developmental Intermediates from Lymph Nodes and Tonsils Aharon G. Freud and Michael A. Caligiuri Abstract Accumulating data indicate that human natural killer (NK) cells undergo terminal maturation in secondary lymphoid tissues (SLTs) including lymph nodes (LNs) and tonsils. In addition, recent studies have revealed that maturing NK cells progress through at least five functionally discrete stages of development within SLTs. These discoveries provide unique possibilities for researchers to investigate the natural processes governing human NK cell development, as they exist in vivo, through analysis of NK cell maturational intermediates found in situ. Herein we describe a detailed, yet simple, four-step protocol for the viable enrichment and purification of human NK cell developmental intermediates from LNs and tonsils. Key words: CD34, lymph node, tonsil, secondary lymphoid tissue, lymphopoiesis, NK development, stages of NK development, NK maturation, purification, isolation, sorting.

1. Introduction Like all other leukocyte populations, natural killer (NK) cells derive from pluripotent hematopoietic progenitor cells (HPCs) through a complicated process of development that involves both differentiation toward the NK cell lineage and maturation into functional competence (1). Traditionally, the process of human NK cell development has been studied almost exclusively in vitro, with researchers culturing purified CD34(+) HPCs in medium plus cytokines in order to derive functionally competent NK cells as well as their immediate precursors. Although such culture systems have provided invaluable insight into numerous aspects of K.S. Campbell (ed.), Natural Killer Cell Protocols, Methods in Molecular Biology 612, DOI 10.1007/978-1-60761-362-6 1, © Humana Press, a part of Springer Science+Business Media, LLC 2010

1

2

Freud and Caligiuri

human NK cell development, they are nonetheless limited by the very fact that it is essentially impossible to wholly recapitulate in vitro what naturally occurs in vivo. Recent discoveries from our laboratory and from others indicate that human NK cells undergo terminal maturation in secondary lymphoid tissues (SLTs) including lymph nodes (LNs) and tonsils (2, 3). Moreover, it is now clear that maturing NK cells progress through at least five functionally discrete stages of development within SLTs: stage 1 CD34(+)CD117(−)CD94(−)CD16(−), stage 2 CD34(+) CD117(+)CD94(−)CD16(−), stage 3 CD34(−)CD117(+) CD94(−)CD16(−), stage 4 CD34(−)CD117(+/−)CD94(+)CD 16(−), and stage 5 CD34(−)CD117(−)CD94(+/−)CD16(+) (4, 5). These findings provide unique possibilities for researchers to investigate the natural processes governing human NK cell development as they exist in vivo, through analysis of NK cell maturational intermediates found in situ. The process of isolating human NK cell developmental intermediates is relatively straightforward, and we have developed a simple, four−step protocol for the viable enrichment and purification of these cells directly from LNs and tonsils using commercially available reagents. The first step in this protocol involves the generation of SLT mononuclear cell (SLTMC) single-cell suspensions from gross tissue. Subsequently, the NK-lineage cells, which are all, by definition, CD3(−)CD19(−), are greatly enriched from SLTMCs via the depletion of total CD3(+) T cells and CD19(+) B-lineage cells that generally constitute >90% of the SLTMCs. Following the depletion step, stage 1 and stage 2 CD34(+) populations are separated from stage 3–5 cells via positive selection of CD34 HPCs. Lastly, each of the stage 1–5 cells may be sorted to purity, and we provide a useful staining procedure and sorting scheme for this purification.

2. Materials 2.1. Generation of SLTMC Single-Cell Suspensions

1. Dulbecco’s phosphate buffered saline (PBS) 1× (without calcium chloride or magnesium chloride) at room temperature (Invitrogen). 2. Ficoll-PaqueTM Plus density gradient at room temperature (GE Healthcare BioSciences). 3. 100− × 20 mm-style polystyrene cell culture dish (Corning Inc.). 4. 70−␮m cell strainers (BD Biosciences).

Purification of Human NK Precursors

3

5. Small surgical scissors and forceps. 6. 3-ml syringe (Becton Dickinson). 7. 50-ml centrifuge tubes (Greiner Bio-One). 2.2. Depletion of T and B Cells from SLTMCs

1. Midi selection magnet(s) with stand (Miltenyi Biotech). 2. LD depletion columns (Miltenyi Biotech). 3. Anti-human CD3 microbeads (Miltenyi Biotech). 4. Anti-human CD19 microbeads (Miltenyi Biotech). 5. De-gassed PBS+: PBS plus 0.5% fetal bovine serum (FBS) plus 2 mM EDTA (Invitrogen). Keep at 4◦ C or on ice (see Note 1). 6. 50-ml centrifuge tubes. 7. 15-ml centrifuge tubes (Greiner Bio-One). 8. 70-␮m cell strainers.

2.3. Positive Enrichment of CD34(+) HPCs

1. Mini selection magnet with stand (Miltenyi Biotech). 2. MS selection columns packaged with plungers (Miltenyi Biotech). 3. PBS+. 4. CD34 progenitor isolation kit including Fc-blocking antibody (Ab), anti-CD34 hapten Ab, and anti-hapten microbeads (Miltenyi Biotech). 5. 5-mL round-bottom tubes with caps (Becton Dickinson) (see Note 2).

2.4. Immunofluorescent Staining for Viable Cell Sorting

1. PBS+. 2. Culture medium consisting of RPMI-1640 + Glutamax plus 10% FBS and antibiotics (Invitrogen). 3. Direct fluorochrome-conjugated antibodies: CD45RA fluorescein isothiocyanate (FITC) (clone HI100), CD94 FITC (clone HP-3D9), CD117 phycoerythrin (PE) (clone 104D2), CD34 allophycocyanin (APC) (clone 581), CD3 APC (clone UCHT1), CD16 APC-Cy7 (clone 3G8) (BD Biosciences). 4. 5-mL round-bottom tubes with caps.

3. Methods It is important to appreciate the fact that human NK cell developmental intermediates, while relatively enriched among total HPCs and NK-lineage cells in SLTs compared to the blood or bone

4

Freud and Caligiuri

marrow compartments, nonetheless collectively constitute only a very minor fraction of the total cellularity of tonsils and LNs (4). Indeed these tissues contain not only a predominance of T and B cells but also numerous other cell types, including mast cells, dendritic cells, myeloid cells, endothelial cells, and nonhematopoietic stroma, which normally reside in SLTs. Therefore, although tonsil specimens may provide an average of one billion total mononuclear cells (with LNs providing far less depending upon size), it is common to obtain only a few thousand stage 1 and stage 2 cells and less than 105 stage 3–5 cells each after each isolation. Such low cell yields may preclude some molecular studies. 3.1. Generation of SLTMC Single-Cell Suspensions

1. The bottom and the lid of the Petri dish are separated from each other and each is filled with 20 mL room temperature PBS. The tonsil or the LN is placed into the bottom of the dish and cut into roughly 1 cm3 pieces using the forceps and scissors. Note that as soon as cuts are made into the tissue, cells can be seen diffusing into the surrounding PBS. If the tissue is encroached by fat (as is often the case with LN specimens), the latter should be removed prior to cutting and discarded as much as possible to avoid losing cells sticking to the fat. 2. A 70-␮m cell strainer is placed into the PBS within the lid of the Petri dish, and the first piece of lymphoid tissue (LT) is then cut as finely as possible over the strainer (being careful not to cut the strainer with the scissors). Using the flat top of the 3-ml syringe plunger, the tissue is gently, yet firmly, mashed against the bottom of the strainer in circular motions so that individual cells pass through into the PBS, whereas fibroadipose pieces of tissue remain within the strainer. The strainer is then discarded (see Note 3). 3. Step 2 is repeated with three more pieces of LT using a new cell strainer for each. 4. The 20mL of PBS-containing strained cells (from a total of four LT pieces) is carefully mixed within the lid with a pipette and then evenly distributed across three 50-ml tubes. 5. The lid of the Petri dish is refilled with 20 mL fresh PBS and Steps 2–4 are repeated until all LT pieces have been mashed and strained. The PBS contained within the bottom of the dish (where the numerous pieces of LT were kept in queue) is then strained and evenly distributed across the three 50ml tubes. All LT cells are now in suspension and divided evenly across three 50-ml tubes.

Purification of Human NK Precursors

5

6. The cells in each 50-ml tube are split across two tubes (thus, there will be six tubes altogether). The volume within each tube is then brought up to 40 mL total with room temperature PBS. 7. Fifteen milliliters of room temperature Ficoll is added to 12 separate 50-ml tubes (see Note 4). Twenty milliliters of cells is then gently and slowly overlaid onto the Ficoll in each of the 12 tubes (see Note 5). 8. The 12 tubes containing cells and Ficoll are centrifuged for 30 min at room temperature, 740 × g with the brake OFF (see Note 6). 9. The mononuclear layers are removed and each is added to a new 50-ml tube (12 tubes altogether). 10. The cells in each tube are washed of the Ficoll by dilution to a total volume of 50 mL with PBS followed by centrifugation for 8 min at 15◦ C, 385 × g with the brake on. 11. The supernatants are aspirated and the cell pellets are combined into one of the 50-ml tubes with a total of 10 mL (see Note 7). The 11 other tubes are then serially washed with 30 mL PBS that is finally added to the 10 mL of combined cells so that all of the SLTMCs are in suspension within a total volume of 40 mL. 12. The cells are thoroughly mixed and a small aliquot is removed for enumeration on a hemocytometer (see Note 8). The SLTMCs can be placed on ice while counting. 3.2. Depletion of T and B Cells from SLTMCs

1. CD3/CD19 depletion involves a one-step staining process followed by depletion across LD columns. 2. The SLTMCs are centrifuged for 8 min at 15◦ C, 385 × g. 3. Following centrifugation, the supernatant is discarded and the cells are resuspended and stained as follows within the 50-ml tube: For every one million cells in the pellet, add 8 ␮L PBS+ plus 1 ␮L CD3 microbeads plus 1 ␮L CD19 microbeads. Thus, the cells will be in a total volume of 10 ␮L/106 cells (see Note 9). 4. The cells are incubated in a 4◦ C refrigerator or cold room (rather than on ice, as per manufacturer’s instructions) for at least 15 min. For total volumes greater than 10 mL, the 50-ml tube should be placed on a nutator mixer or some similar device within a cold room to keep the cells thoroughly mixed in suspension with the microbeads during the incubation period. For samples less than 10 mL, it is sufficient to gently mix the cells via brief vortex or swirling

6

Freud and Caligiuri

the tube in a circular motion by hand for 15 s halfway through the incubation period. 5. During the incubation period, LD columns are placed into Midi magnets as per the manufacturer’s instructions, and the columns are equilibrated with 3 mL PBS+ each(to calculate the total number of LD columns needed for the entire sample, see Note 10). It usually takes 10–12 min for 3 mL to completely flow through each LD column. The LD columns are ready for use once all PBS+ is gone from the reservoir above the magnetic particles within the column and droplets have ceased to form below the column. The magnets and columns should be set up so that a 15-ml centrifuge tube can be placed underneath each column in order to catch the flow-through negative fraction. Every part of this step should be done at room temperature inside a tissue-culture hood to maintain sterility. 6. After the incubation of the cells with the microbeads, it is necessary to wash the cells to remove any unbound microbeads. The cells are diluted with additional PBS+ to a total volume of 50 mL within the 50-ml tube, and then they are centrifuged for 8 min at 15◦ C, 385 × g. The supernatant is discarded and the cells are resuspended in 50 mL PBS+ for a repeat wash step with centrifugation as described above. 7. Following the second wash, the cells are resuspended in ice cold PBS+ for depletion over the equilibrated LD columns. The total volume for depletion is 3 mL multiplied by the total number of LD columns. For example, for 10 LD columns (one billion cells in total), the cells should be resuspended in 30 mL PBS+. Prior to adding the cells to the columns, it is beneficial to filter out any clumps that may have formed during the incubation period. This is done by passing the cells over a 70-␮m cell strainer into a new 50-ml tube. In this case, to avoid losing cells leftover in the old tube, the cells can first be resuspended in a volume less than the final volume needed (for example, 20 mL rather than 30 mL as in the case above), then passed over the cell strainer, and finally the old tube can be washed with 10 mL PBS+ and passed over the same cell strainer to give a total volume of 30 mL (3 mL per LD column). 8. Once the cells are resuspended in the appropriate volume of PBS+, the cells are mixed thoroughly and distributed across the depletion columns. If there are more cells than columns available, the cells may be placed on ice until the next round of depletion using new LD columns.

Purification of Human NK Precursors

7

9. After the cells have completely passed through the columns (i.e., no cells are left in the LD column reservoirs), the 50-ml tube (if empty) is washed with cold PBS+ (3 mL per LD column) and then 3 mL is added per column. This is the first wash, and the 50-ml tube can now be discarded. It is important to not let the columns go dry, so care should be taken to ensure that the wash steps occur very soon after the cells pass through the columns. 10. A second wash of 3 mL PBS+ per LD column is added to each column after the first wash is complete. 11. After the second wash goes through and no more droplets fall from the bottom of the columns, the 15-ml tubes (now containing approximately 11 mL of cells in PBS+) are removed from the magnet apparatus, capped, and centrifuged for 8 min at 15◦ C, 385 × g. 12. During centrifugation, the LD columns can be discarded unless the T and/or B cells within the columns are to be used. In this case, the columns are removed from the magnets and repeatedly plunged with 5 mL PBS+ (four plunges altogether) into a collection tube. 13. After centrifugation of the 15-ml tubes containing Tand B-cell-depleted SLTMCs (Section 3.2, Step 11), the supernatants are removed and the cells are combined into one 15-ml tube. A small aliquot is then taken for enumeration on a hemocytometer and the rest of the cells are placed on ice (see Note 11). 3.3. Positive Enrichment of CD34(+) HPCs

1. The CD34 enrichment involves a two-step staining process followed by positive selection over MS columns. 2. The T-/B-cell-depleted SLTMCs in the 15-ml tube (Section 3.2, Step 13) are centrifuged for 8 min at 15◦ C, 385 × g. The supernatant is discarded, and the cell pellet is resuspended within the 15-ml tube in 2 ␮L PBS+, 1.5 ␮L Fc-blocking Ab, and 1.5 ␮L anti-CD34 hapten Ab, each per million cells in that order. If the total cell number is less than 10 million cells, they should be stained as if there were 10 million (i.e., not less than 20 ␮L PBS+ plus 15 ␮L each antibody in total). 3. The cells are incubated at 4◦ C in a refrigerator for 15 min, with gentle mixing by swirling the tube for 15 s halfway through the incubation. 4. The cells are then washed by adding 5 mL PBS+ followed by centrifugation for 8 min at 15◦ C, 385 × g. 5. After the spin, the supernatant is discarded and the cells are resuspended in 3.5 ␮L PBS+ plus 1.5 ␮L anti-hapten microbeads per million cells in that order.

8

Freud and Caligiuri

6. The cells are again incubated for 15 min at 4◦ C in a refrigerator, as per Section 3.3, Step 3. 7. During this second incubation, one MS column per sample is equilibrated by placing the column on the Mini magnet and adding 1 mL PBS+ to the reservoir. A sterile 5-mL tube is placed directly underneath the column in order to catch the flow-through. This is all done inside a sterile tissueculture hood. 8. Following incubation, the cells are again washed as in Section 3.3, Step 4. 9. After discarding the supernatant, the cells are resuspended in 1 mL PBS+ and placed into the reservoir of the equilibrated MS column that is on the Mini magnet (the 15ml tube is saved for the next step). Because the CD34(+) HPCs are labeled with microbeads, stage 1 and 2 NK cell developmental intermediates are retained within the column (while it is on the magnet), whereas stage 3–5 cells flow through the column and are collected in the 5-mL tube. 10. Once all the fluid has passed through the column, the bottom of the 15-ml tube is rinsed with 0.5 mL PBS+ that is then added to the reservoir above the column. 11. After the PBS+ rinse has passed through the column (ensuring first that drops have ceased to form under the column tip), the 5-mL tube with the CD34(−) fraction is set aside and replaced with a new sterile 5-mL tube, labeled “NK presort.” 12. The CD34(−) cell fraction in the first 5-mL tube is mixed by gentle pipetting and then passed over the column for a second round in order to retrieve any residual CD34(+) cells that may have initially passed through the column. We have observed this to significantly increase overall yields of CD34(+) HPCs. 13. After the cells have passed through the column, the first 5-mL tube is rinsed with 1 mL PBS+ that is then passed over the column as a wash step. 14. After the last of the drops have fallen from beneath the column, the “NK presort” tube is capped and set aside on ice. The MS column is removed from the magnet and plunged with 1 mL PBS+ into a new 5-mL tube, labeled “34 presort.” For best results, the PBS+ should be forced through the column immediately after it has been added to the column. The plunger is removed from the reservoir of the column and another 1 mL PBS+ is plunged through the column into the “34 presort” tube. This is again repeated

Purification of Human NK Precursors

9

twice for a total of four plunges. These plunges should rapidly follow one after the other, and it is recommended that if working with more than one sample, the first column should be plunged all four times before moving to subsequent columns. 15. At this step there are two tubes per sample (“CD34 presort” and “NK presort”), each with ∼4 mL total. The “34 presort” and “NK presort” tubes are then centrifuged for 8 min at 15◦ C, 385 × g. 16. Following the spin, the supernatants are carefully aspirated, leaving ∼100 ␮L liquid. The cell pellets are resuspended in this residual volume, and the tubes can be placed on ice until proceeding with the immunofluorescent staining protocol. 3.4. Immunofluorescent Staining for Viable Cell Sorting

1. Eight new 5-mL tubes are labeled as follows: “34 unstained,” “34 FITC,” “34 PE,” “NK unstained,” “NK FITC,” “NK PE,” “NK APC,” and “NK APC-Cy7.” These

Table 1.1 Antibodies and volumes for immunofluorescent staining prior to sorting Fluorochrome (volume added) Tube label

FITC

PE

APC

APC-Cy7

34 Unstained

–

–

–

–

34 FITC

CD45RA FITC (2.5 ␮L)

–

–

–

34 PE

–

CD117 PE (5 ␮L)

–

–

34 Presort

CD45RA FITC (2.5 ␮L)

CD117 PE (5 ␮L)

CD34 APC (2.5 ␮L)

–

NK unstained

–

–

–

–

NK FITC

CD94 FITC (10 ␮L)

–

–

–

NK PE

–

–

–

NK APC

–

CD117 PE (5 ␮L) –

CD3, CD34 APC (2.5 ␮L each)

–

NK APC-Cy7

–

–

–

CD16 APC-Cy7 (2.5 ␮L)

NK presort∗

CD94 FITC

CD117 PE

CD3, CD34 APC

CD16 APC-Cy7

∗ The volume of each antibody added to the “NK presort” tube depends upon the total cell number (for simplicity, the total cell number from Section 3.2, Step 13 can be used). For every 2 million cells, add 10 ␮L CD94 FITC, 5 ␮L CD117 PE, 2.5 ␮L CD16 APC-Cy7, 2.5 ␮L CD3 APC, and 2.5 ␮L CD34 APC.

10

Freud and Caligiuri

Fig. 1.1. Representative gating scheme for sorting NK cell developmental intermediates from human SLTs. (A) Enriched tonsil CD34(+) cells were stained as indicated in Table 1.1 and analyzed on a FACSVantage flow cytometer. The dot plots are gated as indicated with the following regions: R1, live cells with size and granularity characteristics typical of progenitor cells; R2, CD45RA(+) events; R3, CD34(+)CD117(−) events; R4, CD34(+)CD117(+) events. (B) Tonsil CD3(−)CD19(−)CD34(−) cells were stained as indicated in Table 1.1 and analyzed on a FACSVantage. The dot plots and histogram are gated as indicated with the following regions: R1, live cells within the typical “lymphocyte” gate (small size, low granularity); R2, CD3(−)CD34(−) events; R3, CD117(+)CD94(−) events; R4, CD117(dim/−)CD94(+) events; R5, CD16(−) events; R6, CD16(+) events.

are the control tubes for setting the voltage, amplitude, and compensation parameters on the cell sorter (see Note 12). 2. One hundred microliters of ice cold PBS+ are added to each of the control tubes. 3. Three microliters of cells from the “34 presort” and “NK presort” tubes are then added to each of the corresponding control tubes listed in Section 3.4, Step 1. 4. The cells in each of the tubes are stained with fluorochromelabeled Abs as shown in Table 1.1. We provide the volumes of antibodies typically used in our laboratory, but it is rec-

Purification of Human NK Precursors

11

ommended that these Abs be titrated for optimal staining prior to use as there may be variations from lot to lot. After adding the appropriate volume of each of the Abs, the samples are incubated on ice and covered from light for at least 15 min. 5. Following incubation, the cells are washed with 2 mL PBS+ and then centrifuged for 8 min at 15◦ C, 385 × g.

3.5. Gating Scheme for Sorting Stages 1–5

6. The supernatants are discarded, and the cell pellets are resuspended in the following volumes of culture medium: 300 ␮L for each of the control tubes; 1 mL for the “34 presort” tube; and 3 mL for the “NK presort” tube (see Note 13). The cells are kept on ice and covered from light until they are sorted. 1. Figure 1.1 shows dot plots from a sort for human NK cell developmental intermediates using the staining protocol outlined in Section 3.4. 2. Using the staining protocol outlined herein, stage 1–5 cells are sorted as follows: a. Stage 1 cells: R1 + R2 + R3 (Fig. 1.1A) b. Stage 2 cells: R1 + R2 + R4 (Fig. 1.1A) c. Stage 3 cells: R1 + R2 + R3 (Fig. 1.1B) d. Stage 4 cells: R1 + R2 + R4 + R5 (Fig. 1.1B) e. Stage 5 cells: R1 + R2 + R4 + R6 (Fig. 1.1B) (see Note 14)

4. Notes 1. The SLTMCs are kept in PBS+ solution for most of the protocol following the Ficoll centrifugation step during which the PBS and Ficoll are used at room temperature (see Note 6). For better viability and less chance that the cells will become activated through the procedure, the PBS+ should be kept cold even though many steps of the protocol, aside from incubations and centrifugations, are performed in a sterile hood at room temperature. In addition, to prevent microbubbles from clogging the magnetic columns, the manufacturers recommend that the PBS+ be “de-gassed” prior to use. This can be done by filtering the freshly made PBS+ through a 0.22−␮m vacuum filter (e.g., Steriflip products by Millipore) and leaving the vacuum attached for 30 min after the liquid has been filtered. 2. Both polystyrene and polypropylene 5-mL round-bottom tubes are commercially available. Either can be used for this

12

Freud and Caligiuri

protocol, but compatibility with the cell sorter may depend on the make and model of the latter. Researchers should determine which kind of tube is to be used at their institution. 3. Most of the LTs consist of hematopoietic cells that will go into suspension. Therefore, a good rule of thumb is to continue gently grinding the tonsil or the LN piece against the strainer until all that is left is white, fibroadipose tissue. 4. In our experience, the majority of tonsils provide between 0.5 and 2 × 109 SLTMCs in total. For this number of cells, 12 Ficoll tubes is an appropriate number so as not to overwhelm the Ficoll and have potentially inefficient separation of the mononuclear cells from the fat, debris, endothelial cells, and other non-mononuclear cells. However, for small LN specimens or very large tonsil specimens, individuals may choose to use fewer or more Ficoll tubes, respectively. 5. One must take care to pipette very slowly when initially overlaying the cells upon the Ficoll so as not to mix the cells with the former prior to centrifugation. For those with experience in overlaying blood upon Ficoll, note that blood is much thicker and, thus, less likely to mix with the Ficoll during the overlay compared to LT cells in suspension in PBS. A helpful technique is to angle the 50-ml Ficoll tube at 45◦ and to disperse the cells near the top of the tube very slowly so that the cells run down the side and gently lay over the Ficoll. 6. It is very important to leave the centrifuge break off so that the mononuclear cell layers are not disrupted by an abrupt stop. In addition, the centrifuge should be set at room temperature in order to increase the probability of obtaining very distinct mononuclear layers above the Ficoll. In our experience, centrifugation at cold temperatures results in suboptimal results with hazy mononuclear layers. This is why room temperature PBS and Ficoll are used prior to the centrifugation step. 7. For this step, 10 mL PBS is used to resuspend the cell pellet in the first tube. This cell suspension is then used to resuspend the next cell pellet and so on until all cell pellets have been resuspended and combined in a total of 10 mL within one 50-ml tube. 8. For counting the cells, we generally start with a 40-fold dilution of the cells by first removing 10 ␮L from the mixed cell sample, diluting this 1:10 with PBS, and then diluting the cells again 1:4 with trypan blue. This prevents an excess waste of trypan blue, which is also toxic to cells at high exposure.

Purification of Human NK Precursors

13

9. With so many cells in one 50-ml tube, the cell pellet is very large. To ensure proper resuspension, it is recommended to first use a 5-mL pipette (smaller volume pipettes have smaller diameter tips through which the cells are forced) with 3–5 mL PBS+ and to repeatedly spray the cells against the side of the tube multiple times until there are no visible clumps. Subsequently, additional PBS+ can be added up to the desired volume. This procedure is recommended in general for resuspending large cell pellets. 10. As per the manufacturer’s instructions, each LD column can efficiently deplete up to a maximum of 1 × 108 cells in total. In our laboratory, to determine the number of LD columns to use, we round the total number of cells to the nearest 108 and divide by 108 . For example, for a sample with 9.3 × 108 cells, nine LD columns are used. Note: Because one Midi magnet can hold only one LD column, it is beneficial to purchase multiple Midi magnets and stands so that cells can be passed over the LD columns in parallel rather than in series. This is very important in terms of saving time, because one round of depletion over an LD column, including the initial cell suspension followed by two sequential washes, takes roughly 45 min. With 10 Midi magnets, one can deplete a total of one billion SLTMCs in 45 min, whereas it would take 7.5 h with the use of only one magnet for the same total number of SLTMCs. 11. It is common for the cell yield following T- and B-cell depletion to be approximately 1% of the initial SLTMC count. As such, the cell pellets are much smaller (1–2 mm in diameter) after the depletion step. Nonetheless, the NKlineage cells are greatly enriched within this cell mixture. 12. This staining protocol is applicable for sorting human NK cell developmental intermediates using a FACSVantage or a FACSAria cell sorter from BD Biosciences. Alternative staining protocols may be required if using different sorting machines. 13. In our practice, we prefer to put the cells in culture medium (on ice) prior to the sort in the event that there may be unforeseen delays in sorting the cells. If it is anticipated that the cells will be on ice only for a short amount of time prior to sorting, it is fine to put the cells in PBS+ rather than medium. In addition, we have found that it is preferable to dilute the cells as much as reasonably possible so that there is less chance of cell clumping that could clog the sorter. 14. The minimal CD3(−)CD16(+) phenotype is not sufficient to identify stage 5 NK cells, because non-NK-lineage cells,

14

Freud and Caligiuri

including monocytes and granulocytes, can express CD16 within SLTs. In contrast, the CD3(−)CD94(+)CD16(+) phenotype does unequivocally identify CD56(+) NK cells in these tissues (data not shown). Given that the vast majority of stage 5 cells express CD94, for practical purposes we sort stage 5 cells as CD3(−)CD34(−)CD94(+)CD16(+) and forgo the rare stage 5 cells with background level fluorescent intensity staining of CD94.

Acknowledgments The authors would like to thank Tamra Brooks for her assistance with the chapter. Funding for the Caligiuri laboratory is provided by the National Cancer Institute (P30 CA16059, CA68458 and CA95426) (M.A.C.). References 1. Colucci F., Caligiuri M. A., Di Santo J. P. (2003) What does it take to make a natural killer? Nat Rev Immunol 3, 413–25. 2. Ferlazzo G., Thomas D., Lin S. L., Goodman K., Morandi B., Muller W. A., Moretta A., M¨unz C. (2004) The abundant NK cells in human secondary lymphoid tissues require activation to express killer cell Ig-like receptors and become cytolytic J Immunol 172, 1455–62. 3. Freud A. G., Becknell B., Roychowdhury S., Mao H. C., Ferketich A. K., Nuovo G. J., Hughes T. L., Marburger T. B., Sung

J., Baiocchi R. A., Guimond M., Caligiuri M. A. (2005) A human CD34(+) subset resides in lymph nodes and differentiates into CD56bright natural killer cells Immunity 22, 295–304. 4. Freud A. G., Caligiuri M. A. (2006) Human natural killer cell development Immunol Rev 214, 56–72. 5. Freud A. G., Yokohama A., Becknell B., Lee M. T., Mao H. C., Ferketich A. K., Caligiuri M. A. (2006) Evidence for discrete stages of human natural killer cell differentiation in vivo J Exp Med 203, 1033–43.

Chapter 2 In Vitro Development of Human Killer–Immunoglobulin Receptor-Positive NK Cells Frank Cichocki and Jeffrey S. Miller Abstract The in vitro culture system outlined in this chapter allows for the delineation of events that occur during the development of CD34+ hematopoietic precursor cells into mature KIR+ human NK cells. This system can also be utilized to study the effects of gene overexpression or knockdown on the process of NK cell differentiation through retroviral transduction and long-term culture. The necessary soluble factors and contact-dependent conditions for in vitro human NK cell development have been worked out in our laboratory over the past 16 years. Key words: Human NK cell development, lymphocyte differentiation from hematopoietic precursors

1. Introduction Both mouse and human natural killer cells recognize transformed and virally infected cells and influence the direction of the adaptive immune response in infectious settings (1). However, there are several notable differences between mouse and human NK cells with respect to differentiation markers, making developmental comparisons between the species difficult (2–4). To evaluate and manipulate human natural killer cell differentiation, a robust ex vivo cell culture system is necessary. To this end, our laboratory has developed a long-term culture system for studying human NK cell development from primitive progenitor cells (5–9). In 1992, we were the first to show that primitive progenitors from adult bone marrow can give rise to functional NK cells K.S. Campbell (ed.), Natural Killer Cell Protocols, Methods in Molecular Biology 612, DOI 10.1007/978-1-60761-362-6 2, © Humana Press, a part of Springer Science+Business Media, LLC 2010

15

16

Cichocki and Miller

when cultured in contact with human bone marrow stroma (5). In 1994, we found that in the absence of factors known to support NK cell differentiation, direct contact with human allogeneic stroma is critical for NK cell differentiation (6). In 1998, after an explosion of new data describing novel NK cell receptors recognizing class I MHC, we found that our in vitro model of NK cell differentiation supported the acquisition of KIR (specifically KIR3DL1 in this initial description) (7). Limitations of these initial studies included the heterogeneity and variability of primary human stroma, low cloning frequency when plating primitive stem cell populations, and the inability to support differentiation at the single cell level. These issues raise the possibility of starting progenitor contamination. Therefore, definitive study of NK cell precursors was technically difficult. In 1999, we pioneered the use of a novel murine stromal cell line, called AFT024, which was derived from day 14 gestational fetal liver cells immortalized by a retrovirus containing a temperature-sensitive SV40 T antigen. The most important finding for the success of NK cell development cultures is that AFT024 could officially support NK cell differentiation of human cells at the single cell level. Using this system, we have shown that KIR and NKG2/CD94 receptors are acquired late in NK cell development (8). In 2008, we further improved this system by comparing NK cell development on a novel murine cell line called EL08-1D2 cloned from a culture of embryonic liver at day 11 of gestation. EL08-1D2 was chosen for these studies because of its ability to support generation of human hematopoietic progenitors from CD34+ umbilical cord blood cells without the addition of any cytokines. Use of EL08-1D2 identified several novel properties of this stromal feeder. First, when IL-15 was eliminated from the culture medium, there was an accumulation of CD56− NK cell precursors defined as CD34+ /CD7− , CD34+ /CD7+ , and CD34− /CD7+ . The role of IL-3 and IL-3 plus Flt3 ligand was established, and c-kit ligand and IL-7 appear to add efficiency to the system but are not absolutely required for NK cell differentiation. EL08-1D2 was superior to AFT024 for supporting differentiation of NK cell precursors, NK cell commitment, the acquisition of KIR, and overall proliferation (9). EL08-1D2 has been the stromal feeder of choice in our laboratory based on its ability to recapitulate the acquisition of class I recognizing receptors and developmental intermediates which may be important in NK cell maturation (10). There has been support in the literature for NK cell differentiation cultures in the absence of stroma. In review of this literature, our stromal-based cultures seemed to better allow the acquisition of NK cell receptors where NK cell differentiation in the absence of stroma but in the presence of high concentrations of human cytokines allows NK cell commitment with poor expression of

In Vitro Development of Human Killer–Immunoglobulin Receptor-Positive

17

KIR (11). Our data support the notion that unique signals from stroma are important in the acquisition of NK cell receptors. Further proof is the finding that CD56bright KIR− cells can transition into CD56dim KIR+ cells on EL08-1D2 and human IL-15 (12). In summary, methods for NK cell differentiation cultures using EL08-1D2 will be described here. Other stromal cell lines, such as OP9 and MS5, have been used in the literature for similar purposes and could possibly be substituted here but direct comparisons with EL08-1D2 have not been performed in our laboratory (13, 14). CD34+ hematopoietic progenitors from umbilical cord blood are highly efficient in NK cell development in terms of both cloning efficiency and proliferation. Other hematopoietic stem cell sources such as bone marrow, peripheral blood progenitors, and fetal liver have been tested in our laboratory and can be substituted here if desired, but their efficiency may differ from the bulk of our work using human umbilical cord blood progenitors.

2. Materials This protocol includes the experimental procedures that our laboratory has developed to retrovirally transduce CD34+ cells prior to long-term culture in order to study the effects of gene overexpression or knockdown. If the user does not desire to carry out retroviral transduction, Section 2.3 and steps 24–42 in Section 3.3 can be excluded from the protocol. 2.1. EL08-1D2 Cell Culture and Irradiation

1. EL08-1D2 cells: mouse embryonic liver cells. These cells were obtained from E. Dzierzak at Erasmus University MC, Rotterdam, the Netherlands. 2. EL08-1D2 medium: 202.5 ml ␣-MEM medium (Gibco, R M5300 medium (StemCarlsbad, CA), 250 ml Myelocult Cell Technologies Inc., Vancouver, BC, Canada), 37.5 ml fetal calf serum (see Note 1), 5 ml Penicillin + Streptomycin, 5 ml 100× Glutamax (Gibco), 35.0 ␮l ␤-mercaptoethanol (0.143 M stock: 1:100 in H2 O), and hydrocortisone at a final concentration of 10−6 M. Hydrocortisone must be added fresh at the time of use. This medium (without hydrocortisone) can be stored at 4◦ C for up to 1 month. Note that these additives should be included on top of what is already R M5300 medium. present in the Myelocult 3. Ultrapure water with 0.1% gelatin (Chemicon Intl., Billerica, MA). 4. Trypsin/EDTA (Gibco).

18

Cichocki and Miller

5. Flat-bottom 96-well cell culture plates (Becton Dickinson, Franklin Lakes, NJ). 6. Multi-channel pipetter. 7. Cesium irradiator. 2.2. CD34+ Cell Isolation from Umbilical Cord Blood

R 1. Histopaque -1077 (Sigma-Aldrich, St. Louis, MO).

2. Ammonium chloride solution (StemCell Technologies Inc.). R (Beckman Coulter, Fullerton, 3. Coulter Particle Counter CA).

4. PBS/0.3% BSA. 5. Direct CD34 Progenitor Cell Isolation Kit, human (Miltenyi Biotech, Oberlin, CA). R LS separation columns (Miltenyi Biotech). 6. MACS R pre-separation filters (Miltenyi Biotech). 7. MACS R magnetic stand and magnets (Miltenyi Biotech). 8. MACS

9. Cycling Medium: Iscove’s Modified Dulbecco’s Medium (with L-glutamine), 20% fetal bovine serum, 1% Penicillin + Streptomycin with 20 ng/ml IL-7, 20 ng/ml c-kit ligand, 20 ng/ml Flt3 ligand, and 20 ng/ml thrombopoietin. This medium can be stored at 4◦ C for up to 1 month. 10. 24-well cell culture plates (Becton Dickinson). 2.3. Retroviral Transduction and Cell Sorting of CD34+ Cells

1. Four separate 3 ml aliquots of retroviral supernatant (see Note 2). 2. Transwells with sterile 6-well culture plates − 0.4 ␮M PTFE membrane, 24 mm insert (Corning Incorporated, Corning, NY). 3. Tweezers. 4. RetroNectin (r-Fibronectin) (Otsu, Shiga, Japan). 5. PBS/2% BSA. 6. Iscove’s Modified Dulbecco’s Medium (with L-glutamine) without fetal bovine serum (Gibco). 7. Cycling medium (see Section 2.2). 8. APC-conjugated anti-CD34 monoclonal antibody (clone 8G12, mouse IgG1 ; BD Biosciences) 9. APC-conjugated IgG1 isotype control monoclonal antibody (BD Bioscience). 10. 5 ml polystyrene round-bottom tubes with cell-strainer cap (BD Biosciences). 11. A fluorescence-activated cell sorter.

In Vitro Development of Human Killer–Immunoglobulin Receptor-Positive

2.4. Long-Term Culture of Sorted CD34+ Cells on the EL08-D12 Stromal Line

19

1. Basal Culture Medium: A 2:1 (vol:vol) mix of Dulbecco’s Modification of Eagle’s Medium (DMEM) with 4.5 g/l glucose, L-glutamine, and sodium pyruvate/Ham’s F12 Medium. This medium mixture is then supplemented with 24 ␮M 2-mercaptoethanol, 50 ␮M ethanolamine, 20 mg/l ascorbic acid, 50 ␮g/l sodium selenite, 1% penicillin + streptomycin and 20% heat-inactivated human AB serum (Valley Biomedical, Inc., Winchester, VA). The following cytokines must also be added: 10 ng/ml IL-15, 5 ng/ml IL3, 20 ng/ml IL-7, 20 ng/ml c-kit ligand, and 10 ng/ml Flt3 ligand. This medium can be stored at 4◦ C for up to 1 month. 2. Multi-channel pipetter.

3. Methods 3.1. General Culture Conditions for EL08-1D2 Cells

1. Coat sterile cell culture flasks (75 cm2 or 150 cm2 ) with enough sterile ultrapure water with 0.1% gelatin to cover the bottom of the flask. Let the flask sit in the culture hood at room temperature for 10 min. The gelatin water is necessary to promote adherence of the EL08-1D2 cells. 2. After 10 min, remove the ultrapure water with 0.1% gelatin and gently rinse the bottom of the culture flask with PBS. Remove the PBS. 3. Plate cells at approximately 4000 cells/cm2 on gelatincoated flasks in an appropriate volume of EL08-1D2 medium (12 ml for a 75 cm2 flask or 24 ml for a 150 cm2 flask) containing 20% conditioned EL08-1D2 medium (see Note 3). 4. Culture cells in an incubator set to 32◦ C 5% CO2 . This is important, as the cells grow in a temperature-sensitive manner. 5. Once the cells are 95–99% confluent, they can be split into new flasks. 6. To split cells, remove the spent medium, filter, and store at −20◦ C. Add enough 1× trypsin/EDTA to cover the bottom of the flask and incubate at room temperature until the cells start to lift off the bottom of the flask. Add 5 ml of 0.2 ␮m-filtered EL08-1D2 conditioned supernatant to neutralize the trypsin and collect cells in a 15 ml centrifuge tube. 7. Count cells using a hemocytometer. Remove the desired number of cells (approximately 4000 cells/cm2 ) to seed a new flask(s).

20

Cichocki and Miller

8. Centrifuge at 550 × g for 4 min and decant supernatant. 9. Resuspend cells in freshly prepared EL medium and gently transfer to a flask(s) freshly coated with gelatin water. 10. Return flask(s) to a 32◦ C 5% CO2 incubator. 3.2. Irradiation of EL08-1D2 Cells

1. Use a multi-channel pipette to add 50 ␮l of ultrapure water with 0.1% gelatin per well into the desired number of flatbottom 96-well cell culture plates. Let plates incubate for 10 min at room temperature. 2. After 10 min, remove the ultrapure water with 0.1% gelatin and gently rinse the bottom of the culture plates with PBS. Remove the PBS. 3. Resuspend EL08-1D2 cells in fresh EL08-1D2 medium (with 20% conditioned medium and hydrocortisone) at a concentration of 104 cells/ml. 4. Add 103 cells per well (100 ␮l) to the gelatin-coated and rinsed plates. Incubate plates at 32◦ C 5% CO2 for 3–5 days until the plates are 95–99% confluent. 5. Irradiate confluent plates with 30 Gy of radiation and let the plates incubate at 32◦ C 5% CO2 for at least 2 h before use. Irradiated plates can be kept for up to 1 week before use in an NK cell developmental assay.

3.3. CD34+ Cell Isolation and Retroviral Transduction

Day 1 R -1077 and PBS to room temperature. 1. Warm Histopaque 2. Transfer umbilical cord blood into a sterile container and add 50 ml PBS to dilute the blood. R -1077 to the necessary number of 3. Add 20 ml Histopaque 50 ml centrifuge tubes. R 4. Slowly layer 30 ml diluted blood over the Histopaque 1077.

5. Centrifuge at 550 × g for 30 min at room temperature with the brake off. 6. Slowly aspirate the plasma layer leaving approximately 1/2 R inch of liquid above the lymphocyte/Histopaque -1077 interface. 7. Harvest the lymphocytes from the lymphocyte interface and transfer to a new 50 ml centrifuge tube. Fill up to 50 ml with PBS and centrifuge at 850 × g for 5 min. 8. Gently decant the supernatant and resuspend cells in 10 ml ice-cold ammonium chloride solution. Incubate the cells on ice for 10 min. 9. Centrifuge for 5 min at 850 × g. Gently decant supernatant and resuspend cells in 20 ml PBS.

In Vitro Development of Human Killer–Immunoglobulin Receptor-Positive

21

10. Determine total cell number using a Coulter Particle R (see Note 4). Counter 11. Wash the cells by adding another 30 ml of PBS to the cell suspension. Centrifuge for 5 min at 850 × g. 12. Gently decant the supernatant and resuspend cells in 300 ␮l PBS/0.3% BSA per 108 cells. Keep the PBS/0.3% BSA cold throughout the isolation procedure. 13. Label cells for isolation by adding 100 ␮l FcR blocking reagent for every 108 cells. Next, add 100 ␮l CD34 Microbeads for every 108 cells. Incubate for 30 min on ice in the dark. 14. Add 30 ml of PBS/0.3% BSA and centrifuge for 5 min at 850 × g. 15. Gently decant the supernatant and resuspend the cells in 0.5 ml of PBS/0.3% BSA. R LS column per cord blood donor on 16. Set up one MACS R R  pre-separation a MACS Magnetic Stand. Place a MACS filter on top of each column and equilibrate the column with 3 ml of PBS/0.3% BSA. R pre-separation 17. Pipette the labeled cells into the MACS filter and let the cells pass through the column. There is no need to save the effluent, as the CD34+ cells will be retained within the column.

18. Wash the column three times with PBS/0.3% BSA. R 19. Remove the MACS pre-separation filter, pull the column off the magnet, and transfer the column to a 15 ml centrifuge tube. Apply 5 ml of PBS/0.3% BSA and forcefully plunge the liquid through the column to purge the labeled cells. R LS column. It is 20. Repeat steps 16–19 with a new MACS necessary to double-column the labeled cells in order to obtain a CD34+ population with a high level of purity.

21. Count the total number of cells in the CD34+ fraction using a hemocytometer. 22. Centrifuge cells for 4 min at 550 × g at 4◦ C (see Note 5). 23. Gently decant the supernatant and resuspend the cell pellet in Cycling Medium at a concentration of 5 × 105 cells/ml. Transfer the cells in Cycling Medium to a 6-well or 24-well cell culture plate and place in an incubator set to 37◦ C 5% CO2 . If cells are resuspended in less than 2 ml of medium (1 × 106 cells or less), they should be added to a well of a 24-well plate. Larger volumes should be added to a 6-well plate. The number of cells isolated will vary widely between

22

Cichocki and Miller

umbilical cord blood donors. Let the cells proliferate in the Cycling Medium for 3 days (see Note 6). Day 4 24. Three days after the CD34+ cell isolation, reconstitute one vial of RetroNectin at a concentration of 0.05 mg/ml according to the manufacturer’s instructions. 25. Transfer two transwells per umbilical cord donor to a 6-well culture plate using a pair of sterilized tweezers. 26. Coat transwells in the 6-well plate with 0.1 mg of RetroNectin (2 ml of 0.05 mg/ml RetroNectin per well). Incubate plate for 2 h at room temperature. 27. Remove RetroNectin from both the top and the bottom of the transwells. 28. Wash by adding 2 ml of PBS/2% BSA to the transwell. Let the transwells sit for 1/2 h at room temperature (see Note 7). 29. Remove PBS/2% BSA from both the top and the bottom of the transwells. 30. Add 2 ml of Iscove’s Medium (without FBS) to each transwell and let the transwell sit for 1/2 h at room temperature. 31. Remove all Iscove’s Medium from the top and the bottom of the transwells. 32. Add 2 × 105 CD34+ cells (at a concentration of 4 × 105 cells per ml) to each transwell along with 3 ml of the desired viral supernatant. Do not let the virus sit at room temperature longer than necessary before adding it to the transwell. 33. Place plates in an incubator set to 37◦ C 5% CO2 for 6 h. 34. Remove the viral supernatant from the bottom of the transwells, taking care to disturb the cells on the top of the transwell as little as possible. 35. Place the transwell plate back into the incubator for 5 min to allow the rest of the viral supernatant to pass through the transwell. 36. Remove the remaining viral supernatant from the bottom of the transwells. 37. Slowly add 3 ml Cycling Medium to the top of each transwell and place the transwell plate back in the incubator overnight. Day 5 38. Remove the Cycling Medium from the bottom of the transwells and allow all of the medium to pass through. Remove the remaining medium from the bottom of the transwells. 39. Add 3 ml of the same type of viral supernatant used during the previous day for transduction to each transwell.

In Vitro Development of Human Killer–Immunoglobulin Receptor-Positive

23

40. Place plates in an incubator set to 37◦ C 5% CO2 for 6 h. 41. Repeat steps 34–37 in this section. Day 7 42. Harvest the cells from each transwell by pipetting up and down a few times and washing the top of the transwells with 1 ml of medium taken from the bottom of the transwells. Place the cells into a 15 ml centrifuge tube. 43. Remove 100 ␮l from each GFP control sample and put aside on ice for an isotype control. 44. Count cells using a hemocytometer. Generally, you should expect to observe a two- to threefold increase in cell number over the course of the retroviral transduction due to expansion in the Cycling Medium. 45. Centrifuge cells at 550 × g for 4 min. 46. Gently decant the supernatant and add 10 ␮l of anti-CD34 APC-conjugated monoclonal antibody per 1 × 106 cells. 47. Add 1.5 ␮l of APC-conjugated mouse IgG1 antibody to the 100 ␮l of cell set aside for the isotype control. 48. Incubate samples in the dark on ice for 30 min. 49. Wash off excess antibody by adding 5 ml of PBS/0.3% BSA and centrifuging at 550 × g for 4 min. 50. Gently decant supernatant and resuspend stained cells in approximately 0.3 ml of PBS/0.3% BSA. 51. Pass the labeled cells through the strainer cap and into a 5 ml FACS tube. 52. Collect the double-positive GFP+ /CD34+ population using a fluorescence-activated cell sorter (see Note 8 and Fig . 2.1A). 53. Resuspend sorted GFP+ CD34+ cells at a concentration of 500 cells per ml in Basal Culture Medium. 54. Remove the 96-well plates containing irradiated EL08D12 cells from the incubator and gently remove the EL08D12 medium using a multi-channel pipette. 55. Add 100 ␮l of sorted cells in Basal Culture Medium to each well containing irradiated EL08-D12 cells. This will result in a plating of 50 transduced CD34+ cells per well. 56. Place plates in an incubator set to 37◦ C and 5% CO2 . 3.4. Long-Term Culture of Sorted CD34+ Cells on the EL08-1D2 Stromal Line

1. After 7 days have passed, add 100 ␮l of Basal Culture Medium to each well and place the plates back into the incubator. 2. After 14 days have passed, carefully remove 150 ␮l of medium from each well without disturbing the cells on the bottom of the wells. Add 150 ␮l of Basal Culture Medium to

24

Cichocki and Miller

Fig. 2.1. (A) Representative phenotypes of retrovirally transduced CD34+ cells harvested at the time of sorting. A fairly conservative CD34+ /GFP+ gate should be selected to collect this population for long-term culture. The CD34− cells in this plot represent cells that have already begun to differentiate during the cycling phase of the protocol. These cells should not be collected. (B) Representative phenotypes of long-term cultures of CD34+ cells differentiated on EL08-1D2 stroma in the presence of 10 ng/ml IL-15, 5 ng/ml IL-3, 20 ng/ml IL-7, 20 ng/ml c-kit ligand, and 10 ng/ml Flt3 ligand. CD56 and KIR (mixture of DX9, EB6, GL183, and FES172 monoclonal anti-KIR antibodies conjugated with a common fluorochrome) expressions on cells harvested from day 14 and day 21 cultures are shown. As this figure illustrates, the major transition of hematopoietic cells into mature NK cells occurs between 2 and 3 weeks after plating.

each well. At this time point, a proportion of the cells in culture begin to express CD56, but do not express significant levels of KIR as determined by FACS analysis. 3. After 21 days, change the medium again by removing 150 ␮l of medium from each well and adding back 150 ␮l of Basal Culture Medium. At this time point, a significant number of cells in culture express both CD56 and KIR as determined by FACS analysis (see Note 9 and Fig. 2.1B). 4. Full NK cell maturation will be observed after 21 or 28 days in culture, and this is usually used as an endpoint for in vitro NK cell development experiments (see Note 10).

4. Notes 1. The fetal calf serum used for EL08-D12 cell culture is specially designed for the maintenance of stromal cell lines from mouse embryonic tissues and is distributed by StemCell Technologies, Inc. Other types of serum should not be substituted without comparative testing. 2. For this protocol, it is necessary to prepare ahead of time two 3 ml aliquots of retroviral supernatant prepared using a control GFP construct for each umbilical cord blood donor. You will also need two 3 ml aliquots of retroviral supernatant prepared using a construct containing GFP and your gene-of-interest for each umbilical cord donor.

In Vitro Development of Human Killer–Immunoglobulin Receptor-Positive

25

We have consistently used Murine Stem Cell Virus (MSCV) promoter constructs for gene overexpression in CD34+ cells. We package MSCV retroviral particles in the 293kj cell line using the PCL packaging plasmid as previously described (15). 3. The “conditioned” EL08-1D2 medium refers to the supernatant from EL08-1D2 culture flasks that are at or near confluency. This medium should be harvested, 0.2 ␮mfiltered, and frozen at −20◦ C for storage up to 6 months. Conditioned medium should be added to the EL Medium immediately before use at a 20% concentration. 4. Because of the high cell counts per unit of umbilical cord blood, counting with a hemacytometer will be less accurate. We also recommend that the user add 5–6 drops of Zap-oglobinTM II Lytic Reagent (Beckman Coulter) to the Isotone reagent before counting to obtain a more accurate lymphocyte count. 5. There is considerable variability in CD34+ cell yields between umbilical cord donors. This may be because of age from procurement to use or due to individual unit variability. Final cell counts generally fall between 3 × 105 cells and 1.5 × 106 cells. It is our experience that fresh umbilical cord blood units are more efficient in NK cell differentiation cultures than cryopreserved cells which work but with slightly less efficiency. 6. If gene modification with an EGFP marker is considered, it is important to setup the experiment so that you can sort the cells using a fluorescence-activated cell sorter 6 days after the CD34+ cell isolation. 7. Because of variability in the manufacturing of the transwells, the RetroNectin solution may or may not pass through individual transwells. This will not affect the experiment. However, if the RetroNectin does not pass through, we recommend adding 1 ml of PBS/0.3% BSA to the bottom of the transwell and 1 ml of PBS/0.3% BSA to the top of the transwell during the washing step to soak the transwell so that viral supernatant can pass through. 8. The percentage of GFP+ /CD34+ cells in culture depends on the size of the retroviral construct and the potency of the virus. We routinely observe that, when using conservative gating, 1–5% of cells are GFP+ /CD34+ at the time of the sort. 9. A reduction in the number of EL08-D12 cells will usually be observed after 21 days in culture due to their lysis by NK cells. The EL08-D12 stromal layer will disappear completely at later time points.

26

Cichocki and Miller

10. There is considerable variability between umbilical cord donors with respect to the rates of differentiation and proliferation in culture, which may be due to the intrinsic sensitivity of individual donors to the cytokines used in culture. Therefore, we recommend that at least four to six donors are analyzed and compared for each experiment. References 1. Colucci, F., Caligiuri, M.A., and Di Santo, J.P. (2003) What does it take to make a natural killer? Nat. Rev. Immunol. 3, 413–425. 2. Huntington, N.D., Vosshenrich, C.A., and Di Santo, J.P. (2007) Developmental pathways that generate natural-killer-cell diversity in mice and humans. Nat. Rev. Immunol. 7, 703–714. 3. Kim, S., Iizuka, K., Kang, H.P., Dokun, A., French, A.R., Greco, A., and Yokoyama, W.M. (2002) In vivo developmental stages in murine natural killer cell development. Nat. Immunol. 3, 523–528. 4. Freud, A.J., Yokohama, A., Becknell, A., Lee, M.T., Mao, H.C., Ferketich, A.K., and Caligiuri, M.A. (2006) Evidence for discrete stages of human natural killer cell differentiation in vivo. J. Exp. Med. 203, 1033–1043. 5. Miller, J.S., Verfaille, C., and McGlave, P. (1992) The generation of human natural killer cells from CD34+ /DR– primitive progenitors in long-term bone marrow culture. Blood. 80, 2182–2187. 6. Miller, J.S., Alley, K.A., and McGlave, P. (1994) Differentiation of natural killer (NK) cells from human primitive marrow progenitors in a stroma-based long-term culture system: identification of a CD34+ 7+ NK progenitor. Blood. 9, 2594–2601. 7. Miller, J.S., McCullar, V., and Verfaillie, C.M. (1998) Ex vivo culture of CD34+ /Lin− /DR− cells in stromaderived soluble factors, interleukin-3, and macrophage inflammatory protein-1␣ maintains not only myeloid but also lymphoid progenitors in a novel switch culture assay. Blood. 12, 4516–4522. 8. Miller, J.S., and McCullar, V. (2001) Human natural killer cells with polyclonal lectin and immunoglobulinlike receptors develop from single hematopoietic stem cells with preferential expression of NKG2A and KIR2DL2/L3/S2. Blood. 98, 705–713.

9. McCullare, V., Oostendorp, R., PanoskaltsisMortari, A., Yun, G., Lutz, C.T., Wagner, J.E., and Miller, J.S. (2008) Mouse fetal and embryonic liver cells differentiate human umbilical cord blood progenitors into CD56negative natural killer cell precursors in the absence of interleukin-15. Exp. Hematol. 36, 598–608. 10. Grzywacz, B., Kataria, N., Sikora, M., Oostendorp, R.A., Dzierzak, E.A., et al. (2006) Coordinated acquisition of inhibitory and activating receptors and functional properties by developing human natural killer cells. Blood. 108, 3824–3833. 11. Yu, H., Fehniger, T.A., Fuchshuber, P., Thiel, K.S., Vivier, E., et al. (1998) Flt3 ligand promotes the generation of a distinct CD34+ human natural killer cell progenitor that responds to interleukin-15. Blood. 92, 3647– 3657. 12. Cooley, S., Xiao, F., Pitt, M., Gleason, M., McCullar, V. et al. (2007) A subpopulation of human peripheral blood NK cells that lacks inhibitory receptors for self-MHC is developmentally immature. Blood. 110, 578–586. 13. Nakayama, N., Fang, I., and Elliott, G. (1998) Natural killer and B-lymphoid potential in CD34+ cells derived from embryonic stem cells differentiated in the presence of vascular endothelial growth factor. Blood. 1, 2283–2295. 14. Giarratana, M.C., Verge, V., Schmitt, C., Tertho, J.M., et al. (2000) Presence of primitive lymphoid progenitors with NK or B potential in ex vivo expanded bone marrow cell cultures. Exp. Hematol. 28, 46–54. 15. Chiorean, E.G., Dylla, S., Olsen, K., Lenvik, T., Soignier, Y., et al. (2003) BCR/ABL alters the function of NK cells and the acquisition of killer immunoglobulin-like receptors (KIR). Blood. 101, 3327–3533.

Chapter 3 Subset Analysis of Human and Mouse Mature NK Cells Yoshihiro Hayakawa, Daniel M. Andrews, and Mark J. Smyth Abstract Identification of natural killer (NK) cell subsets has gained attention with the recent discovery that mature mouse NK cells comprise two distinct stages. Delineation of the stages is performed using the markers CD27 and CD11b on gated NK cells. The significance of this finding is underpinned by recent discoveries that mature human NK cells can also be discriminated by differential expression of CD27. This chapter will describe the methods required for the purification of lymphocytes from blood and other organs and the delineation of NK cell subsets by flow cytometry. Key words: Flow cytometry, natural killer, murine, human, CD27, CD11b, CD56.

1. Introduction Peripheral NK cells mediate numerous functions including the cytotoxic control of tumors and virus-infected cells as well as the production of cytokines such as IFN-␥ and TNF-␣ (1). Principally, NK cells have been regarded as mediators of natural host immune resistance. More recently it has been demonstrated that NK cells can act to regulate the generation of the adaptive immune response. Clearly, a high level of complexity underlies mature NK cells in secondary lymphoid and peripheral organs. The first dissection of NK cells into subsets arose through the work of Lewis Lanier in 1983 (2). However, it was not until 1989 that a universal method of human NK cell subset identification was identified (3). It is now widely accepted that human mature NK cells are comprised of CD56dim and CD56high cells; however, this antigen is not expressed on mouse K.S. Campbell (ed.), Natural Killer Cell Protocols, Methods in Molecular Biology 612, DOI 10.1007/978-1-60761-362-6 3, © Humana Press, a part of Springer Science+Business Media, LLC 2010

27

28

Hayakawa, Andrews, and Smyth

NK cells, and thus the translation of biological information concerning NK cells in the mouse has been problematic. While the development of mouse NK cells from precursors has been quite widely studied (4), the identification of mature stages has been largely ignored. Developmentally, mouse NK cells were defined as mature once expression of the integrin CD11b (Mac1) was detected (5). Subsequently, work from our laboratory showed that NK cells could be further identified on the basis of TRAIL expression in adult livers (6). This underlying suggestion of heterogeneity within mature murine NK cells was ultimately demonstrated by delineating two stages of mature NK cells on the basis of CD27 and CD11b (7). Significantly, human NK cells have also been shown to differentially express CD27 (8, 9), making comparative interpretations of the subsets functionality more clear. In the mouse, CD161+ (NK1.1) CD3− (or TCR␣␤− ) NK cells can be divided into CD27lo CD11blo , CD27hi CD11blo , CD27hi CD11bhi , and CD27lo CD11bhi stages. The CD27lo CD11blo stage remains poorly characterized and may be the most immature, but differentiation has been shown to proceed from CD27hi CD11blo through CD27hi CD11bhi to CD27lo CD11bhi (7). The most mature CD27lo NK cell subset possesses a higher threshold to stimulation and appears to be tightly regulated by the expression of NK cell inhibitory receptors (Ly-49/NKG2A). Comparatively, the CD27hi NK cell subset displays a greater effector function, exhibits a distinct tissue distribution and responsiveness to chemokines, and interacts productively with dendritic cells. Importantly, we and others have verified that CD27hi and CD27lo subsets with distinct cell surface phenotypes also exist in human peripheral blood (8, 9). These findings clearly reclassify mature NK cells into distinct stages and begin to discern their specific role in immune responses. The protocol for enriching lymphocytes from specific organs and delineating mature NK cell subsets by flow cytometry is described below.

2. Materials 2.1. Human PBMC Enrichment

1. Ficoll-Paque PLUS (Amersham Biosciences, Piscataway, NJ, USA). Store at 4◦ C. 2. Phosphate-buffered saline (PBS): Prepare stocks at 3.2 mM Na2 HPO4 , 0.5 mM KH2 PO4 , 1.3 mM KCl, and 135 mM NaCl. Adjust to pH 7.4 with HCl if necessary. Store at room temperature. R . 13 × 75 mm, 4 ml. 3. BD Vacutainer

4. 15 ml centrifuge tubes (BD Falcon, Bedford, MA, USA).

Subset Analysis of Human and Mouse Mature NK Cells

2.2. Mouse Lymphocyte Enrichment

29

1. Roswell Park Memorial Institute (RPMI) 1640 medium (Invitrogen, Carlsbad, CA, USA) supplemented with 2% fetal bovine serum (FBS, Mt Wellington, Auckland, NZ). Store at 4◦ C. 2. Type IV collagenase (Worthington Biochemicals, Lakewood, NJ, USA) dissolved in PBS at 1 mg/ml. Store at 4◦ C. 3. Extraction buffer: RPMI-2% FBS supplemented with Type IV collagenase (10 ␮g/ml). 4. ACK lysis buffer: 8.29 g/L NH4 Cl, 1 g/L KHCO3 , 0.0367 g/L Na2 -EDTA. Adjust to pH 7.2–7.4 with HCl. Store at room temperature. 5. Percoll (Amersham Biosciences, Piscataway, NJ, USA). Store at 4◦ C. 6. 10× PBS: Prepare stocks at 32 mM Na2 HPO4 , 5 mM KH2 PO4 , 13 mM KCl, and 1.35 M NaCl. Store at room temperature. 7. 15 and 50 ml centrifuge tubes (BD Falcon, Bedford, MA, USA). 8. 40 ␮m cell strainers (BD Falcon, Bedford, MA, USA). 9. Spoon sieves (see Note 1).

2.3. Flow Cytometry Reagents

1. PBS 2% FBS: PBS supplemented with 2% fetal bovine serum (FBS, Mt Wellington, Auckland, NZ). Store at 4◦ C. Keep sterile. 2. Blocking buffer: Add at 2 ␮g/ml anti-Fc␥RII/Fc␥RIII monoclonal antibody (clone 2.4G2) to PBS 2% FBS. Prepare fresh each time. Store at 4◦ C. 3. 7AAD or Fluorogold (Sigma, St Louis, MO). Store at 4◦ C. Keep in darkness to prevent photobleaching. 4. 24-well tissue culture plates and 96-well U-bottom plates (BD Falcon). 5. 1.2 ml microtiter tubes (Molecular BioProducts, USA). 6. Monoclonal antibodies: Anti-human: CD161 (DX12, BD Pharmingen), CD3 (SK7, BD Pharmingen), CD27 (LG.3A10 or LG.7F9, eBioscience), CD56 (MEM188 or B159, BD Pharmingen). Anti-mouse: NK1.1 (PK136, eBioscience), CD3 (145-2C11, BD Pharmingen), CD27 (LG.7F9, eBioscience), CD11b (M1/70, eBioscience). Store at 4◦ C. Keep in darkness to prevent photobleaching.

30

Hayakawa, Andrews, and Smyth

3. Methods NK cells are lymphocytes and thus have some characteristics of T and B cells. In studying NK cells by flow cytometry it is critically important to exclude any populations that may share a common marker. For instance, NKT cells in the mouse also express NK1.1 (albeit at a lower level than T cells) making it possible to accidentally include this cell population during analysis. The following protocol will outline how to ensure analysis is restricted to NK cells only and will provide details for their enrichment from peripheral organs. For obvious reasons our understanding of mouse NK cells in lymphoid and peripheral organs is much better, thus the detection of human NK cells in this chapter will be restricted to blood. Although the process for isolation of lymphocytes from mouse organs is, for the most part, generic, several aspects require some attention and will be addressed in the notes section. 3.1. Enrichment of Human PBMC from Whole Blood

R 1. 2 ml of whole blood is drawn into a BD vacutainer and supplemented with 2 ml of PBS (see Note 2).

2. The blood and PBS should be mixed. 3. Invert the Ficoll-Paque to ensure that it is mixed. Remove 3 ml and add to a 15 ml centrifuge tube. 4. Overlay the diluted blood sample onto the Ficoll-Paque ensuring that the blood and Ficoll-Paque do not mix. 5. Centrifuge the cells at 400× g for 30 min at room temperature. 6. Aspirate the supernatant to leave the lymphocytes intact (see Note 3). 7. Transfer the lymphocytes to a clean centrifuge tube (see Note 4). Add at least three volumes of PBS to the lymphocytes. 8. Centrifuge at 300× g for 7 min at room temperature. 9. Aspirate supernatant and wash as per steps 7 and 8. 10. Aspirate supernatant and resuspend cells in PBS 2% FBS.

3.2. Isolation of Lymphocytes from Mouse Organs

1. Pre-warm extraction buffer (see Section 2.2). Add extraction buffer to 24-well plate appropriate to organ. For spleen and lung add 1 ml, for lymph nodes add 200 ␮l. Livers should not be extracted in the presence of collagenase (see Note 5). 2. Enrichment of lymphocytes from the liver requires perfusion of this organ. Expose the femurs (upper leg bones), viscera,

Subset Analysis of Human and Mouse Mature NK Cells

31

and chest of euthanized mice and cut the left ventricle of the heart. Expose the portal vein by moving the ventral liver lobe up toward the chest. Insert a 25G needle into the portal vein and perfuse the liver with 5–10 ml of PBS. Successful perfusion will result in the liver becoming beige in color. 3. Remove all organs and chop into small pieces (excluding the femurs) in extraction buffer (remembering not to use collagenase for the liver). Spleen, lung, and lymph nodes should be incubated at 37◦ C for 25 min with agitation. 4. For livers, chop into small pieces and wash through a metal spoon sieve by using the barrel of a 10 ml syringe to disrupt the liver across the sieve. Transfer cells into a 50 ml centrifuge tube. 5. The ends of the femurs should be cut with a scalpel. Using a 25 g syringe, aspirate the marrow with 5 ml of PBS. Aspirate through a 40 ␮m cell strainer inserted into a 50 ml tube to remove large chunks and wash through with 4 ml of PBS. 6. For spleen, lymph node, and lung transfer digested tissue and media into a 40 ␮m cell strainer inserted into a 50 ml tube. Wash wells of 24-well plate with 1 ml of PBS and add to tube containing organ. Disrupt organ through sieve using barrel of 10 ml syringe. Wash cells through strainer using 8 ml of PBS (2 × 4 ml at a time). 7. Centrifuge cells from all organs at 300× g for 7 min. 8. For the liver aspirate supernatant, add 10 ml of PBS and wash again as per step 7. 3.2.1. Preparation of Lymphocytes from Spleen, Lymph Node, and Bone Marrow

1. Once centrifuged, cells from the lymph node can be resuspended in blocking buffer (see Note 6). Maintain at 4◦ C. 2. Lyse red blood cells in bone marrow and spleen using ACK lysis buffer. Resuspend spleen or marrow cells in 1 ml of buffer for 2 min at room temperature. Stop lysis by addition of 4 ml of PBS and immediately centrifuge at 300× g for 7 min. 3. Resuspend cells in blocking buffer (see Note 6). Maintain at 4◦ C.

3.2.2. Preparation of Lymphocytes from Liver

1. Prepare a 37.5% Percoll solution. Adjust Percoll to 90% (mouse osmolarity) using 10× PBS. (i.e., add 5 ml of 10× PBS to 45 ml of Percoll). Dilute 90% Percoll to 37.5% using PBS (i.e., add 37.5 ml of 90% Percoll to 62.5 ml of PBS. Adjust volumes accordingly for scaling up) (see Note 7). 2. Resuspend aspirated liver preparations (Section 3.2, step 8) in 1 ml of 37.5% Percoll using a P1000 (see Note 8). Transfer cells from 50 ml centrifuge tube into a 15 ml centrifuge tube.

32

Hayakawa, Andrews, and Smyth

Top up to 15 ml with remaining Percoll and centrifuge at 690× g for 12 min at room temperature. 3. Following centrifugation lymphocytes will pellet, while hepatocytes will float to the top. Aspirate the hepatocytes and supernatant, leaving the lymphocytes intact. 4. Lyse red blood cells in ACK lysis buffer as per Section 3.2.1, step 2 (see Note 9). 5. Resuspend cells in blocking buffer (see Note 6). Maintain at 4◦ C.

3.2.3. Preparation of Lymphocytes from Lung

1. Prepare a 90% Percoll solution as described in Section 3.2.2, step 1. Use this solution to prepare 45% and 67.5% solutions (i.e., for the 45% solution add 4 ml of 90% Percoll to 4 ml of PBS and for the 67.5% solution add 6 ml of 90% Percoll to 2 ml of PBS. Adjust volumes accordingly for scaling up) (see Note 7). 2. Resuspend aspirated lung preparations in 1 ml of 67.5% Percoll using a P1000 (see Note 8). Transfer cells from 50 ml centrifuge tube into a 15 ml centrifuge tube. Add further 2 ml of 67.5% Percoll to make a final volume of 3 ml. Carefully overlay 2 ml of 45% Percoll onto the cell suspension. Avoid mixing of the 45% and 67.5% layer. 3. Centrifuge at 800× g for 15 min at room temperature. Ensure that brake and acceleration are turned to minimum. 4. After centrifugation, lymphocytes will form a band at the interface of the 45% and 67.5% gradients. Aspirate the 45% gradient, leaving the lymphocyte band intact. 5. Transfer the lymphocytes to a 15 ml centrifuge tube and add PBS to fill the tube. 6. Centrifuge at 300× g for 7 min. 7. Lyse red blood cells in ACK lysis buffer as per Section 3.2.1 part 2. 8. Resuspend cells in blocking buffer (see Note 6). Maintain at 4◦ C.

3.3. Preparation/ Blocking of Cells for Staining with Fluorescent Antibodies

1. Cells should be counted and aliquoted in blocking buffer at 106 cells per stain (see Note 10). 2. For multiple samples use a 96-well U-bottom plate. Add 106 cells per well and maintain at 4◦ C for at least 30 min to ensure appropriate blocking of the Fc receptors. 3. Centrifuge plate at 200× g for 4 min. and aspirate supernatant.

Subset Analysis of Human and Mouse Mature NK Cells

3.4. Staining of Cells with Fluorescent Antibodies to Detect NK Cells

33

1. Prepare a cocktail of antibodies comprising CD161, CD3, CD27, and CD56. Add an appropriate volume (see Note 11) per well and resuspend cells using a pipette. 2. Maintain cells at 4◦ C for at least 1 h.

3.4.1. Detecting Human NK Cell Subsets

3.4.2. Detecting Mouse NK Cell Subsets

1. Prepare a cocktail of antibodies comprising NK1.1, CD3, CD27, and CD11b. Add an appropriate volume (see Note 11) per well and resuspend cells using a pipette. 2. Maintain cells at 4◦ C for at least 1 h.

3.4.3. Labeling Dead Cells for Exclusion on the Flow Cytometer

1. Centrifuge labeled cells at 200× g for 4 min. 2. Aspirate supernatant and resuspend cells in 50–100 ␮l of either PBS 2% FBS with 2 ␮g/ml 7AAD or PBS 2% FBS with 2 ␮g/ml Fluorogold. Leave for 1 min at room temperature. FACS instruments with a 355 nm laser (such as the LSR-II) can use Fluorogold. Most bench top instruments (such as the CANTO) do not have this laser and must use the 488 nm laser to excite 7AAD. 3. Centrifuge cells at 200× g for 4 min. 4. Resuspend cells in 200–400 ␮l of PBS 2% FBS. Transfer to 1.2 ml microtiter tubes (see Note 12). Maintain in the dark and keep cool before acquiring samples on the flow cytometer.

3.5. Acquisition and Gating of Human and Mouse NK Cells by Flow Cytometry

1. Ensure that correct PMT voltages and compensation are applied to the cytometer. 2. In the first instance set an acquisition plot of forward scatter height (FSC-H) vs. forward scatter area (FSC-A) and place a gate around singlet cells (Fig. 3.1A). 3. Within the singlet population set an acquisition plot of FSCA vs. 7AAD or Fluorogold (consult with your FACS operator for correct channels on appropriate instruments). Set a gate around cells that have excluded the viability die. 4. On the live cells, set an acquisition plot of NK1.1 vs. CD3 (for mouse) or CD161 vs. CD3 (for human). In each instance set a gate around the NK1.1+ /CD3− or CD161+ /CD3− cells (Figs. 3.1 and 3.2). 5. On the NK cells set an acquisition plot of CD27 vs. CD11b (for mouse) or CD27 vs. CD56 (for human). Mouse NK cell subsets are identified as immature (CD27− /CD11b− , CD27+ /CD11b− ) and mature and can be subdivided into CD27+ /CD11b+ and CD27− /CD11b+ (Fig. 3.1).

34

Hayakawa, Andrews, and Smyth

Fig. 3.1. The gating procedure and use of fluorescent minus one staining (FMO) to detect mouse NK cell subsets. (A) Singlet cells (to exclude cell doublets, which can appear as cells expressing both fluorochromes) are gated on the 45◦ angle of a plot between FSCA and FSC-H prior to gating on live cells (represented here are fluorogold (FG)-negative cells). The live population is then shown with NK1.1 and CD3, NK cells are gated as NK1.1+ and CD3− . (B) In order to set the gates properly for the subsets FMO are used. The first panel shows gated NK cells (NK1.1+ /CD3− ) on which only CD11b is stained, while in the second panel only CD27 is stained. This allows the quadrant gates to be set appropriately on CD11b+ and CD27+ . The final panel in part B shows the four major subsets (CD27− CD11b− a, CD27+ /CD11b− b; CD27+ /CD11b+ c; CD27− /CD11b+ d). The quadrant gates have been set correctly by the use of the FMO setup tubes.

Human NK cell subsets are identified as CD56dim /CD27− , CD56hi /CD27− , and CD56hi /CD27+ (Fig. 3.2). See Note 13 for choice of fluorochromes and setting gates appropriately.

Fig. 3.2. The gating procedure and use of fluorescent minus one staining (FMO) to detect human NK cell subsets. Singlet cells (to exclude cell doublets, which can appear as cells on the line of identity between FSC-A and FSC-H) are gated on the 45◦ angle prior to gating on live cells (represented here are fluorogold-negative cells) as shown in Fig. 3.1. The live population is then shown in a plot of CD161 vs. CD3, and NK cells are gated as CD161+ and CD3− . In order to set the gates properly for the subsets, FMO are used. The second panel shows gated NK cells (CD161+ /CD3− ) on which only CD56 is stained, while in the third panel only CD27 is stained. This allows the gates to be set appropriately on CD56+ and CD27+ subsets. The final panel shows the three major subsets (CD56dim /CD27− ; CD56hi /CD27− ; CD56hi /CD27+ ). Box gates have been set correctly by the use of the FMO setup tubes. CD27hi /CD56hi cells are gated by aligning the vertical edges of the CD27 box with the CD56hi box.

Subset Analysis of Human and Mouse Mature NK Cells

35

6. Set the cytometer to acquire samples. For analysis of NK cell subsets it is appropriate to collect at least 2000 NK cell events from each organ (Fig. 3.3).

Fig. 3.3. Representative density plots showing mouse NK cells and their subsets in the spleen, liver, lung, bone marrow, and lymph node. (A) Identification of NK cells by using NK1.1 and CD3 on preparations from the spleen, liver, lung, and bone marrow and lymph node. (B) Identification of the NK cell subsets in the different organs. The subsets of CD3− NK1.1+ NK cells are differentially represented across a range of lymphoid and non-lymphoid organs.

4. Notes 1. Spoon sieves with a pore size of around 1–2 mm are most appropriate. Any larger than this and the liver in particular will not be sufficiently homogenized prior to centrifugation. 2. Although mouse blood can be treated in the same manner as human blood, in order to enrich murine lymphocytes this protocol is not necessary. The simpler method for mouse blood is to harvest blood directly into a 1 ml centrifuge tube into which 20 ␮l of 50 mM EDTA has been added. The blood/EDTA is then mixed to prevent coagulation and the cells pelleted by centrifugation. At this point it is possible to then process as per Section 3.2.1, step 2. 3. Always aspirate gently and leave around 500 ␮l of serum remaining above the lymphocyte band. 4. When harvesting lymphocytes from a band within a gradient it is best to aspirate from slightly above the cells. Try to remove as little of the gradient mixture (in this case Ficoll) as this will add to the density of the subsequent wash. If too much gradient and not enough PBS is in the wash step then the lymphocytes will not pellet.

36

Hayakawa, Andrews, and Smyth

5. Although it is possible to remove lymphocytes from spleen and lymph node by mechanical disruption between glass slides we have found this process to be less efficient and more damaging to the cells than collagenase digestion. Digesting organs in collagenase removes cells from the stroma, thereby increasing yields. Although collagenase may increase yields from liver and lung, collagenase should never be used when analyzing liver NK cells for CD27. CD27 expression in the liver is extremely sensitive to collagenase digestion resulting in NK cell preparations in which CD27 cannot be detected. 6. Lymph nodes are free from red blood cells and thus do not need ACK lysis. Naive lymph nodes and lungs have very few cells (˜106 ), thus volumes for resuspending should be carefully controlled in order to have accurate cell counts. The most appropriate volume for resuspending of bone marrow, spleen, and livers for accurate counting is 1 ml. Our laboratory normally resuspends lymph nodes in 200 ␮l of blocking buffer. For trypan blue exclusion counting a 1:10 to 1:20 dilution is acceptable while a 1:500 dilution is fine for use in a Coulter Counter. Using the volumes described here will result in high cell densities. This is important when transferring the cells to a 96-well U-bottom plate. The maximum volume in these wells is around 200 ␮l. Thus, resuspending the organs in a higher density in blocking buffer allows their transfer directly from the centrifuge tube to the 96-well plate without having to spin the cells and increase their density before transferring to the 96-well plate. 7. Percoll needs to be at room temperature when spinning. Always prepare fresh the morning of the experiment, prior to organ collection. As stock Percoll is kept at 4◦ C it is advisable to place the prepared gradients in a 37◦ C water bath while collecting organs. 8. The use of a P1000 pipette is more appropriate as this will better homogenize the pellet than larger pipettes (i.e., a 5–10 ml pipette). A more homogenized pellet will allow better separation of the lymphocytes from the hepatocytes or lung parenchyma. 9. When the lymphocytes are in ACK lysis buffer transfer them to a fresh tube. If the cells are pelleted in the same tube that the Percoll gradient was used for, some hepatocytes that can stick to the top of the tube may pellet with the lymphocytes following centrifugation of the ACK lysed cells. These hepatocytes can block the flow cytometer.

Subset Analysis of Human and Mouse Mature NK Cells

37

10. Cell numbers used in the FACS protocol must always be equivalent in order to ensure reproducibility of the FACS staining. If doing multiple stains it may only be possible to stain 105 cells (especially in the lung where only 106 total cells are recovered). 11. While most commercial outlets claim that 100 ␮l per well of 106 cells/cocktail is necessary we routinely use 20 ␮l per well on 106 cells. This is beneficial as it reduces reagent use by fivefold. It is always best to titrate amounts of antibodies used to optimize staining intensity and thereby assure that adequate gating can be achieved. 12. The use of microtiter tubes can greatly facilitate large experiments. These tubes can be inserted into a 96-well plate or 200 ␮l tip box. In contrast to the larger centrifuge tubes, the 1.2 ml tubes are small enough for all 12 ends of a multichannel pipette to fit. Thus, transfer of 12 wells at a time is possible. 13. When analyzing NK cell subsets the choice of fluorochromes is important. Our laboratory routinely uses NK1.1-PE (2 ␮g/ml), CD3-PECy5.5 (2 ␮g/ml), CD27APC (1 ␮g/ml), and CD11b-FITC (1 ␮g/ml). It is important to note that many mouse strains do not have NK1.1, including BALB/c. In order to identify NK cells in BALB/c mice it is necessary to use DX5 in replacement of NK1.1. DX5 should be used quite concentrated as the antigen is expressed weakly and DX5lo cells (which are also CD3− ) are not mature NK cells and may make analysis difficult. More recently an antibody to NKp46 has become available. This antigen is restricted to NK cells and is found in all mouse strains; however, it is yet to replace NK1.1 for FACS identification. The use of NK1.1-PE results in a bright signal for NK cells and is then easily discriminated from T cells and NKT cells (can be confirmed with CD1d tetramer). We do not encourage the use of similar fluorochromes for analysis of CD27 and CD11b. For example the use of CD27-APC and CD11b-APC-Cy7 is problematic. As these fluorochromes have very similar excitation and emission characteristics it is difficult to compensate them accurately. This makes it difficult to accurately discriminate the three NK cell stages. While APC-Alexa750 does not bleed into the APC channel as much, a similar problem regarding the subsets is still observed. In order to gate the subsets accurately it is necessary to include Fluorescent Minus One tubes (FMO). These tubes will contain either NK1.1/CD3/CD27 or NK1.1/CD3/CD11b. The use of these FMO tubes will allow quadrant gates to be set accurately.

38

Hayakawa, Andrews, and Smyth

Acknowledgments The authors would like to thank Ms. Anabel Silva for her assistance with developing the protocols for identifying human NK cell subsets with CD27. This work was supported by a National Health and Medical Research Council of Australia Program Grant and Doherty Fellowship (DMA, MJS). References 1. Smyth, M. J., Hayakawa, Y., Takeda, K., and Yagita, H. (2002) New aspects of naturalkiller-cell surveillance and therapy of cancer. Nat Rev Cancer 2, 850–61. 2. Lanier, L. L., Le, A. M., Phillips, J. H., Warner, N. L., and Babcock, G. F. (1983) Subpopulations of human natural killer cells defined by expression of the Leu-7 (HNK1) and Leu-11 (NK-15) antigens. J Immunol 131, 1789–96. 3. Ellis, T. M., and Fisher, R. I. (1989) Functional heterogeneity of Leu 19”bright”+ and Leu 19”dim”+ lymphokineactivated killer cells. J Immunol 142, 2949–54. 4. Huntington, N. D., Vosshenrich, C. A., and Di Santo, J. P. (2007) Developmental pathways that generate natural-killer-cell diversity in mice and humans. Nat Rev Immunol 7, 703–14. 5. Kim, S., Iizuka, K., Kang, H. S., Dokun, A., French, A. R., Greco, S., and Yokoyama, W. M. (2002) In vivo developmental stages

6.

7.

8.

9.

in murine natural killer cell maturation. Nat Immunol 3, 523–8. Takeda, K., Cretney, E., Hayakawa, Y., Ota, T., Akiba, H., Ogasawara, K., Yagita, H., Kinoshita, K., Okumura, K., and Smyth, M. J. (2005) TRAIL identifies immature natural killer cells in newborn mice and adult mouse liver. Blood 105, 2082–9. Hayakawa, Y., Huntington, N. D., Nutt, S. L., and Smyth, M. J. (2006) Functional subsets of mouse natural killer cells. Immunol Rev 214, 47–55. Silva, A., Andrews, D. M., Brooks, A. G., Smyth, M. J., and Hayakawa, Y. (2008) Application of CD27 as a marker for distinguishing human NK cell subsets. Int Immunol 20, 625–30. Vossen, M. T., Matmati, M., Hertoghs, K. M., Baars, P. A., Gent, M. R., Leclercq, G., Hamann, J., Kuijpers, T. W., and van Lier, R. A. (2008) CD27 defines phenotypically and functionally different human NK cell subsets. J Immunol 180, 3739–45.

Chapter 4 Assessing Licensing of NK Cells A. Helena Jonsson and Wayne M. Yokoyama Abstract Natural killer (NK) cells express receptors to detect and kill target cells based on expression of target cell surface molecules. Through a process termed NK cell licensing, only NK cells that express inhibitory receptors (e.g., Ly49 receptors in the mouse) for self-major histocompatibility complex (MHC) class I molecules become functionally competent to be triggered through their activation receptors. To determine the licensing status of particular Ly49+ murine NK cell subsets, splenocytes are stimulated with plate-bound anti-NK1.1 monoclonal antibody in the presence of brefeldin A and then assessed for NK cell activation on a single-cell basis using intracellular cytokine interferon-␥ staining and flow cytometry. Key words: Natural killer cells, NK cells, NK cell licensing, NK cell education, Ly49, inhibitory receptors, self-tolerance, NK cell activation.

1. Introduction Natural killer (NK) cells express receptors to detect and kill target cells that display cell surface ligands for these receptors (1). To accomplish this task, NK cells use activation and inhibitory receptors to balance the ability to recognize “missing” or “induced” self with the need to maintain self-tolerance (2, 3). Through an NK cell education process termed licensing, only NK cells that express inhibitory receptors for self-major histocompatibility complex (MHC) class I molecules become functionally competent to be triggered through their activation receptors (Fig. 4.1) (4–6). Conversely, NK cells that lack inhibitory receptors for selfMHC or develop in an environment devoid of MHC class I are unresponsive to most conventional NK cell activation receptor stimuli and hence are termed “unlicensed” (4). While the initial K.S. Campbell (ed.), Natural Killer Cell Protocols, Methods in Molecular Biology 612, DOI 10.1007/978-1-60761-362-6 4, © Humana Press, a part of Springer Science+Business Media, LLC 2010

39

40

Helena Jonsson and Yokoyama

Fig. 4.1. Licensing of NK cells. A developing NK cell stochastically expresses some but not all inhibitory NK cell receptors. In a host with an MHC environment where the NK cell is able to bind MHC class I via one or more of its inhibitory receptors (shown as matching black Ly49 and MHC class I molecules), the NK cell receives a licensing signal. This licensing signal requires an intact cytoplasmic Ly49 tail, including the ITIM. An NK cell that cannot engage MHC class I with any of its inhibitory receptors remains unlicensed, i.e., hyporesponsive. Upon stimulation via their activation receptors ex vivo, a higher percentage of licensed NK cells produce cytokines as compared to unlicensed NK cells.

studies of licensing were performed in mice, recent studies suggest a similar process occurs in humans (7, 8). In mice, the major NK cell inhibitory receptors that bind selectively to specific classical MHC class I alleles include members of the C-type lectin-like Ly49 family, such as Ly49A, C, I, and G2. For example, Ly49A binds H-2Dd but not H-2Kb , whereas the more promiscuous Ly49C binds H-2Kb well and H2Dd relatively weakly (9). Furthermore, inhibitory Ly49 receptors are expressed in a stochastic or probabilistic fashion such that each individual NK cell expresses a random assortment of the genome-encoded Ly49 molecules (10, 11). Because of these expression and binding patterns of Ly49 receptors, a significant portion of the NK cell population in any given mouse may be unable to engage self-MHC class I expressed on normal cells (4, 12). The unlicensed state of these cells ensures

Assessing Licensing of NK Cells

41

their self-tolerance since they are unable to respond to “missing self” stimuli. Notably, unlicensed cells are not hyporesponsive due to exhaustion because phorbol ester and ionomycin stimulation induces abundant cytokine production by these cells comparable to licensed NK cell populations (4). Finally, unlicensed NK cells can be induced to become functionally responsive in an MHCindependent manner by other stimuli, such as cytokine exposure in vivo or in vitro (4). In the laboratory, a licensed NK cell population is currently revealed by response to activation receptor cross-linking. AntiNK1.1 (Nkrp1c) has been most widely used though licensed NK cells can also respond to cross-linking of Ly49D, NKG2D, CD16, and YAC-1 targets (4). The use of brefeldin A and intracellular cytokine staining enables single-cell detection of activation and surface receptor expression using flow cytometry. The frequency of cytokine production by a given subset reflects its licensing status (Fig. 4.1).

2. Materials 1. Mice, including suitable positive and negative control strains (see Note 1). 2. R10 culture medium: RPMI-1640 supplemented with 10% heat-inactivated (45 min, 56◦ C) fetal bovine serum (FBS), 2 mM supplemental L-glutamine, 10 U/mL penicillin (optional), 10 ␮g/mL streptomycin (optional), 0.1 mM 2mercaptoethanol. Store at 4◦ C. 3. Phosphate-buffered saline (PBS): 137 mM NaCl, 8 mM Na2 HPO4 , 2.7 mM KCl, 1.5 mM KH2 PO4 , 0.5 mM MgCl2 , pH 7.2. Store at room temperature. 4. FACS staining buffer: 1× PBS (see above) supplemented with 1% FBS and 0.2% sodium azide (see Note 2). Store at 4◦ C. 5. Tissue culture-treated 6-well plates (see Note 3). 6. Purified PK136 (anti-NK1.1) antibody (#HB-191 at American Type Culture Collection (ATCC), Manassas, VA). 7. Brefeldin A – GolgiPlug (BD Biosciences, San Jose, CA) contains 0.1% BFA in DMSO (see Note 4). Store in 50 ␮L aliquots at 4◦ C (optional). 8. Phorbol ester and ionomycin – as optional positive control for cytokine production. Store aliquoted stock solutions of phorbol ester and ionomycin in DMSO at 0.5 and 1 mg/mL, respectively, at –20◦ C. Aliquot volumes of 10 and 20 ␮L, respectively, are convenient.

42

Helena Jonsson and Yokoyama

9. BD Cytofix/Cytoperm Fixation/Permeabilization Kit (BD Biosciences). Includes a formaldehyde- and saponin-based fixation and permeabilization reagent as well as a 10x concentrate of Perm/Wash buffer containing saponin to maintain permeabilization. 10. Staining tubes or plates (see Note 5). 11. Cell lineage antibodies (Table 4.1). These should include antibodies to gate out CD3+ cells and ideally also CD19+ cells. There should also be an antibody to positively gate on NK cells, such as a monoclonal antibody directed at NKp46, NK1.1 (in C57BL/6 mice), or integrin ␣2 (i.e., mAb clone DX5).

Table 4.1 Recommended fluorochrome selections for four-color flow cytometry Channel (FACSCalibur)

Staining setup #1

Staining setup #2

FL1

FITC-conjugated anti-Ly49 mAb

FITC-conjugated anti-IFN␥ mAb

FL2

PE-conjugated anti-IFN␥ mAb

Biotinylated anti-Ly49 mAb; PE-conjugated streptavidin

FL3

PerCP Cy5.5-conjugated anti-CD3 and anti-CD19 mAbs

PerCP Cy5.5-conjugated antiCD3 and anti-CD19 mAbs

FL4

APC-conjugated anti-NK1.1 mAb

APC-conjugated anti-NK1.1 mAb

12. Antibody monospecific for the inhibitory NK cell receptor of interest (see Note 6). 13. Antibody against murine interferon-␥ (clone XMG1.2). 14. Fc␥RII/III block, e.g., 2.4G2 supernatant (#HB-197 at ATCC). 14. 40 ␮m nylon mesh through which to filter cells after staining.

3. Methods 3.1. Preparation of Reagents 3.1.1. Prepare Splenocyte Suspension 3.1.2. Prepare Antibody-Coated Plates

1. For murine NK cells, harvest spleens and prepare RBC-lysed single-cell suspensions at a concentration of 107 cells/mL in cold, fresh R10 medium. 1. Dilute purified PK136 (anti-NK1.1) antibody to concentrations of 5, 2, and 0.5 ␮g/mL, respectively, in PBS (see Notes 7 and 8).

Assessing Licensing of NK Cells

43

2. Add 1 mL of antibody dilution to each well of polystyrene 6-well plates, making one well per stimulation dose per mouse. For example, for 10 mice, make 10 wells of 5 ␮g/mL PK136, 10 wells of 2 ␮g/mL PK136, and 10 wells of 0.5 ␮g/mL PK136. 3. Also make uncoated wells (one per mouse) by adding PBS alone or purified isotype control antibody diluted in PBS, for unstimulated control samples. 4. Incubate the plates in 37◦ C 5% CO2 incubator for 90 min. 5. Rinse the plate 3× with PBS. Do not let the wells dry at any time. Immediately add cells to the wells as described below. 3.2. Stimulation of NK Cells 3.2.1. Plate-Bound Monoclonal Antibody Stimulation of Murine Splenocytes

1. Add 107 splenocytes (i.e., 1 mL cell suspension) to each well. Gently swirl the plates to assure uniform distribution of cells over the bottom of the well. 2. If including a phorbol ester and ionomycin positive control, dilute them together in R10 medium to concentrations of 0.5 and 4 ␮g/mL, respectively. For example, dilute 5 ␮L of 0.5 mg/mL phorbol ester and 20 ␮L of 1 mg/mL ionomycin into 5 mL of R10 medium. Add 1 mL of this dilution to 1 mL of cell suspension. 3. Incubate the cells in a 37◦ C 5% CO2 incubator for 1 h (see Note 9). 4. After 1 h of stimulation, add Brefeldin A (BD GolgiPlug) as follows: dilute 50 ␮L of 0.1% BFA stock into 450 ␮L of R10 medium for 100× working stock, and add 12 ␮L working stock to each sample to each sample for a final concentration of 1.2×. Mix well by gently swirling the plate. 5. Incubate the cells for an additional 7 h in a 37◦ C 5% CO2 incubator (see Note 10). 6. Stop the stimulation by adding 5 mL cold FACS staining buffer and storing the plates at 4◦ C until ready to proceed to cell harvest (usually overnight) (see Note 11).

3.2.2. Cell Harvest

1. Harvest cells by pipeting up and down repeatedly with a 5 or 10 mL pipet. Collect the cells into a 15 mL tube. Wash the well with an additional 5 mL FACS staining buffer. Collect the wash into the same 15 mL tube. 2. Spin tubes at 500g for 5 min at 4◦ C. Discard supernatant. 3. Resuspend pellet in the remaining buffer and transfer the cell suspension to a FACS tube or deep-well plate (see Note 5). 4. Also set up unstained and single-stain control tubes or wells using pooled residual stimulated cells or fresh splenocytes (see Note 12).

44

Helena Jonsson and Yokoyama

3.3. Immunofluorescent Staining of Stimulated Cells 3.3.1. Surface Staining

1. If using a biotinylated anti-Ly49 receptor monoclonal antibody, dilute this antibody in Fc ␥R block (e.g., 2.4G2 supernatant) and add to the cells (see Note 6). If the anti-Ly49 receptor antibody is directly conjugated, proceed to Step 4. Incubate the cells on ice for 30 min. 2. Wash twice in 1 mL FACS staining buffer. Spin tubes or plates at 500g after each wash. After second wash, resuspend cells in 100 ␮L FACS buffer. 3. Make antibody master mix by combining anti-CD3, CD19, and -NKp46 antibodies and either the directly conjugated anti-Ly49 receptor or a secondary reagent (e.g., streptavidin-PE) in Fc␥R block (2.4G2 supernatant) in a total volume of 70 ␮L per tube or well. Aliquot the master mix to each tube and pipet or vortex to mix. Also make single-stain controls. Incubate the cells on ice for 30 min. 4. Wash twice in 1 mL FACS staining buffer. Spin tubes or plates at 500g after each wash.

3.3.2. Intracellular Cytokine Staining

1. After aspirating or decanting the supernatant from the last wash of the surface staining, vortex the plate or tubes to loosen the pellets. 2. Add 250 ␮L Cytofix/Cytoperm to each sample and pipet repeatedly to mix. Incubate on ice for 25 min, shaking the tubes or plate after 12 min. 3. Add 1 mL of Perm/Wash buffer to each sample and incubate 5 min on ice. 4. Spin at 500g for 5 min and discard supernatant. 5. Add 1 mL of Perm/Wash buffer and resuspend the pellet by pipeting. Spin for 5 min at 500g and discard supernatant. 6. Make a master mix of anti-IFN␥ antibody diluted in Perm/Wash buffer to a final volume of 50 ␮L per sample. Aliquot the master mix to each of the wells and pipet or vortex to mix. Incubate the cells 30 min on ice. 7. Wash twice in 1 mL Perm/Wash buffer. 8. Add 300 ␮L of FACS staining buffer, resuspend the pellet by pipeting, and filter each sample through 40 ␮m nylon mesh into new FACS tubes. 9. Store FACS tubes at 4◦ C, protected from light, until ready to analyze by flow cytometry (see Note 13).

3.4. Flow Cytometric Data Collection and Analysis 3.4.1. Data Collection

1. Arrange the flow cytometer settings for post-collection compensation (see Notes 14 and 15). 2. For best results, collect flow cytometry data for at least 106 live cells (approximately 15,000 NK cells, gated as NKp46+

Assessing Licensing of NK Cells

45

Fig. 4.2. Analysis of licensing assay data. To quantify licensing, graph NK cells (NK1.1+ CD3− CD19− ) on a dot plot with Ly49 on the vertical axis and IFN␥ on the horizontal axis and make a quadrant gate to frame the four populations. In this example, the middle plot displays NK cells from a B10.D2 (H-2d ) mouse, whereas the bottom plot displays NK cells from a C57BL/10 (H-2b ) mouse. The frequency of IFN␥ production by Ly49+ NK cells is calculated using the formula UR/(UR+UL)∗ 100. A high frequency of IFN␥ production (usually 12–25%) indicates licensing of that Ly49+ population. The “licensing ratio” describes the increased propensity of Ly49A+ NK cells to produce IFN-␥ relative to Ly49A− NK cells and is the result of dividing the percentage of Ly49A+ NK cells that produce IFN␥ by the percentage of Ly49A− NK cells that produce IFN␥. If the Ly49 receptor does not mediate licensing, the licensing ratio is approximately 1 (usually 0.8– 1.2). If the Ly49 receptor mediates licensing, the licensing ratio will be over 1.5.

CD3− CD19− ). This often means collecting all events until the sample runs out. 3.4.2. Analyze Flow Cytometry Data

1. Gate on live cells using a forward scatter and side scatter dot plot. Subsequently, gate on NK cells (NKp46+ CD3− CD19− ) (see Note 16).

46

Helena Jonsson and Yokoyama

2. Display NK cells on a dot plot with Ly49 on the vertical axis and IFN␥ on the horizontal axis and make a quadrant gate to frame your four populations (Fig. 4.2). The quadrants will henceforth be referred to as UR (upper right), UL (upper left), LL (lower left), and LR (lower right). 3. Calculate the frequency of IFN␥ production by Ly49+ NK cells using the formula UR/(UR+UL)∗ 100. A high frequency of IFN␥ production (usually 12–25%) indicates licensing of that Ly49+ population. 4. Calculate the “licensing ratio” of the Ly49+ NK cell population using the formula (UR/(UR+UL))/(LR/(LR+LL)). If the Ly49 receptor does not mediate licensing, the licensing ratio is approximately 1 (usually 0.8–1.2). If the Ly49 receptor mediates licensing, the licensing ratio will be over 1.5 (see Note 17). Positive and negative control samples are very important in these calculations (see Note 1).

4. Notes 1. Including appropriate positive and negative control strains is very important. For Ly49C+ NK cells, a C57BL/6 control mouse is a good positive control for licensing. For experiments addressing licensing through Ly49A, a B10.D2 mouse (H-2d , Stock #000463, Jackson Labs, Bar Harbor, ME) or a H-2Dd -transgenic mouse is an appropriate positive control. In most cases, a MHC class I-deficient mice (e.g., ␤2 m−/− ) is a good negative control for cytokine production: NK cells from these mice should produce little if any cytokine in this assay, although they can respond to high concentrations of cross-linking antibody. Nonetheless, licensed NK cells respond better than unlicensed NK cells regardless of strength of activation (4). It is also often useful to include a MHC-sufficient strain that is known not to license the Ly49+ population of interest. For example, a B10.D2 mouse (H-2d ) can serve as an unlicensed control for studies of Ly49C+ NK cells, as can a C57BL/6 mouse (H-2b ) for studies of Ly49A+ NK cells. In addition, the use of age-matched mice is important as NK cell activity changes with age in early adult mice. Dates of birth should ideally be matched to within 2–3 weeks. Mice younger than 7 weeks and older than 16 weeks may have poor NK cell responses and should be avoided. 2. In addition to preventing contamination, azide in the FACS staining buffer blocks internalization or shedding of antibody–receptor complexes.

Assessing Licensing of NK Cells

47

3. This assay is normally performed in 6-well plates using 107 splenocytes in a 1 mL reaction volume. However, the assay can be scaled down if necessary for other applications where cell number is limited. 4. Monensin can be used as an alternative to brefeldin A, if desired. 5. For staining of stimulated cells, conventional 12 × 75 mm round bottom culture tubes work well. For processing of larger numbers of samples, deep-well conical-bottom plates (e.g., 2.2 mL Mark II plates (#AB-0932), ABgene, Surrey, UK) may be more convenient. Washes can then be performed quickly and easily with a P1000 multichannel pipetman. Traditional 96-well conical bottom plates are less ideal for staining several million cells per well as they can only hold small volumes and therefore will require extra wash steps. 6. The anti-Ly49 monoclonal antibody (mAb) must be selected with care. For best results, the mAb should be monospecific, such as clones 4LO33 (anti-Ly49C) and JR9 (anti-Ly49A). Certain mAbs, such as clone A1 (antiLy49A), are very sensitive to cis engagement of Ly49 with MHC class I expressed on the same cell, which results in decreased mAb reactivity (4, 13). These mAb clones should be avoided because it may be difficult or impossible to distinguish the relevant Ly49+ and Ly49– populations. If alternative mAb clones are not available, carefully optimize the choice of fluorochromes and staining conditions. For example, biotinylated 4LO33 (anti-Ly49C) followed by PE-conjugated streptavidin produces a Ly49C+ population that is clearly distinguishable from the Ly49C− cells, while staining with FITC-conjugated 4LO33 does not. 7. The cell suspensions should be ready to go by the time the plates are washed. The timing usually works out right if the purified PK136 antibody is added to the plates immediately after RBC lysis of the splenocytes. The 90-min incubation of the plates provides enough time to count the cells and resuspend them to 107 cells/mL. 8. If enough cells are available, do one well of each antibody concentration for each mouse. If cell number is a limitation, omit the 0.5 and 5 ␮g/mL PK136 stimulations. 9. Incubation of less than 1 h before adding BFA results in significantly lower yield of intracellular cytokines. 10. After the addition of BFA, each hour of incubation increases the amount of intracellular cytokines. However, incubations longer than 8 h may increase cell death due to

48

Helena Jonsson and Yokoyama

BFA toxicity. Seven hours of incubation with BFA allows for ample accumulation of intracellular cytokines without any noticeable cytotoxic effects. 11. The overnight incubation at 4◦ C is not a necessary step but is merely for convenience. 12. Single-stain controls should be made for each fluorochrome. Since NK cells are rare among splenocytes, avoid anti-NK1.1 or anti-Ly49 antibodies for these controls. Instead, use fluorochrome-conjugated antibodies directed at more common markers such as CD19 or CD8 that are expressed at similar levels to markers of interest in NK cells. The unstained control should consist of cells pooled from the remainder of the stimulation. The single-stained controls can be made of these cells as well or from fresh splenocytes. Alternatively, compensation control beads (BD Biosciences) can be used in place of cells. 13. Since the cells are fixed, flow cytometric data collection can be postponed for up to 1 week. However, best results are obtained by analyzing the cells within 24 h of staining. 14. Licensing assays require the use of a flow cytometer that can distinguish at least four colors. Table 4.1 lists two fluorochrome combinations that work well on BD FACSCalibur machines. If simultaneous analysis of multiple Ly49 populations is desired, the fluorochromes must be selected with care to avoid compensation issues since the NK cell subset populations are so rare. 15. If possible, perform compensation after data collection (14). Many contemporary flow cytometry analysis software programs (e.g., FlowJo, Treestar, Ashland, OR) feature easy-to-use post-collection compensation functions. 16. If using NK1.1 to define the NK cell population, be aware that plate-bound PK136 stimulation partially inhibits subsequent staining with fluorochrome-conjugated antiNK1.1 antibodies. In this situation, use a free-form polygon gate to gate on all NK1.1+ cells, even those exhibiting intermediate staining. (NK cells with lower NK1.1 expression are often the most robust cytokine-producing cells.) Also, stimulating with a lower concentration (e.g., 2 ␮g/mL or lower) of plate-bound PK136 may make the NK1.1+ population more distinct. Alternatively, use antibodies against a different marker, such as NKp46 or integrin ␣2 (DX5), to identify NK cells. 17. The licensing ratio is dependent on the “background” IFN-␥ production by NK cells, i.e., the percentage of IFN␥+ cells among NK cells that lack the receptor of inter-

Assessing Licensing of NK Cells

49

est. This ratio may therefore falsely indicate no licensing of a given Ly49 subset if the Ly49-negative subset expresses another receptor capable of self-MHC recognition. Care must be taken when using licensing ratios, especially between murine strains and Ly49/MHC combinations that differ from the Ly49A-H-2Dd and Ly49C-H2 Kb interactions already documented. Absolute frequencies of IFN␥ production may be preferable in these situations. It is also possible that if the NK cells had been previously stimulated in vivo or in vitro with cytokines or other stimuli, then it may be difficult to show an MHCdependent licensing effect because the stimuli may induce unlicensed NK cells to become functionally competent (4).

References 1. Yokoyama, W. M. (2008) Chapter 17. Natural killer cells, in Fundamental Immunology. (Paul, W. E., ed.), Lippincott-Raven, New York. pp. 483–517. 2. Diefenbach, A., and Raulet, D. H. (2003) Innate immune recognition by stimulatory immunoreceptors. Curr Opin Immunol 15, 37–44. 3. Ljunggren, H. G., and Karre, K. (1990) In search of the ‘missing self’: MHC molecules and NK cell recognition. Immunol Today 11, 237–244. 4. Kim, S., Poursine-Laurent, J., Truscott, S. M., Lybarger, L., Song, Y. J., Yang, L., French, A. R., Sunwoo, J. B., Lemieux, S., Hansen, T. H., and Yokoyama, W. M. (2005) Licensing of natural killer cells by host major histocompatibility complex class I molecules. Nature 436, 709–713. 5. Yokoyama, W. M., and Kim, S. (2006) How do natural killer cells find self to achieve tolerance? Immunity 24, 249–257. 6. Yokoyama, W. M., and Kim, S. (2006) Licensing of natural killer cells by self-major histocompatibility complex class I. Immunol Rev 214, 143–154. 7. Anfossi, N., Robbins, S. H., Ugolini, S., Georgel, P., Hoebe, K., Bouneaud, C., Ronet, C., Kaser, A., DiCioccio, C. B., Tomasello, E., Blumberg, R. S., Beutler, B., Reiner, S. L., Alexopoulou, L., Lantz, O., Raulet, D. H., Brossay, L., and Vivier, E. (2004) Expansion and function of CD8+ T cells expressing Ly49 inhibitory receptors specific for MHC class I molecules. J Immunol 173, 3773–3782. 8. Kim, S., Sunwoo, J. B., Yang, L., Choi, T., Song, Y. J., French, A. R., Vlahiotis,

9.

10.

11.

12.

13.

14.

A., Piccirillo, J. F., Cella, M., Colonna, M., Mohanakumar, T., Hsu, K. C., Dupont, B., and Yokoyama, W. M. (2008) HLA alleles determine differences in human natural killer cell responsiveness and potency. Proc Natl Acad Sci U S A 105, 3053–3058. Hanke, T., Takizawa, H., McMahon, C. W., Busch, D. H., Pamer, E. G., Miller, J. D., Altman, J. D., Liu, Y., Cado, D., Lemonnier, F. A., Bjorkman, P. J., and Raulet, D. H. (1999) Direct assessment of MHC class I binding by seven Ly49 inhibitory NK cell receptors. Immunity 11, 67–77. Raulet, D. H., Vance, R. E., and McMahon, C. W. (2001) Regulation of the natural killer cell receptor repertoire. Annu Rev Immunol 19, 291–330. Pascal, V., Stulberg, M. J., and Anderson, S. K. (2006) Regulation of class I major histocompatibility complex receptor expression in natural killer cells: one promoter is not enough! Immunol Rev 214, 9–21. Fernandez, N. C., Treiner, E., Vance, R. E., Jamieson, A. M., Lemieux, S., and Raulet, D. H. (2005) A subset of natural killer cells achieves self-tolerance without expressing inhibitory receptors specific for selfMHC molecules. Blood 105, 4416–4423. Doucey, M. A., Scarpellino, L., Zimmer, J., Guillaume, P., Luescher, I. F., Bron, C., and Held, W. (2004) Cis association of Ly49A with MHC class I restricts natural killer cell inhibition. Nat Immunol 5, 328–336. Herzenberg, L. A., Tung, J., Moore, W. A., Herzenberg, L. A., and Parks, D. R. (2006) Interpreting flow cytometry data: a guide for the perplexed. Nat Immunol 7, 681–685.

Chapter 5 Use of Stem Cell Radiation Chimeras to Analyze How Domains of Specific Proteins Impact on Murine NK Cell Development In Vivo Rebecca H. Lian and Vinay Kumar Abstract Although the use of mutant mice has been extremely useful in identifying those proteins and molecules specifically required for the development of NK cells, the establishment of a well-defined protocol to replicate in vitro the major steps corresponding to the process of NK cell differentiation and maturation has enabled us to dissect the molecular events governing certain aspects of NK cell development. This chapter describes a protocol that combines both the use of mutant mice and the in vitro bone marrow (BM) culture system for examining the role of proteins and their putative signaling domains in NK cell development. BM-derived Lin–c-kit+ stem cells expressing the protein of interest are first cultured for 6 days in a cocktail of cytokines that promote lymphoid development. The semi-differentiated cells are then transplanted into mice to complete their development in vivo. While all hematopoietic lineages can develop from these transplanted cells, we focus primarily on assessing the effect of the protein on the production of NK cells, as well as the acquisition of Ly49 receptors. The most prevalent advantage of this method is the ability to potentially link signaling regulators to known aspects of NK cell development. Key Words: Stem cell chimeras, natural killer cells, Ly49, IL-15R␣.

1. Introduction The biological importance of certain protein receptors or signaling molecules is most often revealed through studies using mutant mice (1). Likewise, in the process of identifying the necessary components needed for natural killer (NK) cell development, molecules such as lymphotoxin, IL-15, IL-15R, IL-2R, Id2, IRF1, Ets-1, and many others were, through the use of genetically deficient mice, categorically described as indispensable for the normal production and maturation of NK cells (2). However, K.S. Campbell (ed.), Natural Killer Cell Protocols, Methods in Molecular Biology 612, DOI 10.1007/978-1-60761-362-6 5, © Humana Press, a part of Springer Science+Business Media, LLC 2010

51

52

Lian and Kumar

linking the biochemical nature of a particular region within a surface receptor or intracellular protein to its specific developmental function in vivo would require generating multiple mutant mice, each transgenically expressing a point mutation or truncation in the protein of interest. Although a viable option, we have instead employed the use of generating radiation chimeras in which donor whole bone marrow cells are substituted with genetically modified stem cells. Traditional radiation chimeras are generated by transplanting whole bone marrow (BM) cells from one donor mouse into a lethally irradiated recipient mouse and then monitoring the developmental path of the donor cells in the milieu of the recipient. In this chapter, we describe a method to study how mutated forms of a protein that is known to be essential in NK development affect the process of differentiation and development of NK progenitors that express those proteins. For obvious reasons, the BM cells for transplantation must originate from a donor that is null for the protein of interest. Then, because whole BM cells contain a mixed population of cells undergoing various stages and lineages of development, purified undifferentiated stem cells are a preferred choice as donors for transplantation. Also, if the protein being analyzed is required for NK cell production, the BM population of these mice would, therefore, most likely lack NK progenitors, leaving the stem cells as the best candidates for study. Finally, the use of stem cells lets us manipulate the system to enrich for only those cells that are successfully expressing cDNA encoding the protein of interest. In this protocol Lin–c-kit+ stem cells are purified from the whole bone marrow of a knock-out (KO) mouse, cultured in vitro under conditions that initiate NK cell differentiation (3, 4), retrovirally transduced with cDNA encoding various mutated forms of the knock-out protein, and consequently re-established in a recipient wild-type mouse. The NK cell progenitors whose development was promoted by the in vitro differentiation system will continue their maturation in vivo in a normal bone marrow milieu, but under the influence of a mutated receptor that is expressed on the progenitors. Because fully matured NK cells acquire various NK lineage markers, including Ly49 receptors (5, 6), the effect of the mutated receptor on proper NK cell development and production can be easily assessed by analyzing the splenocytes of the chimeric mice by flow cytometry for these markers.

2. Materials 1. RNA extraction reagent such as Trizol (www.invitrogen. com) or the RNeasy Plus kit (www1.qiagen.com).

Use of Stem Cell Radiation Chimeras

53

2. A first-strand cDNA synthesis kit (www.fermentas.com). 3. pCR2.1-TOPO cloning kit (Invitrogen). 4. pMIGR1-GFP retroviral vector is available from most labs that routinely perform retrovirus-based transfection methods. Use the standard protocol for requesting and obtaining reagents from other research labs – e-mailing a request usually suffices. 5. Retrovirus packaging cell line: we use PlatE cells, but any 293-based packaging cell line (such as GP293, 293FT, or Phoenix) can be used as long as the vectors encoding the gag, pol, or env genes are available for co-transfection with the retroviral vector. 6. Reagents for PlatE cell maintenance and transfection: 0.25% Trypsin-1 mM EDTA solution, blastocidin, and puromycin (both from InvivoGen, www.invivogen.com); ExGen500 In Vitro Transfection Reagent (Fermentas), sterile 150 mM NaCl. 7. Mice: Mice that are deficient in the protein of interest (knock-out, KO) that is on the C57BL/6J background (CD45.2, wild type), 4–6 weeks old (JAX, stock no. 000664) to provide donor stem cells; B6.SJLPtprca Pepcb /BoyJ congenic mice (CD45.1), approximately 6 weeks old (JAX, stock no. 002014) to serve as recipient animals. After transplantation, donor cells of KO mice can be differentiated from host congenic mouse cells by their respective CD45.2 or CD45.1 protein marker. 8. Flavored pediatric Bactrim (Trimethoprim-Sulfoxaxole) can be obtained from the animal housing facility veterinarian or staff and administered to mice before and after irradiation. 9. QuadroMACSTM and MiniMACSTM cell separation units and the MACS MultiStand (Miltenyi Biotec; www.miltenyibiotec.com). 10. LS and MS columns (Miltenyi Biotec). 11. Streptavidin (SA) magnetic microbeads or particles for binding to biotinylated antibodies (Miltenyi Biotec, cat# 130-048-102 or BD Biosciences, cat# 557811) and antiFITC magnetic microbeads (Milteny Biotec). 12. MACS buffer: 1× Dulbecco’s phosphate-buffered saline (DPBS) containing 2 mM EDTA and 0.5% bovine serum albumin. 13. DMEM-10 medium: Dulbecco’s Modified Eagle’s Medium (DMEM) containing 4,500 mg/L D-glucose with 10% fetal bovine serum, 100 units/ml penicillin,

54

Lian and Kumar

100 ␮g/ml streptomycin, 10 mM HEPES, 1 mM sodium pyruvate, 0.1 mM MEM nonessential amino acids, and 50 ␮M 2-mercaptoethanol. 14. Differentiation medium: DMEM-10 medium containing 1 ␮g/ml indomethacin, 0.5 ng/ml murine IL-7, 30 ng/ml murine stem cell factor (SCF), and 20 ng/ml human Flt3ligand. 15. Antibodies: Conjugated monoclonal antibodies (mAbs) recognizing CD45.1, CD45.2, NK1.1 (clone PK136), CD3 (clone 2C11), Ly49A (any clone), Ly49C/I (clone 5E6), Ly49G2 (clone 4D11), and Ly49D (clone 4E5). Other mAbs used for purification of bone marrow (BM)-derived stem cells include biotinylated forms of anti-CD11b (Mac-1), anti-Gr-1, anti-B220, antierythroid cells (TER-119), and anti-CD2. All mAbs can be purchased from BD Biosciences or eBioscience (www.ebioscience.com) at stock concentrations of 0.5 mg/ml. Also needed will be a biotinylated mAb that recognizes the protein of interest if it is expressed on the cell surface. 16. Cytokines and growth factors: Recombinant murine IL7, Flt3-Ligand (Flt3L), and stem cell factor (SCF) can be purchased from Invitrogen or from R&D Systems (www.rndsystems.com). 17. Flow cytometry: Cell acquisition is performed using the FACS Diva software on the FACSCanto (BD Biosciences). Sorting of BM stem cells is accomplished using the MoFlo (Beckman Coulter, Miami, FL) and data are analyzed using FlowJo software (www.flowjo.com). 18. 2.4G2 hybridoma supernatant (cell line from ATCC, cat# HB-197). 19. Other reagents: 0.2% NaCl and 1.6% NaCl for lysis of erythrocytes; hexadimethrine bromide (Sigma-Aldrich) for enhancing retroviral infection.

3. Methods 3.1. Cloning Your Gene into the Retroviral Expression Vector pMIGR1-GFP

The pMIGR1-GFP retroviral vector contains an IRES sequence directly upstream of the GFP gene. Therefore, subcloning your gene upstream of the IRES sequence will permit the expression of the GFP marker in tandem with your gene of interest (see Fig. 5.1). Due to this bi-cistronic system of expression, transfected or transduced cells that are GFP+ are, therefore, more than

Use of Stem Cell Radiation Chimeras

55

Fig. 5.1. The pMIGR1-GPF vector for transfection of purified Lin–c-kit+ stem cells. It is 6056 bp in size, contains an internal ribosomal entry site (IRES), and a multiple cloning site at BglII (1406) to EcoRI (1424).

likely to also express the gene that is subcloned upstream of the IRES-GFP sequence. The GFP marker also facilitates easy and rapid monitoring of transfection and transduction efficiencies, as well as permits enrichment of positive cells by cell sorting using GFP fluorescence without the need for antibody staining. 1. Amplify your gene of interest (wild type or mutated) by PCR and insert the cDNA into any sub-cloning vector such as pCR3.2-TOPO (Invitrogen) using TA cloning methods according to the manufacturer’s instructions. To generate point or deletion mutants, it is easiest to incorporate the mutation sites into the PCR primers used for cloning. 2. Sequence the cDNA to verify that the gene has been correctly cloned with respect to length and PCR proofreading accuracy. 3. Excise the gene and insert into the pMIGR1 expression vector containing an IRES-controlled GFP gene. 4. Digest the vector and gene at the appropriate sites and sequence the final construct to verify that the gene has been correctly inserted with respect to orientation. 3.2. Preparing Viral Supernatant Using PlatE Packaging Cells

PlatE cells have been genetically manipulated and already contain vectors encoding all the genes necessary for virus particle packaging – gag, pol, and env(7). To maintain only a population of useful (positive) cells, they are cultured in the presence of antibiotics until just 24 h prior to transfection.

56

Lian and Kumar

3.2.1. Culture of PlatE Cells

1. To maintain cultures, grow cells in DMEM-10 medium containing 1 ␮g/ml of puromycin and 10 ␮g/ml blastocidin to keep cells under selective pressure. These antibiotics are necessary to prevent the outgrowth of vector-negative PlatE cells. Prolonged maintenance of these cells beyond 6 weeks is not recommended since the cultures may phenotypically and/or functionally drift. 2. When the cells are confluent, divide them at a 1:10 dilution.

3.2.2. Transfection of cDNA into PlatE Cells Using the ExGen500 Reagent

1. 24 h before transfection, plate 2 × 106 PlatE cells in 10 ml of DMEM-10 medium without puromycin or blastocidin into a 100 mm tissue culture dish. Optimum density should be achieved 12–16 h postplating. 2. The next day, remove the medium from the dish containing the cells and add 3 ml of fresh DMEM-10 medium. 3. For each dish of PlatE cells to be transfected, prepare one sterile Eppendorf (microfuge) tube containing a mixture of 5 ␮g of cDNA and 300 ␮l sterile 150 mM NaCl. Add 16.45 ␮l of ExGen500 reagent into the DNA/salt solution and mix gently for 10 s. Let the mixture incubate for 10 min at room temperature. Following this, add the DNA/salt/ExGen500 mixture dropwise and evenly onto the PlatE cells. Gently rock the plate back and forth to evenly distribute the DNA solution. If possible, spin the plate for 5 min at 280 × g at room temperature. Incubate these cells overnight in 37◦ C/5% CO2 . For production of mock virus supernatant, add nuclease-free water instead of cDNA. 4. The day after transfection, remove the medium from the cells and add 3 ml of fresh DMEM-10 medium to the culture dish. Return cells to the incubator. 5. To verify that the transfected PlatE cells are expressing your gene of interest, harvest a small sample of the cells 48–72 h posttransfection and analyze for GFP expression by flow cytometry (see Note 1). 6. Harvest supernatant from cells whose GFP expression levels are equivalent to or more than 90%. This supernatant contains virus particles and will be used for infection of mouse stem cells (see Note 2). 7. Filter the supernatant through a sterile 45 ␮m syringe filter to remove any cellular debris, aliquot into sterile tubes, and store at −80◦ C to −135◦ C. Supernatants can be thawed just prior to each use. Repeat freezing and thawing is not recommended.

Use of Stem Cell Radiation Chimeras

3.3. Preparing NK Progenitors from Mouse Bone Marrow for Retroviral Infection and Transplantation 3.3.1. Harvesting Whole Bone Marrow Cells from Mice

57

1. Euthanize mice whose bone marrow cells are to be harvested for transplanting into a recipient. The donor mice are those that are deficient for the protein of interest and usually KO mice of the CD45.2 genotype. 2. Remove both the large femoral bones from the mice. Scrape away all muscle tissue and place all the cleaned bones into a sterile ceramic mortar containing 5–10 ml of DMEM-10 medium. 3. Using a sterile pestle, carefully grind the bones to release the cells in the marrow space. 4. Place a 70 ␮m cell strainer (BD Falcon) over the mouth of a sterile 50-ml tube, and using a pipet, transfer the cell solution into the tube through the strainer. This step filters out the bone debris. 5. Add 5–10 ml of sterile 1X PBS to the mortar dish and swirl to rinse the bones. Transfer the cell solution into the same 50-ml tube, again filtering through the cell strainer. 6. Continue adding fresh 1× PBS to the 50-ml tube through the strainer until the tube is filled. 7. Centrifuge at 500 × g for 10 min at 4◦ C to pellet the cells. 8. Loosen the pellet by vortexing. Lyse the erythrocytes by adding 10 ml of sterile 0.2% NaCl solution followed immediately by equal volumes of 1.6% NaCl solution. The first salt solution is hypotonic and quickly lyses the erythrocytes, whereas the second salt solution is hypertonic and is required to restore osmotic balance to the cell environment. Delay in adding the hypertonic salt solution may result in lysis of the other cell types, thus lowering your yield. 9. Fill the remaining volume in the tube with DMEM-10 medium. 10. Centrifuge to pellet the cells. Resuspend the pellet in 5–10 ml of sterile azide-free 2.4G2 hybridoma supernatant and remove the cell debris by filtering through a cell strainer. 11. Count cells to determine total cell count. 12. Incubate the cells (resuspended in the 2.4G2 supernatant) at 4–8◦ C for 20 min.

3.3.2. Purifying Lin– c-kit+ Stem Cells

1. Centrifuge to pellet the whole BM cells that were incubating in 2.4G2 hybridoma supernatant. 2. Decant the 2.4G2 supernatant and loosen the cell pellet. To the pellet add 2–3 ␮l of each of the following biotinylated mAbs (taken directly from the stock without

58

Lian and Kumar

dilution): anti-CD2, anti-CD3, anti-Gr-1, TER119, antiMac-1 (CD11b), anti-NK1.1, and anti-B220. At this time, 2–3 ␮l of FITC-labeled anti-c-kit mAb is also added to the cells. 3. Incubate cells with mAbs on ice for 30 min. 4. Wash cells 1× with sterile ice-cold MACS buffer by centrifugation. 5. Resuspend the pellet with 90 ␮l of MACS buffer for every 1 × 107 (or fewer) total cells. Add 10 ␮l of SA microbeads (Miltenyi Biotec) or SA particles (BD Biosciences) for every 1 × 107 (or fewer) total cells. 6. Mix thoroughly and incubate cells with beads on ice for 20–30 min or at 4◦ C for 12–15 min. 7. During this incubation, prepare the LS column (use more than 1 column if cell number is >3 × 108 total) by placing it on the QuadroMACSTM magnet and rinsing it with 5 ml of de-gassed ice-cold MACS buffer. Discard the flowthrough buffer. 8. When the cells have completed their incubation, wash them once with excess MACS buffer and resuspend the pellet in 3 ml of de-gassed MACS buffer. 9. Add the cell solution to the previously rinsed LS column(s) and allow the cell mixture to flow through the column bed. Collect the flow-through – this fraction contains the Lin– cells. 10. Wash the column 3× with MACS buffer and collect the flow through from each wash. 11. To collect the c-kit+ population, centrifuge the flowthrough to pellet the Lin– cells. 12. To the pellet, add 90 ␮l of MACS buffer and 10 ␮l of antiFITC microbeads for every 1 × 107 (or fewer) total cells. 13. Mix thoroughly and incubate at 4◦ C for 20 min. 14. While cells are incubating, prepare the MS column (may need to use two columns depending on the total cell number) by attaching it to the MiniMACSTM separation magnet and rinsing it with 1 ml of de-gassed MACS buffer. Discard the flow-through. 15. After incubation, collect the cells by washing with excess volumes of MACS buffer. 16. Resuspend the pellet in 0.5 ml MACS buffer and transfer the mixture to the previously rinsed MS column(s). Allow the cell mixture to flow through the column bed. Discard the flow-through – it contains the c-kit− cells. 17. Wash the column 3× with MACS buffer and discard the flow-through from each wash.

Use of Stem Cell Radiation Chimeras

59

18. Transfer the MS column to a clean (sterile) 5-ml tube, add 1 ml of DMEM-10 medium and immediately elute the c-kit+ cells using the column plunger. 19. Centrifuge and resuspend in DMEM-10 medium. Count and resuspend cells at 8 × 105 –1.5 × 106 cells/ml of Differentiation medium. Distribute 2 ml of the cell solution into each well of a 24-well sterile tissue culture plate with lid and incubate overnight at 37◦ C with 5% CO2 . 3.3.3. Retroviral Infection of Lin–c-kit+ Stem Cells

1. Carefully remove 1 ml of medium from each well of stem cells. 2. Add 1 ml of viral supernatant stock to each well of cells. As a control, add 1 ml of mock virus (no particles) supernatant to a few wells. 3. To each well also add 10 mM HEPES and 2 ␮g/ml hexadimethrine bromide, also known as polybrene, and mix gently. Polybrene is a cationic compound that enhances the interaction between virus particles and the cell surface membrane by neutralizing repulsion charges between the virus and the cell membrane. 4. Carefully wrap the plate and centrifuge at 1,300 × g for 90 min at 30◦ C (spinfection step). 5. Resuspend cells by gentle mixing and incubate at 37◦ C and 5% CO2 for 2–4 h. 6. Harvest cells, pooling them according to their condition, and centrifuge as usual. 7. Resuspend the cells in Differentiation medium and reculture them in a clean 24-well plate at 2 ml per well. 8. Incubate at 37◦ C and 5% CO2 overnight. 9. Repeat the virus spinfection steps the next day, and if desired, the following day as well for a total of three rounds of infection (see Note 3). 10. On day 5, GFP expression can be assessed by flow cytometry. GFP+ cells can be enriched by FACS sorting (see Note 4). 11. After enrichment, culture the cells for one additional day in Differentiation medium prior to transplantation into lethally irradiated recipients.

3.4. Generating Radiation Chimeras 3.4.1. Transplanting Whole BM Cells

Since the idea is to use stem cells instead of whole bone marrow cells for transfer, it is important to make a side-by-side comparison of mouse chimeras generated from each donor cell type. The goal here is to show that using Lin–c-kit+ stem cells that had been cultured in vitro for 6 days (see Fig. 5.2A) is equivalent

60

Lian and Kumar

Fig. 5.2. Reciprocal radiation chimeras whose donor cells are obtained from cultured stem cells are equivalent to those from total bone marrow preparations. FACS analysis of freshly isolated spleen cells obtained after 2, 4, or 8 weeks after transplantation. Data are shown as percent positive of cells that were gated on the CD3− NK1.1+ lymphocyte population. Insets show the percent splenic NK cells of each chimera examined. (A) A total of 10 × 106 Lin–c-kit+ stem cells purified from whole BM of WT or IL-15R␣KO mice were transduced with “mock” virus supernatant and cultured for 6 days in SCF, IL-7, and Flt3L before transplanting into lethally irradiated recipient mice. In (B), 10 × 106 total BM cells were harvested and immediately transplanted into recipients.

to using freshly isolated whole BM preparations as donor cells (see Fig. 5.2B). These experiments conclusively eliminate the possibility that a failure to reconstitute NK cells is caused by the method of using cultured stem cells as donor cells. Also, to verify that the mutated proteins are effectively and properly expressed in the cell, transduce the cDNA constructs encoding the mutations into purified stem cells (or a cell line deficient for your protein) and assay by flow cytometry (surface protein) or Western Blotting (intracellular protein). For example, in our own experiments, we were able to demonstrate that our transduced cells correctly expressed

Use of Stem Cell Radiation Chimeras

61

Fig. 5.3. FACS analysis of 58−/− cells expressing cloned WT or cytoplasmic mutants of IL-15R␣ (Y227F, 16 and 30). Successful transduction was measured by GFP fluorescence with surface expression of WT and mutant IL-15R␣ also detected by staining with anti-mouse IL-15R␣ mAb.

either wild-type or various cytoplasmic mutant forms of the IL15R␣ gene on the cell surface and at fairly equivalent levels (see Fig. 5.3). 1. 7–10 days before transplantation, begin a regimen of Bactrim suspension (5–250 ml of drinking water) for the intended recipient mice, which in most cases will be the CD45.1 congenic mice. All aspects of drug administration should be approved by your institution’s veterinarian and comply with all guidelines specified by your animal facility. 2. 24 h before transplantation, lethally irradiate (10 Gy) recipient mice. Prepare at least three mice per experimental condition, including three control mice (to receive PBS instead of cells) (see Note 5). 3. On the day of transplantation, harvest donor whole nucleated BM cells (depleted of erythrocytes), which, in most cases will be the gene KO mice expressing the CD45.2 marker. 4. Resuspend the cells in sterile PBS to a concentration of 5 × 107 –1 × 108 cells/ml. 5. Using a tuberculin syringe and 27-gauge needle, inject 100 ␮l of the cell solution into the tail veins of the lethally irradiated recipient mice. The absolute number of cells injected into each mouse will be 5 × 106 –1 × 107 cells. Equivalent numbers should be injected for each

62

Lian and Kumar

experimental condition. Continue with treatment of Bactrim for two more weeks after transplantation. 6. Analyze the chimeric mice 8 weeks post-BM transfer, at which time >96% of the splenic leukocytes are donor derived. 7. For details on analysis, see Section 3.5. 3.4.2. Transplanting BM-Derived Stem Cells

1. Prepare recipient mice as before (see previous section). 2. On the day of transplantation, harvest Lin–c-kit+ stem cells previously isolated from donor mice that have been cultured with SCF, Flt3L, and IL-7 for 5 days, infected with virus particles and enriched for GFP expression (see Note 6). 3. Wash the cells several times with sterile PBS to remove traces of cytokines. 4. Resuspend the cells in sterile PBS to a concentration of 1 × 107 –5 × 107 cells/ml. 5. Inject 100 ␮l of the cell solution into the tail veins of lethally irradiated recipient mice. The absolute number of cells injected into each mouse will be 1 × 106 –5 × 106 cells. Equivalent numbers should be injected for each experimental condition. 6. Analyze the chimeric mice 8 weeks after stem cell transfer, at which time >96% of the splenic leukocytes are donor derived. 7. For details on analysis, see Section 3.5.

3.5. Analyzing the Radiation Chimeras by Flow Cytometry

1. Harvest spleen cells from the chimeric mice and lyse erythrocytes as usual. 2. Pre-incubate nucleated cells with the 2.4G2 (anti-FcR) hybridoma supernatant and aliquot 5 × 105 cells into each analysis tube. 3. Follow the standard protocol of antibody staining for flow cytometry. 4. To each tube of cells, add the following fluorochromeconjugated mAbs: anti-CD3, anti-NK1.1, anti-CD45.2, and one anti-Ly49 mAb. For example, a combination of antiCD3 PE-Cy7, anti-NK1.1 PE, anti-CD45.2 PE-Cy5.5, and anti-Ly49G2 FITC can be used for one of the tubes. One tube of cells should only contain the first three mAbs, and instead of an anti-Ly49 mAb, use an isotype-matched control mAb. 5. For analysis, quantitate Ly49 expression on cells after gating on the population that is CD3− , NK1.1+ , and CD45.2+ (see Note 7). Figure 5.4 shows results that are representative of

Use of Stem Cell Radiation Chimeras

63

Fig. 5.4. FACS analysis of freshly isolated splenocytes harvested from chimeric mice that were transplanted 8 weeks prior with Lin–c-kit+ stem cells transduced with vector alone, wild-type, or mutated forms of IL-15R␣. Data show percent splenic NK cells (inset) and Ly49 expression of cells gated on CD3− NK1.1+ CD45.2+ for each mouse. The results here indicate that IL-15R␣-mediated signals are important for NK cell production as well as the expression of certain Ly49 molecules.

these experiments. Here, the flow cytometric data indicate the importance of the cytoplasmic domains of the IL-15R␣ protein in NK production and Ly49 expression.

4. Notes 1. Harvesting a small sample of the transfected PlatE cells can be easily done by using a 200–1000 ␮l pipetter and sterile tip to draw 500 ␮l of culture medium while scraping across the bottom of the dish. Since these cells are used as a quick and rough measure to survey the extent of transfection, there is no need to count the cells prior to flow cytometric analysis. Just a quick wash in PBS containing 2% FBS and filtration through a 70 ␮m sieve is sufficient preparation for analysis. 2. The efficiency of transduction can sometimes fluctuate and, therefore, result in the production of insufficient numbers of GFP+ progenitor cells for successful transfer into mice. One extremely important factor that should not be overlooked is the “potency” of the virus supernatant for transduction. If supernatant is collected on days when fewer than 70% of transfected PlatE cells are GFP+ , the supernatant may not contain high enough titers of virus particles to be effective.

64

Lian and Kumar

Therefore, transduction efficiency and quality is determined by the concentration of virus particles in the supernatant. Given this, supernatants can be concentrated using special filters. However, three easier and less expensive methods would be (1) to harvest the supernatant when more than 90% of PlatE cells are GFP+ , (2) to enrich the supernatant by incubating the cells for a longer period with medium prior to harvesting, and (3) to enrich the supernatant by reducing the volume of medium added to the PlatE cells following transfection. In terms of quality, freshly harvested supernatants are the most effective, followed by freshly thawed stocks of supernatant. Supernatant stocks kept at 4◦ F will quickly lose their potency. Repeated freezing and thawing of stocks are also not recommended. 3. Three rounds of infection typically yield 50–65% GFP+ cells. However, we have in the past obtained improved results by following a more labor-intensive protocol involving the use of fibronectin (FN) (8, 9). For that method, pre-coat the 24well plates with 20 ␮g of sterile FN (from BD Biosciences) per well and incubate 2–3 h at room temperature. After that, remove the FN solution and add PBS containing 2% BSA to the same wells for 30 min at room temperature to block nonspecific binding sites. Remove the PBS/BSA solution, wash once with DMEM-10 medium, followed by adding 1 ml of virus supernatant stock to each treated well. Incubate 1 h at room temperature. Remember to add 1 ml of “mock” virus supernatant to one or more wells as control. Finally add 1 ml of cells that were pre-washed and resuspended in 2X Differentiation medium containing 20 mM HEPES and 40 ␮g/ml of polybrene. The rest of the protocol is the same as described in Section 3.3.3, step 4. 4. Magnetic-based methods can be used only if the protein of study is a receptor (i.e., expressed on the surface) for which antibodies are available. 5. Lethally irradiated mice that only receive PBS and not BM cells will usually die within 10 days. 6. To obtain the results in Fig. 5.2A, transplant with stem cells previously infected with mock virus supernatant, i.e., contains no virus particles. An even better control is to use cells infected with virus particles expressing the empty pMIGR1GFP vector. 7. Except when performing reciprocal transfers, in most cases the KO mouse-derived donor cells express the CD45.2 marker, and the recipient cells express the CD45.1 marker of the congenic mouse strain. As such, staining the splenocytes with anti-CD45.2 will identify the donor cells co-expressing

Use of Stem Cell Radiation Chimeras

65

your wild-type or mutated protein. Although all the donor cells presumably originated from GFP+ stem cells, the differentiated population has undergone several cycles of division, during which time GFP expression is extensively diluted and often undetectable in the final analysis. References 1. Williams, N. S., Klem, J., Puzanov, I. J., Sivakumar, P. V., Schatzle, J. D., Bennett, M., and Kumar, V. (1998) Natural killer cell differentiation: insights from knockout and transgenic mouse models and in vitro systems Immunol Rev 165, 47–61. 2. Lian, R. H., and Kumar, V. (2002) Murine natural killer cell progenitors and their requirements for development Semin Immunol 14, 453–60. 3. Williams, N. S., Moore, T. A., Schatzle, J. D., Puzanov, I. J., Sivakumar, P. V., Zlotnik, A., Bennett, M., and Kumar, V. (1997) Generation of lytic natural killer 1.1+, Ly49- cells from multipotential murine bone marrow progenitors in a stroma-free culture: definition of cytokine requirements and developmental intermediates J Exp Med 186, 1609–14. 4. Williams, N. S., Klem, J., Puzanov, I. J., Sivakumar, P. V., Bennett, M., and Kumar, V. (1999) Differentation of NK1.1+, Ly49+ NK cells from flt3+ multipotent marrow progenitor cells J Immunol 163, 2648–56.

5. Dorfman, J. R., and Raulet, D. H. (1998) Acquisition of Ly49 receptor expression by developing natural killer cells J Exp Med 187, 609–18. 6. Roth, C., Carlyle, J. R., Takizawa, H., and Raulet, D. H. (2000) Clonal acquisition of inhibitory Ly49 receptors on developing NK cells is successively restricted and regulated by stromal class I MHC Immunity 13, 143–53. 7. Morita, S., Kojima T., and Kitamura, T. (2000) Plat-E: an efficient and stable system for transient packaging of retroviruses Gene Ther 7, 1063–6. 8. Moritz, T., Dutt, P., Xiao, X., Carstanjen, D., Vik, T., Hanenberg, H., and Williams, D. A. (1996) Fibronectin improves transduction of reconstituting hematopoietic stem cells by retroviral vectors: evidence of direct viral binding to chymotryptic carboxy-terminal fragments Blood 88, 855–62. 9. Bajaj, B., Behshad, S., and Andreadis, S. T. (2002) Retroviral gene transfer to human epidermal keratinocytes correlates with integrin expression and is significantly enhanced on fibronectin Hum Gene Ther 13, 1821–31.

Chapter 6 Use of Transfected Drosophila S2 Cells to Study NK Cell Activation Michael E. March, Catharina C. Gross, and Eric O. Long Abstract Determining the contribution of individual receptors to natural killer (NK) cell function is complicated by the multiplicity of activating and inhibitory NK cell receptors. Mammalian target cells typically express a variety of ligands for NK cell receptors. Engagement of NK cell receptors by antibodies may not mimic activation by natural ligands. To define requirements for activation and dissect the contribution of receptors to NK cell function, we have generated Drosophila Schneider line 2 (S2) cell transfectants expressing ligands for NK cell receptors. The evolutionary distance between Drosophila and mammals greatly reduces the potential of recognition of insect cell molecules by mammalian NK cells. Here, we present methods for maintenance and transfection of S2 cells, as well as protocols for their use in NK cell assays. Key words: Drosophila, S2, insect cells, NK cells, activation, inhibition, transfection.

1. Introduction The activation of natural killer (NK) cells is controlled by combinations of activating and inhibitory receptors expressed by NK cells. Integration of the signals downstream from a wide variety of receptors determines whether or not the NK cells respond (1– 4). Higher expression of activating ligands on target cells induces activation of NK cells, whereas higher expression of inhibitory ligands prevents it. Human NK cells express a wide variety of activating receptors, which respond to a wide variety of ligands. The natural cytotoxicity receptors (NCRs) NKp46, NKp44, and NKp30 and receptors NKG2D, 2B4 (CD244), CD2, LFA1 (CD11a/CD18), and DNAM-1 (CD226) are some of the K.S. Campbell (ed.), Natural Killer Cell Protocols, Methods in Molecular Biology 612, DOI 10.1007/978-1-60761-362-6 6, © Humana Press, a part of Springer Science+Business Media, LLC 2010

67

68

March et al.

receptors known to deliver activating signals to NK cells. Engagement of combinations of these receptors can activate NK cell functions such as adhesion, cytotoxicity, and cytokine secretion (5–7). NK cells are also potent mediators of antibody-dependent cellular cytotoxicity (ADCC) through the low-affinity Fc ␥RIII (CD16). Inhibitory receptors are expressed on NK cells in a clonally distributed manner, where individual NK cells may express single or multiple inhibitory receptors in an essentially random manner. Inhibitory receptors in humans include members of the killer cell Ig-like receptor (KIR) family, which recognize alleles of HLA class I, members of the CD94/NKG2 family that recognize HLA-E, and several other inhibitory receptors that bind to non-MHC ligands (4, 8). The response of NK cells to signals from individual receptors, or combinations of receptors, is often studied by masking the receptors with antibodies in order to block a response to a given target cell (9–13) or by cross-linking the receptors on the surface of NK cells with antibodies in order to provoke a response in the absence of target cells (7, 14). Such experiments can be very informative; however, antibodies typically have much higher affinities for receptors than the natural ligands do and may fail to stimulate those receptors in the physiological way. This is particularly true for receptors such as integrins that undergo conformational changes upon ligand binding (15, 16), which may not occur following antibody cross-linking. Cell lines that are susceptible or resistant to NK cell cytotoxicity are also used to probe NK cell responses. Ectopic expression of ligands in cell lines has revealed much about the contributions of both activating and inhibitory receptors to the regulation of NK cells (9, 13, 17, 18). Interpretation of such experiments can be complicated by the expression of many, often uncharacterized, ligands on target cell lines. For example, the EBV-transformed B cell line 721.221 expresses ICAM-1 (CD54), ICAM-2 (CD102), and ICAM-3 (CD50) (ligands for the integrin LFA-1), CD48 (ligand for 2B4), and B7 (CD80, a ligand of CD28) (19). The erythroblastoid cell line K562 expresses ULBP2 and MICA/B (ligands for NKG2D) (20, 21) along with ICAM-1 and ICAM-2 (22). Additionally, the Fc␥R+ mouse cell line P815, which is often used to stimulate NK cells with antibodies via redirected lysis (10, 11), expresses the mouse ICAM-1, which binds to human LFA-1 (23). In addition, P815 cells are more sensitive to lysis by NK cells that express high levels of NKp46, suggesting that they may express ligands for this NCR (24). Ligands for the NCRs have not been fully characterized, and the pattern of expression on many NK target cell lines is unknown. These uncertainties created a need for a target cell line in which expression of ligands is more easily characterized and controlled. A cell line derived from an evolutionarily distant species,

Transfected Insect Cells as Targets of NK Cells

69

which would also be easy to transfect and maintain, would be a useful surrogate target cell. Schneider line 2 (S2) cells fulfill these requirements. The S2 cell line was derived from latestage embryos of Drosophila melanogaster (25). The cells grow in a loose monolayer. S2 cells are easily transfected. They can be adapted to growth in serum-free medium, suitable for the purification of recombinant secreted proteins. S2 cells have been used to produce and purify proteins in sufficient quantities for structural and biochemical studies (26). Expression of exogenous proteins is often accomplished by transfection with the pRmHa3 plasmid, which uses the Drosophila metallothionein gene promoter for inducible expression, allowing high expression even for proteins that prove deleterious to the growth of S2 cells (27). Transfected S2 cells have been used previously to investigate the minimal requirements for T-cell activation and costimulation. S2 cells expressing peptide-loaded MHC class I, either alone or in combination with ICAM-1 and B7, were used as antigen-presenting cells (APCs) for the stimulation of na¨ıve T cells expressing the 2C transgenic TCR (28, 29). The results demonstrated that signaling through the TCR alone is not sufficient to activate na¨ıve CD8+ T cells and that B7 and ICAM-1 could provide the required co-stimulation for activation. Optimal co-stimulation occurred with both B7 and ICAM-1 expressed on the same MHC class I+ S2 cell. Besides confirming the two-signal hypothesis for T-cell activation, these data also demonstrated that co-engagement of the TCR with a co-stimulatory receptor is sufficient for T-cell activation. The use of S2 cells as APCs for T cells indicated that this Drosophila cell line could be used to reconstitute a sensitive target cell for NK cells. We have successfully used S2 cells to investigate the response of primary human NK cells to stimulation and inhibition through individual receptors (5, 30, 31). A notable advantage of this system is that resting NK cells, freshly isolated from human blood, can be used directly in functional assays with S2 cell transfectants, without further manipulation. In most assays examining degranulation or cytokine secretion, untransfected S2 cells induce minimal or no response from either primary NK cells or NK cell lines. Expression of individual ligands by stable transfection is sufficient to induce responses such as adhesion (5) and granule polarization (30, 31), whereas multiple ligands may be required to induce other responses such as degranulation (Y. Bryceson, unpublished). Transmembrane proteins expressed in S2 cells are glycosylated, which may be important for recognition by some NK cell receptors. S2 cells are easily transfected, either transiently or stably, and can be substituted for mammalian target cells in many NK cell assays with little or no modification. We will present here methods for stable transfection of S2 cells, methods for isolation of cells expressing the transfected protein,

70

March et al.

as well as specific protocols for the use of S2 cells in assays for NK cell function.

2. Materials 2.1. Care and Culture of S2 Cells 2.1.1. Standard Culture

1. S2 culture medium: Schneider’s Drosophila medium (Invitrogen, Carlsbad, CA) supplemented with 10% fetal bovine serum (FBS) (Hyclone, Logan, UT; complement inactivated by heating at 56◦ C for 30 min). 2. 75 cm2 tissue culture flasks (Costar, Corning, Lowell, MA). 3. Freezing medium: Fresh S2 cell culture medium mixed with an equal volume of conditioned medium (the medium in which S2 cells have been grown) supplemented with 10% dimethyl sulfoxide (DMSO) (Sigma Aldrich, St. Louis, MO). Alternatively, freezing medium can be composed of 90% FBS supplemented with 10% DMSO. 4. Reagent for inducible expression: Expression from metallothionein promoter-based plasmids is induced by the addition of cupric sulfate (CuSO4 , Sigma Aldrich) to the cell culture to a final concentration of 1 mM. A 100 mM CuSO4 stock solution is made in deionized water and then sterilized by filtration through a 0.45 ␮m filter.

2.1.2. Adaptation of S2 Cells to Culture in Serum-Free Conditions

1. Bio Whittaker Insect-Xpress medium (Lonza, Basel Switzerland). 2. Iscove’s Modified Dulbecco’s Medium (Invitrogen) supplemented with 0.2% BSA (see Note 1). 3. Fetal Bovine Serum (Hyclone). 4. 75 cm2 tissue culture flasks (Costar, Corning). 5. 125 ml Disposable Erlenmeyer Flask with 0.2 ␮m PTFE Vented Closure (Bellco Glass Inc, Vineland, NJ). 6. Orbital shaker capable of handling 125 ml flasks, at 27◦ C.

2.2. Transfection of S2 Cells 2.2.1. Calcium Phosphate Transfection

1. HEPES buffered saline (HBS) for calcium phosphate transfection: 50 mM HEPES, 1.5 mM Na2 HPO4 , and 280 mM NaCl, in water, adjusted to pH 7.1, sterilized by filtration through a 0.45 ␮m filter. 2. CaCl2 for calcium phosphate transfection: 2 M CaCl2 (Sigma Aldrich) in water, sterilized by filtration through a 0.45 ␮m filter. 3. 6-well tissue culture plates (Costar, Corning).

Transfected Insect Cells as Targets of NK Cells

71

4. pAc5.1-V5-His: Plasmid for constitutive expression in S2 cells using the Drosophila Actin 5 promoter (requires a licensing agreement with Invitrogen). 5. pRmHa3: Plasmid for inducible expression in S2 cells using the metallothionein promoter (available to not-for-profit labs from the Drosophila Genome Research Center, Bloomington, IN). 6. Plasmids for selection of transfectants, which include pNeoFly (27), pCoBlast (Invitrogen), and pCoHygro (Invitrogen). 7. Agents for selection of stable transfectants: Geneticin (G418 sulfate, Invitrogen), Puromycin dihydrochloride (Sigma Aldrich), Hygromycin B (Roche Diagnostics GmbH, Basel Switzerland), Blasticidin S HCl (Invitrogen). 8. 6 ml polypropylene, round bottom tubes, sterile (Falcon, BD Biosciences, San Jose, CA). 9. 15 ml polystyrene conical screw cap tubes (Sarstedt, Newton, NC). 2.2.2. Transfection with Cellfectin

1. Schneider’s Drosophila medium (Invitrogen) ± 10% FBS. 2. Cellfectin Transfection Reagent (Invitrogen). 3. Expression and drug resistance plasmids and agents for selection of stable transfectants (see Section 2.2.1). 4. 6-well tissue culture plates (Costar, Corning). 5. 6 ml polypropylene, round bottom tubes, sterile (Falcon, BD Biosciences).

2.2.3. Enrichment of Expressing Cells

1. Antibodies for the transfected proteins. For fluorescenceactivated cell sorting (FACS), antibodies directly coupled to fluorescent labels may be used when available. Alternatively, unlabeled primary antibodies can be stained with fluorescently labeled secondary antibodies. For enrichment of cells expressing a given protein using magnetic beads coated with goat anti-mouse antibodies, unlabeled primary antibodies are recommended. 2. Buffer for FACS sort: PBS supplemented with 2% heatinactivated FBS, 2 mM EDTA, 100 U/ml Penicillin, and 100 ␮g/ml Streptomycin. 3. FACS tubes: 5 ml polystyrene tubes, 12 × 75 mm (Falcon, BD Biosciences). 4. Dynabeads Goat anti-mouse IgG and selection magnet (Invitrogen). 5. 14 ml Polypropylene Round Bottom Tubes (Falcon, BD Biosciences).

72

March et al.

6. Flat bottomed 96-well plates (Costar, Corning). 2.3. S2 Cells as Targets in NK Cell Assays

1. Anti-CD56 mAb (clone NCAM 16.2) PE (BD Bioscience).

2.3.1. CD107a Degranulation Assay

1. PKH26 Red Fluorescent Cell Linker Kit (Sigma Aldrich).

2.3.2. Conjugate Assay

3. Hank’s Buffered Saline Solution (Invitrogen) supplemented with 5% FBS.

2. Anti-CD107a mAb (clone H4A3) FITC (BD Bioscience).

2. CellTracker Green CMFDA (Invitrogen).

4. IMDM + 10% FBS. 5. FACS tubes: 5 ml polystyrene tubes, 12 × 75 mm (BD Biosciences). 6. 0.5% Paraformaldehye (16% solution from Electron Microscopy Sciences, Hatfield, PA, diluted in phosphatebuffered saline (PBS)). 2.3.3. Perforin Polarization Assay

1. 15 ml polystyrene conical screw cap tubes (Sarstedt). 2. Hank’s Buffered Saline Solution (HBSS) (Invitrogen) supplemented with 3% FBS. 3. Poly-D-Lysine coated 2-well Culture Slides (BD BioCoat, BD Biosciences). 4. 4% Paraformaldehye (16% solution Microscopy Sciences diluted in PBS).

from

Electron

5. Normal donkey serum (Jackson ImmunoResearch Laboratories, West Grove, PA) reconstituted in 10 ml water. 6. Triton X-100 (Sigma Aldrich). 7. Permeablization buffer: 10% normal donkey serum and 0.5% Triton X-100 in PBS. 8. Staining buffer: 3% normal donkey serum and 0.5% Triton X-100 in PBS. 9. CellTracker Green CMFDA (Invitrogen). 10. Anti-perforin antibody, clone ␦G9 (Thermo Scientific, Rockford, IL). 11. Alexa Fluor 568 conjugated goat anti-mouse secondary antibody (Invitrogen). 12. Prolong Gold antifade reagent (Invitrogen). 2.3.4. Flow Cytometry-Based Cytotoxicity Assay

1. PKH67 Green Fluorescent Cell Linker Kit (Sigma Aldrich). 2. 15 ml polystyrene conical screw cap tubes (Sarstedt). 3. Iscove’s Modified Dulbecco’s Medium with and 25 mM HEPES (Invitrogen).

L -Glutamine

4. Propidium iodide solution, 1 mg/ml (Sigma Aldrich).

Transfected Insect Cells as Targets of NK Cells

73

5. FACS tubes: 5 ml polystyrene tubes, 12 × 75 mm (BD Biosciences)

3. Methods 3.1. Care and Culture of S2 Cells 3.1.1. Standard Culture

1. S2 cells grow at or slightly above room temperature. Additional carbon dioxide (CO2 ) is not required. Incubators, without CO2 injection, can be used to maintain an optimal temperature of approximately 27◦ C. Alternatively, S2 cells will grow reasonably well in a drawer or any location at room temperature that is protected from light. 2. S2 cells should be thawed at room temperature. Since S2 cells undergo a heat shock at 37◦ C, thawing the cells in a warm water bath will reduce viability. 3. Cell cultures should be maintained between 1 × 106 and 2.5 × 106 cells/ml in Schneider cell medium supplemented with 10% FBS. Cells grow best when diluted no more than 1:3 at each passage. Maintenance of a healthy culture of growing cells requires a 1:2 passage every other day (see Note 2). 4. Freezing of S2 cells is most successful when large numbers of cells from a healthy culture are frozen, such as 20–30 million cells per vial. S2 cells can be frozen in a variety of freezing media. We have used mainly 45% fresh culture medium combined with 45% conditioned S2 cell medium (the medium in which the cells have been grown) supplemented with 10% DMSO. Alternatively, a simple mixture of 90% FBS supplemented with 10% DMSO has been used. 5. Proteins can be expressed by transfection with either a constitutive or an inducible promoter-containing plasmid. Plasmid pAc5.1 (Invitrogen) drives expression with a Drosophila Actin 5 promoter, which results in constitutive expression. Plasmid pRmHa3 uses a metallothionein promoter, the activity of which is induced with 1 mM CuSO4 . A total of 48 h of induction with 1 mM CuSO4 is sufficient to yield maximal expression of transfected proteins. A fraction of the cell culture is reserved, without induction, for continued culture. Constitutive expression is more convenient, as the transfected proteins are always expressed and variability introduced by induction is avoided. Expression from a constitutive promoter is often more uniform over time than induced expression, which often yields heterogeneous expression levels. Constitutive expression can also be beneficial in transient transfections, where cells already dealing with the stress of the transfection may be adversely affected

74

March et al.

even further by the induction in cupric sulfate. Inducible expression is advantageous when a transfected protein is toxic to the S2 cells. Induced expression is also often used for large-scale protein production in S2 cells, where constitutive expression may slow the expansion of the cultures. The cells can be expanded without expressing the protein, and expression can be induced in 24–48 h. 6. Co-expression of multiple proteins in transfected S2 cells can be achieved either by co-transfection of all the required plasmids at once or by sequential transfections (see Notes 3 and 4). 7. Expression of MHC class I molecules in S2 cells requires co-transfection with ␤2 -microglobulin. Furthermore, exogenous peptides have to be loaded onto MHC class I because insect cells do not have the endogenous peptide loading machinery (e.g., TAP transporter, chaperones) (32). Peptide loading is very efficient because “empty” class I–␤2 microglobulin dimers are readily expressed at the surface of S2 cells at 27◦ C. Appropriate peptides are added to the S2 cell culture at a final concentration of 1 ␮m 1 day prior to the assay with NK cells. 8. Using S2 cells in NK cell assays is often a matter of simply replacing the conventional target cells in the assay with S2 cells. However, there are two important considerations. First, insect cells undergo heat shock and will eventually die at 37◦ C. Second, S2 cells exposed to human serum that has not been heat inactivated are killed in less than 1 h. Each of those issues can be addressed as follows. The viability of S2 cells following incubation at 37◦ C is not significantly reduced for at least 3 h. The assays described here using S2 cells at 37◦ C do not exceed 3 h. S2 cells grow happily at 27◦ C in heat-inactivated bovine serum. However, our primary human NK cells are usually cultured in human serum, which is not routinely heat inactivated. To avoid exposure of S2 cells to human serum, the simplest approach is to wash and resuspend NK cells in medium containing FBS prior to the assays with S2 cells. Although, in our hands, primary human NK cells survive and proliferate best in medium containing 10% human serum, they remain functional in medium containing FBS in short-term assays (see Note 1). 3.1.2. Adaptation of S2 Cells to Culture in Serum-Free Conditions

S2 cells used for production of proteins are often cultured in serum-free conditions in order to simplify the purification of the protein of interest. Serum-free culture of S2 cells has an additional advantage, of interest when using S2 cells as target cells in NK cell assays. Drosophila does not synthesize sterols. Its sole source of cholesterol is from the diet. Culture of S2 cells in serum-free

Transfected Insect Cells as Targets of NK Cells

75

medium is therefore a method to deplete the cells of cholesterol without using compounds such as ␤-methylcyclodextrin. Therefore, S2 cells grown in serum-free conditions may be a useful tool to investigate the role of cholesterol-dependent membrane microdomains on the behavior of ligands of NK cell receptors. A previous report indicated that the inclusion of MICA in lipid rafts on target cells is necessary for the NKG2D-dependent response of NK cells (33). 1. S2 cells to be adapted should be thawed from frozen stocks and cultured in 75 cm2 flasks in Insect-Xpress medium supplemented with 5% FBS. Cells should be expanded until a healthy growing culture at 2 × 106 cells/ml is obtained. 2. Pellet 150 × 106 cells at 135g. Resuspend in 30 ml InsectXpress + 5% FBS and transfer the cell suspension to a 125 ml disposable Erlenmeyer Flask. 3. Expand the cells by culturing on a shaker at 140 rpm at 27◦ C for 2–3 days. 4. Pellet 150 × 106 cells at 135g. Resuspend in 30 ml InsectXpress + 2.5% FBS and transfer the cell suspension to a 125 ml disposable Erlenmeyer Flask. 5. Again, expand the cells by culturing on a shaker at 140 rpm at 27◦ C for 2–3 days. 6. Perform three more cycles of pelleting 150 × 106 cells, resuspending in Insect-Xpress medium, and expanding the cells in shaking culture for 2–3 days, using medium that contains 1% FBS, then 0.5% FBS, and finally no serum. 7. Cells should be cultured in serum-free medium for a further 2 weeks before use in an assay (see Note 1). 3.2. Transfection of S2 Cells

For most NK cell assays, it is preferable to use S2 cells that express ligands at a fairly uniform level. To this end, stable transfectants of S2 cells are usually generated, from which the desired expressing cells can be enriched or cloned, when necessary. Two protocols for the generation of stable S2 cell lines are presented here. In some cases, transient transfections of S2 cells can be used. Transient expression is advantageous to test expression from new plasmids, for preliminary functional assays (where appropriate), and for the rapid screening of a very large number of transfectants. However, there is less control over protein expression levels in transient transfections, and there will usually be a sizeable fraction of the S2 cells that do not express the transfected proteins (see Note 5).

3.2.1. Calcium Phosphate Transfection

1. For transfection, S2 cells should be growing logarithmically, at a density around 2 × 106 /ml, with high viability. Cells can be counted on a hemacytometer with trypan blue staining.

76

March et al.

2. The day before transfection, spin down 3 × 106 cells per transfection in a 15 ml conical tube at 215g. Resuspend the cells at 1 × 106 cells/ml and plate 3 ml (3 × 106 cells) in a well of a 6-well plate. Incubate overnight at 27◦ C. 3. Set up the transfection mix. In one 6 ml tube, aliquot 300 ␮l of HBS. In a second 6 ml tube, mix 36 ␮l of 2 M CaCl2 with 19 ␮g of sterile DNA and sterile water to a total volume of 300 ␮l. This mixture should contain 18 ␮g of the plasmid expressing the protein of interest and 1 ␮g of the desired selection plasmid (see Notes 6 and 7). Vortex the tube containing the HBS at a moderate setting. While vortexing, slowly add the CaCl2 /DNA mixture dropwise to the HBS. This process should take approximately 1 min. Incubate the mixture at room temperature for 30 min. 4. Add the DNA mixture to the S2 cells that had been plated the day before. The mixture should be added to the medium on the cells in a dropwise manner. Gently swirl the plate in order to mix the medium and DNA. Incubate the cells overnight with the transfection mixture at 27◦ C. 5. Pipet the cells into a 15 ml conical tube. Wash the well with 5 ml of S2 medium in order to recover adherent cells. Pellet the cells at 215g and wash the cells once with 10 ml of medium. Plate the cells into a fresh well in a 6-well plate in 3 ml medium and incubate for 48 h at 27◦ C. 6. Add the appropriate selection (see Note 8). Incubate the cells for up to 2 weeks, until the resistant cells expand through the selection. The drug-resistant cells should be assayed for protein expression by flow cytometry. 3.2.2. Cellfectin Transfection

1. The day before the transfection, plate 1 × 106 cells into a well of a 6-well plate in 3 ml of Schneider’s Drosophila medium + 10% FBS. Culture the cells overnight. 2. Prepare the transfection mixture. a. In a 6 ml polypropylene tube, prepare a mixture consisting of 0.5 ␮g of the drug resistance plasmid and 5 ␮g of each protein expression plasmid in 100 ␮l of serum-free Schneider’s Drosophila medium (see Notes 6 and 7). The DNA should be sterile. b. In a second tube, add 10 ␮l Cellfectin Reagent to 90 ␮l serum-free Schneider’s Drosophila medium. c. Add the Cellfectin-containing medium to the tube containing the DNA mixture. Mix gently. d. Incubate the combined mixture at room temperature for 15 min.

Transfected Insect Cells as Targets of NK Cells

77

e. Add 800 ␮l serum-free Schneider’s Drosophila medium to the mixture. 3. Gently aspirate the medium from the S2 cells. Carefully wash the cells once with 2 ml serum-free Schneider’s Drosophila medium. Aspirate the wash medium. 4. Add the Cellfectin/DNA mixture to the cells. Incubate the cells for 18 h at 27◦ C. 5. Replace the DNA-containing medium with Schneider’s Drosophila medium + 10% FBS. Incubate for 48 h at 27◦ C. 6. At this point, the cells can be analyzed by flow cytometry for the expression of proteins from constitutively expressing plasmids, and protein expression from inducible plasmids can be induced with 1 mM CuSO4 and examined after another 48 h. 7. To select stable transfectants, the cells from Step 5 should be harvested and collected by centrifugation and resuspended in 3 ml Schneider’s Drosophila medium + 10% FBS containing the required selection reagent (see Note 8). Cells can be replated into the same well of the 6-well plate in which the transfection was performed. The cells should be incubated at 27◦ C, and the medium exchanged for fresh growth medium every 7 days, until stably transfected cells begin to expand. 3.2.3. Enrichment of Expressing Cells

Cells that have been selected through drug resistance often contain only a subpopulation of cells that express the protein(s) of interest. Although this amount of expression is sometimes sufficient for use in certain assays, it is usually preferable to obtain a more uniform profile of expression. 1. Expressing cells can be enriched with magnetic beads. We have had success with Dynabeads from Invitrogen, simply following the manufacturer’s protocol. This type of enrichment works best with relatively high levels of expression. The S2 cells must be stained with a saturating amount of primary antibody and must be kept cold (preferably on ice) for the duration of the procedure. S2 cells are phagocytic. If the S2 cells are allowed to phagocytose magnetic beads at room temperature, the enrichment for expression will be contaminated with non-expressing cells. 2. Expressing cells can be isolated by FACS. Again, it is critical to obtain the brightest antibody staining possible to ensure the removal of non-expressing cells. When expression of a protein is marginal, it is recommended that only the brightest cells are selected. 3. After enrichment of expressing cells by magnetic beads or FACS, it is necessary to regularly monitor expression of the transfected proteins. We have found that these enriched pop-

78

March et al.

ulations tend to lose expression during culture, presumable due to selective expansion of cells with lower expression. More stable, uniform expression can be obtained by deriving clones from the resistant populations, immediately following transfection or after enrichment by beads or sorting. To clone S2 cell transfectants, plate cells in medium with selection in 96-well plates at decreasing cell concentrations beginning at 10 cells per well. Threefold dilutions of the cells down to a concentration of 0.3 cell per well can be plated. Proliferating clones will usually be visible within 2 weeks. Clones should be expanded into larger volumes of medium until sufficient cells are available to screen by flow cytometry. Multiple clones, often with varying amounts of protein expression, should be selected for experimentation (see Note 9). 3.3. S2 Cells as Targets in NK Cell Assays

The usefulness of the S2 cells relies on the ability to use them in a variety of assays investigating NK cell function. We have successfully performed a variety of flow cytometry- and microscopybased assays, using the S2 cells as target cells to investigate responses of human NK cells to a variety of ligands for NK cell receptors. We have used S2 cells to monitor NK cell degranulation as measured by the surface expression of lysosome-associated membrane protein 1 (LAMP-1, also called CD107a) (31), binding of NK cells to target cells in a conjugation assay (5), clustering of activating and inhibitory receptors at NK–S2 cell synapses (34, 35), signaling induced in NK cells by contact with S2 cells (36), and the intracellular polarization of perforin-containing granules to the site of contact with target cells (31). We have also used S2 cells as target cells in a modified cytotoxicity assay. Of particular note, we have generated, through Anaspec (San Jose, CA), a rabbit polyclonal antiserum against an S2 cell membrane preparation. Addition of this serum to S2 cells provides a potent, physiological stimulus to human NK cells for ADCC through CD16 (31). Untransfected S2 cells provide negative controls for each of these assays.

3.3.1. CD107a Assay

The detection of NK cell degranulation through monitoring of the surface expression of CD107a has rapidly become the method of choice. This assay works well with S2 target cells. For a thorough description of the CD107a assay, please refer to the chapter in this issue by Bryceson et al., as well as our previous work (31). Due to a relatively high auto-fluorescence of S2 cells, it is important to obtain a bright staining of NK cells with the CD56 antibody in order obtain a stringent gate for NK cells. This may require a two-step staining protocol, using a secondary antibody against the CD56 antibody. Note that the E:T ratio used with S2 cells is 1:5 instead the 1:2 used for other target cells.

Transfected Insect Cells as Targets of NK Cells

3.3.2. Conjugation Assay

79

1. This assay has been adapted from Burshtyn et al. (37) (see Chapter 7 in this volume), which is also described for use with mammalian target cells in a chapter in this issue. If necessary, induce expression in S2 cells transfected with a metallothionein-based plasmid with 1 mM CuSO4 48 h before the conjugation assay. If the S2 cells express MHC molecules, those molecules will need to be loaded with a compatible peptide. Peptides should be added to the cells 1 day before the assay to a final concentration of 1 ␮M. Expression of ligands on the S2 cells should always be monitored by flow cytometry on the day of the assay (see Note 10). 2. Each combination of NK cell and target cell should be assayed for conjugation at a series of time points. We routinely perform 0, 5, 10, 20, and 40 min time points. With S2 cell targets, each point in the assay will include 1 × 105 NK cells and 4 × 105 target cells, for a final E:T ratio of 1:4. In the following steps, be sure to label sufficient cells for the assay. 3. Label the S2 cells with the red PHK26 dye, according to the manufacturer’s protocol, summarized here (see Note 11). a. Pellet the required number of S2 cells by centrifugation at 300g. b. Count the cells. Use about 5 × 106 for labeling (scale up as necessary); put 5 × 106 cells into a new tube and bring them up to 10 ml total volume with serum-free medium. Pellet the cells again. c. While the cells are in the centrifuge, prepare the fluorescent dye. For labeling, you need 100 ␮l of “2× dye” for every 106 cells; 2× dye for 5 × 106 cells is 588 ml of Diluent C (supplied with the labeling kit) + 12 ␮l of PKH26. d. Aspirate the medium from the cells. Resuspend the target cells in 100 ␮l of Diluent C for every 106 cells. Add the 2× dye to the cells, 100 ␮l for each 1 × 106 cells. e. Incubate 5 min at room temperature. Add an equal volume of serum to the cells and incubate 1 min at room temperature. f. Wash the cells twice with Schneider cell medium + 10% FBS. After the second wash, resuspend the cells in 5 ml of medium + 10% FBS. Incubate the cells at 27◦ C for 2 h. 4. Label the NK cells with Celltracker Green. The cells should be incubated, in their standard medium at a concentration of 10 × 106 cells/ml, with 1 ␮g/ml of the Celltracker Green for 30 min. The cells should be washed twice with their standard medium and then allowed to rest in 10 ml of medium in a 15 ml tube for 30 min at 37◦ C.

80

March et al.

5. Wash both the NK and the S2 cells twice with HBSS + 5% FBS. Resuspend the cells in HBSS with 5% FBS and count them. Pellet the cells and resuspend the NK cells at a concentration of 1 × 106 cells/ml and the S2 cells at a concentration of 4 × 106 cells/ml using ice cold HBSS with 5% FBS. 6. To a series of FACS tubes (one tube for each time point), add 1 × 105 NK cells and 4 × 105 of the appropriate S2 cells (100 ␮l each), for an E:T ratio of 1:4. Vortex the cells quickly to mix. 7. Pellet the cells, at 4◦ C, for 3 min at 20g. 8. Incubate the tubes in a 37◦ C water bath for the desired lengths of time. When each time point is reached, vortex the appropriate tubes and then fix the cells by adding 1 ml of 0.5% paraformaldehyde. 9. Acquire samples on a flow cytometer, using stained NK and S2 cells alone as compensation controls. Conjugate pairs are represented by green and red double-stained events. Results are typically represented as the percentage of all NK cells that are in conjugates relative to the total number of NK cells. 3.3.3. Perforin Polarization Assay

NK cells respond to contact with sensitive target cells by polarizing their perforin-containing lytic granules to the site of contact with the target cell. This polarization can be detected through fluorescent microscopy. Certain S2 cell transfectants have been shown to induce polarization of granules in human NK cells (30, 31). 1. NK cells and S2 cells will ultimately be mixed at a 1:1 E:T ratio, with 1 × 106 of each cell type per sample. The required number of each transfected S2 cell should be labeled, in standard medium at a concentration of 10 × 106 cells/ml, with 1 ␮g/ml of Celltracker Green for 30 min. The cells should be washed twice in standard medium and then allowed to rest in 1 ml of medium in a 15 ml tube for 30 min. Wash twice with 10 ml HBSS + 3% FBS. Resuspend at 10 × 106 cells/ml. 2. Pellet the required number of NK cells. Wash twice with 10 ml HBSS + 3% FBS. Resuspend at 10 × 106 cells/ml. 3. In a 15 ml conical tube, combine 1 × 106 NK cells and 1 × 106 target cells (100 ␮l each). Vortex briefly to mix the cells. 4. Pellet the cells at 20g for 3 min. 5. Incubate in a 37◦ C water bath for 20 min. 6. Transfer the cells to poly-D-lysine coated culture slides. Resuspend the cells gently to preserve cell:cell contacts.

Transfected Insect Cells as Targets of NK Cells

81

Incubate the cells on the slides for 60 min at room temperature. 7. Aspirate the medium. Wash the cells gently twice with 1 ml PBS. Fix the cells by adding 1 ml 4% paraformaldehyde. Incubate for 20 min at room temperature. 8. Wash the fixed cells twice with PBS. 9. Permeablize the cells with 0.5 ml of 10% normal donkey serum (NDS) and 0.5% Triton X-100 in PBS. Incubate at room temperature for 30 min with gentle shaking. 10. Stain the cells with 3 ␮g/ml anti-perforin antibody (clone ␦G9) in 3% NDS, 0.5% Triton X-100 in PBS. Incubate for 1 h at room temperature with gentle shaking. 11. Wash three times, for 5 min each, with 1 ml PBS. 12. Stain with a fluorescently labeled goat anti-mouse secondary antibody in 3% NDS and 0.5% Triton X-100 in PBS. We typically use an Alexa Fluor 568 conjugated secondary antibody at a concentration of 1 ␮g/ml. Incubate for 1 h at room temperature with gentle shaking. 13. Wash three times, for 5 min each, with 1 ml PBS. 14. Aspirate the PBS and allow the slides to air dry. 15. Seal the samples on slides under a coverslip, using a drop of Prolong Gold antifade reagent as a mounting medium. 16. The cells should be imaged with a fluorescent microscope. We have acquired images on a Zeiss LSM510 confocal microscope. Z-stacks are acquired and used to create 3dimensional reconstructions of the cell contacts. Contacts between a perforin-containing NK cell and a Celltracker Green labeled target cell are identified, and the perforin is scored as polarized or not polarized by the visual appearance of the cell:cell conjugate. In most experiments, NK cells in which granules have polarized are obvious. Large numbers of cell contacts must be analyzed for rigorous analysis. In order to acquire sufficient data, acquire microscope images under relatively low magnification (e.g., multiple 40× fields). We typically analyze 150–200 contacts for each transfected S2 cell analyzed. 3.3.4. Cytotoxicity Assay by Flow Cytometry

One situation where S2 cells cannot be simply substituted for conventional target cells is cytotoxicity assays. Most cytotoxicity assays require that target cells be loaded with a radioactive or fluorescent vital label, like chromium-51, Europium, or calcein. Lysis of target cells is monitored by the release of the vital label into the medium. S2 cells label poorly with these reagents, and they have a high rate of spontaneous release. These factors combine to leave only a very small window between spontaneous

82

March et al.

and maximal release, making it very difficulty to detect specific NK cell-mediated cytotoxicity. Killing of S2 cells is therefore best measured using a flow cytometry-based assay, in which the S2 cells are labeled with a green lipophilic dye, mixed with NK cells, and then stained with propidium iodide to detect cytotoxicity. 1. If necessary, induce expression in S2 cells transfected with a metallothionein-based plasmid with 1 mM CuSO4 48 h before the cytotoxicity assay. If the S2 cells express MHC molecules, those molecules will need to be loaded with a compatible peptide. Peptides should be added to the cells the day before the assay to a final concentration of 1 ␮M. Expression of ligands on the S2 cells should always be monitored by flow cytometry on the day of the cytotoxicity assay (see Note 10). Untransfected S2 cells must be included as a control in the assay (see Note 12). 2. Label S2 cells with the green PKH67 dye, according to the manufacturer’s protocol, as described in Step 3 of Section 3.3.2 (see Note 11). 3. During the incubation, spin down the NK cells and wash with Iscove’s medium containing 10% FBS. The number of NK cells required will depend on the effector to target ratios being used. Each point requires 20,000 targets cells and is done in duplicate. Typically, for S2 cells, the following E:T ratios are performed: 30:1, 10:1, 3:1, 1:1, and target cells alone (spontaneous release). Set up the dilutions of NK cells. Ultimately, the target cells will be resuspended at a concentration of 20,000 cells in 100 ml, and the NK cells will be set up in dilutions containing the appropriate number of NK cells in 100 ml. For example, for the 30:1 points, the NK cells will need to be at a concentration of 6 × 105 cells per 100 ml. 4. Prepare FACS tubes. The cytotoxicity assay will be conducted directly in the FACS tubes. Label 2 tubes for duplicates of every target/dilution combination. 5. Pipet 100 ␮l of the NK cell dilutions into appropriately labeled tubes. Place these tubes in the 37◦ C incubator until the 2-h incubation of the labeled target cells is complete. 6. When the 2-h incubation of the labeled target cells is complete, bring the volume of the target cells up to 10 ml with Schneider’s medium containing 10% FBS, and spin down the cells at 215g. Wash the cells twice more with 10 ml of serum-containing medium. 7. Resuspend the target cells in 4 ml of Schneider cell medium with 10% FBS. Count the cells.

Transfected Insect Cells as Targets of NK Cells

83

8. Transfer 800,000 cells to a new tube and bring the volume up to 4 ml, resulting in a concentration of 20,000 cells in every 100 ␮l. 9. Add 100 ␮l (20,000 cells) to each tube with the NK cells. Vortex each tube. Then spin the tubes at 20g for 5 min. 10. Put the tubes at 37◦ C for 3 h. 11. Add propidium iodide to each tube to a final concentration of 50 ␮g/ml and vortex gently to mix. 12. On the flow cytometer, acquisition should be set up so that the green target cells, very bright in FL1, are gated on and the PI profile for those cells is displayed. Acquire at least 2000 gated target cells. 13. The percentage of cells that are positive for both the PKH67 dye and propidium iodide is the percentage of S2 cells lysed in each sample. Values from duplicate samples are usually averaged. The percentage specific lysis is calculated according to the formula [100 × (% PI -positive cells – % Spontaneous Death)/(100 – % Spontaneous Death)] where spontaneous death is determined with a sample containing only labeled S2 cells.

4. Notes 1. In functional assays involving S2 cells adapted to serumfree conditions, medium supplemented with 0.2% BSA must be used in place of serum-containing medium. Addition of serum to adapted cells will allow the recovery of cholesterol-containing microdomains. 2. S2 cells in a healthy culture are semi-adherent. When passaged into new tissue culture plates or flasks, most of the cells will adhere to the plastic fairly firmly. The cells can be detached from the plastic by repeated washes with medium using a pipet. After several passages in the same flask, the cells will not adhere as much, although they will proliferate and remain trypan blue negative. When in suspension, healthy S2 cells will grow individually, with minimal clustering or clumping. Expression of transfected proteins can be greatly affected by the health of the cells. Cells in overgrown cultures will often express lower levels of protein, whether expression is driven from a constitutive or an inducible promoter. S2 cultures in poor health can often be rescued by seeding cells into fresh medium at 1 × 106 cells/ml and allowing the cells to grow

84

March et al.

to 2.5 × 106 cells/ml. Once the cells are proliferating at a normal rate, protein expression will often recover to previous levels. 3. We have successfully co-transfected S2 cells with six different proteins simultaneously. It is likely that an even greater number of proteins could be simultaneously transfected. 4. Any given protein is usually expressed at a similar level in independently transfected S2 cells. For instance, expression of ICAM-1 is almost always high, whereas other molecules have proven more difficult to express. In the latter case, it is preferable to isolate S2 cells expressing a decent level of the troublesome protein after a first round of transfection and to “super-transfect” plasmids for other proteins in a second round of transfection using a different drug resistance plasmid. For example, S2 cells expressing CD48 were isolated first, prior to sequential transfections for expression of additional proteins. 5. Transient expression of proteins can be achieved by transfection with either calcium phosphate or Cellfectin method. Proteins will be expressed almost immediately following the addition of the transfection mixtures, with maximal expression occurring at 48–72 h. Either protocol should be followed to Step 5 (Section 3.2.1 for calcium phosphate transfection or Section 3.2.2 for Cellfectin transfection), where the transfection mixtures are removed by washing, and the cells are cultured for 48 h. If induction is required, 1 mM CuSo4 should be added 24 h after the washes, and induction should proceed for a further 24–48 h. 6. There are a number of plasmids using different promoters that can be used to obtain expression of exogenous proteins in Drosophila cells. We have used pRmHa3 and pAc5.1 plasmids successfully, as described in Step 5 of Section 3.1.1. Our cDNA constructs for expression in S2 cells have contained the entire coding sequence of the protein of interest, including the signal sequence. Where possible, 5’and 3’-untranslated regions of the mRNA were included in the expression constructs. It has not been necessary to alter the cDNA sequences in order to get good expression of the various human and mouse proteins in S2 cells. 7. The plasmids used for expression of selectable drug resistance must have promoters for constitutive expression. Commercial plasmids exist for selection of S2 cells with blasticidin or hygromycin (Invitrogen), as well as plasmids made by individual researchers for selection with puromycin and neomycin. It is not necessary to use expres-

Transfected Insect Cells as Targets of NK Cells

85

sion vectors that carry both a selectable marker and the coding region of the protein of interest. It is much more convenient to co-transfect plasmids containing the selectable marker together with the plasmids for protein expression. S2 cells are transfected by the calcium phosphate method with 19 ␮g of DNA, of which 18 ␮g is the expression plasmid(s) and 1 ␮g is the selection plasmid. In cases where expression of multiple exogenous proteins is desired, the plasmids are mixed in equal amounts to a total of 18 ␮g of DNA (for example, 6 ␮g each of three different plasmids). 8. The appropriate drug concentration for killing of untransfected S2 cells and selection of resistant S2 cells must be determined experimentally by serial dilutions. We have had success selecting resistant S2 cells with 1 mg/ml G418, 6 ␮g/ml puromycin, and 300 ␮g/ml hygromycin. Puromycin and hygromycin kill untransfected cells within 7 days. After their addition, the transfected S2 cell culture can be left alone until resistant cells grow, which usually occurs in 2 weeks. As G418 kills cells much more slowly, it is recommended to monitor live cell numbers. If the number reaches 3 × 106 cells/ml the cells should be removed from the medium by centrifugation and re-plated at 1 × 106 cells/ml in fresh medium containing G418. Protein expression may not be detectable for 2 weeks in G418 selection. 9. Selection of transfected clones is a time-consuming process, usually requiring an additional month after resistant cells have been obtained. To accelerate the process by several weeks, one can isolate expressing cells by FACS or by enrichment with antibodies coupled to magnetic beads. However, cell sorting will usually result in a broader range of expression than cell cloning and will often result in the loss or a reduction of protein over time. Therefore, it is necessary to regularly monitor expression in a sorted population and to repeat cell sorting when necessary. 10. Expression of proteins on transfected S2 cells should always be verified on the day of an assay. Expression level is dependent on the condition of the cells and can vary widely. 11. Alternatively, the S2 cells can be labeled with Celltracker dyes (Celltracker Green as a replacement for PKH67 and Celltracker Orange for PKH26). The cells should be incubated, in their standard medium at 10 × 106 cells/ml, with 4 ␮g/ml of the Celltracker reagent for 30 min at room temperature. The cells should be washed twice with their standard medium and then allowed to rest in 10 ml of medium

86

March et al.

in a 15 ml tube at room temperature for 30 min. Celltracker dyes are more convenient to use than the lipophilic PKH dyes. Additionally, the S2 cells occasionally display reduced viability after labeling with the PKH reagents. However, S2 cells labeled with the PKH dyes often display more uniform fluorescence than cells labeled with Celltracker, simplifying the FACS analysis. The appropriate label for a given application should be determined experimentally. 12. It is common to encounter problems with high killing of untransfected S2 cells in cytotoxicity assays. To minimize such occurrences, it is essential to seed S2 cells at optimal numbers and to allow them to proliferate for at least a week prior to the assay. However, nonspecific killing of untransfected S2 cells may also be due to constitutive release of perforin and granzymes by activated NK cells. We have not found a reliable solution to this problem. Assays for NK cell degranulation and for granule polarization in NK cells are very reliable and focus on the functions of interest (i.e., NK cell responses) rather than the fate of S2 cells. Nevertheless, target cell lysis may occasionally be a useful parameter of NK cell function, as it requires the combination of granule polarization and degranulation (31).

References 1. Chiesa, S., Tomasello, E., Vivier, E., and Vely, F. (2005) Coordination of activating and inhibitory signals in natural killer cells. Mol. Immunol. 42, 477–84. 2. Lanier, L. L. (2005) NK cell recognition. Annu. Rev. Immunol. 23, 225–74. 3. Moretta, L., and Moretta, A. (2004) Unravelling natural killer cell function: triggering and inhibitory human NK receptors. EMBO J. 23, 255–9. 4. Long, E. O. (2008) Negative signaling by inhibitory receptors: the NK cell paradigm. Immunol. Rev. 224, 70–84. 5. Barber, D. F., and Long, E. O. (2003) Coexpression of CD58 or CD48 with intercellular adhesion molecule 1 on target cells enhances adhesion of resting NK cells. J. Immunol. 170, 294–9. 6. Bryceson, Y. T., and Long, E. O. (2008) Line of attack: NK cell specificity and integration of signals. Curr. Opin. Immunol. 20, 344–52. 7. Bryceson, Y. T., March, M. E., Ljunggren, H. G., and Long, E. O. (2006) Synergy among receptors on resting NK cells for the activation of natural cytotoxicity and cytokine secretion. Blood 107, 159–66.

8. Lanier, L. L. (2008) Up on the tightrope: natural killer cell activation and inhibition. Nat. Immunol. 9, 495–502. 9. Bauer, S., Groh, V., Wu, J., Steinle, A., Phillips, J. H., Lanier, L. L., and Spies, T. (1999) Activation of NK cells and T cells by NKG2D, a receptor for stress-inducible MICA. Science 285, 727–9. 10. Pende, D., Parolini, S., Pessino, A., Sivori, S., Augugliaro, R., Morelli, L., Marcenaro, E., Accame, L., Malaspina, A., Biassoni, R., Bottino, C., Moretta, L., and Moretta, A. (1999) Identification and molecular characterization of NKp30, a novel triggering receptor involved in natural cytotoxicity mediated by human natural killer cells. J. Exp. Med. 190, 1505–16. 11. Pessino, A., Sivori, S., Bottino, C., Malaspina, A., Morelli, L., Moretta, L., Biassoni, R., and Moretta, A. (1998) Molecular cloning of NKp46: a novel member of the immunoglobulin superfamily involved in triggering of natural cytotoxicity. J. Exp. Med. 188, 953–60. 12. Shibuya, A., Campbell, D., Hannum, C., Yssel, H., Franz-Bacon, K., McClanahan, T., Kitamura, T., Nicholl, J., Sutherland,

Transfected Insect Cells as Targets of NK Cells

13.

14.

15.

16.

17.

18.

19.

20.

21.

G. R., Lanier, L. L., and Phillips, J. H. (1996) DNAM-1, a novel adhesion molecule involved in the cytolytic function of T lymphocytes. Immunity 4, 573–81. Moretta, A., Vitale, M., Bottino, C., Orengo, A. M., Morelli, L., Augugliaro, R., Barbaresi, M., Ciccone, E., and Moretta, L. (1993) P58 molecules as putative receptors for major histocompatibility complex (MHC) class I molecules in human natural killer (NK) cells. Anti-p58 antibodies reconstitute lysis of MHC class I-protected cells in NK clones displaying different specificities. J. Exp. Med. 178, 597–604. Garni-Wagner, B. A., Purohit, A., Mathew, P. A., Bennett, M., and Kumar, V. (1993) A novel function-associated molecule related to non-MHC-restricted cytotoxicity mediated by activated natural killer cells and T cells. J. Immunol. 151, 60–70. Kim, M., Carman, C. V., and Springer, T. A. (2003) Bidirectional transmembrane signaling by cytoplasmic domain separation in integrins. Science 301, 1720–5. Shimaoka, M., Xiao, T., Liu, J. H., Yang, Y., Dong, Y., Jun, C. D., McCormack, A., Zhang, R., Joachimiak, A., Takagi, J., Wang, J. H., and Springer, T. A. (2003) Structures of the alpha L I domain and its complex with ICAM-1 reveal a shape-shifting pathway for integrin regulation. Cell 112, 99–111. Ciccone, E., Pende, D., Viale, O., Than, A., Di Donato, C., Orengo, A. M., Biassoni, R., Verdiani, S., Amoroso, A., Moretta, A., et al. (1992) Involvement of HLA class I alleles in natural killer (NK) cell-specific functions: expression of HLA-Cw3 confers selective protection from lysis by alloreactive NK clones displaying a defined specificity (specificity 2). J. Exp. Med. 176, 963–71. Vitale, M., Sivori, S., Pende, D., Moretta, L., and Moretta, A. (1995) Coexpression of two functionally independent p58 inhibitory receptors in human natural killer cell clones results in the inability to kill all normal allogeneic target cells. Proc. Natl. Acad. Sci. U S A 92, 3536–40. Tangye, S. G., Cherwinski, H., Lanier, L. L., and Phillips, J. H. (2000) 2B4-mediated activation of human natural killer cells. Mol. Immunol. 37, 493–501. Cao, W., Xi, X., Wang, Z., Dong, L., Hao, Z., Cui, L., Ma, C., and He, W. (2008) Four novel ULBP splice variants are ligands for human NKG2D. Int. Immunol. 20, 981–91. Li, C., Ge, B., Nicotra, M., Stern, J. N., Kopcow, H. D., Chen, X., and Strominger, J. L. (2008) JNK MAP kinase activation is required for MTOC and granule polar-

22.

23.

24.

25.

26.

27.

28.

29.

30.

87

ization in NKG2D-mediated NK cell cytotoxicity. Proc. Natl. Acad. Sci. U S A 105, 3017–22. Akella, R., and Hall, R. E. (1992) Expression of the adhesion molecules ICAM-1 and ICAM-2 on tumor cell lines does not correlate with their susceptibility to natural killer cell-mediated cytolysis: evidence for additional ligands for effector cell beta integrins. Eur. J. Immunol. 22, 1069–74. Johnston, S. C., Dustin, M. L., Hibbs, M. L., and Springer, T. A. (1990) On the species specificity of the interaction of LFA-1 with intercellular adhesion molecules. J. Immunol. 145, 1181–7. Sivori, S., Pende, D., Bottino, C., Marcenaro, E., Pessino, A., Biassoni, R., Moretta, L., and Moretta, A. (1999) NKp46 is the major triggering receptor involved in the natural cytotoxicity of fresh or cultured human NK cells. Correlation between surface density of NKp46 and natural cytotoxicity against autologous, allogeneic or xenogeneic target cells. Eur. J. Immunol. 29, 1656–66. Schneider, I. (1972) Cell lines derived from late embryonic stages of Drosophila melanogaster. J. Embryol. Exp. Morphol. 27, 353–65. Matsumura, M., Saito, Y., Jackson, M. R., Song, E. S., and Peterson, P. A. (1992) In vitro peptide binding to soluble empty class I major histocompatibility complex molecules isolated from transfected Drosophila melanogaster cells. J. Biol. Chem. 267, 23589–95. Bunch, T. A., Grinblat, Y., and Goldstein, L. S. (1988) Characterization and use of the Drosophila metallothionein promoter in cultured Drosophila melanogaster cells. Nucleic Acids Res. 16, 1043–61. Cai, Z., Brunmark, A., Jackson, M. R., Loh, D., Peterson, P. A., and Sprent, J. (1996) Transfected Drosophila cells as a probe for defining the minimal requirements for stimulating unprimed CD8+ T cells. Proc. Natl. Acad. Sci. U S A 93, 14736–41. Cai, Z., Brunmark, A. B., Luxembourg, A. T., Garcia, K. C., Degano, M., Teyton, L., Wilson, I., Peterson, P. A., Sprent, J., and Jackson, M. R. (1998) Probing the activation requirements for naive CD8+ T cells with Drosophila cell transfectants as antigen presenting cells. Immunol. Rev. 165, 249–65. Barber, D. F., Faure, M., and Long, E. O. (2004) LFA-1 contributes an early signal for NK cell cytotoxicity. J. Immunol. 173, 3653–9.

88

March et al.

31. Bryceson, Y. T., March, M. E., Barber, D. F., Ljunggren, H. G., and Long, E. O. (2005) Cytolytic granule polarization and degranulation controlled by different receptors in resting NK cells. J. Exp. Med. 202, 1001–12. 32. Jackson, M. R., Song, E. S., Yang, Y., and Peterson, P. A. (1992) Empty and peptidecontaining conformers of class I major histocompatibility complex molecules expressed in Drosophila melanogaster cells. Proc. Natl. Acad. Sci. U S A 89, 12117–21. 33. Eleme, K., Taner, S. B., Onfelt, B., Collinson, L. M., McCann, F. E., Chalupny, N. J., Cosman, D., Hopkins, C., Magee, A. I., and Davis, D. M. (2004) Cell surface organization of stress-inducible proteins ULBP and MICA that stimulate human NK cells and T cells via NKG2D. J. Exp. Med. 199, 1005–10.

34. Faure, M., Barber, D. F., Takahashi, S. M., Jin, T., and Long, E. O. (2003) Spontaneous clustering and tyrosine phosphorylation of NK cell inhibitory receptor induced by ligand binding. J. Immunol. 170, 6107–14. 35. Schleinitz, N., March, M. E., and Long, E. O. (2008) Recruitment of activation receptors at inhibitory NK cell immune synapses. PLoS ONE 3, e3278. 36. Riteau, B., Barber, D. F., and Long, E. O. (2003) Vav1 phosphorylation is induced by beta2 integrin engagement on natural killer cells upstream of actin cytoskeleton and lipid raft reorganization. J. Exp. Med. 198, 469–74. 37. Burshtyn, D. N., Shin, J., Stebbins, C., and Long, E. O. (2000) Adhesion to target cells is disrupted by the killer cell inhibitory receptor. Curr. Biol. 10, 777–80.

Chapter 7 Natural Killer Cell Conjugate Assay Using Two-Color Flow Cytometry Deborah N. Burshtyn and Chelsea Davidson Summary This flow cytometry-based method is a quick way to detect adhesion of NK cells to target cells. The two cell types are labeled with distinct fluorescent dyes and following co-incubation, the number of NK cells firmly adhered to target cells is quantified using two-color flow cytometry. Key words: Natural killer cells, adhesion, cell conjugates, flow cytometry.

1. Introduction Intercellular adhesion is required for NK cells to specifically lyse target cells. The formation of a stable conjugate between the NK cell and its target cell allows for signal transduction and ultimately polarized degranulation at the point of contact with the target cell. The adhesion of NK cells to target cells requires the specific interaction of adhesion molecules such as the integrin LFA-1 on NK cells and ICAM on target cells (1). Integrins such as LFA-1 can provide the initial tethering of one cell to another as well as produce stronger binding following activation of the integrin by various stimuli (chemokines, antigen receptor signaling). The latter occurs when signals provided by receptors that lead to cellular activation also modify the ability of integrins to bind to their ligands by enhancing their affinity and/or avidity for their ligands. This process of an activating receptor upregulating the binding of an integrin is known as inside-out signaling and leads to tightly bound cell:cell conjugates. The assay described here measures the formation of such tightly bound conjugates between NK cells K.S. Campbell (ed.), Natural Killer Cell Protocols, Methods in Molecular Biology 612, DOI 10.1007/978-1-60761-362-6 7, © Humana Press, a part of Springer Science+Business Media, LLC 2010

89

90

Burshtyn and Davidson

and their targets. Although we originally developed this assay as a means to optimize conditions for performing live cell imaging on NK:target cell conjugates, the assay has proved useful for studying the molecules involved in adhesion and for detecting signals involved in forming and inhibiting conjugation. We and others have performed this assay, or variations of this assay, with a variety of natural killer cells including human NK cell lines (YTS, NK92), ex vivohuman-derived IL-2-activated NK populations and clones, mouse NK cells, and rat RNK cells (2–7). The flow cytometry method described herein is versatile, avoids the tedious task of counting conjugates by microscopy required by older methods, and does not require specialized equipment beyond the flow cytometer. The cell populations under study are labeled using nonspecific fluorescent membrane dyes. Once labeled, the NK and target cells are incubated together to allow for conjugation, agitated to disperse loosely associated cells, and then fixed. Finally, the conjugates are analyzed by twocolor flow cytometry to quantitatively measure the degree of conjugation.

2. Materials 2.1. Fixative Solution

1. Add 0.5 g of paraformaldehyde to 100 ml PBS. 2. Gently warm on a heating plate while gently stirring with stir bar until completely dissolved. Do not boil. 3. Allow to cool to room temperature. 4. Store at 4◦ C for up to 1 week. Prior to experiment, chill on ice.

2.2. Cell Labeling

1. SIGMA Fluorescent cell linker kits: PKH67-GL (Green) and PKH26-GL (Red) (see Note 1). The kits contain dye in ethanol and a diluent (Diluent C) and are stored at 4◦ C protected from the light according to the manufacturer’s instructions. 2. Serum-free medium. 3. Assay medium: 5% FBS in NK cell culture medium supplemented with 1 mM L-glutamine (see Note 2). 4. FBS. 5. 15 ml conical polypropylene centrifuge tubes.

2.3. Conjugation

1. Standard 3 ml polystyrene flow cytometry tubes (Falcon 2052). 2. 0.5% paraformaldehyde/PBS. 3. Flow cytometer.

NK/Target Cell Conjugate Assay

91

3. Methods 3.1. Cell Labeling

1. Prewarm Diluent C and dye to room temperature. 2. If crystals have formed in the dye stock, warm in 37◦ C water bath until dissolved. 3. Pellet 4 × 106 NK cells and 4 × 106 target cells separately in 15 ml conical centrifuge tubes (see Note 3). 4. Resuspend in 10 ml of serum-free medium. 5. For each cell type, retain 2 × 105 unlabeled cells in a FACs tube. 6. Pellet remaining cells. 7. Resuspend cells in 0.4 ml of Diluent C (room temperature). 8. Prepare 0.4 ml of 20 ␮M of each dye in Diluent C (8 ␮l of stock to 392 ␮l of Diluent C). 9. Add 0.4 ml of NK cells to the PKH67 dye and pipette up and down gently to mix. 10. Add 0.4 ml of target cells to the PKH26 dye and pipette up and down gently to mix. 11. Incubate 5 min at room temperature (22–25◦ C), resuspending the cells at 2 and 4 min by gently flicking the tube. 12. Add 2 ml FBS to stop the labeling and incubate 1 min at room temperature. 13. Wash cells twice by adding 10 ml room temperature assay medium and centrifuging 4 min at 300g. 14. Resuspend in 10 ml assay medium. 15. Incubate with the lid loose for at least 1 h at 37◦ C/5% CO2 to allow the excess dye to bleed out. 16. Perform a viable cell count. Cells should be greater than 90% viable at this stage. 17. Pellet and resuspend target cells in assay medium at 2 × 106 live cells/ml and the NK cells at 1 × 106 live cells/ml. For more information regarding the E:T, see Note 4. 18. Proceed immediately to conjugation assay.

3.2. Cell Conjugation

1. Dispense 100 ␮l of each labeled cell type into a FACs tube to be used for setting up the flow cytometry parameters. 2. Add 300 ␮l of ice cold 0.5% paraformaldehyde to the unlabeled and labeled control samples and store on ice. 3. For each desired time point (e.g., 0, 30 s, 1, 2, 5, and 10 min) aliquot 100 ␮l of target cells into FACs tubes (see Note 5).

92

Burshtyn and Davidson

4. Add 100 ␮l of NK cells. 5. Spin at 20 × g for 1 min (see Note 6). 6. Incubate at 37◦ C in a water bath for desired time points (see Notes 7 and 8). 7. At the end of each time point, vortex sample at high speed for 3 s (see Note 9). 8. Immediately fix by adding 300 ␮l of ice-cold 0.5% paraformaldehyde. 9. Store samples at 4◦ C protected from light until analysis. 10. Analyze by flow cytometry as soon as possible (see Note 10). Use untreated NK and target cells to set up the for-

Fig. 7.1. Illustration for gating on cell:cell conjugates. YTS cells were labeled with PKH67 and 721.221 cells with PKH26. The top panels shows the forward and side scatter plots for the mixture of the two cells at the 0 and 10 min time points. The large viable gate shown was used to capture the conjugates that have shifted to the right and up for the corresponding analysis shown in the lower panels. The numbers in the upper and lower right quadrants are the event counts that are used to calculate the % YTS cells in conjugates. For this example, the background conjugation works out as 72/(4679+72)∗ 100% = 1.5% which increased at 10 min to 1271/(2594+1271) = 33%. Note that the apparent decrease in total “YTS” events is likely due to aggregates forming that are counted as a single event and leads to an underestimate of the actual % conjugation.

NK/Target Cell Conjugate Assay

93

ward and side scatter parameters and set an acquisition gate to collect at least 20,000 live cell events. Be sure to use a wide forward scatter gate to capture conjugates, which may have higher forward scatter (see Fig. 7.1). Use the untreated and labeled cells to set up the sensitivity and compensation for the FL-1 (PKH67) and FL-2 (PKH26) channels correspondingly and acquire the samples. 11. Set up the analysis leaving a generous margin for single color events when setting the quadrants. Calculate the % of conjugated NK cells = 2-color events/(2-color events + FL-1+ events) × 100%. If desired, the background value can be subtracted from each experimental point which for some cell combinations can be a significant value.

4. Notes 1 Our experience has been extensively with the PKH series of dyes from Sigma and EGFP expressing cells (2, 3, 6, 8). However, this assay is adaptable to many dye pairs (such as a variety of CellTracker dyes from Molecular Probes/Invitrogen) or live cell fluorochromes (i.e., EGFP) that are compatible with flow cytometry, provided they are stably associated with the cells (5, 7, 9, 10). It is important when using the PKH dyes not to shorten the “bleed out” incubation time as this will lead to nonspecific transfer of dye from one cell type to the other through the medium. Also, after long time points of conjugation (e.g., >1 h), we do observe some transfer of the dye from one cell type to another which may result from conjugates that have dissociated. As these are dim events, allowing a wide window for the single color events will exclude these from being scored as bona fide two-color events. 2. We find the cells perform best when they are not shocked by any changes. Therefore, we use the same medium to perform the washes and assay as used to culture the cells. Typically we culture our NK lines in Iscoves medium. If you are using RPMI or other media, it is best to supplement with HEPES to prevent the pH from rising sharply when working in small volumes on the bench. The assays were optimized in medium similar to that typically used in our laboratory for cytotoxicity assays, hence 5% serum. The rate and plateau values of conjugation may be affected by higher or lower amounts of serum but we have not investigated this ourselves.

94

Burshtyn and Davidson

3. The NK and target cells should be taken from log phase growth cultures with high viability. When certain NK cell lines become overgrown, the degree of conjugation falls off in parallel to their cytotoxicity. This is particularly obvious with NK92 and YTS cells beyond 7 × 105 cells/ml. Experiment to experiment reproducibility is best when the cells used have been cultured to a similar density each time. Primary NK cells from very dense cultures work fine as long as they are still in rapid growth. If using NK populations or NK clones, the extent of binding will vary with the donor as well as for each clone from a given donor due to the heterogenous nature of these cells. Extremely large target cells can be problematic as it is difficult to properly gate on the NK cells and NK cells in conjugates by forward and side scatter at the same time. 4. The E:T outlined in this protocol is 1:2 as we have found this to be optimal for several NK:target cell combinations. However, titration of the E:T from 2:1 to 1:5 will reveal where the binding is saturated. This should be done by altering the concentration of the targets while holding the concentration of the effectors constant and performing the assay in the same volume. Be careful to keep E:T ratios consistent between samples since small differences in ratios can greatly impact upon the results. 5. Some combinations of NK and targets will produce a considerable amount of background binding at the “zero” time point without incubation at 37◦ C. This binding is occurring before and during the centrifugation step that in reality provides considerable time for binding. This background can be reduced dramatically by incubating the cells on ice for 10 min prior to mixing the cells. However, many NK cell lines and primary NK cells temporarily lose their lytic function when exposed to low temperatures. We found that YTS cells and RNK cells were able to function well in standard chromium release assays following incubation on ice, and therefore, we do incubate these cells on ice prior to beginning the assay, while we avoid this for NK92 and primary NK cells and alternatively prepare them at room temperature. 6. The rationale for this step is to facilitate contact between the cells by concentrating them at the bottom of the tube. The spin is done very gently to avoid shocking the membranes by collision with the plastic or form a tightly packed pellet. 7. Each NK–target cell combination may exhibit a slightly different rate of binding. Generally we find that maximal conjugation has occurred by 10 min and then after 1 h

NK/Target Cell Conjugate Assay

95

begins to subside particularly when the target cells are lysed by the NK cell. When performing long time course experiments, we find it convenient to perform the assay in reverse by mixing all the samples on ice, transferring them all to the 37◦ C water bath, and then removing the samples from shortest to longest time points. Also, handling of too many samples will increase variability at the early time points. 8. The kinetics of conjugation are affected by temperature, occurring more rapidly at 37◦ C than at lower temperatures (2). For reproducibility between samples, we found round bottom tubes to be the best to maximize cell-to-cell contact and allow rapid and even heating of the sample when warmed to 37◦ C. Similarly, it is best to use a recirculating water bath to ensure even heating in order to obtain the most reproducible results particularly for early time points. 9. The vortexing step following the co-incubation prior to fixation is done to break apart loosely associated cells. Without the vortex we observe high binding for many cell:cell combinations. However, too strong or too long a vortex can cause the cells to shear resulting in patches of effector or target cell membrane attached to the other cell type that can confound the flow cytometric analysis. These can readily be visualized by fluorescence microscopy. Consistent vortexing is important to obtaining reproducible results and therefore when learning the technique it is helpful to perform duplicate samples to ensure reproducibility. 10. Typically we analyze the samples promptly. However, we have stored samples in the fixative overnight without a loss of signal or resolution.

Acknowledgments This work was supported by CIHR and AHFMR. References 1. Matsumoto, G., Nghiem, M. P., Nozaki, N., Schmits, R., and Penninger, J. M. (1998) Cooperation between CD44 and LFA1/CD11a adhesion receptors in lymphokineactivated killer cell cytotoxicity. J Immunol 160, 5781–5789. 2. Burshtyn, D. N., Shin, J., Stebbins, C., and Long, E. O. (2000) Adhesion to target cells is disrupted by the killer cell inhibitory receptor. Curr Biol 10, 777–780.

3. Standeven, L., Carlin, L. M., Borszcz, P., Davis, D. M., and Burshtyn, D. N. (2004) The actin cytoskeleton controls the efficiency of Killer cell Ig-like Receptors (KIR) accumulation at inhibitory Natural Killer cell immune synapses. J Immunol 173, 5617–5626. 4. Vyas, Y. M., Maniar, H., Lyddane, C. E., Sadelain, M., and Dupont, B. (2004) Ligand binding to inhibitory killer cell Ig-like

96

Burshtyn and Davidson

receptors induce colocalization with Src homology domain 2-containing protein tyrosine phosphatase 1 and interruption of ongoing activation signals. J Immunol 173, 1571–1578. 5. Yusa, S., and Campbell, K. S. (2003) Src homology region 2-containing protein tyrosine phosphatase-2 (SHP-2) can play a direct role in the inhibitory function of killer cell Iglike receptors in human NK cells. J Immunol 170, 4539–4547. 6. Osman, M. S., Burshtyn, D. N., and Kane, K. P. (2007) Activating Ly-49 receptors regulate LFA-1-mediated adhesion by NK cells. J Immunol 178, 1261–1267. 7. Back, J., Chalifour, A., Scarpellino, L., and Held, W. (2007) Stable masking by H-2Dd cis ligand limits Ly49A relocalization to the site of NK cell/target cell contact. Proc Natl Acad Sci U S A 104, 3978–3983.

8. Borszcz, P. D., Peterson, M., Standeven, L., Kirwan, S., Sandusky, M., Shaw, A., Long, E. O., and Burshtyn, D. N. (2003) KIR enrichment at the effector-target cell interface is more sensitive than signaling to the strength of ligand binding. Eur J Immunol 33, 1084–1093. 9. Yusa, S., Catina, T. L., and Campbell, K. S. (2004) KIR2DL5 can inhibit human NK cell activation via recruitment of Src homology region 2-containing protein tyrosine phosphatase-2 (SHP-2). J Immunol 172, 7385–7392. 10. Li, C., Ge, B., Nicotra, M., Stern, J. N., Kopcow, H. D., Chen, X., and Strominger, J. L. (2008) JNK MAP kinase activation is required for MTOC and granule polarization in NKG2D-mediated NK cell cytotoxicity. Proc Natl Acad Sci U S A 105, 3017– 3022.

Chapter 8 Studying NK Cell/Dendritic Cell Interactions Mathias Lucas, Cedric Vonarbourg, Peter Aichele, and Andreas Diefenbach Abstract Although NK cells were originally identified as “naturally” active cells believed to follow a cellautonomous activation program, it is now widely accepted that NK cells need to interact with dendritic cells for their full functional activation and for their homeostasis. In this chapter, we will provide an experimental guide to the analysis of NK cell/DC interactions in vitro and in vivo. We have put special emphasis on the recently developed mouse models allowing the inducible and specific ablation of various subsets of DCs and other myeloid cells. Key words: LCMV, Listeria monocytogenes , toll-like receptors, dendritic cells, macrophages, diphtheria toxin receptor.

1. Introduction Natural killer (NK) cells were discovered as lymphocytes that can spontaneously kill certain tumor target cells (1–3). These early findings have led to the proposal that NK cells are innate immune cells which follow a cell-autonomous activation and effector program once confronted with an appropriate target cell expressing stimulatory ligands. In contrast to T cells of the adaptive immune system, NK cells were widely believed to be independent of signals provided by other cell types for activation or priming. However, freshly isolated NK cells from mice and humans show only minimal effector functions (cytotoxicity, cytokine production) when incubated in vitro with tumor target cells or when directly trigK.S. Campbell (ed.), Natural Killer Cell Protocols, Methods in Molecular Biology 612, DOI 10.1007/978-1-60761-362-6 8, © Humana Press, a part of Springer Science+Business Media, LLC 2010

97

98

Lucas et al.

gered via their stimulatory receptors (4, 5). These data suggested that resting NK cells might depend on additional signals for their activation. Indeed, most investigators treat mice with tolllike receptor (TLR) ligands, ligands for the cytoplasmic helicase protein melanoma differentiation-associated gene 5 (MDA5) or cytokines (e.g., type I interferons, IFN-I or their synthetic inducers) or culture NK cells ex vivo in the presence of cytokines (e.g., IL-2, IL-15) to elicit detectable effector functions prior to in vitro analyses of NK cell responses (4, 6, 7). These findings have led to more detailed investigations into whether NK cells might be dependent on signals from other cells for their full functional activation. A role for myeloid cells in the induction of NK cell responses in vivo has been considered by many reports (8–16). These initial studies showed that in vitro co-cultures of stimulated or infected bone marrow (BM)-derived dendritic cells (DCs) or macrophages with NK cells resulted in NK cell activation. In other studies, BMderived DCs were injected into mice which led to enhanced NK cell activity against tumors or virally infected cells (12, 17, 18). Until recently, it was not possible to study NK cell activation in the absence of DCs or macrophages in vivo because mouse models specifically lacking DCs or macrophages were unavailable. Studies using depletion of myeloid cells from mice by injection of depleting antibodies were hard to interpret as most of the phenotypic markers used are also expressed by other hematopoietic cells. The systemic depletion of DCs can be achieved by the injection of clodronate liposomes (19). However, uptake of clodronate liposomes requires phagocytic activity of the cell resulting in the depletion of both DC and macrophage populations. In the last years, various genetically modified mouse models became available allowing for the inducible and specific ablation of myeloid cell subsets. All these models employ tissue-specific expression of the avian diphtheria toxin receptor (DTR) (20). As mouse cells are insensitive to diphtheria toxin (DT) from Corynebacterium diphtheriae, only those cells expressing the ectopic DTR are sensitive to DT. By now, various studies are available that have analyzed NK cell function in mice lacking DCs. Collectively these data demonstrate that NK cell function in response to TLR-mediated stimulation and to various pathogens strictly depends on the presence of conventional DCs (5, 21–26). On the following pages, we will provide protocols for the analysis of NK cell priming or activation by myeloid cells in vitro. Furthermore, we discuss how the available mouse models allowing for the ablation of various myeloid cell subsets can be employed to analyze NK cell function in vivo and how they can be further manipulated to reveal the molecular pathways required for the priming of NK cells by myeloid cell populations in vivo.

Studying NK Cell/Dendritic Cell Interactions

99

2. Materials 2.1. Mice

1. C57BL/6 (B6; expressing the CD45.2 antigen) 2. B6-Ly5.2/Cr (C57BL/6-Ly-5a congenic mice, expressing the CD45.1 antigen) 3. CD11c DTR transgenic mice (B6.FVB-Tg(ItgaxDTR/EGFP)57Lan/J) (27) or other mouse strains allowing for the depletion of myeloid cell subsets (Table 8.1). 4. H-2 K-BCL-2: Bcl-2 transgenic mice under the control of the H-2 K promoter (28). 5. Il15 −/− (C57BL/6NTac-IL15tm1Imx )(29) 6. All mice are maintained under specific pathogen-free conditions and used at 8–16 weeks of age.

2.2. Cell Lines

1. MC57 (H-2b ) is a C57BL/6-derived methylcholanthreneinduced fibrosarcoma cell line used to titer LCMV (Section 3.6) (30). 2. NCTC clone 929 (L-929; Connective tissue, mouse; ATCC CCL-1) 3. YAC-1 (ATCC TIB-160) is a T-cell lymphoma used for the NK cell activation assays (Section 3.8) (31). 4. RMA-S is a Tap-2-defective cell line used for the NK cell activation assays (Section 3.8) (32). 5. RMA-S-H60 cells are RMA-S cells retrovirally transduced with a ligand of the NKG2D receptor, H60 (5, 33). These cells are used as target cells in the NK cell activation assays (Section 3.8).

2.3. Solutions

1. Minimum essential medium (MEM) cell culture medium: MEM (Invitrogen) containing Earle’s Salts and nonessential amino acids (NEAA) is complemented with 5% fetal calf serum (FCS), 200 mg/l glutamine (Sigma), 50,000 U/l penicillin, and 50 mg/l streptomycin (Invitrogen), filter sterilized (0.22 ␮m), and stored a 4◦ C. 2. 2× MEM: MEM powder (Invitrogen) is dissolved in 5 l ddH2 O (instead of 10 l) to obtain a 2× concentration. 2× MEM is complemented with 10% FCS, 400 mg/l glutamine, 100,000 U/l penicillin, and 100 mg/l streptomycin, filter sterilized (0.22 ␮m), and stored at 4◦ C. 3. Dulbecco’s Modified Eagle Medium (DMEM) cell culture medium: DMEM (Invitrogen) containing 4.5 g/l glucose and L-glutamine is complemented with 5–10% FCS (this percentage may vary depending on the cultured

100

Lucas et al.

cells), 200 mg/l glutamine, 50,000 U/l penicillin, 50 mg/l streptomycin, 10 mg/l gentamycin sulfate (Biowhittaker), and 50 ␮M ␤-mercaptoethanol (Sigma), filter sterilized (0.22 ␮m), and stored at 4◦ C. 4. Brain Heart Infusion (BHI) medium: 1 l deionized water; 37 g BHI powder (Oxoid). Heat at 100◦ C for 15–20 min in order to completely dissolve the powder. Autoclave and store at 4◦ C. 5. BHI Agar: 1 l deionized water supplemented with 16 g Bacto Agar (BD); 37 g BHI powder. Heat at 100◦ C for 15– 20 min in order to completely dissolve the powder. Autoclave and store at 4◦ C. 6. Methylcellulose solution: Mix 1 vol. of 2% methylcellulose (Sigma) in ddH2 O with 1 vol. 2× MEM. 7. BHI containing 0.05% Triton X-100 (Amresco). 8. Dulbecco’s phosphate buffered saline (DPBS; Invitrogen) containing 0.5% Triton X-100. 9. Trypsin-EDTA: 0.05% Trypsin-EDTA (Invitrogen). 10. ACK lysis buffer (red blood cell lysing buffer): 1 l deionized water supplemented with 8.025 g NH4 Cl, 1 g KHCO3 , and 200 ␮l 0.5 M EDTA (pH 8.0); sterilize using 0.22 ␮m filter and store at 4◦ C. 11. BD Pharm Lyse buffer (BD Biosciences). 12. FACS buffer: DPBS (Invitrogen) complemented with either 2% FCS or 0.5% BSA (Sigma), filter sterilized (0.22 ␮m), and stored a 4◦ C. 13. MACS buffer: DPBS (Invitrogen) complemented with 0.5% BSA and 2 mM EDTA (Invitrogen), filter sterilized (0.22 ␮m), and stored a 4◦ C. 14. PBS/EDTA: DPBS containing 5 mM EDTA (pH 8.0). 15. Monensin (GolgiStop, BD Bioscience). 16. Brefeldin A (Sigma): stock solution of 5 mg/ml in ethanol. 17. Cytofix/Cytoperm solution (BD Bioscience). 18. FACS permeabilization buffer: FACS buffer containing 0.5% saponin (Sigma). 19. FACS fixing solution: DPBS (Invitrogen) containing 4% formaldehyde (dilution of a 37% formaldehyde solution in PBS). 20. 4% Paraformaldehyde: Heat 50 ml ddH2 O to 60◦ C (glass beaker); add 4 g paraformaldehyde (Merck). Stir (stirring bar) and add few drops (2–3) of 1 N NaOH until the solution clears (if not, check 70. That said, very low numerical intensity values are generally not reflective of true signal (similar to “autofluorescence” in FACS). Options for creating a threshold over which fluorescence intensities are to be included in analyses include:

Analysis of the NK Cell Immunological Synapse

147

(1) selecting an objective florescence intensity value; (2) selecting a percentage of fluorescence intensity (i.e., the top 40% of fluorescence intensities; and (3) selecting a specified number of standard deviations above the mean fluorescence intensity. Although there is some degree of relativity present in the latter two approaches, they are especially useful in time-lapse experiments, where the fluorescence may decay over time. In this case, the relative intensity differences will generally be maintained and can be still exploited for quantitative analysis. In all cases comparison to controls is essential in quantitation. These include staining controls as well as biological controls. By using each, true increases, decreases, and alterations in intensity, as well as area/volume containing those intensities can be measured. Note that during colocalization analysis, the threshold for measuring one molecule may be different than the threshold for measuring another. 10. In comparing fluorescent regions as described in the procedures and above in Note 9, it is important that sufficient numbers of cells be evaluated. There is no set number of cells that must be evaluated for an experiment to be valid. What is appropriate, however, is that the number of cells included for analysis represents a sample that is sufficiently powered to enable statistical testing of the hypothesis. In order to decide upon the number of cells that should be evaluated for a given experiment, it is best to perform a sample size calculation using values obtained from a pilot or related experiments. 11. Importantly, once a threshold is set it needs be applied to all analyses within a given experiment uniformly. Different software packages have different options for identifying the pixels that contain fluorescence of multiple fluorophores that fall above the user-defined threshold (in Volocity, this can be performed using the “Intersect Objects” command). These algorithms will also provide the mean intensity of each fluorophore in that colocalized region. As there will likely be multiple colocalized regions within a single cell, all of the individual regions can be joined into a single region to contain the total colocalized area along with its mean intensity (in Volocity this can be performed using the “Join Objects” command). 12. When defining intensity thresholds in live cell experiments, it can be helpful to use percentage intensity- or standard deviation-based thresholds, as in longer experiments the absolute fluorescence of a fluorophore can fade, but the differences remain intact.

148

Sanborn et al.

References 1. Axelrod, D. (2001) Total internal reflection fluorescence microscopy in cell biology. Traffic 2, 764–774. 2. Axelrod, D. and Davidson, M.W. (2008) Specialized Microscopy Techniques – Total Internal Reflection Fluorescence Microscopy. Available online at www. olympusmicro.com/primer/techniques/ fluorescence/tirf/tirfintro.html. 3. Orange, J. S. (2008) Formation and function of the lytic NK-cell immunological

synapse. Nat Rev Immunol. 2008 Sep; 8(9): 713–725. 4. Banerjee, P.P. and Orange, J.S., submitted. 5. Banerjee, P. P., Pandey, R., Zheng, R., Suhoski, M. M., Monaco-Shawver, L., and Orange, J. S. (2007) Cdc42-interacting protein-4 functionally links actin and microtubule networks at the cytolytic NK cell immunological synapse. J Exp Med 204, 2305–2320.

Chapter 10 Measuring Intracellular Calcium Signaling in Murine NK Cells by Flow Cytometry Alexander W. MacFarlane IV, James F. Oesterling, and Kerry S. Campbell Abstract This chapter describes a method by which activating receptor-mediated calcium signaling can be measured in individual murine NK cells using a flow cytometer fitted with a UV laser. One major advantage of this method is that the calcium response of the minority NK cell population and even smaller NK cell subpopulations can be measured simultaneously from a mixture of freshly prepared total splenocytes without resorting to prior cell sorting or expansion in culture. Briefly, cells are harvested and stained to mark the populations of interest, then loaded with indo-1 AM dye and analyzed on the flow cytometer. After an appropriate baseline is established, the cells are treated with a biotinylated antibody to activating receptors, which are subsequently cross-linked by addition of streptavidin. The increase in intracellular calcium is quantified by measuring a shift in the indo-1 emission spectrum that takes place when the dye becomes bound to calcium. Key words: Calcium signaling, indo-1, flow cytometry, NK1.1.

1. Introduction Increased levels of cytosolic calcium are critical to numerous lymphocyte functions including proliferation, metabolism, apoptosis, migration, cytotoxicity, and the formation of an immunological synapse (1). This protocol builds on the work of Valittuti and Dessing, who previously described a methodology for measuring calcium signaling in individual cells within a T-cell population in response to target cell engagement in this series (2). Here we expand upon the method to demonstrate NK cell calcium mobilization in response to soluble ligands and demonstrate how the relative responsiveness of various developmental subpopulations K.S. Campbell (ed.), Natural Killer Cell Protocols, Methods in Molecular Biology 612, DOI 10.1007/978-1-60761-362-6 10, © Humana Press, a part of Springer Science+Business Media, LLC 2010

149

150

MacFarlane IV, Oesterling, and Campbell

can be resolved. Cytosolic calcium concentrations are normally tightly regulated in resting cells. The increased intracellular calcium concentration in response to activating receptor engagement takes place in two stages. Calcium is initially released from the endoplasmic reticulum over a short period of time, followed by a sustained influx from the extracellular environment. The data collected is a convolution of these two processes, but since the early time points are dominated by release of intracellular calcium stores and the later time points by influx from the extracellular medium, it should be possible to see a specific effect on either of the two processes. In this method, the cells are first stained with antibodies to cell surface markers that enable the identification of NK cells and any subsets of interest, then loaded with indo1 AM dye and stimulated through activating receptors. Calcium flux is initiated by cross-linking activating receptors that are first engaged by a biotinylated antibody and subsequently clustered together by addition of streptavidin. Changes in the intracellular calcium concentration are quantified by a shift in the indo1 emission peak from 485 nm (indo-blue) for unbound dye to 405 nm (indo-violet) when the indo-1 molecule is bound to calcium. Mean intracellular calcium concentration is quantified in terms of the ratio of 405/485 nm indo-1 emission peaks. Measuring the change in emission ratio allows comparisons between individual cells within the population that may not be loaded with equivalent amounts of indo-1 dye.

2. Materials 1. Erythrocyte lysis buffer: 125 mM KHCO3 , 1 mM Na2 EDTA.

NH4 Cl,

10 mM

2. Serum-free RPMI-1640 medium (Life Technologies, Rockville, MD). 3. Complete RPMI medium: RPMI-1640 medium, 10% FBS (Hyclone), 100 ␮g/ml penicillin/streptomycin, 2 mM Lglutamine, 10 mM HEPES, pH 7.4, 1 mM MEM sodium pyruvate, and 50 ␮M 2-mercaptoethanol (all from Life Technologies). 4. Cell permeant indo-1 AM dye (Invitrogen, Eugene OR): prepare a 2.5 mM stock in DMSO. Store protected from light at –20◦ C (see Note 1). 5. Pluronic F-127 20% solution in DMSO (Invitrogen). 6. Conjugated monoclonal antibodies: anti-CD3␧PerCP/Cy5.5, anti- anti-CD122-FITC [or anti-CD49b-

Measuring Intracellular Calcium Signaling in Murine NK Cells by Flow Cytometry

151

FITC (DX5)], anti-CD11b-APC/Cy7, and biotinylated anti-NK1.1 (PK136) (all from BioLegend, San Diego, CA). 7. Propidium iodide 1 mg/ml (Invitrogen) (see Note 2). 8. Purified streptavidin (Sigma). 9. 40 ␮m cell strainer (Falcon, BD Biosciences). 10. 10 ml syringe. 11. Flow cytometer: Our studies used a BD FacsVantage SE flow cytometer with the FACSDiVa Option fitted with a Coherent Innova Model 302C Krypton Laser. The laser is operated in multiline UV mode with emission lines ranging from 337.5 to 356.4 nm and a Chroma 440 DCLP splitter/indo-1 filter set measuring emission peaks at 405 and 485 nm. 12. FlowJo software, version 8.7.1 (TreeStar) and appropriate computer.

3. Methods 3.1. Cell Preparation

1. Prepare single cell suspensions of mouse splenocytes by mashing spleens through a 40 ␮m nylon cell strainer with a rubber-tipped 10 ml syringe plunger and rinse the cell strainer with a total of 10 ml of complete RPMI culture medium. 2. Spin the cells at 500 g for 5 min and resuspend in ice cold erythrocyte lysis buffer. Incubate for 3 min on ice, then pellet cells at 500 g for 5 min. 3. Immediately resuspend in 10 ml of cold complete RPMI medium and count. 4. Spin down the cells and resuspend at a dilution of 20 million cells in 1 ml of complete RPMI-1640. 5. Add appropriate staining titer of each of the fluorochromeconjugated antibodies and incubate on ice for 20 min. Single color compensation controls and an unstained control should be prepared at this time as well (see Notes 3 and 4). 6. While the cells are being stained, add 1 ␮l of 20% Pluronic F-127 solution to every 9 ␮l of indo-1 AM stock solution needed and warm at 37◦ C for 5–10 min in a shaking heat block to ensure that it is mixed properly (see Note 1). Protect from light during all preparation procedures.

152

MacFarlane IV, Oesterling, and Campbell

7. Rinse cells twice with serum-free RPMI-1640 medium and resuspend in 4 ml of serum-free RPMI-1640 medium that has been pre-warmed to 37◦ C and place in a 37◦ C water bath. Cell concentration should be about 5 million/ml. 8. Allow 10 min for the cells to warm at 37◦ C, then add 1– 1.5 ␮l of the indo/Pluronic solution prepared in Step 5 to each ml of cell suspension and mix thoroughly by inverting the tube several times. If comparing different cell populations, it is critical to assure uniform indo-1 loading between the different cell samples, so exactly the same conditions should be used for each sample and cells should be mixed thoroughly after addition of the dye. 9. Incubate the cells for 30 min at 37◦ C under foil to protect from light. Resuspend the cells every 10 min during this incubation. Then centrifuge the cells at 500 g for 5 min and resuspend in 2 ml of complete RPMI-1640 medium. If propidium iodide is being used to exclude dead cells, then add at a final concentration of 200 ng/ml. 3.2. Flow Cytometry

Maintain labeled cells in the dark at room temperature (not on ice) at all times prior to analysis on the flow cytometer. Do not store labeled cells on ice, since this will diminish the intensity of subsequent calcium responses. 1. Prewarm a 500 ␮l cell sample of cell suspension in a 5 ml FACS tube at 37◦ C for 5 min prior to analysis on the flow cytometer. 2. Run the 500 ␮l sample through the flow cytometer at about 2000–2500 events/s for 1–2 min to establish the baseline indo-1 signal that represents the basal intracellular calcium concentration. The cell sample should be maintained at 37◦ C throughout the time course of analysis. Although calcium mobilization can be measured at room temperature, the response will be suboptimal as compared to analysis at 37◦ C. 3. Remove the sample and add 4 ␮g of biotinylated anti-NK1.1 (PK136) mAb and return the sample to the instrument (see Note 5). 4. Allow the cells to run for another minute to determine if the antibody stimulates the cells in the absence of streptavidin. We have not found this to be true for PK136, but it may occur when other receptors or antibodies are used. 5. Remove the sample once again, add 8 ␮g of streptavidin, and return the sample to the instrument to collect data for an additional 4–5 min (see Note 6). An increase in indoviolet (405 nm) signal and a decrease in indo-blue (485 nm) signal are indicative of a stimulation-induced increase in

Measuring Intracellular Calcium Signaling in Murine NK Cells by Flow Cytometry

153

intracellular calcium concentration. If longer analysis is desired, an increased starting volume of cells (750–1000 ␮l) should be used and antibody/streptavidin concentrations should be adjusted accordingly. 3.3. Data Analysis

We have used FlowJo software (version 8.7.1) to analyze our data. It is necessary to create a derived parameter to display the indo-violet (405 nm) intensity divided by the indo-blue (485 nm) intensity. This is accomplished by navigating through the following menu tabs: Platform -> Derive Parameters -> Define new or change. Here you can create the new parameter and give it a name. Make sure the “display with linear scale” box is checked and start with lower and upper axis limits of 0.4 and 1.2, respectively. Once the parameter is created you can view the mean relative cytosolic calcium concentration within the gated population by highlighting the desired NK cell gate in the FlowJo work space and choosing Platform -> Kinetics from the menu tabs. In Fig. 10.1A, we set the axis limits to 0.5 and 0.9, while choosing to display the mean fluorescent intensity acquired over the integration time of each data point. Since the actual signal is changing slowly compared with the fluctuations in the data, it is reasonable to increase the signal–to-noise ratio by averaging adjacent points. Figure 10.1B shows the same data with a moving average applied. This replaces each data point with the average value of the data point with several points on either side. The Gaussian smoothing shown in Fig. 10.1C is a moving average that weights adjacent points more heavily than those that are further away. This results in less smoothing than the unweighted moving average, but is also less likely to obscure rapidly changing details or distort the actual rates of ascent and decline. Another approach is to plot the percent of cells with an indo-violet to indo-blue ratio above a baseline threshold value. This provides information about what fraction of a given population or subpopulation is undergoing calcium flux. Figure 10.1D shows this with a threshold of 0.65 and moving average smoothing. One of the important advantages of this method is the ability to distinguish populations and subpopulations of cells without sorting or other mechanical fractionation. Figure 10.2 shows how NK cells can be distinguished from other cell types and the different magnitudes of calcium flux within each gated subpopulation. The first gate, shown in Fig. 10.1A, is applied to a plot of forward scatter height versus forward scatter area. This is referred to as a singlet gate because it excludes instances where more than one cell is contained in a droplet that is analyzed by the cytometer. Figure 10.2B shows the lymphocyte gate, which excludes cell fragments, the majority of dead and dying cells, and large aggregates. As shown in Fig. 10.2C, NK cells are gated as CD122+ (or

154

MacFarlane IV, Oesterling, and Campbell

Fig. 10.1. Mean calcium flux data from NK cells stimulated with biotinylated antiNK1.1 mAb (added at 60 s) streptavidin (at 120 s) are visualized by several smoothing methods. Plotting the ratio of mean indo-violet/indo-blue emission values at individual time points provides values corresponding to relative calcium concentration of cells within the population over the time course of the experiment. (A) No smoothing. (B) Moving average. (C) Gaussian smoothing. (D) Plot of indo-violet (V)/ indo-blue (B) events that are above a threshold ratio value of 0.65. Time in seconds is shown in the x-axis.

Measuring Intracellular Calcium Signaling in Murine NK Cells by Flow Cytometry

155

Fig. 10.2. Electronic gating to distinguish NK cells and subsets of NK cells from other lymphocytes and cellular debris. (A) To assure a single cell analysis, events containing one cell are selected based on linear correlation between forward scatter area (FSC-A) and forward scatter height (FSC-H). (B) Predominantly viable cells are distinguished from dead cells and small particles based on their forward scatter height and side scatter area (SSC). (C) NK Cells are gated as CD3 (TCR)– , CD122+ . (D) Subpopulations of NK cells are defined by their expression levels of CD11b immature (CD11blow ) and mature (CD11bhigh ). (E) Calcium flux measurements of the subsets defined in (D) above with the thin line representing immature NK cells and the thicker line representing mature NK cells.

CD49b+ if substituted), CD3− . Once the NK population has been selected, the cells can be further subdivided according to expression of CD11b into successive maturation stages that are shown in Fig. 10.2D. A comparison of the kinetics of these two subpopulations shown in Fig. 10.2E reveals greater functionality in the mature subset.

156

MacFarlane IV, Oesterling, and Campbell

4. Notes 1. The acetoxymethyl (AM) ester derivative of indo-1 is uncharged and can cross the plasma membrane into cells. Once inside the cells, esterases cleave the side chain to create a charged form of the parent indo-1 molecule that is retained in the cytoplasm. Pluronic F-127 is non-ionic detergent with low cytotoxicity that aids solubility of indo-1AM in aqueous solutions for improved loading of cells. 2. Propidium iodide or 7AAD can be used to exclude dead cells from the analysis, but if the emission channel is needed for another antibody it is also possible to gate populations of cells that have greater than 98% viability by using the forward scatter and side scatter gates. This is possible because dead and dying cells tend to be smaller and more granular, so they exhibit less forward scatter and greater side scatter than viable cells. 3. Avoid staining with antibodies directed toward NK cellactivating receptors that can be triggered by antibody engagement. 4. One should be cautious if considering the use of Cascade Blue, Alexafluor 405, or other fluorophores in the violet/blue emission range as gating antibodies, because their emission spectra may overlap with that of indo-1. 5. NK1.1 is a member of the NKR-P1 family of receptors and is only expressed on NK cells from certain mouse strains, such as C57Bl/6, but not 129 or Balb/c. 6. The balance of biotinylated antibody to streptavidin determines the degree of receptor aggregation, which is critical for the experiment to succeed. Biotinylated antibody and streptavidin concentrations should be titrated to determine the optimal concentrations whenever a new activating receptor is chosen to initiate the calcium flux. This may require significant effort to achieve the proper combination. If not enough streptavidin is used, then the receptors will be poorly cross-linked and fail to stimulate the cells. Using too much streptavidin can also result in poor cross-linking if each biotin moiety binds a single streptavidin molecule, instead of multiple biotin molecules binding to each streptavidin tetramer. Another option is to use fluorophore-conjugated streptavidin as a cross-linking agent to assess degree of cross-linking and surface levels of receptor being engaged. The conditions for biotinylated anti-NK1.1 and streptavidin have been optimized in our hands using this experimental design, in which the NK cells make up on average about 3% of the total splenocytes being analyzed.

Measuring Intracellular Calcium Signaling in Murine NK Cells by Flow Cytometry

157

Acknowledgments We would like to thank Dr. Richard (Randy) Hardy for assistance in setup of the flow cytometry equipment and for suggestions to improve the manuscript. Supported by National Institutes of Health grants R01-CA083859, R01-CA100226 (K.S.C.), T32AI007492 (A.W.M.), and Centers of Research Excellence grant CA06927 (FCCC). The research was also supported in part by the FCCC Blood Cell Development and Cancer Keystone Program and an appropriation from the Commonwealth of Pennsylvania. Its contents are solely the responsibility of the authors and do not necessarily represent the official views of the National Cancer Institute. References 1. Clapham, D.E. (2007) Calcium Signaling. Cell 131, 1047–1058. 2. Valitutti, S. and Dessing, M. (2000) Measurement of Calcium Mobilization

Responses in Killer Cell/Target Conjugates by FACS Analysis. Methods Mol Biol 121, 305–311.

Chapter 11 Intracellular Staining for Analysis of the Expression and Phosphorylation of Signal Transducers and Activators of Transcription (STATs) in NK Cells Takuya Miyagi, Seung-Hwan Lee, and Christine A. Biron Abstract Cytokines stimulate biological responses by activating intracellular signaling pathways. We have been adapting flow cytometric techniques to measure the levels of expression and activation of signaling molecules within mixed populations containing NK cells and to characterize their differences within NK cell subpopulations. Approaches for evaluating the total levels of the signal transducers and activators of transcription STAT1 and STAT4, of STAT1 in cells expressing IFN␥, and of the type 1 interferon (type 1 IFN) activation by phosphorylation, i.e., induction of pSTAT1 and pSTAT4, have been developed. The results of experiments using these techniques have demonstrated that an unusual feature of NK cells is high basal expression of STAT4 but reduced STAT1 levels. The condition predisposes for pSTAT4 activation by type 1 IFNs. The work has also shown, however, that total STAT1 levels are induced during viral infections as a result of IFN exposure, and that this change acts to promote the activation of STAT1 but limit both the activation of STAT4 and IFN␥ expression. The intracellular staining approaches used for the studies described here have utility in characterizing other mechanisms regulating cytokine-mediated signaling, and defining additional pathways shaping cellular responses to cytokines. Key words: STAT1, STAT4, pSTATs, intracellular staining, type 1 IFN.

1. Introduction 1.1. Cytokine Signaling

A range of cytokines use signal transducers and activators of transcription molecules (STATs) as intracellular intermediaries for eliciting cellular responses. A class of cytokines using STATs to signal is the type 1 interferon (IFN) family comprised of a single ␤ and multiple ␣s. The factors are elicited in response to a variety of stimuli including viral infections (1, 2). As a result of binding to specific receptors, type 1 IFNs induce particular

K.S. Campbell (ed.), Natural Killer Cell Protocols, Methods in Molecular Biology 612, DOI 10.1007/978-1-60761-362-6 11, © Humana Press, a part of Springer Science+Business Media, LLC 2010

159

160

Miyagi, Lee, and Biron

kinases to activate STATs by phosphorylation (3, 4). The best understood signaling pathway downstream of the type 1 IFN receptor uses STAT1 and STAT2, and this pathway is linked to the induction of antiviral defense mechanisms. Type 1 IFNs, however, have been reported to induce a wide range of biological functions and have important immunoregulatory effects (1, 2). Some of these are paradoxical, including enhancing and inhibiting IFN␥ production as well as blocking and promoting cell proliferation (1, 5–8), but the mechanisms regulating selection of the subset functions are poorly understood. There are a total of seven STATs, 1 through 6 with two different 5s. Type 1 IFNs have been reported to conditionally activate all of these, including STAT4 (1, 9). Their negative effects on IFN␥ expression and cell proliferation are dependent on STAT1 (5, 6, 10), whereas STAT4 contributes to IFN␥ expression (7, 11, 12). Thus, conditions affecting the relative accessibility of STAT4 and STAT1 have the potential to contribute to the shaping of cellular responses to type 1 IFNs. 1.2. Approaches for Evaluating Access to Signaling Pathways

Until recently, evaluation of the availability and activation of intracellular signaling pathways has been limited to biochemical Western blot analyses using proteins extracted from cell populations. The technique does allow determination of total protein and phospho-protein levels within samples and has the advantage of revealing the molecular weights of proteins detected. Using this approach, we have analyzed samples from mouse splenic leukocytes to show that there is an inverse correlation between type 1 IFNs’ ability to activate STAT4 with total STAT1 levels (7). This method, however, does not allow characterization of differences in responses within mixed cell subpopulations, and the numbers of purified cells required can present a challenge for certain NK cell studies. Flow cytometric or fluorescent-activated cell sorting (FACS) techniques can evaluate multiple parameters within mixed cell subsets. The approach has become very powerful because of the development of (1) specific monoclonal antibodies against a variety of cell determinants and cytokines; (2) an increasing range of fluorochromes with different excitation and emission spectra for coupling to the monoclonal antibodies; and (3) improved instrumentation (13). In addition to being used to evaluate the molecules expressed on the surfaces of cell subsets, flow cytometry is being used to characterize mixed cell subsets in regard to the range and levels of cytokines they can be induced to express in their cytoplasm. This is possible because of the development of different permeabilization protocols allowing intracellular access of antibodies detecting cytokines. Thus, there are indications that flow cytometric techniques may provide opportunities for studying intracellular expression of signaling molecules in different cells within mixed populations.

Intracellular Staining for Analysis of the Expression and Phosphorylation

161

1.3. Technical Challenges to Flow Cytometric Approaches for Measuring STATs

Because of our interest in defining the intracellular signaling pathways activated by type 1 IFNs in NK cells, this group committed to developing protocols to evaluate the levels of STAT1, STAT4, pSTAT1, pSTAT4, IFN␥, and/or combinations of these within NK cells (14). This approach required identification of multiple cell surface determinants with multiple intracellular molecules. Challenges included the availability of specific antibodies and/or specific antibodies for use with fluorochromes having emission spectra that could be separated. In addition, depending on their activation state, STATs can be found in both the cytoplasm and the nucleus. Therefore, staining and permeabilization/fixation methods that could be used in combination needed to be identified to allow detection of the different STATs or pSTATs within cells without destroying the fluorescent function of the fluorochromes and/or the antigenic determinants being detected by the monoclonal antibodies. The methods developed are based on the following: commercially available protocols for staining incorporated nuclear analogues (BD Biosciences), earlier work from our group examining STAT1 levels in T cells (10), reports for detecting pSTATs (15–20), testing using wild type (WT), STAT1-deficient and STAT4-deficient cells (14), and comparisons of results to those obtained with Western blotting (14). Commercially available antibodies to detect pSTATs facilitated the work (BD Biosciences). Antibodies from a variety of sources were screened for intracellular staining of total STAT1 and STAT4 proteins. Monoclonal antibodies detecting full-length STAT1 and STAT4 were being commercially produced and identified (BD Biosciences) and were first used with specific secondary antibodies. Eventually custom reagents with directly conjugated fluorochromes were made. Methanol permeabilization was optimal for intracellular detection of the STATs. It did, however, present problems for detection of cytoplasmic cytokines and the use of particular fluorochromes to identify cell surface markers.

1.4. Methods Developed

The protocols were developed with the goal of characterizing the signaling pathways and responses to type 1 IFN exposure in vivo. The focus was on measuring total STATs, pSTATs, and IFN␥ levels immediately after isolation of the cells from uninfected mice and/or from mice at different times after viral infection. It became clear, however, that optimal protocols for detecting STATs were not compatible with detecting intracellular cytokines. Thus, a second method was developed for identifying STAT1 and IFN␥ within the same cells. Finally, the detection of pSTAT activation in vivo was possible in only low frequencies of cells from immunocompetent mice. This is likely to be in part as a result of the need to capture cells just after cytokine exposure because the biochemical evidence indicates that phosphorylated forms of

162

Miyagi, Lee, and Biron

the STATs are short lived. In addition, however, the experiments carried out also demonstrated that a dynamic regulation of the STAT levels contributed to the selection of particular STATs for activation in vivo. This latter point was proven by examining cellular responsiveness for type 1 IFN activation of STAT1 or STAT4 under control ex vivo conditions after isolation, and by using cells from mice mutated in STAT1 or STAT2. The three methods are presented below: (1) detecting total STATs or total STAT1 with pSTAT4 within freshly isolated NK cells; (2) detecting STAT1 with IFN␥ immediately after isolation; and (3) detecting ex vivo responsiveness to type 1 IFNs with the examination of single parameter pSTATs or total STAT1 with pSTAT4.

2. Materials 2.1. Mice

2.2. General

All protocols require the preparations of splenic leukocytes from wild-type (WT) mice and mice genetically deficient for STAT1 or STAT2. They are available on the 129 background (21, 22) (Taconic Labs). For NK studies, it is best to use mice at 4–9 weeks of age. As indicated, mice can be treated to induce an immunological response. For the protocols detailed below, mice were either uninfected (D0) or infected intraperitoneally with lymphocytic choriomeningitis virus (LCMV) (10, 14). 1. 6-well tissue culture plate (BD Biosciences). 2. 96-well-V-bottom assay plate (Costar). 3. 15 ml polypropylene conical tubes (BD Falcon). 4. 24-well tissue culture plate (BD Biosciences). 5. FACS tube: 1.2 ml polypropylene U-bottom tube (Costar). 6. Sterile frosted glass microscope slides (Fisher Scientific). 7. Nylon mesh (Sefar America). 8. Red Blood Cell Lysing Buffer (Sigma). 9. RPMI-1640 medium (GIBCO). 10. Assay Medium: RPMI-1640 containing 10% FBS, with 1× Penicillin–Streptomycin and 10 mM Hepes Buffer (GIBCO) at pH 7.4. 11. Brefeldin A (Sigma), dissolved in DMSO at 10 mg/ml, aliquoted and stored at −20◦ C. 12. Staining Buffer: PBS containing 2% fetal bovine serum (FBS, Hyclone).

Intracellular Staining for Analysis of the Expression and Phosphorylation

163

13. Goat Block: PBS containing 20% FBS and 10% goat serum (Sigma). 14. 2.4G2 antibody (anti-Fc␥RIII/II; BioXcell) at a working concentration of 1 mg/ml. 15. Cytofix/Cytoperm Buffer (BD Biosciences). 16. Perm Wash Buffer (BD Biosciences). 17. DNase I (Sigma), dissolved in PBS, aliquoted at 1 mg/ml concentration of stock solution and stored at −80◦ C. Dilute with PBS to 300 ␮g/ml for use. 18. Methanol (Fisher Scientific). 2.3. Stimulation

1. In vivo: 2 × 104 plaque-forming units (PFUs) of LCMV. 2. Ex vivo: Recombinant murine IFN␤ (a gift from Biogen Idec, specific activity of 2 × 109 U/mg). Other type 1 IFNs, i.e, IFN␤ or IFN␣, can be used (PBL InterferonSource).

2.4. Flow Cytometric Application and Fluorochromes

There are a number of instruments available for FACS, and these have increasing flexibility in extending parameters for measurement. The methods described here, however, were developed using a FACSCalibur (BD Biosciences) with two lasers having outputs at 15 mW of 488 and 635 nm wavelengths. The results were analyzed using the CellQuest Pro Software (BD Biosciences). The experiments required staining with either three or four different antibodies identified by three or four fluorochromes that could be (1) excited at these wavelengths, (2) result in emission wavelengths distinguishable using available filters, and (3) purchased from commercial sources either directly conjugated to, or available for use in secondary detection steps with, the antibodies needed. In the end, five different fluorochromes were used: fluorescein (FITC), Alexa Fluor 647 (Alexa 647), phycoerythrin (PE), allophycocyanin (APC), and peridinin chlorophyll protein (PerCP). The combinations of four that can be used in individual tests are FITC, PerCP, PE, and Alexa 647; or FITC, PerCP, PE, and APC (see Note 1).

2.5. Antibodies

The reagents detecting mouse leukocyte cell surface markers were developed against mouse determinants. Because NK cells were being identified, two reagents are required for cell surface staining to identify the NK cells and exclude T cells. In the detailed methods, CD49b expression is used to identify NK cells and CD3ε expression is used to exclude T cells. As discussed in the Notes, however, other combinations of antibodies can be used, including expression of NK1.1 to identify NK cells from appropriate strains of mice and expression of TCR␤ to exclude T cells. In contrast to the reagents identifying murine leukocyte subsets, many of the reagents being developed to detect total STATs or pSTATs

164

Miyagi, Lee, and Biron

were prepared against the human molecules, but cross-react with mouse because of the strong homologues between the species. Care should be taken to verify specific reactivity with mouse STATs or pSTATs when new reagents are used. The detailed methods use a custom prepared anti-STAT1 conjugated to PE to detect total STAT1. An anti-STAT4 mouse monoclonal antibody of the IgG1 isotype is identified using a monoclonal antimouse IgG1. This secondary reagent specifically identifies the primary anti-STAT4 when (1) the other reagents used are as listed in the methods because no other is a mouse IgG1 and (2) the splenic leukocytes are prepared at times during infection that precede induction of B-cell isotype switching. The methods describe the use of antibodies recognizing STAT forms phosphorylated at particular tyrosine residues. STATs can be phosphorylated at different tyrosine or serine residues, and the targets of phosphorylation may vary depending on the conditions of stimulation. Thus, modifications of methods to evaluate other sites of phosphorylation maybe of interest to investigators, but will require careful consideration of the biochemical literature. All of the antibodies purchased for flow cytometry and used in the methods are available from BD Biosciences. Some of them can also be purchased from eBioscience. The amounts identified are based on the lots used for developing the protocols. Generally, 30–500 ngs of a particular antibody are used for a single test with 2 × 106 splenic leukocytes. All new lots should be evaluated by titration. 2.5.1. Detection of Total STAT1, Total STAT4, or Total STAT1 with pSTAT4

1. Commercial-specific antibodies: FITC-anti-CD49b (clone DX5); Biotin-anti-CD49b (clone DX5); PerCP-anti-CD3ε (clone 145-2C11); Streptavidin-APC; purified anti-STAT4 (clone 8); FITC-anti-mouse IgG1 (clone A85-1); Alexa 647-anti-STAT4 pY693 (clone 38/pSTAT4); all purchased from BD Biosciences. 2. Customized antibody: PE-anti-STAT1 (C-terminal clone 42), prepared and conjugated with fluorescence dye by BD Biosciences, at a working stock of 2 ␮g/ml. 3. Isotype Controls: PE-Mouse IgG2b (clone 27–35) for PEanti-STAT1; purified mouse IgG1 (clone MOPC-21) for purified anti-STAT4; Alexa 647-Mouse IgG2b (clone 27– 35) for Alexa 647-anti-STAT4 pY693; all purchased from BD Biosciences.

2.5.2. Detection of Total STAT1 with IFN␥

1. Commercial-specific antibodies: FITC-anti-CD49b (clone DX5); PerCP-anti-CD3ε (clone 145-2C11); APC-antiIFN␥ (clone XMG1.2); all antibodies purchased from BD Biosciences.

Intracellular Staining for Analysis of the Expression and Phosphorylation

165

2. Customized antibody: PE-anti-STAT1 (C-terminal clone 42), prepared and conjugated with fluorescence dye by BD Biosciences, at a working stock of 2 ␮g/ml. 3. Isotype Controls: PE-Mouse IgG2b (clone 27–35) for PESTAT1; APC-Rat IgG1 (clone R3-34) for APC-anti-IFN␥; from BD Biosciences. 2.5.3. Detection of Type 1 IFN Responsiveness for pSTAT1 or pSTAT4 Activation or Total STAT1 with pSTAT4 Activation

1. Commercial-specific antibodies: FITC-anti-CD49b (clone DX5); PerCP-anti-CD3ε (clone 145-2C11); PE-antiSTAT1 pY701 (clone 4a); Alexa 647-anti-STAT4 pY693 (clone 38/p-STAT4); all purchased from BD Biosciences. 2. Customized Antibody: PE-anti-STAT1 (C-terminal clone 42), prepared and conjugated with fluorescence dye by BD Biosciences, at a working stock of 2 ␮g/ml. 3. Isotype controls: PE-Mouse IgG2a (clone MOPC-173) for PE-anti-STAT1 pY701; Alexa 647-Mouse IgG2b (clone 27– 35) for Alexa 647-anti-STAT4 pY693; all purchased from BD Biosciences.

3. Methods 3.1. Cell Preparation

Prepare splenic leukocytes (see Note 2) by a regular method (see Note 3). Spleen processing is carried out in a 6-well tissue culture plate, with RPMI medium 1640. Sterile frosted glass microscope slides and nylon mesh, 15 ml polypropylene conical tubes, and Red Blood Cell Lysing Buffer are used.

3.2. Detection of Total STAT1, Total STAT4, or Total STAT1 with pSTAT4

The methods provided below evaluate intracellular levels of total STATs or both STAT1 and pSTAT4 expression immediately after isolation of cells from mice (Fig. 11.1A). They all use methanol permeabilization. The methanol permeabilization is required for detection of STAT4 or pSTATs. The steps detailed will result in samples having been stained with (1) FITCanti-CD49b, PerCP-anti-CD3ε, and PE-anti-STAT1 antibodies; (2) biotin-anti-CD49b antibody detected with streptavidin-APC, PerCP-anti-CD3ε antibody, anti-STAT4 antibody detected with FITC-conjugated anti-mouse IgG1 antibody; and (3) FITC-antiCD49b, PerCP-anti-CD3ε, PE-anti-STAT1, and Alexa 647-antiSTAT4 pY693 antibodies. 1. Resuspend the cells to 2 × 107 cells/ml in cold Staining Buffer. 2. Use 96-well-V-bottomed plate and load 100 ␮l of cell suspension per well (or 2 × 106 cells per test). 3. Centrifuge at 700 g for 3 min with low brake. Remove buffer from plate by flicking plate.

166

Miyagi, Lee, and Biron

4. To prevent non-specific binding of antibodies to Fc receptors during staining, add 200 ␮l of Goat Block with 0.25 ␮l of 2.4G2 antibody to each well and mix well (e.g., mix up and down 10 times) (see Note 4). Incubate for 15 min at 4◦ C. Centrifuge as described above and remove buffer from plate. 5. To label the cells with NK cell surface marker, add 50 ␮l of Staining Buffer containing 0.5 ␮l of FITC-anti-CD49b (for test leading to STAT1 detection) or biotin-conjugated anti-CD49b antibody (for test leading to STAT4 detection) (see Notes 1 and 5) to each well and mix well. Incubate for 15 min at 4◦ C. 6. To wash the cells, add 150 ␮l of Staining Buffer to each well, mix up and down four times. Then centrifuge and remove buffer. 7. To fix the cells, add 100 ␮l of Cytofix/Cytoperm to each well and mix well under the fume hood (see Note 6). Incubate for 20 min at 4◦ C. 8. Wash once with freshly prepared Perm Wash Buffer. 9. To permeabilize the cell, add 200 ␮l of pre-chilled (−20◦ C) pure methanol (see Note 7) to each well and mix well under the fume hood. Incubate for 15 min on ice under the fume hood (see Note 8). 10. Spin down the cells by centrifugation. 11. Flick plate and wash two times with Staining Buffer. 12. Add 50 ␮l of Staining Buffer containing 0.5 ␮l of PerCPconjugated anti-CD3ε antibody (see Notes 1 and 9), with 15 ␮l of PE-conjugated anti-STAT1 antibody (for STAT1 detection), or 0.5 ␮l of streptavidin-APC and 2 ␮l of antiSTAT4 antibody (for STAT4 detection), to each well. For control staining, add the same amount of corresponding isotype controls. In the case of simultaneous staining for STAT1 and pSTAT4, use 15 ␮l of PE-conjugated antiSTAT1 antibody and Alexa 647-conjugated anti-STAT4 pY693 antibody or the same amount of corresponding isotype controls to tests that have been first labeled with FITC-anti-CD49b antibody. 13. Mix well and incubate for 20 min at room temperature. For STAT1 staining, proceed to the step 16. 14. Wash once with Staining Buffer. 15. For STAT4 staining, add 50 ␮l of Staining Buffer containing the secondary antibody, 0.5 ␮l of FITC-conjugated anti-mouse IgG1 antibody to each well and mix well. Incubate for 15 min at 4◦ C.

Intracellular Staining for Analysis of the Expression and Phosphorylation

167

16. Wash once with Staining Buffer. 17. Resuspend in 250 ␮l of Staining Buffer and then transfer to FACS tubes. The samples are ready to be acquired using FACSCalibur (BD Biosciences) with the CellQuest Pro software (BD Biosciences) (see Note 10). An example result in the single staining of total STAT1 or STAT4 is shown in Fig. 11.1B. Another example result in the staining of both total STAT1 and pSTAT4 is shown in Fig. 11.1C.

Fig. 11.1. Methods for identifying intracellular total STATs or total STAT1 with pSTAT4. (A) Leukocytes are prepared from spleens as defined in Section 3.1 and stained to identify the CD49b+CD3ε− NK cells as defined in Section 3.2. Intracellular levels of STAT1, STAT4, or pSTAT4 are evaluated using methanol permeabilization. Total cells are evaluated based on gates set by forward and side scatter. Flow cytometric gates set on CD49b+CD3ε− cell subset allows examination of the NK cell subset. Representation of results with STAT levels in histograms can reveal either general shifts in intensity or mixed populations of positive and negative cells. Dot plots of total STAT1 levels with pSTAT4 reveals diminishing STAT4 activation associated with increasing concentrations of STAT1. (B) Histograms show total levels of STAT1 and STAT4 evaluated separately in total and NK cells prepared from WT mice either uninfected (D0) or LCMV infected for 1.5 or 2.5 days (D1.5 or D2.5). (C) Dot plots show levels of STAT1 and pSTAT4 in NK cell subsets at D0, D1.5, and D2.5 after infection as evaluated in WT, STAT2-, and STAT1-deficient mice. (Panels B and C were originally published in The Journal of Experimental Medicine, 2007, 204, 2382–2396. © Miyagi et al., 2007. doi:10.1084/jem.20070401.)

3.3. Detection of Total STAT1 with IFN␥

The methods presented here evaluate the levels of total STAT1 as compared to IFN␥ expression in individual cells (Fig. 11.2A). Because detection of intracellular cytokines is not possible after treatment with methanol, the fixation and permeabilization method used is dependent on Cytofix/Cytoperm (BD Biosciences) (see Note 11). At this time, it appears that the only signaling molecule detectable by this approach is STAT1. The steps detailed will result in samples having been stained with FITC-antiCD49b, PerCP-anti-CD3ε, PE-anti-STAT1, and APC-anti-IFN␥ antibodies.

168

Miyagi, Lee, and Biron

Fig. 11.2. Methods for detecting total STAT1 and IFN␥ within individual cells. (A) Leukocytes are prepared from spleens, treated with brefeldin A, and stained to identify the CD49b+CD3ε− NK cells as defined in Section 3.3. Intracellular levels of STAT1 along with IFN␥ are evaluated using fixation/permeabilization with Cytofix/Cytoperm. Flow cytometric gates are set on CD49b+CD3ε− cell subset to identify the NK cell subset. Dot plots of total STAT1 levels with IFN␥ reveals increases in STAT1 expression associated with decreasing cytokine expression. (B) Dot plots show NK cells levels of STAT1 and IFN␥ expression at D0, D1.5, and D2.5 after infection as evaluated in WT, STAT2-, STAT1-deficient mice. (Panels B was originally published in The Journal of Experimental Medicine, 2007, 204, 2382–2396. © Miyagi et al., 2007. doi:10.1084/jem.20070401.)

Intracellular Staining for Analysis of the Expression and Phosphorylation

169

1. Prepare cells and resuspend at 1 × 107 cells/ml in Assay Medium containing brefeldin A (see Note 12) in a 15 ml conical tube. 2. Incubate for 4 h at 37◦ C in a CO2 incubator, with mixing after 2 h by tapping the bottom of the tube. 3. To prepare cells for flow cytometry, use a 96-well-Vbottomed plate and load 200 ␮l of cell suspension per well (or 2 × 106 cells per test). Centrifuge and then flick buffer from plate. 4. Wash once with Staining Buffer. 5. To prevent non-specific binding of antibodies to Fc receptors, add 200 ␮l of Goat Block with 0.25 ␮l of 2.4G2 Ab to each well and mix well (see Note 4). Incubate for 15 min at 4◦ C. Centrifuge and then flick buffer from plate. 6. To label the cells with surface markers, add 50 ␮l of Staining Buffer containing 0.5 ␮l of FITC-conjugated antiCD49b antibody and 0.5 ␮l of PerCP-conjugated antiCD3ε antibody to each well and mix well. Incubate for 15 min at 4◦ C. 7. Wash once with Staining Buffer. 8. To fix and permeabilize the cells, add 100 ␮l of Cytofix/Cytoperm to each well and mix well under a fume hood. Incubate for 20 min at 4◦ C. 9. Wash once with freshly prepared Perm Wash Buffer. 10. Add 100 ␮l of DNase solution to each well and mix well. Incubate (preferably with a lid such as aluminum foil) for 1 h at 37◦ C in an incubator. 11. Wash once with the Perm Wash Buffer. 12. Add 50 ␮l of the Perm Wash Buffer containing 0.5 ␮l of APC-conjugated anti-IFN␥ antibody and 0.5 ␮l of PEconjugated anti-STAT1 antibody, or corresponding isotype controls, to each well. Mix well and incubate for 20 min at room temperature. 13. Wash once with the Perm Wash Buffer. 14. Resuspend in 250 ␮l of Staining Buffer and then transfer to FACS tubes. The samples are ready to be acquired using FACSCalibur with the CellQuest Pro software. An example result is shown in Fig. 11.2B. 3.4. Detection of Type 1 IFN Responsiveness for pSTAT1 or pSTAT4 Activation or Total STAT1 with pSTAT4 Activation

The methods presented below evaluate changes in responsiveness to type 1 IFN for activation of STAT1 or STAT4 resulting from conditioning during different ex vivo treatments (Fig. 11.3A). Thus, they require the ex vivo exposure of isolated populations to cytokines prior to staining. The steps detailed will result in samples having been stained with (1)

170

Miyagi, Lee, and Biron

FITC-anti-CD49b, PerCP-anti-CD3ε, and PE-anti-STAT1 pY701 antibodies; (2) FITC-anti-CD49b, PerCP-anti-CD3ε, and Alexa 647-anti-STAT4 pY693 antibodies; and/or (3) FITCanti-CD49b, PerCP-anti-CD3ε, PE-anti-STAT1, and Alexa 647-anti-STAT4 pY693 antibodies. 1. Resuspend the prepared cells to 2 × 107 cells/ml in Assay Medium. 2. Use 24-well-flat-bottomed plate and load 500 ␮l of cell suspension per well. To clear receptors of cytokines bound in vivo and allow cells to return to basal states, incubate for 4 h at 37◦ C in an incubator before testing for responsiveness to type 1 IFNs for STAT1 or STAT4 activation. 3. Add 500 ␮l of Assay Medium with or without a type 1 IFN (e.g., recombinant murine IFN␤ at a final concentration of 10,000 U/ml) for stimulated or unstimulated cells, respectively, to each well. Mix and incubate with a lid for 90 min at 37◦ C in a CO2 incubator. 4. To prepare cells for transfer and staining, mix the content in each well by pipetting up and down 10 times on ice. 5. For the analysis of flow cytometry, use 96-well-V-bottomed plate and load 200 ␮l of medium with stimulated or unstimulated cells to each well (or 2 × 106 cells per test). Centrifuge and then flick buffer from plate. 6. Wash once with cold Staining Buffer. 7. To prevent non-specific binding of antibodies for surface staining, add 200 ␮l of Goat Block with 0.25 ␮l of 2.4G2 Ab to each well and mix well (see Note 4). Incubate for 15 min at 4◦ C. Centrifuge and then flick buffer from plate. 8. To label the cells with NK cell surface marker, add 50 ␮l of Staining Buffer containing 0.5 ␮l of FITC-conjugated anti-CD49b antibody to each well and mix well. Incubate for 15 min at 4◦ C. 9. Wash once with Staining Buffer. 10. To fix the cells, add 100 ␮l of Cytofix/Cytoperm to each well and mix well under a fume hood. Incubate for 20 min at 4◦ C. 11. Wash once with freshly prepared Perm Wash Buffer. 12. To permeabilize the cells, add 200 ␮l of pre-chilled (−20◦ C) pure methanol to each well and mix well under a fume hood. Incubate for 15 min on ice. 13. Spin down the cells by centrifugation. 14. Flick and wash two times with Staining Buffer. 15. Add 50 ␮l of Staining Buffer containing 0.5 ␮l of PerCP-conjugated anti-CD3ε antibody with 15 ␮l of

Intracellular Staining for Analysis of the Expression and Phosphorylation

171

PE-conjugated anti-STAT1 pY701 antibody and/or Alexa 647-conjugated anti-STAT4 pY693, or the same amount of corresponding isotype controls, to each well. If the experiment is to detect total STAT1 and pSTAT4, add 50 ␮l of Staining Buffer containing 0.5 ␮l of PerCP-conjugated anti-CD3ε antibody with 15 ␮l of PE-conjugated antiSTAT1 antibody and Alexa 647-conjugated anti-STAT4 pY693, or the same amount of corresponding isotype controls, to each well. Mix well and incubate for 20 min at room temperature. 16. Wash once with Staining Buffer. 17. Resuspend in 250 ␮l of Staining Buffer and then transfer to FACS tubes. The samples are ready to be acquired using FACSCalibur with the CellQuest Pro software. Examples of single parameter analyses of pSTAT1 and pSTAT4 are shown in Fig. 11.3B and of total STAT1 with pSTAT4 analysis are shown in Fig. 11.3C.

Fig. 11.3. Methods for evaluating type 1 IFN responsiveness with pSTAT1 or pSTAT4 activation. (A) Leukocytes are prepared from spleens, rested in culture to clear cytokine receptors, and treated with type 1 IFN to induce STAT activation. The cells are then stained to identify the CD49b+CD3ε− NK cells as per Section 3.4. Intracellular levels of pSTAT1 or pSTAT4 are evaluated using methanol permeabilization. Total cells are evaluated based on gates set by forward and side scatter. Flow cytometric gates set on CD49b+CD3ε− cells allows examination of the NK cell subset. Dot plots of total STAT1 levels with pSTAT4 reveals differences in STAT4 activation associated with increasing concentrations of STAT1. (B) Ex vivo responsiveness of total and NK cells prepared from WT mice on D0, D1.5, and D2.5 of LCMV infection for STAT1 or STAT4 activation in response to type 1 IFN treatment. (C) Evaluation of ex vivo responsiveness of NK cells to type 1 IFN treatment in association with STAT1 levels on D0 and D2.5 of LCMV infection. Populations for analysis were prepared from WT, STAT2-, and STAT1-deficient mice. (Panels B and C were originally published in The Journal of Experimental Medicine, 2007, 204, 2382–2396. © Miyagi et al., 2007. doi:10.1084/jem.20070401.)

172

Miyagi, Lee, and Biron

4. Notes 1. The PE and PerCP fluorochromes are sensitive to methanol and cannot be used to label cell surface determinants prior to this treatment for cell permeabilization. This may require staining of cell determinants with PE- or PerCPconjugated antibodies after methanol treatment, but it is not always possible to do so because antigenic determinants can also be sensitive to fixation and permeablization treatments. Thus, it is necessary to evaluate expression before and after treatments. 2. This protocol can be adapted for leukocytes from murine liver, and also for human mononuclear cells from peripheral blood with appropriate surface markers (data not shown). 3. The basic protocol for preparation of leukocytes from murine spleen is as follows. Sacrifice mice and remove spleen. Place spleens in a 6-well tissue culture plate containing 5 ml of RPMI medium on ice. Grind spleen between the rough surfaces of frosted glass slides. Filter cell suspension through a nylon mesh and transfer into a 15-ml centrifuge tube. Wash the well with additional 5 ml of cold RPMI medium. Centrifuge at 300 g for 10 min at 4◦ C, and discard supernatant. Disturb pellet by tapping the bottom of the tube, add 1 ml of Red Blood Cell Lysing buffer and vortex briefly. Incubate for 1 min at room temperature. Fill the tube to 10 ml with cold RPMI medium. Filter cell suspension again through nylon mesh and transfer into a 15ml centrifuge tube and count cells. Centrifuge at 300 g for 10 min at 4◦ C, discard supernatant, and resuspend cells at 2 × 107 cells/ml in cold RPMI medium. 4. The combination of goat serum and the 2.4G2 antibody, with directed against receptors for immunoglobulins (FcRs), in the Goat Block is required because it is difficult to inhibit non-specific binding of the antibodies for flow cytometry to the FcRs, particularly when the cells are isolated from infected mice. 5. The APC- as well as FITC- and biotin-conjugated but not PE-conjugated anti-CD49b antibodies (eBioscience) were resistant to methanol exposure. Therefore, APCconjugated anti-CD49b antibody can be adapted in this protocol. Using cells from C57BL/6 background mice, the FITC-, APC-, or biotin-conjugated but not PEconjugated anti-NK1.1 antibodies (PK136) were also resistant to methanol exposure. Moreover, a FITC-, APC-,

Intracellular Staining for Analysis of the Expression and Phosphorylation

173

or biotin-conjugated but not PE-conjugated anti-TCR␤ (clone H57-597) antibodies were found to be resistant to methanol exposure. 6. To fix cells, we take advantage of the Cytofix/Cytoperm from BD because it contains paraformaldehyde (PFA). We found that 4% PFA provided almost identical results with those by Cytofix/Cytoperm so that 4% PFA can be used to fix cells at this step instead of Cytofix/Cytoperm. As PFA is an evaporating toxin, this step must be performed under a fume hood. 7. The treatment with cold pure methanol for the permeabilization of cells was found to provide the best results for intracellular staining of total STATs and pSTATs. The treatment with Cytofix/Cytoperm alone, which is normally used for intracellular staining of cytokines, did not demonstrate a sufficient level of intracellular total STAT4 or pSTATs. 8. Pure methanol is also an evaporating toxin. Thus, this step must be performed under a fume hood. 9. In the methods, anti-CD3ε staining is done after methanol treatment because the fluorochromes available for the combinations required in individual tests limits the commercially available regent to PerCP-anti-CD3ε, and because the CD3ε determinant is still detectable after methanol treatment. 10. At least 100,000 events should be collected within the leukocytes gated for analysis of intracellular molecules in NK cells because the cells are generally at lower than 5% of the total populations. 11. The treatment of methanol following Cytofix/Cytoperm treatment did not allow detection of intracellular IFN␥ expression. It was possible, however, to analyze both STAT1 and IFN␥ because in contrast to STAT4 or the pSTATs, STAT1 levels could be identified following fixation/permeabilization with Cytofix/Cytoperm and DNase treatment. This treatment was already reported to be useful to identify intracellular STAT1 level (10). 12. The treatment with brefeldin A is required for detecting the intracellular cytokine under these conditions. The best final concentration and time of incubation was found to be 5–10 ␮g/ml and 4–6-h of incubation, respectively, in this system.

174

Miyagi, Lee, and Biron

Acknowledgments The authors thank D. Ashley Feldman for review of the manuscript. This work was supported by RO1 grants CA041268 and AI055677 from the National Institutes of Health, a Canadian Institutes of Health Research Fellowship, and funding from the Shinya Foundation. T.M.’s current address is Department of Gastroenterology and Hepatology, Osaka University Graduate School of Medicine, Osaka 565-0871, Japan. References 1. Garcia-Sastre, A., and Biron, C.A. (2006) Type 1 interferons and the virus-host relationship: a lesson in detente. Science 312, 879–882. 2. Biron, C.A. and Sen, G.C. (2007) Innate Immune Responses to Viral Infection. In: Fields Virology Fifth Edition. Knipe D.M. and Howley P.M., eds. Walter Kluwer/Lippincott, Williams & Wilkins, pp. 249–278. 3. Stark, G.R., Kerr, I.M., Williams, B.R., Silverman, R.H., and Schreiber, R.D. (1998) How cells respond to interferons. Annu Rev Biochem 67, 227–264. 4. Platanias, L.C. (2005) Mechanisms of type-Iand type-II-interferon-mediated signalling. Nat Rev Immunol 5, 375–386. 5. Bromberg, J.F., Horvath, C.M., Wen, Z., Schreiber, R.D., and Darnell, Jr., J.E. (1996) Transcriptionally active Stat1 is required for the antiproliferative effects of both interferon alpha and interferon gamma. Proc Natl Acad Sci U S A 93, 7673–7678. 6. Nguyen, K.B., Cousens, L.P., Doughty, L.A., Pien, G.C., Durbin, J.E., and Biron, C.A. (2000) Interferon alpha/beta-mediated inhibition and promotion of interferon gamma: STAT1 resolves a paradox. Nat Immunol 1, 70–76. 7. Nguyen, K.B., Watford, W.T., Salomon, R., Hofmann, S.R., Pien, G.C., Morinobu, A., Gadina, M., O Shea, J.J., and Biron, C.A. (2002) Critical role for STAT4 activation by type 1 interferons in the interferongamma response to viral infection. Science 297, 2063–2066. 8. Tanabe, Y., Nishibori, T., Su, L., Arduini, R.M., Baker, D.P., and David, M. (2005) Cutting edge: role of STAT1, STAT3, and STAT5 in IFN-alpha beta responses in T lymphocytes. J Immunol 174, 609–613.

9. Brierley, M.M., and Fish, E.N. (2002) Review: IFN-alpha/beta receptor interactions to biologic outcomes: understanding the circuitry. J Interferon Cytokine Res 22, 835–845. 10. Gil, M.P., Salomon, R., Louten, J., and Biron, C.A. (2006) Modulation of STAT1 protein levels: a mechanism shaping CD8 Tcell responses in vivo. Blood 107, 987–993. 11. Kaplan, M.H., Sun, Y.L., Hoey, T., and Grusby, M.J. 1996. Impaired IL-12 responses and enhanced development of Th2 cells in Stat4-deficient mice. Nature 382:174–177. 12. Lawless, V.A., Zhang, S., Ozes, O.N., Bruns, H.A., Oldham, I., Hoey, T., Grusby, M.J., and Kaplan, M.H. (2000) Stat4 regulates multiple components of IFN-gammainducing signaling pathways. J Immunol 165, 6803–6808. 13. Herzenberg, L.A., and De Rosa, S.C. (2000) Monoclonal antibodies and the FACS: complementary tools for immunobiology and medicine. Immunol Today 21, 383–390. 14. Miyagi, T., Gil, M.P., Wang, X., Louten, J., Chu, W.M., and Biron, C.A. (2007) High basal STAT4 balanced by STAT1 induction to control type 1 interferon effects in natural killer cells. J Exp Med 204, 2383–2396. 15. Fleisher, T.A., Dorman, S.E., Anderson, J.A., Vail, M., Brown, M.R., and Holland, S.M. (1999) Detection of intracellular phosphorylated STAT-1 by flow cytometry. Clin Immunol 90, 425–430. 16. Uzel, G., Frucht, D.M., Fleisher, T.A., and Holland, S.M. (2001) Detection of intracellular phosphorylated STAT-4 by flow cytometry. Clin Immunol 100, 270–276. 17. Krutzik, P.O., and Nolan, G.P. (2003) Intracellular phospho-protein staining techniques for flow cytometry: monitoring single cell signaling events. Cytometry A 55, 61–70.

Intracellular Staining for Analysis of the Expression and Phosphorylation 18. Krutzik, P.O., Irish, J.M., Nolan, G.P., and Perez, O.D. (2004) Analysis of protein phosphorylation and cellular signaling events by flow cytometry: techniques and clinical applications. Clin Immunol 110, 206–221. 19. Krutzik, P.O., Clutter, M.R., and Nolan, G.P. (2005) Coordinate analysis of murine immune cell surface markers and intracellular phosphoproteins by flow cytometry. J Immunol 175, 2357–2365. 20. Irish, J.M., Hovland, R., Krutzik, P.O., Perez, O.D., Bruserud, O., Gjertsen, B.T., and Nolan, G.P. (2004) Single cell profiling

175

of potentiated phospho-protein networks in cancer cells. Cell 118, 217–228. 21. Meraz, M.A., White, J.M., Sheehan, K.C., Bach, E.A., Rodig, S.J., Dighe, A.S., Kaplan, D.H., Riley, J.K., Greenlund, A.C., Campbell, D., Carver-Moore, K., DuBois, R.N., Clark, R., Aguet, M., and Schreiber, R.D. (1996) Targeted disruption of the Stat1 gene in mice reveals unexpected physiologic specificity in the JAK-STAT signaling pathway. Cell 84, 431–442. 22. Park, C., Li, S., Cha, E., and Schindler, C. (2000) Immune response in Stat2 knockout mice. Immunity 13, 795–804.

Chapter 12 A Model System for Studying NK Cell Receptor Signaling Lukasz K. Chlewicki and Vinay Kumar Abstract Study of NK cell receptor signaling in mouse NK cells has been difficult since there are no clones of murine NK cells. We describe here a model system that overcomes this problem. This system allows the study of many aspects of NK cell receptor function with complete control over the variables that may affect activity such as cis versus trans ligand engagement, homotypic interactions, multiple target types, receptor number, receptor-ligand affinity, and signaling adaptor molecule expression. Although we give examples only for 2B4, Ly49C, and CD48, any NK cell receptors could be studied using these methods. Since many NK cell receptors such as 2B4, CD48, and the Ly49 family can be expressed in T cells, this model system allows the study of not only NK cells but also T cells with NK cell receptors. A standardized system for determining the regulation of NK cell receptor signaling can be important for understanding the anti-tumor activities of NK cells. Key words: Natural killer (NK) cells, T-cell receptor (TCR), target cell, interleukin-2 (IL-2), signaling, activation, inhibition, cross-linking, cis, trans, homotypic.

1. Introduction Natural Killer (NK) cells are lymphocytes which can kill target cells that have become infected with viruses or those that are tumorigenic (1). Lysis operates in a manner similar to cytotoxic T cells (CD8+), where cytolytic molecules from granules including various serine proteases play the predominant role. NK cell surface receptors have been known to share signaling molecules with T cells (reviewed in (2)). Unlike humans, there are no clones of mouse NK cells; hence the mechanism of signaling in NK cells at the clonal level is not completely understood. Activating receptors can recruit small adaptor molecules, such as DAP12, that K.S. Campbell (ed.), Natural Killer Cell Protocols, Methods in Molecular Biology 612, DOI 10.1007/978-1-60761-362-6 12, © Humana Press, a part of Springer Science+Business Media, LLC 2010

177

178

Chlewicki and Kumar

contain immunoreceptor tyrosine-based activation motifs (ITAMs) (3−9). NK cell ITAMs are then phosphorylated by Lck or Fyn, which initiates T-cell-like signaling pathways (10–15). Inhibitory receptors can express immunoreceptor tyrosine-based inhibition motifs (ITIMs) that can recruit negative regulators of NK cell function such as SHP-1, SHP-2, and SHIP (15–17). The 58−/− cell line is a derivative of the DO-11.10.7 mouse T-cell hybridoma. It does not express TCR-␣ and -␤ chains (18) and is negative for CD4, CD8, CD2, 2B4, CD48, and NKG2D, but it expresses CD3 and is H-2K positive (data not shown). Many different NK cell receptors such as 2B4 and the Ly49 family can be expressed on T cells (19, 20), and since 58−/− cells seem to be devoid of many different NK cell receptors, they can be used as a tool to investigate NK cell receptor signaling. The mechanisms involving TCR signaling have been studied extensively; therefore the approach we have taken is to transfect TCR into 58−/− cells along with various NK cell receptors. This allows the investigator to evaluate the function of NK cell receptors when the activating or triggering receptor is TCR. Activation using the TCR, as opposed to some other NK triggering receptor, allows for precise control over stimulation with either cross-linking antibodies, peptide MHC expressed on target cells, or immobilized soluble monomers (21). Target cells can consist of other transfected 58−/− cells (that normally express Kk and Dk , unable to stimulate the 2C TCR) or EL4 cells (Kb with loaded 2C agonist SIYRYGGL peptide, which can stimulate the 2C TCR). Activation through the 2C TCR in this model system results in the secretion of IL-2, which we have shown can be modulated by the stimulation of co-transfected NK cell receptors such as 2B4, CD48, and Ly49C (21–23). Here we outline the generation of fully functioning NK cell receptor expressing 58−/− effector cells and provide several methods for testing their function. 1.1. Generation of Stable Cell Lines in 58−/− and EL4 by Transfections

In order to study the receptor of choice, stable cell lines should be generated in 58−/− cells since this ensures more consistent expression of receptors when compared to transient transfections. We outline the generation of stable cell lines in both 58−/− and EL4 by transfections using cationic lipid reagents.

1.2. Testing Receptor Function by Cross-Linking and Target Cells

Transfected 58−/− cell lines should be screened for the ability to secrete IL-2 by stimulating the triggering receptor, in this case TCR, using either a cross-linking antibody or target cells expressing the TCR ligand. Use of the immobilized ligands allows precise control of local concentration of ligands, which is otherwise difficult to control in purely cell-based systems. Anti-CD3 and anti-CD28 cross-linking have been used extensively in TCR signaling (numerous published reports). Anti-V␤8 (anti-TCR) cross-linking is used in this method instead of anti-CD3 since the

NK Cell Receptor Signaling

179

anti-TCR antibody is more specific for the transfected triggering receptor. A protocol outlining the use of peptide-loaded target cells is also described. 1.3. IL-2 Cytokine ELISA

In addition to killing infected cells, T cells have the ability to secrete the cytokine IL-2. Changes in levels of IL-2 secretion can be monitored by a standard sandwich ELISA to determine the function of the NK cell receptor of interest under various conditions. We expect inhibitory receptors to decrease IL-2 secretion and activating or co-receptors to increase IL-2 secretion.

1.4. Cis Versus Trans Stimulation

An NK cell can encounter multiple cells simultaneously, yet can differentiate these encounters to perform the appropriate functions to each target. The methods presented in this section allow an investigator to test several possible scenarios that an effector cell may encounter in vivo. Under cis conditions, the triggering receptor and the receptor of interest are ligated near each other spatially, mimicking the situation when ligands for both receptors are expressed closely on the same target cell or surface. A subset of cis stimulation, referred to as shared-cis, occurs when the ligand for both triggering receptor (TCR) and protein to be tested is the same molecule, such as class I MHC. Under trans-conditions, the ligand for the TCR or triggering receptor is presented on one surface and the ligand for the receptor to be studied is presented on a different surface. This scenario more closely resembles an encounter that an NK cell may have two target cells, each bearing different ligands.

1.5. Homotypic Interactions

NK cell receptors of the SLAM family, such as SLAM (CD150) (24), engage themselves in a homotypic manner on target cells (25, 26). By using a soluble antibody co-incubation step, homotypic interactions can be identified.

1.6. Intracellular Signaling Experiments

Given the robust growth and ease of transduction of 58−/− , various intracellular signaling studies can be performed. We outline a method using a retroviral transduction system based on the PlatE packaging cell line (27), which both guarantees a higher expression than standard stable cell lines (over-expression) and enables a faster throughput when screening many different signaling adaptor molecules. Over-expression is confirmed using RT-PCR.

2. Materials 2.1. Generation of Stable Cell Lines in 58−/− and EL4 by Transfections

1. Effector cell line: 58−/− (provided by Dr. David Kranz, University of Illinois) or other cell line capable of IL-2 cytokine secretion. 2. Target cell lines: EL4 (ATCC # TIB-39), RMA/S (available from Dr. Vinay Kumar, University of Chicago), or other appropriate target cell line.

180

Chlewicki and Kumar

3. RPMI growth medium: RPMI 1640 medium (Sigma, St. Louis, MO) supplemented with 10% fetal bovine serum (FBS, Sigma), 5 mL non-essential amino acids solution (NEAA, HyClone, Logan, UT), 5 mL sodium pyruvate solution (HyClone), 5 mL Penicillin–Streptomycin solution (HyClone), 5 mL 1 M HEPES (HyClone). 4. Plasmid-containing hEF1␣ promoter (see Note 1), pEF6, or pBUDCE4.1 (Invitrogen, Carlsbad, CA). 5. Optimem-I serum-free medium (Invitrogen). 6. Cationic lipid reagent such as Lipofectamine 2000 (Invitrogen) for 58−/− and EL4 cells. 7. Blasticidin (Invivogen, San Diego, CA), ready to use liquid (10 mg/mL), final working concentration is 5–10 ␮g/mL (1:2000–1:1000). Aliquot and freeze at −20◦ C, do not freeze thaw, and do not store in a frost-free freezer. An enzyme box can be used to store in standard freezer. Shortterm storage at 4◦ C. 8. Puromycin (Invivogen), ready to use liquid (10 mg/mL), final working concentration is 1 ␮g/mL (1:10,000). Longterm storage −20◦ C, short-term storage 4◦ C. 9. G418 (Gemini Biosciences, Woodland, CA), reconstitute to final potency of 50 mg/mL in 1X Dulbecco’s phosphate buffered saline (D-PBS) and filter sterilize (0.22 ␮M). Typical G418 powders are usually about 0.725–0.75 mg potency per mg powder. Final working concentration of active product should be 500 ␮g/mL to 1 mg/mL. This needs to be optimized for each cell line. Store at 4◦ C. 10. Freeze Medium: 45% final FBS, 45% final culture medium for cell line, 10% final DMSO; store at 4◦ C. 2.2. Testing Receptor Function by Cross-Linking and Target Cells

1. Stable 58−/− transfectants, EL4, or RMA/S targets. 2. SIYRYGGL peptide (custom synthesis). 3. RPMI growth medium (see above). 4. Coat Buffer (1X): 26.807 g Na2 HPO4 -7H2 O in 1 L of water, pH 9.0 (No NaCl). 5. PBS (10X): 2.57 g NaH2 PO4 -H2 O, 22.49 g Na2 HPO4 7H2 O, and 87.65 g NaCl per liter of water, pH 7.4. 6. Blocking Buffer: 1X PBS with 1% BSA (10 mg/mL), filter sterilize and store in aliquots at −20◦ C. 7. Wash Buffer: 1X PBS with 0.05% Tween-20, filter sterilize half for coating and store at 4◦ C, the rest is used as an ELISA wash buffer and does not need sterilization (see Note 2)

NK Cell Receptor Signaling

181

8. EIA/RIA-certified high binding 96-well assay plates, sterile with lid (#07-200-721, Corning), these plates work best for coating, since each well is raised and separated from the next by gaps, which greatly reduces cross-contamination between wells. 9. Purified anti-V␤8 cross-linking antibody clone F23.1 (BD Pharmingen, San Jose, CA) or other appropriate purified cross-linking antibody such as 2C11, the anti-CD3 antibody. 10. Mouse IgG control antibody (Jackson Immunoresearch, West Grove, PA). 11. 5 mL polypropylene tubes (BD Falcon). 2.3. IL-2 Cytokine ELISA

1. Coat, wash, and blocking buffers from Section 2.2. 2. EIA/RIA-certified 96-well assay plates non-sterile, no lid (#07-200-39, Corning). 3. Streptavidin-HRP (Jackson Immunoresearch). 4. 1X TMB substrate solution (eBiosciences, San Diego, CA). 5. Stop solution: 3 M Sulfuric acid. 6. Plate reader that reads an absorbance at or near 450 nm. 7. IL-2 matched pair ELISA antibodies: Purified anti-mouse IL-2 clone JES6-1A12 and biotin-conjugated anti-mouse IL-2 clone JES6-5H4 (eBiosciences).

2.4. Cis Versus Trans Stimulation

1. TCR+ 2B4+ , TCR+ Ly49C+ 58−/− effector cell lines. CD48− EL4, CD48+ EL4 target cell lines. 2. RPMI growth medium from Section 2.1. 3. 5 mL polypropylene tubes (BD Falcon). 4. Coat, wash, and blocking buffers from Section 2.2. 5. EIA/RIA-certified high binding 96-well plates with lids (#07-200-721, Corning). 6. Purified anti-V␤8 cross-linking antibody, clone F23.1 (BD Pharmingen), mouse anti-2B4 antibody clone 2B4 (BD Pharmingen). Anti-Ly49C blocking antibody clone 5E6 (BD Pharmingen). 7. Purified class I MHC monomers SIYR/Kb , OVA/Kb (custom synthesis). 8. SIYR (SIYRYGGL) and OVA (SIINFEKL) peptides (custom synthesis).

2.5. Homotypic Interactions

1. TCR+ CD48+ 58−/− effector cell lines, CD48− EL4, CD48+ EL4, 2B4+ EL4 target cell lines. 2. RPMI growth medium from Section 2.1.

182

Chlewicki and Kumar

3. 5 mL polypropylene tubes (# 14-959-11A, BD Falcon). 4. Coat, wash, and blocking buffers from Section 2.2. 5. EIA/RIA-certified high binding 96-well plates (Corning). 6. Purified anti-V␤8 cross-linking antibody, clone F23.1 (BD Pharmingen), mouse anti-CD48 antibody clone HM48-1 (BD Pharmingen). 2.6. Intracellular Signaling Experiments

1. Retroviral transduction: pMIG, pMIGR1 (addgene plasmid# 9044 or available from Dr. Vinay Kumar, University of Chicago) or other retroviral expression plasmid, packaging cell line Plat-E, Polybrene (stock 4 mg/mL, Sigma), 1 M HEPES, cDNA of signaling molecules to be tested. 2. Dulbecco’s modified eagle medium (DMEM, Sigma) supplemented in the same manner as complete RPMI-1640 medium (Section 2.1) with the addition of 50 ␮M 2mercaptoethanol (final concentration). 3. Phosphate buffered saline (PBS). 4. RNA extraction: TRIzol Reagent (Invitrogen), 2-propanol, chloroform, and RNase, DNase-free water. 5. RT-PCR primers for signaling molecules to be tested. 6. First strand cDNA synthesis kit (Fermentas). 7. KOD Hot starts DNA polymerase (EMD Biosciences, Gibbstown, NJ). 8. TAE buffer, Agarose.

3. Methods The 58−/− cell line is quick growing and has been previously grown in RPMI 1640, DMEM, and Advanced-MEM (Invitrogen) media. Optimal transfection results have always been obtained using RPMI 1640 medium. If murine Ly49C function is to be studied, Advanced-MEM medium (Invitrogen) can be used and supplemented with 2% murine serum isolated from B6 mice. Avoid growing 58−/− cells to confluency, since this can adversely affect both cytokine secretion and transfection/transduction efficiencies. 58−/− usually grows in suspension, but can adhere lightly to cell culture plates; it should be passaged between 1:10 and 1:50. As with all receptor work, trypsin−EDTA should never be used to dislodge adherent cells. EL4 target cells grow quicker and to higher densities (2–4 million cells/mL) than 58−/− and should be maintained in complete RPMI 1640 medium. EL4 can be passaged very thin

NK Cell Receptor Signaling

183

and tolerate overcrowding better than 58−/− . As with all transfected cell lines, care should be taken to freeze back as many early passages as possible and not culture cells longer than 6–8 weeks before thawing an earlier stock. The use of this model system depends on the transfected receptor of interest to modulate TCRmediated IL-2 secretion, so check the parental cell line for IL-2 secretion before commencing transfections. 3.1. Generation of Stable Cell Lines in 58−/− and EL4 by Transfection

A stable variant of 58−/− expressing the 2C T-cell receptor was previously generated (21) and requires 1 mg/mL final concentration of the G418 antibiotic for selection. However, any TCR or known triggering receptor capable of activating the T-cell receptor signaling machinery can be used (see Note 3). The following protocol can also be used to transfect proteins into EL4. EL4 requires double the concentration of Blasticidin to kill nonexpressing cells and has higher stable transfection efficiency than 58−/− (personal observation). 1. Check 58−/− cells for expression of receptors of interest by flow cytometry. 2. On the day of transfection, plate 2 million 58−/− cells per transfection in 2 mL of fresh RPMI (complete) medium in each well of a 6-well plate or a 35 mm dish (see Note 4). 3. Following manufacturer’s suggestions, mix 4 ␮g plasmid DNA (see Note 5) with 250 ␮L Optimem-I in a sterile microfuge tube. Mix 10 ␮L of Lipofectamine 2000 with 250 ␮L Optimem-I in a separate tube. Let it stand for 5 min at R.T. 4. Combine contents of two tubes, mix gently, and incubate 20 min at R.T. 5. Add mixture drop wise on top of cells. 6. Gently rock plate back and forth several times to spread transfection medium. 7. Incubate plate/s in cell culture incubator at 37◦ C and 5% CO2 overnight. 8. The following day, expand each well (from 6-well plate) into a new 100 mm plate using 10 mL fresh complete RPMI medium, and incubate overnight. 9. 45–48 h post-transfection, expand the 12.5–100 mL with complete RPMI medium and add selective antibiotics for final 1 mg/mL G418 (for TCR expression) and 5–10 ␮g/mL Blasticidin (for gene to be tested, see Note 6). Aliquot 1 mL into each well of four 24-well plates. Incubate in cell culture incubator until colonies are visible (usually 6–10 days, Note 7).

184

Chlewicki and Kumar

10. Circle and pick colonies (100 ␮L setting on pipette) when densely packed colonies (1–2 mm in size) appear and transfer to wells of a new 24-well plate in 1 mL selection medium. 11. Analyze for expression either by flow cytometry or by western blot after 24–48 h. 12. Expand positive colonies to 6-well plates and test for receptor signaling by receptor cross-linking or with target cells (see Note 8). 13. Clones testing positive for both surface expression and cytokine secretion should be expanded and several aliquots should be frozen in freeze medium and stored at −80◦ C for several days, followed by transfer to liquid nitrogen (see Note 9). 3.2. Testing Receptor Function Using a Cross-Linking Antibody or Target Cells

3.2.1. Receptor Cross-Linking

The following protocol assumes that the murine 2C TCR is used as the triggering receptor for IL-2 secretion. For 2C, antiV␤8.1/2 (F23.1) and anti-V␤8.2 (F23.2) provide a robust signal, comparable to the anti-CD3 antibody 2C11 and are more specific for the transfected TCR (Fig. 12.1A). As with all quantitative experiments, full titrations should be performed with each batch of purified antibody/reagents to determine concentrations required to achieve a 1/2 maximal IL-2 secretion (see Note 10). 1. Dilute the anti-V␤8 (or anti-CD3) antibody to a final concentration of 0.1–0.5 ␮g/mL (determined previously to be the 1/2 maximal concentration) in coat buffer (see Note 11). 2. Aliquot 50 ␮L per well, incubate at least 2 h or preferably overnight at 4◦ C. 3. Block using 100 ␮L sterile PBS-1% BSA, at least 30 min, but not more than 2 h. 4. During blocking step, count cells. 5 × 104 cells are needed per well for optimal IL-2 secretion. A higher number of cells will decrease the availability of ligand in a cross-linking experiment. However, low cell numbers may not yield sufficient IL-2 for detection by ELISA (personal observation). 5. Aliquot appropriate amount of cells, spin down cells, and decant supernatant. 6. Resuspend cell pellet in RPMI growth medium (no selection antibiotics) to 500,000 cells/mL density, which yields 5 × 104 cells for every 100 ␮L per well, and set aside. Small 5 mL polypropylene tubes are ideal when working with multiple cell lines.

NK Cell Receptor Signaling

185

7. Dump out coat and block solution and smack dry on paper towels. 8. Wash plate with sterile wash buffer and smack dry. 9. Add 100 ␮L complete RPMI medium (no selection antibiotics) to each well of plate, use 50 ␮L instead for each well if using a blocking antibody. 10. If using a blocking antibody, add 50 ␮L of a blocking antibody solution (at 40 ␮g/mL, 4× strength) to each well. To wells not receiving blocking antibody, add isotype control IgG antibody. Omit control IgG if no blocking antibody is being used (see Note 12). 11. Aliquot 100 ␮L of each cell suspension per well, for a final volume of 200 ␮L. 12. Incubate plate at 37◦ C and 5% CO2 in cell culture incubator for 26–30 h. 13. Harvest 50 ␮L supernatants (see Note 13) and place directly on ELISA plates pre-coated with capture antibody, and follow protocol in Section 3.3. 14. Plate can be frozen and stored at −20◦ C until ready to assay. 3.2.2. Using Target Cells

Since the triggering receptor is a T-cell receptor, target cells must first be loaded with the appropriate concentration of peptide, which must be determined using serial dilution titrations (Fig. 12.1B). Experiments should always be performed in triplicates. 1. Count target cells and aliquot out the appropriate number. We will initially use a 1:1 effector:target ratio for 58−/− to EL4, therefore 5 × 104 targets per well will be required.

Fig. 12.1. TCR+ 58−/− transfectants secrete IL-2 (measured as Absorbance at 450 nm) (A) TCR+ 58−/− effector cells were stimulated by plate-bound anti-TCR antibodies. Anti-CD3 (2C11), anti-V␤8 (F23.1), and anti-V␤8.2 (F23.2) antibodies were immobilized on plates at various concentrations and TCR+ effectors were added and monitored for IL-2 secretion. (B) TCR+ 58−/− effector cells were stimulated by EL4 target cells loaded with SIYRYGGL peptide titrated at various concentrations and assayed for IL-2 secretion.

186

Chlewicki and Kumar

2. The final assay plate volume will be 200 ␮L per well as with cross-linking. Resuspend target cells in 50 ␮L RPMI growth medium per 5 × 104 cells. This leaves room for peptide, blocking antibodies, and target cells for the remainder 150 ␮L/well. 3. Load EL4 target cells with peptide: Dilute SIYR peptide to 4 × 10−5 M (40 ␮M, this is 4× strength since the sample will be diluted fourfold final in the well) in a fresh polypropylene tube and serially dilute to desired concentrations in RPMI growth medium (see Note 11). 4. Add 50 ␮L of the 4× peptide solution to each well. 5. Add 5 × 104 target cells per well (50 ␮L). 6. If using blocking antibody, add 50 ␮L of a blocking antibody solution (at 40 ␮g/mL, 4×, diluted in RPMI growth medium) to each well. To wells not receiving blocking antibody, add the same quantity of isotype control IgG antibody. Omit control IgG if no blocking antibody is being used. 7. Incubate plate at 37◦ C and 5% CO2 in cell culture incubator for 1 h. 8. Take plate out from incubator and add 5 × 104 effector cells diluted in 50 ␮L of RPMI growth medium per well. 9. Incubate plate at 37◦ C and 5% CO2 in cell culture incubator for 26–30 h. 10. Harvest supernatants and perform ELISA as with crosslinking experiment. 3.3. IL-2 Cytokine ELISA

To detect the secretion of IL-2 by the effector cell lines, perform a standard IL-2 sandwich ELISA. Since the absorbance scale is not linear, a twofold serial dilution of soluble murine IL-2 should be performed on the ELISA plate to quantify. 1. Dilute the IL-2 capture antibody to 2 ␮g/mL final (1:250 of commercial stock) in coat buffer and add 50 ␮L to each well of a 96-well ELISA/EIA/RIA plate. Incubate at RT for 2 h or overnight at 4◦ C (see Note 14). 2. Add 100 ␮l Blocking buffer to each well (see Note 15). Incubate at RT for 1 h. 3. Dump out coat and block solution and smack dry on paper towels. 4. Wash plate with wash buffer and smack dry. 5. Add 50 ␮L of supernatants from experiment to each well. Incubate 2–4 h at RT or overnight at 4◦ C. 6. Remove supernatants from all the wells and smack dry on paper towels. Be careful not to cross-contaminate between wells.

NK Cell Receptor Signaling

187

7. Add 150–200 ␮l wash buffer to each well and then remove as previously. Wash three times. 8. Dilute biotinylated IL-2 sandwich antibody 1:250 in blocking buffer and add 50 ␮l per well. Incubate for 1 h at RT 9. Remove secondary antibody and wash three times. 10. Dilute Streptavidin-HRP 1:1000 and add 50 ␮l per well. Incubate 45 min to 1 h at RT. 11. Remove Streptavidin-HRP and wash four times. 12. Add 50 ␮l 1X TMB substrate solution to each well. Gently tap plate to mix. 13. Put plate on white sheet of paper and wait for blue color to develop (length of time will vary from 10 s−10 min depending on samples, shorter for cross-link, longer for targets). 14. Stop reaction using 50 ␮l of stop solution per well. Gently tap plate to mix. Color will then turn yellow instantly. 15. Read plate on plate reader at 450 nm and analyze data. 16. Since not all cell lines secrete the same level of IL-2, data should be normalized to percent change of IL-2 secretion (of control) when comparing different cell lines. When a single cell line or multiple cell lines with similar IL-2 secretion are used, it is fine to express the data as just raw IL-2 secretion or absorbance at 450 nm. 3.4. Cis Versus Trans Stimulation 3.4.1. Cis Stimulation

Cis stimulation can be replicated experimentally by using target cells bearing ligands for TCR and NK cell receptors or by immobilizing the ligands, either stimulating antibodies or soluble proteins, on the same surface of a plate as indicated in Fig. 12.2A. As an example, we generated 58−/− effector cell lines expressing both 2B4 and 2C TCR and co-cultured the cells with EL4 target cells expressing SIYR/Kb and CD48, which are ligands for the TCR and 2B4, respectively (28–30). This example shows that 2B4 can inhibit TCR-mediated IL-2 secretion in cis with target cells (Fig. 12.2B). The use of a target cell line negative for CD48 and a blocking antibody shows that these effects are specific and are recommended controls. Since immobilization of ligands allows more precise control over ligand concentration compared to using target cells, it is preferred when quantification is important (see Note 16). An example using an anti-TCR antibody along with MHC monomers (for Ly49C+ , TCR+ effectors) is given in Fig. 12.2C. In this example, stimulation of Ly49C, by OVA/Kb monomers leads to a decrease in TCR-mediated IL-2 secretion using the anti-V␤8 antibody F23.1. The following method is similar to method 3.2.1 with an added ligand for the receptor to be tested.

188

Chlewicki and Kumar

Fig. 12.2. Cis stimulation. (A) Cis presentation of two ligands using either a target cell or by immobilization on the surface of a plate. (B) TCR+ 2B4+ 58−/− or control TCR+ 58−/− cells were stimulated by CD48− or CD48+ EL4 target cells loaded with SIYRYGGL peptide in the presence or absence of blocking antibody and assayed for changes in IL-2 secretion. IL-2 secretion is measured as a percentage change of control of IL-2 secreted by CD48− EL4 targets. (C) TCR+ Ly49C+ 58−/− or control TCR+ 58−/− cells were stimulated by 0.3 ␮g/mL anti-TCR and increasing concentrations of OVA/Kb monomers immobilized on the surface of plates and assayed for changes in IL-2 secretion. (D) In sharedcis stimulation, TCR+ Ly49C+ 58−/− or control TCR+ 58−/− cells were stimulated by SIYRYGGL peptide-loaded EL4 cells in the presence or absence of 5E6 blocking antibody and assayed for IL-2 secretion.

1. Dilute the anti-V␤8 antibody to final concentration determined previously to yield a 1/2 maximal IL-2 secretion (usually 0.1–0.5 ␮g/mL) in coat buffer; we will use 0.5 as our example. Be sure to make enough to cover all the wells that will be used to titrate the antibody/protein specific for the receptor that is being studied. Since an ELISA plate has 8 × 12 rows, decide on 7 or 11 different concentrations of the other antibody or protein to use, leave 1 row aside to use as a control having only triggering receptor (anti-TCR) antibody and IgG control. Check manufacturer’s specifications for the total amount of IgG/well that can bind to your plates. If you exceed this number, triggering receptor antibody will compete for binding sites in the wells, altering your results (see Note 17). 2. Using coat buffer containing anti-TCR antibody at a fixed concentration, perform serial dilutions of antibody or protein to test (see Note 18). For this example, we use 4 ␮g/mL as our high point and perform twofold serial dilutions. Leave the last tube only with anti-TCR antibody.

NK Cell Receptor Signaling

189

3. Make up the difference to ∼4.5 ␮g/mL total (anti-TCR plus antibody/protein to test) using isotype IgG control antibody so that anti-TCR antibody in each well has an equal chance of adhering to bottom of well. 4. Add 50 ␮L solution to each well, let incubate at least 2 h RT or overnight at 4◦ C. 5. Follow from step 7 of standard cross-linking method 3.2.1. The 2C TCR and Ly49C bind at different sites of SIYR/Kb (31, 32), thus the same ligand could potentially be presented to both Ly49C and the TCR simultaneously. In our example, Ly49C (on TCR+ , Ly49C+ effectors) can inhibit TCR-mediated IL-2 secretion when engaging SIYR/Kb loaded EL4 targets (Fig. 12.2D). This method of presentation uses the target cell protocol in Section 3.2.2. As with all the methods outlined here, it is recommended to use blocking antibodies to confirm the specificity of the interaction. 3.4.2. Trans Stimulation

Although it is possible to use two target cells simultaneously (Fig. 12.3A), we feel that the difficulty in controlling the growth rate of three cell types could present additional variables

Fig. 12.3. Trans stimulation. (A) Trans presentation of two ligands using either two target cells or by expressing one ligand on a target cell and the other immobilized on the surface of a plate. (B) TCR+ 2B4+ 58−/− or control TCR+ 58−/− cells were stimulated by immobilized anti-TCR cross-linking antibody and CD48+ or CD48− EL4 target cells (no peptide loaded) in the presence or absence of blocking antibodies and assayed for changes in IL-2 secretion. (C) TCR+ Ly49C+ 58−/− or control TCR+ 58−/− cells were stimulated by immobilized anti-TCR cross-linking antibody and OVA peptideloaded EL4 cells and assayed for changes in IL-2 secretion. Control IL-2 secretion was about 3000 pg/mL or an A450 of about 1.2 for all the samples tested.

190

Chlewicki and Kumar

making analysis more difficult. Removing serum to arrest the growth phase also inhibits the secretion of IL-2 (personal observation). To get around this problem the stimulating anti-TCR antibody can be immobilized on plates, while the ligand for either NK cell receptor 2B4 or Ly49C can be expressed on target cells, therefore spatially separating the two ligands. This method combines the protocols of both the cross-linking and the target cell stimulation methods from Sections 3.2.1 and 3.2.2. An example of this method for both 2B4 and Ly49C expressing transfectants is given in Fig. 12.3B and C. In both of these scenarios, no inhibitory signal is generated, even at high concentrations of ligand (see Note 19). 3.5. Homotypic Interactions

To study homotypic interactions, stable cell lines expressing TCR along with the receptor of interest are cultured in the presence of cross-linking anti-TCR antibody and blocking antibodies for this other receptor. This assures that interactions occur only amongst neighboring effector cells. When TCR+ effector cell lines were generated with stable CD48 expression, blocking antibody alone (devoid of any targets) yielded a decrease in IL-2 secretion. Since 58−/− does not normally express ligands for CD48, this suggests homotypic interactions from neighboring effector cells are responsible for this effect (Fig. 12.4A). To confirm this effect and analyze how this homotypic interaction compares to the known heterotypic interaction of 2B4−CD48, we generated separate EL4 cell lines expressing high levels of either CD48 or 2B4 and cultured them with the TCR+ CD48+ effectors. Under these conditions the CD48−CD48 homotypic interaction is not as strong as the CD48−2B4 heterotypic interaction (using 2B4+ EL4 targets) (Fig. 12.4B).

Fig. 12.4. Homotypic interactions. (A) TCR+ CD48+ 58−/− or control TCR+ 2B4+ 58−/− cells were stimulated by immobilized anti-TCR and anti-IgG control antibodies in the presence or absence of blocking antibodies and assayed for changes in IL-2 secretion. (B) TCR+ CD48+ 58−/− cells were cultured with CD48+ EL4, 2B4+ EL4, or control EL4 cells (all loaded with SIYR peptide) in the presence or absence of blocking antibodies and assayed for changes in IL-2 secretion. IL-2 secretion was normalized to treatment with control EL4 targets.

NK Cell Receptor Signaling

3.6. Intracellular Signaling Experiments 3.6.1. Retroviral Transduction of 58−/−

191

Transient transfections of 58−/− are not recommended for overexpression studies since 58−/− divides rapidly and does not transiently express constructs well (personal observation). The common pMIG/MIGR1 retroviral transduction system using the PlatE packaging cell line produces excellent over-expression results, with typical transduction efficiencies from 45 to 95% and at higher levels than transfection. These can easily be sorted for GFP+ populations by FACS. 1. Maintain PlatE cells in complete DMEM supplemented with 50 ␮M 2-mercaptoethanol, 1 ␮g/mL Puromycin and 10 ␮g/mL Blasticidin. 2. 24 h before transfection, plate ∼2–3 million cells per 100 mm dish in complete DMEM medium, incubate overnight at 37ºC and 5% CO2 . 3. Remove medium and add 3 mL fresh complete DMEM medium without selection antibiotics. 4. In each microfuge tube, combine 5 ␮g DNA in 300 ␮L 150 mM NaCl and mix. 5. Add 16.45 ␮L ExGen 500, vortex for 10 s, and incubate for 10 min at RT. 6. Add mixture on top of cells drop wise, rock back and forth to spread out. 7. Incubate overnight at 37ºC and 5% CO2 . 8. Remove supernatant and add 3 mL of fresh complete DMEM medium and incubate overnight at 37ºC and 5% CO2 . 9. Harvest supernatant in the morning (now 41–46 h posttransfection), add 3 mL fresh medium, and incubate until the end of the day. 10. Harvest supernatant, add fresh medium, and incubate overnight at 37ºC and 5% CO2 . 11. Harvest supernatant in the morning and discard the cells. 12. Filter supernatant with a 0.45 ␮M syringe filter and freeze at −80◦ C (long term) or store at 4◦ C for no longer than several days if to be used soon. 13. Prepare 58−/− cells: Keep cells in log growth phase before cells are overcrowded and before medium turns yellow. Otherwise passage cells again. 14. Plate 1 million cells/mL in each well of a 24-well plate (see Note 20). 15. Add 2 ␮L of Polybrene (stock at 4 mg/mL) and 20 ␮L of 1 M HEPES. 16. Add 1 volume of viral supernatant (1:1 mL cells) and mix.

192

Chlewicki and Kumar

17. Centrifuge at 30ºC at 1300 g for 90 min. 18. Loosen cells by pipetting. 19. Incubate overnight at 37ºC and 5% CO2 . 20. Replace medium with fresh medium and expand culture to 6-well plates. 21. Check for GFP expression at 48 h post-transduction and sort if necessary. 22. Perform experiments as outlined earlier. 3.6.2. RT-PCR for Expression

After transduction, the expression of the transduced gene should be confirmed either by western blot for protein expression or by RT-PCR for transcript level and compared to the parental cell line with a mock transduction of GFP-containing vector control. 1. Pellet 1 × 107 cells by centrifugation, aspirate medium, wash once with 1 × PBS. 2. Add 1 mL TRIzol for every 1 × 107 cells, pipette up and down to shear precipitate and obtain smooth, wellsuspended solution. 3. Incubate 5 min at RT or freeze cells at −80◦ C for up to 1 month. 4. Add 0.2 mL chloroform per 1 mL TRIZOL. 5. Shake tubes vigorously by hand for 15 s and incubate at RT for 2–3 min. 6. Centrifuge samples at 12,000 g for 15 min at 4◦ C. 7. Recover upper phase into a new tube. 8. Add 0.5 mL 2-propanol per 1 mL TRIZOL (to upper phase in new tube) and incubate for 10 min at RT. 9. Centrifuge at 12,000 g for 10 min at 4◦ C. 10. Remove supernatant (aspirate). 11. Wash with 1 mL 75% ethanol. 12. Centrifuge at no more than 7500 g for 5 min at 4◦ C. 13. Aspirate very carefully. 14. Air dry pellet for 5–15 min. 15. Dissolve RNA in 20–60 ␮L DEPC-water or RNase-, DNase-free water. 16. Measure absorbance at 260 nm. 17. Perform reverse transcription according to Fermentas kit instructions. Usually we use about 2 ␮g of total RNA for each reaction. 2 ␮L of this will subsequently be used for the PCR (see Note 21).

NK Cell Receptor Signaling

193

Fig. 12.5. Cytoplasmic signaling studies. TCR+ 2B4+ 58−/− cells were stimulated by immobilized anti-TCR cross-linking antibody and increasing concentrations of anti-2B4 cross-linking antibody and assayed for changes in IL-2 secretion. Clones were transduced to over-express EAT-2A (EAT-2A ex) or the control GFP plasmid. Equal quantities of cDNA were amplified to estimate relative level of EAT-2A transcript, GAPDH was amplified as a control.

18. Determine conditions for PCR and perform according to the protocol provided with the KOD polymerase, trying to optimize the annealing temperature that is best for the primer pair. The user should optimize the annealing temperature and primers to achieve one solid band, also the number of cycles may need to be adjusted to 30 or fewer so, as not to reach saturation (see Note 22). 19. Make serial dilutions of PCR product and resolve on TAEAgarose gel. After expression of the adaptor molecule/construct is confirmed, proceed with experiments as outlined earlier. In our example we over-expressed the SAP-like adaptor molecule EAT-2A and showed that it helped increase TCR-mediated IL-2 secretion by receptor cross-linking (Fig. 12.5). This effect is similar, but weaker in magnitude than what we reported previously using the adaptor molecule SAP, which is very similar to EAT-2A (33).

4. Notes 4.1. Generation of Stable Cell Lines in 58−/− and EL4 by Transfections

1. A number of different promoters have been used in 58−/− cells. However, the hEF1␣ promoter offers the most reliability. Expression profiles with this promoter can range several orders of magnitude as measured by flow cytometry. CMV-based promoters from non-retroviral vectors are not reliable in this cell line (personal observation). Always include a Kozak initiation sequence before your ATG to assure expression. Epitope tags such as myc, 6×-His, HA, etc., are not recommended initially for receptors, since they may interfere with ligand binding. Unless you are certain, do not use them. A myc-tag on Ly49C does not interfere

194

Chlewicki and Kumar

with MHC binding on target cells (personal observation). Caution should be used on cytoplasmic proteins if the tag is placed near any functional domains. A myc-tag, however, enables rapid isolation of proteins by immunoprecipitation. 2. 1X PBS with either BSA or with Tween contaminates quickly. Both of these should be kept sterile for use in cell culture coating plates. However, PBST used for ELISA washes does not need to be sterile. DO NOT add Na-Azide to any diluted reagent that will be used for ELISA, as this will completely deactivate the HRP. 3. 2C is chosen since it recognizes EL4 target cells loaded with the superagonist peptide SIYRYGGL. Any TCR capable of recognizing a target cell loaded with a peptide of choice that can be controlled by the user may be used, including activating NK cell receptors. 4. This protocol uses a cationic lipid reagent, but 58−/− can also be transfected using electroporation; however, cells should be cultured in RPMI medium with 1% DMSO for 24 h post-electroporation. Cationic lipid transfections are used in lieu of retroviral transduction so as to leave the cells GFP negative to allow later over-expression of adaptor molecules. One should also use unique antibiotics for each gene or simply inactivate one of the resistance genes with a restriction enzyme if a co-transfection is to be used. For co-transfections with two unique drugs, use 2 ␮g of each DNA for each well of a 6-well format. For co-transfections with an inactivated gene, use 4 ␮g total DNA with a 10:1 ratio of inactivated: active resistance gene. ExGen500 (Fermentas) may also be used with reduced transfection efficiency. 5. Plasmid DNA should have a 280/260 ratio of at least 1.8. It is not necessary to perform a Maxi-prep (Qiagen), since several mini-preps will provide ample, highly purified DNA for multiple transfections. Mini-prep kits from Qiagen, Fermentas, and Eppendorf all yield highly purified DNA suitable for transfections. 6. The exact concentration of selection antibiotics varies greatly with culture media composition and cell density and should first be determined by plating about 100,000 cells per well in a 24-well plate at various concentrations of antibiotics. In addition, 2-mercaptoethanol in the medium (at 50–100 ␮M final) usually requires double the concentration of Blasticidin and is omitted in our formulation. The lowest concentration of antibiotic that kills all cells within 24 h (for Blasticidin and Puromycin) is the amount to use for selection. 7. Avoid picking colonies that appear after 12–14 days, since the effectiveness of Blasticidin diminishes in culture after this

NK Cell Receptor Signaling

195

period. Also avoid letting the medium turn yellow at any time during the course of culture since this may dramatically decrease the ability of the cells to secrete IL-2. 8. As with any cell line for generating stables, multiple lines should be isolated and tested before any conclusions are drawn. Always attempt to use the largest pool possible of stable cell lines to assure accuracy. We usually test between 15 and 30 separate cell lines at various surface expression profiles before proceeding with experiments. 9. Not all cell lines will secrete IL-2, even if the parental line did. If no cell lines secrete IL-2, check for triggering receptor expression and another batch of cross-linking antibody. If still no luck, thaw a new aliquot of parental cell line. Secretion of IL-2 is lost from time to time and certain cell lines may lose and reacquire the ability to secrete IL-2. 4.2. Testing Receptor Function by Cross-Linking and Target Cells

10. A 1/2 maximal concentration of antibody for triggering receptor should be used when one wants to study the effects that treatment has on IL-2 secretion since this will give the highest signal-to-noise ratio in the experiment. Maximal concentrations should not be used since a receptor or adaptor molecule providing a co-activating signal may show no response, likewise too low of a concentration will mask an inhibitory signal. Anti-CD3 should not be used when EL4 targets are employed, since these cells also stain positive for CD3. Antibody purity/activity is critical and should be analyzed for every new batch of antibody. Custom-purified antibodies (in-lab) have consistently performed better than commercial antibodies (two- to fourfold) in these sensitive assays and are preferred (personal observation). 11. Twofold serial dilutions give more points; however, many wells are required. ∼3.16-fold serial dilutions can be used to give two points per log (i.e., 3.162 = ∼10). 12. Blocking antibodies can be added to any of the steps, as long as the blocking antibody is added at least before or with the effector cells. Blocking antibody may be preincubated with effector cells if specific for transfected receptor or with target cells if specific for a target cell ligand. 10 ␮g/mL final blocking concentration is usually sufficient to block. Not all blocking antibodies are effective at blocking their respective receptor completely. For example, the commercially available anti-2B4 antibody does not completely block 2B4−CD48 interactions (34). Higher concentrations may be required or blockade of the ligand may be necessary.

196

Chlewicki and Kumar

13. It is not necessary to centrifuge the supernatants since cells will have settled to the well bottom during the experiment; however, it will not hurt to do so or to transfer supernatants to new plates before freezing. 4.3. IL-2 Cytokine ELISA

14. A capture antibody concentration of 0.2–10 ␮g/mL is usually used. Most standard binding ELISA plates will saturate closer to 5 ␮g/mL (check each spec sheet), therefore higher concentrations can actually lower signal since they can bind soluble antigen left over and be washed away during the wash steps. 15. It is not necessary to dump out the coat step, since the concentration of soluble proteins in 1% BSA of 10 mg/mL is much higher than the single well binding capacity. However, the user can choose to dump this out before blocking.

4.4. Cis Versus Trans Stimulation

16. It is very important to only titrate one antibody or protein at a time, otherwise data analysis is very difficult. Usually one would want to use a constant concentration of crosslinking antibody for the triggering receptor and then vary the concentration of the proteins or cross-linking antibodies for the receptor to be tested. One should aim to stimulate the triggering receptor with a concentration that is near the 1/2 maximal required for stimulation, as this concentration allows for a very sharp rise or fall in IL-2 secretion indicative of activation or inhibition, respectively. Always have proper controls for every parameter tested, including a control for spontaneous IL-2 release. 17. Each well must contain the same amount of total IgG or protein, otherwise differences may appear in the wells simply arising from competition for binding spots. If the manufacturer states that only 4.5 ␮g/mL can bind per well, each well should contain 4.5 ␮g/mL total IgG maximum, this includes both antibody for triggering receptor (TCR) and the antibody or protein that will be tested along with control IgG to make up the difference. 18. Antibodies should be at the highest purity and activity possible since these experiments use low concentrations to see dramatic effects, batch differences have been detected before in antibodies of up to —three- to fourfold. To achieve the highest possible replication of results, always test each batch of antibody. If growing hybridoma supernatants yourself, be sure to use serum that has been stripped of any possible bovine IgG. 19. Trans stimulation can also be achieved by immobilizing an antibody/ligand for the receptor of interest and target cells for TCR.

NK Cell Receptor Signaling

4.5. Intracellular Signaling Experiments

197

20. A higher density of cells helps increase retroviral transduction. Transductions should be performed when cells are still actively dividing. 21. It is far cheaper to purchase the kit components separately; however, we list it here to make it easier for novices. 22. One should aim to have as clean a PCR product as possible, with one intense band on the gel. By using 30 or fewer cycles, the largest difference between samples can be obtained, which helps data analysis. In samples with around 35 cycles, the difference between high and low expressers is not as dramatic, since PCR reactions have plateaued.

Acknowledgments We would like to thank David M Kranz for providing the 58−/− hybridoma and the F23.2 antibody, Cox Terhorst for providing the EAT-2 cDNA, and Hans Schreiber for providing SIYRYGGL peptide. References 1. Trinchieri, G. (1989) Biology of natural killer cells. Adv Immunol 47, 187–376. 2. Lanier, L. L. (2003) Natural killer cell receptor signaling. Curr Opin Immunol 15, 308–314. 3. Lanier, L. L., Corliss, B. C., Wu, J., Leong, C., and Phillips, J. H. (1998) Immunoreceptor DAP12 bearing a tyrosine-based activation motif is involved in activating NK cells. Nature 391, 703–707. 4. Smith, K. M., Wu, J., Bakker, A. B., Phillips, J. H., and Lanier, L. L. (1998) Ly-49D and Ly-49H associate with mouse DAP12 and form activating receptors. J Immunol 161, 7–10. 5. Lanier, L. L., Corliss, B., Wu, J., and Phillips, J. H. (1998) Association of DAP12 with activating CD94/NKG2C NK cell receptors. Immunity 8, 693–701. 6. McVicar, D. W., Taylor, L. S., Gosselin, P., Willette-Brown, J., Mikhael, A. I., Geahlen, R. L., Nakamura, M. C., Linnemeyer, P., Seaman, W. E., Anderson, S. K., Ortaldo, J. R., and Mason, L. H. (1998) DAP12mediated signal transduction in natural killer cells. A dominant role for the Syk protein-tyrosine kinase. J Biol Chem 273, 32934–32942. 7. Campbell, K. S., and Colonna, M. (1999) DAP12: a key accessory protein for relaying signals by natural killer cell receptors. Int J Biochem Cell Biol 31, 631–636.

8. Gosselin, P., Mason, L. H., Willette-Brown, J., Ortaldo, J. R., McVicar, D. W., and Anderson, S. K. (1999) Induction of DAP12 phosphorylation, calcium mobilization, and cytokine secretion by Ly49H. J Leukoc Biol 66, 165–171. 9. Wu, J., Cherwinski, H., Spies, T., Phillips, J. H., and Lanier, L. L. (2000) DAP10 and DAP12 form distinct, but functionally cooperative, receptor complexes in natural killer cells. J Exp Med 192, 1059–1068. 10. Lowin-Kropf, B., Kunz, B., Schneider, P., and Held, W. (2002) A role for the src family kinase Fyn in NK cell activation and the formation of the repertoire of Ly49 receptors. Eur J Immunol 32, 773–782. 11. Gadue, P., Morton, N., and Stein, P. L. (1999) The Src family tyrosine kinase Fyn regulates natural killer T cell development. J Exp Med 190, 1189–1196. 12. Marti, F., Xu, C. W., Selvakumar, A., Brent, R., Dupont, B., and King, P. D. (1998) LCKphosphorylated human killer cell-inhibitory receptors recruit and activate phosphatidylinositol 3-kinase. Proc Natl Acad Sci U S A 95, 11810–11815. 13. Brumbaugh, K. M., Binstadt, B. A., Billadeau, D. D., Schoon, R. A., Dick, C. J., Ten, R. M., and Leibson, P. J. (1997) Functional role for Syk tyrosine kinase in natural killer cell-mediated natural cytotoxicity. J Exp Med 186, 1965–1974.

198

Chlewicki and Kumar

14. Adunyah, S. E., Wheeler, B. J., and Cooper, R. S. (1997) Evidence for the involvement of LCK and MAP kinase (ERK-1) in the signal transduction mechanism of interleukin15. Biochem Biophys Res Commun 232, 754–758. 15. Binstadt, B. A., Brumbaugh, K. M., Dick, C. J., Scharenberg, A. M., Williams, B. L., Colonna, M., Lanier, L. L., Kinet, J. P., Abraham, R. T., and Leibson, P. J. (1996) Sequential involvement of Lck and SHP-1 with MHC-recognizing receptors on NK cells inhibits FcR-initiated tyrosine kinase activation. Immunity 5, 629–638. 16. Binstadt, B. A., Billadeau, D. D., Jevremovic, D., Williams, B. L., Fang, N., Yi, T., Koretzky, G. A., Abraham, R. T., and Leibson, P. J. (1998) SLP-76 is a direct substrate of SHP-1 recruited to killer cell inhibitory receptors. J Biol Chem 273, 27518–27523. 17. Tangye, S. G., Lazetic, S., Woollatt, E., Sutherland, G. R., Lanier, L. L., and Phillips, J. H. (1999) Cutting edge: human 2B4, an activating NK cell receptor, recruits the protein tyrosine phosphatase SHP-2 and the adaptor signaling protein SAP. J Immunol 162, 6981–6985. 18. Letourneur F. M. B. (1989) Derivation of a T cell hybridoma variant deprived of functional T cell receptor alpha and beta chain transcripts reveals a nonfunctional alpha-mRNA of BW5147 origin. Eur J Immunol 12, 2269–2274. 19. McMahon, C. W., and Raulet, D. H. (2001) Expression and function of NK cell receptors in CD8+ T cells. Curr Opin Immunol 13, 465–470. 20. Ugolini, S., and Vivier, E. (2000) Regulation of T cell function by NK cell receptors for classical MHC class I molecules. Curr Opin Immunol 12, 295–300. 21. Holler, P. D., Chlewicki, L. K., and Kranz, D. M. (2003) TCRs with high affinity for foreign pMHC show self-reactivity. Nat Immunol 4, 55–62. 22. Holler, P. D. and Kranz, D. M. (2003) Quantitative Analysis of the Contribution of TCR/pepMHC Affinity and CD8 to T Cell Activation. Immunity 18, 255–264. 23. Holler, P. D., Lim, A. R., Cho, B. K., Rund, L. A., and Kranz, D. M. (2001) CD8(-) T cell transfectants that express a high affinity T cell receptor exhibit enhanced peptidedependent activation. J Exp Med 194, 1043–1052.

24. Wang, N., Morra, M., Wu, C., Gullo, C., Howie, D., Coyle, T., Engel, P., and Terhorst, C. (2001) CD150 is a member of a family of genes that encode glycoproteins on the surface of hematopoietic cells. Immunogenetics 53, 382–394. 25. Engel, P., Eck, M. J., and Terhorst, C. (2003) The SAP and SLAM families in immune responses and X-linked lymphoproliferative disease. Nat Rev Immunol 3, 813–821. 26. Sidorenko, S. P., and Clark, E. A. (2003) The dual-function CD150 receptor subfamily: the viral attraction. Nat Immunol 4, 19–24. 27. Morita, S., Kojima, T., and Kitamura, T. (2000) Plat-E: an efficient and stable system for transient packaging of retroviruses. Gene Ther 7, 1063–1066. 28. Udaka, K., Wiesmuller, K. H., Kienle, S., Jung, G., and Walden, P. (1996) Self-MHCrestricted peptides recognized by an alloreactive T lymphocyte clone. J Immunol 157, 670–678. 29. Latchman, Y., McKay, P. F., and Reiser, H. (1998) Identification of the 2B4 molecule as a counter-receptor for CD48. J Immunol 161, 5809–5812. 30. Brown, M. H., Boles, K., van der Merwe, P. A., Kumar, V., Mathew, P. A., and Barclay, A. N. (1998) 2B4, the natural killer and T cell immunoglobulin superfamily surface protein, is a ligand for CD48. J Exp Med 188, 2083– 2090. 31. Degano, M., Garcia, K. C., Apostolopoulos, V., Rudolph, M. G., Teyton, L., and Wilson, I. A. (2000) A functional hot spot for antigen recognition in a superagonist TCR/MHC complex. Immunity 12, 251–261. 32. Dam, J., Guan, R., Natarajan, K., Dimasi, N., Chlewicki, L. K., Kranz, D. M., Schuck, P., Margulies, D. H., and Mariuzza, R. A. (2003) Variable MHC class I engagement by Ly49 natural killer cell receptors demonstrated by the crystal structure of Ly49C bound to H-2 K(b). Nat Immunol 4, 1213– 1222. 33. Chlewicki, L. K., Velikovsky, C. A., Balakrishnan, V., Mariuzza, R. A., and Kumar, V. (2008) Molecular basis of the dual functions of 2B4 (CD244). J Immunol 180, 8159– 8167. 34. Clarkson, N. G., Simmonds, S. J., Puklavec, M. J., and Brown, M. H. (2007) Direct and indirect interactions of the cytoplasmic region of CD244 (2B4) in mice and humans with FYN kinase. J Biol Chem 282, 25385– 25394.

Chapter 13 Expression of cDNAs in Human Natural Killer Cell Lines by Retroviral Transduction S. M. Shahjahan Miah and Kerry S. Campbell Abstract Human NK-like cell lines are difficult to transfect using standard mammalian expression vectors and conventional transfection protocols, but they are susceptible to retroviral transduction as a means to introduce cDNAs. Our laboratory has exploited this technique to study a number of receptors in human NK cell lines. The method utilizes a bicistronic retroviral vector that co-expresses either drug resistance or enhanced green fluorescent protein (EGFP) in parallel with the gene of interest. After a single infection with recombinant retrovirus, transduced NK cells can be sorted for expression of EGFP or the transduced cell surface marker. Alternatively, cells expressing the transduced cDNAs can be selected for by treatment with neomycin, puromycin, or hygromycin. Using this method, the sorted/selected cells uniformly express the gene of interest and the expression is stable for many weeks of culture. Key words: Retroviral transduction, NK cell lines, EGFP.

1. Introduction A number of transformed human natural killer (NK)-like cell lines have been adapted to culture and provide valuable models for studying NK cell function and signal transduction. Most of these cell lines lack expression of many normal NK cell surface receptors, especially killer cell Ig-like receptors (KIR), CD94/NKG2 heterodimers, and CD16. Therefore, it is very attractive to express these receptors in the transformed NK cell lines to examine molecular functions. Unfortunately, the available NK cell lines are highly resistant to transfection with standard mammalian expression vectors. K.S. Campbell (ed.), Natural Killer Cell Protocols, Methods in Molecular Biology 612, DOI 10.1007/978-1-60761-362-6 13, © Humana Press, a part of Springer Science+Business Media, LLC 2010

199

200

Shahjahan Miah and Campbell

In the late 1990s, several groups successfully expressed cDNAs in NK-like cell lines by retroviral transduction. Amphotropic retroviral transduction was first successfully used to introduce the IL-2 cDNA into the human NK-92 cell line (1). Cohen et al. subsequently introduced the ecotropic receptor into the human YTS cell line, which permitted susceptibility to transduction with mouse ecotropic retrovirus [(2) and further described in chapter 17 by Mandelboim in this volume]. Amphotropic retroviruses can infect most mammalian cells, including human, and can therefore be biohazardous to laboratory personnel. On the other hand, ecotropic virus can only infect mouse and rat cells, and hence, working with ecotropicsensitive YTS cells has the advantage of avoiding some biosafety issues. Nonetheless, many versions of replication-incompetent amphotropic retrovirus have been engineered, and these strains are not particularly dangerous if carefully handled under BSL2 biosafety conditions, which are achievable in most modern biology laboratories (see Note 1). Our laboratory has exploited amphotropic retroviral transduction to introduce a number of cDNAs into a variety of NK-like cell lines, including NK-92, NKL, NK3.3, and KHYG-1 (3–7), and our optimized transduction protocol is detailed in this chapter (see Note 2). Retroviral vectors derived from Moloney murine leukemia virus (MMLV) are the most widely used and allow the delivery of genes to dividing mammalian cells. The expression of a cloned gene of interest is strongly promoted by the flanking longterminal repeat (LTR) elements within these vectors, and the vectors integrate into the cell’s chromosomes, thereby establishing long-term, stable protein expression after a single transduction procedure. Retroviral infection generates a polyclonal transduced population, since the distinct random chromosomal integration events occur in multiple clones. The polyclonal nature of the transduced population thereby dilutes potential bias that may be introduced by influencing an integrated vector’s promoter on genes adjacent to the integration site when studying monoclonal transfected populations. To allow purification of transduced cells that express the gene of interest, retroviral vectors may also encode selectable markers, such as neomycin-, puromycin-, or hygromycin-resistance genes, or a fluorescent marker, especially enhanced green fluorescence protein (EGFP). Standard mammalian expression vectors contain independent transcription units for the selectable marker and the gene of interest. A number of bicistronic retroviral vectors have been developed; however, that contain an internal ribosome entry site (IRES), which allows both the marker and the gene of interest to be expressed independently from a single transcript. Our retroviral transduction of NK cell lines has utilized a system developed and made readily available by Dr. Garry Nolan

Retroviral Transduction of Human NK Cell Lines

201

(Stanford University, Stanford, CA). Detailed information about this system can be found at www.stanford.edu/group/nolan/. This system utilizes the retroviral vector pBMN-IRES-EGFP, which co-expresses EGFP and the Phoenix-amphotropic packaging cell line. The Phoenix-amphotropic packaging lines encode three major retroviral elements: (1) pol, which functions as a reverse transcriptase, RNase H, and integrase, (2) gag, which is a large protein that is processed into viral matrix and core structures, and (3) the envelope (env) protein, which exists in the lipid layer and determines viral tropism. When Phoenix cells are transfected with the pBMN plasmid, these elements package a replication-incompetent retrovirus that is secreted into the culture medium and used to infect the NK cell lines.

2. Materials 1. Phoenix-amphotropic retroviral packaging cell line: The cell line can be obtained from Dr. Garry Nolan. It is based on the 293T cell line, a human embryonic kidney fibroblast that is transformed with adenovirus E1a and carries a temperature-sensitive T antigen co-selected with neomycin. The Phoenix-ampho cell line was created by introducing genes producing gag-pol and env for infection of most mammalian cells, including human. Gag-pol and envelope use different promoters to minimize their interrecombination potential (see Note 3). 2. NK-like cell lines: NK-92 (ATCC #CRL-2407), NKL (obtained from Dr. Marco Colonna, Washington University, St. Louis, MO), NK3.3 (obtained from Dr. Jacki Kornbluth, St. Louis University School of Medicine, St. Louis, MO), and KHYG-1 (obtained from Health Science Research Resources Bank, Japan Health Sciences Foundation, Osaka, Japan, #JCRB0156). 3. pBMN-IRES-EGFP retroviral vector (also obtained from Dr. Garry Nolan) with cDNA of interest sub-cloned into appropriate restriction sites. 4. Stbl2 bacteria (Invitrogen) (see Note 4). 5. Complete RPMI medium: RPMI 1640 medium (Life Technologies, Rockville, MD) containing 10% heatinactivated fetal bovine serum (FBS) (HyClone Laboratories, Logan, UT), 2 mM L-glutamine, 100 ␮g/ml penicillin, 100 ␮g/ml streptomycin, 10 mM HEPES (pH 7.4), 1 mM sodium pyruvate (all supplements from Life Technologies or Mediatech, Manassas, VA), and 50 ␮M 2mercaptoethanol.

202

Shahjahan Miah and Campbell

6. Folic acid stock solution (2 mM): Mix 0.221 g folic acid (Gibco) in 250 ml ddH2 O. Autoclave, wrap in foil, and store at 4◦ C. The final preparation will be a suspension and needs to be warmed to 70◦ C for 10 min prior to addition to the complete ␣-MEM preparation. 7. Myo-inositol stock solution (20 mM): Mix 0.901 g myoinositol (Sigma-Aldrich, St. Louis, MO) in 250 ml ddH2 O, sterile filter, and store at 4◦ C. 8. Complete ␣-MEM: ␣-minimal essential medium (MEM; Life Technologies, Rockville, MD) containing 10% heatinactivated FBS (HyClone Laboratories, Logan, UT), 10% heat-inactivated horse serum (Life Technologies), 2 mM Lglutamate, 100 ␮g/ml penicillin, 100 ␮g/ml streptomycin, 1 mM sodium pyruvate, 100 ␮M 2-mercaptoethanol, 2.5 ␮M folic acid (Sigma-Aldrich), 200 ␮M myo-inositol (Sigma-Aldrich), and 100–500 U/ml recombinant human IL-2 (Roche). 9. OPTI-MEM reduced serum medium (Gibco/Invitrogen). 10. Recombinant human IL-2, available from a variety of commercial sources, such as Roche (Basel, Switzerland) that was generously provided by the National Cancer Institute Biological Resources Branch (Frederick, MD). 11. Freezing medium: 94% heat-inactivated FBS supplemented with 6% DMSO. 12. Cryogenic freezing vials (Nalgene). 13. NALGENETM cryo freezing container (Nalgene). 14. LipofectamineTM reagent and PlusTM reagent (Invitrogen). 15. Polybrene (1,5-dimethyl-1,5-diazaundecamethylene polymethobromide and hexadimethrine bromide; Sigma): Polybrene is dissolved at a stock concentration 5 mg/ml in PBS, subsequently filtered through a 0.2 ␮m filter, and stored for several weeks at +4◦ C or long term at −80◦ C. The working concentration of polybrene is 10 ␮g/ml. 16. Plasticware: T25 and T75 culture flasks (Nunc, Denmark), 15 ml conical centrifuge tubes (BD Falcon, USA), and sterile disposable 5 and 10 ml pipettes (FisherBrand). 17. Neomycin (Fisher Scientific) was dissolved in HEPES buffer, pH 7.2, at a concentration of 50 mg/ml and used to select transduced cells at a final concentration of 1.25 mg/ml. 18. Puromycin (EMD/Calbiochem, San Diego, CA) dissolved in DMSO at 5 mg/ml stock concentration and was used to select transduced cells at a concentration of 2.5 ␮g/ml.

Retroviral Transduction of Human NK Cell Lines

203

3. Methods 3.1. Cell Culture 3.1.1. Culture of Phoenix-Amphotropic Cells (see Note 5) 3.1.2. Culture of NK-Like Cell Lines

1. Phoenix cells are cultured in complete RPMI medium in 25 mm flasks (set horizontally) maintained at 37◦ C in humidified 7% CO2 atmosphere. 2. When cells reach about 80% confluence, they should be split 1:10 into 1:20 every 3–4 days with fresh culture medium. Several different NK cell lines, including KHYG-1, NK-92, NK3.3, and NKL, can be cultured under the following conditions. Cultures should be replaced with freshly thawed stocks every 4–6 weeks to maintain biological uniformity that can drift upon long-term culture. 1. NK cell lines are cultured in 50 ml of fresh complete ␣MEM medium in T75 flasks (standing vertically) at 37◦ C in a humidified 7% CO2 atmosphere. 2. Cells are passaged between 1:5 and 1:10 into fresh medium with human IL-2 every 4 days. Optimal growth is achieved by seeding new cultures with 4 million cells per 50 ml.

3.1.3. Freezing Cell Lines

To assure the optimal viability of cell lines, they should be harvested from log-phase cultures prior to freezing. NK cell lines grow in suspension, whereas Phoenix cells are adherent and easily detach from the tissue culture flask by gentle shaking. 1. Collect cells in a centrifuge tube and centrifuge the cells at 500 × g for 5 min. 2. Remove the medium, resuspend in fresh medium, and count the cells. 3. Centrifuge again at 500 × g for 5 min. 4. Remove supernatant and resuspend in freezing medium at 2–3 × 106 cells per ml. 5. Transfer 1 ml to a 2 ml cryogenic freezing vial, put the vial in a NALGENETM cryo freezing container at room temperature, and transfer to −70◦ C overnight. 6. Transfer vials to liquid nitrogen on the following day for long-term storage.

3.1.4. Thawing Cells

1. Remove vial from liquid nitrogen and thaw rapidly in a 37◦ C water bath. 2. Immediately after complete thaw, add 1 ml of warm culture medium (37◦ C) to the freezing vial and transfer this solution to 15 ml sterile conical screw cap tube containing 13 ml of warm culture medium.

204

Shahjahan Miah and Campbell

3. Centrifuge tube at 500 × g for 5 min. 4. Remove the supernatant and resuspend the cell pellet by flicking the tube. Add warm culture medium and transfer to a culture flask. Expand the culture at 37◦ C in humidified 7% CO2 atmosphere. 3.2. Retroviral Transduction of Human NK-Like Cell Lines (see Note 2) 3.2.1. Transfection of Phoenix-Amphotropic Cells

The first step of retroviral transduction is to clone your gene of interest into the pBMN-IRES-EGFP vector (see Note 4) and use this engineered vector to prepare recombinant retrovirus by transfecting into the Phoenix-amphotropic packaging cell line. The Phoenix-amphotropic cell line should be maintained at less than 80% confluence, and cultures should be replaced with freshly thawed stocks every 6–8 weeks (see Note 3). Day 1: 1. Plate 0.1–0.2 × 106 Phoenix-amphotropic cells per 6 ml complete RPMI medium per well in a 6-well plate 24 h prior to transfection. Phoenix cells should be about 70–80% confluent on the day of transfection (see Note 5). Day 2: 2. In a 1.5 ml microfuge tube, mix at least 4 ␮g pBMN-IRESEGFP vector containing cDNA of interest with 10 ␮l Plus Reagent and bring total volume to 100 ␮l with the addition of pre-warmed reduced serum OPTI-MEM. 3. In a separate microfuge tube, add 8 ␮l Lipofectamine to 92 ␮l pre-warmed reduced serum OPTI-MEM. 4. Incubate both samples at RT for 15 min. 5. After incubation, mix the contents of both tubes together for a total volume of 200 ␮l and incubate at RT for another 15 min. 6. Wash Phoenix-amphotropic cells once by aspirating the culture medium and gently adding 6 ml pre-warmed reduced serum OPTI-MEM along the side of the well. Add the medium slowly, because Phoenix cells do not adhere tightly and the added medium should not detach the cells. 7. After the DNA and transfection reagents have finished incubating, add 800 ␮l pre-warmed reduced serum OPTIMEM for a total volume of 1 ml. 8. Gently aspirate the wash medium from the culture well and gently add the 1 ml transfection reaction slowly to the Phoenix cells by releasing along the side of the well with a pipette. 9. Incubate the plate at 37◦ C in 7% CO2 atmosphere for at least 3 h.

Retroviral Transduction of Human NK Cell Lines

205

10. After 3 h of incubation add 4 ml pre-warmed complete RPMI medium to the transfected well and incubate plate at 37◦ C, 7% CO2 overnight. Day 3: 11. Remove medium by aspiration and wash cells once with 5 ml pre-warmed reduced serum OPTI-MEM. 12. Add 2 ml reduced serum OPTI-MEM and incubate at 37◦ C, 7% CO2 for 24 h (add 1.3 ml OPTI-MEM if using 2 wells for single transfection). 3.2.2. Collection of Virus and Transduction of NK Cell Line

This section describes the generation and manipulation of biohazardous retrovirus. Therefore, BSL2 biosafety procedures should be followed throughout the following steps. Upon finishing this section, incubate all disposed plasticware (including pipettes) under the UV light of a biohazard hood for at least 1 h to destroy the retroviral contamination. 1. On day 4, the retroviral supernatant is ready for harvesting. Collect supernatant (containing virus) into a 15 ml centrifuge tube. 2. Centrifuge the tube at 500 × g for 5 min to remove any remaining cells or filtering through a 45 ␮m filter (see Note 6). 3. Transfer virus into a new 15 ml centrifuge tube. 4. Add 20 ␮l Plus Reagent to virus suspension and incubate at RT for 15 min. 5. After incubation add 8 ␮l Lipofectamine to virus suspension and incubate at RT for another 15 min. 6. During the virus/Plus/Lipofectamine incubation, wash NK cells once with OPTI-MEM by spinning at 500 × g for 5 min. Count the cells and transfer 0.5 × 106 cells to a 15 ml tube. 7. Resuspend NK cells to be transduced with 2 mL viruscontaining supernatant and transfer to a single well of a 12-well plate. Centrifuge the plate at 700 × g for 30 min (see Note 7). 8. Incubate the plate at 37◦ C in a humidified 7% CO2 atmosphere for at least 3 h. 9. Centrifuge the plate again at 700 × g for 30 min, incubate again for 3 h at 37◦ C, and then proceed to step 10 or 11. 10. Transfer cells from each well of the 12-well plate to a separate T25 flask and add 8 mL of fresh IL-2-containing ␣-MEM culture medium to each flask. Incubate the culture at 37◦ C in a humidified 7% CO2 atmosphere until cells are confluent and proceed to step 12.

206

Shahjahan Miah and Campbell

11. Alternatively after step 9, to maximize transduction efficiency, add 2 mL of fresh IL-2-containing ␣-MEM culture medium to the virus-infected NK cells in well(s) of the 12well plate (total 4 ml of medium) and incubate overnight at 37◦ C in a humidified 7% CO2 atmosphere. Then transfer cells from the plate to the T25 culture flask (most of the cells will detach from plastic and can be easily transferred) and add another 6 ml of fresh complete ␣-MEM containing IL-2. Incubate the culture at 37◦ C in a humidified 7% CO2 atmosphere until cells are confluent and proceed to step 12. 12. When cells are ready (∼6 days after transduction, see Note 8) sort the transduced population for EGFP or for the expression of the transduced surface marker by flow cytometry (see Note 9). Alternatively, select transduced cells by treatment with antibiotics (if using a vector containing an antibiotic resistance gene; see Note 10). The fraction of transduced cells depends on the cell line used. The resulting cell population will retain expression for many weeks of culture. 13. Expand the transduced NK cell population and freeze several vials as described in Section 3.1.3. Discard growing transduced NK cell populations after 4–6 weeks of culture and replace with a newly thawed aliquot of frozen stock (see Note 11).

4. Notes 1. Importantly, constructs encoding potential oncogenes should be avoided when working with retroviral, adenoviral, or lentiviral expression systems. 2. This is a general protocol for transducing human NK cell lines but can be applied with little variation to other cell lines. 3. An important feature of the Phoenix cell lines is hightransfection efficiency using conventional transfection methods (e.g., including calcium phosphate or lipidbased techniques). In our hands, approximately 60–90% of Phoenix-amphotropic cells can be transiently transfected with Lipofectamine reagents, depending on the construct introduced. 4. The Stbl2 bacterial cells are suitable for the cloning of unstable inserts such as LTR-containing retroviral sequences or direct repeats, and for optimal performance, bacteria should be grown at 30◦ C.

Retroviral Transduction of Human NK Cell Lines

207

5. The transfection efficiency of Phoenix-amphotropic retroviral packaging cells depends on their health and growth status, which must be maintained by regular passage. If the cells are 100% confluent, transfection is very inefficient, so never let the cells reach confluence. 6. The retroviral supernatant can be used immediately for transduction of target cells or kept on ice if used within several hours. Otherwise, retroviral supernatant may be frozen at −80◦ C, resulting in a minimal loss of viral titer. 7. Our usual centrifugation period is 30 min, but increasing the centrifugation time up to 90 min can increase transduction efficiency in some cell lines. 8. Depending on growth rate, cells are generally sorted 6–8 days after transduction. Transduction efficiency can depend upon the number of cells infected, the construct used, and the NK cell line to be transduced. Starting with a higher number of cells usually requires relatively shorter times to be ready for sorting. 9. We have successfully used this transduction protocol to express cDNAs in the following human NK-like cell lines at the indicated efficiency of transduction: KHYG-1 (20– 50%), NK-92 (5–15%), NKL (15–30%), and NK3.3 (5– 15%). KHYG-1 cells are highly susceptible to retroviral transduction and can be successfully transduced by adding polybrene (10 ␮g/ml) instead of using Lipofectamine reagents at steps 4–6 in Section 3.2.2. To improve transduction efficiencies, more NK-92 or NK3.3 cells can be infected in steps 6–7 of Section 3.2.2. 10. Starting on day 2 after transduction with retrovirus containing an antibiotic resistance gene, the transduced cells should be selected with antibiotics for 5 days. 11. For testing any biological effect in transduced cells, the results should always be compared in cells from separate transduction procedures using the same construct. This will assure that the impact is not unique to the cells derived from a specific transduced population.

Acknowledgments We would like to thank all previous members of the Campbell Laboratory for establishing and optimizing this technique, Dr. Amanda Purdy for review of the chapter, and Dr. Garry Nolan for reagents. Supported by National Institutes of Health grants R01-CA083859, R01-CA100226 (K.S.C.), T32CA009035 (S.M.S.M.), and Centers of Research Excellence grant

208

Shahjahan Miah and Campbell

CA06927 (FCCC). The research was also supported in part by an appropriation from the Commonwealth of Pennsylvania. Its contents are solely the responsibility of the authors and do not necessarily represent the official views of the National Cancer Institute. References 1. Nagashima, S., R. Mailliard, Y. Kashii, T. E. Reichert, R. B. Herberman, P. Robbins, and Whiteside, T. L. (1998) Stable transduction of the interleukin-2 gene into human natural killer cell lines and their phenotypic and functional characterization in vitro and in vivo. Blood 91:3850–3861. 2. Cohen, G. B., Gandhi, R. T., Davis, D. M., Mandelboim, O., Chen, B. K., Strominger, J. L., and Baltimore, D. (1999) The selective downregulation of class I major histocompatibility complex proteins by HIV-1 protects HIV-infected cells from NK cells. Immunity 10:661–671. 3. Yusa, S., Catina, T. L., and Campbell, K. S. (2002) SHP-1- and phosphotyrosineindependent inhibitory signaling by a killer cell Ig-like receptor cytoplasmic domain in human NK cells. J Immunol 168: 5047–5057. 4. Kikuchi-Maki, A., Yusa, S., Catina, T. L., and Campbell, K. S. (2003) KIR2DL4 is an

IL-2-regulated NK cell receptor that exhibits limited expression in humans but triggers strong IFN-gamma production. J Immunol 171:3415–3425. 5. Yusa, S., Catina, T. L., and Campbell, K. S. (2004) KIR2DL5 can inhibit human NK cell activation via recruitment of Src homology region 2-containing protein tyrosine phosphatase-2 (SHP-2). J Immunol 172:7385–7392. 6. Alvarez-Arias, D. A., and K. S. Campbell. (2007) Protein kinase C regulates expression and function of inhibitory killer cell Iglike receptors in NK cells. J Immunol 179: 5281–5290. 7. Miah, S. M., Hughes, T. L., and Campbell, K. S. (2008) KIR2DL4 differentially signals downstream functions in human NK cells through distinct structural modules. J Immunol 180: 2922–2932.

Chapter 14 Lentiviral Gene Transduction in Human and Mouse NK Cell Lines Ram Savan, Tim Chan, and Howard A. Young Abstract Natural killer (NK) cells play a vital role in the control of cancer and microbial infections. A major hinderance in studying NK cells is the resistance of these cells to gene transfer. Considering over-expression and gene knockdown studies are crucial tools to study the biology of cells, technologies suitable for transfering genes into NK cells are invaluable. Among various technologies available for gene transfer, lentiviral-mediated transduction has been successful in introducing genes into NK cells. We have standardized methods of lentiviral infection in human and mouse NK cell lines. We obtain transduction efficiencies of 15% in the NK-92 cell line and 30–40% in LNK, YT, and DERL7 cell lines. This method allows efficient and stable introduction of genes and shRNAs into NK cell lines. Key words: Natural killer cells, lentivirus, transduction, flow cytometry, viral titration, human, mouse, NK-92, LNK.

1. Introduction Natural killer (NK) cells are large, granular lymphocytes which play a vital role in tumor immunosurveillance and combating microbial infections (1). These cells eliminate targets using multiple mechanisms and also recruit and amplify inflammatory responses. NK cells are able to lyse class I MHC-negative targets by releasing perforin, granzymes, and TNF ligands upon stimulation. These effector functions of NK cells are gradually acquired during their development and maturation. These cells are also known to recruit other immune cells to the sites of infection by releasing chemokines and cytokines. To understand the biology of NK cells and dissect specific gene functions, tools facilitating over-expression of genes or knockdown of genes by siRNA are invaluable. Resistance of NK cells to exogenous gene transfer is K.S. Campbell (ed.), Natural Killer Cell Protocols, Methods in Molecular Biology 612, DOI 10.1007/978-1-60761-362-6 14, © Humana Press, a part of Springer Science+Business Media, LLC 2010

209

210

Savan, Chan, and Young

a major hinderance in understanding NK cell biology and the potential use of these cells in immunotherapy. Human and mouse NK cells (primary cells and cell lines) are refractory to gene transfer. Methods like calcium phosphate precipitation (CaPO4 ), liposome reagents, and electroporation techniques have resulted in very low rates of gene transfer (2). Nucleofection technology (Amaxa, Inc.) has been the most promising among nonviral technologies that has shown some success in gene transfer in NK cell lines (3, 4). Several viral-mediated gene transfer protocols have been reported, but these approaches have had only variable success (5–7). For example, gene introduction with vaccinia virus altered the phenotype of NK cells (6) while adenoviral vectors have shown to be ineffective because NK cells do not possess appropriate receptors to mediate attachment of the virus to the cell. In contrast, Schroers et al. (8) showed that chimeric fiber-modified Ad5/F35 adenoviral vector efficiently infected primary human NK cells. Recently, ALAK (IL2-activated NK) cells transduced by adenovirus have been used in tumor therapy in mice (9, 10). Use of retroviral transduction in primary NK cells has resulted in partial success, primarily because of the requirement of multiple rounds of infection to introduce genes (11). Among the viral-mediated gene transfer systems, lentivirus transductions have been the most efficient gene delivery systems available for hematopoietic cells and hard to transfect cells. Recently, Tran and Kung (7) demonstrated an average of 40% transduction efficiency with primary murine NK cells. Furthermore, they show that lentiviral transduction does not affect the viability, function, and phenotype of NK cells. Lentiviral vectors (LV) are increasingly being utilized as a tool for introducing and obtaining stable expression of transgenes in both dividing and nondividing cells. Self-inactivating replicationincompetent lentiviral particles are generated by co-expressing the packaging elements of the virus along with the vector genome in the most commonly used virus-producing cell line, human embryonic kidney cell 293 or 293T (containing the SV40 large T antigen). The packaging elements of the HIV-1-based lentiviral vectors are from the HIV-1 genome. These HIV vector systems can be divided into three generations based on the progressive deletions of the packaging system from the parent vector. The first LV generation system contained all the HIV-1 genes except for the envelope in the parent vector. In the second generation, HIV genes including vpu and nef were deleted and the LV packing elements are provided as separate vectors. The third generation LV system offers maximum biosafety features by having the gag, pol, and rev genes provided as separate vectors. For this latest system, the envelope gene is from a heterologous virus, vesicular stomatitis virus (VSV), thus resulting in a pseudotyped virus with a broad host range.

Lentiviral Gene Transduction in Human and Mouse NK Cell Lines

211

The first sucessful report utilizing lentiviral vectors in NK cells described the transduction of primary murine NK cells (7). As the lentiviral transduction system allows efficient and stable integration of transgenes, NK cell lines can be utilized to develop in vitro models to study and further understand the biology of NK cells. Several human and mouse cell lines have been developed that can be used to study the function of NK cells. Among the human NK cell lines, NK-92 (12), YT (13), NKL (14), and DERL7 (15) are the most commonly used. NK-92 was originally derived from a male non-Hodgkin s lymphoma patient (12). This IL-2dependent NK cell line kills target cells and displays characteristics of activated NK cells. NK-92 produces copious amounts of IFN-␥ upon stimulation by cytokines individually or in combination (16), making it an ideal model for studying gene regulation in NK cells. DERL7 (CD56+ , CD3− , TCR␥␦− ) is a newly described cell line from a non-Hodgkin’s lymphoma patient, which posses both NK and T-cell surface markers (15). There is only one NK cell line derived from mice, designated LNK (CD132+ , CD16+ , CD3− , IgM− ), which is an IL-2-dependent NK line derived from liver lymphocytes of BALB/c nude mice (17). We have developed protocols which facilitate efficient and stable introduction of genes and short-hairpin RNA (shRNAs) into NK cell lines. The transduction efficiencies obtained are around 15% in NK-92 cells and 30–40% in LNK, YT, and DERL7 cells.

2. Materials 2.1. Plasmid Preparation

1. Plasmids: pLKO.1, pGIPZ, and pTRIPZ (Open Biosystems, www.openbiosystems.com); pMD2.G (Plasmid 12259; Addgene, www.addgene.org); and psPAX2 (Plasmid 12260; Addgene). 2. Bacterial strains: Vectors from Open Biosystems are generally transformed in Escherichia coli DH5␣. To prevent recombination of the plasmid with genomic DNA, use E. coli strains that lack the recombinase gene (recA). We recommend the use of commercially available electro- or chemicalcompetent E. coli strains like Stbl2 or Stbl4. 3. Luria–Bertani (LB) medium (1 L): Tryptone 10 g, yeast extract 5 g, and NaCl 10 g. Adjust volume to 1 L distilled water, sterilize by autoclaving, and store at room temperature (RT). 4. SOC medium (1 L); tryptone 20 g, yeast extract 5 g, 5 M NaCl 2 mL, 1 M KCl 2.5 mL, 1 M MgCl2 10 mL, 1 M

212

Savan, Chan, and Young

MgSO4 10 mL, and 1 M glucose 20 mL. Adjust volume to 1 L distilled water, sterilize by autoclaving, and store at RT. 5. Antibiotics: Prepare ampicillin stock solution at 50 mg/mL in distilled water and store at −20◦ C. 6. Plasmid purification kits: Purchase mini- and maxi-prep kits from any commercial source. Endo-free maxi-prep plasmid purification kits are recommended to obtain transfection quality DNA free of endotoxin. 2.2. Transfection Reagents

1. 2 M calcium chloride (CaCl2 ; Invitrogen, Walkersville, MD, USA). 2. 2× HEPES (2.5 M, Invitrogen).

2.3. Cell Lines and Culture Media

1. NK-92: This human NK cell line can be purchased from ATCC (American Type Culture Collection, Rockville, MD, USA; CRL-2047). NK-92 cells are cultured in RPMI 1640 medium (Cambrex, Walkersville, MD, USA) supplemented with 10% fetal bovine serum (FBS), 100 IU/mL penicillin, 50 ␮g/mL streptomycin, and 2 mM L-glutamine (PSG, Cambrex), recombinant human IL-2 (1000 IU/mL; PeproTech, Rocky Hill, NJ), and recombinant human IL-15 (100 ng/mL; PeproTech). 2. DERL7 (CD56+ , CD3− , TCR␥␦− ): This human cell line was kindly provided by Dr. L. Del Vecchio (A Cardarelli Hospital, Italy) and cultured in RPMI 1640 medium supplemented with 20% FBS, 100 IU/mL penicillin, 50 ␮g/mL streptomycin and 2 mM L-glutamine, recombinant human IL-2 (1000 U/mL), and stem cell factor (250 ng/mL; PeproTech). 3. LNK cell line: LNK is cultured in RPMI 1640 medium supplemented with 10% FBS, 100 IU/mL penicillin, 50 ␮g/mL streptomycin and 2 mM L-glutamine, 5% HEPES, 5% sodium pyruvate and 1× beta-mercaptoethanol, and recombinant human IL-2 (1000 U/mL). 4. 293FT cell line (Invitrogen): This human embryonic kidney line transfected with the SV40 large T antigen is cultured in DMEM supplemented with 10% FBS, 100 IU/mL penicillin, 50 ␮g/mL streptomycin and 2 mM L-glutamine, 5% HEPES, 5% sodium pyruvate, and 1% gentamicin. 5. 3T3 cell line: This mouse fibroblast cell line is cultured in DMEM supplemented with 10% FBS, 100 IU/mL penicillin, 50 ␮g/mL streptomycin and 2 mM L-glutamine, 5% HEPES, and 5% sodium pyruvate. 6. Trypsin−EDTA: 2.5% trypsin and 0.5 M ethylenediaminetetraacetic acid (EDTA).

Lentiviral Gene Transduction in Human and Mouse NK Cell Lines

213

7. PBS−EDTA: Phosphate-buffered saline (pH 7.4) and 0.1 M EDTA. 8. Antibiotics: (a) Puromycin: prepare stock solution at 10 mg/mL in PBS and store at −20◦ C. (b) Geneticin (G418): prepare stock solution at 50 mg/mL in PBS and store at −20◦ C. 9. Polybrene (Sigma): Prepare stock solution at 8 mg/mL in distilled water and filter sterilize. Aliquot (200 ␮L) and store at −20◦ C. 2.4. Virus Production and Purification

1. 10 cm collagen-coated tissue culture grade plates (we have routinely utilized BD Biocoat plates, but other brands may also be suitable) 2. Syringe-driven filter unit: Use low protein binding 0.45 ␮M pore size filters 3. 10 mL disposable syringe 4. 1.8 mL cryo-vials

3. Method 3.1. Preparation of Plasmids

1. Mix 100 ng of plasmid DNA with chemical-competent E. coli (DH5␣, Top10, Stbl2, or Stbl4) in a 1.5 mL polypropylene tube and incubate on ice for a minimum of 30 min. 2. Heat shock the bacteria−plasmid mixture at 42◦ C for 30 s and immediately incubate on ice for 2 min. 3. Add 500 ␮L of SOC medium to the bacteria−plasmid mixture and incubate in a shaker incubator (≈300 rpm) for 1 h. 4. Spread 100 ␮L of transformed bacteria onto LBA (LBagar) containing ampicillin (100 ␮g/mL) and incubate overnight at 37◦ C bacterial incubator. 5. Pick at least 10 colonies from each plate and inoculate in 5 mL LB medium containing ampicillin (100 ␮g/mL) and incubate in a shaker incubator for 6 h. 6. Aliquot 3 mL of the bacteria from above (step 5) and extract plasmid using any commercially available mini-prep kit and verify plasmid by restriction enzyme digestion. 7. Aliquot 1 mL from step 5 and grow in 300 mL of LB medium (supplement with 100 ␮g/mL ampicillin) for 12−16 h at 37◦ C. 8. Centrifuge the bacteria at 6000×g for 15 min.

214

Savan, Chan, and Young

9. Extract plasmid using any commercial endotoxin-free maxi-prep kit. 10. Verify by restriction digestion and store plasmid at 4◦ C at a final concentration of 1 mg/mL in TE buffer. 3.2. Virus Production (see Note 1)

Day 1 (seeding 293FT cells) 1. Seed 10 mL of 2.5 × 105 /mL of 293FT cells in complete DMEM medium on a collagen-coated 10 cm plate and incubate overnight at 37◦ C in a CO2 incubator (see Note 2). Day 2 (calcium phosphate transfection) 2. The cells should be 60–80% confluent the following day (100% confluency must be avoided). Gently remove medium and replace with pre-warmed 3T3 complete DMEM medium and return the culture dish to the incubator for a minimum of 4 h. 3. Mix lentiviral vector, gag/pol (psPAX2), and VSVG (pMD2.G) DNAs in a ratio of 2:2:1. We use 21 ␮g of the parent lentiviral vector, 21 ␮g of psPAX2, and 10 ␮g of VSVG (see Note 3) in a 14 mL polypropylene tube (Falcon). 4. Add 36 ␮L of 2 M CaCl2 solution and cell culture grade water to the plasmid mix to a final volume of 300 ␮L and vortex it for a second or gently flick the mixture. 5. It is necessary to next bubble air slowly into the mixture. We routinely do so with a Pasteur pipette. While the bubbling is ongoing, add 300 ␮L of 2× HBSS into the tube in a drop-wise manner. This process should be slow and lasts 1–2 min. 6. Incubate the above mixture for 30 min at RT (see Note 4). 7. Add the mixture onto the 293FT plate drop-wise and mix the contents on the plate by gently swirling the plate. Return the plate to the CO2 incubator at 37◦ C. 8. After 4–6 h, gently remove the medium from the plate and wash with 1× PBS. Add 10 mL of complete DMEM and incubate for 30–48 h at 37◦ C in a CO2 incubator. Day 3 (viral harvest) 9. Harvest the supernatant from the dish in a 15 mL capped polypropylene tube (Falcon). Centrifuge the tube at 360×g for 5 min to pellet cellular debris. 10. Pass the supernatant through 0.45 ␮M syringe-driven filter unit (see Note 5). 11. Aliquot the filtered supernatants into 1.8 mL cryo-vials and store them at −80◦ C.

Lentiviral Gene Transduction in Human and Mouse NK Cell Lines

3.3. Viral Titer 3.3.1. Biological Titering (see Note 6)

215

Day 1 1. The day before titering, plate 3T3 cells at a density of 2 × 105 cells/well of a 6-well plate and incubate in a CO2 incubator overnight (see Note 7). Cell confluency should be approximately 30–50% the following day. Day 2 2. Thaw the lentiviral sample (see Note 8) and prepare 10-fold serial dilutions in complete DMEM medium of the LV supernatant from 10−2 to 10−6 in sterile 5 mL polypropylene tubes (Falcon) with a minimal final volume of 2 mL for each dilution. Mix by gently pipetting the solution; DO NOT VORTEX. 3. Aspirate the medium from the 6-well overnight cultured plates and add 1 mL of the diluted virus-containing medium to their respective wells. Also include a control well. 4. Add polybrene to final concentration of 8 ␮g/mL to each well and incubate overnight. Day 3 5. Remove the infection medium from each well and replace with 2 mL of complete medium containing 10% FBS. Day 4 6. Remove the medium from the well and replace with complete medium containing 2 ␮g/mL puromycin to begin selecting transduced cells (see Note 9). Days 5–13 7. Every 3–4 days, remove the medium and replace with fresh complete medium containing 2 ␮g/mL puromycin Day 14 8. After 10 days on selection medium, the control well should have no cells left and the sample well should have some discrete puromycin-resistant colonies observed. 9. Remove the medium and wash the wells with PBS twice. 10. Add 1 mL Crystal Violet solution (1% w/v in 10% ethanol) and incubate at room temperature for 10 min. 11. Remove crystal violet solution and wash the wells with PBS twice. 12. Record the number of blue colonies observed in each of the wells with respect to the appropriate dilution. 13. Calculate the titer as transducing units (TU/mL) as follows: TU/mL = (number of discrete colonies/dilution factor)/volume of infection medium used.

216

Savan, Chan, and Young

3.3.2. qPCR-Based Titering (see Note 10)

14. Utilizing 150 ␮L of the viral supernatant, purify the viral RNA using the Nucleospin RNA Virus kit (Macherey-Nagel), according to the manufacturer’s protocol. Elute the RNA from the column using 50 ␮L RNasefree water. 15. Remove contaminating vector DNA from the sample by treating the sample with DNAse I (8 U/reaction) and incubating the tube at 37◦ C for 30 min followed by 5 min at 70◦ C. 16. Meanwhile, prepare 10-fold dilutions of the RNA Control template (Clontech, Mountain View, CA, USA) and your samples. You will use 2 ␮L for each reaction done in duplicates. 17. Prepare a Master Reaction mix containing 50 nM of the forward and reverse primer, 1× SYBR green, 1× QTaq DNA polymerase, 1× reaction buffer, and 1× qRT Mix (Clontech) with a final volume of 23 ␮L/sample. Add 2 ␮L of the diluted RNA control template and sample to the reaction. 18. Using a 7300 ABI Prism qPCR thermocycler, program the reaction as per manufacturer’s protocol and run the reaction/analysis. 19. Determine the copy number based upon the raw copy number obtained from the qRT-PCR standard and factoring the dilutions and amount of sample used to obtain the copy number/mL. Utilize an infectivity coefficient (as determined by a biological titering assay) to relate the copies/mL to obtain the transforming units/mL.

3.4. Transduction of Human NK Cell Lines (NK-92 and DERL7)

Day 1 1. Plate the NK cell line in appropriate complete culture medium at a density under 1.0 × 106 cells/mL (see Note 11). Day 2 2. Count the cells and resuspend in appropriate medium at 1.0 × 106 cells/mL. 3. Stimulate the lines with human IL-2 (100 U/mL) and human IL-12 (PeproTech; 100 ng/mL) for 2 h before transduction (see Note 12). 4. Following cytokine stimulation, readjust the concentration of the cells to 1.0 × 105 cells/mL using NK cell medium and plate 1 mL in each well of a 12-well plate. 5. Add viral supernatant for a multiplicity of infection between 20 and 100 of the titered virus to the medium-containing

Lentiviral Gene Transduction in Human and Mouse NK Cell Lines

217

cells and polybrene (8 ␮g/mL; see Note 13) and incubate overnight at 37◦ C. 6. Alternatively, spinoculation by centrifugation of cells and virus-containing supernatant at 360×g for 90 min at 32◦ C yields good transduction efficiency. Day 3 7. Remove the medium by spinning the cells at 360×g for 5 min at RT and resuspend the cells in 0.5 mL of NK cell medium and 0.5 mL of conditioned NK cell medium (see Note 14). Days 5–10 (selection of transgene-positive cells) 8. Cells can be selected by mammalian antibiotic resistance or YFP-positive cells depending on the marker used (see Fig. 14.1). We have titrated the required puromycin and hygromycin concentrations to be 2 ␮g/mL when using NK-92 or DERL7 cells (see Note 9).

Fig. 14.1. Flow cytometric analyses of yellow fluorescent protein (YFP)-positive cells in NK-92 and DERL7 cells postlentiviral transduction compared to controls.

9. Once selected for positive selection marker, choose up to five clones and check for knockdown (for shRNA) or transgene expression and select the clone with the desired expression level (see Note 15).

218

Savan, Chan, and Young

3.5. Transduction of Mouse NK Cell Line (LNK)

Day 1 1. Plate the LNK cell line in appropriate medium at a density under 1.0 × 105 cells/mL on a 6-well plate (see Note 16). Day 2 2. Stimulate the lines with high dose of IL-2 (1000 U/mL) for 2 h before transduction. 3. Add viral supernatant for a multiplicity of infection between 20 and 100 of the titered virus to the medium-containing cells and polybrene (8 ␮g/mL) and incubate overnight at 37◦ C. Day 3 4. Remove the medium and add 0.5 mL of fresh NK cell medium and 0.5 mL of conditioned NK cell medium (see Note 14). Days 5–10 (selection of transgene-positive cells) 5. Cells can be selected by mammalian antibiotic resistance or green fluorescent protein (GFP)-positive cells or other criteria depending on the marker used. We have titrated puromycin and hygromycin to be effective at killing non-transduced cells at a concentration of 2 ␮g/mL (see Note 9). 6. Once selected for the positive selection marker, select up to five clones and check for knockdown (for shRNA) or transgene expression and select the clone with desired expression level (see Note 15).

4. Notes 1. You must obtain approval from your institutional biosafety committee (IBC) before producing and/or utilizing lentivirus in the laboratory. All the vectors and the packaging systems should be listed in the IBC. Safety hoods, incubators, and centrifuges used for lentivirus work should be clearly indicated in the laboratory. 2. Refer to the manual from the commercial supplier for subculturing conditions for 293FT cells. Always use low passage number of 293FT cells for virus production. 293FT with high passage number will result in significantly lower viral titers. Do not allow the cell line to become overconfluent at any time. 3. The ratio and amount of plasmids are variable and should be optimized based on the titers obtained.

Lentiviral Gene Transduction in Human and Mouse NK Cell Lines

219

4. The mixture will form fine precipitates. However, if they form large precipitates, disperse the precipitates using a Pasteur pipette. 5. You could further concentrate the virus by ultracentrifugation to get a higher titer of virus (18). 6. The method for titering will depend upon the marker available within the LV system utilized. Biological titering will only be useful for LV systems with a selectable marker such as puromycin or blasticidin. If a reporter gene (e.g., green fluorescent protein or DsRed) is present, a flow cytometry-based method is needed. Titering, using qPCR-based methods, will work on all systems regardless of the selection or reporter gene present, but only provides the number of copies of genes present in the sample, and does not reflect the biological infectivity of the lentivirus. Therefore, if qPCR-based methods are utilized, an infectivity coefficient needs to be determined with the addition of a biological titering assay to obtain a ratio between the qPCR titer and the biological titer. 7. Various cell lines may be used for titering the lentivirus such as HeLa (human cervical cancer; ATCC CCL-2) or HT1080 (human fibrosarcoma; ATCC CCL-121). These cells are easy to work with, grow attached to the plate/flask, and rapidly divide. It is generally not recommended to use the packaging cell line (i.e., 293T cells) for titering, as carryover plasmid from the transfection may result in an overestimate of the titer obtained. 8. Lentiviruses should not be thawed more than three times. Viral titers are known to decrease with each thaw. 9. The actual concentration for puromycin selection will need to be determined for the different cell lines used as the sensitivity of alternate cell lines to puromycin may differ from that reported here. 10. This protocol shown below is based upon Clontech’s protocol for determining the lentivirus titer via RNA isolation. There are also other qPCR-based methods that can be done by isolating DNA and/or RNA at various steps through virus production and using gene-specific primers for the lentiviral vectors such as LV1, LV2, GAG, etc. (18, 19). 11. NK cells grow well at a density of 0.5–1.0 × 106 cells/mL. In our experience, we see that they do not transduce well if they are plated at a lower or higher density.

220

Savan, Chan, and Young

12. Stimulation of DERL7 cell line with human IL-2 (100 U/mL) and stem cell factor (250 ng/mL) results in better transduction efficiencies. 13. Retronectin can be used as an alternative to polybrene as human NK cells express VLA-4 and VLA-5, cell surface molecules which mediate adhesion to fibronectin (20). 14. Conditioned NK cell medium is filter-sterilized medium from log-phase culture of NK-92 or DERL-7 cell lines. This medium provides additional factors for the growth and survival of infected NK cells. 15. Expression or knockdown of the gene can be assayed by western blot or real-time PCR. 16. LNK cells are adherent cells and can be dislodged from the culture dish by washing once with PBS and then incubating the cells in PBS containing 1 mM EDTA for 5 min at 37◦ C in a CO2 incubator. The cells should then be washed with PBS to remove EDTA from the cells.

Acknowledgments This research was supported by the Intramural Research Program of the National Cancer Institute − Center for Cancer Research (NCI-CCR), National Institutes of Health. We thank Dr. Morihiro Watanabe for his help and advice in the development of lentivirus protocols for infection of NK cells. We thank Dr. Geraldine O’Connor for critical review of the chapter. The authors do not endorse any particular commercial products mentioned in this chapter nor should the specific product designations be interpreted as an endorsement of the products by the US Government.

References 1. Trinchieri, G. (1989) Biology of natural killer cells. Adv Immunol 47, 187–376. 2. Grund, E. M. and Muise-Helmericks, R. C. (2005) Cost efficient and effective gene transfer into the human natural killer cell line, NK92. J Immunol Methods 296, 31–36. 3. Maasho, K., Marusina, A., Reynolds, N. M., Coligan, J. E., and Borrego, F. (2004) Efficient gene transfer into the human natural killer cell line, NKL, using the Amaxa nucleofection system. J Immunol Methods 284, 133–140.

4. Schoenberg, K., Trompeter, H. I., and Uhrberg, M. (2008) Delivery of DNA into natural killer cells for immunotherapy. Methods Mol Biol 423, 165–172. 5. Jiang, W., Zhang, J., and Tian, Z. (2008) Functional characterization of interleukin15 gene transduction into the human natural killer cell line NKL. Cytotherapy 10, 265–274. 6. Kirwan, S., Merriam, D., Barsby, N., McKinnon, A., and Burshtyn, D. N. (2006) Vaccinia virus modulation of natural killer cell function by direct infection. Virology 347, 75–87.

Lentiviral Gene Transduction in Human and Mouse NK Cell Lines 7. Tran, J., and Kung, S. K. (2007) Lentiviral vectors mediate stable and efficient gene delivery into primary murine natural killer cells. Mol Ther 15, 1331–1339. 8. Schroers, R., Hildebrandt, Y., Hasenkamp, J., Glass, B., Lieber, A., Wulf, G., and Piesche, M. (2004) Gene transfer into human T lymphocytes and natural killer cells by Ad5/F35 chimeric adenoviral vectors. Exp Hematol 32, 536–546. 9. Goding, S., Yang, Q., Mi, Z., Robbins, P. D., and Basse, P. H. (2007) Targeting of products of genes to tumor sites using adoptively transferred A-NK and T-LAK cells. Cancer Gene Ther 14, 441–450. 10. Goding, S. R., Yang, Q., Knudsen, K. B., Potter, D. M., and Basse, P. H. (2007) Cytokine gene therapy using adenovirally transduced, tumor-seeking activated natural killer cells. Hum Gene Ther 18, 701–711. 11. Guven, H., Konstantinidis, K. V., Alici, E., Aints, A., Abedi-Valugerdi, M., Christensson, B., Ljunggren, H. G., and Dilber, M. S. (2005) Efficient gene transfer into primary human natural killer cells by retroviral transduction. Exp Hematol 33, 1320–1328. 12. Gong, J. H., Maki, G., and Klingemann, H. G. (1994) Characterization of a human cell line (NK-92) with phenotypical and functional characteristics of activated natural killer cells. Leukemia 8, 652–658. 13. Yodoi, J., Teshigawara, K., Nikaido, T., Fukui, K., Noma, T., Honjo, T., Takigawa, M., Sasaki, M., Minato, N., Tsudo, M., et al. (1985) TCGF (IL 2)-receptor inducing factor(s). I. Regulation of IL 2 receptor on a natural killer-like cell line (YT cells). J Immunol 134, 1623–1630. 14. Robertson, M. J., Cochran, K. J., Cameron, C., Le, J. M., Tantravahi, R., and Ritz, J. (1996) Characterization of a cell line,

15.

16.

17.

18. 19.

20.

221

NKL, derived from an aggressive human natural killer cell leukemia. Exp Hematol 24, 406–415. Di Noto, R., Pane, F., Camera, A., Luciano, L., Barone, M., Lo Pardo, C., Boccuni, P., Intrieri, M., Izzo, B., Villa, M. R., Macri, M., Rotoli, B., Sacchetti, L., Salvatore, F., and Del Vecchio, L. (2001) Characterization of two novel cell lines, DERL-2 (CD56+/CD3+/Tcry5+) and DERL-7 (CD56+/CD3-/ TCRgammadelta-), derived from a single patient with CD56+ nonHodgkin s lymphoma. Leukemia 15, 1641–1649. Hodge, D. L., Schill, W. B., Wang, J. M., Blanca, I., Reynolds, D. A., Ortaldo, J. R., and Young, H. A. (2002) IL-2 and IL-12 alter NK cell responsiveness to IFN-gammainducible protein 10 by down-regulating CXCR3 expression. J Immunol 168, 6090–6098. Tsutsui, H., Nakanishi, K., Matsui, K., Higashino, K., Okamura, H., Miyazawa, Y., and Kaneda, K. (1996) IFN-gammainducing factor up-regulates Fas ligandmediated cytotoxic activity of murine natural killer cell clones. J Immunol 157, 3967–3973. Salmon, P., and Trono, D. (2006) Production and titration of lentiviral vectors. Curr Prot Neurosci Chapter 4, Unit 4 21. Sastry, L., Johnson, T., Hobson, M. J., Smucker, B., and Cornetta, K. (2002) Titering lentiviral vectors: comparison of DNA, RNA and marker expression methods. Gene Ther 9, 1155–1162. Gismondi, A., Morrone, S., Humphries, M. J., Piccoli, M., Frati, L., and Santoni, A. (1991) Human natural killer cells express VLA-4 and VLA-5, which mediate their adhesion to fibronectin. J Immunol 146, 384–392.

Chapter 15 Introduction of shRNAs into Human NK-Like Cell Lines with Retrovirus Amanda K. Purdy and Kerry S. Campbell Abstract Natural killer (NK) cell lines are difficult to transfect using standard techniques, which limits the ability to establish long-term knockdown of proteins with short-hairpin (sh)RNAs. We have developed a method to stably knockdown protein expression in human NK-like lines by introducing shRNAs in retroviral vectors. After a single transduction with retrovirus, shRNA-containing cells can be selected with drug treatment or sorted for enhanced green fluorescent protein (EGFP) expression. With this method, protein expression can be stably decreased to less than 10% of wild-type levels. Key words: Retroviral transduction, shRNAs, NK-like cell lines.

1. Introduction Primary NK cells are difficult to transfect with standard vectors under a variety of conditions proven successful in other cell types (1). These cells are also not generally amenable to viral or retroviral infection (2), although recent successes with lentivirus (see protocols by Kung and Savan/Young in this volume) are overcoming this technical hurdle. In contrast, NK-like cell lines are more permissive to gene transfer by transfection (3, 4) and especially by retroviral transduction (2), thereby allowing for overexpression and knockdown of genes of interest. The technique of RNA interference (RNAi) has revolutionized modern cell biology by enabling researchers to selectively eliminate the expression of specific mRNAs. To achieve shortterm knockdown of the target mRNA expression (and subsequent K.S. Campbell (ed.), Natural Killer Cell Protocols, Methods in Molecular Biology 612, DOI 10.1007/978-1-60761-362-6 15, © Humana Press, a part of Springer Science+Business Media, LLC 2010

223

224

Purdy and Campbell

protein expression), double-stranded short-interfering (si)RNAs of 21–23 nucleotides (nt) matching specific sequences in the target mRNA are introduced into cells by electroporation or lipofection. Alternatively, short-hairpin (sh)RNAs can be expressed in cells (see below) and processed by the Dicer endonuclease complex to generate sustained expression of double-stranded 21–23-nt siRNAs, achieving long-term, stable knockdown of target mRNAs (5). The siRNAs hybridize with target mRNAs to tag them for degradation by the RNAi-induced silencing complex (RISC), which contains an endoribonuclease (6). The stable expression of shRNAs in cell lines can be achieved by transfection with certain vectors or transduction with retroviral or lentiviral vectors. Constructs cloned into these vectors generally consist of a 60-nt oligo that, when expressed in cells, is processed to generate a 19-nt siRNA with uridine overhangs (Fig. 15.1). Expression of the shRNA construct is driven by the polymerase III HI promoter, which produces a small RNA transcript, lacking a poly-A tail, and can be processed into a standard siRNA molecule.

Fig. 15.1. Schematic of a 60-nt siRNA-generating oligo. The ds oligo, flanked by restriction sites for cloning at the 5’ and 3’ ends, is composed of a 19 nucleotide sense sequence of the siRNA, a 9 nucleotide hairpin spacer sequence followed by the complementary 19 nucleotide anti-sense sequence of the siRNA. This figure was reproduced with permission from Oligoengine.

Here, we describe a method for knocking down proteins of interest in NK-like cell lines by the co-expression of two different retroviral vector-based shRNAs. This protocol can be modified to include co-expression of up to three distinct shRNAs simultaneously in the same NK cell line.

2. Materials 2.1. Generating shRNA-Containing Retroviral Vectors

1. Buffered saline: 100 mM NaCl and 50 mM HEPES, pH 7.4, in water 2. shRNA oligos: design an oligo targeting a specific sequence in the mRNA of interest. We have successfully designed

Stable shRNA-Dependent Knockdown in Human NK Cell Lines

225

shRNAs that knocked down the expression of several proteins using the online software provided by Oligoengine (www.oligoengine.com; Seattle, WA). The software scans an input mRNA sequence to predict the optimal target siRNAs. Upon choosing a particular target sequence, the software designs a double-stranded oligo consisting of a sense and an anti-sense strand with an intervening hairpin sequence. Each oligo also contains appropriate restriction site overhangs (Bgl II and Hind III) for cloning. These oligos can be purchased directly from Oligoengine. 3. pSuperior.retro.puromycin, pSuperior.retro.neomycin, and pSuperior.retro.neomycin.gfp vectors (Oligoengine) (see Note 1) 4. Agarose (Invitrogen, Carlsbad, CA) 5. SV Gel and PCR Clean Up System (Promega, Madison, WI) 6. Restriction enzymes: Bgl II, Hind III, EcoRI, and associated buffers (Invitrogen) 7. T4 ligase and 5× ligase buffer (Invitrogen) 8. Stbl2-competent bacterial cells (Invitrogen) 9. Liquid LB supplemented with 0.4% glucose and 50 ␮g/ml ampicillin 10. Wizard Plus SV Miniprep kit (Promega) 11. High-purity plasmid purification system (Marlingen, Ijamsville, MD) 2.2. Generating Retrovirus and Transduction of NK-Like Cell Lines

1. Phoenix-amphotropic retroviral packaging cell line (a gift from Dr. Garry Nolan, Stanford University, Stanford, CA) 2. Complete RPMI medium: RPMI-1640 medium (Mediatech, Herndon, VA) containing 10% heat-inactivated fetal bovine serum (FBS; HyClone, Logan, UT), 2 mM Lglutamate, 100 IU/ml penicillin (Mediatech), 100 ␮g/ml streptomycin (Mediatech), 50 mM HEPES, and 50 ␮M 2mercaptoethanol (ME) 3. Complete ␣-MEM: ␣-minimum essential medium (MEM; Life Technologies, Rockville, MD) containing 10% heat-inactivated FBS, 10% heat-inactivated horse serum (Invitrogen), 2 mM L-glutamate, 100 IU/ml penicillin, 100 ␮g/ml streptomycin, 1 mM sodium pyruvate (SigmaAldrich, St. Louis, MO), 200 ␮M myoinositol (SigmaAldrich), 2.5 ␮M folic acid (Sigma-Aldrich), 1× nonessential amino acids (Mediatech), and 100 ␮M 2-ME 4. OPTI-MEM reduced serum medium (Invitrogen) 5. Recombinant IL-2, available from a variety of commercial sources, such as Roche (Basel, Switzerland) was

226

Purdy and Campbell

generously provided by the NCI Biological Resources Branch (Frederick, MD). 6. 6- and 12-well culture plates (Fisher Scientific) 7. Plus Reagent and lipofectamine (Invitrogen) 8. 15 ml conical tubes (BD Falcon, USA) 9. Puromycin (Calbiochem, San Diego, CA) dissolved in DMSO at 5 mg/ml stock concentration 10. Neomycin dissolved in HEPES buffer, pH 7.2, at a stock concentration of 50 mg/ml (Fisher Scientific)

3. Methods 3.1. Designing shRNAs

There are numerous free programs available for designing shRNAs (Oligoengine, Dharmacon, etc.). For this protocol, Oligoengine software was used, since the interface was user-friendly, GenBank sequences could be uploaded directly, and both the secondary structure and the nucleotide usage were considered in the selection algorithm. We suggest designing a minimum of four shRNAs for each gene of interest, since we encountered an approximate 50% success rate. For knocking down SHP-2 phosphatase in the human NK-like cell line, KHYG-1, the expression of a single shRNA resulted in ∼50% decrease in wild-type protein levels, while co-expression of two shRNAs decreased levels by >90% (7). Avoid designing shRNAs that target common domains within gene families (e.g., phosphatase domain), as these will be less specific for the gene of interest and could nonspecifically suppress other members of the gene family. The final sequence should be tested for homology with other mRNAs with the BLAST program (http://blast.ncbi.nlm.nih.gov/Blast.cgi), and sequences with high homology to other known sequences should be abandoned. In addition, do not limit shRNAs to just one region of the mRNA. Finally, generate scrambled versions of each shRNA (having the same nucleotides as a shRNA targeting the gene of interest, in a randomly scrambled order) for use as controls. Scrambled control sequences should also not crossreact with known mRNAs in the targeted species, as assessed by a BLAST search. Alternatively, predesigned/pretested shRNAs of many genes cloned into retroviral vectors can be purchased from a commercial provider (e.g., Santa Cruz Biotechnology Inc., Santa Cruz, CA; Sigma-Aldrich, St. Louis, MO; Invitrogen).

3.2. Generating shRNA-Containing Vectors

The procedure for generating shRNA constructs was adapted from the Oligoengine pSuperior protocol. The protocol can be adapted for other specific vector and restriction enzyme combinations.

Stable shRNA-Dependent Knockdown in Human NK Cell Lines

227

1. To generate double-stranded shRNAs, combine 1 ␮l of each single-stranded oligo (both sense and anti-sense strands with compatible restriction site overhangs at 3 mg/ml) in 48 ␮l of buffered saline. Heat the oligo mix to 95◦ C for 10 min, and then slowly cool to RT (see Note 2). Once RT is reached, shRNAs can be immediately used or stored at 4◦ C for future use. 2. Digest pSuperior.retro vectors with BglII and Hind III restriction enzymes for 1 h at 37◦ C. 3. Separate the digested vector on a 1% agarose gel and purify with Promega Gel Extraction kit according to the manufacturer’s instructions. 4. Ligate shRNAs (1 ␮l of 1:100 dilution) into the purified linearized vector at 14◦ C overnight (see Notes 3 and 4). 5. Transform Stbl2 recombination-deficient competent cells (see Note 5) with 8 ␮l of the ligation reaction. 6. Plate the bacteria on LB culture plates containing 50 ␮g/ml ampicillin and grow at 30◦ C overnight. 7. Pick at least six colonies and culture each in 5 ml liquid LB with 0.4% glucose and 50 ␮g/ml ampicillin at 30◦ C overnight (see Note 6). 8. Isolate bacterial DNA with the Promega Miniprep kit, digest with Eco RI and Hind III for 1 h at 37◦ C, and separate digests on 1% agarose gels. Positive colonies are identified by the presence of a 300 bp band upon digestion (see Note 7). 9. Confirm the orientation of the shRNA insert by sequencing. 3.3. Transfection of Phoenix-Ampho Cells

It is important for both the Phoenix-amphotropic and the NKlike cell lines to be freshly passed a day prior to transfection and transduction, respectively. This protocol works well for KHYG-1, NKL, NK3.3, and NK-92 cell lines (7). Day 0: 1. One day prior to transfection, plate 5 ml Phoenixamphotropic cells into a 6-well culture plate (9 ml of a confluent Phoenix culture +21 ml of complete RPMI medium) and culture overnight in a 37◦ C incubator with 7% CO2 atmosphere (see Notes 8 and 9). Day 1: 2. For each transfection reaction, combine 4 ␮g of each shRNA-containing plasmid, 10 ␮l Plus Reagent, and OPTI-MEM to a final volume of 100 ␮l and incubate for 15 min at RT.

228

Purdy and Campbell

3. In a separate tube, combine 8 ␮l lipofectamine with 92 ␮l OPTI-MEM for each reaction and incubate for 15 min at RT. 4. After 15 min, combine the plasmid and lipofectamine solutions and incubate for 15 min more. 5. During the incubation, wash Phoenix-amphotropic cells once with 6 ml OPTI-MEM (see Note 10). 6. Add 800 ␮l of OPTI-MEM to the plasmid/lipofectamine solution (total volume is now ∼1 ml). Remove the wash solution from Phoenix-amphotropic cells and cover the cells with the plasmid/lipofectamine solution. 7. Transfect cells for at least 3 h at 37◦ C and then cover with 4 ml complete RPMI medium and incubate overnight. Day 2: 8. On the next day (afternoon), remove the medium, wash once with 6 ml OPTI-MEM, and cover cells with 2 ml OPTI-MEM. Day 3: 9. Collect the Phoenix cell supernatant containing the retrovirus and transfer into a 15 ml conical tube (see Note 11). Remove any cells by centrifugation at 500×g for 3 min. Transfer the viral supernatant to a new tube containing 20 ␮l of Plus Reagent. 10. Incubate the viral supernatant for 15 min at RT. 11. After 15 min, add 8 ␮l of lipofectamine and incubate 15 min longer (see Note 12). 12. During incubation, prepare NK cells for transduction by washing once with OPTI-MEM. 13. Transfer 5 × 105 NK cells to a 15 ml conical tube and pellet cells by centrifugation 500×g to remove the medium. 14. Resuspend the NK cells in the viral supernatant (about 2 ml), transfer to a 12-well cell culture plate, and spin at 700×g for 30 min at RT. 15. Incubate cells for at least 6 h at 37◦ C. 16. After incubation, cover cells with ∼2 ml complete ␣MEM with 50 U/ml recombinant IL-2 and incubate cells overnight at 37◦ C (see Note 13). Day 4: 17. Remove supernatant and continue to culture in fresh complete ␣-MEM with 50 U/ml recombinant IL-2 (see Note 14). 3.4. shRNA Selection

Selection scheme will depend upon which vectors (puromycin, neomycin, or gfp) were used.

Stable shRNA-Dependent Knockdown in Human NK Cell Lines

229

Day 6: 1. After resting cells for 2 days following transduction, add fresh complete ␣-MEM with 50 U/ml recombinant IL-2 and/or 2.5 ␮g/ml puromycin, 1.25 mg/ml neomycin (see Notes 15–17). Day 7: 2. Remove the old medium and add fresh complete ␣MEM with 50 U/ml recombinant IL-2 and/or 2.5 ␮g/ml puromycin, 1.25 mg/ml neomycin (see Notes 18). Day 9: 3. Remove the old medium and add fresh, drug-free complete ␣-MEM with 50 U/ml recombinant IL-2. 4. Check knockdown at the protein level by Western blotting or at the mRNA level by RT-PCR. For SHP-2 and SHP-1, protein knockdown was observed after 3 days of drug treatment (earliest time point analyzed), but knockdown did not become stable and consistent until 7 days after drug treatment was concluded. 5. Compare cellular protein or mRNA levels in cells transduced to express single shRNAs versus cells expressing the scrambled shRNAs or empty pSuperior vector.

4. Notes 1. Depending upon your selection scheme (selection with antibiotics versus GFP expression), it is important to choose the appropriate retroviral vector(s). With this system, one can express and select for up to three separate shRNAs at the same time (one shRNA in puromycin, neomycin, and gfp vectors). We have had success using each shRNA-containing vector singly or all together. 2. It is easiest to use a heat block for this step, turning it off after the 95◦ C incubation and allowing the block to cool to RT. 3. It is recommended to have a 1:3 ratio of vector to insert for the ligation reaction. Some PCR and gel extraction kit elution buffers disrupt accurate determinations of the DNA concentration by absorbance spectroscopy. To avoid this issue, measure the concentration of the insert and digested vector on an agarose gel using a DNA ladder for the concentration control. 4. To decrease the number of false-positive colonies, digest the ligation reaction with Bgl II for 1 h at 37◦ C. The Bgl II

230

Purdy and Campbell

site is destroyed upon successful cloning of the shRNA pair; therefore, vectors containing the shRNA insert will not be cut. 5. To prevent recombination at retroviral LTRs, always use recombination-defective bacteria (e.g., Stbl2) when working with retroviral vectors, and always culture bacteria at 30◦ C. 6. Grow colonies for longer times to compensate for the reduced growth temperature. 7. False positives that stem from re-ligation of empty vector will have a ∼1 kb band in the Oligoengine system. 8. Phoenix cells should be ∼80% confluent for optimal transfection. Never allow the cells to reach full confluence or transduction efficiency will suffer. 9. Several varieties of the Phoenix packaging cells are available (e.g., amphotropic, ecotropic, polytropic), which differ primarily in the expression of the viral envelope proteins that mediate viral entry into target cells (8). Use Phoenixamphotropic cells when generating virus for transduction of human cells, Phoenix-ecotropic for murine cells. 10. Phoenix cells are semi-adherent and can easily detach from the plate. Exercise caution when manipulating the cells or when changing the medium. Add medium very slowly to the side of the 6-well plate, keeping the pipette tip horizontal to the plane of the cells. Never add medium directly on top of the cells. 11. Although the retrovirus is replication defective and should be noninfectious, be sure to UV treat all contaminated glass and plasticware for at least 1 h before discarding. The use of retroviral technology requires standard class BSL2 biohazard safety precautions and approval by the local biohazard safety committee in your institution. 12. Retrovirus can be used immediately or can be stored at −80◦ C for up to 4 months with an estimated potency loss of ∼50%. We have had success using virus that was frozen to co-transduce KHYG-1 cells with two vectors simultaneously when subsequently selected in medium containing the appropriate combination of antibiotics. 13. To increase viability, let cells detach from the 12-well plate overnight instead of forcing cell detachment with pipetting. 14. Transduction efficiency varies widely among known NK-like cells lines. For example, NK-92 cells had ∼4% transduction efficiency with this protocol, while KHYG-1 had almost 30% efficiency for single vector transductions.

Stable shRNA-Dependent Knockdown in Human NK Cell Lines

231

15. To improve drug selection, expand cells to multiple wells if overgrown. 16. Determine the optimal concentration of drug needed to kill each NK-like cell line being used. For transduction of KHYG-1, NK-92, and NKL cells, puromycin was used at 2.5 ␮g/ml and G418 at 1.25 mg/ml. 17. To ensure proper drug selection, include a well of nontransduced cells during selection. 18. Puromycin-induced death was observed by days 1–2 of drug treatment and days 3–5 for neomycin.

Acknowledgments We would like to thank Drs. Lauren O’Donnell and S. M. Shahjahan Miah for helpful comments on the chapter. Supported by National Institutes of Health grants R01-CA083859, R01CA100226 (K.S.C.), T32-CA009035 (A.K.P.), and Centers of Research Excellence grant CA06927 (FCCC). The research was also supported in part by an appropriation from the Commonwealth of Pennsylvania. Its contents are solely the responsibility of the authors and do not necessarily represent the official views of the National Cancer Institute. References 1. Goldfarb, R. H., Whiteside, T. L., Basse, P. H., Lin, W. C., Vujanovic, N., and Herberman, R. B. (1994) Natural killer cells and gene therapy: potential of gene transfection for optimizing effector cell functions and for targeting gene products into tumor metastases. Nat Immunol 13, 131–140. 2. Nagashima, S., Mailliard, R., Kashii, Y., Reichert, T. E., Herberman, R. B., Robbins, P., and Whiteside, T. L. (1998) Stable transduction of the interleukin-2 gene into human natural killer cell lines and their phenotypic and functional characterization in vitro and in vivo. Blood 91, 3850–3861. 3. Liu, J. H., Wei, S., Blanchard, D. K., and Djeu, J. Y. (1994) Restoration of lytic function in a human natural killer cell line by gene transfection. Cell Immunol 156, 24–35. 4. Tran, A. C., Zhang, D., Byrn, R., and Roberts, M. R. (1995) Chimeric

5. 6. 7.

8.

zeta-receptors direct human natural killer (NK) effector function to permit killing of NK-resistant tumor cells and HIVinfected T lymphocytes. J Immunol 155, 1000–1009. Tijsterman, M., Ketting, R. F., and Plasterk, R. H. (2002) The genetics of RNA silencing. Annu Rev Genet 36, 489–519. Hannon, G. J. (2002) RNA interference. Nature 418, 244–251. Purdy, A. K. and Campbell, K.S. (2009) SHP-2 expression negatively regulates NK cell function. J Immunol in press. Danos, O. and Mulligan, R. C. (1988) Safe and efficient generation of recombinant retroviruses with amphotropic and ecotropic host ranges. Proc Natl Acad Sci U S A 85, 6460–6464.

Chapter 16 Introduction of shRNAs into Primary NK Cells with Lentivirus Sam K.P. Kung Abstract Natural killer (NK) cells are lymphocytes that provide an important line of defense against viruses and tumors. Technical hurdles in genetic modifications of primary NK cell ex vivo had limited our studies of protein function(s) in NK cell differentiation, acquisition of self-tolerance, and induction of antitumor responses. We used VSV-G-pseudotyped, EGFP-expressing lentiviral vectors to develop an efficient gene transfer system to modify gene expression in primary murine NK cells with or without prior IL-2 activation. Lentiviral vector transduction did not impair NK cellular viability, phenotype, or functions. We also demonstrated the use of this system in modifying differentiating NK cells derived from lentiviral-transduced murine hematopoietic progenitor cells. Furthermore, the same transduction protocol is amendable to delivery of short-hairpin RNA (shRNA) for specific gene silencing. Collectively, our approach in genetic engineering of primary murine NK cells will prove useful in studying basic NK cell biology and in exploring therapeutic potentials of NK cells in inbred and transgenic mouse models. Key words: Lentiviral vectors, natural killer cells, LAK, mouse, transduction, gene therapy.

1. Introduction The availability of a simple and efficient methodology that supports long-term and stable transduction of genetic materials into primary NK cells and NK progenitors will empower us the ability to acquire greater mechanistic understanding of NK cell differentiation and receptor function(s) both in vitro and in vivo. From a therapeutic point of view, the ability to manipulate NK cells directly to express novel receptors or genetically engineered NK receptors will allow us to enhance NK target recognition and redefine NK target specificities.

K.S. Campbell (ed.), Natural Killer Cell Protocols, Methods in Molecular Biology 612, DOI 10.1007/978-1-60761-362-6 16, © Humana Press, a part of Springer Science+Business Media, LLC 2010

233

234

Kung

A major challenge in genetically engineering primary NK cells is to introduce stable genetic modification of NK cells efficiently with minimal in vitro manipulation, and without any impairment in cellular viability, phenotype, or functions. Our laboratory used replication-incompetent HIV-1 lentiviral vectors to develop a complementary strategy to the existing methods to deliver either a transgene (1) or a short-hairpin RNA (shRNA) into primary NK cells. Lentiviral vectors transduce and integrate into host genomes without the need of cell division (2–4). They will therefore support persistent expression of introduced genetic materials in modified NK cells (progenitors). In addition, as they can transduce both nondividing and dividing cells in vitro and in vivo (4–6), they may also be used to circumvent the need of an in vitro activation step in NK cell transduction and/or to improve transduction efficiency of primary human and mouse NK cells. Here, we describe a single-step lentiviral transduction of primary murine NK cells that supported efficient transduction and stable expression of the transferred genetic materials without any apparent compromise in the cellular viability, phenotype, or functions (1).

2. Materials 2.1. DNA Plasmids

1. Second-generation lentiviral vector packaging system (Fig. 16.1): Three-plasmid components are needed to produce replication-incompetent lentiviral particles. It includes a gene transfer vector of choice to transduce NK cells to either over-express a transgene [e.g., FUGW or pRRLSIN.cPPT.PGK-GFP, Addgene, Cambridge, MA; SIN18RhMLV-Cppt2E (7, available from our laboratory upon request)] or to down-regulate endogenous gene expression via RNA interference (e.g., pLKO or FG12) (Addgene), a packaging plasmid (pCMV-dR8.2dvpr) (Addgene), and an envelope plasmid (pCMV-VSVG) (Addgene) for pseudotyping. 2. Stbl3 (Invitrogen, Carlsbad, CA)-competent cells for transformation. 3. Plasmids are isolated using commercial plasmid purification kits such as Qiagen Plasmid Midi Kit (Qiagen, Valencia, CA) and Nucleobond Plasmid Midi Kit (Clontech, Mountain View, CA).

2.2. Viral Vector Preparation, Concentration, and Transduction

1. Iscove’s Modified Dulbecco’s Medium (IMDM) (HyClone, Logan, UT) supplemented with 10% fetal bovine serum (FBS) (HyClone) and 1% penicillin/streptomycin/ glutamate (PSG) (Invitrogen).

Introduction of shRNAs into Primary NK Cells with Lentivirus

235

Fig. 16.1. Lentiviral vector production. A Second-generation three-plasmid transfection system for replicationincompetent lentiviral vector productions. Three plasmid components are needed to produce replication-incompetent lentiviral particles in 293T cells: (1) gene transfer vector of choice, (2) a packaging plasmid, and (3) an envelope plasmid for pseudotyping (see Section 2.1). Viral particles are collected in the culture supernatant and can be further concentrated by ultracentrifugation. B Examples of lentiviral vectors for NK cell transduction. Lentiviral vectors that use internal promoters RhMLV (rhesus-adapted Murine Leukemia Virus LTR) (SIN18-RhMLV-cppt2E), PGK (phosphoglycerate kinase) (pRRL.SIN.cPPT.PGK.-GFP.WPRE), or UbiC (Ubiquitin-C) (FUGW) are used in transgene over-expression in NK cells. In shRNA-expressing lentviral vectors (FG12 and pLKO.1), U6 promoter is used to direct RNA Polymerase III transcription of the shRNA. Reporter genes (EGFP or puromycin-resistant gene) are expressed under a separate transcription cassette driven by an internal UbiC or PGK promoter. Note that the use of FG12 requires cloning of an expression cassette that contains U6 RNA pol III promoter and the shRNA target sequence of interest. Suggested restriction sites for cloning are indicated. C Schematic representation of a 5 −3 oligonucleotide that contains a shRNA target sequence for cloning (see Section 3.5). Synthesize the complementary 3 −5 oligonucleotide for annealing. Selection of G as the initiating nucleotide facilitates U6 promoter transcription. The shRNA is followed by a polyT termination sequence for RNA Polymerase III.

2. Phosphate-buffered saline (PBS), pH 7.4, autoclaved. 3. Chloroquine (Sigma, St. Louis, MO) is reconstituted in autoclaved PBS at 10 mM stock concentration. The solution is sterilized by 0.22 ␮m filter cup, stored in aliquots at −20◦ C. 4. Hexadimethrine bromide (Polybrene) (Sigma) is reconstituted in autoclaved PBS at 1 mg/ml stock concentration, sterilized using 0.22 ␮m filter cup, and stored in aliquots at −20◦ C. 5. 2 M CaCl2 (Tissue culture grade, Sigma) solution is prepared by dissolving CaCl2 salt in autoclaved distilled water. ◦ It is sterilized by 0.22 ␮m filter and stored at 4 C.

236

Kung

6. 2× HEPES-buffered saline (HBS) (100 ml): 50 mM HEPES (1.19 g) 280 mM NaCl (1.63 g) 1.5 mM Na2 HPO4 Pre-prepared 0.25 M Na2 HPO4 stock (0.6 ml) NaOH (5 M) and (1 M) for pH adjustment to 7.12 All chemicals are purchased from Sigma. Bring volume up to 100 ml. Filter solution with 0.22 ␮m filter. Store at ◦ 4 C. 7. Beckman ultracentrifuge with a SW41 or SW28 rotor. Polyallomer centrifuge tubes (Beckman, Palo Alto, CA). 8. Puromycin (Calbiochem, San Diego, CA, USA) is dissolved in PBS at 2 mg/ml stock solution. 9. 0.05% trypsin−EDTA (Invitrogen) in autoclaved PBS. 10. Sterilized 50 ml polypropylene conical tubes, T175 flask (BD Falcon, Franklin Lakes, NJ) 11. 293T cells (GenHunter, Nashville, TN, USA). 12. HIV-1 p24 ELISA (PerkinElmer Life Sciences, Inc., Waltham, Massachusetts, USA). 2.3. Natural Killer Cells Purification, Activation, and Expansion In Vitro

1. Roswell Park Memorial Institute 1640 (RPMI 1640) culture medium: RPMI 1640 (HyClone) supplemented with 10% FBS, 1% PSG, and 1.6 mM 2-mercaptoethanol (Sigma). 2. EasySep Mouse NK Negative Selection Kit (StemCell Technologies, Vancouver, British Columbia, Canada). 3. IL-2 (Peprotech, Rocky Hill, NJ, USA) is dissolved in autoclaved distilled water at 0.1 mg/ml, further diluted with RPMI 1640 culture medium to obtain a stock solution of 100 U/␮l. The cytokine is stored in aliquots at −70◦ C until use. A final concentration of 1000 U/ml is used in NK cultures.

2.4. Isolation of Progenitor Stem Cells (PSCs) and NK Differentiation In Vitro

1. Progenitor stem cells are enriched using an EasySep Mouse Progenitor Stem Cell Negative Selection kit (StemCell Technologies). 2. Cytokines IL-2, IL-7, and IL-15, stromal cell factor (SCF), and Flt3L (Peprotech) are reconstituted according to manufacturer’s instruction. 3. NK differentiation-conditioned medium: RPMI 1640 supplemented with 10% FBS, 1% PSG, 1.6 mM 2-ME, 0.5 ng/ml of mIL-7, 30 ng/ml of SCF, and 50 ng/ml Flt3L.

Introduction of shRNAs into Primary NK Cells with Lentivirus

237

4. RPMI 1640 culture medium containing 30 ng/ml of IL-15. 5. 24-, 48-well tissue culture plate (BD Falcon).

3. Methods 3.1. VSV-GPseudotyped Lentiviral Vector Production 3.1.1. Preparing 293T Cells for Transfection on Day 0 (see Note 1)

Adherent 293T cells are collected by trypsin treatment and gentle tapping of the flask (Fig. 16.1). 1. Aspirate and discard the IMDM culture medium from the T175 flask. 2. Add 15 ml of PBS to rinse the adherent cells. 3. Aspirate and discard PBS. 4. Add 3 ml trypsin−EDTA to the flask to cover all the adherent 293T cells. Leave the flask at room temperature for 5 min. 5. Gently tap the bottom of the flask to dislodge the cells into suspension. 6. Add 10 ml of IMDM culture medium to the flask. 7. Pipette up and down a few times to prepare single-cell suspension. 8. Centrifuge cells at 400 × g for 5 min to collect cells. 9. Resuspend cells in 10 ml IMDM culture medium. Count cells. 10. Culture 2 × 107 293T cells in 25 ml IMDM culture medium in a T-175 flask on the day before transfection (see Note 2).

3.1.2. Three Plasmids Transfection of 293T Cells on Day 1

1. Aspirate to remove the IMDM culture medium from the 293T cell culture flask. Feed them with 25 ml fresh warm medium. 2. Prepare the DNA mix in a 50 ml conical tissue culture centrifuge tube. A total of 30 ␮g of plasmid DNA should be used for each T175 flask (VSVG 5 ␮g, packaging 12.5 ␮g, and vector 12.5 ␮g). Add autoclaved ddH2 O to a final volume of 977 ␮l (see Notes 3 and 4). 3. Add 133 ␮l of 2 M CaCl2 into the DNA mix, vortex well, and leave on ice for 5 min. 4. Add 1110 ␮l of 2× HBS solution drop by drop, while simultaneously mixing the solution by vortex. Mix again and rest it on ice for 20 min. 5. Flip the T175 flask upside down to drain the culture medium to the other side of the flask that does not have

238

Kung

cells (to avoid detaching 293T cells during the mixing of DNA precipitates). Add the mixed DNA precipitates into the culture medium drop by drop and mix gently (see Note 5). 6. Add 100 ␮l of chloroquine (10 mM) to the culture. 7. Incubate the culture for 6–8 h at 37◦ C, 5% CO2 . 8. After 6–8 h incubation, change medium with 35 ml of fresh medium to remove the transfection mix. Do not touch cells for 36 h (see Note 6). 9. On Day 3, harvest the culture supernatant as virus and keep at 4◦ C. Feed the culture again with 25 ml of fresh IMDM culture medium. 10. On day 4, collect supernatant again. 3.1.3. Concentrating VSV-Pseudotyped Lentiviral Vectors (to 100 ×)

1. Centrifuge viral supernatant at 500 × g for 5 min, 4◦ C, to remove cell debris. 2. Filter the supernatant through 0.45 ␮m filter cups. 3. Transfer 35 ml of viral supernatant into a polyallomer ultracentrifuge tube for SW28 rotor (see Note 7). 4. Centrifuge at 50,000×g (16,000 rpm) at 4◦ C for 90 min in a Beckman ultracentrifuge to collect viral particles at the bottom of the ultracentrifuge tube. 5. Remove the supernatant by decantation and aspirate residual supernatant with pipetman. Add IMDM culture medium to a total volume of 350 ␮l. 6. Use parafilm to seal the ultracentrifuge tubes and leave them at 4◦ C overnight for the viral vector pellet to resuspend. 7. Use a P-1000 pipetman to pipette up and down the viral particles three times to resuspend the viral particles in the next morning. Store the viral particles in aliquots at −70◦ C. Avoid freeze-and-thaw. 8. Viral particles and wastes should be handled according to the biosafety regulation and guidelines from your institution (see Note 8).

3.2. Determination of Viral Titers

For the lentiviral vectors that express a puromycin-resistant gene or proteins (surface or cytosolic) detectable by flow cytometry. 1. Plate 5 × 104 cells/well of 293T cells in a 24-well tissue culture dish a day before titration. With single doubling time, it will become 105 cells on the day of transduction. 2. On day of titration dilute virus (keep virus on ice as much as possible): 100× → 10× (10 ␮l virus in 90 ␮l IMDM medium)

Introduction of shRNAs into Primary NK Cells with Lentivirus

239

10×→ 1× (40 ␮l 10× in 360 ␮l IMDM medium) 1×→ 1/10× (30 ␮l 1× in 270 ␮l IMDM medium) 3. Add polybrene to the diluted viruses (typically 1×, 1/10×) at a final concentration of 8 ␮g/ml. Mix well. 4. Remove tissue culture medium from the wells. Add 250 ␮l of the diluted viruses to each well of 293T cells. Include 1–2 mock transduction control wells (tissue culture medium plus polybrene). 5. Incubate at 37◦ C in a tissue culture incubator for 2 h. 6. Remove virus-containing medium. Add 1 ml fresh IMDM medium and continue to culture for 3 days. 7. For the vectors that express a puromycin-resistant gene, add puromycin-containing tissue culture medium at 8 ␮g/ml on day 2 for an additional 2 days. Leave a mocktransduced control well without puromycin treatment to determine total cell number. Determine the number (and percentage) of cells that are resistant to puromycin for each viral dilution tested. 8. For vectors that express surface or intracellular proteins detectable by flow cytometry. Harvest cells and determine the percentage of the transduced cells that express the gene of interest: For surface proteins, use specific mAb in standard surface staining procedures for flow cytometry. For intracellular proteins, use specific mAb in intracellular staining procedures for flow cytometry. For fluorescence proteins, such as EGFP, the transduced cells can be analyzed in flow cytometry directly (see Note 9). 9. Use the data from the dilution that provided 5–20% transduced cells. Calculate the titer based on this equation: Titer (IU/ml) = (1 × 105 cells) × (% of the transduced cells) × (4) × (dilution factor). 10. For vectors that cannot be titered directly in the above assays, a HIV p24 ELISA is recommended. The p24 concentration is not a direct measurement of the infectious particles but will be useful in normalizing the same amount of viral particles (single or multiple) to be used in the transduction. 3.3. Primary NK Cell Transduction

1. Isolate primary NK cell from splenocytes that are free of T and NKT cell (see Note 10).

3.3.1. Adding Viral Vector Particles to NK Cells

2. Purified NK cells can be used directly for transduction or cultured as LAK cells in IL-2 (1000 U/ml) before transduction. 3. Place 2.0 × 105 cells into a 1.5 ml screw-cap centrifuge tube.

240

Kung

4. Pellet cells at 1700 × g for 10 min in a microcentrifuge. 5. Remove supernatant. 6. Resuspend cells in 0.25 ml viral mixture [virus + 8 ␮g/ml polybrene + complete RPMI medium (10% FBS, 1% PSG, 1.6 mM 2-mercaptoethanol)] at MOI of 20 (see Note 11). 7. Transduction can be carried out by using either the “spin” or the “no-spin” protocol as described below (see Notes 12–15). 3.3.2. No-Spin Protocol

1. Loosely tighten the cap on centrifuge tube (from step 6). 2. Incubate cells at 37◦ C and 5% CO2 for 2 h. 3. Pellet cells at 1700 × g for 10 min. 4. Remove viral supernatant. 5. Resuspend in 1 ml 1,000 U/ml IL-2.

complete

medium

containing

6. Transfer to 48-well culture plate. 7. Incubate cells at 37◦ C and 5% CO2 for 3 days before analysis. 3.3.3. Spin Protocol

1. Transfer the resuspended cells into a 48-well culture plate. (see Note 16). 2. Centrifuge plate at 900 × g for 1 h at room temperature. 3. Remove viral supernatant. 4. Resuspend in 1 ml of complete RPMI medium containing 1,000 u/ml IL-2. 5. Incubate cells at 37◦ C and 5% CO2 for 3 days before analysis. Examples of primary NK LAK transduction are shown in Fig. 16.2. Lentiviral vector transduction of primary NK cells is efficient, stable, and applicable to different mouse strains (e.g., B6, SJL/J) (see Note 17).

3.4. Transduction of Bone Marrow Progenitor Stem Cells (PSCs) for NK Differentiation In Vitro

1. Obtain single-cell suspension of bone marrow cells from C57BL/6 mice. 2. Enrich for progenitor stem cells using an EasySep Mouse Progenitor Stem Cell Negative Selection kit (StemCell Technologies). 3. 0.3 × 106 PSCs are placed into each well of a 48-well culture plate (BD Falcon). Transduction is conducted at a MOI of 20 immediately after enrichment of progenitor stem cells using the spin protocol. 4. After transduction, remove virus-containing supernatant. Culture the transduced progenitor stem cells in NK differentiation-conditioned medium (RPMI 1640 supplemented with 10% FBS, 1% PSG, 1.6 mM 2-ME, 0.5 ng/ml of mIL-7, 30 ng/ml of stromal cell factor (SCF), and 50 ng/ml Flt3L.

Introduction of shRNAs into Primary NK Cells with Lentivirus

241

Fig. 16.2. Stable and efficient transduction of primary NK LAKs. A Day 3 IL-2-activated B6 LAK or Day 7 IL-2activated SJL/J LAK cells were transduced with SIN18-RhMLV-cppt2E lentiviral vectors at MOI of 20 with the spin protocol. The transduced cells were analyzed for EGFP expression under fluorescence microscopy on days 3, 6, and 10 post-transduction. Mock-transduced NK LAK cells were negative controls for EGFP expression. B A representative flow cytometric analysis of the percentage of EGFP-expressing B6 NK cells on day 6 post-transduction. Percentage of EGFP-expressing NK cells remained relatively unchanged despite their active proliferation during this 10 day posttransduction period. C Down-regulation of SHP-1 phosphatase protein expression in B6 LAK. Day 5 IL-2-activated LAK cells were transduced with EGFP shRNA control or SHP-1 shRNA vectors at MOI = 40 IU/cell on two consecutive days using the “spin” protocol as described. Transduced cells were incubated in IL-2-supplemented medium for 3 days posttransduction. Cells were selected for puromycin resistance (at 24 mg/ml) for 24 h. The selected cells were stained intracellularly with anti-SHP-1/anti-rabbit-AlexaFluor antibodies and then analyzed by flow cytometry 3 days post-puromycin selection.

5. On day 3 post-transduction, add 0.5 ml of fresh NK differentiation-conditioned medium. 6. On day 5, pellet cells to remove old medium from the culture. Add 0.5 ml of complete RPMI medium containing 30 ng/ml of IL-15 to culture the cells. 7. On day 8, add 0.5 ml of fresh IL-15-containing medium. 8. On day 10, replace old medium with fresh RPMI medium containing 30 ng/ml of IL-15 and 1000 U/ml of IL-2. 9. On day 14, analyze the differentiating NK cells in flow cytometry. These NK cells can be further differentiated in the presence of a stromal cell line. 3.5. A Variation of the Common Theme − Introduction of shRNA

RNA interference (RNAi) is an innate cellular process that involves multiple RNA−protein interactions (8, 9). Its gene silencing activity is activated when a double-stranded RNA

242

Kung

molecule of greater than 19 duplex nucleotides enters the cells, causing degradation of both the dsRNA and the single-stranded RNA (endogenous mRNA) of identical sequences (8, 9). We and others have previously established a single lentiviral vector system to stably express simultaneously both a reporter gene (EGFP or puromycin-resistant gene) and a short-hairpin RNA (shRNA) to induce RNA interference (RNAi) in cell lines and primary T cells (10–12). Using the lentiviral transduction protocol we described in Section 3.3, we are successful in introducing shRNA into primary NK cells for specific gene silencing. We use the U6 promoter to direct RNA Polymerase III transcription of the shRNA. The sequence encoded in the dsRNA contributes to the potency of the dsRNA in degradation of target mRNA (13). 1. Determine the “optimal” 21-mer target sequence in your gene (see Note 18). 2. Synthesize forward and reverse shRNA oligonucleotides that contain the overhanging restriction enzyme sites for cloning, 21 “sense” bases identical to the target gene, a loop, and 21 “antisense” bases that are complementary to the “sense” bases. The shRNA is followed by a polyT termination sequence for RNA Polymerase III (see Note 19, Fig. 16.1C). 3. Anneal the oligonucleotides and clone into the restriction enzyme sites of lentiviral vectors containing a U6 pol III promoter (e.g., pLKO) (see Note 20). 4. Produce VSV-G-pseudotyped lentiviral particles as described in Section 3.1. 5. Isolate primary NK cells and culture them in IL-2supplemented RPMI medium (1000 U/ml) for 3–5 days. 6. Transduce NK LAK cells with the shRNA-containing lentiviral particles at a MOI of 20, using the “spin” protocol as described in Section 3.3. Use a lentiviral vector that contains an irrelevant target gene sequence (e.g., luciferase) or scrambled RNA sequence for specificity control. Culture the transduced cells in IL-2-supplemented RPMI medium. 7. Repeat the transduction on the next day to increase the transduction efficiency if necessary. 8. For lentiviral vectors that express EGFP reporter protein (e.g., FG12), the transduced LAK cells were cultured in IL2-supplemented medium (1000 U/ml) for 2–3 days before carrying out the analyses or cell sorting in flow cytometry. 9. For lentiviral vectors that express puromycin resistance gene, culture the transduced LAK in IL-2-supplemented medium (1000 U/ml) for 3 days before the addition of selection medium (i.e., puromycin) at a concentration of 24 ␮g/ml for an additional 24 h. Analyze the puromycin-resistant

Introduction of shRNAs into Primary NK Cells with Lentivirus

243

LAK cells for phenotype and functions or expand these selected cells further in IL-2-supplemented medium (see Note 20). An example of the shRNA-mediated silencing of SHP-1 phosphatase in primary NK LAK is shown in Fig. 16.2C.

4. Notes 1. The condition of the cells is very important for optimal transfection, which in turn determines the viral titers obtained in the preparation. They should be growing exponentially and evenly distributed. We do not use 293T cells that have been in passage for more than 3 months even though they look “normal.” 2. We aim at obtaining 70% confluency of 293T cells on the day of transfection (and determined that seeding 18–20 × 106 293T cells a day before the transfection would yield 70% confluency). As culture condition (such as the FCS) used in different laboratories may vary, the number of 293T cells used to seed a T175 the day before transfection may need to be adjusted empirically. 3. 30 ␮g of total DNA mix is used in transfecting a T175 flask of 293T cells. In making a few flasks of the same virus, we simply multiply the amount of each DNA plasmid by the number of flasks (up to 3 flasks) to be mixed in a single 50 ml conical centrifuge tube. Multiple 50 ml centrifuge tubes are used if we will prepare more than 3 flasks of the same virus. 4. New lentiviral vectors for more specific or tighter control in transgene or shRNA expression have been developed in different laboratories. Packaging systems (such as the third-generation packaging system which involves four plasmids transfection for maximal biosafety) or commercial lentiviral packaging mix are also available. It is important to check the compatibility of the packaging system and the gene transfer lentiviral vector that you are interested in. 5. 293T cells are highly susceptible to transfection. We used calcium phosphate and 2× HBS transfection protocol because they are cheap and easy to prepare in the laboratory. It can be replaced by other commercially available transfection reagents.

244

Kung

6. On days 2 and 3 post-transfection, VSVG-induced syncytium formation may be observed in the transfected 293T cultures. This is normal. 7. We do not sterilize polyallomer centrifuge tubes and parafilm before use. We open a new box inside a biosafety cabinet and handle it under standard aseptic tissue culture practice. We do not observe any contamination. 8. We transplanted autologous G-CSF/SCF-mobilized CD34+ cells that were transduced with lentiviral vectors expressing EGFP into myeloablated rhesus macaques. The safety of these replication-incompetent vectors was assessed in vivo for more than 4 years post-transplant. The animals remain healthy with no evidence of circulating HIV-1 viruses nor hematopoietic abnormalities/malignancies (14). 9. Viral titers: Using EGFP-expressing lentiviral vectors in flow cytometric analysis, we routinely obtained 105 –106 Infectious Units/ml viral titer (before ultracentrifugation). 10. NK cells can be isolated by either negative selection or positive selection kits (StemCell, AutoMACS). If a DX5positive selection kit is used, a complement depletion of Thy1.2+ cells is needed to remove contaminating NKT cells. 11. MOI (multiplicity of infection) is defined as a ratio of the number of infectious virus particles to target cells. We observed that MOIs of 20–40 are usually optimal for a single round of NK transduction; however, we suggested a titration of MOI be performed in each laboratory. 12. We found that primary NK cells can be transduced efficiently either by the “no-spin” or by the “spin” protocol. However, the “spin” protocol is more effective when a lower MOI is used (1). 13. Lentiviral vectors can transduce ex vivo-purified NK cells, as well as lymphokine-activated cells; however, we observed that cytokine activation of NK cells enhanced transgene expression (1). 14. Lentiviral vectors use internal promoters to drive transgene expression or shRNA expression. Their promoter activities in primary NK cells are therefore critical in determining the efficiency of transgene expression (1) or shRNA knockdown. We found that the CMV promoter is weak in mouse NK cells. We recommended PGK, Ubiquitin, or a rhesus-derived MLV promoter (7) for trans-

Introduction of shRNAs into Primary NK Cells with Lentivirus

245

gene expression in primary NK cells. U6 promoter is a promoter of choice in driving shRNA expression. 15. If needed, transduction can be repeated on the second day to improve transduction efficiency (1). 16. Transduction can also be done in 24-well culture plates. Up to 4.0 × 105 cells per well can be used in the transduction. 17. The genetically modified NK cells maintained stable EGFP transgene expression in vitro and can be further expanded in IL-2-supplemented culture medium. Lentiviral transduction does not affect NK surface phenotypes or functions (apoptosis, cytokine production, and cytotoxicity) (1). The transduction protocol is amendable to human NK cells. 18. Design of shRNA sequence. Selection of suitable 21-mer targets in your gene is the first step toward efficient gene silencing. 18.1. Guideline for designing siRNAs with effective gene silencing (adapted from www.rnaiweb.com/RNAi/siRNA Design): 1. Targeted regions on the cDNA sequence of a targeted gene should be located at least 25 nucleotide downstream of the start codon (ATG). 2. Search for sequence motif AA(N19)TT or NA(N21), or NAR(N17)YNN, where N is any nucleotide, R is purine (A, G), and Y is pyrimidine (C, U). 3. Avoid sequences with >50% G+C content. 4. Avoid stretches of 4 or more nucleotide repeats. 5. Avoid sequences that share a certain degree of homology with other related or unrelated genes. 18.2. Examples of commercial companies and web sites available to assist the design of siRNA target sequences. 1. http://www.ambion.com/techlib/misc/ siRNA finder 2. http://jura.wi.mit.edu/bioc/siRNAext/ 18.3. Examples of commercial companies that offer a collection of ready-to-test shRNA-containing lentiviral vectors.

246

Kung

1. Sigma (www.sigmaaldrich.com/life-science/ functional-genomics-and-rnai/shrna/trcshrna-products) 2. OpenBiosystems (www.openbiosystems.com) 19. We used 5 -TTCAAGAGA-3 or 5 -CTCGAG-3 loop sequence in the constructs (see Fig. 16.1C). 20. Under the optimal transduction protocol, the choice of shRNA sequences determines the potency of shutdown. We usually generate a few lentiviral constructs that express different shRNA sequences against the target gene and test them on NK cells to determine their knockdown potency (flow, western blots, qPCR). 21. We found that 24 ␮g/ml of puromycin is the minimal concentration of puromycin that results in complete NK cell death within 24 h of culture in our laboratory. However, depending on NK cultures and batch to batch preparations of puromycin stock, the concentration of puromycin used in the selection may vary. We suggest that an optimal concentration of puromycin be predetermined in each laboratory.

Acknowledgments This work has been supported by the Establishment and Operating grants from Manitoba Health Research Council, Manitoba Institute of Child Health, Canada Foundation for Innovation and Natural Sciences and Engineering Research Council (to S.K.P.K). Some of the constructs were obtained from the Biomedical Functionality Resource established under the support of Dean Strategic Research Fund at University of Manitoba. S.K.P.K. is a Basic Science Career Development Research Awardee of the Manitoba Medical Service Foundation supported with funds provided by the Manitoba Blue Cross. The author declares there is no conflict of interest. References 1. Tran, J. and Kung, S. K. (2007). Lentiviral vectors mediate stable and efficient gene delivery into primary murine natural killer cells. Mol. Ther. 15:1331–1339. 2. Case, S. S., Price, M. A., Jordan, C. T., Yu, X. J., Wang, L., Bauer, G., Haas, D. L., Xu, D., Stripecke, R., Naldini, L., Kohn, D. B., and Crooks, G. M. (1999). Stable transduction of quiescent CD34(+)CD38(-) human

hematopoietic cells by HIV-1-based lentiviral vectors. Proc. Natl. Acad. Sci. U S A 96:2988–2993. 3. Miyoshi, H., Smith, K. A., Mosier, D. E., Verma, I. M., and Torbett, B. E. (1999). Transduction of human CD34+ cells that mediate long-term engraftment of NOD/SCID mice by HIV vectors. Science 283:682–686.

Introduction of shRNAs into Primary NK Cells with Lentivirus 4. Sutton, R. E., Wu, H. T., Rigg, R., Bohnlein, E., and Brown, P. O. (1998). Human immunodeficiency virus type 1 vectors efficiently transduce human hematopoietic stem cells. J. Virol. 72:5781–5788. 5. Blomer, U., Naldini, L., Kafri, T., Trono, D., Verma, I. M., and Gage, F. H. (1997). Highly efficient and sustained gene transfer in adult neurons with a lentivirus vector. J. Virol. 71:6641–6649. 6. Kafri, T., Blomer, U., Peterson, D. A., Gage, F. H., and Verma, I. M. (1997). Sustained expression of genes delivered directly into liver and muscle by lentiviral vectors. Nat. Genet. 17:314–317. 7. Kung, S. K., An, D. S., and Chen, I. S. (2000). A murine leukemia virus (MuLV) long terminal repeat derived from rhesus macaques in the context of a lentivirus vector and MuLV gag sequence results in highlevel gene expression in human T lymphocytes. J. Virol. 74:3668–3681. 8. Hannon, G. J. (2002). RNA interference. Nature 418:244–251. 9. Fire, A. (1999). RNA-triggered gene silencing. Trends Genet. 15:358–363. 10. Qin, X. F., An, D. S., Chen, I. S., and Baltimore, D. (2003). Inhibiting HIV-1 infec-

11.

12.

13.

14.

247

tion in human T cells by lentiviral-mediated delivery of small interfering RNA against CCR5. Proc. Natl. Acad. Sci. U S A 100: 183–188. An, D. S., Xie, Y., Mao, S. H., Morizono, K., Kung, S. K., and Chen, I. S. (2003). Efficient lentiviral vectors for short hairpin RNA delivery into human cells. Hum. Gene Ther. 14:1207–1212. Stewart, S. A., Dykxhoorn, D. M., Palliser, D., Mizuno, H., Yu, E. Y., An, D. S., Sabatini, D. M., Chen, I. S., Hahn, W. C., Sharp, P. A., Weinberg, R. A., and Novina, C. D. (2003). Lentivirus-delivered stable gene silencing by RNAi in primary cells. RNA 9: 493–501. Reynolds, A., Leake, D., Boese, Q., Scaringe, S., Marshall, W. S., and Khvorova, A. (2004). Rational siRNA design for RNA interference. Nat. Biotechnol. 22(3):326–330. Kung, S. K., An, D. S., Bonifacino, A., Metzger, M. E., Ringpis, G. E., Mao, S. H., Chen, I. S., and Donahue, R. E. (2003). Induction of transgene-specific immunological tolerance in myeloablated nonhuman primates using lentivirally transduced CD34+ progenitor cells. Mol.Ther. 8: 981–991.

Chapter 17 Methods to Identify and Characterize Different NK Cell Receptors and Their Ligands Dikla Lankry, Roi Gazit, and Ofer Mandelboim Abstract Different cellular immune responses are modulated by the cross talk between activating and inhibitory signaling pathways initiated via different cell surface receptors. Similarly, the killing of NK cells is controled by multiple activating and inhibitory surface receptors. In humans, the major NK triggering receptors, identified so far, include NKp80, 2B4 NKG2D, and CD16 and the natural cytotoxic receptors (collectively named NCRs) include NKp46, NKp44, and NKp30. The two major families of MHC-specific inhibitory receptors identified in humans are the Ig superfamily (KIR and LIR) and the C-type lectin (CD94/NKG2A) receptor superfamily. The different inhibitory receptors show diverse specificity and discriminate between different class I MHC proteins. Much is known about the function and expression patterns of the different NK cell receptors, but the ligand identity of several of the activating NK cell receptors is yet to be discovered. This chapter introduces several research tools that can be used to uncover the identities of different ligands for NK cell receptors. Key words: NK cell clones, Ig-fused proteins, BW assay, cytotoxicity assays, point mutations.

1. Introduction Human natural killer (NK) cells are bone marrow-derived lymphocytes that comprise 5–15% of the peripheral blood lymphocytes (1). As part of the innate immune system, NK cells are designed to kill a broad spectrum of infectious agents and tumors without prior specific stimulation. Each NK cell is capable of recognizing and destroying multiple targets. This broad specificity is mediated by multiple activating and inhibitory surface receptors and intracellular signal transduction molecules (2). However, the ligands for many of the NK cell receptors are still unknown. To K.S. Campbell (ed.), Natural Killer Cell Protocols, Methods in Molecular Biology 612, DOI 10.1007/978-1-60761-362-6 17, © Humana Press, a part of Springer Science+Business Media, LLC 2010

249

250

Lankry et al.

study the NK cell receptors, we must first separate the NK cells from the other peripheral blood mononuclear cells (3). Once we obtained high purity of NK cells, we can identify the different receptors expressed by the different NK cell clones and further study the variations in the different killing potentials of these specific clones. Two different methods described in this section help us identify target cells that express potential NK cell receptor ligand(s): (1) The fusion Ig protein approach (4) allows us to easily scan many target cells for a potential ligand(s). (2) The BW assay (5, 6), which is a cell-to-cell interaction assay, utilizes the NK cell receptor expressed on the cell surface in its accurate conformation to detect presence of a ligand on target cells. In addition, to study the function of a specific NK cell receptor with regard to killing, we can use NK tumor lines such as the YTS cell line (6). Using this system we can study the function of the native NK cell receptor or generate mutations and study their effects on the activity of a particular receptor. In vitro methods are essential tools to study NK cell receptors, but an in vivo system is more accurate and informative. Therefore, we constructed a knockout mouse lacking the NCR1 receptor, and its entire NK cell repertoire is marked with GFP. Using this system, we can easily isolate the NK cells and study the importance of the NCR1 receptor for the immune response executed by NK cells (7)

2. Materials 2.1. General Reagents

1. RPMI-1640 medium (GIBCO, Invitrogen) 2. Dulbecco’s Modified Eagle Medium (DMEM; GIBCO, Invitrogen) 3. Supplements for complete media preparations: 1% sodium pyruvate solution (Biological Industries),1% penicillin/streptomycin solution (Biological Industries), 1% L-glutamine solution (Biological Industries), 1% nonessential amino acids solution (Biological Industries), and 10% fetal bovine serum (FBS; PERBIO) 4. 96 U-well/96 V-well/96 flat (F)-well/6-well/24-well culture plates (no particular manufacturer necessary) 5. PCR primers (SIGMA) 6. PBS (1× prepared from 10× stock; Biological Industries)

2.2. Culturing of Primary NK Clones

1. Heparin (ROTEXMEDICA) 2. Ficoll-Paque PLUS (GE Healthcare) 3. RPMI-8866 cell line (ATCC)

Methods to Identify and Characterize Different NK Cell Receptors and Their Ligands

251

4. Gamma irradiation source 5. PHA−L (phytohemagglutinin-L; Roche) 6. AutoMACS separator (Miltenyi Biotec) 7. Human NK cell isolation kit: biotin Ab cocktail and antibiotin microbeads (Miltenyi Biotec) 8. AutoMACS Medium (1 L): 4 mL EDTA (final concentration of 2 mM), pH 8 and 5 mL FBS in PBS 9. NK cell medium: 350 mL DMEM, 150 mL F-12 Ham‘s medium (GIBCO, Invitrogen), 50 mL human serum (SIGMA), 5 mL of sodium pyruvate solution, penicillin/streptomycin solution, L-glutamine solution and nonessential amino acids solution (Biological Industries), and 2 mL IL-2, final concentration of 500 U/mL (500,000 U/mL stock; Boehringer Mannheim GmbH) 10. Human anti-CD56/PE mAb (BioLegend) 11. Human anti-CD3/FITC mAb (BioLegend) 12. Negative control mouse IgG1/RPE (Dako) 13. Negative control mouse IgG1/FITC (Dako) 14. Anti-CD3 mAb (T3D Hybridoma) 2.3. Producing Fusion Ig Proteins

1. Restriction enzymes: Hind III, Bam HI (Fermentas) 2. TOP10/P3 bacteria (Invitrogen) 3. HiYield Plasmid mini kit (RBC Bioscience) 4. PureLink HiPure Plasmid maxi prep kit (Invitrogen) 5. Ampicillin 100 mg/mL solution (Roche) 6. Tetracycline 1 mg/mL solution (Boehringer Mannheim GmbH) 7. LB broth base (Invitrogen) 8. COS-7 cells (ATCC) 9. TransIT − LT1 Transfection reagent (MIRUS Bio.) 10. 500 mL Filter system − 0.22 ␮m CA (Corning Incorporated) 11. LPM medium, Low Protein Medium BSA-Free (Biological Industries) 12. 10% sodium azide stock solution 13. Protein A/G column − HiTrap Healthcare)

TM

Protein A/G HP (GE

14. Pharmacia P-1 peristaltic pump with adjustable flow rate up to 500 mL/h (Pharmacia Biotech) 15. 0.1 M glycine stock solution, pH 2.7 16. 1 M Tris−HCl, pH 8.8

252

Lankry et al.

17. 20% ethanol solution 18. SnakeSkin Pleated Dialysis Tubing (PIERCE) 2.4. BW Zeta Assay

1. BW cells (ATCC) 2. Electroporation instrument: BioRad Gene Pulser 3. Gene Pulser cuvettes − 0.4 cm (BioRad) 4. Neomycin 10 mg/mL solution: Geneticin G-418 sulfate (GIBCO) 5. MaxiSorp plate (Nunc) 6. Binding buffer: 0.1 M Na2 HPO4 , pH 9.0 (0.1 M NaH2 PO4 ) 7. PBS/Tween20 0.05% solution: 0.05% polyoxyethylene 20 sorbitan monolaurate (Tween20; J.T. Baker) in PBS 8. Blocking buffer: PBS containing 10% FBS or 3% BSA 9. Purified rat anti-mouse IL-2 antibody, 0.5 mg/mL (BioLegend) 10. Biotin rat anti-mouse IL-2, 0.5 mg/mL (BioLegend) 11. Streptavidin−HRP, 1 mg/mL (Jackson Immunoresearch) 12. Chromogen mation)

tetramethylbenzidine

(TMB;

Dakocyto-

13. Optic reader PowerWave ×S (BioTek) 2.5. Killing Assay

1.

S-methionine, 7.5 mCi/500 ␮L (Institute of Isotopes Co. Ltd) 35

2. P815 mastocytoma cell line (ATCC) 3. Modified RPMI-1640 (SIGMA)

medium,

lacking

methionine

4. Opaque Opti-plates (Perkin Elmer) 5. Scintillation liquid: MicroSCINT

TM

40 (Perkin Elmer)

6. Top seal A (Perkin Elmer) 7. Microplate scintillation and luminescence TM (␤ counter): Packard Top Count − NXT 2.6. Mutations and Transfection into YTS Cells

counter

1. YTS-ECO cells (ATCC) 2. BOSC cell line (ATCC) 3. 100% ethanol supplemented with 10% of 3 M sodium acetate solution, pH 5.3 4. Puromycin diHCl (Calbiochem) 5. Hexadimethrine bromide (polybrene; Aldrich)

Methods to Identify and Characterize Different NK Cell Receptors and Their Ligands

253

3. Methods 3.1. Culturing of Primary NK Cell Clones

In order to study the NK cell receptors and their interactions with different target cells, it is important to separate the NK cells from the peripheral blood mononuclear cells (PBMC). Working with NK populations of high purity will help to define the specific NK cell receptor interactions and activities.

3.1.1. Isolation of Human PBMCs from Peripheral Blood

Work in a sterile environment. 1. Use a 30 mL syringe washed with heparin to obtain 10 mL of fresh blood. Transfer the blood into 50 mL tube and dilute 1:2 with pre-warmed RPMI medium, 37◦ C (see Note 1). 2. Add 13 mL of Ficoll-Paque Plus into a new 50 mL tube and gently load 30 mL of the diluted blood on top of it (Ficoll:medium ratio approximately 1:2). Centrifuge for 30 min at 515×g at room temperature, brake off. PBMC will be localized between the Ficoll layer and the plasma layer as a white ring. 3. Take out most of the plasma layer, leave about 5 mL above the PBMC ring, and use 10 mL pipette to collect the PBMC ring and transfer into a new 50 mL tube. 4. Wash the PBMC with pre-warmed RPMI medium, 37◦ C, centrifuge for 5 min at 515×g at 4◦ C, and resuspend the cells with 10 mL pre-warmed RPMI medium, 37◦ C.

3.1.2. Preparation of Feeder Cells

Work with 96 U-well plates 1. Use Ficoll gradient in order to separate PBMC from two different donors (see Section 3.1.1 and Note 1). 2. Mix together 5×106 cells/plate from each donor (total of 10×106 ) and 5×105 cells/plate of the EBV-transformed Bcell lymphoma RPMI-8866 (ATCC). 3. Irradiate the cells with 6000 rad. 4. Wash the cells with pre-warmed RPMI medium, 37◦ C, and centrifuge for 5 min at 515×g at 4◦ C. 5. Discard supernatant, resuspend the pellet with NK medium, 100 ␮L/well (total of 10 mL/plate), and add 1 mg/mL PHA-L (final concentration of 0.1 ␮g/well). Seed 100 ␮L/well of the feeder cells in 96 U-well plates and incubate in a 37◦ C, 5% CO2 incubator until the NK separation process is completed (see Section 3.1.3). Seed about 8−12 96 U-well plates.

254

Lankry et al.

3.1.3. Separation of Human NK Cells from PBMCs by Using the AutoMACS Instrument.

Keep working on ice! These steps are performed according to the manufacturer’s instructions (Miltenyi). 1. Collect 1×106 PBMCs and wash with cold autoMACS medium. Centrifuge for 5 min at 515×g at 4◦ C (see Note 2). 2. Discard supernatant and gently resuspend the pellet with 40 ␮L of cold autoMACS medium. Add 10 ␮L biotin Ab cocktail, gently pipette, and incubate on ice for 10 min. 3. Add 30 ␮L of cold autoMACS medium and 20 ␮L of antibiotin microbeads, gently pipette, and incubate on ice for 15 min. 4. Wash the cells by adding 5 mL of cold autoMACS medium and centrifuge for 5 min at 515×g at 4◦ C. 5. Discard all supernatant by using a 200 ␮L micropipette and resuspend the pellet with 500 ␮L of cold autoMACS medium. 6. Separate the NK cells by using the autoMACS instrument based on a negative selection (according to the manufacturer’s instructions); select the “depletes” program and use one tube to collect the negative fraction (NK cells) and another to collect the positive fraction (mainly T cells). 7. Count the number of NK cells in the negative fraction. About 10% NK cells should be obtained. 8. Validate NK cell phenotype (CD56+ CD3− ) by FACS after staining the cells with anti-CD56/PE and anti-CD3/FITC. In a separate sample, use mouse IgG1/RPE and mouse IgG1/FITC as negative controls. Alternatively, identify the NK cells with anti-NKp46 mAb.

3.1.4. NK Cell Cloning and Culturing

NK cells are cloned and cultured in 8–12 96 U-well plates 1. Dilute the NK cells in NK cell medium according to the desirable number of cells per well. Seed different amounts of NK cells: two plates with 8 cells/well, 2 plates with 4 cells/well, 2 plates with 2 cells/well, 2 plates with 1 cells/well, and 2 plates with all of the remaining NK cells (the total plate). Add 100 ␮L NK cell medium per well. 2. Incubate the plates for 1 week in a 37◦ C, 5% CO2 incubator. 3. Following 1 week incubation, repeat step 3.1.2 and add irradiated feeder cells in 50 ␮L/well instead of 100 ␮L. 4. Two options to obtain polyclonal NK cell lines: a. After 2–3 days from the second addition of feeder cells, plates that were seeded with a large number of NK cells (the total plates) will become yellow, which represents cell growth. Collect the medium from the well by using

Methods to Identify and Characterize Different NK Cell Receptors and Their Ligands

255

a multichannel micropipette into a 10 mL culture dish. Then, by using a 10 mL pipette, reseed the cells into 24well plates, 2 mL/well. Validate NK cell phenotype by FACS staining (see Section 3.1.3 and step 8). b. After 2–3 days from the second addition of feeder cells, single wells in the different plates will become yellow. Collect each clone into a new 96-U plate and validated NK cell phenotype by FACS staining, using anti-CD56 and anti-CD3 or anti-NKp46 antibodies. To obtain NK cell lines, at least 20 CD3− and CD56+ clones should be mixed and grown together in 24-well plates containing 2 mL NK cell medium. 5. Once every 3–4 days, NK cells should be split by mixing each well with a 1000 ␮L micropipette and transferring 1 mL into a new 24-well plate. Then add 1 mL NK cell medium to each well for a final volume of 2 mL/well (see Note 3). 6. In order to obtain NK cell clones: a. Mix the desired NK cell clone well and split the well into two new wells of a 96 U-well plate by transferring 100 ␮L to each well. Then add 100 ␮L NK cell medium to each well for a final volume of 200 ␮L/well. b. After 2–3 days, split each well until you have 4–8 wells of each clone. c. Validate NK cell phenotype by FACS staining, using antiCD56 and anti-CD3 or anti-NKp46 antibodies. 3.2. Producing Fusion Ig

3.2.1. Generation of the Genetic Constructs

COS-7 cells are transfected with an expression vector carrying a gene encoding the extracellular domain of a desired receptor attached to the Fc portion of an IgG1 antibody. This method allows us to create an easy screening method for target cells that express the appropriate ligand(s) for a particular receptor. 1. Use PCR in order to amplify the leader peptide and the entire extracellular portion of the relevant receptor by using primers containing restriction enzyme sites for Hind III at the 5 and Bam HI at the 3 ends (see Notes 4 and 5 and Fig. 17.1). 2. Insert the fragment by ligation into an Ig expression plasmid (Homemade plasmid Cs -4Fc by Brian Seed, Harvard, Boston), which contains the Fc portion of human IgG1 (see Notes 6 and 7). Following ligation, transform the plasmid into TOP10/P3-competent bacteria and seed on LB plates containing ampicillin and tetracycline solutions (1:4000 and 1:100 for final concentrations of 0.025 and 0.01 mg/mL, respectively) (see Note 8).

256

Lankry et al.

Fig. 17.1. Fusion Ig protein production. A. Schematic representation of the fusion Ig protein production. B. Generation of the genetic constructs: amplification of the extracellular portion by PCR and insertion into the Ig fusion vector.

3. Collect 5–10 bacteria colonies and extract plasmids by using a mini prep kit. 4. Analyze gene insertion by restriction enzymes and gel electroporation. 5. Upon identifying the desired colony, produce the plasmid in large quantities by using a Maxi prep kit. 6. Send the plasmid for sequence analysis in order to ensure correct sequence and reading frame coordinates with the Fc portion. 3.2.2. Generation of the Ig-Fused Receptors in COS-7 Cells

1. Seed COS-7 cells in 6-well plates (75,000– 100,000 cells/well) in complete DMEM medium (2 mL /well) and incubate overnight in a 37◦ C, 5% CO2 incubator (see Note 9). 2. DNA precipitation: a. Precipitate 180 ␮g DNA by adding 2.5× DNA volume of 100% ethanol containing 10% sodium acetate 3 M, pH=5.3. Vortex in order to homogenize the fluids. b. Incubate overnight at −20◦ C or 2 h at −70◦ C. c. Centrifuge for 30 min at 20,000×g at 4◦ C. d. Discard the supernatant and add 1 mL of 70% ethanol. e. Centrifuge for 30 min at 20,000×g at 4◦ C. f. Work in a sterile environment. Discard the supernatant with a 1000 ␮L micropipette, and avoid touching the

Methods to Identify and Characterize Different NK Cell Receptors and Their Ligands

257

DNA sediment. Leave the tube open in order to dry ethanol remnants. g. Resuspend the DNA with 360 ␮L pre-warmed 37◦ C DMEM medium. h. Transfer the resuspended DNA into a new 50 mL tube. 3. DNA transfection: Work under sterile conditions. a. Transfer 36 mL DMEM medium into a new 50 mL tube and gently drop 720 ␮L of TransIT − LT1 Transfection reagent on top of it, incubate at room temperature for 5 min. b. Following incubation, resuspend the medium, and by using a 10 mL pipette, gently drop the entire 36 mL on top of the resuspend DNA. Incubate at room temperature for 15 min. c. Following incubation, resuspend the medium and use a 200 ␮L micropipette to gently drop 100 ␮L of the DNAcontaining medium on top of each well containing COS7 cells. Upon completing this procedure for each 6-well plate, gently swirl the plate in circular movements in order to evenly distribute the transfection reagent. d. Incubate for 6 h or overnight in a 37◦ C, 5% CO2 incubator. e. Following incubation, discard the medium from each well by using a 10 mL pipette. Gently add 2 mL of complete LPM medium lacking serum to each well (see Note 10). f. Incubate for 48 h in a 37◦ C, 5% CO2 incubator. g. Following incubation, collect the medium from each well into 50 mL tubes and gently add to each well 2 mL of new complete LPM medium lacking serum. Work under sterile conditions (see Note 11). h. Incubate for an additional 48 h in a 37◦ C, 5% CO2 incubator. i. Following incubation, collect the medium from each well into 50 mL tubes and discard the 6-well plates. Working in a sterile environment is no longer necessary! 4. Filtering the collected medium. a. Centrifugate the collected medium for 5 min at 515×g at 4◦ C to discard remaining cells. b. Following centrifugation, filter the medium with a 0.22 ␮m 500 mL filter and add a final concentration of 0.05% sodium azide. c. Transfer the medium into a 1 L beaker and keep at 4◦ C (see Note 12).

258

Lankry et al.

5. Protein purification using a protein A/G column. These steps should be performed in a 4◦ C room and all media/buffers should also be at 4◦ C. a. Pump properties: speed − 5 mL/min, rate: ×10. b. Apply 30 mL of 0.1 M glycine HCl, pH 2.7, elution buffer onto the column. c. Apply 30 mL of DMEM medium onto the column. d. Apply the medium containing the fusion protein onto the column e. Apply 30 mL of DMEM medium onto the column. f. Protein elution: transfer 10 mL of 0.1 M glycine HCl, pH 2.7, elution buffer collect the eluate into a 50 mL tube containing 1 M Tris−HCl, pH 8.8 (25 ␮L/1 mL of elution buffer). Collect at least three fractions of 10 mL each, depending on column size. Column preservation: g. Apply 30 mL of 0.1 M glycine HCl, pH 2.7. h. Apply 30 mL of double-distilled water (DDW). i. Apply 30 mL 20% ethanol. 6. Protein dialysis: a. Transfer the three fractions into three different dialysis bags (5–10 cm of dialysis tubing each) and place them into a beaker containing 2 L of cold PBS buffer in a room at 4◦ C overnight under stirring conditions. The following day, discard the PBS, add a new 2 L of PBS, and repeat the same procedure. b. The following day, transfer each fraction into a new 50 mL tube and check if the pH = 7–7.5. Determine protein concentration (by spectrophotometer, or any other appropriate protein method). c. Aliquot the proteins to 0.5–1 mL in microfuge tubes and store at −20◦ C. 3.3. BW Zeta Assay

3.3.1. Generation of Receptor-CD3-Zeta Fusion Constructs

The mouse T lymphoma cell line, BW, is transfected by electroporation with an expression vector carrying genes encoding the extracellular domain of a desired receptor and transmembrane and cytoplasmic domains of the mouse CD3-zeta chain. This method enables the generation of an in vitro system in which the activity of the receptor is measured. 1. Use two-step PCR reactions in order to amplify the entire extracellular (EC) portion of the relevant receptor fused to the transmembrane and tail of mouse CD3-zeta chain (see Fig. 17.2 and Note 13).

Methods to Identify and Characterize Different NK Cell Receptors and Their Ligands

259

Fig. 17.2. Generation of CD3-zeta-fusion constructs. A two-step PCR reaction is used in order to construct a receptor − mouse CD3-zeta fusion protein. In the first PCR reaction, amplify the receptor extracellular portion and the mouse CD3-zeta TM and cytoplasmic tail. In the second PCR reaction, attach the two segments together and the construct is inserted into the appropriate vector.

a. In the first step, amplify each of the fragments (EC portion of the receptor and the transmembrane and tail of CD3-zeta). The EC portion of the receptor should be amplified by using a 5 primer containing the appropriate restriction site and beginning of the leader sequence and a 3 primer corresponding to the last 20 bp of the EC domain of the receptor and the first 9 bp of the transmembrane of CD3-zeta in the same codon reading frame. The CD3-zeta portion should be amplified similarly by using a 3 primer containing the appropriate restriction site and a 5 primer containing the last 9 bp of the EC domain of the receptor and the first 20 bp of the transmembrane region of CD3-zeta. b. In the second step, these two fragments are mixed and amplified by PCR using the 5 primer of the receptor and the 3 primer of CD3-zeta. 2. Insert the fragment by ligation into the pCDNA3 plasmid or any other expression plasmid (see Note 14). 3. Following ligation, transform appropriate competent bacteria with the plasmid and seed them onto LB plates containing the appropriate antibiotics (in accordance with the plasmid used), such as ampicillin solution (1:1000) for final concentrations of 0.1 mg/mL. 4. Collect 5–10 bacteria colonies and extract plasmids by using a mini prep kit. 5. Analyze the gene insert by restriction enzyme digest and gel electrophoresis. 6. Once the desired colony is identified, produce the plasmid in large quantities and purify using a Maxi prep kit.

260

Lankry et al.

7. Sequence the plasmid in order to ensure correct reading frame and sequence of the EC − CD3-zeta fusion construct. 3.3.2. Stable Transfections by Electroporation

Perform these steps under sterile conditions. 1. Precipitate 100 ␮g of the CD3-zeta fusion construct DNA (see Section 3.3.1) and resuspend the DNA with 800 ␮L RPMI medium. 2. Collect 10×106 BW cells into a new 50 mL tube. 3. Centrifuge for 5 min at 515× at 4◦ C. 4. Discard the supernatant and resuspend cells in the 800 ␮L of DNA-containing RPMI medium. 5. Incubate the cells and electroporation cuvette separately for 5–10 min on ice. 6. Transfer the cells into the electroporation cuvette and perform electroporation (250 V, 500 ␮F). 7. Transfer the cells from the electroporation cuvette into a new 50 mL tube. Add 50 mL of RPMI medium and centrifuge for 5 min at 515×g at 4◦ C. 8. Discard supernatant and resuspend cells with 40 mL complete RPMI medium. 9. Seed the cells into 24-well plates at 1 mL/well. Also seed one control well which contains 250,000 untransfected BW cells (these should die under antibiotic selection). 10. Incubate for 24 h in a 37◦ C, 5% CO2 incubator. 11. Following incubation, add 1 mL/well of complete RPMI medium containing neomycin, selection medium, reaching a final concentration of 5 mg/mL G418 (see Note 15). If using a plasmid containing a different antibiotic selection gene, calibrate the antibiotic concentration required to kill the untransfected BW cells. 12. Every 2–3 days, remove 1 mL medium from each well and add 1 mL of fresh selection medium to achieve a final concentration of 5 mg/mL G418 (see Note 16). Medium refreshment should be performed until no sign of growing cells is evident (about 3–4 times). 13. Upon emergence of neomycin-resistant clones, analyze cells for protein expression by using FACS staining or any other appropriate method. 14. Grow the cells in the appropriate selective medium until stable expression is achieved (see Note 17).

3.3.3. IL-2 Secretion Assay

1. Collect transfected BW cells (effector cells; 50,000 cells/well) into a new 50 mL tube (see Note 18). 2. Centrifuge for 5 min at 515×g at 4◦ C.

Methods to Identify and Characterize Different NK Cell Receptors and Their Ligands

261

3. Resuspend cells with complete RPMI medium (100 ␮L containing 50,000 cells/well) and keep on ice until obtaining target cells. 4. Collect target cells into a new 50 mL tube. 5. Irradiate target cells with 3000 rad. 6. Add 40 mL of RPMI medium to each target cell and centrifuge for 5 min at 515×g at 4◦ C. 7. Resuspend target cells with complete RPMI medium (100 ␮L/well). 8. Seed transfected BW cells and target cells [maintaining consistent effector cell concentration, alter target cell concentration to achieve different effector to target (E:T) ratios, either 1:1 or 2:1, are recommended] in 96 F-well plates and incubate in a 37ºC, 5% CO2 incubator for 24 or 48 h (see Note 19). 9. Following incubation, centrifuge for 5 min at 515×g at 4◦ C and collect 100 ␮L of the supernatant for cytokine analysis (see Note 20). 3.3.4. ELISA for Cytokines

1. Coat MaxiSorp plate with 0.1 ␮g/well of purified rat antimouse IL-2 in 50 ␮L/well binding buffer. 2. Incubate overnight at 4ºC or 1 h at 37ºC. 3. Discard the supernatant and wash plate four times with PBS/Tween20 0.05%. 4. Add 200 ␮L/well blocking buffer (PBS/10% FCS or 3% BSA). 5. Incubate at room temperature for 2 h. 6. Discard the supernatant. 7. Add the collected supernatant from Section 3.3.3 and incubate for 4 h at room temperature or overnight at 4◦ C. 8. Discard the supernatant and wash plate four times with PBS/Tween20 0.05%. 9. Add 100 ␮L/well of biotinylated rat anti-mouse IL-2 (0.1 ␮g/well) diluted in blocking buffer containing 0.05% Tween20 and incubate for 1 h at room temperature (see Note 21). 10. Discard the supernatant and wash plate x6 with PBSx1/Tween20 0.05%. 11. Add 100 ␮L/well of streptavidin−HRP (0.1 ␮g/well) diluted in blocking buffer and incubate for 30 min at room temperature. 12. Discard the supernatant and wash plate six times with PBS/Tween20 0.05%.

262

Lankry et al.

13. Add 100 ␮L/well of TMB substrate and read the results using an Optic reader at a wavelength of 650 nm. 3.4. Cytotoxicity Assays

3.4.1. 35 S-Methionine Release Cytotoxicity Assays

NK cells are incubated with radioactive-labeled target cells. This method allows us to monitor the killing of the target cells by NK cells. By measuring the radioactivity levels released into the supernatants, we can determine efficiency of the killing. Furthermore, by using redirected killing, which is mediated by antibodies and an Fc-receptor-expressing target cell, we can define the function of a specific receptor and whether this is an inhibitory, activating, or co-stimulatory receptor. 1. Radioactive labeling of target cells: a. Collect target cells (1×106 cells/well of a 6-well plate). b. Centrifuge for 5 min at 515×g at 4◦ C. c. Resuspend cells with complete modified RPMI-1640 medium (lacking methionine), which contains [35 S]Methionine (1 ␮L of [35S]-methionine stock/1 mL complete modified RPMI medium). d. Seed the cells into a 6-well plate and incubate overnight in a 37◦ C, 5% CO2 incubator. e. Following incubation, collect the labeled target cells into a new 50 mL tube and add 40 mL of RPMI medium (normal RPMI-1640 medium from here onward). f. Centrifuge for 5 min at 515×g at 4◦ C. g. Resuspend the cells with 2–5 mL complete RPMI medium and determine cell number. 2. Seed target cells (5000 cells/well) diluted in complete RPMI medium (100 ␮L/well) into a 96 U-well plate. Seed 3 wells of each target cell alone (5000 cells/well) to determine spontaneous radioactive release and another 3 wells with target cells alone (5000 cells/well) to determine the total radioactive release (see Notes 22 and 23). 3. Seed different amounts of effector NK cells (various E:T ratios) diluted with complete RPMI medium (100 ␮L/well) to the same wells into which the target cells were seeded (see Note 24). Each E:T ratio condition should be seeded in triplicate wells. Do not seed effector cells into the wells dedicated for spontaneous and total release. 4. Incubate in a 37◦ C, 5% CO2 incubator for at least 5 h. 5. Following incubation, add 100 ␮L of 0.1 M NaOH to each total release well (medium color turns red) (see Note 25). 6. Centrifuge for 5 min at 515×g at 4◦ C.

Methods to Identify and Characterize Different NK Cell Receptors and Their Ligands

263

7. Collect 50 ␮L of cell supernatants (be careful not to touch the cells at the bottom of the well) and transfer to opaque Opti-plates. 8. Add 150 ␮L scintillation liquid to each well, cover with plastic sticker (Top seal A), and incubate in the dark at room temperature overnight. 9. Analyze results by using a ␤-counter (see Note 26). 3.4.2. Redirected Cytotoxicity Assays

1. Use the Fc-receptor-expressing P815 mouse mastocytoma cell line as target cells (see Fig. 17.3).

Fig. 17.3. Redirected cytotoxicity assays. P815 cells (mouse lymphoblast-like mastocytoma cell line), which express Fc-␥ receptors, are incubated on ice for 1 h with 0.2 ␮g/␮L antibodies which recognizes an activating receptor (gray) or an inhibitory receptor (black). The Fc portion of the IgG antibody binds to the Fc-␥ receptor expressed on the P815 cells. Once NK cells are added, the appropriate receptor on the NK cell is cross-linked and thus is activated. This allows the specific examination of a particular receptor.

2. Label target cells as described in Section 3.4.1. 3. Seed target cells (5000 cells/well) diluted with complete RPMI medium (50 ␮L/well) into a 96 U-well plate. Seed 3 wells with target cells (5000 cells/well) for spontaneous radioactive release and another 3 wells with target cells (5000 cells/well) for total radioactive release (see Notes 22 and 23). 4. Prior to NK cell addition, incubate P815 cells with the relevant mouse IgG antihuman NK cell receptor mAb (0.2 ␮g/␮L diluted in 50 ␮L/well complete RPMI medium) for 1 h on ice (this allows the binding of antibodies to the P815 cells via their Fc-receptors and thus mediates cross-linking of the desired receptor on the NK cell’s surface) (see Notes 27 and 28). 5. Continue working according to Section 3.4.1 and steps 3–9.

264

Lankry et al.

3.5. Expression of Receptors in YTS Cells

3.5.1. Generation of Point Mutant Constructs

YTS cell lines are transformed NK-like cells, which do not express all NK cell receptors. Hence, we can express a specific NK cell receptor and study its function in these cells. The insertion of a point mutation can determine the importance and function of a single amino acid in the receptor’s sequence. 1. Two-step PCR reactions can be used in order to insert a point mutation into the desired receptor (see Fig. 17.4 and Notes 29):

Fig. 17.4. Generation of point mutated constructs. A two-step PCR reaction is used in order to insert a point mutation. In the first PCR reaction, amplify the 5 and 3 portion of the gene with the appropriate primers, which contain the mutation. In the second PCR reaction, attach the two segments together, and the construct is inserted into the appropriate vector. If the mutation insertion is a part of the 5 primer or the 3 primer, a one-step PCR reaction can be performed.

a. In the first step, amplify each of the fragments by PCR (upper and lower portion of the receptor). The upper portion of the receptor should be amplified by using a 5 primer containing the appropriate restriction site and a 3 primer corresponding to the 20 bp surrounding the targeted mutation. The lower portion should be amplified similarly by using a 3 primer containing the appropriate restriction site and a 5 primer corresponding to the 20 bp of the area with the targeted mutation. b. In the second step, these two fragments are mixed and amplified by PCR using the 5 - and 3 -end primers of the receptor. 2. Insert the fragment by restriction digestion and ligation into a retroviral plasmid, such as pBabe (see Note 30). 3. Following ligation, transform the plasmid into appropriate competent bacteria, such as Stbl2 (Invitrogen) to avoid vector recombination events. Seed transformed bacteria onto LB plates containing the appropriate antibiotics (in accordance with the plasmid used), such as ampicillin (1:1000 for final concentrations of 0.1 mg/mL).

Methods to Identify and Characterize Different NK Cell Receptors and Their Ligands

265

4. Collect 5–10 bacteria colonies and extract plasmids by using a mini prep kit. 5. Analyze the gene insertion by restriction enzyme digestion and gel electrophoresis. 6. Once the desired colony is identified, produce the plasmid in large quantities by growing a large volume culture and purify using a Maxi prep kit. 7. Send the plasmid for sequence analysis in order to ensure correct reading frame, sequence and introduction of mutation. 3.5.2. Retroviral Transduction into YTS Cells

YTS cells cannot be transfected via electoporation or with other conventional methods of transfection. To avoid using human retroviruses, these cells were transfected with an ecotropic receptor and hence were given the name YTS-Eco. Protein expression in YTS-Eco cells is now also possible by using mouse retroviruses produced in BOSC packaging cells. 1. Seed BOSC cells in 10 mL culture dish (1.7×106 cells/plate in 10 mL complete DMEM medium). 2. Incubate overnight in a 37ºC, 5% CO2 incubator. 3. DNA precipitation: a. Precipitate 10 ␮g DNA by adding 2.5× of DNA volume of 100% ethanol containing 10% sodium acetate 3 M, pH 5.3 (see Note 31). b. Incubate overnight at −20◦ C or 2 h at −70◦ C. c. Centrifuge for 30 min at 20,000×g at 4◦ C. d. Discard the supernatant and add 1 mL of 70% ethanol. e. Centrifuge for 30 min at 20,000×g at 4◦ C. f. Working under sterile conditions, discard the supernatant with a 100 ␮L micropipette and avoid touching the DNA sediment. Leave the tube open in order to air dry ethanol remnants. g. Resuspend the DNA with 100 ␮L DMEM medium prewarmed to 37◦ C (see Note 32). 4. DNA transfection performed under sterile conditions: a. Transfer 400 ␮L DMEM medium into a new microfuge tube and gently drop 30 ␮L of TransIT−LT1 transfection reagent on top of it. Incubate at room temperature for 5 min. b. Using a 200 ␮L micropipette, resuspend and gently drop the 100 ␮L of DNA on top of the 400 ␮L of medium containing the transfection reagent (see Notes 33 and 34). Incubate at room temperature for 15 min.

266

Lankry et al.

c. Following incubation, resuspend the medium and use a 200 ␮L micropipette in order to transfect the BOSC cells. Gently drop all 500 ␮L of the DNA-containing medium on top of the 10 mL medium in the culture dish. Use circular movements while dropping in order to cover the entire plate. Before incubation gently swirl the plate further in circular movements in order to evenly distribute the transfection reagent in each dish. d. Incubate the cells for 48 h in a 37ºC, 5% CO2 incubator. 5. Mouse retrovirus collection and YTS-Eco cell transduction: a. Following 48 h of culture, gently collect the culture medium into a new 50 mL tube. Try to avoid collecting cells! The supernatant contains the ecotropic retrovirus. b. Centrifuge for 30 min at 14,000 rpm at 4◦ C. c. Using a 10 mL pipette, gently transfer the supernatant into a new 50 mL tube and avoid touching the cell sediment. d. Add 400,000 YTS-Eco cells into a new 15 mL tube. e. Centrifuge for 5 min at 515×g at 4◦ C. f. Discard the supernatant, resuspend cell pellet with 1.5–2 mL of virus-containing medium, add 1 ␮L of polybrene solution (5 ␮g/␮L stock), and mix. g. Seed the cells in a 96 U-well plate by adding 150 ␮L of cell suspension/well and centrifuge for 1.5 h at 515×g at 32◦ C (see Note 35). h. Incubate the cells for 5–6 h in a 37ºC, 5% CO2 incubator. i. Following incubation, collect the cells into a new 15 mL tube. j. Centrifuge for 5 min at 515×g at 4◦ C. k. Discard the supernatant, resuspend the cells in 1 mL of complete RPMI medium, seed into a well of a 24-well plate, and culture for 24 h in a 37ºC, 5% CO2 incubator. l. After 24 h incubation, add 1 mL complete RPMI medium containing G418 and puromycin (for final concentrations of 1.6 mg/mL G418 and 0.7 ␮g/mL puromycin, add 1 mL of medium containing 3.2 mg/mL G418 and 1.4 ␮g/mL puromycin, for the pBabe plasmid; for other plasmids, the antibiotic concentrations need to be established). m. Every 2–3 days, remove 1 mL medium from each well and add 1 mL of fresh selection medium (final concentrations of 1.6 mg/mL G418 and 0.7 ␮g/mL puromycin). Medium refreshment should be performed until no sign of growing cells is evident (about 3–4 times).

Methods to Identify and Characterize Different NK Cell Receptors and Their Ligands

267

n. Upon emergence of G418- and puromycin-resistant clones, analyze cells from individual wells for protein expression by using FACS staining or other appropriate method. o. Transfer positive clones into 6-well plates, and after 2–3 days, transfer the cells into 10 mL culture dishes. Grow the cells in the appropriate selection medium (final concentrations of 1.6 mg/mL G418 and 0.7 ␮g/mL puromycin) until stable expression is achieved (see Note 36). 3.6. Gene Targeting in Mice for Study of NK Cells(7)

The ability to directly target a gene of interest is one of the most powerful tools for in vivo research, as recently recognized with the Nobel prize to Mario R. Capecchi, Martin J. Evans, and Oliver Smithies (http://nobelprize. org/nobel prizes/medicine/laureates/2007/adv.html). Knocking a gene out (KO) is an ultimate test for its essential role; introduction of a reporter further allows precise detection of the cells that express it, and more sophisticated manipulations involving introducing specific recombinase sites and inducible elements open additional options to study the functional roles of genes in specific cell types and at defined times. In this section we describe a general strategy toward designing and generating a genetic model that can help in the study of NK cells in vivo.

3.6.1. Choose Your Gene

Simply said, this is the most important decision one has to take in this procedure, as one gene may yield a significant and interesting phenotype, while another might be redundant. Deep search into the literature is essential; notably many genes have multiple names that make this search a bit tricky.

3.6.2. Has This Gene Been Targeted Before?

As the number of genes is finite, and more genes are being targeted every day, it is highly possible that the gene you have picked was already targeted. If a paper was published, you may yet decide to take a different strategy [e.g., target different part of the gene, introduce a mutation (knock-in), or make a conditional KO]. Importantly, there are several large-scale efforts that have and continue to target many genes in ES cells, finding targeted cells may save you time.

3.6.3. Learn the Locus

Some genes are small and simple, while some contain many exons and spread over a long genomic region. Nevertheless, it is possible to target any gene with the right plan. Genome browsers at the NCBI (http://www.ncbi.nlm.nih.gov/ sites/entrez?db=genomeprj&cmd=Retrieve&dopt=Overview& list˙uids=169), UCSC (http://genome.ucsc.edu/cgi-bin/ hgGateway?org=mouse), and Sanger institute EMBL-EBI

268

Lankry et al.

(http://www.ensembl.org/Mus musculus/index.html) are of great help. We personally found the UCSC to have wide-ranging data nicely organized and an incorporated BLAST algorithm, while different advantages might be found in the others. Be sure to consider the known and the possible alternative splicing forms that might be expressed. If you choose to target just part of a gene, consider the possibility that the remaining parts will still be expressed and functional (either normal or abnormal). In particular, targeting only initial exons may fail due to skipping over them or use of an alternative start codon, thus resulting in no detectable phenotype. Avoid repeat elements that appear many times in the genome, as they will reduce the frequency of targeting and may prevent southern analysis. Be sure to plan precise analysis and confirm it experimentally. 3.6.4. Making a Targeting Construct

A general schematic construct with homologous arms, positive and negative selections, is shown (Fig. 17.5). Use genomic DNA from the same strain as your ES cells to assure a perfect match (different strains have variations in their genomes that may reduce efficiency of homologous recombination and complicate southern analysis). Classical cloning, PCR, and restriction enzymes are well established and sufficient, while more advanced recombineering and BAC approaches may have advantages, once familiarized. In any case, the size of the homologous arms should be in the order of several thousand bp. Within this range, shorter fragments are easier to handle, but longer ones are believed to increase the frequency of homologous recombination. Utilization of pre-made plasmids that already have positive and negative selection (and other elements if desired) will save some time. Sequencing of the whole construct is highly recommended.

Fig. 17.5. Gene targeting scheme. A gene locus, targeting vector and targeted locus are presented schematically. Restriction sites for one enzyme are presented above the original locus by the letter R and the fragment lengths by the black lines between them. Additional restriction sites for the same enzyme are present in the targeting vector and, upon correct targeting, will change the sizes of fragments detected by the 3 and 5 probes (shown above the targeted represent exons (numbered above in the original locus) and dashed lines indicate the area of homologous locus). is for a possible reporter, and is for antibiotic selection cassette (which also has recognirecombination. tion sequences for a recombinase before and after it, which would allow its removal in the mice by expression of the recombinase, such as Cre).

Methods to Identify and Characterize Different NK Cell Receptors and Their Ligands

269

3.6.5. Targeting the Gene in Mouse ES Cells

First make sure that you start with ES cells that are capable of potently establishing germ line chimeric mice (7). If your laboratory is not experienced with ES cell culturing, it is better to collaborate with someone with experience or use a professional service. The linearized construct is electroporated into the ES cells, which are then cultured under antibiotic selection. Positive clones are picked and expanded to allow screening for correctly targeted ones. It is important to maintain proper culture conditions throughout, since even a slight differentiation may impair the ability of ES cells to generate germ line chimeric mice.

3.6.6. Generating Chimeric Mice

Confirmed ES cell transfectants bearing the correctly targeted construct are either injected into blastocytes or aggregated with early morula. Both techniques require practical experience, and getting help from specialized centers is important. Using different mouse strains that differ in coat color from that of the ES cells allows easy identification of chimeric mice.

3.6.7. Obtaining Germ Line Transmitted Offspring

Founder chimeras are mated with divergent coat color mice to identify ES cell-derived offspring. However, one must also consider the genetic background necessary for further experiments. It is possible to first mate chimeric founders with another strain to identify germ line-positive mice and then mate them with the strain of choice. Backcrossing onto a desired genetic background should be started as soon as possible (literally by mating the chimeric founder with the strain of interest).

3.6.8. Verify the Correct Gene Targeting in the Mice

Characterization of newly generated mice is essential. Southern analysis for correct targeting of the gene locus should be performed, as well as northern analysis or RT-PCR for mRNA expression and protein detection by western or any additional method that applies. There are numerous examples of reasonable approaches that generated a mouse with no phenotype that were later realized to suffer from a technical problem. Having mice with verified targeting should be established as soon as possible in your new model.

3.6.9. Backcross onto a Pure Genetic Background

For NK cell studies, it is essential to backcross at least —seven to nine generations onto a chosen background mouse strain to assure homozygosity in both the NK cell gene locus (especially the Ly49 genes, which are polymorphic between mouse strains and can account for variations in NK cell responsiveness between mouse strains) and the MHC class I locus (to assure Ly49 ligands are also consistent). Several published reports of NK cell defects in knockout mouse studies have subsequently been attributed to lack of adequate backcross to a homozygous background in these gene loci, thereby preventing adequate comparisons with wild-type control mice of that background.

270

Lankry et al.

4. Notes 1. When using a buffy coat, dilution should be 1:3. 2. If separating more than 1×106 cells, multiply all added amounts accordingly, for example, for 2×106 , add 80 ␮L of cold autoMACS medium and 20 ␮L biotin Ab cocktail (multiply by 2). 3. NK cell splitting should be performed when the well becomes yellow. If cell splitting is done ahead of time, NK cell division will be stopped. 4. cDNA of the relevant gene should be constructed from mRNA of cells that express the gene. Expression should be validated by FACS staining or by any other appropriate method. 5. Restriction enzymes should be chosen according to the enzyme restriction map of the relevant gene and plasmid. Other restriction sites can be used if different vectors are utilized. 6. Design your construct so that the extracellular portion of the desired receptor will be in frame with the Fc portion. 7. The Ig plasmid is a low-copy vector; therefore, the large prep should be cultured in a large volume of 500 mL LB medium containing ampicillin and tetracycline solutions (1:4000 and 1:100, respectively). [It is now possible to work with a high copy Ig plasmid, which can be transfected into regular competent bacteria (see Fig. 17.3).] 8. TOP10 Escherichia coli cells carrying the p3 plasmid are designed for transformation of vectors that encode the synthetic supF gene (tyrosine tRNA suppressor). The p3 plasmid is a low-copy number, 60 kb plasmid that carries the drug resistance markers for kanamycin, tetracycline, and ampicillin. The kanamycin gene is fully active and is used to select for cells carrying p3. The tetracycline and ampicillin genes carry amber mutations that render the genes inactive during normal growth and replication of the bacteria. Upon transformation of a vector carrying the suppressor F gene (such as pcDNA1.1 or pCDM8), the amber mutations in the tetracycline and ampicillin genes on the p3 plasmid are suppressed and the E. coli are resistant to these antibiotics. 9. To obtain a sufficient amount of a single-fusion Ig protein, harvest from 60- × 6-well plates, for a total of 360 wells, is recommended. All calculations are for 360 wells.

Methods to Identify and Characterize Different NK Cell Receptors and Their Ligands

271

10. Medium should be gently dropped on the well’s side walls and not directly on top of the cells, since the cells are easily detached from the plate. 11. The medium at this step contains the Ig fusion protein. The collected medium should be yellow, which indicates cell growth and fusion Ig secretion. 12. Medium from the first and second collection should be pooled together. 13. Restriction enzymes should be chosen according to the enzyme restriction map of the relevant gene and plasmid. 14. The pcDNA3 plasmid is a high-copy vector; therefore, the large prep should be in a volume of 200–250 mL LB containing ampicillin (1:1000) for final concentrations of 0.1 mg/mL. 15. For final concentration of 5 mg/mL add 1 mL of medium containing 10 mg/mL G418. 16. For final concentration of 5 mg/mL add 1 mL of medium containing 5 mg/mL G418. 17. Analyze cell for protein expression after each cell transfer from one plate to another since cells sometimes tend to lose expression once transferred to larger volumes. 18. Use normal BW cells as a negative control. 19. Prepare triplicates for each E:T ratio of each target cell. 20. Supernatant can be transferred onto an ELISA plate coated with the relevant antibody or transferred to a new 96 F-well plate and kept at −20◦ C until further analysis. Once analysis is possible, thaw the supernatant at room temperature and transfer onto a MaxiSorp plate coated with the relevant antibody. 21. Importantly, the two anti-IL-2 mAb should recognize different epitopes. 22. Work with triplicates for each target cell, for each effector to target ratio, and for each experiment. 23. Spontaneous and total radioactive release determinations are important for the result analysis and calculations. It enables the comparison between different experiments. 24. Do not add NK cells to the spontaneous and total release wells!!! Equalize the volume in each spontaneous release well to 200 ␮L by adding complete RPMI medium. The volume in each total well will be normalized to 200 ␮L later on by adding 100 ␮L of 0.1 M NaOH. 25. 0.1 M NaOH should be prepared fresh while plates are being centrifuged (use 1 M NaOH stock solution).

272

Lankry et al.

26. The experiment is considered to be valid if the spontaneous release does not exceed 10–30% of the total release, meaning that the target cells were in good condition. Results are calculated as percentages according to the following equation: Target − Spontaneous  ∗ 100 = % Specific Lysis Total − Spontaneous 27. If adding two different antibodies or more, dilute each antibody so that total/sum antibodies volume/well will be 50 ␮L. 28. Depending on killing levels, one can determine if the receptor is activating, inhibitory, or co-stimulatory. An activating receptor is expected to exhibit high killing percentage following antibody cross-linking. Inhibitory receptors are expected to cause low killing percentage following combined addition of both activating and inhibitory receptor antibodies as compared to addition of activating receptor antibody alone. A co-stimulatory receptor is expected to cause low killing percentage following antibody crosslinking alone, which resembles killing levels of control antibody cross-linking or no antibody. On the other hand, co-stimulatory receptor antibody should increase killing percentage when added in combination with an activating receptor antibody, as compared to adding the activating receptor antibody alone. 29. Restriction enzymes should be chosen according to the enzyme restriction map of the relevant gene and plasmid. 30. The pBabe plasmid is a high-copy plasmid; therefore, the large prep bacterial culture should be grown in a volume of 200–250 mL LB containing ampicillin (1:1000), for final concentrations of 0.1 mg/mL. pBabe can be obtained from Dr. Garry Nolan, Stanford University. 31. It is best to work with triplicates. Therefore, 40 ␮g should be precipitated, which would include 10 ␮g for each individual transfection and 10 ␮g extra. 32. If triplicate transfections are being performed, resuspend the 40 ␮g of DNA with 300 ␮L of DMEM medium (100 ␮L/10 ␮g DNA). Ignore the extra 10 ␮g and consider as triplicates only. Also prepare three different microfuge tubes with 400 ␮L medium containing the transfection reagent, which would include one for each individual sample. 33. In contrast to the previously described transfections, be sure to drop the DNA on top of the medium containing the transfection reagent here and not the other way around.

Methods to Identify and Characterize Different NK Cell Receptors and Their Ligands

273

34. If 40 ␮g of DNA was resuspended with 300 ␮L DMEM medium, mix well and add 100–400 ␮L of DMEM medium plus transfection reagent ×3. 35. This is an important step for ensuring high transfection efficiency. 36. Analyze cells for protein expression following each cell transfer from one plate to another. Cells tend to lose expression once transferred to larger volumes. References 1. Katz, G., Gazit, R., Arnon, T. I., GonenGross, T., Tarcic, G., Markel, G., Gruda, R., Achdout, H., Drize, O., Merims, S., and Mandelboim, O. (2004) MHC class I-independent recognition of NK-activating receptor KIR2DS4. J Immunol 173, 1819–1825. 2. Biassoni, R., Cantoni, C., Pende, D., Sivori, S., Parolini, S., Vitale, M., Bottino, C., and Moretta, A. (2001) Human natural killer cell receptors and co-receptors. Immunol Rev 181, 203–214. 3. Mandelboim, O., Reyburn, H. T., ValesGomez, M., Pazmany, L., Colonna, M., Borsellino, G., and Strominger, J. L. (1996) Protection from lysis by natural killer cells of group 1 and 2 specificity is mediated by residue 80 in human histocompatibility leukocyte antigen C alleles and also occurs with empty major histocompatibility complex molecules. J Exp Med 184, 913–922. 4. Mandelboim, O., Malik, P., Davis, D. M., Jo, C. H., Boyson, J. E., and Strominger, J.

L. (1999) Human CD16 as a lysis receptor mediating direct natural killer cell cytotoxicity. Proc Natl Acad Sci U S A 96, 5640–5644. 5. Richardson, J., Reyburn, H. T., Luque, I., Vales-Gomez, M., and Strominger, J. L. (2000) Definition of polymorphic residues on killer Ig-like receptor proteins which contribute to the HLA-C binding site. Eur J Immunol 30, 1480–1485. 6. Baba, E., Erskine, R., Boyson, J. E., Cohen, G. B., Davis, D. M., Malik, P., Mandelboim, O., Reyburn, H. T., and Strominger, J. L. (2000) N-linked carbohydrate on human leukocyte antigen-C and recognition by natural killer cell inhibitory receptors. Hum Immunol 61, 1202–1218. 7. Nagy, A., Rossant, J., Nagy, R., AbramowNewerly, W., and Roder, J. C. (1993) Derivation of completely cell culture-derived mice from early-passage embryonic stem cells. Proc Natl Acad Sci U S A 90, 8424–8428.

Chapter 18 Generating NK Cell Receptor-Fc Chimera Proteins from 293T Cells and Considerations of Appropriate Glycosylation Alon Zilka, Michal Mendelson, Benyamin Rosental, Oren Hershkovitz, and Angel Porgador Abstract The use of recombinant receptors as a scientific tool has become widespread in many research fields. Of particular interest are the natural killer (NK) receptors that play a major role in the immune response against tumors and virus-infected cells. We present here (i) a detailed protocol for the production and purification of soluble recombinant NK cell receptors tagged with human IgG1-Fc (thus termed receptorFc chimera or receptor-Ig fusion protein) and (ii) a protocol for cell staining with these recombinant receptor-Fc chimeras. As these recombinant proteins are produced in eukaryotic cells, we further discuss the glycosylation pattern of these receptors that might interfere with their ligand-binding phenotype. Key words: Natural killer (NK), natural cytotoxicity receptors (NCRs), recombinant receptor-Fc chimera, transient transfection, glycoprotein.

1. Introduction Employment of recombinant immunoreceptor-Fc chimeras as a scientific tool was first reported nearly two decades ago [e.g., the elucidation of the CD28/CTLA4 and B7 interactions (1, 2)]. Today, these chimeras have become imperative for basic and applied scientific research in various fields including studies of natural killer (NK) cell receptors. When compared to anti-ligand mAbs, the receptor-Fc chimera approach can present advantages for studies involving unknown ligands, receptors that recognize a multitude of ligands, and agonist studies. We have employed the chimera approach for the identification of ligands to natural K.S. Campbell (ed.), Natural Killer Cell Protocols, Methods in Molecular Biology 612, DOI 10.1007/978-1-60761-362-6 18, © Humana Press, a part of Springer Science+Business Media, LLC 2010

275

276

Zilka et al.

cytotoxicity receptors (NCRs). Several companies (e.g., R&D, Autogen bioclear) sell recombinant NCR-Fc chimeras, yet we prefer to produce them in-house. This is not simply because of financial reasons but mostly due to a need for stringent quality and batch control; in particular, we consider the glycosylation state of these recombinant glycoproteins to be very important and therefore produce them in eukaryotic expression systems (see Notes 1 and 2). For the NCR-Fc chimera, we have utilized (i) the ectodomains of the NCRs mostly with the membrane linker region (3–6) and (ii) a human (h) IgG1 Fc (CH3 + CH2 + hinge sequences) as published previously (7) (see Note 3 for the amino acid sequence) and also used within commercial products. We have employed transient transfection (HEK-293T) or have generated stable transfectants (CHO-K1), using pCDNA3.1-based vectors (4, 5, 8, 9). For transient transfections we have also employed pDC409- or piSV-based vectors. Below, we describe a protocol for production of NCR-Fc chimeras in HEK-293T cells, purification of the chimeric protein, and use in flow cytometrybased experiments. The yield ranges between 1 and 5 mg of purified product per 1 L of supernatant from transiently transfected HEK-293T cells.

2. Materials 1. Plasticware: 25 and 10 mL pipettes, 100 mm tissue culture plates, 0.22 ␮m filters (1 L, 500 and 250 mL), and T25, T75, and T175 tissue culture flasks. 2. 5% CM DMEM: 460 mL DMEM medium containing glutamine, 25 mL fetal bovine serum (FBS; heat-inactivated), 5 mL 1 M HEPES, 5 mL nonessential amino acids, and 5 mL sodium pyruvate (all solutions from Gibco BRL, Carlsbad, CA, USA). We do not add antibiotics. 3. HEK-293T cells (American Type Culture Collection). 4. BIO-CHO (+): 500 mL BIO-CHO-1 (Biological Industries, Beit-Haemek, Israel), 5 mL BIOGRO-CHO (Biological Industries), and 5 mL L-glutamine ×100 (Gibco BRL). This is a medium without serum that is intended for collection of the supernatant from transfected cells. 5. 2 M CaCl2 : resuspend 5.88 g of CaCl2 ·2H2 O in 20 mL sterile DDW. Sterilize by passing through a 0.2 ␮m syringe filter (Corning, Acton, MA, USA). Keep in a 50 mL tube sealed with parafilm. 6. HBS (HEPES-buffered saline) ×2: in 90 mL sterile DDW, dissolve 1.6 g NaCl, 0.074 g KCl, 0.027 g Na2 HPO4 ·2H2 O, 0.2 g dextrose, and 5 mL 1 M HEPES

NK Cell Receptor-Fc Chimera Proteins

277

solution (Gibco BRL). Mix thoroughly until dissolved, add sterile DDW up to 100 mL, and filter through a 0.22 um filter. The final concentrations of the ingredients are as follows: 280 mM NaCl, 10 mM KCl, 1.5 mM Na2 HPO4 ·2H2 O, 12 mM dextrose, and 50 mM HEPES. This solution is stable for only 3 months, so make sure to note the date on the bottle. pH of the solution should be 7.05, otherwise titer (important). 7. 20% ETOH. 8. D-PBS (Dulbecco’s phosphate-buffered saline; Gibco BRL). 9. 25% trypsin-EDTA (Gibco BRL). 10. Akta Explorer FPLC system (Pharmacia Biotech, GMI Inc., Ramsey, MN, USA). 11. HitrapTM protein G HP column (GE Healthcare, Uppsala, Sweden). 12. Amicon Ultra-15 dialyzer (Millipore, Carrigtwonhill, Ireland). 13. 0.1 M glycine (pH 2.7). 14. Tris−HCL (1 M solution, pH 9). 15. PBA×10: dissolve BSA (Sigma, cat. A-7030) and sodium azide in D-PBS to a final concentration of 5 and 0.5%, respectively. Dilute PBA×10 in D-PBS to get PBA×1 (final concentrations: 0.5% BSA and 0.05% sodium azide). 16. D-PBS + 1 mM EDTA. 17. APC/PE-conjugated F(ab )2 -goat-antihuman IgG-Fc secondary antibody with minimal cross-reaction to bovine and mouse IgG (Jackson Immuno Research, West Grove, PA, USA). 18. PI (propidium iodide) stock (1 mg/mL).

3. Methods 3.1. Transient Transfection Method for Production of Recombinant Receptor-Fc 3.1.1. Preparation for Transfection

All media and solutions are sterile; use sterile technique throughout the following procedures. Log all activities from medium preparation onto the batch record of the specific production. 1. Defrost an aliquot of HEK293T cells (P2-passage freeze). Centrifuge cells and resuspended in 10 mL of 5% CM DMEM. Then add to a T25 flask and transfer to a 37◦ C, 5% CO2 incubator for cell growth.

278

Zilka et al.

2. Cell expansion: upscale from a T25 flask to T75 and T175 flasks: a. During cell expansion, harvest flasks with a confluence below 80%, thus keeping the cells in a log phase. b. Carefully decant the used medium from the flask. c. Add 10 mL D-PBS to the flask without pouring it on the cells so that they will not fall. d. Wash carefully and decant the D-PBS. Add trypsin solution for 1 min (1 mL to each T75 or 2 mL to each T175) to detach the cells from the flask without tapping. e. Add 10 mL 5% CM DMEM and transfer to the bigger flask (1:3 ratio from T25 to T75 and from T75 to T175; 25 mL in T75, and 50 mL in T175). 3.1.2. Day 1 of Transfection

1. Trypsinize the cells in the T175 flasks as above. Transfer cells (about 12.5 mL of trypsin and medium) from each T175 flask to a different 50 mL tube. Using a different tube will prevent cross-contamination, should it happened in one of the flasks. 2. Count the cells in a hemocytometer. Add 5% CM DMEM to the 50 mL tube, so the cells will be at a final concentration of 0.75×106 cells/mL. After counting the cells from all flasks (each flask in a different 50 mL tube) decide how many 100 mm plates to use. For large-scale protein prep, use fifty to sixty 100 mm plates. Take into account the available medium, plates, cells, and DNA per plate: for each plate you need 20 mL BIO-CHO (+), 2 ␮g DNA, and 0.75×106 cells. 3. Prepare the needed number of 100 mm plates and apply 9 mL of 5% CM DMEM to each plate with a 25 mL pipette. Then add 1 mL cells, stir the plate gently and allow the cells to adhere overnight.

3.1.3. Day 2 of Transfection

All calculations listed are for fifty 100 mm plates. 1. Check the 100 mm plates for a confluence of 40–50%, not more. 2. In a 50 mL tube, mix 3.1 mL 2 M CaCl2 with 22.75 mL sterile DDW (∼0.25 M CaCl2 final concentration). Then add 105.5 ␮g of plasmid DNA and mix. Let it stand for a minimum of 1 min. 3. Prepare 10 sterile 1.5 mL microfuge tubes on a stand inside the sterile hood. 4. Aliquot 490 ␮l HBS×2 to each microfuge tube. 5. Then add 490 ␮l of the CaCl2 −plasmid mix to the HBS×2: very slowly − drop by drop to create precipitation. Do so for all of the 10 microfuge tubes.

NK Cell Receptor-Fc Chimera Proteins

279

6. Collect the CaCl2 + plasmid + HBS without pipetting, using a 1 mL pipetman and add it to the cells drop by drop. 7. Stir the medium gently and incubate the plates in the 37◦ C, 5% CO2 incubator. 3.1.4. Day 3 of Transfection

1. Take out the BIO-CHO (+) medium and warm it to 37◦ C. (If you want to change medium in the morning, leave it overnight on the bench.) 2. Take out the plates from the incubator and organize them for efficient work. Raise a plate at one side and aspirate the DMEM from the other side. Remove the medium from 10 plates before applying the BIO-CHO (+). Discard the used DMEM medium. 3. Apply 10–13 mL of BIO-CHO (+) onto the side of the raised empty plate. Do not perform it too quickly, as cells can detach from the plate, and they will not re-adhere under these conditions. Gently return to the incubator.

3.1.5. Days 5 and 8 of Transfection

1. Take out the BIO-CHO (+) medium and warm it to 37◦ C. (If you want to change medium in the morning, leave it overnight on the bench.) 2. Take out the plates from the incubator and organize them for efficient work. Raise the plate at one side and aspirate the BIO-CHO (+). Filter the collected medium through a 500 mL 0.22 ␮m filter apparatus. Collect the medium from 10 plates before applying the fresh BIO-CHO (+). 3. Apply 10–13 mL of fresh BIO-CHO (+) onto the one sideraised empty plate. Do not perform it too quickly, as cells can detach from the plate and they will not re-adhere under these conditions. Gently return to the incubator. 4. Take the collected and filtered BIO-CHO (+) medium, seal the cap with parafilm, and label the bottle as collection I with date and protein name. Store at 4◦ C for the purification step. 5. On day 8, collect the medium from the plates as above, but filter the collected medium through a 1 L 0.22 ␮m filter apparatus and re-filter the day 5-collected medium (collection I) into the same 1 L apparatus. Store at 4◦ C for the purification step, yet better to purify ASAP. After collecting the medium, discard the plates.

3.2. Protein G-Based Purification of NCR-Fc Chimera on FPLC

The following protocol is generalized for a Akta Explorer FPLC (Pharmacia Biotech), although other equipment can be substituted. 1. Filter the protein-containing medium in a 0.2 um filter onto the same day it is loaded on the FPLC. As a rule, all

280

Zilka et al.

solutions (except the 20% ETOH) should be sterilized by filtration before they are loaded on the FPLC. 2. Install a 5 mL HitrapTM protein G HP column on the FPLC. 3. Wash the system and the column with 10 column volumes (CV) of DDW. 4. Change the loading buffer to D-PBS and the elution buffer to glycine 0.1 M pH 2.7 and rewash the system and the column with 10CV D-PBS. 5. Load the filtered protein-containing medium onto the column at a max flow of 5 mL/min. 6. Wash the column with 10CV of D-PBS to remove unbound proteins. 7. Install cap-less eppendorfs containing 80 ␮L of Tris−HCL 1 M pH 9 on the fraction collector. 8. Elute the recombinant protein with a linear gradient of glycine 0.1 M pH 2.7. 9. Collect the fractions containing the recombinant protein (as seen on the 280 uv OD graph) and dialyze on an Amicon Ultra-15 with the appropriate molecular weight cutoff (MWCO) against D-PBS. 10. Wash the FPLC system and the column with DDW and then with 20% ETOH. 11. Measure protein concentration and keep in aliquots at −20o C. 3.3. Cell Staining with Fusion Proteins (Receptor-Fc Chimera) Using Flow Cytometry

Keep cells and reagents on ice throughout all this procedure. 1. For floating/loosely attached cell, remove cells from the flask by pipetting with D-PBS or tapping the flask. For adherent cells, remove the medium and incubate the cells with 10 mL D-PBS + EDTA 1 mM for a few minutes. Tap the flask lightly to remove the cells. 2. Transfer the cells to a 15 mL tube and centrifuge 300×g for 6 min at 4◦ C. 3. Re-suspend the pellet in PBA, count the cells, and dilute to final concentration of 105 cells/100 ␮L. 4. Plate the cells in 96-well U-bottom plate, 100 ␮L in each well. 5. Centrifuge the plate 1100×g for 3 min at 4◦ C. 6. Discard the supernatant by flipping and gently blotting the inverted plate onto a paper towel. 7. Add 20 ␮L PBA to each well to prevent cells from drying.

NK Cell Receptor-Fc Chimera Proteins

281

8. Gently vortex the plate while holding the cover until the cell pellet breaks. 9. Prepare the fusion protein as follows: for each well, use 1–10 ␮g of fusion protein (depend upon the recombinant receptor), diluted in D-PBS to a final volume of 72 ␮L, plus 8 ␮L of PBA×10. 10. Add fusion proteins to the wells with a multichannel pipetman. Incubate on ice for 2 h. 11. Add 130 ␮L of PBA with a multi-pipetman to each well and centrifuge 1100×g for 3 min at 4◦ C. Discard the supernatant by flipping and gently blotting the inverted plate onto a paper towel. Do not vortex the plate! 12. Add a secondary antibody at 1:60 or 1:75 dilution in 50 ␮L PBA and incubate on ice for 30 min (cover with aluminum foil). We recommend APC/PE-conjugated F(ab )2 -goatantihuman IgG-Fc with minimal cross-reaction to bovine and mouse IgG as a secondary reagent. 13. Add 150 ␮L of PBA with a multichannel pipetman to each well and centrifuge 1100×g for 3 min at 4◦ C. Discard the supernatant by flipping and gently blotting the inverted plate onto a paper towel. 14. Add 200 ␮L of PBA with a multichannel pipetman to each well, transfer the samples to FACS tubes, and analyze the samples using a flow cytometry instrument. If staining the cells with PI, use 5 ␮L of PI stock +195 ␮L of PBA

4. Notes 1. The utilization of recombinant proteins including receptorFc chimeras for scientific research and biotechnology industry is associated with the concern for proper glycosylation (10). The importance of the posttranslational modification of proteins with N- or O-linked oligosaccharides is well documented by their implication in numerous biological phenomena (11, 12). In particular, the phenomenon of altered glycosylation interfering with ligand binding of a glycoprotein is well established (13, 14). The recombinant glycoproteins produced by the pharmaceutical industry are analyzed regularly for their glycan content in accordance with FDA regulations. However, insufficient analysis of glycan content is performed for recombinant glycoproteins produced solely for research purposes. We have shown that for NCRs, conjugated glycans can be involved (i) directly − e.g., the interaction of influenza hemagglutinin

282

Zilka et al.

with NKp46 and NKp44 (3, 6, 15) and (ii) indirectly − e.g., excessive N-glycosylation of NKp30 inhibits its binding to tumor membrane-associated heparan sulfate (8). The effect of N-glycosylation on the ability or NKp30 to bind to cellular ligands is not clearly evident for NKp46 and NKp44 (data not shown). This could be due to the location of the N-glycans, which are restricted to the membrane linker hinge region in NKp44 and NKp46, yet are located within the Ig-like domain for NKp30 (8). Therefore, excessive N-glycosylation within the domain could induce nonreversible glycan structuredictated changes in protein folding that take place in the ERGolgi (16). Alternatively, excessive N-glycosylation within the domain could mask the site for binding of the cellular ligands. 2. We showed that the choice of cell line for production of the NCR-Fc chimera has a prominent effect on the conjugated glycoforms (8). Moreover, the growth conditions of the producing cell line could alter the glycosylation phenotype. We compared NKp30-Fc produced in CHO-K1 cells grown in bioreactor or in tissue culture flasks and observed excessive glycosylation for the NKp30-Fc from the bioreactor-grown cells. That growth conditions induce changes of glycosylation in recombinant glycoproteins is well documented (17). Therefore, we recommend checking and comparing each new batch for size, purity, and monomer−dimer ratio by SDS-PAGE analysis (with and without ␤-ME) and for sialic acid content by 2D (IEF/SDS-PAGE) electrophoresis. If we have indications for excessive N-glycosylation, we further assess the batch glycoforms in the Center for Glycobiology at Ben Gurion University. 3. Amino acid sequence of the hIgG1 Fc region (CH3 + CH2 + hinge sequences): DPEPKSSDKTHTCPPCPAPEFEGAPSVFLFPPKPKDTL MISRTPEVTCVVVDVSHEDPEVKFNWYVDGVEVHNA KTKPREEQYNSTYRVVSVLTVLHQDWLNGKEYKCKV SNKALPAPIEKTISKAKGQPREPQVYTLPPSRDELTKN QVSLTCLVKGFYPSDIAVEWESNGQPENNYKTTPPVL DSDGSFFLYSKLTVDKSRWQQGNVFSCSVMHEALHN HYTQKSLSLSPGK

Acknowledgments This work was supported by a grant from the United States-Israel Binational Science Foundation (AP).

NK Cell Receptor-Fc Chimera Proteins

283

References 1. Linsley, P. S., Brady, W., Grosmaire, L., Aruffo, A., Damle, N. K., and Ledbetter, J. A. (1991) Binding of the B cell activation antigen B7 to CD28 costimulates T cell proliferation and interleukin 2 mRNA accumulation. J Exp Med 173, 721–730. 2. Linsley, P. S., Brady, W., Urnes, M., Grosmaire, L. S., Damle, N. K., and Ledbetter, J. A. (1991) CTLA-4 is a second receptor for the B cell activation antigen B7. J Exp Med 174, 561–569. 3. Arnon, T. I., Lev, M., Katz, G., Chernobrov, Y., Porgador, A., and Mandelboim, O. (2001) Recognition of viral hemagglutinins by NKp44 but not by NKp30. Eur J Immunol 31, 2680–2689. 4. Bloushtain, N., Qimron, U., Bar-Ilan, A., Hershkovitz, O., Gazit, R., Fima, E., Korc, M., Vlodavsky, I., Bovin, N. V., and Porgador, A. (2004) Membrane-Associated Heparan Sulfate Proteoglycans Are Involved in the Recognition of Cellular Targets by NKp30 and NKp46. J Immunol 173, 2392–2401. 5. Hershkovitz, O., Jivov, S., Bloushtain, N., Zilka, A., Landau, G., Bar-Ilan, A., Lichtenstein, R. G., Campbell, K. S., Kuppevelt, T. H., and Porgador, A. (2007) Characterization of the Recognition of Tumor Cells by the Natural Cytotoxicity Receptor, NKp44. Biochemistry 46, 7426–7436. 6. Mandelboim, O., Lieberman, N., Lev, M., Paul, L., Arnon, T. I., Bushkin, Y., Davis, D. M., Strominger, J. L., Yewdell, J. W., and Porgador, A. (2001) Recognition of haemagglutinins on virus-infected cells by NKp46 activates lysis by human NK cells. Nature 409, 1055–1060. 7. Aruffo, A., Stamenkovic, I., Melnick, M., Underhill, C. B., and Seed, B. (1990) CD44 is the principal cell surface receptor for hyaluronate. Cell 61, 1303–1313. 8. Hershkovitz, O., Jarahian, M., Zilka, A., Bar-Ilan, A., Landau, G., Jivov, S., Tekoah, Y., Glicklis, R., Gallagher, J. T., Hoffmann, S. C., Zer, H., Mandelboim, O., Watzl, C., Momburg, F., and Porgador, A. (2007) Altered glycosylation of recombinant NKp30

9.

10.

11.

12.

13.

14.

15.

16. 17.

hampers binding to heparan sulfate: a lesson for the use of recombinant immunoreceptors as an immunological tool. Glycobiology, epub Nov 15. Zilka, A., Landau, G., Hershkovitz, O., Bloushtain, N., Bar-Ilan, A., Benchetrit, F., Fima, E., van Kuppevelt, T. H., Gallagher, J. T., Elgavish, S., and Porgador, A. (2005) Characterization of the heparin/heparan sulfate binding site of the natural cytotoxicity receptor NKp46. Biochemistry 44, 14477–14485. Jenkins, N., Parekh, R. B., and James, D. C. (1996) Getting the glycosylation right: implications for the biotechnology industry. Nat Biotechnol 14, 975–981. Gabius, H. J. (2006) Cell surface glycans: the why and how of their functionality as biochemical signals in lectin-mediated information transfer. Crit Rev Immunol 26, 43–79. Rudd, P. M., Elliott, T., Cresswell, P., Wilson, I. A., and Dwek, R. A. (2001) Glycosylation and the immune system. Science 291, 2370–2376. Kaneko, Y., Nimmerjahn, F., and Ravetch, J. V. (2006) Anti-inflammatory activity of immunoglobulin G resulting from Fc sialylation. Science 313, 670–673. Krapp, S., Mimura, Y., Jefferis, R., Huber, R., and Sondermann, P. (2003) Structural analysis of human IgG-Fc glycoforms reveals a correlation between glycosylation and structural integrity. J Mol Biol 325, 979–989. Ho, J. W., Hershkovitz, O., Peiris, M., Zilka, A., Bar-Ilan, A., Nal, B., Chu, K., Kudelko, M., Kam, Y. W., Achdout, H., Mandelboim, M., Altmeyer, R., Mandelboim, O., Bruzzone, R., and Porgador, A. (2008) H5-type influenza virus hemagglutinin is functionally recognized by the natural killer-activating receptor NKp44. J Virol 82, 2028–2032. Molinari, M. (2007) N-glycan structure dictates extension of protein folding or onset of disposal. Nat Chem Biol 3, 313–320. Werner, R. G., Noe, W., Kopp, K., and Schluter, M. (1998) Appropriate mammalian expression systems for biopharmaceuticals. Arzneimittelforschung 48, 870–880.

Chapter 19 Identification of NK Cell Receptor Ligands Using a Signaling Reporter System Yoshie-Matsubayashi Iizuka, Nikunj V. Somia, and Koho Iizuka Abstract NK cell responses are regulated by a balance of inhibitory and activating signals, reflecting the net effect of interactions between receptors and ligands on target and effector cell surfaces. The identification of ligands for orphan NK cell receptors is key to enhancing our understanding of NK cell biology. Here we describe a strategy (protocol) for the identification of ligands for orphan NK cell receptors using signaling reporter cells in combination with a virus rescue system. Key words: expression cloning, NK cell receptor, retroviral expression vector, retrovirus rescue, repackaging of retrovirus.

1. Introduction Expression cloning requires screening and detection procedures with as little background noise in the system as possible. Where this is unavoidable, multiple rounds of screening and detection may reduce noise (reiterated screening). Success further depends on the specificity and sensitivity of baits (or assays) that may also contribute to reducing background noise. Here we describe a protocol for the identification of orphan NK cell receptor ligands using a signaling reporter cellular assay as bait. The expression cloning protocol is composed of two components: screening candidate cells expressing putative ligand(s) with subsequent expression cloning and enriching candidate cDNA by repackaging the integrated cDNA in a retrovirus vector (retrovirus rescue). The success of T cell expression cloning methodology has established the feasibility of using cellular reporter assays as highly specific and sensitive baits. This methodology was developed to K.S. Campbell (ed.), Natural Killer Cell Protocols, Methods in Molecular Biology 612, DOI 10.1007/978-1-60761-362-6 19, © Humana Press, a part of Springer Science+Business Media, LLC 2010

285

286

Iizuka, Somia, and Iizuka

identify a specific peptide for a T cell receptor (TCR) by using a T cell hybridoma harboring a reporter construct with a tandem repeat of the NFAT-binding site from the IL-2 promoter region following a LacZ sequence (NFAT-LacZ reporter gene) (1). The cytoplasmic tail of the CD3␨ chain was shown to be sufficient for TCR signaling by crosslinking of CD8/zeta chimeric receptors (2). Many NK cell receptors contain immunoreceptor tyrosinebased inhibitory motifs (ITIMs) or immunoreceptor tyrosinebased activating motifs (ITAMs). NK cells utilize many of the same signaling molecules involved in the TCR activation pathway. In this context, a reporter cell expressing a chimeric receptor consisting of the ectodomain of an orphan receptor with a CD3␨ cytoplasmic domain and an integrated NFAT-LacZ reporter gene can interact with cells expressing putative ligand and induce LacZ expression with specificity and high sensitivity (3) (Fig. 19.1). This reporter system allows us to detect ligand expression on a target cell surface as a positive functional readout, regardless of the native receptor function or structure. The specificity of signaling is confirmed by the negative responses of reporter cells expressing the orphan receptor without a cytoplasmic signaling chain. A signaling reporter system utilizing a chimeric receptor achieves functional specificity through cellular binding followed by a functional signaling readout. Thus, reporter cells serve as an effective screening tool for cells expressing putative ligands in expression cloning strategies. Upon identification of a cell line or cell types that stimulate reporter cells, a cDNA library within a retrovirus expres-

Fig. 19.1. Reporter assay system. Reporter cell (J7) harbors a reporter construct with three tandem repeats of the NFAT-binding sites from the IL-2 promoter followed by a lacZ sequence (1). The chimeric reporter construct − Nkrp1f ectodomain, Ly49A transmembrane (TM), and mouse CD3␨ cytoplasmic domain − is transduced into J7 reporter cells. Upon ligand (Clrg) recognition, reporter cells induce lacZ expression.

Identification of NK Cell Receptor Ligands Using a SignalingReporter System

287

sion vector can be generated from ligand-expressing cells. During screening, negative cell lines that fail to stimulate reporter cells will also be identified. These ligand-negative cell lines can be transduced with the cDNA library generated from ligand-positive cells, hence enabling ligand display and cloning using a reporter cell line. One of the most important and critical steps in expression cloning using a retroviral expression vector is to control the number of cDNA-containing vectors integrated into a recipient cell. Knowing this number enables an evaluation of how many clones should be screened from a given size of the cDNA library. As a rule you need to screen three times the base number of the library to get coverage, i.e., for a library of 106 clones, 3 × 106 integrants need to be analyzed. An estimate of the number of cDNA vector integrates is important not only for the evaluation of library coverage but also for recovering the candidate cDNA. Single integrants make the recovery and re-analysis of potential ligand cDNAs a facile process. Multiple integrations can be easily avoided by lowering the titer of infection. However, this strategy is not ideally applied to screening a cDNA library in a ligand-negative and a reporter cell assay, because lowering the multiplicity of infection (MOI) significantly increases the number of plates required for screening. On the other hand, higher MOI increase the number of integrants per cell and reduces the number of screening plates. How-

Fig. 19.2. Schematic representation of recovering stably integrated helper-free retrovirus by repackaging. Stably integrated packaging signal ()-containing helper-free viruses (GFP in this scheme) in NIH 3T3 or BW5147 (3T3-GFP and BW-GFP, respectively) were rescued by introduction of MMLV env and gag-poleither by transient transfection or by stable transduction with a lentivirus vector. IP, IRES-Puro; IB, IRES-blasticidin.

288

Iizuka, Somia, and Iizuka

ever, this approach requires a greater effort to establish which cDNA is responsible for reporter signaling. We address this problem by recovering multiple-integrated cDNAs from infected cells by repackaging them as infectious virions. To this end, we introduce Moloney murine leukemia virus (MMLV) env and MMLV gag-pol genes by either transient transfection or transduction of these genes by lentiviral vectors (Fig. 19.2). By repeating the screening and enriching the pool of cDNA for the ligand, candidate cDNA can be identified. Due to unknown nature of putative ligand and orphan receptor, a specific description of the protocol for each receptor and ligand is not applicable. Thus, in Section 3, we describe a general procedure for this cloning strategy. In Section 4, we describe an established library and reporter for Nkrp1f and the library containing its ligand cDNA, Clrg (3). We also describe potential caveats and troubleshooting related to this example receptor and ligand that may be instructive for other receptor–ligand pairs.

2. Materials

2.1. Cell Culture and Reagents for Transduction and Transfection

1. RPMI-1640 medium supplemented with 2 mM glutamine, 1 mM pyruvate, 50 ␮M 2-mercaptoethanol (2-ME), 100 U/ml penicillin, 100 ␮g/ml streptomycin, and 10% fetal bovine serum. 2. Appropriate medium for target cells expressing the candidate ligand. 3. Polybrene (Sigma-Aldrich). 4. FuGENE 6 (Roche)

2.2. CPRG Assays

1. Z-buffer (PBS pH 7.4 containing 100 mM 2-ME, 9 mM MgCl2 , and 0.125% NP-40). Store at room temperature. Stable at least 3 weeks. 2. 200 × CPRG stock solution (30 mM chlorophenol red galactoside; Calbiochem). Cover with foil, store at 4◦ C. 3. Z-buffer containing CPRG: dilute CPRG stock solution 200-fold in Z-buffer for a final concentration of 150 ␮M CPRG. 4. Stop Buffer (H2 O, 300 mM glycine, 15 mM Na2 EDTA)

2.3. Microplate Reader 2.4. Cell Lines

Several commercial options are available. We use ␮Quant (BIOTEK Instruments). 1. Reporter cell line expressing orphan receptor-CD3␨ chimeric construct and reporter cell line expressing orphan

Identification of NK Cell Receptor Ligands Using a SignalingReporter System

289

receptor without the cytoplasmic signaling motifs as negative control (see Note 1). We use BWZ-36 (mouse reporter cell line) or J7 (Jurkat cell-derived reporter cell line) as reporter cells (1, 4). 2. Slate of cell lines or primary cells to screen with the reporter cells (see Note 2). 3. Ecotropic and/or amphotropic packaging cell lines (see Note 3). We use 293T cell-based package cell lines. We use Plat-E for ecotropic virus production and Nick-A for amphotropic virus production (4, 5). 2.5. cDNA Library and Expression Vectors Containing gag-pol and env Genes

1. cDNA library generated from mRNA from the cell line stimulating reporter cells but not control reporter cells. For cDNA library construction, please refer to the protocol in each cDNA library construction kit. A MLV-derived retroviral expression vector should be used for this protocol. Here, we use the pMX vector (5). 2. pMX-GFP to use as positive control and to estimate the average number of integrated viral vector in recipient cells. 3. Mammalian expression vectors coding for MMLV env and gag-pol. We prepared eco-env and gag-pol cDNAs from MMLV genome (NC 001501) by PCR and inserted a Kozak sequence to the cDNAs. Similarly, we prepared ampho-env (M33470) with a Kozak sequence. We inserted these cDNAs into pEF-BOS vector (6) (see Note 4).

3. Methods 3.1. Screening the Slate of Cell Lines and Primary Cells

1. Culture a total of 1 × 105 reporter cells overnight with 1 × 105 target cells in a 96-well plate. Incubate them in a CO2 incubator overnight. 2. Spin down the plate for 5 min at 500 g. Stop centrifuge with minimal or no brakes. 3. Remove medium by flipping the plate. Place the plate on paper towels for 5 s to remove extra medium. 4. Add 100 ␮l of 1 × Z-buffer containing CPRG. 5. Place the plate for 4 h at 37◦ C in a regular and non-culture incubator. 6. Add 100 ␮l of stop buffer to terminate the enzyme reaction. 7. Measure the quantitative accumulation of chlorophenol red on the microplate reader by determining the absorption at 595 nm, using 635 nm as reference wavelength.

290

Iizuka, Somia, and Iizuka

8. During the screening process, cell lines negative for ligand will produce a negative readout on co-cultivation with reporter cells. One of these cell lines will be the recipient cell for the cDNA library (see Note 5). 3.2. Characterization of Ligand Expression by the Cell Line Stimulating the Reporter Cells

Before construction of the cDNA library from a ligand-positive cell line, we suggest careful examination of the amplitude of reporter response in the reporter cell assay. If augmented reporter cell responses are observed, i.e., a greater amplitude, it suggests that ligand expression is increased at the cell surface, possibly due to increased mRNA for the ligand or possibly co-expression of another ligand that can stimulate reporter cell responses. Therefore, a cDNA library generated from these cells may increase the chance of identifying the ligand. Various stimulating reagents that may induce or increase ligand expression, such as cytokines and LPS, may also need to be tested. Special caution must be paid to the reagents that can directly stimulate TCR signaling because the remaining reagents in the system may directly stimulate reporter cell responses. This can be easily controlled by applying the reagent onto the reporter cell and assaying for reporter expression. Construct a cDNA library using mRNA from the identified ligand-positive cells line. The cDNA should be cloned into a retroviral expression vector and the base number of independent cDNA clones should be quantified.

3.3. Determine the Infection Efficiency of Recipient Cells with Marker Viruses

Since the infection efficiency of retroviral vectors can vary depending on the cell type, it is important to quantify the infection efficiency of the ligand-negative recipient cells (see Note 6). 1. Prepare supernatant containing retroviral vectors made from pMX-GFP. 2. Infect the recipient cells with undiluted and serial dilution pMX-GFP supernatant in the presence of polybrene at 10 ␮g/ml on day 0. Incubate for 6 h. Add the same amount of fresh medium into the wells. 3. On day 1, harvest cells, wash out virus supernatant, and incubate cells in fresh medium. 4. On day 2, analyze the GFP expression by FACS. 5. With serially diluted viral vector supernatants, we observe a condition where infection efficiency (i.e., the number of cells infected) correlates with the dilution of the vector (i.e., the linear responses), but the mean fluorescence intensity (MFI) of GFP is constant. We assume that these cells harbor a single GFP vector. In our experience, most cell lines have single GFP integration when the infection efficiency is less than 5–15%. By comparing MFI of GFP between single

Identification of NK Cell Receptor Ligands Using a SignalingReporter System

291

virus integration and multiple integrations, one can estimate the approximate number of GFP virus integrations. For a 20-fold increase in MFI compared to single GFP cell populations, we estimate that this cell population has a corresponding 20-fold increase in integrated cDNA. 3.4. Determine the Sensitivity of Reporter Cell Assay

Determine the sensitivity of reporter cell assay by serially diluting the percentage of ligand-positive and recipient cells in a total of 1 × 105 target cells in a 96-well plate. If you can detect one positive cell out of 100 recipient cells (1% positive cells), the sensitivity is 100.

3.5. Expression Cloning of Ligand with Reporter Cells

1. On day 0, transfect the cDNA library into the packaging cell line. As a control virus for vector production and to track infection efficiency, transfect with pMX-GFP in a duplicate plate of packaging cells. 2. On day 2, harvest the supernatant containing the library viral vectors and infect recipient cells as in Section 3.3. Infect GFP virus into recipient cells as control. 3. On day 4, analyze the GFP expression on a per-cell basis by fluorescence cytometry and calculate the number of cells infected. Assume a similar infection for the library vectors. 4. Plate the number of library-infected cells in each well of a 96-well plate corresponding to the number for reporter cell sensitivity determined in Section 3.4. Determine the number of plates to cover the library size (see Note 7). 5. Maintain library-infected cells in 96-well plates by changing 50% of the medium (see Note 8). When cells become 80% confluent, duplicate all the plates. 6. When duplicated plates are semi-confluent, perform reporter cell assays on one set of the duplicated plates by adding 1 × 105 reporter cells. It is not necessary to count the cell number of each recipient cell well. 7. Positive wells will be identified as described in Section 3.1. 8. Cells from the duplicated positive well will be expanded and applied to the reporter assays in a quantitative manner (i.e., 1 × 105 cells in a 96-well plate) together with noninfected recipient cells as a negative control. Once the positive well is confirmed by the reporter cells, make backup frozen vials of these bulk populations. 9. Establish a positive clone or oligo-clonal cell population by limiting dilution. 10. Introduce into the positive clonal cells the MMLV env and gag-pol genes either by transient transfection or stable transduction with a lentivirus vector.

292

Iizuka, Somia, and Iizuka

11. Harvest the supernatant of positive clonal cells that have been either transfected or transduced with the env and gagpol genes. The supernatant should contain the retrovirus encoding cDNA for the ligand. Infect the next round of recipient cells with the supernatant (see Note 9). 12. Two days after the infection, perform the reporter cell assay with infected cells. Confirm that the infected cells can stimulate reporter cells. Plate cells into 96-well plates and perform a second screening. 13. After establishing positive clonal cells in the second round of screening, isolate the integrated cDNA by PCR using primers specific to the expression vector. If oligo-bands are observed, sequence them and confirm the specificity with reporter cells after transducing the candidate cDNA(s) into recipient cells. If multiple PCR bands are observed, further repackaging of cDNA and a third round of screening may be required to simplify the downstream analysis.

4. Notes 1. The Nkrp1f chimeric receptor cell was generated with the Nkrp1 ectodomains and CD3␨ cytoplasmic domains. To avoid difficulties in expression of putative activation receptors and a confounding association with signaling adaptor molecules that contain charged transmembrane (TM) residues, we replace the TM domain with that from Ly49A (Fig. 19.1). For orphan receptors with a type I structure, we replace the TM domain with that from CD8. A bulk population of reporter cells is perfectly suitable for the initial screening of ligand-expressing cells. We sometimes observe that reporter cells with low levels of receptor expression become dominant after a long period of culture. Therefore, it is advisable to establish clonal reporter cells expressing a high level of receptor for use in expression cloning procedures following the identification of ligand-positive cells. Such clonal reporter cells will make the assay condition stable, in addition to increasing the sensitivity of the assay. 2. Cell lines are preferable as a source of mRNA for cDNA library construction. The size of the cDNA library from heterogeneous cell populations will not reflect the distribution and abundance of mRNA from a homogeneous cell population. 3. We utilize FuGENE 6 for plasmid transfection according to the manufacturer’s instruction. We utilize 2 × 106

Identification of NK Cell Receptor Ligands Using a SignalingReporter System

293

packaging cells, 9 ␮l of FuGENE6, and 3.5 ␮g of plasmid DNA in a 10 cm dish with 8 ml medium. We strongly recommend establishing a transfection condition using GFP expression in pMX or the vector utilized for the cDNA construction, because we have observed significant variability in transfection efficiency among laboratories. We use freshly prepared supernatant containing the virus because the freezing of supernatant decreases the viral titer. Filtration of virus supernatant also decreases the titer. Unless the packaging cell line is capable of stimulating reporter cells, we utilize the unfiltered supernatant after a brief centrifuge (500 g, 10 min in a table top centrifuge). 4. Alternatively, recombinant lentiviruses or adenoviruses expressing Env and Gag-Pol can be utilized. For lentivirus expression, an expression plasmid for VSV-G and the delta8.9 plasmid (7) are required. The choice of the method for retrovirus rescue is dependent on several factors related to the recipient cells: efficiency of the gene delivery method and expression level of Env and Gag-Pol achieved by the vector system. When recipient cells are adherent cells or harbor large T antigen, we utilize env and gag-pol in a pEF-BOS vector (6). When we repackaged pMX-GFP from NIH3T3 cells by transfecting these vectors, we observed that more than 40% of 3T3 cells expressed GFP after infection from the day 2 supernatant (following transfection) (Fig. 19.2A and data not shown). When cells are non-adherent and refractory to conventional transfection, we prefer to use a lentivirus vector, pEF-SIN (8), containing env-IRES-puro and gag-pol-IRESblasticidin (Fig. 19.2A). Puromycin and blasticidin selection is required for efficient repackaging of the pMX-GFP when we utilize BW5147 cells as recipient cells (data not shown). When we repackaged pMX-GFP from BW-GFP by sequential infection and selection, we observed that 5% of BW5147 cells expressed GFP from the day 2 supernatant (data not shown). These data indicate that each cell type can repackage the integrated virus cDNA with different efficiency. Please note that a regular retrovirus vector cannot be used in this strategy unless the expression vector has been modified to a “self-inactivating” (SIN) form (9). Similar repackaging methodology using adenovirus (Ad) has also been described and high virus titer recovery is reported (10). Infection efficiency of adenovirus is dependent on the expression of coxsackie-adeno receptor (CAR). When CAR expression is low in the recipient cells, Ad containing an

294

Iizuka, Somia, and Iizuka

Arg-Gly-Asp (RGD) motif mutant in the fiber-knob region (11, 12) may be an efficient alternative vector to overcome this problem. 5. The choice of recipient cell is critical. Ideal characteristics of a recipient cell line for this expression cloning strategy are (1) the cell line can integrate many virus cDNAs without compromising proliferation ability and (2) the cell line can be efficiently transfected or virally infected for expression of env and gag-pol genes for repackaging. Please refer to Notes 4, 7, and 9 for further detail. The biggest caveat in expression cloning is that the ligand may be a complex of molecules. To account for this possibility, a parental cell line capable of stimulating reporter cells and a derivative mutant cell line incapable of stimulating reporter cells can be searched for or generated. The derivative mutant cell line can be used as recipient cell due to the prediction that simultaneous mutations in the genes of the ligand complex are unlikely, unless the complex genes are on the same loci. Alternatively, using a similar cell type to the ligand-expressing cell may resolve this problem. Successful expression cloning of a ligand complex has been reported for the DX-5 antigen with a mAb in this manner (13). 6. According to Poisson distribution, when MOI of 1 (1 viral particle per cell) is used to infect a population of cells, the probability that a cell will not get infected is 37%, and the probability that it be infected by a single particle is 37%, by two particles is 18%, by three particles is 6%, and so on. However, we observe variability among cells utilized for infection. As a practical approach to estimate the average number of integrated viral vector in recipient cells, we utilize FACS with a GFP vector as surrogate marker. Hence, we assume that the titer of GFP vector preparation is similar to the one of retroviral cDNA library vector generated under the same conditions. Please note that our estimate is based on the assumption that the number of integrated GFP vector and the MFI of GFP expression are in a linear phase when measured on the FACS. Clonal analysis followed by a Southern blotting or quantitative PCR may be an alternative method (14). 7. Estimate the number of cDNA that can be screened in a 96-well plate. For example, if the sensitivity is 100 and the estimated number of integration is 10, then approximately 96,000 cDNA clones can be screened in a 96-well plate (100/sensitivity per well × 10/integration per cell × 96 well). Another consideration is the doubling time of recipient cells. If the doubling time is 24 h, it should be assumed that the 100 cells on day 2 represent the 25 cells on the day

Identification of NK Cell Receptor Ligands Using a SignalingReporter System

295

of infection. Plate the number of cells corresponding to the sensitivity into each well in the number of 96-well plates required to cover the library size. We utilized the cDNA library for Nkrp1f ligand, which contained 2.8 × 106 independent inserts (bi-directional insertion). When we utilized 293T cells as recipient cells, we observed that a 293T cell can harbor four pMX-GFP inserts with our supernatant containing pMX-GFP viruses (data not shown). When we plated 100 cells per well in 20 of 96-well plates, we observed two positive wells with J7Nkrp1f reporter cells and confirmed the insertion of Clrg cDNA by PCR and sequencing (data not shown). When we utilized BW-5147 cells as recipient cells, we observed that a BW-5147 cell can harbor eight pMX-GFP inserts with our supernatant containing pMX-GFP viruses (data not shown). When we plated 100 cells per well in 20 of 96-well plates, we observed 13 positive wells with J7-Nkrp1f reporter cells and confirmed the insertion of Clrg cDNA by PCR and sequencing (data not shown). 8. We utilize a multi-channel pipette for changing the medium. We change tips between removing and adding medium for each plate but not for each well. We set a constant ori-

Fig. 19.3. Recovery of stably integrated helper-free retrovirus as infectious helper-free virus. Clonal 293T cells stimulating Nkrp1f reporter cells were established from a positive well in the first round of screening and were transfected with pBOS-AmphoEnv and pBOS-gp. Two days after the transfection, supernatant were harvested, serially diluted, and used for infection with 293T cells. Two days after the infection, 1 × 105 of infected 293T cells were applied to Nkrp1f reporter cells assays.

296

Iizuka, Somia, and Iizuka

entation for this procedure. By doing so, we are able to detect potential cross-contamination if we observe a gradual decline of reporter responses in the orientation. 9. As was already mentioned in Note 4, the titer of repackaged viruses varies depending on the recipient cells. A high repackaging efficiency (i.e., high virus titer) may be observed: We established oligo-clonal 293T cells capable of stimulating Nkrp1f reporter cells from the first screening process described in Note 6. To rescue integrated virus in an amphotropic envelope, we transfected them with ampho-env and gag-pol in a pEF-BOS vector. Supernatant was harvested 2 days after transfection. We infected 293T cells with serially diluted supernatant. We observed that Nkrp1f reporter cells detected ligand-expressing cells infected even with 16 times diluted supernatant (Fig. 19.3). When low repackaging efficiency (i.e., low virus titer) is observed, we recommend the following protocol to concentrate retroviruses (15): spin filtered supernatant at 6000×g for 16 h at 4◦ C and resuspend in a cell culture medium or Hank’s balance salt solution (HBSS). With this method, we observed at least 10 times the concentration of the virus supernatant. References 1. Sanderson, S., and Shastri, N. (1994) LacZ inducible, antigen/MHC-specific T cell hybrids. Int Immunol 6, 369–376. 2. Irving, B. A., and Weiss, A. (1991) The cytoplasmic domain of the T cell receptor zeta chain is sufficient to couple to receptorassociated signal transduction pathways. Cell 64, 891–901. 3. Iizuka, K., Naidenko, O. V., Plougastel, B. F., Fremont, D. H., and Yokoyama, W. M. (2003) Genetically linked C-type lectinrelated ligands for the NKRP1 family of natural killer cell receptors. Nat Immunol 4, 801–807. 4. Ito, D., Iizuka, Y. M., Katepalli, M.P., Iizuka, K. (2009) Essential role of the Ly49A stalk region for immunological synapse formation and signaling. Proc Natl Acad Sci USA. 106(27), 11264–11269. 5. Morita, S., Kojima, T., and Kitamura, T. (2000) Plat-E: an efficient and stable system for transient packaging of retroviruses. Gene Ther 7, 1063–1066. 6. Mizushima, S., and Nagata, S. (1990) pEFBOS, a powerful mammalian expression vector. Nucleic Acids Res 18, 5322. 7. Lois, C., Hong, E. J., Pease, S., Brown, E. J., and Baltimore, D. (2002) Germline transmis-

8.

9.

10.

11.

sion and tissue-specific expression of transgenes delivered by lentiviral vectors. Science 295, 868–872. Cui, Y., Golob, J., Kelleher, E., Ye, Z., Pardoll, D., and Cheng, L. (2002) Targeting transgene expression to antigenpresenting cells derived from lentivirustransduced engrafting human hematopoietic stem/progenitor cells. Blood 99, 399–408. Yu, S. F., von Ruden, T., Kantoff, P. W., Garber, C., Seiberg, M., Ruther, U., Anderson, W. F., Wagner, E. F., and Gilboa, E. (1986) Self-inactivating retroviral vectors designed for transfer of whole genes into mammalian cells. Proc Natl Acad Sci U S A 83, 3194–3198. Bhattacharya, D., Logue, E. C., Bakkour, S., DeGregori, J., and Sha, W. C. (2002) Identification of gene function by cyclical packaging rescue of retroviral cDNA libraries. Proc Natl Acad Sci U S A 99, 8838–8843. Dmitriev, I., Krasnykh, V., Miller, C. R., Wang, M., Kashentseva, E., Mikheeva, G., Belousova, N., and Curiel, D. T. (1998) An adenovirus vector with genetically modified fibers demonstrates expanded tropism via utilization of a coxsackievirus and adenovirus

Identification of NK Cell Receptor Ligands Using a SignalingReporter System receptor-independent cell entry mechanism. J Virol 72, 9706–9713. 12. Nagi, P., Vickers, S. M., Davydova, J., Adachi, Y., Takayama, K., Barker, S., Krasnykh, V., Curiel, D. T., and Yamamoto, M. (2003) Development of a therapeutic adenoviral vector for cholangiocarcinoma combining tumor-restricted gene expression and infectivity enhancement. J Gastrointest Surg 7, 364–371. 13. Arase, H., Saito, T., Phillips, J. H., and Lanier, L. L. (2001) Cutting edge: the mouse NK cell-associated antigen recognized by DX5 monoclonal antibody is CD49b (alpha 2 integrin, very late antigen-2). J Immunol 167, 1141–1144.

297

14. Kustikova, O. S., Wahlers, A., Kuhlcke, K., Stahle, B., Zander, A. R., Baum, C., and Fehse, B. (2003) Dose finding with retroviral vectors: correlation of retroviral vector copy numbers in single cells with gene transfer efficiency in a cell population. Blood 102, 3934–3937. 15. Bowles, N. E., Eisensmith, R. C., Mohuiddin, R., Pyron, M., and Woo, S. L. (1996) A simple and efficient method for the concentration and purification of recombinant retrovirus for increased hepatocyte transduction in vivo. Hum Gene Ther 7, 1735–1742.

Chapter 20 Determining Ligand Specificity of Ly49 Receptors Kerry J. Lavender and Kevin P. Kane Abstract Ly49 receptors in rodents, like KIR in humans, play an integral role in the regulation of NK cell activity. Some inhibitory Ly49 are known to interact with specific MHC I alleles to maintain tolerance to self tissues, and NK activation is triggered upon the loss of inhibitory signals due to pathological downregulation of self MHC I. Although a virally encoded ligand has been identified that can trigger NK cytotoxicity through an activating Ly49, some activating Ly49 also recognize MHC I and the role of most activating receptors in NK effector function remains poorly defined. As many Ly49 remain orphan receptors, we describe methods to unambiguously discern receptor–ligand pairs. Additionally, we describe a method for the mutagenesis of Ly49 and MHC ligands that can be used to define the motifs conferring receptor specificity for their ligands. Further elucidation of Ly49 ligands is required to continue to define the role of Ly49 in regulating NK cell effector function and may give vital clues to the role of KIR in human health and disease. Key words: Rodent, MHC I, Ly49, mutagenesis, transfection, 51 Cr release assay.

1. Introduction Ly49 are C-type lectin-like receptors found in mice and rats that regulate the balance of NK cell activation and tolerance. These receptors are functional homologues to the immunoglobulinlike KIR receptors that regulate NK cell activity in humans and other primates. Ly49 are encoded as a multi-gene family and inbred rodent strains can differ with respect to their complement of genes encoding activating and inhibitory type Ly49 (1–3). To date, more than 20 Ly49 genes have been described in the mouse, 13 of which are inhibitory and 8 that are activating (4). Furthermore, 25 Ly49 have been identified in the rat including 13 inhibitory, 8 activating, and 5 bi-functional receptors (1). K.S. Campbell (ed.), Natural Killer Cell Protocols, Methods in Molecular Biology 612, DOI 10.1007/978-1-60761-362-6 20, © Humana Press, a part of Springer Science+Business Media, LLC 2010

299

300

Lavender and Kane

Ligands for some Ly49 have been identified (5–10) and like the few ligands identified for KIR receptors (11–15), these primarily consist of a single or small repertoire of specific MHC I alleles. Additionally, a mouse cytomegalovirus-encoded MHC I homologue, m157, has been identified as the ligand for the activating Ly49H and inhibitory Ly49I receptors of specific mouse strains and may suggest that similar viral ligands could exist in human systems (16). Despite some progress in identifying Ly49 ligand specificity, many Ly49, particularly those in the rat, remain orphan receptors. As there are many parallels between Ly49 and KIR functionality in NK cell biology, continued elucidation of Ly49 receptor−ligand pairs in rodents can provide valuable models to further understand the significance of KIR receptors in human health and disease. Although identifying ligands for Ly49 can be approached broadly using activated effector and target cells from inbred or congenic strains, it is nearly impossible to deduce specific interactions from this system as most targets will express more than one potential MHC I ligand. Additionally, the complex signals derived through the interaction of MHC I and the variegated Ly49 repertoires expressed by NK cells in any individual animal can result in complicated and potentially ambiguous interpretations of ligand specificity. To remove such ambiguity, we describe a system involving the transfection of potential receptor−ligand pairs into a specific RNK-16 NK cell effector−YB2/0 target system followed by a description of a 51 Cr release assay for direct and specific recognition. The RNK-16/YB2/0 system was originally described by J. Ryan et al. (17), and our description here is an update and expansion of the approach. The effector−target system displays a moderate level of background cytotoxicity that allows the detection of increased cytotoxicity upon ligand recognition by an activating receptor, as well as suppressed cytotoxicity upon ligand recognition by an inhibitory receptor. Furthermore, the isolation of MHC I and Ly49 DNA for transfection allows modifications of the sequence, including the addition of tags to facilitate or confirm the successful transfection and potentially the surface expression of molecules that cannot be identified by specific antibody. We give one example here of the C-terminal fusion of EGFP to MHC I that we have used to facilitate screening for positive transfectants. Additionally, we describe a method for mutagenesis of either the ligand or Ly49 to identify the epitopes conferring specificity, as such knowledge can aid in the description of motifs that can lead to the identification of other potential ligands for Ly49. Finally, slight alterations to the methods we discuss here can allow the identification of virally encoded or induced ligands for Ly49, by modifying the assays to include infected target cells or target cells transfected to express virally encoded ligands.

Identifying Ly49 Ligands

301

2. Materials 2.1. Mutagenesis of Ly49 and MHC I

1. DNA plasmid containing the MHC I or Ly49 insert to be mutated (see Note 1). 2. Mutagenesis primers (1.25 ␮g/␮l in 10 mM Tris−Cl, pH 8.5 (see Note 2). 3. PfuTurbo DNA polymerase (2.5 U/␮l) and 10× buffer (Stratagene, La Jolla, CA). 4. dNTP mix (100 mM) (Stratagene, La Jolla, CA) (see Note 3). 5. RNAse/DNase-free H2 O (Invitrogen, Carlsbad, CA). 6. DpnI restriction enzyme (10 U/␮l) (New England BioLabs, Ipswich, MA). 7. Max-efficiency DH5␣ competent cells (Invitrogen, Carlsbad, CA). 8. Agar plates containing antibiotic selection appropriate to your vector, for example, ampicillin (50 ␮l/ml) for pC1-neo and BSR␣EN or kanamycin (30 ␮l/ml) for pEGFP-N3. 9. Luria-Bertani (LB) Broth. 10. 50% glycerol in RNase/DNase-free H2 O.

2.2. Transfection of YB2/0 and RNK-16 Cells

1. Endo-free DNA preps of the plasmid containing the molecule to be transfected (see Notes 1 and 4). 2. RNK-16 growth medium: RPMI-1640 (Invitrogen/Gibco, Carlsbad, CA) supplemented with 10% FBS (Hyclone, Logan, UT), 1% of penicillin/streptomycin, 1% Lglutamine, and 0.1% mercaptoethanol (Invitrogen/Gibco, Carlsbad, CA). 3. YB2/0 growth medium: Dulbecco’s modified Eagle’s medium, (DMEM) (Invitrogen/Gibco, Carlsbad, CA) supplemented with 10% FBS (Hyclone, Logan, UT), 1% L-glutamine, 1% sodium pyruvate, and 1% penicillin/streptomycin (Invitrogen/Gibco, Carlsbad, CA). 4. Unsupplemented DMEM and RPMI-1640 medium (Invitrogen/Gibco, Carlsbad, CA). 5. Geneticin (G418) (50 mg/ml) (Invitrogen/Gibco, Carlsbad, CA). 6. Electroporation cuvettes (4 mm) and electroporator (BioRad, Hercules, CA). 7. Black-walled, flat and clear-bottomed, tissue culture-treated 96-well plates (Thermo Scientific, Asheville, NC).

302

Lavender and Kane

2.3. Screening MHC I−EGFP and Ly49 Transfectants

1. FLA-5100 imaging system (Fujifilm). 2. Phosphate-buffered Carlsbad, CA).

saline

(PBS)

(Invitrogen/Gibco,

3. ChromPure Rat or Mouse IgG (Jackson Immunoresearch, West Grove, PA) or serum can be obtained directly from animals and maintained as frozen stocks. 4. Purified primary antibody specific to the transfected molecule. 5. FITC- or PE-conjugated AffiniPure F(ab’)2 fragment donkey anti-mouse (or rat) secondary antibody (Jackson Immunoresearch, West Grove, PA). 6. 37% formaldehyde solution. 7. 10% fetal bovine serum (FBS) (Hyclone, Logan, UT). 8. Dimethyl sulfoxide (DMSO) (Pierce Biotechnology, Inc., Rockford, IL). 2.4. Assay of Recognition

1. RNK-16 growth medium: RPMI-1640 (Invitrogen/Gibco, Carlsbad, CA) supplemented with 10% FBS (Hyclone, Logan, UT), 1% of penicillin/streptomycin, 1% Lglutamine, and 0.1% mercaptoethanol (Invitrogen/Gibco, Carlsbad, CA). 2. YB2/0 growth medium: Dulbecco’s modified Eagle’s medium (DMEM) (Invitrogen/Gibco, Carlsbad, CA) supplemented with 10% FBS (Hyclone, Logan, UT), 1% L-glutamine, 1% sodium pyruvate, and 1% penicillin/streptomycin (Invitrogen/Gibco, Carlsbad, CA). 3. Wash medium: RPMI-1640 (Invitrogen/Gibco, Carlsbad, CA) supplemented with 3% fetal bovine serum (FBS) (Hyclone, Logan, UT). 4. Lead shielding (Marshield, Burlington, ON) (see Note 11). 5. 1 mCi/ml Na51 CrO4 (51 Cr) (Perkin Elmer, Woodbridge, ON, Canada). 6. 25 and 100 ␮l multichannel pipette(s). 7. Scintillation fluid (OptiPhase SuperMix, Perkin Elmer, ON, Canada). 8. TriLux Microbeta and flexible 96-well microplates (Perkin Elmer, Waltham, MA).

3. Methods 3.1. Mutagenesis of Ly49 and MHC I

1. Thaw your DNA template, primers, 10× PfuTurbo reaction buffer and dNTPs (see Note 3). Centrifuge briefly and place on ice.

Identifying Ly49 Ligands

303

2. In a PCR tube add a. 5 ␮l of 10× PfuTurbo reaction buffer b. 50 ng of DNA template c. 1 ␮l of each mutagenesis primer d. 1 ␮l of dNTP mix e. RNase/DNase-free H2 O to a final volume of 50 ␮l f. 1 ␮l PfuTurbo DNA polymerase (see Note 3) g. Overlay with mineral oil if required for your thermocycler Place in thermocycler and cycle as follows. Segment

Cycles

Temperature, ºC

Time

1

1

95

30 s

95

30 s

55

1 min

68

2 min/kb of plasmid length

2

12–18

∗

∗ 12

cycles for point mutations, 16 for single amino acid changes, and 18 for multiple amino acid insertions or deletions.

3. After cycling, cool the reaction to ≤37ºC and add 1 ␮l of DpnI (see Note 3) to the reaction tube. Ensure to insert the pipette tip below the mineral oil overlay, if used. 4. Mix gently by pipetting up and down several times, spin for 1 min in a microcentrifuge, and incubate at 37ºC for 2 h to overnight. 5. The next day, transform max-efficiency DH5␣ cells using 4 ␮l of the DNA reaction mixture. Streak on plates containing the appropriate selection antibiotic for your vector and incubate overnight at 37ºC. 6. Select two to three colonies for overnight expansion in 5 ml of LB broth plus selection antibiotic (for example, 30 ␮l/ml kanamycin or 50 ␮l/ml ampicillin). 7. Next day create glycerol stocks in a 1.5 ml microtube using 700 ␮l of culture and 300 ␮l of 50% glycerol, store at −80ºC. 8. Purify plasmid from the remainder of the culture and perform DNA sequencing to ensure the correct mutation has been introduced. 9. Make Endo-free preps of successful mutants using the stored glycerol stock. 3.2. Transfection of YB2/0 and RNK-16 Cells

1. Culture YB2/0 cells in appropriate growth medium. 2. Culture RNK-16 cells in appropriate growth medium (see Note 5).

304

Lavender and Kane

3. On the day of transfection remove cells in log-phase growth from 100 mm tissue culture plates using cell scrapers for YB2/0 or gentle washing for RNK-16 cells (see Note 6). 4. Harvest 2×106 cells per transfection for YB2/0 and 4×106 cells for RNK-16. 5. Resuspend each transfection in 0.4 ml cold unsupplemented DMEM (YB2/0) or RPMI-1640 (RNK-16). 6. Add 10 ␮l (YB2/0) or 20 ␮l (RNK-16) of sterile plasmid (see Note 7) per transfection, mix gently, and transfer into a 4 mm electroporation cuvette. Place briefly on ice and proceed immediately to transfection. 7. Turn the Bio-Rad electroporator off and on a few times. Set the electroporator to 0.18 V and the capacitance extender to 960 ␮F. 8. Insert each cuvette snugly into the cuvette holder ensuring good metal-to-metal contact. Depress the double red buttons on the front of the machine simultaneously until a beep is heard indicating electroporation is complete. 9. Replace the cuvette onto ice and as soon as possible add the contents of the cuvette to 7 ml of growth medium dropwise in a 100 mm tissue culture-treated plate. 10. Aspirate the medium the day after transfection and resuspend the cells in 10 ml growth medium supplemented with 1 mg/ml G418. Count and dilute YB2/0 cells to 2×104 /ml in G418-supplemented growth medium, making enough for two 96-well plates (25 ml) though more plates may be added for stubborn transfections. RNK-16 cells are left undiluted. 11. Plate cells at 100 ␮l/well into flat-bottomed, tissue culturetreated 96-well plates: RNK-16 cells into standard plates and YB2/0 cells into black-walled plates. 12. After 4 days, add an additional 100 ␮l of G418supplemented growth medium to each well. 13. After an additional 5 days check wells for growth. Up to 30% of YB2/0 wells will show growth in a successful transfection; additional growth suggests the wells may not be clonal. RNK-16 cell transfection is much more difficult and successful transfection efficiency can be 1–3% or less. RNK-16 cells often take 2–4 weeks to exhibit growth and additional G418 medium may be added to avoid excessive evaporation. 3.3. Screening for MHC I–EGFP and Ly49 Transfectants

1. Aspirate medium from wells containing actively growing YB2/0 clones that have become confluent in the well (see Note 8) and proceed to the FLA-5100 Imaging System.

Identifying Ly49 Ligands

305

2. Scan the plate for the presence of EGFP fluorescence. Positive wells will appear in a uniform dark color if cells are evenly distributed and confluent in the well. 3. Transfer EGFP-positive YB2/0 cells and actively growing RNK-16 cells from 96-well plates into 1 ml of G418supplemented growth medium (see Section 3.2) in 24-well plates. Ensure clones have reached sufficient density prior to transfer (see Note 5). 4. When cells are confluent in the 24-well plates transfer three-fourths of the cells from each well into 96well V-bottom plates for antibody staining. Resuspend the remaining cells in G418 medium for continued growth. 5. Prepare a single well of 1×106 untransfected cells to serve as a negative control. 6. Wash the cells to be stained once with 100 ␮l/well PBS. Centrifuge plates at 450×g for 4 min using low brake. Gently flick the liquid from plates into an appropriate receptacle to remove PBS from the pellet. 7. Add 50 ␮l/well of blocking serum or IgG. Use serum/IgG from the species that will not react with your secondary antibody. For example, if your secondary antibody is an anti-mouse antibody you need to block with rat serum/IgG. Dilute commercial IgG 1:250 in PBS. Serum is used as a 4% solution in PBS. 8. Incubate at room temperature for 15 min then wash two times as described in step 6. 9. Add antibody specific to the transfected molecule. Most antibody preparations are highly concentrated and are diluted in cold PBS as recommended by the manufacturer just prior to use. Prepare enough antibody to allow for 50 ␮l/well. 10. Incubate for 15 min on ice then wash two times as described in step 6. 11. Add fluorochrome-conjugated secondary antibody diluted 1:100 in cold PBS making enough for 50 ␮l/well. Since MHC I−EGFP constructs will fluoresce in the FITC channel regardless of surface expression, a fluorochrome other than FITC must be used to detect surface expression of the transfected molecule. Most flow cytometers are capable of detecting FITC and PE fluorescence making PE the recommended choice. 12. Incubate in the dark at 4ºC for 15 min then wash two times as described in step 6.

306

Lavender and Kane

13. Fix cells in 400 ␮l of a freshly made 2% formaldehyde−PBS solution and transfer to FACS tubes. 14. Proceed to FACS analysis. Compare fluorescence (surface expression) of each transfectant to the negative control. Successful MHC I transfectants will have expression between 1 and 2 (or more) logs higher than the negative control. Successful Ly49 transfectants are lower, around 1 log higher than the negative control (see Note 9). 15. Continue to grow transfectants with good surface expression in selection medium and discard those with no or inadequate expression. 16. Freeze down 10 vials of successfully transfected clones at 4×106 cells in 1 ml of 5% DMSO in FBS. Choose 3–5 MHC I clones with differing levels of surface expression to allow comparisons with other clones of interest and with your positive control. Not all RNK-16 clones will necessarily exhibit similar levels of cytotoxicity despite similar surface expression of the transfected Ly49. Therefore, all successful RNK-16 transfections should be preserved until they can be tested for cytotoxicity. 3.4. Assay of Recognition

1. Thaw RNK-16 effectors and YB2/0 targets. Include untransfected RNK-16 and YB2/0 cells as negative controls. Also include a known Ly49/MHC I recognition pair as a positive control; preferably a known ligand for each Ly49 receptor being used. 2. Grow transfected target and effector cells in G418supplemented growth medium as described in Section 3.2. Change to G418-free growth medium 48 h prior to assay. Exactly 2.4×105 target cells are required for each effector in the assay and 1.5×106 effector cells are required for each target in the assay. Prepare a slight excess. Ideally RNK16 cells will be 90% confluent on the day of the assay (see Note 5). 3. Harvest target cells (see Note 10) and resuspend in 85 ␮l of FBS. 4. In a designated area for radioactive work add Na51 CrO4 following the handling protocols for gamma radiation at your facility (see Notes 11 and 12). Be careful to shield yourself appropriately and to properly contain and dispose of radioactive material. 5. Incubate for 1 h in a 37ºC water bath gently agitating every 15 min. 6. While the targets are incubating prepare your effector cells. Harvest RNK-16 cells by gently washing them off the plate. Wash and resuspend in 5 ml of wash medium.

Identifying Ly49 Ligands

307

7. Count cells and resuspend at 2.5×106 /ml in RNK-16 growth medium. 8. Prepare 1:2 dilutions in RNK-16 growth medium from 25×105 /ml to 1.56×104 /ml. Set aside until targets are ready. RNK-16 growth medium containing 1.5×106 cells/600 ␮l is required for each target cell type in the assay, always prepare slightly more than is required. 9. Once the target cell incubation is complete wash targets four times in 10 ml of wash medium. 10. After the final wash carefully aspirate all remaining wash medium using a pipetman and resuspend target cell pellet in 2 ml of RNK-16 growth medium. 11. Count targets using a hemocytometer designated for radioactive work, being careful to contain and properly dispose of all radioactive materials. 12. Calculate the number of target cells required and resuspend at 1×105 targets/ml in RNK-16 growth medium. Twenty-four wells of 1×104 targets/100 ␮l are required for each effector in the assay, prepare slightly more than that is required. 13. Plate each target into two rows of a V-bottom 96-well plate, aliquoting 100 ␮l/well (1×104 cells/well). 14. Plate each effector cell dilution, as prepared in step 8, into triplicate wells for each target following the final effector:target ratios as shown in the plate layout in Table 20.1. Aliquot 100 ␮l/well of growth medium into the spontaneous (S) and total (T) release wells. 15. Centrifuge plates at 250×g for 3 min with low brake and incubate for 4 h at 37ºC. 16. After incubation, centrifuge plates at 450×g for 5 min using no brake. 17. Harvest 25 ␮l of supernatant from each well into microbeta plates. Total release (T) wells are resuspended prior to harvest. 18. Add 100 ␮l of scintillation fluid to harvested samples.

Table 20.1 Format of effector to target cell ratios of individual wells in a 96-well plate 25:1

25:1

25:1

12:1

12:1

12:1

6:1 6:1 6:1 3:1 3:1 3:1

1.5:1 1.5:1 1.5:1 0.78:1 0.78:1 0.78:1 S

S

S

T

T

T

308

Lavender and Kane

19. Seal plates and shake for 10 min on a plate shaker. 20. Place plates in the microbeta counter and count for 1 min/sample. 21. Calculate % specific lysis using the formula % specific lysis =

observed release−spontaneous release × 100 total release − spontaneous release

22. Reduced or increased specific lysis in a set of wells would indicate that the MHC class I molecule in those target cells effectively engaged with the inhibitory or activating Ly49, respectively, on the effector cells to function as a ligand. Specific lysis equivalent to that of untransfected effector cells would indicate that the MHC class I molecule on the targets was not an effective ligand.

4. Notes 1. We recommend cloning of MHC I into either pCIneo (Promega, Madison, WI) or into pEGFP-N3 (Clontech, Mountain View, CA). The pEGFP-N3 vector allows for the creation of a MHC I molecule that is Cterminally fused to EGFP and facilitates screening of positive transfectants. Mutagenesis of the start codon of EGFP to isoleucine (ATG to ATA) aids the translation of an intact fusion molecule. Cloning Ly49 into BSR␣EN (Dr. A. Shaw, Washington University, St. Louis, MO) is highly recommended for successful transfection of RNK-16 cells. 2. Design primers containing the desired mutation and that anneal to the identical sequence on opposite strands. Ideally primers should a. contain the desired mutation in the middle of the primer with about 10–15 bases of correct sequence on either side, b. have a GC content of ≥40% and terminate at each end in one or more C or G bases, c. have a Tm ≥ 78ºC (Tm = 81.5 + 0.41(%GC) − 675/primer length in base pairs), d. be FPLC or PAGE purified. 3. We suggest aliquoting dNTPs into smaller lots to reduce the number of freeze−thaw cycles. All enzymes (polymerases, restriction enzymes, etc.) should be removed

Identifying Ly49 Ligands

309

from −20◦ C just prior to use and immediately replaced. Do not vortex enzymes. 4. RNK-16 and YB2/0 are devoid of mouse and human molecules but YB2/0 naturally express MHC I of the rat LOU strain and RNK-16 express NK receptors of the F344 rat strain. This must be considered when choosing receptor/ligand combinations for testing. 5. RNK-16 cells are very sensitive to cell density. When passaging, plate at 3×105 cells/ml and passage every 2–3 days. Over-dilution can result in cell death. Conversely, allowing RNK-16 to become too dense can reduce both viability and cytotoxic function. 6. Actively growing log-phase cells that have not been passaged for more than a week transfect most efficiently. Do not use confluent cells. 7. Transfections are more successful with Endo-Free DNA, particularly for RNK-16 cells. Inhibitory Ly49 are typically easier to transfect than activating receptors. We usually set up —two to three transfections for each inhibitory Ly49 and —five to six for each activating Ly49. A single MHC I transfection using YB2/0 cells is sufficient. 8. Wells containing growing YB2/0 clones often contain a single large clump of cells. It is important to resuspend these clumps the day before screening to redistribute the cells on the well floor and maximize the fluorescence intensity across the well. All wells will appear positive if the culture medium is not sufficiently removed. 9. Assessing whether surface expression of ligand and receptor is adequate for recognition is an important factor to consider. Matching expression levels of potential ligands on YB2/0 with those of known ligands on YB2/0 cells is one possibility. This cannot be done when assessing Ly49 expression levels on RNK-16 cells. A better technique for assessing the triggering capacity of a MHC I or Ly49 transfectant is to use ADCC or reverse ADCC, respectively. This is particularly important for RNK-16 transfectants, which can vary widely in their level of cytotoxicity independent of receptor expression. 10. Do not use trypsin for adherent cells as this will remove the surface ligand. Use cells scrapers or Versene. 11. Most institutions have guidelines and regulations that must be met before investigators can be licensed for radionuclide use. Obtain information from your institution on proper shielding, waste management, and regular monitoring for contamination.

310

Lavender and Kane

12. The half-life of 51 Cr is 27 days. For a 1 mCi/ml stock less than 2 weeks old 100 ␮l is sufficient for 5–25×106 target cells, if more than 2 weeks old 150 ␮l is required. Using 51 Cr that is greater than 1 month old is not recommended.

Acknowledgments This work was supported by Canadian Institutes of Health Research grant 43864. References 1. Nylenna, O., Naper, C., Vaage, J., Woon, P., Gauguier, D., Dissen, E., Ryan, J., and Fossum, S. (2005) The genes and gene organization of the Ly49 region of the rat natural killer cell gene complex. Eur. J. Immunol. 35, 261–272. 2. Proteau, M., Rousselle, E., and Makrigiannis, A. (2004) Mapping of the BALB/c Ly49 cluster defines a minimal natural killer cell receptor gene repertoire. Genomics 84, 669–677. 3. Wilhelm, B., Gagnier, L., and Mager, D. (2002) Sequence analysis of the ly49 cluster in C57BL/6 mice: a rapidly evolving multigene family in the immune system. Genomics 80, 646–661. 4. Carlyle, J., Mesci, A., Fine, J., Chen, P., B´elanger, S., Tai, L., and Makrigiannis, A. (2008) Evolution of the Ly49 and Nkrp1 recognition systems. Semin. Immunol. 20, 321–330.. 5. Lian, R., Y. Li, S. Kubota, Mager, D., and Takei, F. 1999. Recognition of class I MHC by NK receptor Ly-49C: identification of critical residues. J. Immunol. 162, 7271–7276. 6. Silver, E., Gong. D., Chang, C., Amrani, A., Santamaria, P., and Kane, K. (2000) Ly-49P activates NK-mediated lysis by recognizing H-2Dd. J. Immunol. 165, 1771–1781. 7. Karlhofer, F., Ribaudo, R., and Yokoyama, W. (1992) MHC class I alloantigen specificity of Ly-49+ IL-2-activated natural killer cells. Nature 358, 66–70. 8. Silver, E., Gong, D., Hazes, B., and Kane, K. (2001) Ly-49 W, an activating receptor of nonobese diabetic mice with close homology to the inhibitory receptor Ly-49G, recognizes H-2D(k) and H-2D(d). J. Immunol. 166, 2333–2341. 9. Nakamura, M., Linnemeyer, P., Niemi, E., Mason, L., Ortaldo, J., Ryan, J., and Sea-

10.

11.

12.

13.

14.

15.

man, W. (1999) Mouse Ly-49D recognizes H-2Dd and activates natural killer cell cytotoxicity. J. Exp. Med. 189, 493–500. Silver, E., Lavender, K., Gong, D., Hazes, B., and Kane, K. (2002) Allelic variation in the ectodomain of the inhibitory Ly-49G2 receptor alters its specificity for allogeneic and xenogeneic ligands. J. Immunol. 169, 4752–4760. Pende, D., Biassoni, R., Cantoni, C., Verdiani, S., Falco, M., di Donato, C., Accame, L., Bottino, C., Moretta, A., and Moretta, L. (1996) The natural killer cell receptor specific for HLA-A allotypes: a novel member of the p58/p70 family of inhibitory receptors that is characterized by three immunoglobulinlike domains and is expressed as a 140-kD disulphide-linked dimer. J. Exp. Med. 184, 505–518. Vitale, M., Sivori, S., Pende, D., Augugliaro, R., Di Donato, C., Amoroso, A., Malnati, M., Bottino, C., Moretta, L., and Moretta, A. (1996) Physical and functional independency of p70 and p58 natural killer (NK) cell receptors for HLA class I: their role in the definition of different groups of alloreactive NK cell clones. Proc. Natl. Acad. Sci. U S A 93, 1453–1457. Litwin, V., Gumperz, J., Parham, P., Phillips, J., and Lanier, L. (1994) NKB1: a natural killer cell receptor involved in the recognition of polymorphic HLA-B molecules. J. Exp. Med. 180, 537–543. D¨ohring, C., Scheidegger, D., Samaridis, J., Cella, M., and Colonna, M. (1996) A human killer inhibitory receptor specific for HLAA1, 2. J. Immunol. 156, 3098–3101. Moretta, A., Vitale, M., Bottino, C., Orengo, A., Morelli, L., Augugliaro R., Barbaresi M., Ciccone E., and Moretta, L. (1993) P58 molecules as putative receptors for major histocompatibility complex (MHC) class I

Identifying Ly49 Ligands molecules in human natural killer (NK) cells. Anti-p58 antibodies reconstitute lysis of MHC class I-protected cells in NK clones displaying different specificities. J. Exp. Med. 178, 597–604. 16. Arase, H., Mocarski E., Campbell A., Hill A., and Lanier, L. (2002) Direct recogni-

311

tion of cytomegalovirus by activating and inhibitory NK cell receptors. Science 296, 1323–1326. 17. Ryan, J., Niemi E., and Nakamura, M. (2000) Functional analysis of natural killer cell receptors in the RNK-16 rat leukemic cell line. Methods Mol. Biol. 121, 283–295.

Chapter 21 Probing the Interactions of NK Cell Receptors with Ligand Expressed in trans and cis ´ Jonathan Back, Leonardo Scarpellino, and Werner Held Abstract Certain receptors on natural killer (NK) cells, which are specific for MHC class I (MHC-I) molecules, do not only interact with ligand expressed on opposing cell membranes (in trans) but also interact with those on the same cell membrane (in cis). Cis interactions have been demonstrated for only a small number of cell surface receptors. However, this has not been tested systematically, raising the possibility that additional receptors may be able to bind ligand expressed in cis. Here we describe a number of approaches to evaluate trans and cis binding of the Ly49A NK cell receptor to its H-2Dd ligand. These procedures should facilitate the investigation of cis/trans interactions of other receptor–ligand pairs and simplify the analysis of NK cell receptor variants. Key words: Ly49, MHC class I, cis interaction, receptor masking, tetramer, cellular adhesion, ligand transfer.

1. Introduction NK cells use activating and inhibitory receptors to scan host cells for altered expression of various self-ligands such as MHC-I molecules. MHC-I receptors include the C-type lectin-like Ly49 receptor family in mice and the killer cell immunoglobulin-like receptor (KIR) family in humans (1). Many of these receptors are inhibitory and they ensure that NK cells do not attack host cells expressing MHC-I at a normal surface density while attacking those with low levels of MHC-I. Ly49A represents the prototype MHC-I receptor on mouse NK cells. This receptor binds to H-2Dd (Dd ) or Dk but not to K.S. Campbell (ed.), Natural Killer Cell Protocols, Methods in Molecular Biology 612, DOI 10.1007/978-1-60761-362-6 21, © Humana Press, a part of Springer Science+Business Media, LLC 2010

313

314

Back, Scarpellino, and Held

Db expressed on potential target cells to inhibit NK cell-mediated effector functions (2). Besides binding ligand expressed on other cells (in trans), Ly49A can also bind ligand expressed in the plane of the NK cells’ membrane (in cis) (3). Cis binding reduces the number of Ly49A receptors that can functionally interact with Dd expressed on potential target cells (4). Hence, masking of Ly49A lowers the threshold at which NK cell activation exceeds inhibition. This renders Ly49A NK cells more useful to detect diseased host cells. Besides Ly49A, several other Ly49 family receptors can bind MHC-I ligand expressed in cis (5). Other examples include the structurally distinct, immunoglobulin-like, inhibitory human LILRB2 and the homologous mouse PIR-B receptors that also interact with MHC-I ligand expressed in trans as well as in cis (6) or sialic acid-binding Ig-like lectins (Siglecs), which associate with sialic acid modifications of glycoproteins expressed in the same cell membrane (reviewed in (7)). We have developed a number of tests to determine to what extent the Ly49A receptor is masked by Dd ligand expressed in the plane of the same membrane. Here, we detail the use of soluble ligand to determine Ly49A receptor accessibility prior to and following the destruction of tri-molecular MHC-I ligand complexes, which unmasks Ly49A (see Note 1). In addition to soluble ligand, we describe two approaches to determine whether ligand expression in cis affects ligand binding to membrane-associated trans ligand. Collectively, these procedures should be useful to rapidly determine whether receptor variants retain their ability to bind ligand in trans and in cis. Moreover, they may be adapted to investigate cis/trans interactions of other receptor/ligand pairs.

2. Materials 1. Cells: C1498 (ATCC #TIB-49) (H-2b ) is an immature NK T cell line (8). C1498 cells can be efficiently transfected using electroporation or infected with lentiviruses. Stable cell lines are rapidly established using multiple rounds of magnetic cell sorting (MACS, Miltenyi Biotec, Gladbach, Germany), or using expression vectors, which include a puromycin resistance gene (select using 10 ␮g/mL of puromycin). 2. Stable C1498 transfectants used here express Ly49A, Dd ; Ly49A together with Dd ; or a Dd -EYFP fusion protein (3, 4).

NK Cell Receptors Binding in trans and in cis

315

3. Cell culture medium: RPMI 1640 containing glutamax (Invitrogen, Carlsbad, CA, USA) supplemented with HEPES (10 mM), 2-mercaptoethanol (5×10−5 M), penicillin (50 ␮g/mL), streptomycin (50 ␮g/mL) (all from Invitrogen) and 10% foetal calf serum (FCS). 4. Plasticware: 96-well V-bottom plates (such as Costar); 96-well U-bottom plates (such as Costar); FACS tubes (5 mL round bottom polystyrene tubes, BD Biosciences, San Jose, CA, USA) and 1.4 mL U-bottom tubes (such as #M32022 plus rack from Micronic, Lelystad, The Netherlands); 15 mL V-bottom polypropylene tubes and 25 cm2 cell culture flasks. 5. Tetramers and monoclonal antibodies (mAbs): Dd tetramers were refolded using mouse ␤2 m and the HIV-derived peptide RGPGRAFVTI using standard techniques (9) (see Note 2) and labelled with phycoerythrin (PE); anti-Ly49A (JR9-318) either purified or labelled with fluorescein isothiocyanate (FITC) or biotin; anti-Dd (34-2-12), either PE or FITC labelled; anti-␤2m (19.11) biotinylated (all available from BD Biosciences); streptavidin−PE (Invitrogen Molecular Probes) to detect biotinylated mAbs. 6. FACS buffer: Phosphate-buffered saline (PBS) supplemented with 2% FCS. 7. Flow cytometer (such as FACScan BD Biosciences) together with analysis software (such as CellQuest (BD Biosciences) or FlowJo (Tree Star, Ashland, OR, USA)). 8. Stripping buffer: Citric acid 0.133 M, di-sodium hydrogen phosphate dihydrate (Na2 HPO4 (2H2 O)) 0.066 M. The pH is adjusted to 3.3 before sterile filtration (0.22 ␮m, Millipore, Billerica, MA, USA). The solution is stable for 2–3 months at room temperature; however the pH should be checked before use. Aliquots can be stored at −20◦ C. 9. Neutralization buffer: PBS supplemented with 1% bovine serum albumin (BSA), fraction V (Sigma-Aldrich, Munchen, Germany) and 10 mM HEPES. 10. Fixing solution: Paraformaldehyde (PFA) (1%) in PBS, made fresh. Alternatively, aliquots of 4% PFA in PBS can be stored at −20◦ C. 11. Cell labelling reagents: 5-(and-6)-(((4-chloromethyl) benzoyl) amino) tetramethylrhodamine)-mixed isomers (CMTMR) (Invitrogen Molecular Probes); carboxyfluorescein diacetate succinimidyl ester (CFSE) (SigmaAldrich).

316

Back, Scarpellino, and Held

3. Methods 3.1. Determining Ly49A Receptor Masking Using Soluble Dd Ligand

This procedure allows an estimation of receptor masking by comparing Dd tetramer binding to Ly49A in the presence or absence of Dd ligand expressed in cis. In addition, Dd tetramer binding is assessed prior to and following Ly49A unmasking. For MHC-I ligands, this is achieved by exposing cells to an acidic buffer, which disrupts the tri-molecular MHC-I complexes (see Note 1). 1. Harvest C1498 cells. Half the cells will be treated with acidic buffer, the rest are kept for control staining in cold FACS buffer. 2. Centrifuge 3–10 × 106 cells in a 15 mL polypropylene tube for 5 min at 500×g. Wash the cells twice with 10 mL of PBS, completely remove the remaining PBS. 3. Resuspend the cells in 500 ␮L of stripping buffer, leave for 4 min (not longer!) at room temperature. 4. Add 14 mL of neutralization buffer and immediately centrifuge at 500×g for 5 min. 5. Completely remove the remaining liquid and resuspend the cells in FACS buffer, count cells and determine viability (which should not be affected by acid treatment). Wash once more and resuspend the cells at 5 × 106 cells/mL in FACS buffer. 6. Distribute 5 × 105 cells (stripped or non-stripped) per well into V-bottom 96-well plates. After centrifugation (3 min at 500×g), the cells are stained for flow cytometry in a final volume of 50 ␮L for 30 min on ice. We routinely use 10–20 ␮g/mL of PE-labelled Dd tetramer, but the optimal tetramer concentration has to be determined for each batch. Independent samples are stained with anti-Ly49A mAb (to determine the cell surface density of Ly49A) and with anti-␤2m mAb (to determine whether ␤2m was quantitatively removed by acid treatment) (Fig. 21.1). Background staining is determined using untransfected C1498 cells. Non-stripped, transfected cells are used as positive controls. 7. Add 200 ␮L of FACS buffer, centrifuge 3 min at 500×g, remove the liquid by flicking the plate. 8. Resuspend the cells in 400 ␮L of cold FACS buffer, transfer to 5 mL FACS tubes and run samples on a flow cytometer. 9. Data analysis and quantification: All flow cytometry analysis is done on live cells (identified based on their forward and

NK Cell Receptors Binding in trans and in cis

317

Fig. 21.1. Ly49A accessibility probed with soluble Dd ligand. Histograms show nontransfected (-) C1498 cells or cells transfected with Ly49A or co-transfected with Ly49A and Dd , stained for Ly49A (mAb JR9), ␤2m (mAb 19.11) and with Dd tetramers. Open histograms depict cells that had been treated with an acidic buffer and grey-filled histograms show cells that were left untreated. Numbers depict the mean fluorescence intensity (MFI) of staining with the indicated reagent.

side scatters (FSC/SSC)). Determine the mean fluorescence intensity (MFI) of staining using ␤2m mAb, Ly49A mAb and Dd tetramer. a. Determine the “relative tetramer binding” by calculating the ratio of MFI (tet)/MFI (mAb) for Ly49A transfectants that lack and that express Dd ligand (Fig. 21.1) (see Note 3). b. Determine “receptor masking” by calculating the ratio of MFI (tet, strip)/MFI (tet, non-strip) for Ly49A transfectants that lack and that express Dd ligand (Fig. 21.1) (see Note 4). 3.2. Determining Ly49A Receptor Accessibility Using MembraneAssociated Dd Ligand

While soluble Dd ligand is useful to determine receptor masking, the use of multimeric ligands is not likely to completely reflect receptor binding to monomeric ligands that are expressed on a target cell membrane. This is of importance when analysing receptor variants in which either cis and/or trans binding may be impaired. The assays shown below are used to assess the ability of Ly49A to interact with membrane-bound ligand and to see to what extent Dd expression in cis impacts this interaction. In Section 3.2.1 we describe a FACS-based cell−cell adhesion assay

318

Back, Scarpellino, and Held

(based on (10)) and in Section 3.2.2 we exploit Dd ligand capture from neighbouring cells (11) to assess the capacity of Ly49A to interact with membrane-bound ligand. 3.2.1. Cell Adhesion Assay

1. Prepare a fresh solution of 1 ␮M CMTMR in PBS and of 0.1 ␮M CFSE in PBS. Keep the solutions protected from light. 2. Harvest and count C1498 transfectants. Take 5–10 × 106 cells per labelling and wash once with PBS. 3. Discard the PBS and resuspend in 1 ␮M CMTMR solution (107 cells/mL) or in 0.1 ␮M of CFSE solution (107 cells/mL), incubate for 10 min at 37◦ C. 4. Add 14 mL of cold cell culture medium, centrifuge 5 min at 500×g. 5. Wash the cells two more times and resuspend at 0.5 × 106 cells/mL in cold cell culture medium. 6. Mix 2.5 × 104 (50 ␮L) of CMTMR-labelled cells with 2.5 × 104 (50 ␮L) of CFSE-labelled cells in a U-bottom 96-well plate. 7. Centrifuge the plate for 1 min at 75×g and incubate for 10 min at 37◦ C. Do not resuspend the pellet! 8. Transfer the plate on ice and add directly 100 ␮L/well of 1% PFA in PBS. Gently resuspend the cells by pipetting. In order not to disrupt conjugates use tips where the ends have been cut off. 9. Transfer the resuspended cells to 1.4 mL collection tubes, which contain 200 ␮L of 1% PFA fixation solution. 10. Run the cells immediately on a flow cytometer. 11. Data analysis: Ly49A-expressing cells (CMTMR+) are detected in FL2, Dd cells (CFSE+) are detected in FL1. Conjugates are double positive for CMTMR and CFSE (Fig. 22.2). Determine the percentage of Ly49A cells (CMTMR+) conjugated with CFSE+ cells as a percentage of all CMTMR+ cells (see Note 5).

3.2.2. Ligand Capture Assay

Upon interaction, Ly49A cells acquire Dd ligand from neighbouring cells. In fact acquisition is bi-directional as Dd cells also acquire Ly49A receptor from neighbouring cells (Fig. 21.3A). Ligand transfer can easily be quantified using C1498 cells expressing a Dd -EYFP fusion protein (Dd -EYFP cells) and measuring green (EYFP) fluorescence of Ly49A+ cells. 1. Harvest and count transfectants, include C1498 cells expressing the Dd -EYFP fusion protein.

NK Cell Receptors Binding in trans and in cis

319

Fig. 21.2. Ly49A-mediated cell−cell adhesion. Formation of conjugates between Ly49A-expressing and Dd -expressing C1498 transfectants. C1498 transfectants were labelled with CMTMR or CFSE, mixed and incubated for 10 min at 37◦ C before fixation. Numbers in the density plots indicate the percentage of CMTMR+ cells conjugated with CFSE+ cells as a percentage of all CMTMR+ cells.

2. Mix 105 Ly49A cells with 105 Dd -EYFP cells in a final volume of 8 mL of cell culture medium in a 25 cm2 culture flask. Independently set up 2 × 105 Ly49A, Ly49A Dd cells or Dd EYFP cells. As a specificity control for ligand transfer set up a co-culture of non-transfected C1498 cells with Dd -EYFP cells. 3. Culture cell mixtures in a CO2 incubator (37◦ C, 5% CO2 ). 4. After 48 h harvest the co-cultures (they must be confluent), count, wash and resuspend cells at 5 × 106 cells/mL in FACS buffer. The co-culture time may be reduced. However, the initial cell density must be increased accordingly to reach confluence more rapidly. 5. Distribute 5 × 105 cells per well in V-bottom 96 well plates. After centrifugation (3 min, 500×g), flick the plate and stain the cells with anti-Ly49A−biotin in a final volume of 50 ␮L of FACS buffer for 30 min on ice. 6. Samples are washed once in FACS buffer, and further stained with streptavidin−PE in 50 ␮L of FACS buffer for 20 min on ice. 7. After washing, the cells are resuspended in a total volume of 400 ␮L of cold FACS buffer, transferred to FACS tubes and run on a flow cytometer. 8. Determine the mean fluorescence intensity (MFI) of (Dd -) EYFP (FL1) on gated Ly49A cells (FL2) (see Note 6).

320

Back, Scarpellino, and Held

Fig. 21.3. Ly49A-mediated ligand uptake from environmental cells. (A) C1498 transfectants (acceptor cells) were co-cultured with transfectants expressing a Dd -EYFP fusion protein (donor cells) (lower panel) or were cultured alone (upper panel). The cell mixture was stained for Ly49A and analysed by flow cytometry. Density plots show that Ly49Aexpressing cells (rectangular gate) acquired green fluorescence (Dd -EYFP). Dd -EYFP transfer was not observed when acceptor cells lacked Ly49A and transfer was strongly reduced when Ly49A cells co-expressed Dd . Numbers indicate the mean FL1 (EYFP) fluorescence intensity (MFI) in the indicated gate. The net Dd -EYFP transfer (MFI) was calculated by subtracting the FL1 MFI of acceptor cells cultured alone from that of acceptor cells co-cultured with Dd -EYFP-expressing cells. (B) Confocal microscopy pictures of Dd -EYFP fluorescence, Ly49A staining and transmitted light (maximum intensity projections of series of Z-stack confocal acquisitions) show a Ly49A transfectant conjugated with a Dd -EYFP transfectant. Arrows highlight Dd -EYFP acquired by the Ly49A cell.

4. Notes 1. Acid stripping disrupts tri-molecular MHC-I complexes. Whereas bound peptide and ␤2m are removed, the MHC-I heavy chain remains membrane bound. This procedure was initially developed to extract antigenic peptides bound to MHC molecules from living cells (12). In the case of Siglecs, sialic acid cis ligands can be removed enzymatically using sialidase treatment (13). 2. Binding of several Ly49 receptors to MHC-I is influenced by species-specific residues in ␤2m (5, 14). It is

NK Cell Receptors Binding in trans and in cis

321

therefore important to use mouse ␤2m to refold MHC-I tetramers. While Ly49A binding to Dd is not influenced by the peptide, Ly49I binding to Kd is strongly influenced by bound peptide (5, 15). For commercially available MHC-I tetramers see Beckman Coulter (Fullerton, CA, USA), ProImmune (Oxford, UK), and http://www.niaid. nih.gov/reposit/TETRAMER/overview.html 3. “Relative tetramer binding” takes into account a possible difference in the Ly49A cell surface density between transfectants. A reduction of the MFI (tet)/MFI (mAb) ratio in the presence of ligand indicates that ligand expression masks the receptor (Fig. 21.1). In the example shown (Fig. 21.1), the ratio of MFI (tet) to MFI (mAb) for cells expressing Ly49A in the absence of ligand is 3.5, while it is 1.4 for cells co-expressing Ly49A and Dd . Tetramer binding is thus specifically reduced (2.5-fold) in the presence of Dd ligand, indicating that approximately 60% of Ly49A receptors are masked. The accurate determination of MFI (tet)/MFI (mAb) in the absence and presence of ligand is based on the assumption that antibody binding is not influenced by the expression of ligand in cis. Ly49A staining with mAb JR9-318 (or YEI-48) is actually slightly reduced (1.2- to 1.5-fold) in the presence of Dd . While true receptor downmodulation cannot be completely excluded (16) we believe that this effect is predominantly based on receptor masking. Indeed, mAb staining improves upon acid stripping (Fig. 21.1). This method may thus somewhat underestimate receptor masking. The binding of some mAbs, such as A1, is much more sensitive to the presence of Dd as compared to mAb JR9318. This is almost completely due to the masking of the A1 epitope by Dd expression in cis (unpublished data). Thus what may appear as a reduction of Ly49 cell surface density is actually due to receptor masking and the extent of the effect is dependent on which mAb is used for detection. As another example, the binding of mAb 5E6 to Ly49C is strongly reduced by the presence of Kb in cis, while the binding of mAb 4D12 is much less affected. 4. Acid stripping is controlled by a reduction of staining for ␤2m: as shown in Fig. 21.1, the MFI of ␤2m staining is 66 prior to stripping and 4 after stripping. Next, the MFI of tetramer staining of acid−stripped cells (MFI (tet, strip)) is compared to that of non-stripped cells (MFI (tet, no strip)) (Fig. 21.1). In our example, the ratio of MFI (tet, strip)/MFI (tet, no strip) for Ly49A cells is 0.9. The corresponding ratio for Ly49A Dd cells is 3.0 (Fig. 21.1),

322

Back, Scarpellino, and Held

indicating that the access to the receptor has improved due to acid stripping. The extent of improvement suggests that 66% of Ly49A receptors were masked by Dd . It is important to determine the ratio between MFI (tet, strip)/MFI (tet, no strip) in the absence of known ligand. In this case, a ratio of >1.0 indicates that Ly49A may be masked by an additional ligand. Conversely, a ratio significantly 16 h) at 42◦ C. After hybridization, 200 ␮L of RNase digestion buffer containing a 1:100 dilution of RNase A/RNase T1 enzyme is added to each tube. Digestion buffer without enzyme is added to one of the two tubes containing probe and yeast RNA as a control for probe integrity. Digestion is performed at 37◦ C for 30 min. The reactions are then precipitated with RNase inactivation/precipitation solution and resuspended in gel loading buffer. 6. The protected RNA products are separated on an 8% denaturing polyacrylamide gel alongside a labeled RNA marker (DecadeTM Marker System, Ambion). The gel is dried and exposed for ∼24 h in a PhosphorImager cassette (Molecular Dynamics). The image is visualized using PhosphorImager SI analysis and ImageQuaNT (Molecular Dynamics). 3.3.2. Identification of Transcription Start Sites by 5 -RACE

1. 5 -RACE is performed with the FirstChoice RLM-RACE Kit (Ambion). Ten micrograms of total RNA is dephosphorylated with 2 ␮L calf Intestine Alkaline Phosphatase in 1× CIP buffer for a total volume of 20 ␮L.

Identification and Analysis of Novel Transcripts and Promoters

385

2. After 1 h incubation at 37◦ C, the volume is adjusted to 150 ␮L with 115 ␮L of water and 15 ␮L of ammonium acetate solution. The RNA solution is extracted once with acid phenol/chloroform, and the upper aqueous phase is extracted a second time with 150 ␮L of chloroform. 3. The RNA is precipitated by the addition of 150 ␮L of isopropyl alcohol, chilled on ice for 10 min, and then centrifuged for 20 min at top speed in a refrigerated microcentrifuge. The pellet is carefully washed with 70% ethanol and air dried. 4. The RNA is resuspended in 7 ␮L of water and the decapping reaction is performed with tobacco acid pyrophosphatase (TAP) in a total volume of 10 ␮L containing 1 ␮L of 10× TAP buffer and 2 ␮L of TAP enzyme. The reaction is incubated at 37◦ C for 1 h and then 2 ␮L of the reaction is used immediately for ligation of a 5 RNA linker. 5. The linker ligation reaction contains 2 ␮L processed RNA, 1 ␮L RNA adaptor, 1 ␮L 10× RNA ligase buffer, 2 ␮L T4 RNA ligase, and 4 ␮L water. The reaction is incubated at 37◦ C for 1 h. 6. Reverse transcription is performed with either random decamers or a gene-specific primer. cDNA generated with random decamers can be used to map start sites of several transcripts; however, the increased sensitivity of specific priming is necessary when mapping the start site of rare transcripts. 7. The cDNA is then amplified with a gene-specific 3 primer together with a 5 primer corresponding to the linker added to the 5 end of the RNA. 8. The PCR products are cloned using the StrataClone PCR Cloning System, plated on ampicillin plates, and at least 24 individual clones are selected for sequencing. The sequence of the cDNA clones is compared to the genomic sequence, and the junction between the 5 linker and the homology to KIR genomic sequence indicates the start site of transcription. 3.4. Generation of Luciferase Reporter Plasmids

1. Mononuclear cells are obtained by Ficoll-Hypaque (Sigma Diagnostics) density-gradient centrifugation of peripheral blood from healthy donors. Genomic DNA is isolated using QIAamp DNA Blood Mini Kit (Qiagen). 2. A fragment of the KIR3DL1 core bi-directional promoter (see Note 8) is amplified from human genomic DNA using a gene-specific forward primer starting at −229 (ATAGTGAAGGACGCGAGGTGTC) and a reverse primer

386

Li, Wright, and Anderson

starting at −1 relative to the start codon of the gene (GGTGCTGCCGGTGCAGAC). 3. The PCR reactions are carried out in a 50 ␮L final volume containing 100 ng of genomic DNA and 20 pmol of each R PCR SuperMix (Invitrogen). The primer using Platinum thermal cycling conditions were 35 cycles of 94◦ C for 20 s, 59◦ C for 30 s, and 73◦ C for 5 s. 4. PCR products are cloned into the Invitrogen TOPO TA vector when promoter fragments less than 500 bp are studied (see Note 4). Clones are isolated and sequenced as described in Sections 3.1.3–3.1.5. 5. Bacterial clones containing the desired promoter fragment are grown in 70 mL of LB with 50 ␮g/mL ampicillin shaking at 37◦ C overnight, and plasmid DNA is isolated using a Midi Plasmid Kit (Qiagen). 6. The inserts are excised with either SacI/XhoI to produce one orientation or XhoI/HindIII for the opposite orientation (see Note 9). Forty micrograms of plasmid DNA is digested for 1 h at 37◦ C in a total volume of 100 ␮L containing 80 units of enzyme (4 ␮L of 20 units/␮L enzyme) and 1 ␮g/mL of BSA in reaction buffer compatible with both enzymes used. NEB buffer 1 is used for SacI/XhoI digestion and NEB buffer 2 is used for XhoI/HindIII digestion (all reagents are provided with the enzyme from New England Biolabs). 7. The digested DNA is separated by electrophoresis in a 1.4% agarose gel in 0.5× TBE buffer containing 1× GelRed, nucleic acid gel stain, using a well comb large enough to accommodate the entire 100 ␮L of digested DNA. A 30-mm-wide well is adequate when using a 1-mm-thick gel comb if the agarose gel is poured to a depth of at least 5 mm. 8. The DNA fragments corresponding to KIR3DL1 promoter are visualized under low-intensity UV light and cut from the gel with a clean razor blade, and purified using the QIAquick Gel Extraction Kit (Qiagen). 9. The purified inserts are cloned into the pGL3-basic vector (Promega) digested with the appropriate enzymes to generate constructs in both forward and reverse orientations. A typical ligation reaction for SacI/XhoI cloning contains 2 ␮L of 5× T4 DNA ligase buffer (Invitrogen), 2 ␮L of 50 ng/␮L SacI/XhoI insert, 1 ␮L of 50 ng/␮L SacI/XhoIdigested pGL3-basic vector, 1 ␮L of 20 units/␮L T4 DNA ligase, and 4 ␮L of ddH2 O. The reaction is incubated for 15 min at room temperature.

Identification and Analysis of Novel Transcripts and Promoters

387

10. Chemically competent Escherichia coli cells are transformed with 3 ␮L of the pGL3/insert ligation. Twenty microliters of bacteria are plated on LB–ampicillin and grown overnight. Individual colonies are picked and grown in 70 mL of LB with 50 ␮g/mL of ampicillin shaking at 37◦ C overnight, and plasmid DNA is isolated using a Midi Plasmid Kit (Qiagen). All constructs are verified by sequencing with the pGL3-specific RV3 primer (Promega). 3.5. Cell Transfection, Electroporation, and Luciferase Assays 3.5.1. HEK293 Cell Transfection

1. HEK293 cells are plated at 0.5—1.0 ×105 cells/well in a sixwell plate on the day before transfection (see Notes 10–13). Incubate the cells at 37◦ C in a CO2 incubator overnight. The cell density will reach 3–4 ×105 cells/well when the cells are harvested. 2. For each well, mix 1 ␮g of an individual reporter construct plus 0.01 ␮g of the Renilla luciferase pRL-SV40 control DNA in a 1.5-mL microcentrifuge tube. 3. For each well, add 5 ␮L of Fugene (Roche Diagnostics) to 95 ␮L of DMEM medium without serum. Wait for 5 min before using this mixture. 4. Add the Fugene mixture dropwise to the tube containing DNA. 5. Wait for 15 min and then add directly to cells in a six-well plate. 6. Harvest cells at the desired time after transfection (usually 48 h) and assay luciferase expression using the DualLuciferase Reporter Assay System (Promega).

3.5.2. YT-Indy Cell Electroporation

1. Culture YT-Indy cells to reach a cell density of 5–10 ×105 cells/mL of culture at the time of transfection (see Notes 10–13). 2. Centrifuge the cells in a conical centrifuge tube at 400×g for 5 min at 4◦ C. 3. Resuspend the cell pellet in 10 mL of ice cold PBS, re-pellet. 4. Remove the supernatant by aspiration and resuspend the cell pellet at a concentration of 1 × 107 cells/mL in RPMI medium without added serum, and place on ice. 5. Add 10 ␮L of 1 ␮g/␮L pGL3 construct plus 10 ␮L of freshly prepared 10 ng/␮L Renilla luciferase pRL-SV40 vector to a 1.5-mL microcentrifuge tube and place on ice. 6. Add 0.5 mL of YT-Indy cells to each tube, mix by pipetting. Transfer the mixture into an electroporation cuvette (BioRad) and place on ice.

388

Li, Wright, and Anderson

7. Transfect YT-Indy cells using electroporation with a BTX ECM 830 (Harvard Apparatus) set at 250V, with three pulses of 7 ms at an interval of 100 ms. 8. Remove cells from the cuvette with a sterile pipette and place in a six-well culture plate with 5 mL of culture medium. 9. Incubate at 37◦ C in a CO2 incubator for 8–48 h (48 h gives optimal activity). Luciferase activity is assayed using the Dual-Luciferase Reporter Assay System (Promega). 3.5.3. Preparation of Cell Lysates for Luciferase Assay

1. After 48 h of transfection, transfer cells to a 15-mL conical centrifuge tube and centrifuge the cells at 400×g for 5 min at 4◦ C. 2. Remove the medium from cells and wash once with 10 mL of cold PBS. 3. Use a micropipette to remove any residual PBS from the cell pellet. 4. Dilute 5× passive lysis buffer (Promega) 1:5 using dH2 O to generate 1× passive lysis buffer. 5. Add 500 ␮l of 1× passive lysis buffer and vortex samples. 6. Transfer the mixture to a 1.5-mL microcentrifuge tube, place at room temperature for 15 min. Centrifuge for 30 s to pellet debris.

3.5.4. Luciferase Activity Assay

1. Prepare the Luciferase Assay Reagent II (LAR II) by adding 10 mL of luciferase assay buffer II to a bottle of lyophilized luciferase assay substrate (Promega). Vortex gently for 10 s and put on ice. The LAR II can be stored at −20◦ C for 1 month. 2. Prepare Stop & Glo reagent by adding 20 ␮L of 50× Stop & Glo substrate to 980 ␮L of Stop & Glo buffer (Promega), vortex gently, and put on ice. This reagent must be freshly prepared. 3. Add 100 ␮L of the LAR II reagent to a luminometer tube. 4. Add 20 ␮L of cell lysate, pipetting up and down three to five times to mix, and measure luciferase activity on the luminometer. 5. Take tube out of the luminometer and add 100 ␮L of Stop & Glo solution, vortex 2–3 s. Put the tube back into the luminometer and measure again. Try to be consistent with the timing of this step for each sample. 6. The luciferase activity of the promoter constructs is normalized relative to the activity of the Renilla luciferase produced by the pRL-SV40 control vector. The Renilla activity of the pGL3-basic negative control is divided by the Renilla activity

Identification and Analysis of Novel Transcripts and Promoters

389

of each construct and the firefly luciferase activity is multiplied by this ratio to normalize it relative to the pGL3-basic control. 7. The fold promoter activity relative to empty vector (pGL3basic) is calculated as the ratio of the corrected firefly luciferase activity of each construct relative to the pGL3 activity. 8. Each construct should be tested in at least three independent experiments for both forward and reverse orientations. In all experiments, values are given as mean ±SD (n≥3).

4. Notes 1. The PCR SuperMix is very convenient for rapidly performing PCR reactions under defined conditions. We have found, however, that some lots of SuperMix have low Taq activity; therefore we routinely add 3 ␮L of purified Platinum Taq enzyme (Invitrogen) to each tube of SuperMix when it is thawed and subsequently store it at 4◦ C to avoid repeated cycles of freeze-thawing. 2. When choosing primers for PCR of potential intergenic transcripts, it is important to avoid making primers that span possible RNA splice sites. Therefore, primers should not span CT-rich regions that contain a potential AG splice acceptor or regions that correspond to splice donors (GTNAGT). 3. All PCR primers should be tested for specificity with the NCBI BLAST program (http://blast.ncbi.nlm. nih.gov/Blast.cgi) before ordering. 4. PCR products less than 500 bp are cloned into the Invitrogen TOPO TA vector, whereas the Stratagene system is used for cloning larger fragments. The Stratagene system is designed to accommodate larger fragments, whereas the Invitrogen system is designed such that the open ends of the vector are held together, thereby increasing the efficiency of cloning small fragments but decreasing the cloning of large fragments. 5. The anchor provides an efficient primer binding site at the 3 end of the message and avoids the continuous elongation of the PCR product that occurs when an oligo-dT primer is used to amplify cDNA. 6. When performing 3 -RACE on rare transcripts, we find that it is absolutely necessary to add a linear amplification step, since there is no gene specificity provided by the 3 primers.

390

Li, Wright, and Anderson

7. Radioactive nucleotides and solutions containing them should always be kept behind plexiglass shields to avoid unnecessary exposure to radiation. Lab workers should be properly trained in the safe use of radioisotopes. 8. When choosing fragments for reporter assays, it is important to avoid insertion of competing ATG start codons in front of the luciferase start codon, especially those contained within a Kozak consensus (ACCATGG), since it will decrease the observed promoter activity (17). In addition, the presence of consensus splice donor (GTNAGT) or acceptor (YNYYYNCAG) sequences can affect transcript stability or bypass the luciferase start codon (18). 9. Our laboratory routinely clones all promoter fragments into pGL3 in both orientations in order to determine whether bi-directional promoter activity is present. 10. The efficiency of the transfection should be optimized and reproducible. Individual experiments should contain sufficient cells to transfect all reporter plasmids being analyzed in the study in a single experiment. The use of the Renilla internal control in each transfection allows normalization of transfection efficiency within each experiment and provides a more reliable result for comparison of subsequent experiments. 11. For good transfection efficiency it is crucial that the plasmid DNA is of high quality; the A260/A280 ratio should be at least 1.8. 12. To ensure a high degree of reproducibility, a single large batch of each control vector (pGL3-basic and pRL-SV40) should be produced and used for all experiments. 13. Transfected cells can be analyzed several hours to several days after electroporation. Times shorter than 8 h are usually not long enough for reasonable expression of the protein. Over a period of several days, the cells lose transfected DNA and thus expression from the plasmid will gradually decrease. We have found that 48 h is optimal for most reporter experiments.

Acknowledgments This project has been funded in whole or in part with federal funds from the National Cancer Institute, National Institutes of Health, under contract N01-CO-12400. The content of this publication does not necessarily reflect the views or policies of the Department of Health and Human Services nor does men-

Identification and Analysis of Novel Transcripts and Promoters

391

tion of trade names, commercial products, or organizations imply endorsement by the US government. This research was supported in part by the Intramural Research Program of the NIH, the National Cancer Institute, the Center for Cancer Research. References 1. Lanier, L. L. (2005) NK cell recognition. Annu Rev Immunol 23, 225–274. 2. Moretta, L., and Moretta, A. (2004) Killer immunoglobulin-like receptors. Curr Opin Immunol 16, 626–633. 3. Parham, P. (2005) MHC class I molecules and KIRs in human history, health and survival. Nat Rev Immunol 5, 201–214. 4. Yokoyama, W. M., and Plougastel, B. F. (2003) Immune functions encoded by the natural killer gene complex. Nat Rev Immunol 3, 304–316. 5. Moretta, A., Bottino, C., Pende, D., Tripodi, G., Tambussi, G., Viale, O., Orengo, A., Barbaresi, M., Merli, A., Ciccone, E., and Moretta, L. (1990) Identification of four subsets of human CD3-CD16+ natural killer (NK) cells by the expression of clonally distributed functional surface molecules: correlation between subset assignment of NK clones and ability to mediate specific alloantigen recognition. J Exp Med 172, 1589–1598. 6. Valiante, N. M., Uhrberg, M., Shilling, H. G., Lienert-Weidenbach, K., Arnett, K. L., D Andrea, A., Phillips, J. H., Lanier, L. L., and Parham, P. (1997) Functionally and structurally distinct NK cell receptor repertoires in the peripheral blood of two human donors. Immunity 7, 739–751. 7. Held, W., and Kunz, B. (1998) An allelespecific, stochastic gene expression process controls the expression of multiple Ly49 family genes and generates a diverse, MHCspecific NK cell receptor repertoire. Eur J Immunol 28, 2407–2416. 8. Raulet, D. H., Vance, R. E., and McMahon, C. W. (2001) Regulation of the natural killer cell receptor repertoire. Annu Rev Immunol 19, 291–330. 9. Parham, P. (2006) Taking license with natural killer cell maturation and repertoire development. Immunol Rev 214, 155–160.

10. Saleh, A., Davies, G. E., Pascal, V., Wright, P. W., Hodge, D. L., Cho, E. H., Lockett, S. J., Abshari, M., and Anderson, S. K. (2004) Identification of probabilistic transcriptional switches in the Ly49 gene cluster: a eukaryotic mechanism for selective gene activation. Immunity 21, 55–66. 11. Pascal, V., Stulberg, M. J., and Anderson, S. K. (2006) Regulation of class I major histocompatibility complex receptor expression in natural killer cells: one promoter is not enough! Immunol Rev 214, 9–21. 12. Stulberg, M. J., Wright, P. W., Dang, H., Hanson, R. J., Miller, J. S., and Anderson, S. K. (2007) Identification of distal KIR promoters and transcripts. Genes Immun 8, 124–130. 13. Davies, G. E., Locke, S. M., Wright, P. W., Li, H., Hanson, R.J., Miller, J. S., and Anderson, S. K. (2007) Identification of bi-directional promoters in the human KIR genes. Genes Immun 8, 245–253. 14. Gorman, C. M., Moffat, L. F., and Howard, B. H. (1982) Recombinant genomes which express chloramphenicol acetyltransferase in mammalian cells. Mol Cell Biol 2, 1044–1051. 15. Wood, K. V. (1991) in Chemiluminescence: Current Status (Stanley, P. and Cricka, L., eds.), Wiley, New York, p. 543. 16. Yodoi, J., Teshigawara, K., Nikaido, T., Fukui, K., Noma, T., Honjo, T., Takigawa, M., Sasaki, M., Minato, N., Tsudo, M., Uchiyama, T., and Maeda, M. (1985) TCGF (IL 2)-receptor inducing factor(s). I. Regulation of IL 2 receptor on a natural killerlike cell line (YT cells). J Immunol 134, 1623–1630. 17. Kozak, M. (1986) Point mutations define a sequence flanking the AUG initiator codon that modulates translation by eukaryotic ribosomes. Cell 44, 283. 18. Sharp, P. A. (1981) Speculation on RNA splicing. Cell 42, 643–646.

Chapter 27 Use of Inbred Mouse Strains to Map Recognition Receptors of MCMV Infected Cells in the NK Cell Gene Locus Nassima Fodil-Cornu, Michal Pyzik, and Silvia M. Vidal Abstract Genetically distinct inbred strains of mice that differ in their susceptibility to mouse cytomegalovirus (MCMV) are invaluable for dissecting complex host–pathogen interactions. Their study has allowed the identification of host-resistance loci, including several activating NK cell receptors of major histocompatibility complex (MHC) class I. In this chapter, we provide a practical guide to the genetic mapping and functional characterization of NK cell receptors that control innate immunity against MCMV via specific recognition of infected cells. Key words: MCMV, host resistance, cross, genotyping, mapping, retroviral expression, reporter cell assay.

1. Introduction The complex interaction between NK cells and cytomegaloviruses (CMVs) has been appreciated since the recognition that patients (1) and mutant mice (2) lacking functional NK cells are particularly vulnerable to infection. Shortly after, experimental depletion of NK cells using anti-NK cell-specific antibodies confirmed the role of this lymphocyte subset in host defense. Since then, infection of mice with mouse CMV (MCMV) has provided overwhelming evidence of the critical role of NK cells in the innate anti-viral response. Two major approaches have been used: “reverse genetics,” whereby mice carrying targeted Nassima Fodil-Cornu and Michal Pyzik contributed equally.

K.S. Campbell (ed.), Natural Killer Cell Protocols, Methods in Molecular Biology 612, DOI 10.1007/978-1-60761-362-6 27, © Humana Press, a part of Springer Science+Business Media, LLC 2010

393

394

Fodil-Cornu, Pyzik, and Vidal

mutations in known genes are tested for the response against MCMV. This approach has shown that mutations in genes involved in NK cell maturation, stimulation, or function (cytokine secretion, cytotoxicity, receptor-mediated apoptosis) dramatically alter the response to MCMV and render mutant mice highly susceptible compared to control littermates. “Forward genetics,” whereby natural phenotypic differences in mice are exploited to isolate and understand the genetic determinants of the NK cellmediated response against MCMV; the goal of “forward genetics” is to link a change in phenotype with a change in genotype in mouse candidate genes. Two major genetic loci have been linked to innate resistance to MCMV: NKC (natural killer cell complex) and the H2 (Histocompatibility 2 locus or mouse MHC) on chromosome 6 and 17, respectively. Among others, the NKC encodes for the C-type lectin receptor family expressed at the surface of NK cells, whereas the H2 encodes for the MHC class I molecules which constitute the natural ligands of these NK cell receptors. Among several phenotypically defined loci mapped at the NKC (3), three MCMV resistance loci exist, including Cmv1 and Cmv3, which have been resolved at the gene level as members of the Ly49 NK cell receptor family. Ly49 receptors bind to classical MHC class I or MHC class I-like proteins and present varied repertoire in different inbred mice, delivering inhibitory or activating signals. Activating Ly49 receptors, contrary to inhibitory ITIM-bearing Ly49 receptors, associate with the ITAM-containing DAP12 adapter protein. Cmv1, originally described by Scalzo and co-workers in C57BL/6 mice, is inherited as a single dominant H2-independent locus (4). The resistance allele (Cmv1r ) encodes the activating receptor Ly49H capable of binding an MHC class I-like viral product, m157 (5). In contrast, Cmv3-mediated resistance present in MA/My mice depends on a specific combination of alleles at Ly49 and H2 loci (6). In the MA/My mouse strain, the activating Ly49P receptor is able to recognize target cells but only if the cells are MCMV infected and present the H2k haplotype. Although the details of the Ly49P ligand remain to be fully elucidated, Ly49P and the H2Dk molecules seem to work in a synergistic manner in which both participants are required for viral control. Those two modes of inheritance of MCMV resistance correspond to two different modes of Ly49-mediated recognition of the infected cell. Thus, the “forward genetics” approach allowed to assign a function to activating Ly49 receptors and to identify NK cell-specific mechanisms for the recognition and control of MCMV infection. In this chapter, we provide a practical guide to the identification of Ly49 receptors that define innate resistance to MCMV via specific recognition of MCMV-infected cells.

Use of Inbred Mouse Strains to Map Recognition Receptors

395

In inbred strains resistant to MCMV, NK cells control MCMV growth and eliminate the virus from target organs within the first few days of infection. Such NK cell control is absent in MCMV-susceptible strains, including BALB/c, BALB.K, NOD, and 129S, leading to high viral growth and eventually death. Therefore, a typical forward genetic study involves the mating of MCMV-resistant (R) and MCMV-susceptible (S) mouse strain and the evaluation of progeny cohorts for viral titer in target organs, as was done in an informative (MA/My × BALB/c)F2 cross (Fig. 27.1 and Sections 3.1 and 3.2). To test a possible role of Ly49 genes and/or H2 genes in the resistance phenotype, first polymorphic markers at these loci (Table 27.1 ) are used to genotype progeny mice (Section 3.2). Subsequently, the association between the phenotype and the genotype data is tested statistically using ANOVA (Section 3.3 ). With such analysis it is possible to determine the mode of inheritance of MCMV resistance, which may segregate as an H2-independent or H2-dependent locus. Upon the confirmation of the role of Ly49 and/or H2 genes in the MCMV-resistant phenotype, a functional assay is required to determine the Ly49-mediated mechanism of recognition of the infected cell. One method originally described in Dr. Lewis Lanier’s group involves the expression of Ly49 cDNAs in reporter cells, which are co-cultured with MCMV-infected cells. The triggering of reporter cells, which is monitored by FACS, indicates

 Fig. 27.1. Investigation of the inheritance mode of Ly49-mediated resistance to MCMV: the Cmv3 example. (A) Segregation of MCMV resistance. MCMV titers (expressed as log10 PFU) in the spleens of MCMV-resistant mice (MA/My), MCMV-susceptible mice (BALB/c), and their F1 and F2 progenies. In this cross, 120 F2 mice were generated. The distribution of the spleen viral titer in the F2 progeny is continuous, i.e., characterized by a wide span ranging from resistant to susceptible mice. Such phenotypic distribution indicates that MCMV resistance is complex in this cross, i.e., determined by more than one gene. (B) Genotyping of the Ly49 and H2 loci using the Ly49e and IAA1 markers. Ly49e and IAA1 were amplified by PCR and digested by HincII and PstI (Table 27.1) to detect polymorphisms within the Ly49 and H2 loci between BALB/c and MA/My paternal strains and the heterozygote progenitors. The three possible genotypes for the Ly49 and H2 loci are shown. (C) Contribution of Ly49 and H2 loci to the resistance phenotype. Ly49e genotypes are shown at the bottom; ‘m’ and ‘c’ represent inheritance of MA/My and BALB/c alleles, respectively. H2 genotypes (obtained with the IAA1 genetic marker) are shown at the top; H2k and H2d represents inheritance of MA/My and BALB/c alleles, respectively. Box plots indicate the interquartile range and median (horizontal line) of log10 PFU spleen viral load for each of the nine possible allelic combinations at Ly49 and H2 loci. Using ANOVA and Bonferroni post hoc analysis, we found that there is a significant effect on spleen viral load only when both the H2kk haplotype and homozygosity at the Ly49mm locus from the MA/My strain are present. Likewise, only when the Ly49mm is homozygous for the MA/My genotype, the H2kk has an effect. This result indicates that MCMV resistance depends on the interaction of two loci, Ly49 and H2.

396

Fodil-Cornu, Pyzik, and Vidal

Fig. 27.1. (continued).

the ability of the specific Ly49 receptor to recognize the MCMVinfected cell (Fig. 27.2 and Section 3.5)

2. Materials 2.1. Investigation of the Inheritance Mode of Ly49-Mediated Resistance to MCMV

1. Animals: Mice are purchased from Jackson Laboratories. The strain name, the MCMV phenotype, and the catalog number are listed in Table 27.1.

128468912

128784912 129544491

129849145

D6mit61

D6mit135 Nkg2d

Ly49e

PCR

47707105

29405253 34137861

D17mit68

D17mit101 D17MIT28

PCR PCR

RFLP or PstIa HindIIIb

RFLP HinFI

RFLP HincII

PCR RFLP XbaI

PCR

PCR

143 120

134

167, 96b

374

980

140 300, 300, 100

136

147

Methodology C57Bl/61R

IAA1

D6Ott11

127788235

D6mit52

pb MGI

137

140

ND ND

ND

263a

-