METHODS
IN
MOLECULAR BIOLOGY™
Series Editor John M. Walker School of Life Sciences University of Hertfordshire Hatfield, Hertfordshire, AL10 9AB, UK
For other titles published in this series, go to www.springer.com/series/7651
Two-Dimensional Electrophoresis Protocols Edited by
David Sheehan and Raymond Tyther Department of Biochemistry, University College Cork, Cork, Ireland
Editors David Sheehan Department of Biochemistry University College Cork Cork, Ireland
Raymond Tyther Department of Biochemistry University College Cork Cork, Ireland
ISBN: 978-1-58829-937-6 e-ISBN: 978-1-59745-281-6 ISSN: 1064-3745 e-ISSN: 1940-6029 DOI: 10.1007/978-1-59745-281-6 Library of Congress Control Number: 2008942064 © Humana Press, a part of Springer Science+Business Media, LLC 2009 All rights reserved. This work may not be translated or copied in whole or in part without the written permission of the publisher (Humana Press, c/o Springer 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. Cover Illustration: Figure 4A, Chapter 10, “Proteomic Analysis of Caenorhabditis Elegans” by Pan-Young Jeong, Keun Na, Mi-Jeong Jeong, David Chitwood, Yhong-Hee Shim, and Young-Ki Paik Printed on acid-free paper springer.com
Preface The human genome and other large-scale genome sequencing projects have inevitably led to a focus on the proteins encoded by genes. The field of proteomics has grown enormously as a result and a number of high-throughput technologies have now been developed allowing discovery-led investigations of protein populations rather than more traditional hypothesis-led studies on single proteins. These high-throughput technologies include gene and protein microarrays, the yeast two-hybrid system, and various mass spectrometry methodologies. However, despite developments and improvements in these technologies, two-dimensional electrophoresis (2DE) remains one of the most widely used approaches. This technique was revolutionised by the development of immobilised pH gradient strips which are now commercially available. This has made possible highly reproducible separations of matched samples. Developments in staining, mass spectrometry, and bioinformatics supported these developments and have led to a measure of standardisation in design, execution, and analysis of proteomics experiments. This book began life as a proposed update of the excellent volume 2DE Protocols edited by Andrew Link of the University of Washington at Seattle. However, we realised that 2DE has undergone major development in aspects of its technology in recent years and we were anxious to reflect these in the present volume. We are also conscious that many researchers have now begun to apply proteomics methodologies to a growing range of biological material and we were anxious to include contributions to reflect the challenges posed in sample preparation in less widely used organisms. As with all of this series, the emphasis in this volume is on the presentation of clear protocols suitable for a newcomer to the field. We felt, however, that some aspects merited inclusion of overview review-type articles and a number of these are included at the beginning of the book. The protocols reflect the key steps in a 2DE experiment which include sample preparation, staining, post-translational modification, spot identification, and bioinformatics. We hope especially that newcomers to 2DE will find this volume useful and be encouraged to apply some of the powerful techniques described here to their own research. The editors would like to thank especially the series editor, Prof. John Walker, for his endless patience, enthusiasm, and encouragement throughout this project. We would also like to thank our contributors for their excellent cooperation and generosity in sharing their expertise in this book. Cork, Ireland
David Sheehan Raymond Tyther
v
Contents Preface. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Contributors. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
v xi
OVERVIEW CHAPTERS 1 2 3 4 5
6
7
Two-Dimensional Electrophoresis: An Overview . . . . . . . . . . . . . . . . . . . . . . . . . 3 Richard Smith Solubilization of Proteins in 2DE: An Outline . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 Thierry Rabilloud Selection of pH Ranges in 2DE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 Mireille Starita-Geribaldi Difficult Proteins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47 Ben Herbert and Elizabeth Harry Organelle Proteomics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65 Matthias Plöscher, Bernhard Granvogl, Veronika Reisinger, Axel Masanek, and Lutz Andreas Eichacker Applications of Chemical Tagging Approaches in Combination with 2DE and Mass Spectrometry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83 Alexander Leitner and Wolfgang Lindner Immunoblotting 2DE Membranes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103 Brian McDonagh
PROTOCOLS CHAPTERS 8 9 10
11 12 13 14
Troubleshooting Image Analysis in 2DE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Bettina Levänen and Åsa M. Wheelock Analysis of Bacterial Proteins by 2DE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Philip Cash and Evelyn Argo Proteomic Analysis of Caenorhabditis elegans . . . . . . . . . . . . . . . . . . . . . . . . . . . . Pan-Young Jeong, Keun Na, Mi-Jeong Jeong, David Chitwood, Yhong-Hee Shim, and Young-Ki Paik Protein Extraction for 2DE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Claus Zabel and Joachim Klose Analysis of Proteins from Marine Molluscs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Suze Chora, Maria João Bebianno, and Michèle Roméo Preparation and Analysis of Plastid Proteomes by 2DE . . . . . . . . . . . . . . . . . . . . . Anne von Zychlinski and Wilhelm Gruissem High-Resolution 2DE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Katrin Marcus, Cornelia Joppich, Caroline May, Kathy Pfeiffer, Barbara Sitek, Helmut Meyer, and Kai Stuehler
vii
113 131 145
171 197 205 221
viii
15
Contents
Blue Native-Gel Electrophoresis Proteomics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 241 Kelly Andringa, Adrienne King, and Shannon Bailey
16 2DE for Proteome Analysis of Human Metaphase Chromosomes . . . . . . . . . . . . . 259 Kiichi Fukui and Susumu Uchiyama 17 Microsomal Proteomics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 273 Diana M. Wong and Khosrow Adeli
18 19 20 21 22
23
24 25
26
27 28 29 30 31 32
Prefractionation Using Microscale Solution IEF . . . . . . . . . . . . . . . . . . . . . . . . . . Won-A Joo and David Speicher Diagonal Electrophoresis for Detection of Protein Disulphide Bridges . . . . . . . . . Brian McDonagh High-Resolution Large-Gel 2DE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Claus Zabel and Joachim Klose Silver Staining of Proteins in 2DE Gels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Cécile Lelong, Mireille Chevallet, Sylvie Luche, and Thierry Rabilloud Detection of 4-Hydroxy-2-Nonenal- and 3-Nitrotyrosine-Modified Proteins Using a Proteomics Approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Rukhsana Sultana, Tanea Reed, and D. Allan Butterfield Proteomic Detection of Oxidized and Reduced Thiol Proteins in Cultured Cells . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Sarah L. Cuddihy, James W. Baty, Kristin K. Brown, Christine C. Winterbourn, and Mark B. Hampton Detection of Ubiquitination in 2DE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Brian McDonagh Phosphoproteome Analysis by In-Gel Isoelectric Focusing and Tandem Mass Spectrometry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Sarka Beranova-Giorgianni, Dominic M. Desiderio, and Francesco Giorgianni Detection of Protein Glutathionylation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Elisabetta Gianazza, Ivano Eberini, and Pietro Ghezzi Activity-Based Protein Profiling of Protein Tyrosine Phosphatases . . . . . . . . . . . . Chad Walls, Bo Zhou, and Zhong-Yin Zhang Active Protease Mapping in 2DE Gels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Zhenjun Zhao and Pamela J. Russell Two-Dimensional Difference Gel Electrophoresis . . . . . . . . . . . . . . . . . . . . . . . . . Gert Van den Bergh Protein Expression Profiling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Brian P. Bradley, Bose Kalampanayil, and Michael C. O’Neill C-Terminal Sequence Analysis of 2DE-Separated Proteins . . . . . . . . . . . . . . . . . . Bart Samyn, Kjell Sergeant, and Jozef Van Beeumen Shotgun Protein Analysis by Liquid Chromatography-Tandem Mass Spectrometry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Kazuishi Kubota, Toshiyuki Kosaka, and Kimihisa Ichikawa
291 305 311 339
351
363
377
383
397 417 431 439 455 469
483
Contents
De Novo Sequence Analysis of N-Terminal Sulfonated Peptides After in-Gel Guanidination . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Kjell Sergeant, Jozef Van Beeumen, and Bart Samyn 34 Tryptic Digestion of In-Gel Proteins for Mass Spectrometry Analysis . . . . . . . . . . Mai-Loan Huynh, Pamela Russell, and Bradley Walsh 35 Database Interrogation Algorithms for Identification of Proteins in Proteomic Separations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Patricia M. Palagi, Frédérique Lisacek, and Ron D. Appel 36 Creating 2DE Databases for the World Wide Web . . . . . . . . . . . . . . . . . . . . . . . . Christine Hoogland, Khaled Mostaguir, and Ron D. Appel Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
ix
33
495 507
515 533 541
Contributors KHOSROW ADELI • Molecular Structure and Function, The Hospital for Sick Children Research Institute, and the Department of Biochemistry, University of Toronto, Toronto, ON, Canada KELLY ANDRINGA • Department of Environmental Health Sciences and The Center for Free Radical Biology, University of Alabama at Birmingham, Birmingham, AL, USA RON APPEL • Proteome Informatics Group, Swiss Institute of Bioinformatics and Computer Science Department, Geneva University, Geneva, Switzerland SHANNON BAILEY • Department of Environmental Health Sciences and The Center for Free Radical Biology, University of Alabama at Birmingham, Birmingham, AL, USA JAMES W. BATY • Department of Pathology, Christchurch School of Medicine & Health Sciences, University of Otago, Christchurch, New Zealand MARIA BEBIANNO • Department of Marine Science, University of the Algarve, Faro, Portugal SARKA BERANOVA-GIORGIANNI • Department of Pharmaceutical Sciences, Charles B. Stout Neuroscience Mass Spectrometry Laboratory and Department of Neurology, The University of Tennessee Health Science Center, Memphis, TN, USA BRIAN BRADLEY • Department of Biological Sciences, University of Maryland, Baltimore County, Baltimore, MD, USA KRISTIN K. BROWN • Department of Pathology, Christchurch School of Medicine & Health Sciences, University of Otago, Christchurch, New Zealand ALLAN BUTTERFIELD • Department of Chemistry, Sanders-Brown Center on Aging and Center of Membrane Sciences, University of Kentucky, Lexington, KY, USA PHILIP CASH • Department of Medical Microbiology, University of Aberdeen, Aberdeen, Scotland, UK MIREILLE CHEVALLET • CEA-DSV-iRTSV/LBBSI and UMR CNRS, CEA Grenoble, Grenoble, France DAVID CHITWOOD • Nematology Laboratory, USDA, ARS, BARC-West, Beltsville, MD, USA SUZE CHORA ROSE • University of Nice Sophia-Antipolis, Nice, France Department of Marine Science, University of the Algarve, Faro, Portugal SARAH L. CUDDIHY • Department of Pathology, Christchurch School of Medicine & Health Sciences, University of Otago, Christchurch, New Zealand DOMINIC DESIDERIO • Charles B. Stout Neuroscience Mass Spectrometry Laboratory and Department of Neurology, The University of Tennessee Health Science Center, Memphis, TN, USA
xi
xii
Contributors
IVANO EBERINI • Dipartimento di Scienze Farmacologiche, Università degli Studi di Milano, Milan, Italy LUTZ EICHAKER • Department für Biologie I, Ludwig-Maximilians-Universität, Munich, Germany KIICHI FUKUI • Department of Biotechnology, University of Osaka, Osaka, Japan PIETRO GHEZZI • Istituto di Ricerche Farmacologiche “Mario Negri”, Milan, Italy ELIZABETTA GIANAZZA • Dipartimento di Scienze Farmacologiche, Università degli Studi di Milano, Milan, Italy FRANCESCO GIORGIANNI • Charles B. Stout Neuroscience Mass Spectrometry Laboratory and Department of Neurology, The University of Tennessee Health Science Center, Memphis, TN, USA BERNHARD GRANVOGL • Department für Biologie I, Ludwig-MaximiliansUniversität, Munich, Germany WILHELM GRUISSEM • Institute of Plant Sciences and Functional Genomics Center, ETH Zurich, Zurich, Switzerland MARK B. HAMPTON • Department of Pathology, Christchurch School of Medicine & Health Sciences, University of Otago, Christchurch, New Zealand ELIZABETH HARRY • Institute for the Biotechnology of Infectious Diseases, University of Technology, Sydney, NSW, Australia BEN HERBERT • Institute for the Biotechnology of Infectious Diseases, University of Technology, Sydney, NSW, Australia CHRISTINE HOOGLAND • Swiss Institute of Bioinformatics, Geneva, Switzerland MAI-LOAN HUYNH • Minomic Pty Ltd, Chatswood West, NSW, Australia KIMIHISA ICHIKAWA • Advanced Technology Research Laboratories, Daiichi Sankyo Co., Ltd., Hiromachi, Shinagawa-ku, Tokyo, Japan MI-JEONG JEONG • Yonsei Proteome Research Center, Biomedical Proteome Research Center and Department of Biochemistry, Yonsei University, Seoul, Korea PAN-YOUNG JEONG • Yonsei Proteome Research Center, Biomedical Proteome Research Center and Department of Biochemistry, Yonsei University, Seoul, Korea WON-A JOO • The Wistar Institute, Philadelphia, PA, USA CORNELIA JOPPICH • Medizinisches Proteom-Center, Ruhr-Universität Bochum, Bochum, Germany BOSE KALAMPANAYIL • Department of Biological Sciences, University of Maryland, Baltimore County, Baltimore, MD, USA ADRIENNE KING • Department of Environmental Health Sciences and The Center for Free Radical Biology, University of Alabama at Birmingham, Birmingham, AL, USA JOACHIM KLOSE • Institut für Humangenetik, Universitätsklinikum Charité, Berlin, Germany TOSHIYUKI KOSAKA • Advanced Technology Research Laboratories, Daiichi Sankyo Co., Ltd., Hiromachi, Shinagawa-ku, Tokyo, Japan KAZUISHI KUBOTA • Advanced Technology Research Laboratories, Daiichi Sankyo Co., Ltd., Hiromachi, Shinagawa-ku, Tokyo, Japan ALEXANDER LEITNER • Department of Analytical Chemistry and Food Chemistry, University of Vienna, Vienna, Austria
Contributors
xiii
CÉCILE LELONG • CEA-DSV-iRTSV/LBBSI and UMR CNRS, CEA Grenoble, Grenoble, France BETTINA LEVÄNEN • Karolinska Biomics Center and Department of Medicine, Division of Respiratory Medicine, Karolinska Institutet, Stockholm, Sweden WOLFGANG LINDNER • Department of Analytical Chemistry and Food Chemistry, University of Vienna, Vienna, Austria FREDERIQUE LISACEK • Proteome Informatics Group, Swiss Institute of Bioinformatics, Geneva, Switzerland SYLVIE LUCHE • CEA-DSV-iRTSV/LBBSI and UMR CNRS, CEA Grenoble, Grenoble, France KATRIN MARCUS • Medizinisches Proteom-Center, Ruhr-Universität Bochum, Bochum, Germany AXEL MASANEK • Department für Biologie I, Ludwig-Maximilians-Universität, Munich, Germany CAROLINE MAY • Medizinisches Proteom-Center, Ruhr-Universität Bochum, Bochum, Germany BRIAN MCDONAGH • Department of Biochemistry, University College Cork, Cork, Ireland HELMUT MEYER • Medizinisches Proteom-Center, Ruhr-Universität Bochum, Bochum, Germany KHALED MOSTAGUIR • Swiss Institute of Bioinformatics, Geneva, Switzerland KEUN NA • Yonsei Proteome Research Center, Biomedical Proteome Research Center and Department of Biochemistry, Yonsei University, Seoul, Korea MICHAEL C. O’NEILL • Department of Biological Sciences, University of Maryland, Baltimore County, Baltimore, MD, USA YOUNG-KI PAIK • Yonsei Proteome Research Center, Biomedical Proteome Research Center and Department of Biochemistry, Yonsei University, Seoul, Korea PATRICIA PALAGI • Proteome Informatics Group, Swiss Institute of Bioinformatics, Geneva, Switzerland KATHY PFEIFFER • Medizinisches Proteom-Center, Ruhr-Universität Bochum, Bochum, Germany • Department für Biologie I, Ludwig-Maximilians-Universität, ´´ MATTHIAS PL OSCHER Munich, Germany THIERRY RABILLOUD • Commissariat à l’Energie Atomique, Grenoble, France TANEA REED • Department of Chemistry, Sanders-Brown Center on Aging and Center of Membrane Sciences, University of Kentucky, Lexington, KY, USA VERONIKA REISINGER • Department für Biologie I, Ludwig-Maximilians-Universität, Munich, Germany MICHELE ROMEO • University of Nice Sophia-Antipolis, Nice, France PAMELA J. RUSSELL • Oncology Research Centre, Prince of Wales Hospital, Randwick, and Faculty of Medicine, University of New South Wales, Sydney, NSW, Australia BART SAMYN • Department of Biochemistry Physiology and Microbiology, University of Ghent, Ghent, Belgium KJELLL SERGEANT • Department of Biochemistry Physiology and Microbiology, University of Ghent, Ghent, Belgium
xiv
Contributors
YHONG-HEE SHIM • Department of Bioscience and Biotechnology and Bio/Molecular Informatics Center, Konkuk University, Seoul, Korea BARBARA SITEK • Medizinisches Proteom-Center, Ruhr-Universität Bochum, Bochum, Germany RICHARD SMITH • Department of Medical Physics and Applied Radiation Sciences, McMaster University, Hamilton, ON, Canada DAVID SPEICHER • The Wistar Institute, Philadelphia, PA, USA MIRIELLE STARITA-GERIBALDI • Department des Sciences Biologique UFR d’Ontologie, Pôle Universitaire Saint-Jean d’Angely, Nice, France KAI STUEHLER • Medizinisches Proteom-Center, Ruhr-Universität Bochum, Bochum, Germany RUKSHANA SULTANA • Department of Chemistry, Sanders-Brown Center on Aging and Center of Membrane Sciences, University of Kentucky, Lexington, KY, USA RAYMOND TYTHER • Department of Biochemistry, University College Cork, Cork, Ireland SUSUMU UCHIYAMA • Department of Biotechnology, University of Osaka, Osaka, Japan JOSEF VAN BEEUMEN • Department of Biochemistry Physiology and Microbiology University of Ghent, Ghent, Belgium GERT VAN DEN BERGH • Laboratory of Neuroplasticity and Neuroproteomics, Katholieke Universiteit Leuven, Leuven, Belgium ANNE VON ZYCHLINSKI • Department of Biochemistry, University of Otago, Dunedin, New Zealand CHAD WALLS • Department of Biochemistry and Molecular Biology, Indiana University School of Medicine, Indianapolis, IN, USA BRADLEY WALSH • Minomic Pty Ltd, Chatswood West, NSW, Australia ÅSA WHEELOCK • Karolinska Biomics Center and Department of Medicine, Division of Respiratory Medicine, Karolinska Institutet, Stockholm, Sweden CHRISTINE C. WINTERBOURN • Department of Pathology, Christchurch School of Medicine & Health Sciences, University of Otago, Christchurch, New Zealand DIANA WONG • Molecular Structure and Function, The Hospital for Sick Children Research Institute, and the Department of Biochemistry, University of Toronto, Toronto, ON, Canada CLAUS ZABEL • Institut für Humangenetik, Universitätsklinikum Charité, Berlin, Germany ZHONG-YIN ZHANG • Department of Biochemistry and Molecular Biology, Indiana University School of Medicine, Indianapolis, IN, USA ZHENJUN ZHAO • Institute for Eye Research and Vision Cooperative Research Centre, University of New South Wales, Sydney, NSW, Australia BO ZHOU • Department of Biochemistry and Molecular Biology, Indiana University School of Medicine, Indianapolis, IN, USA
Overview Chapters
Chapter 1 Two-Dimensional Electrophoresis: An Overview Richard Smith Summary Two-dimensional gel electrophoresis (2DE) separates proteins by molecular charge and molecular size. Proteins are first solubilised in a denaturing buffer containing a neutral chaotrope, a zwitterionic or neutral detergent, and a reducing agent. First-dimension isoelectric keywords, focusing, then subjects proteins to a high voltage within a pH gradient. The amphoteric nature of proteins means each migrates to the pH where the net molecular charge is zero. After equilibration, to ensure complete protein unfolding, the second dimension separates by molecular size. Each protein is therefore resolved at a unique isoelectric point/molecular size coordinate. After visualisation by staining proteome changes are revealed by gel image analysis, and protein spots of interest excised and identified by mass spectrometry sequence analysis combined with database comparison. Variations to this procedure include staining or radio-labelling prior to electrophoresis. Although 2DE does have limitations, the most significant being the resolution of membrane and/or hydrophobic proteins, the potential solutions offered by pre-fractionation or adjustments to the electrophoresis regimen mean this technique is likely to remain central to proteomic research. Key words: Protein solubilisation, Interfering substances, Isoelectric focussing, Polyacrylamide slab gel, Imaging, Difference gel electrophoresis, Radio-labelling, Membrane proteins, Review.
1. Proteomics and Two-Dimensional Gel Electrophoresis
The term ‘proteomics’ can refer to the protein complement of a cell at any given time, the complete set of proteins expressed during a cell life cycle (1), or be used in a more specialised context: cell map proteomics ‘…defines all proteins within particular organelles to gain insight into cellular architecture and protein function’, whereas functional proteomics ‘…identifies proteins in a cell, tissue or organism that undergo changes in response to a
David Sheehan and Raymond Tyther (eds.), Methods in Molecular Biology, Two-Dimensional Electrophoresis Protocols, vol. 519 © Humana Press, a part of Springer Science + Business Media, LLC 2009 DOI: 10.1007/978-1-59745-281-6_1
3
4
Smith
specific biological condition’ (2). Proteomic research can be global or targeted, i.e. directed towards all proteins, or restricted to a well-defined group, such as glycoproteins, organelle proteins, or ribosomal proteins, respectively (3). Analysing the proteome also provides insights into gene product localisation, proteolysis, recycling, and sequestration (4). Thus proteomics is more than simply visualising the genome. In fact mRNA and protein expression are poorly related (5, 6), and the extensive and variable degree of post-translational modification means significantly more proteins may be expressed than are accounted for by the genome (7). Two-dimensional (2D) gel electrophoresis (2DE) can be used to address all of the above and is therefore now a central technique in proteomic research. Originally developed in 1975 (8, 9), the procedure has been summarised as isoelectric focusing (IEF) in a pH gradient, followed by separation on sodium dodecyl sulphate (SDS) polyacrylamide slab gels (SDS–PAGE) (10). Thus proteins are separated firstly according to molecular charge and secondly according to molecular mass. Since these are unrelated parameters, individual protein spots are resolved almost uniformly over the second-dimension gel (8), with each located at a unique pH/molecular mass coordinate. Some excellent reviews of 2DE have already been written (1, 11–13), and this chapter sets out to summarise the main principles, considerations, and indeed the limitations of the process, from initial tissue preparation to gel analysis.
2. Principles and Development of TwoDimensional Gel Electrophoresis Methodology 2.1. Protein Preparation and Solubilisation
Prior to electrophoresis complete protein solubilisation is essential (13). This is achieved by mechanical disruption in a solubilisation buffer, the purpose being the breakage of all non-covalent interactions, both between proteins and between proteins and non-proteinaceous molecules, and the reduction of disulphide bonds (14). This disaggregates all macromolecular complexes and reduces and denatures individual proteins or polypeptides (1). Solubilisation should be accomplished quickly, ideally using ice-cold buffer and chilled disruption apparatus. It is also important to prevent proteolysis, protein modification, or the introduction of artefacts (1, 14–16). High-quality reagents are required to make the buffer and solubilisation immediately before IEF is advisable (1). Interfering substances must also be removed or denatured, and protein solubility must be maintained throughout IEF. The composition of 2DE solubilisation buffer, which has remained relatively unchanged since the original investigation
Two-Dimensional Electrophoresis: An Overview
5
(8), is based on a neutral chaotrope, a neutral or zwitterionic detergent, and a reducing agent (1, 14). Urea (8 M) is the most common chaotrope, but because it is not so effective at breaking hydrophobic interactions, membrane proteins are relatively poorly solubilised. Although a urea/thiourea mixture increases the number of individual solubilised proteins (17), particularly membrane proteins (14), it is possible thiourea may reduce the resolution of some of the more acidic proteins. As a result the ideal proportions of urea and thiourea vary according to sample (18). The original detergents, Triton X-100 and Nonidet P-40, have largely been replaced by the zwitterionic cholamidosulphobetaine detergent 3-[3-cholamidopropyl dimethylammonio]1-propane sulphonate (CHAPS) (19) or by alkylsulphobetaine detergents (20), as these also improve membrane protein solubility. However these are less soluble in 8 M urea, and the trade-off for using a lower urea concentration is a reduction in overall protein solubilisation. SDS is also sometimes used and boiling with SDS, followed by dilution with urea or urea/ thiourea; both improve solubilisation and also inhibit proteases. However, to avoid interference during IEF, SDS must be displaced from proteins, by excessive dilution with neutral or zwitterionic detergents (21). The original reductant, 2-mercaptoethanol, has been replaced by dithiothreitol (DTT) or dithioerythreitol (DTE). With more insoluble samples tributyl phosphine (TBP) has also proved useful (22), but this is less stable, less soluble in aqueous solutions, and less able to maintain proteins in a reduced state. Protein solubility is also pH dependent, but care must be taken as high buffer concentrations interfere with IEF. For example, HEPES (being amphoteric) can focus within the pH gradient, thus preventing protein focusing in the resulting area of high conductivity, whilst Tris (being cationic) concentrations can increase at the cathode, thus reducing alkaline protein focusing. The latter can be corrected by using a spermine base (23). However, since the pH for optimal solubility differs between proteins, altering the buffer can result in differences in the final two-dimensional gel. 2.2. Removal of Interfering Substances
A number of non-proteinaceous compounds can interfere with one or other aspect of 2DE. Whereas some, such as hormones or simple sugars, do not cause an appreciable amount of interference (14), others can significantly affect protein solubilisation or final gel visualisation (1, 11, 12). Many small ionic metabolites accumulate at the anode or cathode during first-dimension IEF, thus adversely increasing conductivity and therefore reducing or eliminating focusing. A number of ion-removal methods use protein precipitation and resolubilisation based on, for example, trichloroacetic acid (TCA)
6
Smith
(24), acetone (25), ethanol or chloroform and methanol (26), or TCA and acetone (27–31), but these all risk protein loss (Note: TCA must be removed, by diethyl ether or acetone wash, before IEF). Commercially available de-salt columns offer a convenient solution, and an alternative to salt removal is to minimise salt inclusion from the outset, by using a low ionic strength washing buffer (e.g. Tris-buffered sorbitol). Nucleic acids increase the viscosity of the solubilised protein solution, thus reducing IEF efficiency, particularly in the acidic section, and can ‘clog’ the second-dimension gel pores. Nucleic acids can also combine with ampholytes (if used) to cause ‘streaking’ of the final gel. Both DNA and RNA can be removed by ultracentrifugation and the addition of a basic polyamine (spermine) (14), but this risks protein removal, especially those proteins bound to nucleic acids. The addition of RNase and DNase can reduce nucleic acid contamination, but the high urea concentration reduces enzyme activity. Additionally there is the consideration of introducing additional spots on the final gel. Lipids reduce protein solubility. However ensuring an adequate amount of detergent is present in the solubilisation buffer reduces the influence of lipids. Lipids can also be removed by the use of organic solvent (methanol or acetone) extraction, but again this can result in protein loss (12). Inhibiting protease activity is essential if proteolytic modifications are to be avoided and proteomic changes are to be confidently identified as genuine. Rapid solubilisation, under ice-cold conditions, minimises protease activity, as does boiling in ureafree SDS or adding Tris base (although high Tris concentrations can impede IEF). This can be augmented by the use of protease inhibitors (although their activity can be affected by DTT or DTE), such as phenylmethylsulphonyl fluoride (PMSF) and/or aminoethylbenzenesulphonyl fluoride (AEBSF). However these inhibitors can modify some proteins (16, 32). Interference from phenolic compounds is particular to plant proteomics. These can be removed by TCA protein precipitation and re-solubilisation (33), or adsorption onto polyvinylpyrrolidone (PVP) or polyvinylpolypyrrolidone (PVPP) (25). Alternatively the use of reductants (DTT and DTE) prevents protein oxidation from these compounds. 2.3. First-Dimension Isoelectric Focusing
The net charge of a protein molecule is pH-dependent. IEF exploits this amphoteric property by subjecting proteins contained in a pH gradient to a high voltage. Each protein then migrates to its isoelectric point (pI), i.e. the pH where the net charge is zero and the molecule is therefore immobilised. Originally this was carried out in polyacrylamide tube gels (8, 9, 34) and facilitated by carrier ampholytes, i.e. small amphoteric molecules with a high pH buffering capacity which, once a voltage
Two-Dimensional Electrophoresis: An Overview
7
is applied, ensure a smooth pH gradient is generated. However acrylamide tube gels are inconvenient and carrier ampholytes suffer from batch variation and reduced efficiency at extreme pHs (1). The most significant IEF development is the immobilised pH gradient (IPG) (35–37). IPGs employ ten acrylic derivatives (immobiline reagents): CH2=CH–CO–NH–R, where R = a carboxy or amino residue copolymerised with an acrylic matrix (11) resulting in a stable, consistent pH gradient (35, 38, 39). The greater accuracy of pI resolution by IPG strips was quickly realised (40, 41), as well as the possibility to employ a wider pH range than is possible with carrier ampholytes (1, 11). Additionally IPGs eliminate cathodic drift and can (depending on pH range) accommodate a greater total protein load (42). A wide pH range (4–11) can hold approximately 1 mg of total protein, whereas a narrow range (1 pH unit) can hold 10 mg (12). IPG strips also allow protein loading to be combined with gel rehydration. Typically protein samples are applied throughout the entire gradient. However ‘cup loading’ at the cathode (38) or, more usually, the anode, is required with basic pH gradients: pH 6–10, 7–10, or 6–12 (43). IEF time and temperature are critical. The optimal time is when IEF attains a steady state (37, 43, 44). Insufficient time causes horizontal streaking in the final gel, whilst excessively long IEF results in the phenomena known as electroendosmotic flow, i.e. water loss from the gel. If the temperature is too low, urea crystallisation can impede IEF, whilst IEF at too high a temperature causes protein carbamylation and inconsistent spot location on the final gel (44). A temperature of 20°C is best in terms of protein entry and mobility within the gradient. 2.4. IPG Strip Equilibration
IPG equilibration ensures that proteins are able fully to interact with SDS in the second dimension. Two equilibration buffers are involved: buffer 1 contains glycerol and urea, and reduces the electroendosmotic effect which can interrupt protein transfer to the second dimension (37); and buffer 2 contains iodoacetamide, in place of DTT, to prevent DTT streaking in the second-dimension gel (45). Inadequate equilibration, particularly with buffer 1, reduces protein transfer to the second-dimension gel (46).
2.5. SecondDimension Protein Separation
The second-dimension gel contains the anionic detergent SDS, which completes the protein unfolding initiated by DTT. Once completely unfolded, proteins are ellipsoid and contained in an SDS micelle. In this state the second-dimension gel is able to exploit the linear relationship between the log molecular mass and migration distance (47, 48) and thus separate by protein size. Two categories of second-dimension gel exist: homogenous and gradient. In both gel pore size is controlled by adjusting the
8
Smith
percentages of polyacrylamide and cross-linker. Homogenous gels contain the same percentage of acrylamide and cross-linker throughout, whereas gradient gels contain a variable percentage of cross-linker. The former gives the best resolution within a specific molecular size range, whilst the latter is best for running a wider range of protein sizes (1). Of the buffers used, Laemmli buffer (47) is the most common (49), but Tris–tricine improves the separation of smaller proteins (12). 2.6. Protein Spot Detection
After fixation, in an acetic acid, ethanol and water mixture, protein detection options include various staining techniques or, if radio-labelled proteins are involved, autoradiography, fluorography, and phosphor-imaging (50–52) (For extended reviews see refs.1, 12).
2.6.1. Gel Staining
Protein staining is by far the more common detection method. The least sensitive stain, Coomassie blue, is an anionic triphenylmethane dye which is electrostatically attracted to basic amino acids (53). Although staining 200 ng protein spot−1 or less has been reported (53), more usually 1 μg protein spot−1 is required. Coomassie blue is mass spectrometry compatible and, because of its colloidal nature, causes very little background staining. Ruthenium-based SYPRO stains also bind non-covalently to basic amino acids (54). These mass-spectrometry-compatible stains are more sensitive than Coomassie. Protein identifications are possible with 1–2 ng protein spot−1 (i.e. femtomole range, 55, 56) and have a wider dynamic range (the ability to stain in proportion to, and thus reflect, a wider range of protein quantities), thus greatly improving quantitative analysis of the proteome. Whilst Coomassie and SYPRO stains bind to the protein, negative staining uses the precipitation of insoluble imidazole zinc (57). Precipitation is slower where the gel is occupied by proteins, resulting in a “negative” appearance when imaged. Since there is very little interaction between stain and protein this technique is also mass spectrometry compatible. Sensitivity is between Coomassie and SYPRO staining (approximately 0.1 ng protein spot−1), but the poor linear range means it is less useful for quantitative analysis. Silver staining is the most sensitive, with detection of 0.1 ng protein spot−1 being possible. More than 100 variants of the original protocol (58) now exist (59), and these fall into two categories: silver nitrate with alkaline formaldehyde (60) and the more sensitive silver diamine salt complexed with ammonia (61). Both methods of silver staining are time consuming, and demand cleanliness and a high degree of reagent and water quality (>15 MΩcm). However the biggest disadvantage is that silver stains the surface molecules in a spot which usually renders the protein unsuitable for mass spectrometry analysis. Mass-spectrometry-compatible
Two-Dimensional Electrophoresis: An Overview
9
silver staining is possible (62, 63) but, since this is less sensitive than ammoniacal silver staining (62, 64–67) and has the same reagent and water purity demands, it is less popular than SYPRO staining. Like Coomassie blue, silver staining does not have the dynamic range of SYPRO stain and is only able to resolve a tenfold (as opposed to a 1,000-fold) variation in protein spot volume. 2.6.2. Gel Imaging, Protein Quantification, and Identification
Of the imaging methods available (68) the Charged Coupled Device (CCD) is probably the most versatile (1). When lightsensitive photosites are hit by photons they emit electrons which are counted by the CCD camera (13). Imaging is the first stage in converting analogue to digital data. Gaussian algorithms then quantify each protein spot according to a grey scale which can be analysed using the various two-dimensional gel software packages (see also Chapter “Troubleshooting Image Analysis in 2DE”). Detecting those protein spots which are either induced or deleted (i.e. absolute qualitative changes) is relatively straightforward. In theory, this imaging process should also allow the detection of quantitative changes to the proteome (i.e. changes in spot size, volume, or intensity, which reflect an up- or downregulation of a protein). However, in spite of the fact 2DE and image analysis are recognised as key technologies in measuring protein abundance (13), the question has been raised of whether limitations in dynamic ranges mean staining is strictly qualitative (12). Certainly this concern does have some merits and if twodimensional gels are to be analysed quantitatively the use of an optimal protein load combined with a wider dynamic range stain is definitely advantageous. Nevertheless, as long as a mass-spectrometry-compatible staining method is used, any changes in protein expression, detected by 2DE, can be identified by western blotting (69) (see also Chapter “Multivalent Protein Probes for the Identification and Characterization of Cognate Cellular Ligands for Myeloid Cell Surface Receptors”) and/or peptide analysis of excised protein spots, using mass spectrometry (4), combined with protein sequence database comparison (70)(see Chapter “Database Interrogation Algorithms for Identification of Proteins in Proteomic Separations”).
2.7. Difference Gel Electrophoresis
The key difference with difference gel electrophoresis (DIGE) is that the solubilised proteins are stained prior to IEF, with fluorescent cyanine dye N-hydroxy succininmidyl ester derivatives: Cy3 or Cy5 (71) (see Chapters “High-Resolution 2DE” and “TwoDimensional Difference Gel Electrophoresis”). Since Cy3 and Cy5 have different excitation and emission wavelengths, individual protein samples can be mixed, run on a single gel, and a direct comparison of protein expression made (72–74). The technique
10
Smith
can be made more robust by including a third dye (Cy2) (75). The major advantage of DIGE is the pooling of multiple samples on a single gel, thus reducing gel-to-gel variability as well as gel image analysis time. The technique is especially useful when a single experimental comparison is involved (72). Despite this, DIGE is not without limitations, such as selective labelling of proteins by one of the dyes, but this can be determined by comparing images of homogenous samples stained with each dye separately (76). More importantly it is essential each individual sample contains equal amounts of protein (76). Protein labelling efficiency is also critical, i.e. on ice, a buffer pH of 8–8.5, a minimum protein concentration of 5–10 mg/mL (minimum = 1 mg/mL), and the dye: protein ratio adjusted so each protein is labelled with a single dye molecule (72). Furthermore some protein solubilisation buffer reagents (e.g. Triton-X) can affect Cy labelling and nucleic acids should also be removed (76). However the major significant disadvantage of DIGE is sensitivity. Protein detection by DIGE requires approximately 2–10 ng protein spot−1 (i.e. less sensitive than SYPRO and silver staining) and the use of pooled samples can result in low abundance proteins, particularly those present in only a few individual samples, being diluted out (76). 2.8. Radio-Labelling and Protein Detection
Like staining, the methods of protein detection by radio-labelling also vary in terms of sensitivity and dynamic range. Autoradiography is perhaps the most common radiological detection method since it is possible to use a number of isotopes (32P, 125I, 14C, and 35 S) (16). However detection (i.e. exposure of the gel) can take days or, if a high degree of sensitivity is required, even weeks (16, 77), and another disadvantage is the low dynamic range. Both fluorography and phosphorimaging are more sensitive than autoradiography (12), and both work on the principle of detecting photons of light produced as a result of radio-labelling. With fluorography a scintillant is incorporated into the gel which is then exposed at sub-zero temperatures. With phosphorimaging the X-ray film is replaced with a europium salt phosphor screen (78). Luminescence is then proportional to the amount of radiation, which is in turn proportional to the amount of protein. Apart from increased sensitivity phosphorimaging also has a significantly greater dynamic range. However resolution is not as precise (12), making the visualisation of adjacent proteins more difficult. A radio-labelling/phosphorimaging technique which is analogous to DIGE is differential gel exposure (difExpo) (79). Proteins pre-labelled with 14C or 3H are mixed, run on 2DE, and then transferred to a PVDF membrane. Two phosphorimagers are then used, one sensitive to both isotopes and the other to 3H only. As is the case with DIGE, since these images are from the
Two-Dimensional Electrophoresis: An Overview
11
same gel they are perfectly superimposable and not susceptible to variability between gels.
3. TwoDimensional Gel Limitations and Challenges for Improving Proteomic Analysis
Proteomics, and therefore 2DE, is a discovery, not hypothesisbased discipline (3). Indeed this is perhaps the most fundamentally important feature of this undoubtedly powerful technique. However 2DE gels are limited by the amount of protein which can be loaded. Since the copy numbers of individual proteins can vary by many orders of magnitude it must be accepted that only a proportion of the complete proteome can therefore be resolved on a single gel. This is compounded by difficulty in resolving certain protein types (3, 12). In addition there is an inherent risk of error due to gel-to-gel variation. All have implications in terms of the final gel accurately reflecting the proteome, but selective protein resolution is probably the most significant. Apart from proteins present in low copy numbers, which are obviously less likely to be resolved, three protein groups are notoriously problematic: highly alkaline, extremely high and low molecular sizes, and membrane-bound proteins. Large proteins (150 kDa) tend to be hydrophobic making both solubilisation and IPG entry more difficult. However protein absorption by the pH gradient can be improved by “active re-swelling”, i.e. applying a low voltage (30–50 V) to the IPG strip during IPG rehydration (43, 80). Increasing the equilibration stage also improves transfer to the SDS–PAGE gel (12). Membrane-bound proteins, although not necessarily large, are also hydrophobic. Therefore effective solubilisation is again a major concern. In addition membrane proteins invariably have lower copy numbers than many cytoplasmic proteins which means they are poorly represented on two-dimensional gels. However, collectively, membrane proteins account for approximately 30% of the proteome. Therefore improving their inclusion on two-dimensional gels is regarded as the single largest concern for proteomic research (12). Despite this there are relatively few effective methods for collecting membrane proteins (reviewed in 81) although incorporating a primary solubilisation stage, before the use of a urea-based solubilisation buffer, has been shown to increase membrane protein yield (82). In general methods aimed at enhancing these more difficult-to-resolve proteins are based on improved protein solubilisation, pre-fractionation of protein groups of interest prior to 2DE, or adjustments to the 2DE regime.
12
Smith
TCA precipitation can be used to increase the proportion of alkaline proteins (30, 31, 80). Alkaline-protein-enriched samples can then be focussed using narrow (alkaline) pH range IEF (28, 29) (see later). Hydrophobic and membrane protein solubility is improved by the use of thiourea (83) or by changing the detergent. SDS is particularly suitable, but its incompatibility with IEF negates this advantage. However the zwitterionic detergent amidosulphobetaine is a useful alternative, especially when used in combination with a urea/thiourea mixture (84). More recently, use of the detergent 3-(4-heptyl)phenyl-3-hydroxypropyl) dimethylammoniopropanesulphonate (C7BzO), in combination with thiourea, in the solubilisation buffer and the use of increased SDS concentration in the equilibration buffer has been shown to significantly increase the number of alkaline proteins resolved from plant leafs (85). Pre-fractionation methods include protein separation by an additional electrophoresis procedure, chromatography, sequential extraction using a series of solubilisation buffers, centrifugation using gradients of alkaline buffers, or affinity purification (reviewed in 1, 12). Examples include the enrichment of low abundance proteins (26, 86), low molecular mass proteins (87), and membrane proteins (88–92), and proteins from specific organelles such as the endoplasmic reticulum (92). The alternative to enriching lower abundant proteins is to eliminate or deplete high abundance proteins. For example, commercially available albumin-removal kits are particularly useful for investigating tissue samples which have not been perfused prior to protein solubilisation. Adjustments to the 2DE regime, which are aimed at increasing the amount of the proteome resolved on the gel, tend to involve the IEF stage. Use of non-linear, or flattened, pH gradients allows greater resolution over selected pH ranges whilst still allowing the retention of a full range. This type of IPG strip is particularly useful for improving resolution of alkaline proteins (31), although the pH range can equally be extended within the acidic region (93). Extending the entire IPG separation distance also improves resolution across the entire pH range (94). An alternative approach is to use a series of full-length IPG strips, each containing a very limited pH range, i.e. IPGs of only 1 or 1.5 pH units. The final gel images are then electronically combined. These so-called “zoom” or “ultra-zoom” gels ( 80, 95–97) permit a larger amount of total protein to be resolved, which greatly increases the resolution of lower abundance proteins (46). Improving zoom gel resolution is increased further by combining this methodology with selective pre-fractionation (98). Thus the technology exists to optimise 2DE gels for most applications. The use of these specific methodologies and the appreciation of the general principles and techniques of 2DE are likely to maintain this technique as a leading contributor to proteomic research.
Two-Dimensional Electrophoresis: An Overview
13
References 1. Stochaj, W. R., Berkelman, T., and Laird, N. (2003) Preparative two-dimensional gel electrophoresis with immobilised pH gradients, in Proteins and Proteomics, A Laboratory Manual (Simpson, R. J., ed.), Cold Spring Harbor, New York, pp. 143–218 2. Graves, P. R., and Haystead, T. A. (2003) Proteomics and the Molecular Biologist. In Handbook of Proteomic Methods (Conn, P. M., ed.), Humana, New Jersey, pp. 3–16 3. Speicher, D. W. (2004) Overview of Proteome Analysis. In Proteome analysis: Interpreting the Genome (Speicher, D. W., ed.), Elsevier, Amsterdam, pp. 19–73 4. Pandey, A., and Mann, M. (2000) Proteomics to study genes and genomes. Nature 405, 837–846 5. Anderson, L., and Seilhamer, J. (1997) A comparison of selected mRNA and proteins abundance in human liver. Electrophoresis 18, 53–537 6. Gygi, S. P., Rochon, Y., Franza, B. R., and Abersold, R. (1999) Correlation between protein and mRNA abundance in yeast. Mol. Cell. Biol. 19, 1720–1730 7. Godley, A. A., and Packer, N. M. (1997) The Importance of protein co-and post-translational modifications in Proteome Projects. In Proteome Research: New Frontiers in Functional Genomics (Wilkins, M. R., Williams, K. L., Appel, R. D., and Hochstrasser, D. F., eds.), Springer, Berlin, pp. 65–91 8. O’Farrell, P. H. (1975) High resolution twodimensional electrophoresis of proteins. J. Biol. Chem. 250, 4007–4021 9. Klose, J. (1975) Protein mapping by combined isoelectric focussing and electrophoresis of mouse tissues. A novel approach to testing for induced point mutations in mammals. Humangenetik 26, 231–243 10. Fichmann, J., and Westermeier, R. (1998) 2-D protein gel electrophoresis. An overview. In Methods in Molecular Biology, Vol 112: 2-D Proteome Analysis Protocols (Link, A. J., ed), Humana Press inc, New Jersey 11. Görg, A., Drews, O., and Weiss, W. (2004) Separation of proteins using two-dimensional electrophoresis, in Purifying Proteins for Proteomics, A Laboratory Manual (Simpson, R. J., ed.), Cold Spring Harbor, New York, pp. 391–430 12. Görg, A., and Wiess, W. (2004) Protein profile comparisons of microorganisms, cells and tissues using 2D gels, in Proteome Analysis.
13.
14.
15.
16. 17.
18.
19.
20.
21.
22.
23.
Interpreting the Genome (Speicher, D. W., ed.), Elsevier, Amsterdam, pp. 19–73 Twyman, R. M. (ed.) (2004) Principles of Proteomics. Garland Science / Bios scientific publishers. New York. Rabilloud, T. (1999) Solubilization of proteins in 2-D electrophoresis. An outline. In Methods in Molecular Biology, Vol 112: 2-D Proteome Analysis Protocols (Link, A. J., ed.), Humana Press inc, NJ. Dunn, M. J. (1987) Two-dimensional gel electrophoresis, in Advances in Electrophoresis, Vol. 1 (Chrambach, A., Dunn, M. J., and Radola, B. J., eds.), VCH Weinheim, New York, pp. 1–109 Dunn, M. J. (1993) Gel Electrophoresis: Proteins. BIOS, Oxford. Rabilloud, T., Adessi, C., Giraudel, A., and Lunardi, J. (1997) Improvements of the solubilization of proteins in two-dimensional electrophoresis with immobilized pH gradients. Electrophoresis 18, 307–316 Musante, L., Candiano, G., and Chiggeri, G. M. (1997) Resolution of fibronectin and other uncharacterised proteins by two-dimensional polyacrylamide electrophoresis with thiourea. J. Chromatogr. 705, 351–356 Perdew, G. H., Schaup, H. W., and Selivonchick, D. P. (1983) The use of a zwitterionic detergent in two-dimensional gel electrophoresis of trout liver microsomes. Anal. Biochem. 135, 435–455 Adessi, C., Miege, C., Albrieux, C., and Rabilloud, T. (1997) Two-dimensional electrophoresis of membrane proteins: a current challenge for immobilized pH gradients. Electrophoresis 18, 127–135 Harder, A., Wildgruber, R. Nawrocki, A., Fey, S. J., Larsen, P. H., and Görg, A. (1999) Comparison of yeast cell protein solubilization procedures for two-dimensional electrophoresis. Electrophoresis 20, 826–829 Herbert, B. R., Molloy, M. P., Gooley, A. A., Walsh, B. J., Bryson, W. G., and Williams, K. L. (1998) Improved protein solubility in twodimensional electrophoresis using tributyl phosphine as the reducing agent. Electrophoresis 19, 845–851 Rabilloud, T., Valette, C., and Lawrence, J. J. (1994) Sample application by in-gel rehydration improves the resolution of two-dimensional electrophoresis with immobilized pH gradients in the first dimension. Electrophoresis 15, 1552–1558
14
Smith
24. Guy, C. R., Philip, R., and Tan, Y. H. (1994) Analysis of cellular phosphoproteins by twodimensional electrophoresis. Applications for cell signalling in normal and cancer cells. Electrophoresis 15, 417–440 25. Flengsrud, R., and Kobro, G. (1989) A method for two dimensional electrophoresis of proteins from green plant tissue. Anal. Biochem. 177, 33–36 26. Taylor, R. S., Wu, C. C., Hays, L. G., Eng, J. K., Yates, J. R. I., and Howell, K. E. (2000) Proteomics of rat liver Golgi complex: minor proteins are identified through sequential fractionation. Electrophoresis 21, 3441–3459 27. Damerval, C., de Vienne, D., Zivy, M., and Thiellement, H. (1986) Technical improvements in two-dimensional electrophoresis increase the level of genetic variation detected in wheat-seedling proteins. Electrophoresis 7, 52–54 28. Görg, A., Obermaier, C., Boguth, G., Csordas, A., Diaz, J. J., and Madjar, J. J. (1997) Very alkaline immobilized pH gradients for two-dimensional electrophoresis of ribosomal and nuclear proteins. Electrophoresis 18, 328–337 29. Görg, A. (1999) IPG-Dalt of very alkaline proteins, in Methods in Molecular Biology: 2-D Proteome Analysis Protocols, Vol. 112, pp. 197–209 (Link, A. J., ed.), Humana, New Jersey 30. Wildgruber, R., Reil, G., Drews, O., Parlar, H., and Görg, A. (2002) Web-based twodimensional database of Saccharomyces cerevisiae proteins using immobilized pH gradients from pH 6 to pH 12 and matrix-assisted laser desorption/ionization – time of flight mass spectrometry. Proteomics 2, 727–732 31. Görg, A., Boguth, G., Obermaier, C., and Weiss, W. (1998) Two-dimensional electrophoresis of proteins in an immobilized pH 4–12 gradient. Electrophoresis 19, 1516–1519 32. Olivieri, E., Herbert, B., and Righetti, P. G. (2001) The effect of protease inhibitors on the two-dimensional electrophoresis pattern of red blood cell membranes. Electrophoresis 22, 560–565 33. Grainer, F. (1988) Extraction of plant proteins for two-dimensional electrophoresis. Electrophoresis. 9, 712–718 34. Scheele, G. A. (1975) Two-dimensional gel analysis of soluble proteins. Characterization of guinea pig exocrine pancreatic proteins. J. Biol. Chem. 250, 5375–5385 35. Bjellqvist, B., Ek, K., Righetti, P. G., Gianazza, E., Görg, A., Westmeier, R., and Postel, W. (1982) Isoelectric focussing in immobilized
36.
37.
38.
39.
40.
41.
42.
43.
44. 45.
46.
pH gradients: principles, methodologies and some applications. J. Biochem. Biophys. Methods 6, 317–339 Görg, A., Postel, W., and Günther, S. (1985) Improved horizontal two-dimensional electrophoresis with immobilized pH gradients. Electrophoresis 6, 599–604 Görg, A., Postel, W., and Günther, S. (1988) The current state of two-dimensional electrophoresis with immobilized pH gradients. Electrophoresis 9, 531–546 Corbett, J. M., Dunn, M, J., Posch, A., and Görg, A. (1994) Positional reproducibility of protein spots in two-dimensional polyacrylamide gel electrophoresis using immobilised pH gradient isoelectric focussing in the first dimension: an interlaboratory comparison. Electrophoresis 15, 1205–1211 Blomberg, A., Blomberg, L., Norbeck, J., Fey, S. J., Larsen, P. H., Roepstorff, P., et al. (1995) Interlaboratory reproducibility of yeast protein patterns analysed by immobilized pH gradient two-dimensional gel electrophoresis. Electrophoresis 16, 1935–1945 Righetti, P. G., and Bossi, A. (1997) Review. Isoelectric focusing in immobilized pH gradients: an update. J. Chromatogr. B, 699, 77–89 Righetti, P. G., Gianazza, E., and Bjellqvist, B. (1983) Minireview. Modern aspects of isoelectric focusing: two-dimensional maps and immobilized pH gradients. J. Biochem. Biophys. Methods 8, 8–108 Bjellqvist, B., Sanchez, J. C., Pasquali, C., Ravier, F., Paquet, N., Frutiger, S., et al. (1993) Micropreparative two-dimensional electrophoresis allowing the separation of samples containing milligram amounts of proteins. Electrophoresis 14, 1375–1378 Görg, A., Obermaier, C., Boguth, G., Harder, A., Scheibe, B., Wildgruber, R., and Weiss, W. (2000) The current state of two-dimensional electrophoresis with immobilized pH gradients. Electrophoresis 21, 1037–1234 Görg, A. (1991) Two-dimensional electrophoresis. Nature 349, 545–546 Görg, A., Postel, W., Weser, J., Günther, S., Strahler, S. R., Hanash, S. M., and Somerlot, L. (1987) Elimination of point streaking on silver stained two-dimensional gels by addition of iodoacetamide to the equilibration buffer. Electrophoresis 8, 122–124 Zou, X., and Speicher, D. W. (2002) Comprehensive analysis of complex proteomes using microscale solution isoelectric focussing prior to narrow pH range two-dimensional electrophoresis. Electrophoresis 2, 58–68
Two-Dimensional Electrophoresis: An Overview 47. Laemmli, U. K. (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680–685 48. Weber, K., and Osborne, M. (1969) The reliability of molecular weight determination by dodecyl sulphate-polyacrylamide gel electrophoresis. J. Biol. Chem. 244, 4406–4412 49. Shägger, H., and von Jagow, G. (1987) Tricine-sodium dodecyl-polyacrylamide gel electrophoresis for the separation of proteins in the range from 1 to 1000 kDa. Anal. Biochem. 166, 368–379 50. Rabilloud, T. (2000) Detecting proteins separated by 2-D gel electrophoresis. Anal. Chem. 72, 48A–55A 51. Rabilloud, T., and Charmont, S. (2000) Detection of proteins on two-dimensional electrophoresis gels. In Proteome Research: Two Dimensional Electrophoresis and Identification Methods (Rabilloud, T., ed.), Springer, Berlin, pp. 107–126 52. Patton, W. F. (2002) Detection technologies in proteome analysis. J. Chromatogr. B. Anal. Technol. Biomed. Life Sci. 771, 3–31 53. Neuhoff, V., Arold, N., Taube, N., and Ehrhardt, W. (1988) Improved staining of proteins in polyacrylamide gels including isoelectric gels with clear background at nanogram sensitivity using Coomassie Brilliant Blue G-250 and R-250. Electrophoresis 9, 255–262 54. Lopez, M. F., Berggren, K., Chernokalskaya, E., Lazarev, A., Robinson, M., and Patton, W. (2000) A comparison of silver stain and SYPRO ruby protein gel stain with respect to protein detection in two-dimensional gels and identification by peptide mass profiling. Electrophoresis 21, 3673–3683 55. Gatlin, C. L., Kleeman, G. R., Hays, L. G., Link, A. J., and Yates, J. R. III (1998) Protein identification at the low femtomole level form silver-stained gels using and new fruitless electrospray interface for liquid chromatography microspray and nanospray mass spectrometry. Anal. Biochem. 263, 93–101 56. Jensen, O. N., Mortensen, P., Vorm, O., and Mann, M. (1997) Automation of matrixassisted laser desorption/ionization mass spectrometry using fuzzy logic feedback control. Anal. Chem. 69, 1706–1714 57. Castellanos-Serra, L., Proenza, W., Huerta, V., Moritz, R. L., and Simpson, R. L. (1999) Proteome analysis of polyacrylamide gel-separated proteins visualized by reversible negative staining using imidazole-zinc salts. Electrophoresis 20, 732–737 58. Switzer, R. C., Merril, C. R., and Shifrin, S. (1979) A highly sensitive silver stain for
59.
60.
61.
62.
63.
64.
65.
66.
67.
68.
69.
15
detecting proteins and peptides in polyacrylamide gels. Anal. Biochem. 98, 321–327 Rabilloud, T., Vullard, L., Gilly, C., and Lawrence, J. J. (1994) Silver staining of proteins in polyacrylamide gels: a general overview. Cell. Mol. Biol. 40, 57–75 Merril, C. R., Goldman, D., Sedman, S. A., and Ebert, M. H. (1981) Ultrasensitive stain for proteins in polyacrylamide gels shows regional variation in cerebrospinal fluid proteins. Science 211, 1437–1438 Oakley, B. R., Kirsch, D. R., and Morris, N. R. (1980) A simplified ultrasensitive silver stain for detecting proteins in polyacrylamide gels. Anal. Biochem. 105, 361–363 Shevenko, A., Wilm, M., Vorm, O., and Mann, M. (1996) Mass spectrometric sequencing of protein silver-stained polyacrylamide gels. Anal. Chem. 68, 850–858 Mortz, E., Krogh, T. N., Vorum, H., and Görg, A. (2001) Improved silver staining protocols for high sensitivity protein identification using matrix-assisted laser desorption/ ionization time of flight analysis. Proteomics 1, 359–363 Yan, J. X., Walt, R., Berkelman, T., Harry, R. A., Westbrook, J. A., Wheeler, C. H., and Dunn, M. J. (2000) A modified silver staining protocol for visualization of proteins compatible with matrix-assisted laser desorption/ionization and electrospray mass spectrometry. Electrophoresis 21, 3666–3672 Scheler, C., Lamer, S., Pan, Z., Li, X., Salinkow, J., and Jungblut, P. (1998) Peptide mass fingerprint sequence coverage from differently stained proteins on two-dimensional electrophoresis patterns by matrix assisted laser desorption/ ionization – mass spectrometry (MALDI-MS). Electrophoresis 19, 918–927 Gharahdaghi, F., Weinberg, C. R., Meagher, D. A., Imai, B. S., and Mische, S. M. (1999) Mass spectrometric identification of proteins form silver-stained polyacrylamide gels: a method for the removal of silver ions to enhance sensitivity. Electrophoresis 20, 601–605 Sinha, P., Poland, J., Schnölzer, M., and Rabilloud, T. (2001) A new silver staining apparatus and procedure for matrix/ionization – time of flight analysis of proteins after two-dimensional electrophoresis. Proteomics 1, 835–840 Patton, W. F., Lim, M. J., and Shepro, D. (1999) Image acquisition in 2-D electrophoresis. Methods Mol. Biol. 112, 353–362 Celis, J. E., and Gromov, P. (1999) 2D protein electrophoresis: can it be perfected? Curr. Opin. Biotechnol. 10, 16–21
16
Smith
70. Sanchez, J-C., Chiappe, D., Converset, V., Hoogland, C, Binz, P-A., Appel, R. D., et al. (2001) The mouse SWISS-2D PAGE database: a tool for proteomics of diabetes and obesity. Proteomics 1, 136–163 71. Ünlü, M., Morgan, M. E., and Minden, J. S. (1997) Difference gel electrophoresis: a single gel method for detecting changes in protein extracts. Electrophoresis 18, 2071–2077 72. Tonge, R., Shaw, J., Middleton, B., Rowlinson, R., Rayner, S., Young, J., et al. (2001) Validation and development of fluorescence two-dimensional differential gel electrophoresis proteomics technology. Proteomics 1, 377–396 73. Zhou, G., Li, H., DeCamp, D., Chen, S., Shu, H., Gong, Y., Flaig, M., et al. (2002) 2-D differential in-gel electrophoresis for the identification of esophageal scans cell cancerspecific protein markers. Mol. Cell. Proteom. 1, 117–123 74. Gharbi, S., Gaffney, P., Yang, A., Zvelebil, M. J., Cramer, R., Waterfield, M. D., and Timms, J. F. (2002) Evaluation of two-dimensional differential gel electrophoresis for protein expression analysis of a model breast cancer cell system. Mol. Cell. Proteom. 1, 91–98 75. Alban, A., David, S. O., Bjorksten, L., Anderson, C., Sloge, E., Lewis, S., and Currie, I. (2003) A novel experimental design for comparative two-dimensional gel analysis: two-dimensional gel electrophoresis incorporating a pooled internal standard. Proteomics 3, 36–44 76. Hanlon, W. A., and Griffith, P. R. (2004) Protein profiling using two-dimensional gel electrophoresis with multiple fluorescent tags. In Proteome Analysis: Interpreting the Genome (Speicher, D. W., ed.), Elsevier, Amsterdam, pp. 19–73 77. Link, A. J. (1999) Autoradiography of 2D gels. Methods Mol. Biol. 112, 285–290 78. Patterson, S. D., and Latter, G. I. (1993) Evaluation of storage phosphor imaging for quantitative analysis of 2D gels using the Quat II system. Biotechniques 15, 1076–1083 79. Monribot-Espagne, C., and Boucherie, H. (2002) Differential gel exposure, a new methodology for the two-dimensional comparison of protein samples. Proteomics 2, 229–240 80. Görg, A., Obermaier, C., Boguth, G., and Weiss, W. (1999) Recent developments in two-dimensional gel electrophoresis with immobilized pH gradients: wide pH gradients up to pH 12, longer separation distances and simplified procedures. Electrophoresis 20, 712–717
81. Santoni, V., Rabilloud, T., Doumas, P., Rouqui, D., Mansion, M., Kieffers, S., et al. (1999) Towards the recovery of hydrophobic proteins on two-dimensional electrophoresis gels. Electrophoresis 20, 705–711 82. Schüpbach, J., Ammann, R. W., and Freiburghaus, A. U. (1991). A universal method for two-dimensional polyacrylamide gel electrophoresis of membrane proteins using isoelectric focusing on immobilized pH gradients in the first dimension. Anal. Biochem. 196, 337–343 83. Rabilloud, T. (1998) The use of thiourea to increase the solubility of membrane proteins in two-dimensional electrophoresis. Electrophoresis 19, 758–760 84. Chevallet, M., Santoni, V., Poinas, A., Rouquie, D., Fuchs, A., Kieffer, S., et al. (1998) New zwitterionic detergents improve the analysis of membrane proteins by two-dimensional electrophoresis. Electrophoresis 19, 1901–1090 85. Maserti, B. E., Della Croce, C. M., Luro, F., Morillon, R., Cini, M., and Caltavuturo L. (2007) A general method for the extraction of citrus leaf proteins and separation by 2D electrophoresis: A follow up. J. Chromatogr. B 849, 351–356 86. Fountoulakis, M., Juranville, J. F., Roder, D., Evers, S., Berndt, P., and Langen, H. (1998) Reference map of the low molecular mass proteins of Haemophilus influenzae. Electrophoresis 19, 1819–1827 87. Fountoulakis, M., Berndt, P., Langen, H., and Suter, L. (2007) The rat liver mitochondria proteins. Electrophoresis. 23, 311–328 88. Molloy, M. P., Herbert, B. R., Walsh, B. J., Tyler, M. I., Traini, M., Sanchez, J-C., et al. (1998) Extraction of membrane proteins by differential solubilisation for separation using twodimensional electrophoresis. Electrophoresis 19, 837–844 89. Nouwens, A. S., Cordwell, S. J., Larsen. M. R., Molloy, M. P., Gillings, M., Willcox, M. D., and Walsh, B. J. (2000) Complementary genomics with proteomics: the membrane bproteome of Pseudomonas aeroginosa PAO 1. Electrophoresis 21, 3797–3809 90. Molloy, M. P., Herbert, B. R., Williams, K. L., and Gooley, A. A. (1999) Extraction of Escherichia coli proteins with organic solvents prior to two-dimensional electrophoresis. Electrophoresis 20, 701–704 91. Ferro, M., Seigneurin-Berny, D., Rolland, N., Chapel, A., Salvi, D., Garin, J., and Joyard, J. (2000) Organic solvent extraction as a versatile procedure to identify hydrophobic chloroplast membrane proteins. Electrophoresis 21, 3517–3526
Two-Dimensional Electrophoresis: An Overview 92. Fujiki , Y. , Hubbard , A. L. , Fowler, S. , and Lazarow, P. B. (1982) Isolation of intracellular preparations by means of sodium carbonate treatment: applications to endoplasmic reticulum . J. Cell Biol. 93 , 97 – 102 93. Bjellqvist, B., Sanchez, J. C., Pasquali, C., Ravier, F., Pasquet, N., Frutiger, S., et al. (1993) A nonlinear wide-range immobilised pH gradient for two-dimensional electrophoresis and its definition in a relevant pH scale. Electrophoresis 14, 1357–1365 94. Poland, J., Cahill, M. A., and Sinha, P. (2003) Isoelectric focussing in long immobilized pH gradient gels to improve protein separation and proteome analysis. Electrophoresis 24, 1271–1275 95. Hoving, S., Voshol, H., and van Oostrum, J. (2000) Towards high performance two-
17
dimensional electrophoresis using ultrazoom gels. Electrophoresis 21, 2617–2621 96. Sabounchi-Schutt, F., Astrom, J., Eklund, A., Grunewald, J., and Bellqvist, B. (2001) Detection and identification of human bronchoalveolar lavage proteins using narrowrange immobilized pH gradient Drystrips and paper bridge sample application method. Electrophoresis 22, 1851–1860 97. Westbrook, J. A., Yan, J. X., Wait, R., Welson, S. Y., and Dunn, M. J. (2001) Zooming-in on the proteome: very narrow-range immobilised pH gradients reveal more protein species and isoforms. Electrophoresis 22, 2865–2871 98. Görg, A., Boguth, G., Köpf, A., Reil, G., Parlar, H., and Weiss, W. (2002) Sample preparation with Sephadex isoelectric focusing prior to narrow pH range two-dimensional gels. Proteomics 2, 1652–1657
Chapter 2 Solubilization of Proteins in 2DE: An Outline Thierry Rabilloud Summary Protein solubilization for two-dimensional electrophoresis (2DE) has to break molecular interactions to separate the biological contents of the material of interest into isolated and intact polypeptides. This must be carried out in conditions compatible with the first dimension of 2DE, namely isoelectric focusing. In addition, the extraction process must enable easy removal of any nonprotein component interfering with the isoelectric focusing. The constraints brought in this process by the peculiar features of isoelectric focusing are discussed, as well as their consequences in terms of possible solutions and limits for the solubilization process. Key words: Proteins, Two-dimensional electrophoresis, Solubilization, Isoelectric focusing, Extraction, Review.
1. Introduction The solubilization process for two-dimensional electrophoresis (2DE) has to achieve four parallel goals: 1. Breaking macromolecular interactions in order to yield separate polypeptide chains. This includes denaturing the proteins to break noncovalent interactions, breaking disulfide bonds, and disrupting noncovalent interactions between proteins and nonprotein components such as lipids or nucleic acids. 2. Preventing any artifactual modification of the polypeptides in the solubilization medium. Ideally, the perfect solubilization medium should freeze all the extracted polypeptides in their exact state prior to solubilization, both in terms of amino acid
David Sheehan and Raymond Tyther (eds.), Methods in Molecular Biology, Two-Dimensional Electrophoresis Protocols, vol. 519 © Humana Press, a part of Springer Science + Business Media, LLC 2009 DOI: 10.1007/978-1-59745-281-6_2
19
20
Rabilloud
composition and in terms of posttranslational modifications. This means that all the enzymes able to modify the proteins must be quickly and irreversibly inactivated. Such enzymes include of course proteases, which are the most difficult to inactivate, but also phosphatases, glycosidases, etc. In parallel, the solubilization protocol should not expose the polypeptides to conditions in which chemical modifications (e.g., deamidation of Asn and Gln, cleavage of Asp-Pro bonds) may occur. 3. Allowing the easy removal of substances that may interfere with 2DE. In 2DE, proteins are the analytes. Thus, anything in the cell but proteins can be considered as an interfering substance. Some cellular compounds (e.g., coenzymes, hormones) are so dilute they go unnoticed. Other compounds (e.g., simple nonreducing sugars) do not interact with proteins or do not interfere with the electrophoretic process. However, many compounds bind to proteins and/or interfere with 2DE, and must be eliminated prior to electrophoresis if their amount exceeds a critical interference threshold. Such compounds mainly include salts, lipids, polysaccharides (including cell walls), and nucleic acids. 4. Keeping proteins in solution during the 2DE process. Although solubilization stricto sensu stops at the point where the sample is loaded onto the first-dimension gel, its scope can be extended to the 2D process per se, as proteins must be kept soluble until the end of the second dimension. Generally speaking, the second dimension is an SDS gel, and very few problems are encountered once the proteins have entered this. The one main problem is overloading of the major proteins when micropreparative 2DE is carried out, and nothing but scaling up the SDS gel (its thickness and its other dimensions) can counteract overloading. However, severe problems can be encountered in the isoelectric focusing (IEF) step. They arise from the fact that IEF must be carried out in low ionic strength conditions and with no manipulation of the polypeptide charge. IEF conditions give problems at three stages: (a) During the initial solubilization of the sample, important interactions between proteins of widely different pI and/ or between proteins and interfering compounds (e.g., nucleic acids) may happen. This yields poor solubilization of some components. (b) During the entry of the sample in the focusing gel, there is a stacking effect due to the transition between a liquid phase and a gel phase with a higher friction coefficient. This stacking increases the concentration of proteins and may give rise to precipitation events. (c) At, or very close to, the isoelectric point, the solubility of the proteins comes to a minimum. This can be explained
Solubilization of Proteins in 2DE: An Outline
21
by the fact that the net charge comes close to zero, with a concomitant reduction of the electrostatic repulsion between polypeptides. This can also result in protein precipitation or adsorption to the IEF matrix. Apart from breaking molecular interactions and solubility in the 2DE gel which are common to all samples, the solubilization problems encountered will greatly vary from one sample type to another, due to wide differences in the amount and nature of interfering substances and/or spurious activities (e.g., proteases). The aim of this outline chapter is not to give detailed protocols for various sample types, and the reader should refer to the chapters of this book dedicated to the type of sample of interest. I would rather like to concentrate on the solubilization rationale and to describe nonstandard approaches to solubilization problems. A more detailed review on solubilization of proteins for electrophoretic analyses can be found elsewhere (1).
2. Rationale of SolubilizationBreaking Molecular Interactions
Apart from disulfide bridges, the main forces holding proteins together and allowing binding to other compounds are noncovalent interactions. Covalent bonds are encountered mainly between proteins and some coenzymes. The noncovalent interactions are mainly ionic bonds, hydrogen bonds, and “hydrophobic interactions.” The basis for “hydrophobic interactions” is in fact the presence of water. In this very peculiar (hydrogen-bonded, highly polar) solvent, the exposure of nonpolar groups to the solvent is thermodynamically not favored compared to the grouping of these apolar groups together. Indeed, although the van der Waals forces give an equivalent contribution in both configurations, the other forces (mainly hydrogen bonds) are maximized in the latter configuration and disturbed in the former (solvent destruction). Thus, the energy balance is clearly in favor of the collapse of the apolar groups together (2). This explains why hexane and water are not miscible, and also that the lateral chain of apolar amino acids (L, V, I, F, W, Y) pack together and form the hydrophobic cores of proteins (3). These hydrophobic interactions are also responsible for some protein–protein interactions and for the binding of lipids and other small apolar molecules to proteins. The constraints for a good solubilization medium for 2DE are therefore to be able to break ionic bonds, hydrogen bonds, hydrophobic interactions, and disulfide bridges under conditions compatible with IEF, i.e., with very low amounts of salt or other charged compounds (e.g., ionic detergents).
22
Rabilloud
2.1. Disruption of Disulfide Bridges
Breaking of disulfide bridges is usually achieved by adding to the solubilization medium an excess of a thiol compound. Mercaptoethanol was used in the first 2D protocols (4), but its use does have drawbacks. Indeed, a portion of the mercaptoethanol will ionize at basic pH, enter the basic part of the IEF gel, and ruin the pH gradient in its alkaline part because of its buffering power (5). Although its pK is around 8, dithiothreitol (DTT) is much less prone to this drawback, as it is used at much lower concentrations (usually 50 mM instead of the 700 mM present in 5% mercaptoethanol). However, DTT is still not the perfect reducing agent. Some proteins of very high cysteine content or with cysteines of very high reactivity are not fully reduced by DTT. In these cases, phosphines are very often an effective answer. First, the reaction is stoichiometric, which allows us in turn to use very low concentrations (a few mM). Second, these reagents are not as sensitive as thiols to dissolved oxygen. The most powerful compound is tributylphosphine, which was the first phosphine used for disulfide reduction in biochemistry (6). However, this reagent is volatile, toxic, has a rather unpleasant odor, and needs an organic solvent to make it water miscible. In the first uses of the reagent, propanol was used as a carrier solvent at rather high concentrations (50%) (6). It was however found that DMSO or DMF are suitable carrier solvents, which enable the reduction of proteins by 2 mM tributylphosphine (7). All these drawbacks have disappeared with the introduction of a water-soluble phosphine, tris (carboxyethyl) phosphine, for which 1 M aqueous stock solutions can be easily prepared and stored frozen in aliquots.
2.2. Disruption of Noncovalent Interactions
The perfect way to disrupt all types of noncovalent interactions would be the use of a charged compound that disrupts hydrophobic interactions by providing a hydrophobic environment. The hydrophobic residues of the proteins would be dispersed in that environment and not clustered together. This is just the description of SDS, and this explains why SDS has been often used in the first stages of solubilization (8–11). However, SDS is not compatible with IEF and must be removed from the proteins during IEF (see later). The other way of breaking most noncovalent interactions is the use of a chaotrope. It must be kept in mind that all the noncovalent forces keeping molecules together must be taken into account with a comparative view on the solvent. This means that the final energy of interaction depends on the interaction per se and on its effects on the solvent. If the solvent parameters are changed (dielectric constant, hydrogen bond formation, polarizability, etc.), all the resulting energies of interaction will change. Chaotropes, which alter all the solvent parameters, exert profound effects on all types of interactions. For example, by changing the hydrogen bond structure of the solvent, chaotropes disrupt
Solubilization of Proteins in 2DE: An Outline
23
hydrogen bonds but also decrease the energy penalty for exposure of apolar groups and therefore favor the dispersion of hydrophobic molecules and the unfolding of the hydrophobic cores of a protein (12). Protein unfolding will also greatly decrease ionic bonds between proteins, which are very often not very numerous and highly dependent of the correct positioning of the residues. As the gross structure of proteins is driven by hydrogen bonds and hydrophobic interactions, chaotropes decrease dramatically ionic interactions both by altering the dielectric constant of the solvent and by denaturing the proteins, so that the residues will no longer be positioned correctly. Nonionic chaotropes, as those used in 2DE, however, are unable to disrupt ionic bonds when high charge densities are present (e.g., histones, nucleic acids) (13). In this case, it is often quite advantageous to modify the pH and to take advantage of the fact that the ionizable groups in proteins are weak acids and bases. For example, increasing the pH to 10 or 11 will induce most proteins to behave as anions, so that ionic interactions present at pH 7 or lower turn into electrostatic repulsion between the molecules, thereby promoting solubilization. The use of a high pH results therefore in dramatically improved solubilizations, with yields very close to what is obtained with SDS (14). The alkaline pH can be obtained either by addition of a few mM of potassium carbonate to the urea–detergent–ampholytes solution (14), by the use of alkaline ampholytes (11), or by the use of a spermine-DTT buffer which allows better extraction of nuclear proteins (15). For 2DE, the chaotrope of choice is urea. Although urea is less efficient than substituted urea in breaking hydrophobic interactions (12), it is more efficient in breaking hydrogen bonds, so that its overall solubilization power is greater. However, denaturation by urea induces the exposure of the totality of the proteins hydrophobic residues to the solvent. This increases, in turn, the potential for hydrophobic interactions so that urea alone is often not sufficient to quench completely the hydrophobic interactions especially when lipids are present in the sample. This explains why detergents, which can be viewed as specialized agents for hydrophobic interactions, are almost always included in the urea-based solubilization mixtures for 2DE. Detergents act on hydrophobic interactions by providing a stable dispersion of a hydrophobic medium in the aqueous medium, through the presence of micelles for example. Therefore, hydrophobic molecules (e.g., lipids) are no longer collapsed in the aqueous solvent but will disaggregate in the micelles, provided the amount of detergent is sufficient to ensure maximal dispersion of the hydrophobic molecules. Detergents have polar heads that are able to contract other types of noncovalent bonds (hydrogen bonds, salt bonds for charged heads, etc.). The action of detergents is the sum of the dispersive
24
Rabilloud
effect of the micelles on hydrophobic part of the molecules and the effect of their polar heads on the other types of bonds. This explains why various detergents show very variable effects varying from a weak and often incomplete delipidation (e.g., Tweens) to a very aggressive action where the exposure of the hydrophobic core in the detergent-containing solvent is no longer energetically unfavored and leads to denaturation (e.g., SDS). Of course, detergents used for IEF must bear no net electrical charge, and only nonionic and zwitterionic detergents may be used. However, ionic detergents such as SDS may be used for the initial solubilization, prior to isoelectric focusing, in order to increase solubilization and facilitate the removal of interfering compounds. Low amounts of SDS can be tolerated in the subsequent IEF (10) provided that high concentrations of urea (16) and nonionic (10) or zwitterionic detergents (17) are present to ensure complete removal of the SDS from the proteins during IEF. Higher amounts of SDS must be removed prior to IEF, by precipitation (9), for example. It must therefore be kept in mind that SDS will only be useful for solubilization and for sample entry, but will not cure isoelectric precipitation problems. The use of nonionic or zwitterionic detergents in the presence of urea presents some problems due to the presence of urea itself. In concentrated urea solutions, urea is not freely dispersed in water but can form organized channels (18). These channels can bind linear alkyl chains, but not branched or cyclic molecules, to form complexes of undefined stoichiometry called inclusion compounds. These complexes are much less soluble than the free solute, so that precipitation is often induced upon formation of the inclusion compounds, precipitation being stronger with increasing alkyl chain length and higher urea concentrations. Consequently, many nonionic or zwitterionic detergents with linear hydrophobic tails (19, 20) and some ionic ones (21) cannot be used in the presence of high concentrations of urea. This limits the choice of detergents mainly to those with nonlinear alkyl tails (e.g., Tritons, Nonidet P40, CHAPS) or with short alkyl tails (e.g., octyl glucoside), which are unfortunately less efficient in quenching hydrophobic interactions. Sulfobetaine detergents with long linear alkyl tails have however received limited applications, as they require low concentrations of urea. Good results have been obtained in certain cases for sparingly soluble proteins (22–24), although this type of protocol seems rather delicate owing to the need for a precise control of all parameters to prevent precipitation. Apart from the problem of inclusion compounds, the most important problem linked with the use of urea is carbamylation. Urea in water exists in equilibrium with ammonium cyanate, the level of which increases with increasing temperature and pH (25). Cyanate can react with amines to yield substituted urea.
Solubilization of Proteins in 2DE: An Outline
25
In the case of proteins, this reaction takes place with the α-amino group of the N-terminus and the ε-amino groups of lysines. This reaction leads to artifactual charge heterogeneity, N-terminus blocking, and adduct formation detectable in mass spectrometry. Carbamylation should therefore be completely avoided. This can be easily made with some simple precautions. The use of a pure grade of urea (p.a.) decreases the amount of cyanate present in the starting material. Avoidance of high temperatures (never heat urea-containing solutions above 37°C) considerably decreases cyanate formation. In the same way, urea-containing solutions should be stored frozen (−20°C) to limit cyanate accumulation. Last but not least, a cyanate scavenger (primary amine) should be added to urea-containing solutions. In the case of IEF, carrier ampholytes are perfectly suited for this task. If these precautions are correctly taken, proteins seem to withstand long exposures to urea without carbamylation (26).
3. Solubility During IEF Additional solubility problems often arise during the IEF at sample entry and solubility at the isoelectric point. 3.1. Solubility During Sample Entry
Sample entry is often quite critical. In most 2DE systems, sample entry in the IEF gel corresponds to a transition between a liquid phase (the sample) and a gel phase of higher friction coefficient. This induces a stacking of the proteins at the sample-gel boundary, which results in very high concentration of proteins at the application point. These concentrations may exceed the solubility threshold of some proteins, thereby inducing precipitation and sometimes clogging of the gel, with poor penetration of the bulk of proteins. Such a phenomenon is of course more prominent when high amounts of proteins are loaded onto the IEF gel. The sole simple but highly efficient remedy to this problem is to include the sample in the IEF gel. This process abolishes the liquid–gel transition and decreases the overall protein concentration, as the volume of the IEF gel is generally much higher than the one of the sample. This process is however rather difficult for tube gels in carrier ampholyte-based IEF. The main difficulty arises from the fact that the thiol compounds used to reduce disulfide bonds during sample preparation are strong inhibitors of acrylamide polymerization, so that conventional samples cannot be used as such. Alkylation of cysteines and of the thiol reagent after reduction could be a solution, but many neutral alkylating agents (e.g., iodoacetamide, N-ethyl maleimide) also inhibit acrylamide
26
Rabilloud
polymerization. Owing to this situation, most workers describing inclusion of the sample within the IEF gel have worked with nonreduced samples (27, 28). Although this presence of disulfide bridges is not optimal, inclusion of the sample within the gel has proven of great but neglected interest (27, 28). It must however be pointed out that it is now possible to carry out acrylamide polymerization in an environment where disulfide bridges are reduced. The key is to use 2 mM tributylphosphine as the reducing agent in the sample and using tetramethylurea as a carrier solvent. This ensures total reduction of disulfides and is totally compatible with acrylamide polymerization with the standard Temed/persulfate initiator (T. Rabilloud, unpublished results). This modification should help the experimenters trying sample inclusion within the IEF gel when high amounts of proteins are to be separated by 2DE. The process of sample inclusion within the IEF gel is however much simpler for IPG gels. In this case, rehydration of the dried IPG gel in a solution containing the protein sample is quite convenient and efficient, provided that the gel has a sufficiently open structure to be able to absorb proteins efficiently (15). Coupled with the intrinsic high capacity of IPG gels, this procedure enables to easily separate milligram amounts of protein (15). 3.2. Solubility at the Isoelectric Point
This is usually the second critical point for IEF. The isoelectric point is the pH of minimal solubility, mainly because the protein molecules have no net electrical charge. This abolishes the electrostatic repulsion between protein molecules, which maximizes in turn protein aggregation and precipitation. The horizontal comet shapes frequently encountered for major proteins and for sparingly soluble proteins often arise from such a near-isoelectric precipitation. Such isoelectric precipitates are usually easily dissolved by the SDS solution used for the transfer of the IEF gel onto the SDS gel, so that the problem is limited to a loss of resolution, which however precludes the separation of high amounts of proteins. The problem is however more severe for hydrophobic proteins when an IPG is used. In this case, a strong adsorption of the isoelectric protein to the IPG matrix seems to occur, which is not reversed by incubation of the IPG gel in the SDS solution. The result is severe quantitative losses, which seem to increase with the hydrophobicity of the protein and the amount loaded (29). The sole solution to this serious problem is to increase the protein solubilizing power of the medium used for IEF, by acting both on the chaotrope and on the detergent. As to the chaotrope, it has been shown that using a mixture of urea and thiourea increases protein solubility (30). On a molar basis, thiourea has been shown to be a much stronger denaturant than urea itself (31). Thiourea alone is weakly soluble in
Solubilization of Proteins in 2DE: An Outline
27
water (ca 1 M), so that it cannot be used as the sole chaotrope. However, thiourea is more soluble in concentrated urea solutions (31). Consequently, urea–thiourea mixtures (typically 2 M thiourea and 5–8 M urea, depending on the detergent used) exhibit a superior solubilizing power and are able to increase dramatically the solubility of membrane or nuclear proteins in IPG gels as well as protein transfer to the second-dimension SDS gel (30). The benefits of using thiourea–urea mixtures to increase protein solubility can be transposed to conventional, carrier ampholyte-based focusing in tube gels with minor adaptations. Thiourea strongly inhibits acrylamide polymerization with the standard TEMED/persulfate system. However, photopolymerization with methylene blue, sodium toluene sulfinate, and diphenyl iodonium chloride (32) enables acrylamide polymerization in the presence of 2 M thiourea without any deleterious effect in the subsequent 2DE (33) so that higher amounts of proteins can be loaded without loss of resolution (33). As to the detergent, considerable interest has been shown in this field due to its potential application for the solubilization of membrane proteins (34). It must be kept in mind, however, that the detergents used in denaturing IEF must work in high concentrations of urea. On the one hand, this poses the problem of inclusion compounds, as described earlier. On the other, this highly chaotropic mixture changes dramatically the detergent aggregations parameters (critical micellar concentration, critical micellar temperature) and thus the detergent properties. This can be favorable in some cases, e.g., with deoxychaps which cannot be used in water alone due to its high critical micellar temperature (55°C), while it can be used in 8 M urea where it is fully soluble at room temperature (35). Investigations in the field of detergents for denaturing IEF have concerned the two families that are compatible with IEF, namely zwitterionic detergents and nonionic ones. Sulfobetaines with various hydrophobic parts and/or linkers between the hydrophobic and hydrophilic parts have been synthesized and tested (35–37). Some of them have shown interesting solubilizing properties, such as ASB14 and C7BzO, and are now commercially available. Besides work on this particular detergent family, there has been a renewal of interest in non-ionic detergents, and some of them have been shown to be able to solubilize membrane proteins (38–40). From this, the importance of the detergent/chaotrope couple is clearly highlighted. For example, Triton X100 has been used in conjunction with urea since the very beginning of 2DE and has not been shown to solubilize any membrane protein. However, when Triton X100 is used with a urea/thiourea mixture, it has been shown to solubilize some membrane proteins efficiently (38).
28
Rabilloud
4. Concluding Remarks Although this outline chapter has mainly dealt with the general aspects of solubilization, the main concluding remark is that there is no universal solubilization protocol. Standard urea-reducerdetergent mixtures usually achieve disruption of disulfide bonds and noncovalent interactions. Consequently, the key issues for a correct solubilization are removal of interfering compounds, blocking of protease action, and disruption of infrequent interactions (e.g. strong ionic bonds). These bonds will greatly depend on the type of sample used, the proteins of interest, and the amount to be separated, so that the optimal solubilization protocol can vary greatly from one sample to another. However, the most frequent bottleneck for the efficient 2DE separation of as many and as much proteins as possible does not usually lie in the initial solubilization but in keeping the solubility along the IEF step. In this field, the key feature is the disruption of hydrophobic interactions, which are responsible for most, if not all, of the precipitation phenomena encountered during IEF. This means improving solubility during denaturing IEF will focus on the quest forever more powerful chaotropes and detergents. In this respect, the use of thiourea may prove to be one of the keys to increase the solubility of proteins in 2DE. One of the other keys being the use of as powerful detergent or detergent mixture as possible. In a complex sample, some proteins may be well denatured and solubilized by a given detergent or chaotrope, while other proteins will require another detergent or chaotrope. Consequently, the future of solubilization may well be to find mixtures of detergents and chaotropes able to cope with the diversity of proteins encountered in the complex samples separated by 2DE. It must be kept in mind, however, that this protein diversity may overcome the solubilization power that is achievable with chemicals bearing no electrical charge, as in the case of IEF. When hydrophobic proteins are to be analyzed, it may be a safer approach to use ionic detergents. These have a much higher solubilizing power, as they confer a net electrical charge to the protein–detergent complexes, and the Coulombian repulsion between the protein detergent complexes prevents aggregation and promotes solubilization. The price to pay is to renounce IEF and to use electrophoresis schemes of much lower resolution (41). However, such electrophoresis schemes using only ionic detergents have been shown to be able to deal with very hydrophobic proteins (42).
Solubilization of Proteins in 2DE: An Outline
29
References 1. Rabilloud, T. (1996) Solubilization of proteins for electrophoretic analyses. Electrophoresis 17, 813–829 2. Tanford, C. (1980) The Hydrophobic Effect, 2nd edn, Wiley, New York 3. Dill, K.A. (1985) Theory for the folding and stability of globular proteins. Biochemistry 24, 1501–1509 4. O’Farrell, P.H. (1975) High resolution twodimensional electrophoresis of proteins. J. Biol. Chem. 250, 4007–4021 5. Righetti, P.G., Tudor, G., and Gianazza, E. (1982) Effect of 2 mercaptoethanol on pH gradients in isoelectric focusing. J. Biochem. Biophys. Methods 6, 219–227 6. Ruegg, U.T., and Rüdinger, J. (1977) Reductive cleavage of cystine disulfides with tributylphosphine. Methods Enzymol. 47, 111–116 7. Kirley, T.L. (1989) Reduction and fluorescent labeling of cyst(e)ine containing proteins for subsequent structural analysis. Anal. Biochem. 180, 231–236 8. Wilson, D., Hall, M.E., Stone, G.C., and Rubin, R.W. (1977) Some improvements in two-dimensional gel electrophoresis of proteins. Anal. Biochem. 83, 33–44 9. Hari, V. (1981) A method for the two-dimensional electrophoresis of leaf proteins. Anal. Biochem. 113, 332–335 10. Ames, G.F.L., and Nikaido, K. (1976) Twodimensional electrophoresis of membrane proteins. Biochemistry 15, 616–623 11. Hochstrasser, D.F., Harrington, M.G., Hochstrasser, A.C., Miller, M.J., and Merril, C.R. (1988) Methods for increasing the resolution of two dimensional protein electrophoresis. Anal. Biochem. 173, 424–435 12. Herskovits, T.T., Jaillet, H., and Gadegbeku, B. (1970) On the structural stability and solvent denaturation of proteins. II. Denaturation by the ureas. J. Biol. Chem. 245, 4544–4550 13. Sanders, M.M., Groppi, V.E., and Browning, E.T. (1980) Resolution of basic cellular proteins including histone variants by two-dimensional gel electrophoresis: Evaluation of lysine to arginine ratios and phosphorylation. Anal. Biochem. 103, 157–165 14. Horst, M.N., Basha, M.M., Baumbach, G.A., Mansfield, E.H., and Roberts, R.M. (1980) Alkaline urea solubilization, two-dimensional electrophoresis and lectin staining of mammalian cell plasma membrane and plant seed proteins. Anal. Biochem. 102, 399–408
15. Rabilloud, T., Valette, C., and Lawrence, J.J. (1994) Sample application by in-gel rehydration improves the resolution of two-dimensional electrophoresis with immobilized pH gradients in the first dimension. Electrophoresis 15, 1552–1558 16. Weber, K., and Kuter, D.J. (1971) Reversible denaturation of enzymes by sodium dodecyl sulfate. J. Biol. Chem. 246, 4504–4509 17. Remy, R., and Ambard-Bretteville, F. (1987) Two-dimensional electrophoresis in the analysis and preparation of cell organelle polypeptides. Methods Enzymol. 148, 623–632 18. March, J. (1977) Advanced Organic Chemistry, 2nd edn, McGraw-Hill, London, pp. 83–84 19. Dunn, M.J., and Burghes, A.H.M. (1983) High resolution two-dimensional polyacrylamide electrophoresis. I. Methodological procedures. Electrophoresis 4, 97–116 20. Rabilloud, T., Gianazza, E., Catto, N., and Righetti, P.G. (1990) Amidosulfobetaines, a family of detergents with improved solubilization properties: application for isoelectric focusing under denaturing conditions. Anal. Biochem. 185, 94–102 21. Willard, K.E., Giometti, C., Anderson, N.L., O’Connor, T.E., and Anderson, N.G. (1979) Analytical techniques for cell fractions. XXVI. A two-dimensional electrophoretic analysis of basic proteins using phosphatidyl choline/ urea solubilization. Anal. Biochem. 100, 289–298 22. Clare Mills, E.N., and Freedman, R.B. (1983) Two-dimensional electrophoresis of membrane proteins. Factors affecting resolution of rat liver microsomal proteins. Biochim. Biophys. Acta 734, 160–167 23. Satta, D., Schapira, G., Chafey, P., Righetti, P.G., and Wahrmann, J.P. (1984) Solubilization of plasma membranes in anionic, non ionic and zwitterionic surfactants for iso-dalt analysis: a critical evaluation. J. Chromatogr. 299, 57–72 24. Gyenes, T., and Gyenes, E. (1987) Effect of stacking on the resolving power of ultrathin layer two-dimensional gel electrophoresis. Anal. Biochem. 165, 155–160 25. Hagel, P., Gerding, J.J.T., Fieggen, W., and Bloemendal, H. (1971) Cyanate formation in solutions of urea. I. Calculation of cyanate concentrations at different temperature and pH. Biochim. Biophys. Acta 243, 366–373 26. Bjellqvist, B., Sanchez, J.C., Pasquali, C., Ravier, F., Paquet, N., Frutiger, S., et al. (1993)
30
27.
28.
29.
30.
31.
32.
33.
34.
35.
Rabilloud Micropreparative two-dimensional electrophoresis allowing the separation of samples containing milligram amounts of proteins. Electrophoresis 14, 1375–1378 Chambers, J.A.A., Degli Innocenti, F., Hinkelammert, K., and Russo, V.E.A. (1985) Factors affecting the range of pH gradients in the isoelectric focusing dimension of twodimensional gel electrophoresis: the effect of reservoir electrolytes and loading procedures. Electrophoresis 6, 339–348 Semple-Rowland, S.L., Adamus, G., Cohen, R.J. and Ulshafer, R.J. (1991) A reliable twodimensional gel electrophoresis procedure for separating neural proteins. Electrophoresis 12, 307–312 Adessi, C., Miege, C., Albrieux, C., and Rabilloud, T. (1997) Two-dimensional electrophoresis of membrane proteins: a current challenge for immobilized pH gradients. Electrophoresis 18, 127–135 Rabilloud, T., Adessi, C., Giraudel, A., and Lunardi, J. (1987) Improvement of the solubilization of proteins in two-dimensional electrophoresis with immobilized pH gradients. Electrophoresis 18, 307–316 Gordon, J.A., and Jencks, W.P. (1963) The relationship of structure to effectiveness on denaturing agents for proteins. Biochemistry 2, 47–57 Lyubimova, T., Caglio, S., Gelfi, C., Righetti, P.G. and Rabilloud, T. (1993) Photopolymerization of polyacrylamide gels with methylene blue. Electrophoresis 14, 40–50 Rabilloud, T. (1998) Use of thiourea to increase the solubility of membrane proteins in two-dimensional electrophoresis. Electrophoresis 19, 758–760 Santoni, V., Molloy, M., and Rabilloud, T. (2000) Membrane proteins and proteomics: un amour impossible? Electrophoresis 21, 1054–1070 Chevallet, M., Santoni, V., Poinas, A., Rouquie, D., Fuchs, A., Kieffer, S., et al. (1998)
36.
37.
38.
39.
40.
41.
42.
New zwitterionic detergents improve the analysis of membrane proteins by two-dimensional electrophoresis. Electrophoresis 19, 1901–1909 Rabilloud, T., Blisnick, T., Heller, M., Luche, S., Aebersold, R., Lunardi, J., et al. (1999) Analysis of membrane proteins by two-dimensional electrophoresis: comparison of the proteins extracted from normal or Plasmodium falciparum-infected erythrocyte ghosts. Electrophoresis 20, 3603–3610 Tastet, C., Charmont, S., Chevallet, M., Luche, S., and Rabilloud, T. (2003) Structure-efficiency relationships of zwitterionic detergents as protein solubilizers in two-dimensional electrophoresis. Proteomics 3, 111–121 Luche, S., Santoni, V., and Rabilloud, T. (2003) Evaluation of nonionic and zwitterionic detergents as membrane protein solubilizers in two-dimensional electrophoresis. Proteomics 3, 249–253 Taylor, C.M., and Pfeiffer, S.E. (2003) Enhanced resolution of glycosylphosphatidylinositolanchored and transmembrane proteins from the lipid-rich myelin membrane by twodimensional gel electrophoresis. Proteomics 3, 1303–1312 Lescuyer, P., Strub, J.M., Luche, S., Diemer, H., Martinez, P., Van Dorsselaer, A., et al. (2003) Progress in the definition of a reference human mitochondrial proteome. Proteomics 3, 157–167 MacFarlane, D.E. (1989) Two dimensional benzyldimethyl-n-hexadecylammonium chloride-sodium dodecyl sulfate preparative polyacrylamide gel electrophoresis: a high capacity high resolution technique for the purification of proteins from complex mixtures. Anal. Biochem. 176, 457–463 Burre, J., Beckhaus, T., Schagger, H., Corvey, C., Hofmann, S., Karas, M., et al. (2006) Analysis of the synaptic vesicle proteome using three gel-based protein separation techniques. Proteomics 6, 6250–6262
Chapter 3 Selection of pH Ranges in 2DE Mireille Starita-Geribaldi Summary This chapter describes the technical improvements of the two-dimensional electrophoresis pattern resulting of an optimized pH range in the first dimension. Various types of pH gradients are available. Different strategies can be applied in order to select the pH ranges for the exploration of a proteome. The resulting gels are analysed for their background, resolution, sensitivity in relation with the sample complexity. As the complete dynamic range of protein expression cannot be visualized, the high loading capacity of immobilized narrow pH gradients can be used. The limitations and possible enhancements are discussed. Key words: Basic pH gradient, Immobilized pH gradient, Narrow pH gradient, Overlapping, Resolution, Sensitivity, Two-dimensional gel electrophoresis, Wide pH gradient, Zoom gel, Review.
1. Introduction Two-dimensional gel electrophoresis (2DE) is the only proven method for simultaneously separating thousands of proteins and quantitatively comparing changes of protein profiles in cells, tissues, whole organisms. In a gel more than 2,000 polypeptides spots are present routinely, but probably more than 20,000 protein species are present in the cell. The total number of protein spots can be higher than the number of genes: proteolysis, posttranslational modifications, and mRNA alternative splicing can increase the number of spots. Limitations of 2DE are an inability to separate all protein components and insufficient dynamic range of detection. One way proposed to resolve this problem is to enrich protein of interest via subcellular fractionation (see Chapters “Organelle Proteomics”, “2DE for Proteome Analysis of David Sheehan and Raymond Tyther (eds.), Methods in Molecular Biology, Two-Dimensional Electrophoresis Protocols, vol. 519 © Humana Press, a part of Springer Science + Business Media, LLC 2009 DOI: 10.1007/978-1-59745-281-6_3
31
32
Starita-Geribaldi
Human Metaphase Chromosomes”, and “Microsomal Proteomics”) or chromatographic or electrophoretic (see Chapters “Difficult Proteins” and “Pre-Fractionation Using Microscale Solution IEF”) fractionation of the sample before 2DE. As the first dimension of 2DE has seen considerable recent progress, another way is to separate proteins in “zoom” gels where the number of spots detected is improved by narrowing the pH range. Conventional isoelectric focusing (IEF) in carrier ampholytes (CA-IEF) originated with Svensson (1). The definition of the resolving power of IEF derived by Rilbe (2) is: ⎡ D (dpH / dx) ⎤ Δ(pI) = 3 ⎢ ⎥ ⎣ E (−du / dpH) ⎦
1/2
(1)
Δ(pI) = resolution capacity D = diffusion coefficient of the protein E = field strength (V/cm) dpH/dx = pH gradient du/dpH = mobility slope at pI The two variables are pH gradient width dpH/dx which is minimized and field strength which is maximized. A good resolution is favoured by high field strength and a weak pH gradient. This results in an extremely low Δ(pI), the minimum pI difference needed to resolve two neighbouring bands. A suitable resolution will be obtained for compounds with low diffusion coefficient (D can be limited by the gel) and a high mobility slope at their pI, conditions that are satisfied by all proteins. IEF utilizing carrier ampholytes (CA) for the generation of the pH gradient has been widely used for the separation of proteins and other amphoteric molecules. But the technique has inherent limitations and problems such as distorted protein zones. The so-called ‘cathodic drift’ (3) causes problems when the focusing time is long, e.g. when very narrow pH ranges are used. Disturbance by salt ions transported through the gel exists in IEF where CA are relatively free to move within a gel. When high concentrations of salts are present, conductivity gradually increases. The ions that migrate towards the electrodes produce plateaus and reduce the effective separation distance. A shift from the traditional technique to the immobilized pH gradient (IPG) method has progressively occurred (4). As the groups generating the pH gradient are buffering acrylamide derivatives (immobilines) covalently bound to the gel matrix, this does not cause pH gradient drift. Electroendosmosis is not a problem in IPGs. IPG are reproducible, are not modified by the sample composition, and allow higher resolution than previously possible. This is due mainly to the lower conductivity in the gel resulting from the immobiline
Selection of pH Ranges in 2DE
33
concentrations used compared to CA-based pH gradients. Then, high voltages can be applied as the covalently bound pH gradient is not affected by long focusing times. Proteins with pIs differing as little as 0.001 pH units can be completely resolved compared to 0.01–0.02 with CA-IEF (5). Smooth, stable, and also flat pH gradients are possible. Furthermore, narrow pH intervals can be prepared in long IPG strips which allow high spatial resolution and high protein loading without distortion of the band patterns. This increased reproducibility in the IEF dimension has allowed comparison of inter-laboratory data in 2DE maps (6; see also Chapter “Creating 2DE Databases for the World Wide Web”) and the possibility to carry out a preparative step (7), new tools such as databases were developed. In this chapter we will describe the different types of pH gradients. 2DE in a wide pH gradient is generally the starting point but depends on sample complexity. A modification of the pH slope can be selected to enhance resolution and protein loading in the 2DE pattern. Diverse strategies can be applied for this selection and are illustrated by some examples.
2. pH Gradients in 2DE 2.1. Linear pH Gradients
A series of buffering acrylamide derivatives with the general structure CH2⇒CH–CO–NH–R, where R is a carboxylic acid or a tertiary amino group defined by its pKa value, were used for copolymerization with acrylamide and bisacrylamide in order to produce IPG. In the IPG technique, the non-amphoteric buffering groups generating the pH gradient are covalently linked to the matrix used as an anti-convective support. When different pK values of commercially available immobilines (pK values of 3.6, 4.6, 6.2, 7.0, 8.5, and 9.3) are mixed in appropriate proportions in a gradient maker together with acrylamide, bisacrylamide, TEMED, and ammonium persulphate, a pH gradient is formed that is covalently attached to the polyacrylamide gel. Casting is identical to conventional polyacrylamide pore gradient gels. The buffer solutions are mixed in a gradient mixer allowing the pH gradient to co-vary with a density gradient. Linear mixing is the most widespread way to realize the pH gradient. In practice, an acid-dense solution where glycerol is added and a basic light one are prepared. Different recipes listing immobiline volumes for diverse pH gradients broad and narrow range reaching pH 10 or 10.5 were described by the manufacturer in an application note (8). The corresponding tables have been reported (9) and, if custom-made pH ranges are needed such as ultranarrow (less than
34
Starita-Geribaldi
one pH unit), the Immobiline volumes of the light and dense solution can be deduced by graphic interpolation. Commercial immobilized linear pH gradient wide range (3–10), 4 or 5 units pH range (7–11, 6–11), 3 units pH range (3–6, 4–7, 5–8, 6–9, 7–10), 2 units pH range (6.3–8.3), 1.2 or 1.3 unit pH range (3.9–5.1, 4.7–5.9, 5.3–6.5, 5.5–6.7, 6.2–7.5), and 1 unit pH range (3.5–4.5, 4–5, 4.5–5.5, 5–6) are available in various lengths: 7 cm, 11 cm, 13 cm, 17 or 18 cm, and more recently 24 cm. It must be noticed that few commercially available IPGs with basic pH ranges exist and, in this case, the pH gradients must be home-made. Since the 2DE technique has been revised in a lot of technical parameters among which the size of the gel (46 × 30 cm) to achieve high resolution (10), use of wide pH gradients is in progress. High reproducibility of large format 2-DE gels was achieved (11). For example, 24-cm-long IPGs 4.5–5.5 and 5.5–6.7 pH range were used to study radioresistance-related proteins in rectal cancer with 500 μg of samples separated and longer focusing time than in the classical 3–10 pH range (12). Separations in 24-cm long IPGs, pH 4–7, were used for differential gel exposure (Difexpo) of radiolabelled proteins in vivo, based on the co-electrophoresis on a 2DE gel of two protein samples, one labelled with 14C and the other with a 3H isotope (13). Finally the 3H/14C ratio for each spot was determined by the peak height value on two different imaging plates. A set of two measures from two different experiments on the same spot necessitates well-resolved 2DE maps. Running conditions with 24-cm-long IPGs have been described (14). 2.2. Non-linear pH Gradients
Non-linear (NL) pH gradients can be flattened for only a part of the pH range or give specific pH profiles. They are used with CAIEF and IPGs. The different ways to obtain them are described briefly (15). The gradient slope is altered in different segments in order to optimize the resolution of all components. pH gradient engineering has developed diverse types of NL pH gradients (16). Although logarithmic and polynomial gradients are of limited applicability, concave exponential gradients appeared suitable to give space to proteins spots in complex protein mixtures such as cell lysates and biological fluids and appeared quite useful in many 2DE applications. A series of gradients of 3 or 4 units or wide pH range were derived starting at pH 3, pH 4, or pH 5 as well as the most extended pH 2.5–11 which was optimal for highly heterogeneous samples. Their enhanced resolution was used early in proteomics investigations: IPG pH 3.5–10 NL resolved fairly well trains of spots such as α1-anti-trypsin or α1-anti-chymotrypsin in plasma protein 2DE map (17) and was also used to generate human liver 2DE maps for a two-dimensional gel database by microsequencing (18). Commercial IPG pH 3–10, 3–11, 7–11, 3–5.6 NL strips are available in diverse
Selection of pH Ranges in 2DE
35
lengths. Results can be compared to linear pH gradients of the same pH range or to a more narrow range: the 2-DE reference map of skeletal muscle using an IPG pH 3–10 NL strip in the first dimension revealed 652 spots compared to 697spots in a 4–7 pH range linear IPG (19). IPG 3–10 gave a broad overview, but the IPG 4–7 offered better resolution of the acidic region. 2.3. Basic pH Gradients
Commercially available basic pH range IPGs include 6–9, 7–10, 6–11, 7–11. Some recipes up to pH 10.5 are tabulated (9) and specific laboratory recipes have been reported (20, 21). Above pH 7.5 it is difficult to obtain good resolution, and basic IPGs run with standard protocols also deteriorate in quality with high sample load (22). Anodic cup application of the sample does give improved results, however. Specific focusing times have been described for particular pH ranges such as 34 kVh for IPG 6–9 (20, 21) or 6–11 (21) compared to IPG 6.2–8.2 and 7.5–9.5 focused until 80 kVh. Finally, chemicals involved in IEF can disimprove the pattern: dithiothreitol (DTT) and dithioerythritol (DTE) are weak acids with pK values in the region of pH 8–9 which are then transported out of the basic part of the IPG strip during focusing, resulting in loss of solubility for some proteins and horizontal streaking. Another problem is the appearance of extra spots and non-reproducible trains of spots probably due to variation in the number of oxidized protein thiols. Alkylation improved the 2DE map without a complete elimination of the problems. Poor resolution in the alkaline pH range can be enhanced by increasing the concentration of DTT in the rehydration buffer to 2.5% DTT compared to 1% in the original protocol, and the pattern can be further improved by addition of isopropanol and glycerol (21). The “Clean Gel” paper bridge method (22) can be used to introduce DTT continuously (23) by soaking “CleanGel” electrode strips in rehydration buffer with 3.5% DTT and placing them at the alkaline (cathodic) end of the strip. Oxidation of protein thiol groups with disulphides was proposed to eliminate streaking when narrow range basic gradients are used. This stabilization of cysteines was performed by using hydroxyethyldisulphide (HED) reagent (Destreak reagent from GE Healthcare) instead of the reductant DTT in the rehydration solution (24, 25). The combined use of anodic cup loading and the HED containing solution produced better resolution in both analytical and micropreparative cup loading (26).
2.4. Very Alkaline pH Gradients
pH gradients of ranges up to pH 10 can be made with six commercial immobilines (see Subheading 2.1). The range can be extended between pH 2 and 12 by including other immobilines such as pK 1.0, pK 10.3, and even pK > 13 (27–30). Suitable 2DE patterns were obtained particularly with TCA/acetone precipitation recommended for enrichment and better visualization
36
Starita-Geribaldi
of basic proteins >10. Very alkaline pH gradients were used for 2DE of ribosomal and nuclear proteins (29). By using optimized pH gradients in the ranges 9–12 and 8–12 calculated using the Altland IPGMAKER program (31) and 10–12 range (32–34), these proteins were focused in IPGs to the steady state. By using the most basic immobilines, the gel of the IPG strip is charged at the cathodic end. Reverse electroendosmotic flow takes place with water transport from cathode to anode. Disturbed patterns were corrected by gel composition replacing acrylamide by dimethylacrylamide, using additives such as isopropanol 10%, glycerol 10%, and methylcellulose 0.2% or 16% isopropanol, and adding 10% glycerol to the reswelling solution for IPGs (29). Finally, anodic sample application is mandatory and an increase of the final voltage up to 8,000 V gave optimum results (7, 35, 36). Wide pH gradients in the range 4–12 (27) or 3–12 (28) that give an overview of the total proteome (e.g. protein expression in mouse liver) have been successfully used. The reverse electroendosmotic flow is negligible, and 3–12, 4–12, and even 6–12 can be run in standard conditions. Furthermore, by increasing the separation distance from 18 to 24 cm, resolution is improved (28). Sample application by in-gel rehydration (37, 38) and IEF are performed automatically in one-step procedure in an integrated system (IPGphor).
3. Strategies 3.1 The Initial Experiment 3.1.1. The Theoretical Approach
3.1.2. The Experimental Approach
The statistical distribution of the pIs of water-soluble proteins along the pH axis has been computed and plotted on a histogram (39). The ideal curve was derived taking into account the relative abundance of proteins. Then, the non-linear pH 4–10 IPG interval became one of the standard recipes for 2DE (40). Other examples are the theoretical pI and Mr maps of yeast cell proteins derived from MIPS (http://www.mips.embnet;org/proj/ yeast/) (41) and the theoretical map of Bacillus subtilis displaying the two-thirds of the protein spots located in the neutral and acidic region (42). In this case, the MICADO database (http:// locus.jouy.inra.fr/cgi-bin/genmic/madbase_home.pI) was used, in which the pI region 4–7 should represent the bulk of cytosolic proteins. Furthermore, the number of membrane spanning domains is indicated in this map. It appears also that few proteins are out of the analytical window of 2DE. 2DE in a wide pH gradient is the starting point in nearly all proteomic studies. The choice of the pH gradient depends on sample
Selection of pH Ranges in 2DE
37
complexity. pH ranges 3–10, 3–12, 4–9, 4–7 have been used classically since the beginning of IEF and can be linear or not. 2DE in a wide pH gradient allows analysis of simple proteomes; this is the case for a small genome or when an organelle or affinityselected subfractionation has been prepared. In the case of more complex proteomes, this approach is used to get an overview of the total proteome. It is possible to appreciate the protein concentration at the analytical level, the distribution of the spots, and to determine areas of the gel that require better resolution. A differential analysis between proteomes in two physiological or pathological states in cells (43), tissues (44), or physiological fluids (45) can be performed. Peptide mass fingerprinting (PMF) of the most expressed proteins on semi-preparative gels in NL IPG pH 3–10 has been used in various recent proteomic investigations (46, 47). However narrow pH gradients will be required for identification of minor spots by MS analysis. Starting from the initial experiment, the ways by which one can select other pH gradients in order to get better resolution are: (a) determination of the pI by focusing sample on a CA gel plate, pH 3.5–9.5, at 10°C. A pH gradient measured with a surface pH electrode at the temperature of IEF will allow pI values to be determined with sufficient precision from the band position, but surface measurement is not practicable with IPGs as the conductivity in the strip is very low and the strip very narrow; (b) the use of carbamylated standard pI markers migrating with the sample (48); (c) as the endpoints are known, the pH of the endpoints for the narrow gradient can be deduced after measuring their distance from the edge of the gel with a ruler; nowadays, the development of informatic tools such as image analysis software calculates the pI/Mr of each spot suggesting quickly in which pH zone the gel could be better resolved. Then, three pH units, two pH units, and/or one pH unit can be used sequentially for separation of the sample. Before separation, the sample is dissolved in IEF lysis buffer without interfering substances (49). Experimental details in relation to sample preparation, sample loading, voltage ramps to be used in diverse pH gradients and in different apparatus have been reviewed (7, 50) and protocols detailed (51). Examples of IEF lysis buffers and voltages are described in the legend to Fig. 1. 3.2. Narrowing the pH Zone Progressively
This procedure allows us to conclude what is the best pH range to use, taking into account the background, the number of spots, sensitivity, and applicability to micropreparative purpose or detection of a protein of interest and its isoforms. Reduced background is obtained by flattening the pH gradient in the first dimension and the polyacrylamide gradient in the second dimension (52). With narrow pH gradients and very wide IPGs the overcrowded zones can spread fairly well in high spaced separations.
38
Starita-Geribaldi
Fig. 1. pH ranges used in seminal plasma from a normospermic man 2-DE. Sample (50 μg in IPG pH 3–10) or after desalting (100 μg in IPG pH 4–7, 750 μg in IPG pH 5–8, 500 μg in IPGs pH 4.5–5,5 and 5–6) was applied by in-gel rehydration and separated on Multiphor II (GE Healthcare) with a limitation of 3,500 V. IEF lysis buffer: 7 M urea, 2 M thiourea, 4% CHAPS, 65 mM DTT, 0.8% ampholytes of the corresponding pH range or 1.6% in semi-preparative experiments. (a): initial experiment using 18-cm-long IPG pH 3–10 (GE Healthcare) and gradient voltage 20,000 Vh (54); (b) 2DE using 18-cm-long IPG pH 4–7 (GE Healthcare) and gradient voltage 23,000 Vh; (c): 17-cm-long IPG pH 5–8 (BIO-RAD) and step voltage 150 V during 1 h, 300 V 1 h, 600 V 1 h, 3,500 V to reach 60,000 Vh; (d): 18-cm-long IPG pH 4.5–5 (GE Healthcare) with identical step voltage as in (C) and 60,000 Vh; (e): 18-cm-long IPG pH 5–6 (GE Healthcare) as in (d). SDS equilibration buffer (54–56) is used before second dimension in ExcelGels (GE Healthcare) followed by silver staining (see Chapter “Silver Staining of Proteins in 2DE Gels”). White arrows are spots of epididymal secretory protein E1.
To determine whether narrowing the pH range of the IPG strip would allow detection of additional protein spots, IPG strips of wide, medium, and narrow pH ranges can be compared. By loading equal amounts of proteins from rat heart tissue (100 μg) on the pH 3–10, 5–8, and 5.5–6.7 IPG strips the resulting silver stained 2DE were analysed (53). It appeared that the largest increment in the number of spots detected by the image analysis
Selection of pH Ranges in 2DE
39
software is gained by moving from the medium range to the narrow range strip, increasing the number of spots detected by 87% compared with the number of spots detected in that pH range on the pH 5–8 IPG strip. However progressive protein concentrations can be separated even at the analytical level thanks to the better loading capacity of the narrow pH range IPGs. This was achieved using cup loading anodic side application to Sacchromyces cerevisiae 2DE maps (41) or the in-gel rehydration method used by our group to separate human seminal plasma by 2DE (37, 38) (Fig. 1). In our initial experiment, 50 μg of seminal plasma proteins was allowed to separate in an 18-cm wide pH gradient, pH 3–10, revealing 734 ± 27 polypeptides visible in the corresponding 2DE of seminal plasma of a normospermic man (Fig. 1a). We were able to program a differential analysis between normospermic men and azoospermic patients and listed candidate protein infertility biomarkers in azoospermic patients (54). Better resolution of the seminal plasma 2DE is possible with an IPG strip pH 4–7 (Fig. 1b). This can be loaded with 100 μg protein. However, the protein of interest can be a guide to selecting the optimum pH zone: prostate specific antigen (PSA, Swiss Prot accession code P07288, Fig. 1a) cannot be studied in the 4–7 pH range as only one spot (experimental pI/Mr: 6.77/30.54, Fig. 1b) is placed at the boundary of this 3 units’ pH range IPG strip, the two other isoforms of PSA (experimental pI/Mr: 6.95/30.2 and 7.13/30.77) are outside this range. The train of three PSA spots is better resolved with an IPG pH 5–8 (Fig. 1c). Semi-preparative load (750 μg) is used to enhance detection of other spots of interest. Furthermore, concentration of spots in the acidic zone of the initial experiment with IPG 3–10 in the first dimension is so important that narrowing the pH zone in one unit pH gradient is possible. With 500 μg of proteins of seminal plasma loaded, 438 ± 25 spots could be obtained in a 4.5–5.5 pH range 2DE gel (Fig. 1d) compared to 338 ± 28 in the same fraction of the pH 4–7 IPG (Fig. 1b) and 280 ± 56 spots in the same fraction of the wide pH 3–10 IPG (Fig. 1a) and 637 ± 25 spots are obtained with the pH 5–6 IPG (Fig. 1e) compared to 302 ± 28 in the same fraction of the pH 4–7 IPG (Fig. 1b) and 295 ± 45 in the same pH range in the wide pH 3–10 IPG (Fig. 1a). In one pH gradient unit as in IPG pH 4.5–5.5 or 5–6, preparative work can separate several milligrams of seminal plasma proteins. Isolated spots, trains of spots, or even faint spots could be digested by trypsin and identified by MS and database searching (55, 56). Better resolution can be obtained by progressively narrowing the pH range (41, 55). Narrow pH gradients, also called “zoom” gels, can be used for more specialized work: by decreasing the pH slope, contaminants comigrating with spots of interest such as two of the isoforms of human epididymal secretory protein E1
40
Starita-Geribaldi
(pI/Mr 5.16/23.35 and pI/Mr 5.2/24.81, Swiss Prot accession code Q 15,668) could be separated, and detection of very faint spots such as amyloid p-component (Swiss Prot accession code: P02743) in the azoospermic patient seminal plasma was possible with a one pH unit range in the first dimension (56). A more accurate differential analysis between normospermic men and azoospermic men was then performed. Furthermore the train of clusterin (P10909, Swiss Prot accession code) close to the train of a major protein of the map was masked by vertical streaking, due to overloading of the molecules of the neighbouring proteins in 2DE performed with an IPG pH 3–10. Resolution obtained with an IPG pH 4.5–5.5 overcame this problem and furthermore suggested a specific role of some clusterin isoforms in the pathology of infertility (56). The importance of separating isoforms has also been demonstrated by others (57). 3.3. Dividing the pH Zone Systematically in Narrow pH Zones
Overlapping IPG separations and recombining a 2DE map can be applied even by combining 3 units of pH gradients: 2DE of circulating human monocytes allowed the reproducible detection and quantification of 1,500 spots in the 4–7 pH range and more than 2,000 spots in total by combining 2DE gels in the 4–7, 6–9, and 6–11 pH ranges (58). In order to display proteomes more completely, few studies attempted systematically to determine the maximum possible number of proteins that can be resolved using currently available overlapping narrow pH range IPGs. Overlapping narrow IPGs covering one pH unit with 18-cm separation distance have been proposed for the acidic range up to pH 7 (32). The protein pattern from 697 pre-B lymphoma cell separated by the use of a pH 3–10 IPG contains both acidic and basic proteins (20). A series of narrow pH unit IPG strips with overlapping boundaries were used and home-made gels were required for the basic zones. This approach is interesting when the aim of the study is a global screen of the proteome. The concept of increasing the resolution by dividing the pH 3–10 range into six narrow pH ranges works well in the acidic range of the proteome (20). However, recent improvements can be used above pH 7.5 (see Subheadings 2.3 and 2.4), although this requires extensive efforts to analyse many gels. Recently, visualization of the complex proteome of human heart left ventricle by a series of narrow range IPGs was achieved (59). A composite pattern covers all the proteome. Over a twofold increase in the number of spots has been calculated in human heart proteins with composite gels achieved with successive 3.5–4.5, 4–5, 4.5–5.5, 5–6, 5.5–6.7 narrow zones compared with the number of observed spots on the equivalent pH range of an IPG 3–10 NL 2DE gel (57). Recombining the 2DE map is particularly interesting when isoforms or trains of isoforms appear in two consecutive narrow pH gradients as spots of human epididymal secretory protein E1
Selection of pH Ranges in 2DE
41
present in seminal plasma and separated in the IPGs pH 4.5–5.5 and 5–6 (Fig. 1d, e) (56). 3.4. Limitations
When loading varying amounts of protein, the number of spots detected did not increase linearly as a function of the number of spots. With IPG strips, the limit of linear increase seems to be 200 μg for silver (see Chapter “Silver Staining of Proteins in 2DE Gels”) and 400 μg for SyproRuby in a pH 5–8 IPG strip (53). Overloading 2DE gels has a limited power, a phenomenon more pronounced for narrow IPGs. In systematic studies testing diverse concentrations, some spots merge in some conditions, while others remain at the limit of visualization or can fuse due to overloading (60). Furthermore, 2DE maps obtained with IPG pH 5–6 in the first dimension (Fig. 1e) begin to deteriorate at the ends with 500 μg of seminal plasma proteins, displaying vertical streaking. This confirms protein precipitation near the electrodes previously described for proteins with pI outside the pH range of the IPG strip (61). When the entire 2DE map is electronically reconstructed with narrow-range maps, the technique may give unreliable results because even very narrow IPG strips have to be loaded with the entire cell/tissue lysate or physiological fluid sample which contains proteins of diverse pIs. Proteins that focus in a selected narrow-range IPG interval will be strongly underrepresented as they will be only a small fraction of the entire sample loaded. Aggregation among unlike proteins will cause massive precipitation. Proteins display heterogeneity in size, charge, solubility, relative abundance, and multiple technologies can be required to analyse a wide range of proteins.
3.5. Enhancements
Prefractionation was proposed as a solution to obtain clearer patterns with background enhancement, suppression of vertical streaking, and a greater number of spots. Prefractionation methods can be applied in many ways. TCA/acetone precipitation may be selective (62), affinity purification (see Chapter “Solubilization of Proteins in 2DE: An Outline”) or chromatographic separation have been reported (63). Sequential extraction of cells or membranes has been proposed with strong solubilization solutions at the end of the procedure in order to extract the most difficult proteins (64–66) (see also Chapters “Difficult Proteins”, “Organelle Proteomics”, “Protein Extraction for 2DE”, and “Pre-Fractionation Using Microscale Solution IEF”). Another approach can be subcellular fractionation (67). Prefractionation methods according to pI have been reported such as free-flow IEF (68), IEF in a rotating multichamber device (69), a multifunctional electrokinetic membrane apparatus (70), a multicompartment electrolyser operating with isoelectric membranes (71) and its miniaturized form (72, 73). The advantage of IEF prefractionation is illustrated by selection of serum fractions in a given pI range which can then
42
Starita-Geribaldi
be applied to IPGs of similar pI range allowing 1,000–1,500 μg protein loadings in one unit pH range (74). Finally, a simple and fast sample prefractionation procedure has been proposed using flat-bed IEF in granulated Sephadex gels with conventional IEF instrumentation and prior separation in narrow pH range IPG (61).
4. Concluding Remarks Although competition is increasing from various other technologies, 2DE justifiably maintains a central role in proteomics. The present review has illustrated the role of the pH gradient in the powerful separation power of 2DE for analysis of complex mixtures. Wide pH gradients can give an overview of the total protein expression of cells while overlapping, narrow, and immobilized pH gradients yield greater numbers of spots. In order to increase the resolution and number of spots detected, long immobilized pH gradient strips with narrow pH ranges have been designed. However, we are confronted with the problem that the complete dynamic range of protein expressions in a complex sample cannot be visualized. In order to display many more proteins than already achieved, it seems that the development of prefractionation methods that meet the requirement for proteomics, associated with narrow pH gradients, can give new insights at the detection level. References 1. Svenson, H. (1961) Isoelectric fractionation, analysis and characterization of ampholytes in natural pH gradient. Acta Chem. Scand. 15, 325–341 2. Rilbe, H. (1973) Historical and theoretical aspects of isoelectric focusing. Ann. N. Y.. Acad. Sci. 209, 11–22 3. Righetti, P.G., and Drysdale, J.W. (1973) Smallscale fractionation of proteins and nucleic acids by isoelectric focusing in polyacrylamide gels. Ann. N. Y. Acad. Sci. 209, 163–186 4. Bjellqvist, B., Ek, K., Righetti, P.G., Gianazza, E., Görg, A., Westermeier, R., and Postel, W. (1982) Isoelectric focusing in immobilized pH gradients: principle, methodology and some applications. J. Biochem. Biophys. Methods 6, 317–339 5. Righetti, P.G., Gianazza, E., and Gelfi, C. (1988) Immobilized pH gradients. Trends Biochem. Sci., 13, 335–338
6 . Corbett , J. , Dunn , M.J. , Posch , A. , and Görg , A. (1994) Positional reproducibility of protein spots in two-dimensional polyacrylamide gel electrophoresis using immobilized pH gradient isoelectric focusing in the first dimension: an interlaboratory comparison. Electrophoresis, 15, 1205–1211 7. Görg, A. (2000) The current state of twodimensional electrophoresis with immobilized pH gradients. Electrophoresis, 21, 1037–1053 8. LKB Application note No 324, Bromma, Sweden. 9. Westermeier, R. (2000) IEF in immobilized pH gradient, in Electrophoresis in Practice, 3rd edn, (Westermeier, R., ed.), Wiley-VCH, Weinheim, pp. 223–238 10. Klose, J., and Kobalz, U. (1995) Two-dimensional electrophoresis of proteins: an updated protocol and implications for a functional analysis of the genome. Electrophoresis, 16, 1034–1059
Selection of pH Ranges in 2DE 11. Challapalli, K.K., Zabel, C., Schuchhardt, J., Kaindi, A.M., Klose, J., and Herzel, H. (2004) High reproducibility of two-dimensional electrophoresis. Electrophoresis, 25, 3040–3047 12 . Allal , A.S. , Kähne , T. , Reverdin , A.K. , Lippert , H., Schlegel, W., and Reymond, M.A. (2004) Radioresistance-related proteins in rectal cancer. Proteomics, 4, 2661–2669 13. Monribot-Espagne, C., and Boucherie, H. (2002) Differential exposure, a new methodology for the two-dimensional comparison of protein samples. Proteomics, 2, 229–240 14. Boucherie, H., and Monribot-Espagne, C. (2006) Two-dimensional gel electrophoresis of total yeast proteins in yeast protocols. Methods Mol. Biol., 313, 47–64 15. Gianazza, E., Giacon, P., Sahlin, B., and Righetti, P.G. (1985) Non-linear pH courses with immobilized pH gradients. Electrophoresis, 6, 53–56 16. Righetti, P.G., and Tonani, C. (1991) Immobilized pH gradients simulator- an additional step in pH gradient engineering: II. Nonlinear pH gradients. Electrophoresis, 12, 1021–1027 17. Hughes, G.J., Frutiger, S., Paquet, N., Ravier, F., Pasquali, C., Sanchez, J.C., et al. (1992) Plasma protein map: an update by microsequencing. Electrophoresis, 13, 707–714 18. Hochstrasser, D.F., Frutiger, S., Paquet, N., Bairoch, A., Ravier, F., Pasquali, C., et al. (1992) Human liver protein map: a reference database established by microsequencing and gel comparison. Electrophoresis, 13, 992–1001 19. Yan, J., Harry, R.A., Wait, R., Welson, S.Y., Emery, P.W., Preedy, V.R., et al. (2001) Separation and identification of rat skeletal muscle proteins using two-dimensional gel electrophoresis and mass spectrometry. Proteomics, 1, 424–434 20. Hoving, S., Voshol, H., and van Oostrum, J. (2000) Towards high performance twodimensional gel electrophoresis using ultrazoom gels. Electrophoresis, 21, 2617–2621 21. Hoving, S., Gerrits, B., Voshol, H., Muller, D., Roberts, R.C., and van Oostrum, J. (2002) Preparative two-dimensional gel electrophoresis at alkaline pH using narrow range immobilized pH gradients. Proteomics, 2, 127–134 22. Sabounchi-Schütt, F., Aström, J., Olsson, I., Eklund, A., Grunewald, J., and Bjellqvist, B. (2000) An immobiline dry-strip application method enabling high-capacity two-dimensional gel electrophoresis. Electrophoresis, 21, 3649–3656 23. Görg, A., Boguth, G., Obermaier, C., Posch, A., and Weiss, W. (1995) Twodimensional polyacrylamide gel electrophore-
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
43
sis with immobilized pH gradients in the first dimension (IPG-Dalt): The state of the art and the controversy of vertical versus horizontal systems. Electrophoresis, 16, 1079–1086 Olsson, I., Larsson, K. Palmgreen, R., and Bjellqvist, B. (2002) Organic disulfides as a means to generate streak-free two-dimensional maps with narrow range basic immobilized pH gradient strips as first dimension. Proteomics, 2, 1630–1632 Luche, S., Diemer, H., Tastet, C., Chevallet, M., Van Dorsselaer, A., Leize-Wagner, E., et al. (2004) About thiol derivatization and resolution of basic proteins in two-dimensional electrophoresis. Proteomics, 4, 551–561 Pennington, K., McGregor, E., Beasley, C.L., Everali, I., Cotter, D., and Dunn, M.J. (2004) Optimization of the first dimension for separation by two-dimensional gel electrophoresis of basic proteins from human brain tissue Proteomics, 4, 27–30 Görg, A., Boguth, G., Obermeier, C., and Weiss, W. (1998) Two-dimensional electrophoresis of proteins in an immobilized pH 4–12 gradient. Electrophoresis, 19, 1516–1519 Görg, A., Obermeier, C., Boguth, G., and Weiss, W. (1999) Recent developments in two-dimensional gel electrophoresis with immobilized pH gradients: wide pH gradients up to pH 12, longer separation distances and simplified procedures. Electrophoresis, 20, 712–717 Görg, A., Obermaier, C., Boguth, G., Csordas, A., Diaz, J.-J., and Madjar, J.-J. (1997) Very alkaline immobilized pH gradients for two-dimensional electrophoresis of ribosomal and nuclear proteins. Electrophoresis, 18, 328–337 Görg, A., and Weiss, W. (2000) Two-dimensional electrophoresis with immobilized pH gradients, in Proteome Research: Two-Dimensional Gel Electrophoresis and Identification Methods (Rabilloud, T., ed.), Springer Verlag, Berlin, pp. 57–92 Altland, K. (1990) IPG maker: A program for IBM-compatible personal computers to create and test recipes for immobilized pH gradients. Electrophoresis, 11, 140–147 Righetti, P.G. (1990) in Immobilized pH Gradients: Theory and Methodology, Biomedical Press, Elsevier, Amsterdam Bossi, A., Righetti, P.G., Vecchio, G., and Severinsen, S. (1994) Focusing of alkaline proteases (subtilisins) in pH 10–12 immobilized pH gradients. Electrophoresis, 15, 1535–1540 Bossi, A., Gelfi, C., Orsi, A., and Righetti, P.G. (1994) Isoelectric focusing of histones
44
35.
36.
37.
38 .
39.
40.
41.
42.
43.
44.
45.
Starita-Geribaldi in extremely alkaline immobilized pH gradients. J. Chromatogr. A, 686, 121–128 Wildgruber, R., Reil, G., Drews, O., Parlar, H., and Görg, A. (2002) Web-based two-dimensional database of Saccharomyces cerevisiae proteins using immobilized pH gradients from pH 6 to pH 12 and matrix-assisted laser desorption/ionization-time of flight mass spectrometry. Proteomics, 2, 727–732 Drews, O., Reil, G., Parlar, H., and Görg, A. (2004) Setting up standards and a reference map for the alkaline proteome of the Gram-positive bacterium Lactococcus lactis. Proteomics, 4, 1293–1304 Rabilloud, T., Valette, C., and Lawrence, J.J. (1994) Sample application by in-gel rehydration improves the resolution of two-dimensional electrophoresis with immobilized pH gradients in the first dimension. Electrophoresis, 15, 1552–1558 Sanchez , J.C. , Rouge , V. , Pisteur, M. , Ravier, F., Tonella, L., Moosmayer, M., et al. (1997) Improved and simplified in-gel sample application using reswelling of dry immobilized pH gradients. Electrophoresis, 18, 324–327 Gianazza, E., and Righetti, P.G. (1980) Size and charge distributions of macromolecules in living systems. J. Chromatogr., 193, 1–8 Gianazza, E., Giacon, P., Sahlin, B., and Righetti, P.G. (1985) Non-linear pH courses with immobilized pH gradients. Electrophoresis, 6, 53–56 Wildgruber, R., Harder, A., Obermaier, C., Boguth, G., Weiss, W., Fey, S.J., et al. (2000) Towards higher resolution: two-dimensional electrophoresis of Saccharomyces cerevisiae proteins using overlapping narrow immobilized pH gradients. Electrophoresis, 21, 2610–2616 Buttner, K., Bernhardt, J., Scharf, C., Schmid, R., Mäder, U., Eymann, C., et al. (2001) A comprehensive two-dimensional map of cytosolic proteins of Bacillus subtilis. Electrophoresis, 22, 2908–2935 Yocum, A.K., Busch, C.M., Felix, C.A., and Blair, Y.A. (2006) Proteomics-based strategy to identify biomarkers and pharmacological targets in leukemias with t(4;11) translocations. J. Proteome. Res., 5, 2743–2753 Bajo, M., Fruehauf, J., Kim, S.H., Fountoulakis, M., and Lubec, G. (2002) Proteomic evaluation of intermediary metabolism enzyme proteins in fetal Down’s syndrome cerebral cortex. Proteomics, 2, 1539–1546 Neumann, M., von Bredow, C., Ratjen, F., and Griese, M. (2002) Bronchoalveolar lavage
46.
47.
48.
49.
50.
51.
52.
53.
54.
55.
56.
57.
protein patterns in children with malignancies, immunosuppression, fever and pulmonary infiltrates. Proteomics, 2, 683–689 Guércio, R.A.P., Shevchenko, A., Shevchenko, A., Lopez-Lozano, Paba, J., Sousa, M.V., et al. (2006) Ontogenic variations in the venom proteome of the amazonian snake Bothrops atrox. Proteome Sci., 11, 4–11 Mayr, U., Mayr, M., Yin, X., Begum, S., Tarelli, E., Wait, R., et al. (2005) Proteomic dataset of mouse aortic smooth muscle cells. Proteomics, 5, 4546–4557 Anderson, N.L., and Hickman, B.J. (1979) Analytical techniques for cell fractions. XXIV. Isoelectric point standards for two-dimensional electrophoresis. Anal. Biochem., 93, 312–320 Rabilloud, T. (ed.) (2000) Proteome Research: Two-Dimensional Gel Electrophoresis and Identification Methods. Springer Verlag, Berlin. Görg, A., Weiss, W., and Dunn, M.J. (2004) Current two-dimensional electrophoresis technology for proteomics. Proteomics, 4, 3665–3685 Weiss, W., and Görg, A. (2006) Two-dimensional gel electrophoresis for plant proteomics. Methods Mol. Biol., 112, 235–244 Görg, A., Postel, W., Weser, J., Günther, S., Strahler, J.R., Hanash, S.M., et al. (1987) Horizontal two-dimensional electrophoresis with immobilized pH gradients in the first dimension in the presence of non-ionic detergent. Electrophoresis, 8, 45–51 Li, B.Z, Flint, P.W., and Boluyt, M.O. (2005) Evaluation of several two-dimensional gel electrophoresis techniques in cardiac proteomics. Electrophoresis, 26, 3572–3585 Starita-Geribaldi, M., Poggioli, S., Zucchini, M., Garin, J., Fenichel, P., and Pointis, G. (2001) Mapping of seminal plasma proteins by two-dimensional gel electrophoresis in men with normal and impaired spermatogenesis. Mol. Hum. Reprod., 7, 715–722 Starita-Geribaldi, M., Roux, F., Garin, J., Fenichel, P., and Pointis, G. (2002). ExcelGel precast gels for proteomic analysis of physiological fluids. Life Sci. News, 10, 7–9 Starita-Geribaldi, M., Roux, F., Garin, J., Chevallier, D., Fenichel, P., and Pointis, G. (2003) Development of narrow immobilized pH gradients covering one pH unit for human seminal plasma proteomic analysis. Proteomics, 3, 1611–1619 Westbrook, J.A., Yan, J.X., Wait, R., Welson, S.Y., and Dunn, M.J. (2001) Zooming-in on the proteome: Very narrow-range immobilised pH gradients reveal more protein species and isoforms. Electrophoresis, 22, 2665–2871
Selection of pH Ranges in 2DE 58. Gonzalez-Barderas, M., Gallego-Delgado, J., Mas, S., Duran, M.C., Làzaro, A., Hernandez-Merida, S., et al. (2004) Isolation of circulating monocytes with high purity of proteomic analysis. Proteomics, 4, 432–437 59. Westbrook, J.A., Wheeler, J.X., Wait, R., Welson, S.Y., and Dunn, M.J. (2006) The human heart proteome: two-dimensional maps using narrow-range immobilised pH gradients. Electrophoresis, 27, 1547–1555 60. Corthals, G.L., Wasinger, V.C., Hochstrasser, D.F., and Sanchez, J.C. (2000) The dynamic range of protein expression: a challenge for proteomic research. Electrophoresis, 21, 1104–1115 61. Görg, A., Boguth, G., Köpf, A., Reil, G., Parlar, H., and Weiss, W. (2002) Sample prefractionation with Sephadex isoelectric focusing prior to narrow pH range two-dimensional gels. Proteomics, 2, 1652–1657 62. Görg, A., Boguth, G., Obermeier, C., and Weiss, W. (1998) Two-dimensional electrophoresis of proteins in an immobilized pH 4–12 gradient. Electrophoresis, 19, 1516–1519 63. Fountoulakis, M., Tabacs, M.F., and Takacs, B.J. (1999) Enrichment of low copy-number gene products by hydrophobic interaction chromatography. J. Chromatogr. A, 833, 157–168 64. Weiss, W., Postel, W., and Görg, A. (1992) Application of sequential extraction procedures and glycoprotein blotting for the characterization of the 2-D polypeptide patterns of barley seed proteins. Electrophoresis, 13, 770–773 65. Weiss, W., Vogelmeier, C., and Görg, A. (1993) Electrophoretic characterization of wheat grain allergens from different cultivars involved in bakers’ asthma. Electrophoresis, 14, 805–816 66. Molloy, M.P., Herbert, B.R., Walsh, B.J., Tyler, M.I., Traini, M., Sanchez, J.C., et al. (1998) Extraction of membrane proteins by differential solubilization for separation using
67.
68.
69.
70.
71.
72.
73.
74.
45
two-dimensional gel electrophoresis. Electrophoresis, 19, 837–844 Pasquali, C., Fialka, I., and Huber, L.A. (1999) Subcellular fractionation, electromigration analysis and mapping of organelles. J. Chromatogr. B. Biomed. Sci. Appl., 722, 89–102 Burggraf, D., Weber, G., and Lottspeich, F. (1995) Free flow-isoelectric focusing of human cellular lysates as sample preparation for protein analysis. Electrophoresis, 16, 1010–1015 Hochstrasser, A.C., James, R.W., Pometta, D., and Hochstrasser, D. (1991) Preparative isoelectrofocusing and high resolution 2-dimensional gel electrophoresis for concentration and purification of proteins. Appl. Theor. Electrophor., 1, 333–337 Corthals, G.L., Molloy, M.P., Herbert, B.R., Williams, K., and Gooley, A.A. (1997) Prefractionation of protein samples prior to twodimensional electrophoresis. Electrophoresis, 18, 317–323 Herbert, B.R., and Righetti, P.G. (2000) A turning point in proteome analysis: sample prefractionation via multicompartment electrolyzers with isoelectric membranes. Electrophoresis, 21, 3639–3648 Righetti, P.G., Castagna, A., and Herbert, B.R. (2001) Prefractionation techniques in proteome analysis. Anal. Chem., 73, 320A–326A Zuo, X., Echan, L., Hembach, P., Tang, H.Y., Speicher, K.D., Santoli, D., et al. (2001) Towards global analysis of mammalian proteomes using sample prefractionation prior to narrow pH range two-dimensional gels and using one-dimensional gels for insoluble and large proteins. Electrophoresis, 22, 1603–1615 Zuo, X., and Speicher, D.W. (2002) Comprehensive analysis of complex proteomes using microscale solution isoelectrofocusing prior to narrow pH range two-dimensional electrophoresis. Proteomics, 2, 58–68
Chapter 4 Difficult Proteins Ben Herbert and Elizabeth Harry Summary The degree of protein diversity and dynamic range within organisms means that even the simplest proteome cannot be captured by any single extraction and separation step. New techniques have focused on major protein classes often under-represented in proteome analysis; low abundance, membrane, and alkaline proteins. The last decade has seen considerable technology development in fractionation tools aimed at complexity reduction in many forms. The key outcome of complexity reduction is that each fraction, or sub-proteome, can be studied in more detail, and proteins which would have remained undetected in a total extract are present in sufficient quantities. However, the tools available are fractionations, not amplifications, and like all mining for rare and difficult items, a large amount of starting material is often required. The key shortcomings of many proteome analysis techniques are now well documented. With this knowledge, the best modern proteomics ‘platform’ involves combining multiple protein extractions, gel and chromatographic separations, and multiple MS analysis methods. Key words: Low abundance, High abundance, Two-dimensional electrophoresis, Fractionation, Depletion, Affinity, Membrane proteins, Alkaline proteins, Review.
1. Introduction Difficult proteins are those which the biology suggests are present in the sample, but remain undetected in the analysis. The good news is it is probably not about the analysis method. The better news is there are probably sample preparation methods that will deliver the difficult proteins you seek. In most cases those proteins would have separated and been detected just fine, but they were not present in the analysis because they were excluded during sample preparation.
David Sheehan and Raymond Tyther (eds.), Methods in Molecular Biology, Two-Dimensional Electrophoresis Protocols, vol. 519 © Humana Press, a part of Springer Science + Business Media, LLC 2009 DOI: 10.1007/978-1-59745-281-6_4
47
48
Herbert and Harry
In the late 1990s and the early part of this century there were many discussions about the relative merits of the main proteomics methodologies, mainly comparing two-dimensional electrophoresis (2DE) gels to liquid chromatography–mass spectrometry (LC–MS). In most cases 2DE was considered to be a dinosaur: difficult to use and unable to separate important classes of proteins, namely alkaline, membrane, and low-abundance proteins. Important papers were published which painted an unfavourable picture of 2DE as the preferred tool for proteomics compared with LC–MS as the base of modern strategies for high-throughput protein identification (1–3). The early part of this century did not, however, see the death of 2DE, and it became clear that no single sample preparation, fractionation, or separation technology can provide a comprehensive view of any proteome. The current best practice in proteomics involves accessing at least two independent methods for sample preparation and protein separation. Ideally, the separated proteins are analysed by different MS techniques. This is highlighted by Breci et al. (4) who analysed yeast with a combination of LC–MS and gel methods. They clearly illustrate that different subsets of proteins are identified with each of the techniques used. The technology platforms available to proteomics researchers have also progressed in the last 10 years and, whilst skilled operators will always be important, relative novices can obtain highquality data from both MS and gel-based systems. However, to obtain the best results one must know how to treat a sample to extract the desired proteins and present them to the analysis systems. The degree of protein diversity and dynamic range within organisms means that even the simplest proteome cannot be captured by any single extraction and separation step. The last decade has seen considerable technology development in fractionation tools aimed at complexity reduction in many forms. The key outcome of complexity reduction is that each fraction can be studied in more detail, and proteins which would have remained undetected in a total extract are present in sufficient quantities. However, the tools available are fractionations, not amplifications, and like all mining for rare and difficult items, a large amount of starting material is often required. In this chapter we discuss low-abundance, membrane, and alkaline proteins and look at some sample preparation techniques which enable their analysis. In each case we consider fractionation methods which are applicable to both gel-based and LC-based separations.
Difficult Proteins
49
2. Low-Abundance Proteins Most proteins are of low abundance. For example, 50% of the protein content of yeast is the product of 100 genes (5), and this type of distribution is observed in most eukaryotic cells. This means that the remaining protein is distributed over thousands of gene products (5). Ghaemmaghami et al. (6) determined the expression levels of yeast proteins through immunodetection experiments. The study used genome-wide tagging of open reading frames to create fusion proteins with a common high affinity tag. Each protein was expressed from its natural chromosome location, and western blot detection of the common tag was used to quantify expression levels. The group was able to analyse protein expression levels for 4,251 proteins (83% of named ORFs), which were converted into measurements of protein molecules per cell. The abundance of proteins expressed from the yeast genome can range from less than 50 to more than a million molecules per cell. Importantly, 75% of the proteome was present at less than 5,000 proteins per cell (6). The low-abundance proteins in cells are considered to be some of the most important, including regulatory proteins, signal transduction proteins, and receptors. Some organisms are sufficiently well understood to enable the prediction of a protein’s cellular abundance on the basis of the codon usage bias of the respective genes (7). Proteins arising from genes with codon bias index (CBI) or codon adaptation index (CAI) values of less than 0.2 are considered low abundance in yeast, the model organism used for recent high-throughput protein identification studies (1). Low-abundance proteins are almost never seen in traditional 2DE maps because either overwhelming quantities of abundant soluble proteins obscure their detection or they are not completely solubilized. Most 2DE gel-based proteomic studies employ a ‘one extract-one gel’, approach and the majority of proteins identified in these studies are in high abundance because that is what is seen on a single gel (confirmed by Gygi et al.) (1). The advent of immobilized pH gradients (IPGs; see also Chapters “Two-Dimensional Electrophoresis (2DE): An Overview”, “Solubilization of Proteins in 2DE: An Outline”, “High-Resolution 2DE”) has greatly improved the reproducibility of 2DE gels and made it easier for new users to implement high-quality 2DE technology. Many groups now understand that IPG technology can be used to create narrow (2–3 pH units) and very narrow (approx. 1 pH unit) gradients (see Chapter “Selection of pH Ranges in 2DE”) that will enable many more proteins to be resolved – and many of the most useful narrow gradients are commercially available. In addition, the load capacity of narrow-range
50
Herbert and Harry
IPGs is considerably higher than broad-range IPGs, thus enabling the visualization and identification of previously unseen proteins. However, detection of low-abundance proteins in these fractions also requires removal of abundant proteins from the sample. In the following paragraphs we will outline the technical difficulties of working with unfractionated samples and discuss some fractionation methodologies which can be used for mining below the levels of abundant proteins. A total protein extract from most tissues or organisms usually comprises proteins with isoelectric points (pI) of between pH 3 and pH 12. When this total extract is loaded onto a pH gradient covering only 1 or 2 pH units there are many proteins that do not focus within the IPG. These extraneous proteins cause two main problems. First, although it is possible to load many milligrams of total protein onto the narrow-range IPG, the majority of the proteins will have pIs outside the pH range of the IPG. Even in the highly crowded pH 5–6 region, only 35% of the predicted yeast proteome is present, so a whole extract loaded on a pH 5–6 IPG will focus poorly at loads above a few hundred micrograms. The second problem is the severe disturbance to isoelectric focusing (IEF) caused by the proteins which do not focus within the separation range of the IPG. These proteins migrate to the ends of the IPG strip and precipitate in very concentrated zones, which are conductive because the proteins remain charged (8). This phenomenon can be minimized by loading a small amount of total protein onto the IPG strip; however, this is not useful for automated identification of low-abundance proteins unless a highly specialized (and manual) detection and MS strategy is used (9). To eliminate these two problems and resolve high loads of proteins on narrow- and very-narrow-range IPGs, it is essential to pre-fractionate the protein extract into discrete isoelectric fractions that correspond to the pH ranges of the IPGs to be used for 2DE gel (8, 10, 11) (see Chapter “Pre-Fractionation Using Microscale Solution IEF”). For gel-based separations, an ideal method of preparative isoelectric pre-fractionation is separation in a multi-compartment electrolyser (MCE; Fig. 1). This technology uses the same chemistry as IPGs to define a segmented pH gradient. It consists of purifying to homogeneity proteins in a liquid vein, by capturing them in an isoelectric trap formed by two Immobiline membranes having pIs encompassing the pI of the desired fraction (8). The protein sample for fractionation is prepared, as for IEF in IPGs, and loaded into one chamber of an MCE which is usually configured with between three and six sample chambers. Following preparative IEF each fraction yields protein samples fully compatible with the IPG step, namely devoid of salts and other typical components of biological fluids, such as fatty acids, amines etc. (8). Clearly, sample fractionation is required to
Difficult Proteins
51
Fig. 1. Schematic diagram of an MCE. This is a commercially available small volume (2.5–5 mL/chamber) instrument with no sample recirculation; instead each chamber is stirred using magnetic stirrer bars driven by rotating magnets embedded in the peltier cooling plate. The chambers are sealed using pH-modified membranes which are matched to the endpoints of commonly used narrow-range IPGs. This creates a stepwise pH gradient and complex mixtures are fractionated into pH segments. Not recirculating the chamber contents speeds up the fractionation by continuously exposing the whole sample to the electric field.
reduce the complexity of cellular extracts and enrich for particular classes of proteins. One of the simplest fractionation methods is sequential extraction which takes advantage of the highly soluble nature of many abundant proteins. An initial extract with an aqueous buffer leaves an insoluble pellet to be extracted with stronger reagents (2, 12). The advantage of this method is that all extraction steps can be performed in a single tube, thus minimizing losses. There has been considerable discussion of the technical issues associated with the detection of low-abundance proteins in a number of recent publications (1, 2, 5, 9, 13–15). A major issue concerns the dynamic range of protein concentration that one can expect to encounter in proteomics. For the sake of clarity, it is essential to differentiate between cellular dynamic range and protein abundance in extra-cellular fluids such as plasma. For yeast, the most abundant proteins are present at around 2,000,000 copies per cell which represents 4% of the total protein, whereas the least abundant proteins are present at around 100 copies per cell, a dynamic range of only four orders of magnitude (5). However, in extra-cellular fluids the situation is considerably more difficult, and Corthals et al. (13) have predicted that the dynamic range of proteins in plasma could cover up to 12 orders of magnitude,
52
Herbert and Harry
although this is a rather skewed distribution because of the presence of albumin which represents over 50% of the total protein, by weight. Similarly, it is important to remember that any tissue or even whole organism will not express the whole genome at any one time. In genome-wide expression profiling of Escherichia coli (16) and Saccharomyces cerevisiae (17), transcripts for only 75% of these genomes have been detected by combining data from all phases of the cell cycle under normal culture conditions. In yeast grown to log phase, transcripts for 52% of the genome have been detected and in E. coli grown in rich media at 37°C, around 25% of the genes are expressed at detectable levels (16, 17). However, in a more recent study on yeast, Ghaemmaghami et al. (6) found that 80% of the predicted proteome was expressed under normal growth conditions. Partly, the discrepancy in the predicted proteome is due to differences in the number of ORFs which are considered, i.e. whether all ORFs or only named ORFs are included. Thus, it is clear that, for any proteome project, predictions of expected protein numbers should be considerably reduced from those obtained using the whole genome and the latest ORF predictions should be considered. 2.1. ElectrophoresisBased Fractionation Methods
Preparative electrophoresis, especially IEF, methodologies have recently experienced a revival. This is due to the high resolving power of IEF and the use of pI as a powerful tool in refining shotgun LC–MS data (18). Some of the major techniques are briefly reviewed, but it should be kept in mind that the preferred tools will be those that exploit IEF techniques, since the resulting fractions can be directly interfaced with 2DE.
2.1.1. Continuous Electrophoresis in Free Liquid Films
Compared to gel-based separations, this technique has two main advantages. Much higher sample loads can be applied, and certain artefactual protein modifications induced by free monomers in polyacrylamide gels are eliminated. Present equipment derives from the concepts and instrumentation of Hannig (19), in which an electrolyte solution flows in a direction perpendicular to an electric field and the mixture undergoing separation is added continuously at a small entry point in the flowing medium. Components of the mixture are deflected in diagonal trajectories according to their electrophoretic mobility and can be collected at the bottom of the device into a maximum of 96 fractions. Freeflow electrophoresis (FFE) was born as a technique for purifying cells and sub-cellular organelles, which could be recovered with high purity due to their very low diffusion coefficients. Whilst of use for purification of organelles, FFE is not ideal for the pre-fractionation of proteins. This is due to proteins’ higher diffusion coefficients, as compared with cells and organelles. For protein separation, FFE provides improved separation in the IEF
Difficult Proteins
53
mode (FF-IEF), due to built-in forces impeding entropic peak dissipation. FF-IEF can also be used as the first dimension of a 2DE map, the eluted fractions being directly analysed by orthogonal SDS–PAGE. In use, one advantage of FF-IEF was immediately evident: large proteins such as vinculin (117 kDa) could be well recovered and easily identified. In contrast, recovery of high mass proteins has always been problematic in IPG gels (20) because they are more difficult to resolubilize after IEF. 2.1.2. Rotationally Stabilized Focusing Apparatus: The Rotofor
The rotofor is a device that separates a liquid protein sample into multiple fractions by IEF. The device is assembled using an acrylic tube, separated into 20 sample chambers using liquid-permeable nylon screens. Cation- and anion-exchange membranes are placed against the anodic and cathodic chambers, respectively, so as to prevent diffusion of electrode-associated by-products into the separation chambers. The narrow-pI range fractions can then be used to generate conventional 2DE maps. In recent times the Rotofor has been used as the first dimension of a combined electrophoresis/chromatography two-dimensional fractionation methodology in which each isoelectric fraction was further analysed by hydrophobic interaction chromatography, using nonporous reversed-phase HPLC (21). Each protein peak collected from the HPLC column was digested to peptides with trypsin, subjected to MALDI-TOF MS analysis, and identified by database searching.
2.1.3. Sample Pre-fractionation Via Multi-compartment Electrolysers with Isoelectric Membranes
Multi-compartment electrolysers (MCE) were first introduced as a class of instruments based on conventional IEF in the presence of the soluble, amphoteric, carrier ampholyte buffers. However, MCEs based on immobiline membranes represent a quantum leap over previous techniques (22, 23). This relies on isoelectric membranes that are fabricated with the same acrylic monomers used in IPG gels. The advantages of such a method are immediately apparent. Firstly, such a device produces fractions of the proteome that are fully compatible with first-dimension separation in 2DE maps, a focusing step also based on IPG technology. Secondly, it permits a sample to be fractionated into proteins of precise pI values, such that they fall within the pH gradient of any narrow- or wide-range IPG strip. The chances of protein precipitation occurring will thus be much reduced. In fact, when an entire cell lysate is analysed in a wide gradient, there are few risks of protein precipitation. In contrast, when the same mixture is analysed in a narrow gradient, massive precipitation of all non-isoelectric proteins can occur. Furthermore, due to the fact that only proteins co-focusing in the same IPG interval will be present, much higher sample loads can be used, permitting the detection of low-abundance proteins (Fig. 2).
54
Herbert and Harry
Fig. 2. (a) Unfractionated rat brain extract separated on an 11-cm pH 3–10 IPG. (b) Three MCE fractions of rat brain extract separated on 11-cm pH 3–10 IPGs to illustrate the isoelectric fractionation. The MCE fractions are, from left to right; pH 3–5, 5–6 and 6–11.
The MCE technique continues to undergo improvement. Whilst initial models of the MCE were large, a miniaturized version of the MCE has been reported (8). A remedy to the slow migration of proteins in MCEs because of the sieving effect of isoelectric membranes has also been proposed (24) by Fortis et al., via the introduction of isoelectric beads (25). These beads were made using ionic acrylamide derivative monomers co-polymerized within the pores of a central ceramic hard core, minimizing mass transfer resistance of proteins that are transiently adsorbed onto the beads. As a result, significantly reduced separation time was shown along with a very low electroendosmotic flow. A recent study of yeast membrane and low-abundance proteins illustrates how combining fractionation steps and including MCE enable many low-abundance proteins to be detected. Pedersen et al. (26) employed a membrane preparation method consisting of direct cell lysis in 100 mM sodium carbonate to dissociate many of the peripheral proteins from the membrane which resulted in a high yield of integral membrane proteins (27). After membrane recovery by ultracentrifugation, the supernatant contains most of the abundant, soluble proteins and the pellet is enriched for cell and organellar membrane and lowabundance proteins. From this membrane-enriched fraction, which was MCE fractionated and displayed across three different pH gradients, MALDI-TOF MS was used to identify 780 protein isoforms, representing 323 gene products, including 28% lowabundance proteins [with codon adaptation index (CAI) values <0.2] and 49% of these were membrane or membrane-associated proteins. Figure 3 illustrates the study data compared to the CAI profile of the whole yeast genome. 2.1.4. Miniaturized Isoelectric Separation Devices
A simple variant of preparative IEF called ‘off-gel IEF’ has been described. Just like the multi-compartment separation technique, the system has been devised for the separation of proteins according to their pI and for their direct recovery in solution without adding buffers or ampholytes. The principle is to place a sample
Difficult Proteins
55
Fig. 3. (a) The codon adaptation index (CAI) for the whole yeast genome, divided in columns of 0.1. The X-axis indicates the number of genes in each CAI slice. The left bar represents the genes with CAI between 0.0 and 0.1, which will produce the lowest-abundance proteins. 80% of all yeast genes have CAI values less than 0.2 which indicates they will produce low-abundance proteins. (b) The combined data for 780 protein IDs from Pedersen et al. (24, 25) showing that, when membrane preparation and IEF are combined, it is possible to identify many low-abundance proteins.
in a liquid chamber, positioned to be touching an IPG gel. After separation, only the globally neutral species with pI equal to the pH of the IPG gel remain in solution. In a further extension of this initial work, the system was improved and adapted to a multi-well device, composed of a series of compartments of small volume (100–300 μL) and compatible with current instruments for separation (28). As discussed earlier, a major new use of this type of electrophoretic fractionation involves the separation of protein mixtures or peptide digests as part of an LC–MS strategy. Compared with cation-exchange chromatography, IPG IEF provides higher resolution separation and experimentally derived pI information, which can be used as a filter criterion for tandem mass spectral data validation (18). Importantly, the proteins or peptides are available in liquid fractions, thus eliminating the need for gel-based enzyme digests and peptide extractions. 2.2. Affinity Sample Pre-fractionation Methods 2.2.1. High-Abundance Protein Depletion
For the analysis of serum, plasma, and cerebrospinal fluid a common approach is to remove high-abundance species prior to analysis (29). It is thus possible to load more of the depleted sample and therefore to observe a significantly higher number of species, including many that would have escaped detection in an untreated sample. Several methods are described for removal of high-abundance proteins; all of them derive from very classical approaches in the domain of affinity chromatography. A variety of commercial products are available to deplete albumin and up to 19 other high-abundance proteins using specific immunosorbents – mostly antibodies on chromatographic bead supports. Despite the widespread adoption of depletion strategies, some studies have reported that the removal of high-abundance species
56
Herbert and Harry
causes a dramatic loss of low-abundance proteins due to nonspecific binding. Shen et al. (30) observed that, during human serum albumin (HSA) depletion, another 815 species (not including HSA) were co-depleted as well. When capturing IgGs, another 2,091 species (not including IgG) were also co-depleted, among which 56% were antibody sequences, while the other 44% included low-abundance cytokines and related proteins. Paradoxically, in the depleted sera, only 1,391 proteins could be detected. Of this minority population, 269 had been co-depleted with the albumin fraction and another 568 were in common with the IgG group. When considering that there was a total of 4,590 unique gene products (30), it is ironic that most of them were discovered in the albumin and IgG fractions rather than in the depleted fraction! The issue of co-depletion in affinity depletion prefractionation procedures is now becoming common knowledge. Colantonio et al. (31) discuss the problem of reaching a balance between total depletion of IgGs, with concomitant loss of other proteins, versus the partial removal of IgGs, for minimizing loss of other species. 2.2.2. Dynamic Range Equalization Via Hexapeptide Libraries
The use of combinatorial peptide libraries provides an alternative, non-antibody, strategy for embracing all the protein species in a given proteome. This approach now seems to be a reality, and it is embodied in the concept of protein equalizer technology, commercially available as ProteoMiner™, a combinatorial hexapeptide library bound to chromatographic beads. The original article, outlining the synthesis of the beads and some of their fundamental properties, has recently been published together with some reviews describing some of the very basic concepts (32–34). The ProteoMiner™ approach consists of a solid-phase combinatorial library of hexapeptides, that is synthesized via a short spacer on poly(hydroxymethacrylate) beads. The ligands are represented throughout the beads’ porous structure and can reach a concentration of 50 pmoles per mL of beads of the same hexapeptide distributed throughout the core of the pearl. This amounts to a ligand density of ca. 40–60 μmoles per mL of bead volume (average bead diameter of about 70 μm). Each single bead, thus, has millions of copies of a single, unique ligand and each bead, potentially, has a different ligand from every other bead. Considering that, for the synthesis, the 20 natural amino acids are used, this means that the library contains a population of linear hexapeptides amounting to 206, i.e. 64 million different ligands. Such a vastly heterogeneous population of hexapeptides means that, in principle, an appropriate volume of beads should contain a partner able to interact with just about any protein present in a complex proteome, in a biological fluid or a tissue or cell lysate of any origin.
Difficult Proteins
57
Fig. 4. Fluorescently stained 2DE images of human plasma separated on 11-cm pH 4–7 IPGs and Bio-Rad Criterion 4–12% second-dimension gels. The control gel (left) was loaded with 220 μg of undepleted plasma. The ProteoMiner gel (right) was loaded with 220 μg of plasma proteins eluted from the hexapeptides. This load represents approximately 10% of the eluted protein from 1 mL of starting plasma.
Up to 2004, the only extensive data set available on serum proteins was that of Anderson et al. (35) who published a compilation of 1,175 non-redundant species collated from several sources. Although this was an impressive total, it was quickly superseded by the massive effort of the HUPO Plasma Protein Project (PPP). The PPP was started in 2002 as a network of 35 collaborating laboratories and multiple analytical groups. Via these combined efforts, a core set of 3,020 serum plasma proteins was generated and is now available on public databases (www. bioinformatics.med.umich.edu/hupo/ppp). There have been an additional 6,484 unique proteins identified via only a single peptide (thus with a rather low confidence level, ca. 60–75%) bringing the total to an extraordinary level of 9,504. Sennels et al. (36) have reported results with the human serum proteome, as captured with a single, simple experimental protocol exploiting a diffusible bag of ProteoMiner™ beads. In this experiment 300 mL of serum was processed and subjected to adsorption with the ligand library. Upon elution with three different eluants and analysis via FT-ICR MS, a total of 3,661 unique gene products could be recognized, with a confidence level of 95% (Fig. 4).
3. Membrane Proteins Very few membrane proteins are seen on 2DE gels when conventional sample preparation methods are used. Under such conditions hydrophobic membrane proteins aggregate when they concentrate at their pI and are not readily solubilized by SDS for
58
Herbert and Harry
entry into the second dimension. Rabilloud’s group (CEA, Grenoble, France; see also Chapter “Solubilization of Proteins in 2DE: An Outline”) and others have shown that strong solubilizing agents greatly reduce this problem (27, 37). However, improved solubility conditions alone are not sufficient significantly to increase the recovery of membrane proteins on 2DE gels. Fractionation, in combination with the correct solubilizing reagents, produces sample extracts highly enriched for membrane proteins. Sequential extraction of proteins from a sample by increasing protein solubilizing conditions at each step is an effective strategy for first removing the more abundant soluble proteins and concentrating the less abundant and less soluble membrane proteins. Molloy et al. (12) used sequential extraction and 2DE gels with MALDI-TOF MS to detect 11 integral outer membrane proteins from E. coli. Although these proteins contained up to seven transmembrane domains they were not hydrophobic as determined by their grand average of hydropathy (GRAVY) scores and were readily solubilized after the more abundant soluble proteins had been removed. Another type of sequential extraction where the membrane was stripped of peripheral proteins by incubation in sodium carbonate at pH 11 was used in a study of E. coli, and here 80% of the integral outer membrane proteins were detected on a 2DE gel (27). In their large 2DE gel-based study of yeast membrane proteins, Pedersen et al. used preparative isoelectric focusing in an MCE to obtain acidic, neutral, and alkaline pH fractions of sodium-carbonate-stripped membrane proteins (24). This strategy enabled high protein loads on narrow pH gradients. Analysis of the pH 7.5–10.5 fraction resulted in identification of 237 proteins, representing 93 genes, including proteins from 30 (32%) genes with CAI < 0.2, integral membrane proteins from 20 (22%) genes, and membrane-associated proteins from 10 (11%) genes. All the proteins identified in the pH 7–10.5 fraction were also predicted to have alkaline pI values. In total, Pedersen’s study identified 800 proteins, of which 323 were unique proteins. From the unique proteins, 105 were integral membrane proteins (33%), 54 were membrane-associated proteins, and 90 were low-abundance proteins (28%) with CBI below 0.2. Of the 105 integral membrane proteins, the majority had fewer than four transmembrane domains and no proteins with more than six transmembrane domains were detected. To summarize, the last decade of 2DE-based membrane proteomics – the methods and reagents for analysing membrane proteins on 2DE gels – has developed to the point where large studies are possible. However, these methods require large amounts of starting material and multiple fractionation steps. At this stage, it is not common to detect proteins with more than six transmembrane domains on 2DE gels.
Difficult Proteins
59
Many studies incorporating membrane proteomics conducted in the last few years have adopted a simpler one-dimensional SDS–PAGE (1DE) approach to membrane protein analysis. In this strategy the membrane preparation is solubilized by SDS and separated by 1DE. Entire separation lanes are excised into many small bands, without the need for complete staining. Bands are processed for LC–MS following standard methodology (38). In another powerful illustration of combining multiple extraction, gel and chromatographic separation, and multiple MS analysis methods, Wolff et al. have identified almost 50% of the predicted proteome of Bacillus subtilis (39). They combined 2DE gels and MALDI MS with two-dimensional LC–MS using iTRAQ to analyse the soluble proteins from heat shock experiments. The membrane fractions were analysed by 1DE followed by LC–MS of the excised bands. Using this method they were able to identify 453 proteins in the membrane fraction, of which 232 contained transmembrane domains. This indicates that the sodium carbonate membrane stripping method is not successful in removing all non-transmembrane proteins, a result similar to that found by Pedersen et al. (24). Only 140 out of a total of 1,218 proteins were identified by all three methods used, which is expected because the two-dimensional and LC–MS methods were used to analyse soluble proteins. In a positive outcome for 2DE, they found a good correlation between iTRAQ and 2DE gel quantitation with an R2 of 0.817.
4. Alkaline Proteins Alkaline proteins have been particularly difficult to resolve on 2DE gels for two reasons. Firstly, in sample preparation for 2DE the standard procedure adopted up to the present calls for reduction prior to the IEF/IPG step, followed by a second reduction and alkylation step in the equilibration solution between the first and second dimension, in preparation for the SDS–PAGE step. This protocol is far from being optimal, due to incomplete reduction during IEF and often results in a large number of spurious multimeric spots, due to ‘scrambled’ disulphide bridges among like and unlike chains. Due to the negative charge on the –SH group of typical reducing agents such as DTT, this compound would act as a buffer and will keep migrating inside the pH gradient (towards the anode) and be arrested by protonation, around pH 7. Thus, artefacts arising from incomplete reduction are more often observed in the alkaline portion of the IPG. Even tributyl phosphine, a strong non-thiol reducing agent, does not appear to have the reducing power to maintain all proteins as monomeric
60
Herbert and Harry
polypeptides during IEF. A recent series of papers showed that the number of such artefactual spots can be impressively large in alkaline IPGs, even in the case of simple polypeptides such as the human α- and β-globin chains, which possess only one (α-) or two (β-) –SH groups (40–42). In addition, failure to alkylate proteins prior to IEF can result in a substantial loss of spots on 2DE, probably due to the fact that proteins, at their pI, regenerate disulphide bridges with concomitant formation of macroaggregates which become entangled with and trapped within the polyacrylamide gel fibres. This strongly quenches their transfer in the subsequent SDS–PAGE step. The importance of cysteine alkylation in obtaining high-quality separations in alkaline IPGs has been confirmed by Luche et al. (43). They adopted two strategies for cysteine blocking: classical alkylation with maleimide derivatives and mixed disulphide exchange with an excess of a low molecular weight disulphide. Good resolution of basic proteins was obtained with in-gel rehydration on wide gradients (e.g. 3–10 and 4–12), but anodic cup loading produced the best resolution for narrow-range basic gradients (e.g. 6–12 or 8–12). Our group has further simplified the methodology of reduction and alkylation by using acrylamide monomer as an alkylating agent instead of iodoacetamide (41–43). In fact, at the alkaline pH values at which alkylation is customarily performed, iodoacetamide and acrylamide have similar reactivity rates on ionized –SH groups. Acrylamide does not react with TBP, so the reaction can be carried out in a single step, which results in a significant saving in time. Furthermore, acrylamide does not seem to react with thiourea or amino acid groups other than cysteine. A second major issue which prevents high resolution on alkaline IPGs is the isoelectric distribution of the extract. The majority of cellular extracts are biased towards acidic proteins, and therefore whole-cell extracts do not resolve well on alkaline pH gradients, especially at the high protein loads required for protein characterization. Whilst low protein loads often enable acceptable resolution to be obtained with total extracts on narrow pH-range alkaline gels, many proteins are then not present in sufficient quantity for routine MS analysis. As discussed in Subheading 2, the majority of species in every proteome are present at low abundance, so using low protein loads to obtain high-quality separations would seem counterproductive. One way to solve this is to remove the acidic protein fraction and load higher amounts of the remaining alkaline protein fraction onto 2DE gels. Preparative isoelectric pre-fractionation enables the isolation of a sample enriched for alkaline proteins and high protein loads on alkaline pH gradients (Fig. 5) (8).
Difficult Proteins
61
Fig. 5. Rat brain alkaline MCE fraction from Fig. 2 separated on an 11-cm pH 6–11 IPG. The gel was stained with Sypro Ruby.
5. Concluding Remarks Publication of the human genome (44, 45) has focused attention on complexity and the issue of proteomes, their diversity, and their relationship to genomes. Proteomics technology has matured to the level where the identification of hundreds of proteins in a single experiment is routine and thousands of proteins in a single experiment will be routine within the next decade. The next generation of proteomics will be an integral part of systems biology, studying modifications at the transcriptional, translational, and post-translational levels to define proteome complexity. The number of sequenced genomes has made protein identification a relatively simple task and the focus is now on characterization of protein isoforms. To this end, there are a number of vital methodologies, such as IEF and, to a lesser extent, 2DE, that cannot be omitted without severely limiting the utility of the data. Fractionation at the sub-cellular level or with sequential extraction, and in combination with isoelectric fractionation, are powerful tools for mining below the most abundant 10% of the proteome. The separation of co- and post-translationally modified forms of proteins, which reflect the true functional state of the proteome, is essential, and IEF and 2DE remain key parts of the technology platform.
62
Herbert and Harry
References 1. Gygi, S.P., Corthals, G.L., Zhang, Y., Rochon, Y., and Aebersold, R. (2000). Evaluation of twodimensional electrophoresis-based proteome analysis. Proc. Natl. Acad. Sci. U.S.A. 97, 9390–9395 2. Washburn, M.P., Wolters, D., and Yates, J.R. (2001) Large-scale analysis of the yeast proteome by multidimensional protein identification technology. Nat. Biotechnol. 19, 242–247 3. Patterson, S.D. (2000) Proteomics: the industrialization of protein chemistry. Curr. Opin. Biotechnol. 11, 413–418 4. Breci, L., Hattrup, E., Keeler, M., Letarte, J., Johnson, R., and Haynes, P.A. (2005) Comprehensive proteomics in yeast using chromatographic fractionation, gas phase fractionation, protein gel electrophoresis, and isoelectric focusing. Proteomics 5, 2018–2028 5. Futcher, B., Latter, G.I., Monardo, P., McLaughlin, C.S., and Garrells, J.I. (1999) A sampling of the yeast proteome. Mol. Cell. Biol. 19, 7357–7368 6. Ghaemmaghami, S., Huh, W.K., Bower, K., Howson, R.W., Belle, A., Dephoure, N. et al. (2003). Global analysis of protein expression in yeast. Nature 425, 737–741 7. Kurland, C.G. (1991) Codon bias and gene expression. FEBS Lett. 285, 165–169 8. Herbert, B., and Righetti, P.G. (2000) A turning point in proteome analysis: sample prefractionation via multicompartment electrolysers with isoelectric membranes. Electrophoresis 21, 3639–3648 9. Fey, S.J., and Larsen, P.M. (2001) 2-D or not 2-D. Two-dimensional gel electrophoresis. Curr. Opin. Chem. Biol. 5, 26–33 10. Righetti, P.G., Castagna, A., and Herbert, B. (2001) Prefractionation techniques in proteome analysis. Anal. Chem. 73, 320A–326A 11. Zuo, X., and Speicher, D.W. (2000) A method for global analysis of complex proteomes using sample prefractionation by solution isoelectrofocusing prior to two-dimensional electrophoresis. Anal. Biochem. 284, 266–278 12. Molloy, M.P., Herbert, B.R., Walsh, B.J., Tyler, M.I., Traini, M., Sanchez, J.C. et al. (1998) Extraction of membrane proteins by differential solubilization for separation using two-dimensional gel electrophoresis. Electrophoresis 19, 837–844 13. Corthals, G.L., Wasinger, V.C., Hochstrasser, D.F., and Sanchez, J.C. (2000) The dynamic range of protein expression: a challenge
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
for proteomic research. Electrophoresis 21, 1104–1115 Fountoulakis, M., Takács, M.F., Berndt, P., Langen, H., and Takács, B. (1999) Enrichment of low abundance proteins of Escherichia coli by hydroxyapatite chromatography. Electrophoresis 20, 2181–2195 Fountoulakis, M., Takács, M.F., and Takács, B. (1999) Enrichment of low-copy-number gene products by hydrophobic interaction chromatography. J. Chromatogr. A 833, 157–168 Richmond, C.S., Glasner, J.D., Mau, R., Jin, H., and Blattner, F.R. (1999) Genome-wide expression profiling in Escherichia coli K-12. Nucleic Acids Res. 27, 3821–3385 Velculescu, V.E., Zhang, L., Zhou, W., Vogelstein, J., Basrai, M.A., Bassett, D.E. Jr. et al. (1997) Characterization of the yeast transcriptome. Cell, 88, 243–251 Heller, M., Ye, M., Michel, P.E., Morier, P., Stalder, D., Jünger, M.A. et al. (2005) Added value for tandem mass spectrometry shotgun proteomics data validation through isoelectric focusing of peptides. J. Proteome Res. 4, 2273–2282 Hannig, K. (1978) Continuous free-flow electrophoresis as an analytical and preparative method in biology. J Chromatogr. 159, 183–191 Hoffmann, P., Ji, H., Moritz, R.L., Connolly, L.M., Frecklington, D.F., Layton, M.J. et al. (2001) Continuous free-flow electrophoresis separation of cytosolic proteins from the human colon carcinoma cell line LIM 1215: a non twodimensional gel electrophoresis-based proteome analysis strategy. Proteomics 1, 807–818 Xiao, Z., Conrads, T.P., Lucas, D.A., Janini, G.M., Schaefer, C.F., and Buetow, K.H. (2004) Direct ampholyte-free liquid-phase isoelectric peptide focusing: application to the human serum proteome. Electrophoresis 25, 128–133 Righetti, P.G., Wenisch, E., and Faupel, M. (1989) Preparative protein purification in a multi-compartment electrolyzer with immobiline membranes. J. Chromatogr. 475, 293–309 Righetti, P.G., Wenisch, E., Jungbauer, A., Katinger, H., and Faupel, M. (1990) Preparative purification of human monoclonal antibody isoforms in a multicompartment electrolyzer with Immobiline membranes. J. Chromatogr. 500, 681–696 Pedersen, S.K., Harry, J.L., Sebastian, L., Baker, J., Traini, M.D., McCarthy, J.T. et al. (2003) Unseen proteome: mining below the
Difficult Proteins
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
tip of the iceberg to find low abundance and membrane proteins. J. Proteome Research 2, 303–312 Fortis, F., Girot, P., Brieau, O., Boschetti, E., Castagna, A., and Righetti, P.G. Amphoteric buffering chromatographic beads for proteome prefractionation. I: Theoretical model. Proteomics 5, 620–628 Molloy, M.P., Herbert, B.R., Slade, M.B., Rabilloud, T., Nouwens, A.S., Williams, K.L., and Gooley, A.A. (2000) Proteomic analysis of the Escherichia coli outer membrane. Eur. J. Biochem. 267, 2871–2881 Michel, P.E., Reymond, P., Arnaud, I.L., Josserand, J., Girault, H.H., and Rossier, J.S. (2003) Protein fractionation in a multicompartment device using Off-Gel isoelectric focusing. Electrophoresis 24, 3–11 Hörth, P., Miller, C.A., Preckel, T., and Wenz, C. (2006) Efficient fractionation and improved protein identification by peptide OFFGEL electrophoresis. Mol. Cell. Proteomics 10, 1968–1974 Liu, T., Qian, W.J., Gritsenko, M.A., Xiao, W., Moldawer, L.L., Kaushal, A. et al. (2006) High dynamic range characterization of the trauma patient plasma proteome. Mol. Cell. Proteomics 10, 1899–1913 Shen, Y., Kim, J., Strittmatter, E.F., Jacobs, J.M., Camp II, D.G., Fang, R. et al. (2005) Characterization of the human blood plasma proteome. Proteomics 5, 4034–4045 Colantonio, D.A., Dunkinson, C., Bovenkamp, D.E., and Van Eyk, J.E. (2005) Effective removal of albumin from serum. Proteomics 5, 3831–3835 Thulasiraman, V., Lin, S., Gheorghiu, L., Lathrop, J., Lomas, L., Hammond, D., and Boschetti, E. (2005) Reduction of the concentration difference of proteins in biological liquids using a library of combinatorial ligands. Electrophoresis 26, 3561–3571 Righetti, P.G., Castagna, A., Antonucci, F., Piubelli, C., Cecconi, D., Campostrini, N. et al. (2005) Proteome analysis in the clinical chemistry laboratory: myth or reality? Clin. Chim. Acta 357, 123–139 Righetti, P.G., Boschetti, E., Lomas, L., and Citterio, A. (2006) Protein Equalizer Technology: the quest for a “democratic proteome”. Proteomics 6, 3980–3992 Anderson, L.N., Polanski, M., Pieper, R., Gatlin, T., Tirumalai, R.S., Conrads, T.P.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
63
et al. (2004) The human plasma proteome: a nonredundant list developed by combination of four separate sources. Mol. Cell. Proteomics 3, 311–326 Sennels, L., Salek, M., Lomas, L., Boschetti, E., Righetti, P.G., and Rappsilber, J. (2007) Proteomic analysis of human blood serum using peptide library beads. J Proteome Res. 6, 4055–4062 Santoni, V., Molloy, M., and Rabilloud, T. (2000) Membrane proteins and proteomics: un amour impossible? Electrophoresis 21, 1054–1070 Eymann, C., Dreisbach, A., Albrecht, D., Bernhardt, J., Becher, D., Gentner, S. et al. (2004) A comprehensive proteome map of growing Bacillus subtilis cells. Proteomics 4, 2849–2876 Wolff, S., Otto, A., Albrecht, D., Zeng, J.S., Buttner, K., Gluckmann, M. et al. (2006) Gelfree and gel-based proteomics in Bacillus subtilis: a comparative study. Mol. Cell. Proteomics 7, 1183–1192 Herbert, B., Galvani, M., Hamdan, M., Oliveri, E., McCarthy, J., Pedersen, S. et al. (2001) Reduction and alkylation of proteins in preparation of two-dimensional map analysis: why, when and how? Electrophoresis 22, 2046–2057 Galvani, M., Hamdan, M., Herbert, B., and Righetti, P.G. (2001). Reduction and alkylation of proteins in preparation of twodimensional map analysis: a kinetic study via MALDI-TOF MS. Electrophoresis 22, 2058–2065 Galvani, M., Rovatti, M., Hamdan, M., Herbert, B., and Righetti, P.G. (2001). Protein alkylation in presence/absence of thiourea in proteome analysis: A MALDI-TOF mass spectrometry investigation. Electrophoresis 22, 2066–2074 Luche, S., Diemer, H., Tastet, C., Chevallet, M., Van Dorsselaer, A., Leize-Wagner, E. et al. (2004). About thiol derivatization and resolution of basic proteins in two-dimensional electrophoresis. Proteomics 4, 551–561 Lander, E.S., Linton, L.M., Birren, B., Nusbaum, C., Zody, M.C., Baldwin, J. et al. (2001). Initial sequencing and analysis of the human genome. Nature 409, 860–921 Venter, J.C., Adams, M.D., Myers, E.W., Li, P.W., Mural, R.J., Sutton, G.G. et al. (2001). The sequence of the human genome. Science 291, 1304–1351
Chapter 5 Organelle Proteomics Matthias Plöscher, Bernhard Granvogl, Veronika Reisinger, Axel Masanek, and Lutz Andreas Eichacker Summary The proteome of the cell is at the frontier of being too complex for proteomic analysis. Organelles provide a step up. Organelles compartmentalize the cell enabling a proteome, physiology and metabolism analysis in time and in space. Protein complexes separated by electrophoresis have been identified as the next natural level to characterize the organelles’ compartmentalized membrane and soluble proteomes by mass spectrometry. Work on mitochondria and chloroplasts has shown where we are in the characterization of complex proteomes to understand the network of endogenous and extrinsic factors which regulate growth and development, adaptation and evolution. Key words: Organelle, Proteome, Electrophoresis, Mass spectrometry, Chloroplast, Mitochondria, Review.
1. Introduction All organisms rely on proteins to exist and solve biological problems. Proteins organize themselves into functional metabolic and regulatory pathways localized in a cell, tissue, or whole organism. To address the highly diverse range of protein functions, researchers today seek no less than identification of all the proteins and characterization of all the qualitative and quantitative changes in the cell’s complement of proteins. The information output complements the proteome; the information source is proteomics. A proteome includes information about the protein composition in space and time and about changes induced in proteins by internal or external signals. In proteomics, cells are of special interest since they represent the lowest biological organization state that can be David Sheehan and Raymond Tyther (eds.), Methods in Molecular Biology, Two-Dimensional Electrophoresis Protocols, vol. 519 © Humana Press, a part of Springer Science + Business Media, LLC 2009 DOI: 10.1007/978-1-59745-281-6_5
65
66
Plöscher et al.
maintained in a laboratory in culture. At the level of protein, three distinct steps are performed in proteomic investigations. Single species of proteins or protein complexes are isolated from protein mixtures, proteins are broken down into peptides, and peptides are identified by mass spectrometry (MS). The elucidation of the cell proteome from peptides is a challenging goal. At present, the proteome of a cell is just at the edge of being too complex to be analysed with all the tools and protocols employed in proteomics. Here, organelles provide a crucial step up. Organelles compartmentalize and organize the cell. A focus on organelle proteomes provides a basis to study the organization of the cell proteome in space. The defined developmental programs of the cell then provide the basis to understand the organization of the proteomic data in time. Finally, species-dependent changes of the proteome organelles indicate steps in the molecular evolution of the cell. Since, however, the complexity of the cellular protein complement is still much too high for in-depth proteomic analysis using the technical means we have at hand today, we will follow the natural staircase of cellular sub-compartmentalization. Cells are first fractionated into organelles, then high-quality MS-based proteomic data are obtained from proteins or peptides prepared by either ‘gel-based’ or ‘gel-free’ technology.
2. Isolation of Subcellular Organelles The starting point of any practical approach to carry out organelle proteomics is the selection of a suitable protocol for isolation of the organelle of interest. In most approaches, the tissue of interest is disrupted followed by mechanical destruction of the cell. Since the target of the whole subfractionation procedure is an organelle, often surrounded by a single or double lipid bilayer membrane, the mechanical stability of tissues and cells and the osmotic conditions of the cell matrix are of major importance. Stability varies widely between the different classes of organisms (e.g. plant, animal, fungi, or bacteria) and even within one organism (e.g. flower and stem within plants, or blood cells and bone cells in vertebrates). Forces applied to the organism, the tissue, and the cell to release organelles therefore have to be adopted carefully, and the vast majority of protocols for isolation of organelles start with the crude cutting of the starting material. This step preceding homogenization or grinding is often already performed in a buffer providing isoosmotic conditions for the organelles (1). Since the shear forces applied lead to a progressive destruction of the biological supramolecular architecture, the exposure time as well as the strength of these forces has to be optimized; other-
Organelle Proteomics
67
wise large-scale destruction of organelles cannot be avoided and organelle yield is compromised. As an alternative to mechanical shear forces for tissue disruption, cells may be released from tissues by an enzymatic approach. Peptide bond linkages between cells may be digested using one enzyme or enzyme cocktails, releasing intact cells into the medium. For sorting, individual cells in an isoosmotic medium can be handled like single-celled organisms. Sorting of cells can be performed using physical properties such as size, surface charge, colour, and fluorescence or chemical differences, e.g. specific surface molecules (2). After isolation of the desired cell population, cells are lysed to release organelles. Methods for lysis depend on the structural architecture of the cell. The presence and composition of a cell wall determines whether an osmotic swelling or treatment with detergents is sufficient for the disruption of the outer barrier or if mechanical shear forces have to be applied. In the presence of a cell wall, a particularly gentle way to free organelles from cells is digestion of the cell wall with enzymes to release protoplasts which are subsequently lysed using osmosis or to apply detergents to break up the membrane barrier. The resulting homogenate is the basis for separation of the different species of organelles by gradient centrifugation, free-flow electrophoresis, or immunoisolation (3). With respect to versatility and ease of use, gradient centrifugation has been established as the most widely used approach. The instrumental requirements and costs for maintenance are low, and the method is applicable to almost all types of organelles. The procedure is based on differences in organelle density and size. These differences give a distinct sedimentation pattern in a continuous gradient after centrifugation, theoretically allowing for a separation of the organelles in one step. In practice, the density of all types of organelles differs too little to use continuous gradients compromising yield and purity of the organelle fraction. Therefore, subsequent centrifugation steps with discontinuous gradients are selected to separate organelles either to the top, the bottom, or to an interphase of media exhibiting defined densities (4). As gradient media, most commonly sucrose, percoll, ficoll, and nycodenz are employed (5). The latter three are almost osmotically neutral media for density gradient formation; hypertonic sucrose solutions however can lead to dehydration of organelles like mitochondria and chloroplasts which are enclosed by a sucrose impermeable lipid membrane bilayer. With respect to purity of isolated organelles, methods that rely on the surface charge have been found to work best. Both the chromatographic two-phase partition technology pioneered by Per-Ake Albertsson and the electrophoretic method of freeflow electrophoresis rely on differences in the surface charge of organelles and can both perform separation in a continuous mode
68
Plöscher et al.
(6). In continuous two-phase partitioning, cell homogenate is applied to a mixture of two polymers: polyethylene–glycol and dextran. Partitioning of organelles according to surface properties occurs by repetitive mixing of polymer phases and separation by centrifugation. Major factors influencing separation are polymer molecular weight and concentration, temperature, ionic species, and pH. In free-flow electrophoresis, the homogenate is delivered by a pump to a separation chamber consisting of a buffer film flowing between two electrodes. The organelles are separated by their differing migration speeds in an electric field perpendicular to the flow of the buffer film. The different fractions are collected by pumps at the other end of the separation chamber. Since the handling of the devices was somewhat error prone and the net charges of an organelle can change, e.g. with the current physiological state, this technique has not been applied widely and has been used mostly in combination with other approaches like gradient centrifugation (6, 7). Finally, organelles may be enriched by immunoaffinity capture. The separation is based on recognition of surface epitopes on the outer phase of the organelle. Binding of the antibodies to a molecular support like a chromatographic matrix coupled to magnetic beads or protein A then allows for high specificity detection and high purity enrichment of organelles. Problems occur, since the detection of the epitope has to take place in the native state of the target protein and the antibody would also isolate aggregates of the target organelle with unspecific cell debris (3). These restrictions limit the use of the method to special cases when, after a prefractionation by gradient centrifugation, a preparation of highest purity of an organelle has to be achieved. However, it should not be overlooked that the principle of providing a second dimension of separation based on immunoaffinity or on surface charge in addition to conventional centrifugation purification protocols that rely on size and density has potential for a more precise, deeper, and more thorough investigation of organelle proteomes. In this respect, further information about techniques dedicated to organelle purification is needed as found in Olson et al. (8). Organelle purity is essential for experimental proteome quality. Proteins from other compartments found to co-isolate with an isolated organelle may only contaminate the sample; however, they may also well be true functional components of the isolated organelle and another cell compartment simultaneously, complicating the interpretation of proteome data. Hence, a majority of plant organelle proteomic studies focused on chloroplast and mitochondria, which can be well isolated and the purity characterized (9). For mitochondria, mostly cytochrome c oxidase (COX) is used as an enzymatic marker protein, whereas alkaline pyrophosphatase (PPase) is used to specify plastid purity (10).
Organelle Proteomics
69
Finally, it should be briefly noted that isolation of organelles is not the only means to identify the protein complement of an organelle. Chemical labelling or molecular cloning of a fluorescent protein (FP) onto a protein of interest may provide an excellent means to study localization of a protein of interest to a specific organelle in the cellular background and complement the proteomic investigations at the level of isolated organelles (10– 12). In the case of FPs, subcellular location can be visualized by autofluorescence after gene expression of the tagged protein and in vivo targeting to a specific organelle. Due to high quantum yields, FPs provide sensitive tools to follow biological molecules in real time. However, in order to avoid artefactual localization, fusion proteins have to meet several important criteria (13). First, the fluorescent protein should in general be stable over a wide range of the physiological conditions found in different subcellular compartments (e.g. pH tolerant). The insertion of the FP should minimally influence the conformation of the target protein and preserve native targeting signals and post-translational modification sites. Also, protein expression should occur from its native regulatory sequences to maintain its developmental and/ or tissue-specific regulation.
3. ElectrophoresisBased Proteomics The challenge for any investigation of an organelle proteome is the biochemical complexity, the number, and the dynamic range of proteins. First approaches to cope with this problem used the high capacity of gel-based electrophoretic systems for separation of proteins (see Chapter “Two-Dimensional Electrophoresis (2DE): An Overview”). Therefore, this first-generation approach was called ‘classical proteomics’. Thereafter, the term ‘gel-free’ was coined for all second-generation proteomic approaches that utilized the high capacity of chromatographic methods for peptide selection (14) (see also Chapter “Shotgun Protein Analysis by Liquid Chromatography-Tandem Mass Spectrometry”). To cope with the high number of proteins, two-dimensional electrophoresis (2DE) was developed in the 1970s (15). With this technique, a gold standard for high-resolution separation of up to 2,000 different protein spots per gel and for quantification of proteins down to 1 ng was established (16). Any new gel-based method has to match the standards set by 2DE. Despite its outstanding properties, 2DE has some limitations with respect to the biochemical complexity of organelle proteins. Most critical are under-representation of basic and highly hydrophobic membrane proteins (see Chapters “Solubilization of Proteins in
70
Plöscher et al.
2DE: An Outline”, “Difficult Proteins”, “Protein Extraction for 2DE”, “2DE for Proteome Analysis of Human Metaphase Chromosomes”, “Microsomal Proteomics”, “Pre-Fractionation Using Microscale Solution IEF”). For example, alkaline proteins are the major cause of problems in 2DE of bacterial species (17, 18). Basic proteins are charged positive at neutral pH, and near the cathode spots were found not to focus but to streak over a wider pH range. For precise focusing, wide pH gradients combined with short separation distances and running times were found to be advantageous (14). In addition, application of additional DTT or a replacement of the strip holder after the rehydration step was found to decrease streaking (19, 20). Membrane proteins received highest interest since most drug targets are completely or in part embedded into the lipid bilayer phase of a membrane (21). Due to the low solubilization capacity of non-ionic detergents employed in IEF–PAGE and the tendency of membrane proteins to aggregate around their pI, separation of membrane proteins by classical 2DE did not yield good results (22). A 2DE approach has therefore, in many instances, been abandoned and only the second step comprising separation of proteins according to molecular mass employed. In these approaches, the ionic detergent SDS has been proven to be the most widely applicable detergent for membrane protein solubilization (23–25). However, the resolution of this one-dimensional SDS–PAGE (1DE) approach is limited. Compared to 2DE, in 1DE the number of individual separated spots is about 400-fold less. Therefore, alternative twodimensional separation strategies were developed for the analysis of membrane proteins. Two different two-dimensional separation strategies were followed. First, both dimensions were carried out under denaturing conditions. In this case different ionic detergents like SDS, 16-BAC, or CTAB were used for the first dimension (for review see 26). All these approaches were combined with SDS–PAGE as second dimension. For highly hydrophobic membrane proteins, SDS/SDS–PAGE experiments were shown to yield best results resolving twofold to fivefold more proteins compared to 1DE (27, 28). Second, membrane proteins were separated in their native state in the first dimension, whereas the second dimension was again denaturing. Since many membrane proteins form multi-subunit complexes in their functional state, methods which retain the native state are of special interest. Native complex separation is achieved by solubilization of membrane protein complexes by non-ionic detergents and separation of the solubilized complexes by CN, hrCN-, or BN–PAGE (see Chapter “Blue Native Gel Electrophoresis (BN–PAGE) Proteomics”) (29–31). A special feature of all three methods is direct identification of a protein or protein complexes that are separated in the gel by catalytic activity. This opens a completely new area for proteomic
Organelle Proteomics
71
investigations of organelles, since identification of complex metabolic pathways assembled in protein complexes will be possible. For solubilization of protein complexes non-ionic detergents like digitonin, n-dodecyl-β-D-maltoside, or Triton were most widely used. The separation principle in CN–PAGE relies on the net negative charge of the protein in the pH background of the gel, whereas in BN and hrCN–PAGE negative charges are externally provided to the complexes by Coomassie blue and bile salts, respectively. As a result of the electrophoretic setup, separation of proteins in CN–PAGE is useful for protein complexes with a pI above seven and normally yields a lower resolution than BN or hrCN–PAGE. For analysis of the subunit composition of the complexes, all three methods are combined with SDS–PAGE as a second dimension. Another challenging aspect of gel-based approaches is the limited amount of sample which can be loaded. The dynamic range of gel-based experiments is at 103–104 (16), whereas in cells and organelles, the dynamic range of protein copies per cell covers a range of about 106 (32). Hence, problems encountered in the detection of low abundant proteins were the basis for the development of ‘gel-free’ approaches which, for proteome characterization, was in most cases linked to an MS-based identification of separated peptides.
4. Mass SpectrometryBased Proteomics
Proteomics relies on fast and reliable identification of a large number of different proteins. Here, only MS has been found to enable protein identification at a large scale. In sample preparation for MS analysis, proteolytic digestion of proteins to peptides is the first step (see Chapters “Shotgun Protein Analysis by Liquid Chromatography-Tandem Mass Spectrometry”, “De Novo Sequence Analysis of N-Terminal Sulfonated Peptides After InGel Guanidination”, “Tryptic Digestion of In-Gel Proteins for Mass Spectrometry Analysis”). This step is either achieved before or after separation of proteins by gel-based electrophoresis (33). For peptide identification, currently two different systems are used to ionize and transfer peptides in the gas phase for accurate mass analysis: Matrix-Assisted Laser Desorption/Ionization (MALDI) (34) and ElectroSpray Ionization (ESI) (35). In combination with genome sequencing projects and elaborate bioinformatic data analysis capabilities (see Chapters “Tryptic Digestion of In-Gel Proteins for Mass Spectrometry Analysis” and “Database Interrogation Algorithms for Identification of Proteins in Proteomic Separations”), an efficient system for protein identification by peptide
72
Plöscher et al.
sequencing or peptide mass fingerprint analysis linked to database searching has been established (36). Still, the quality of the MS analysis is compromised by the complexity of the protein complement of a cell or cellular subfraction. To maintain high-quality analysis, protein samples are first separated by 2DE, which results in a low protein complexity of one to three proteins per spot. For these samples, MALDI sources combined with Time-Of-Flight (TOF) MS (MALDI-TOF) are widely used. If the complexity is low, determination of the identity of a protein is software driven, rapid, and simple. By digestion with an endoprotease, peptides are generated from proteins; a number of masses are identified by the mass spectrometer and are written into a list called a peptide mass fingerprint (PMF). The result is compared with a list of theoretical masses generated in silico from a protein database and statistically evaluated (37–39). If there is more than one protein in the sample, results fail to be significant often with increasing complexity of the sample. But also a low number of assigned peptides or missing of the protein in the database leads to no significant matches or false positive identifications. As one possibility to overcome the issue associated with protein complexity in MS, fragment spectra of the single peptide mass signals (MS/MS) may be recorded. The fragmentation of a parent ion (peptide) is the result of collision-induced dissociation (CID). The MS/MS data contain information about the amino acid composition of a peptide and can be evaluated to calculate the de novo sequence (40–42). The advantage of de novo sequencing technology is that the alignment of peptide sequences to signals of fragment ions in the mass spectrum does not require alignment with any database information (40). Still, if the protein corresponding to the peptide sequenced is not found in the database, a homology search algorithm like Fasta (43) may be used to obtain significant and biologically relevant matches. A second step to analyse protein samples with high complexity is to separate the corresponding peptide mixtures by high-performance liquid chromatography (HPLC) systems (see Chapter “Shotgun Protein Analysis by Liquid Chromatography-Tandem Mass Spectrometry”). In these approaches, HPLC systems are equipped with reversed-phase C18 columns which are directly connected to an ESI source MS. Together with a flow rate of down to 50 nL/min, peptide separation and concentration is achieved immediately before the ionization step, resulting in a substantial increase of sensitivity (44, 45). Such ‘online’ approaches are often ‘gel-free’ without any prior gel electrophoretic separation of protein samples. However, although more than one chromatographic step is usually employed, full separation of complex peptide mixtures has not been achieved by one single combination of chromatographic matrices, making this multidimensional protein
Organelle Proteomics
73
identification technology (MudPIT) technologically challenging (46, 47). An additional aspect of functional proteome analysis concerns post-translational modifications (PTMs) which might have a major influence on a protein’s function, activation state, or its cellular localization (see Chapters “Detection of 4-Hydroxy-2-Nonenaland 3-Nitrotyrosine-Modified Proteins Using a Proteomics Approach”, “Proteomic Detection of Oxidized and Reduced Thiol Proteins in Cultured Cells”, “Phosphoproteome Analysis by In-Gel Isoelectric Focusing and Tandem Mass Spectrometry”, “Detection of Protein Glutathionylation”). MS techniques are of prime importance to identify phosphorylation, glycosylation, acetylation, or methylation of proteins (48, 49). Modern MS techniques do not only allow qualitative proteome analysis, but also quantitative techniques have been developed in recent years. For a relative quantification of protein levels, different methods for chemical labelling of proteins and peptides with stable isotopes have been published. Thiol-reactive isotope coded affinity tags (ICAT) may be used. Cysteines in peptides from two different samples are labelled with two different ICAT reagents. One tag contains only hydrogens, the other only deuterium atoms resulting in a distinct mass difference of eight Daltons. The two samples are combined and after MS analysis, relative quantification of the deuterium labelled relative to the hydrogen labelled peptides is possible which is directly proportional to the intensity of the signals of each different labelled peptide (50). Another method is dependent on four isobaric tags which are reactive to amino groups and can be differentiated by MS/MS (iTRAQ). This method allows the simultaneous quantification of up to four samples and, as an additional advantage, the method is not limited to peptides with cysteine residues only (51, 52). In contrast to labelling peptides with isotope coded tags, label-free methods have also been developed for use with LC-MS. These methods depend on reproducible chromatographic separation systems in combination with a high mass resolution and a high mass accuracy. LC-MS/MS is used to generate a chromatogram of the signal intensity of unlabelled peptides (MS) over time and of the signal intensity of peptide fragments (MS/ MS) over the mass-to-charge ratio. With the software DeCyder™ MS (DeCyderMS) the signal intensity can be analysed during an automated detection and a relative quantification of peptides is possible via the 2DE data analysis (53). Another approach combines MS data with MSE data, a method where no precursor ion is isolated separately, but the collision energy is ramped resulting in fragment ions of all peptides simultaneously (54). With this method, not only identification and relative quantification is possible, even an absolute quantification of proteins can be achieved (55). However, how do these methods perform when it comes to
74
Plöscher et al.
identification of an organelle proteome? Let us look at the proteome of two major organelles of the plant and animal cell.
5. The Chloroplast Proteome In a eukaryotic plant cell (see also Chapter “Preparation and Analysis of Plastid Proteomes by 2DE”), chloroplasts are the characteristic organelle where photosynthesis occurs. Chloroplasts can be easily isolated in high amounts from green leaves. As sub-proteomes, the stroma, the thylakoid lumen, the thylakoid, and the envelope membranes can be well differentiated (for reviews see 56, 57) A potential proteome of chloroplasts may be selected in silico from genomic data by using subcellular localization prediction programs such as Predotar (http://urgi.versailles.inra.fr/ predotar/predotar.html) (58), TargetP (http://www.cbs.dtu.dk/ services/TargetP) (59, 60), or WoLF PSORT (http://wolfpsort. seq.cbrc.jp) (61). These in silico approaches define the organelle proteome on the basis of N-terminal presequences that were experimentally indicated to target the nuclear-encoded and cytoplasmically expressed protein to a specific organelle (62, 63). Programs were tested and chloroplast targeting was found closely to match experimental data (64). Also, for the determination of proteins destined for a suborganellar compartment like the membranes or the chloroplast lumen, prediction programs were developed. TMHMM, a program for prediction of transmembrane domains (http://www.cbs.dtu.dk/services/TMHMM-2.0), may be a start to map a membrane proteome, or LumenP, a predictor for the luminal proteome of chloroplast, may help to specify protein identification (65). In an early attempt to identify the chloroplast proteome of Arabidopsis thaliana, Abdallah et al. (66) used ChloroP to predict a number of 1,900–2,500 proteins including proteins encoded by nuclear and chloroplast DNA (http://www. cbs.dtu.dk/services/ChloroP) (66, 67). A more differentiated study was then carried out by Sun et al. (68). Here, experimental datasets published with respect to the localization of proteins within organelle compartments were combined with prediction programs to predict the proteome of chloroplasts of A. thaliana. A number of 4,255 nuclear-encoded non-redundant proteins were found. This number of proteins was differentiated in the soluble proteins of the lumenal proteome (285–291 proteins) and of the stromal proteome (3,387 proteins), the combined integral membrane proteome of thylakoids and inner membranes (520 proteins), and the thylakoid membrane proteins containing a luminal transit peptide (57 proteins) (68).
Organelle Proteomics
75
In order to link theoretical with experimental data, several databases have been built up independently. Results of the study by Sun et al. (68) were integrated in the plastid proteome database (PPDB; http://ppdb.tc.cornell.edu) (69). Another database focused on plastid proteomes that differentiates between the proteome of chloroplasts, etioplasts, and undifferentiated plastids is ‘plprot’ (http://www.plprot.ethz.ch). At the time of publication plprot contains 2,043 protein entries and allows a BLAST search to obtain comparative information on the proteomes of different plastid types (70). Besides plastid-specific databases, a combinatorial database for experimental and predicted datasets to localize subcellular compartmentation of proteins in A. thaliana is available as SUBA (http://www.suba.bcs.uwa.edu.au) (71, 72). This database is a comprehensive accumulation of protein location and further protein characteristics, encompassing the plastid, mitochondrion, peroxisome, nucleus, plasma membrane, endoplasmatic reticulum, vacuole, Golgi, cytoskeleton general structures, and the cytosol. The experimental SUBA datasets are based on MS techniques, visualization of fluorescent proteins attached to selected proteins, and on the analysis of entries in the SwissProt database, and are combined with the use of different prediction tools (71). Although a prediction of the theoretical size of the plastid proteome and sub-proteome provides a value independent of biochemical constrains, this value has to be validated by proteomic results providing evidence that predicted proteins are identified at predicted locations. In one attempt to localize the thylakoid membrane proteome, thylakoids were isolated from barley chloroplasts, proteins were separated by a 2-D Blue Native/SDS– PAGE system, and peptides were identified by nano-ESI-MS/ MS identification. Here, 56 membrane proteins were explicitly detected by de novo sequence analysis (73). Although the approach resulted in a precise determination of the identity of the protein’s localization and of its belonging to a distinct protein complex, it was obvious that especially small hydrophobic membrane proteins were not detected selectively. In another explicit study of thylakoids from A. thaliana, 2DE was combined with organic solvent extraction of hydrophobic membrane proteins, and MALDI-TOF MS analysis was combined with nanoLC-ESI MS/MS for protein identification (69). In this study, 154 proteins were identified, including 83 out of 100 thylakoid membrane proteins known to belong to the photosynthetic apparatus. In another detailed study, 32 proteins of plastid localized plastoglobuli were detected (74). More specialized proteomic analyses of chloroplasts also included post-translational modification of proteins. Phosphorylation was addressed by the group of Aro et al. (75), whereas glycosylation (76) was addressed in a detailed study of proteomic changes during the interaction of the
76
Plöscher et al.
pathogen Fusarium graminearum and its host Triticum aestivum including quantitative changes of the proteome and the role of a glycosylation during infection (76). To further analysis especially of membrane proteomes, a review about the isolation of different plant membranes has been recently published (77).
6. The Mitochondrial Proteome In eukaryote cells, the mitochondrion is the key organelle for cellular respiration. Mitochondrial core functions are oxidative phosphorylation, amino acid metabolism, fatty acid oxidation, and iron-sulphur cluster assembly. Hence, mitochondrial proteins have been highly conserved in all types of organisms during evolution. A number of systematic studies in model organisms contributed to identifying mitochondrial proteins in general and facilitated the identification of human mitochondrial disease genes (78, 79; see also Chapter “Blue Native Gel Electrophoresis (BN–PAGE) Proteomics”). The genome size of mitochondria ranges from 16 kb in humans to more than 500 kb in particular plant species (80) encoding three (Plasmodium falciparum) to 85 proteins (A. thaliana) (81). By comparison, the mtDNA of yeast (Saccharomyces cerevisiae) contains about 86 kb encoding 30 proteins (82). Most of the genes encode key components of vital mitochondrial functions like aerobic respiration and translation (e.g. ATP synthase subunits, cytochrome oxidases, and ribosomal protein genes). If organisms are compared, the number of predicted mitochondrial proteins increases with the size of the genome indicating a relation between genome and proteome complexity of the organelle(s) (81). Furthermore, the number of mitochondriarelated proteins varies subject to cell type and organism. In contrast to the small number of 30 proteins encoded in the yeast mitochondrial genome, at least 800–1,000 proteins with assigned functions occur in the yeast mitochondrion (81, 83, 84). Also in A. thaliana the occurrence of 2,000–3,000 mitochondrial proteins is predicted (60, 81), and about 1,000–4,000 mitochondrial proteins are predicted for the differentiated mammalian cells of humans (81, 85). Hence, there is no doubt that the majority of mitochondrial proteins are encoded in the nucleus and are subsequently imported into the mitochondrion. However, the specific identity of nuclear genes predicted to encode proteins targeted to mitochondria is not completely defined. According to the high number of proteins imported into mitochondria, the ability of programs (e.g. PSORT, TargetP, and
Organelle Proteomics
77
MitoProt) to predict the number of mitochondrial proteins from various eukaryotes has been investigated in detail. However, in contrast to plant chloroplasts, these methods failed to attain sufficient specificity (low number of false positives, but loss of true positives) and sensitivity (high number of false positives) to produce confident estimates. In addition, numerous mitochondrial membrane proteins are known not to contain the characteristic N-terminal sequences found in most matrix proteins. In summary, the individual programs correctly predicted about 60% of the experimentally identified mitochondrial proteins (81, 86), and the algorithms detect solely those proteins that utilize an amino-terminal presequence. Hence, proteins with internal targeting signals were not considered. However, the rising number of experimentally identified proteins which are assigned to mitochondria may now be used as training sets to improve the targeting prediction algorithms. To verify the mitochondrial proteome experimentally, numerous proteome analyses of mitochondria from different species and tissues have been published in the last decade. The application of 1DE in combination with LC-MS/MS analysis of the digested proteins led to identification of ~179 different gene products, which corresponded to about 20% of the yeast mitochondrial proteins (87). Sickmann et al. used a couple of strategies for protein separation and MS analysis (88). After protein separation by 1DE or 2DE, proteins were digested with four different proteases. Resulting peptides were analysed by multidimensional liquid chromatography (MDLC) coupled to ESI-MS or nano-LC-MS/ MS. Utilization of multiple approaches assured wide coverage of proteins with different properties. The combined data resulted in 750 proteins which could be related to yeast mitochondria. Based on the predicted mitochondrial proteins in the literature, it was calculated that the identified proteins represent ~ 90% of all mitochondrial proteins. Another study used a ‘shotgun strategy’ to identify 546 proteins from membranes and matrix fractions of mitochondria (64). Proteins identified in both studies were compared to the MitoP2 database (http://ihg.gsf.de/mitop), which contained 477 entries. However, only 238 proteins were present in all three datasets. This dataset therefore represented about a quarter of the total protein number predicted for mitochondria. The set of newly identified mitochondrial proteins in the proteomic approaches comprised about 100 proteins which were confirmed by GFP localization (89). The mitochondrial proteome of the model plant A. thaliana was also subjected to multiple analyses (for a review see 90). Initial investigations of the protein composition by traditional 2DE separations of a cell culture from A. thaliana revealed ~350 and 800 mitochondrial protein spots (91, 92). Millar et al. analysed about
78
Plöscher et al.
half of the detected spots after tryptic digestion using MALDI-MS (91), and the data analysis resulted in ~90 distinct protein matches from mitochondria. In a second study, only 51 distinct proteins (about 12%) of the 800 resolved spots were successfully identified by immunoblotting, Edman degradation, and MS/MS-analysis of internal peptides (92). However, the spot number detectable in a 2DE gel is not always significant in estimating the number of total proteins of the corresponding proteome. First, the spot number does not correlate to the number of non-redundant proteins, because a single protein often results in multiple protein spots. Hence, the non-redundant set of proteins in these gels can be considerably lower. On the other hand, limitations in resolution of the gel-based approach can lead to overlapped spots and low-abundance proteins are invisible due to the presence of highabundance proteins that limit the loading capacity for separation in 2DE. In addition, the method is not suitable for isolation of highly hydrophobic or alkaline proteins. The largest dataset of mitochondrial proteins from A. thaliana was again obtained by the ‘shotgun’ technique. Direct analysis of peptides, derived from trypsin digestion of complex polypeptide mixtures, by LC-MS/ MS led to the identification of 416 non-redundant proteins (10). About 60% of these were newly identified by this LC-MS/MS approach. The gel-free strategy was superior especially for identification of large, small, and basic proteins. The dataset also contained a significant number of low-abundance proteins involved in DNA synthesis, transcriptional regulation, protein complex assembly, and cell signalling. For human mitochondria, about half of the estimated proteins localized or functionally associated with mitochondria are now known. The proteome of human-heart mitochondria was investigated with two different methods (93, 94). In a gel-based approach, mitochondrial protein complexes were pre-fractionated by density gradient centrifugation followed by 1DE for separation of the complex subunits. After in-gel digestion, resulting peptides were analysed by LC-MS/MS. In a second approach, proteins from mitochondria preparations were solubilized by 8 M urea and subsequently divided into ‘soluble’ and ‘insoluble’ fractions. Both fractions were chemically and enzymatically digested, and data obtained from multidimensional LC-MS/MS experiments were used for protein identification. Using these techniques a total of 680 mitochondrial or mitochondria-associated proteins were identified. Still, despite the strong progress in completion of the mitochondrial proteome, it appears that none of the approaches provided a complete answer to the questions: Which of the proteins belong to the mitochondrial proteome? How similar are the mitochondrial proteomes from different organisms and tissues?
Organelle Proteomics
79
References 1. Yu, C., and Cohen, L. (2004) Tissue sample preparation – not the same old grind. LC-GC Europe 17, 96–103 2. Pappas, D., and Wang, K. (2007) Cellular separations: a review of new challenges in analytical chemistry. Anal Chim Acta 601, 26–35 3. Pasquali, C., Fialka, I., and Huber, L. A. (1999) Subcellular fractionation, electromigration analysis and mapping of organelles. J Chromatogr B Biomed Sci Appl 722, 89–102 4. James Morre, D., and Andersson, B. (1999) Isolation of all major organelles and membranous cell components from a single homogenate of green leaves. Methods Enzymol 228, 412–419 5. Murayama, K., Fujimura, T., Morita, M., and Shindo, N. (2001) One-step subcellular fractionation of rat liver tissue using a Nycodenz density gradient prepared by freezing-thawing and two-dimensional sodium dodecyl sulfate electrophoresis profiles of the main fraction of organelles. Electrophoresis 22, 2872–2880 6. Krivankova, L., and Bocek, P. (1998) Continuous free-flow electrophoresis. Electrophoresis 19, 1064–1074 7. Zischka, H., Weber, G., Weber, P. J., Posch, A., Braun, R. J., Buhringer, D., et-al. (2003) Improved proteome analysis of Saccharomyces cerevisiae mitochondria by free-flow electrophoresis. Proteomics 3, 906–916 8. Olson, K. J., Ahmadzadeh, H., and Arriaga, E. A. (2005) Within the cell: analytical techniques for subcellular analysis. Anal Bioanal Chem 382, 906–917 9. Dunkley, T. P., Dupree, P., Watson, R. B., and Lilley, K. S. (2004) The use of isotope-coded affinity tags (ICAT) to study organelle proteomes in Arabidopsis thaliana. Biochem Soc Trans 32, 520–523 10. Heazlewood, J. L., Tonti-Filippini, J. S., Gout, A. M., Day, D. A., Whelan, J., and Millar, A. H. (2004) Experimental analysis of the Arabidopsis mitochondrial proteome highlights signaling and regulatory components, provides assessment of targeting prediction programs, and indicates plant-specific mitochondrial proteins. Plant Cell 16, 241–256 11. Chalfie, M., Tu, Y., Euskirchen, G., Ward, W. W., and Prasher, D. C. (1994) Green fluorescent protein as a marker for gene expression. Science 263, 802–805 12. Hanson, M. R., and Kohler, R. H. (2001) GFP imaging: methodology and application to investigate cellular compartmentation in plants. J Exp Bot 52, 529–539
13. Tian, G. W., Mohanty, A., Chary, S. N., Li, S., Paap, B., Drakakaki, G., et-al. (2004) Highthroughput fluorescent tagging of full-length Arabidopsis gene products in planta. Plant Physiol 135, 25–38 14. Canas, B., Pineiro, C., Calvo, E., Lopez-Ferrer, D., and Gallardo, J. M. (2007) Trends in sample preparation for classical and second generation proteomics. J Chromatogr A 1153, 235–258 15. O’Farrell, P. H., (1975) High resolution twodimensional electrophoresis of proteins. J Biol Chem 250, 4007–4021 16. Lopez, J. L. (2007) Two-dimensional electrophoresis in proteome expression analysis. J Chromatogr B Analyt Technol Biomed Life Sci 849, 190–202 17. Drews, O., Reil, G., Parlar, H., and Gorg, A. (2004) Setting up standards and a reference map for the alkaline proteome of the Grampositive bacterium Lactococcus lactis. Proteomics 4, 1293–1304 18. Cordwell, S. J., Basseal, D. J., Bjellqvist, B., Shaw, D. C., and Humphery-Smith, I. (1997) Characterisation of basic proteins from Spiroplasma melliferum using novel immobilised pH gradients. Electrophoresis 18, 1393–1398 19. Görg, A., Boguth, G., Obermaier, C., Posch, A., and Weiss, W. (1995) Two-dimensional polyacrylamide gel electrophoresis with immobilized pH gradients in the first dimension (IPGDalt): the state of the art and the controversy of vertical versus horizontal systems. Electrophoresis 16, 1079–1086 20. López, J.-L., and Humphery-Smith, I. (2003) New improvements for high-resolution and high-protein load using basic two-dimensional electrophoresis gels. Genome Letts 2, 93–97 21. Hopkins, A. L., and Groom, C. R. (2002) The druggable genome. Nat Rev Drug Discov 1, 727–730 22. Kashino, Y. (2003) Separation methods in the analysis of protein membrane complexes. J Chromatogr B Analyt Technol Biomed Life Sci 797, 191–216 23. Simpson, R. J., Connolly, L. M., Eddes, J. S., Pereira, J. J., Moritz, R. L., and Reid, G. E. (2000) Proteomic analysis of the human colon carcinoma cell line (LIM 1215): development of a membrane protein database. Electrophoresis 21, 1707–1732 24. Pisareva, T., Shumskaya, M., Maddalo, G., Ilag, L., and Norling, B. (2007) Proteomics
80
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
Plöscher et al. of Synechocystis sp. PCC 6803. Identification of novel integral plasma membrane proteins. FEBS J 274, 791–804 Weiner, J. H., and Li, L. (2007) Proteome of the Escherichia coli envelope and technological challenges in membrane proteome analysis. Biochim Biophys Acta 1778, 1698–1713 Braun, R. J., Kinkl, N., Beer, M., and Ueffing, M. (2007) Two-dimensional electrophoresis of membrane proteins. Anal Bioanal Chem 389, 1033–1045 Rais, I., Karas, M., and Schagger, H. (2004) Two-dimensional electrophoresis for the isolation of integral membrane proteins and mass spectrometric identification. Proteomics 4, 2567–2571 Williams, T. I., Combs, J. C., Thakur, A. P., Strobel, H. J., and Lynn, B. C. (2006) A novel Bicine running buffer system for doubled sodium dodecyl sulfate – polyacrylamide gel electrophoresis of membrane proteins. Electrophoresis, 27, 2984–2995 Krause, F. (2006) Detection and analysis of protein-protein interactions in organellar and prokaryotic proteomes by native gel electrophoresis: (membrane) protein complexes and supercomplexes. Electrophoresis 27, 2759–2781 Wittig, I. and Schagger, H. (2005) Advantages and limitations of clear-native PAGE. Proteomics 5, 4338–4346 Wittig, I., Karas, M., and Schagger, H. (2007) High resolution clear native electrophoresis for in-gel functional assays and fluorescence studies of membrane protein complexes. Mol Cell Proteomics 6, 1215–1225 Corthals,G.L.,Wasinger,V.C.,Hochstrasser,D.F., and Sanchez, J. C. (2000) The dynamic range of protein expression: a challenge for proteomic research. Electrophoresis 21, 1104–1115 Granvogl, B., Ploscher, M., and Eichacker, L. A. (2007) Sample preparation by in-gel digestion for mass spectrometry-based proteomics. Anal Bioanal Chem 389, 991–1002 Karas, M., and Hillenkamp, F. (1988) Laser desorption ionization of proteins with molecular masses exceeding 10,000 daltons. Anal Chem 60, 2299–2301 Fenn, J. B., Mann, M., Meng, C. K., Wong, S. F., and Whitehouse, C. M. (1989) Electrospray ionization for mass spectrometry of large biomolecules. Science 246, 64–71 Shevchenko,A.,Jensen,O.N.,Podtelejnikov,A.V., Sagliocco, F. Wilm, M., and Vorm, O. (1996) Linking genome and proteome by mass spectrometry: large-scale identification of yeast
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
proteins from two dimensional gels. Proc Natl Acad Sci U S A 93, 14440–14445 James, P., Quadroni, M., Carafoli, E., and Gonnet, G. (1993) Protein identification by mass profile fingerprinting. Biochem Biophys Res Commun 195, 58–64 Henzel, W. J., Billeci, T. M., Stults, J. T., Wong, S. C., Grimley, C., and Watanabe, C. (1993) Identifying proteins from two-dimensional gels by molecular mass searching of peptide fragments in protein sequence databases. Proc Natl Acad Sci U S A 90, 5011–5015 Yates, J. R., Speicher, S., Griffin, P. R., and Hunkapiller, T. (1993) Peptide mass maps: a highly informative approach to protein identification. Anal Biochem 214, 397–408 Mann, M., and Wilm, M. (1994) Errortolerant identification of peptides in sequence databases by peptide sequence tags. Anal Chem 66, 4390–4399 Yates, J. R., Eng, J. K., McCormack, A. L., and Schieltz, D. (1995) Method to correlate tandem mass spectra of modified peptides to amino acid sequences in the protein database. Anal Chem 67, 1426–1436 Rowley, A., Choudhary, J. S., Marzioch, M., Ward, M. A., Weir, M., Solari, R. C., and Blackstock, W. P. (2000) Applications of protein mass spectrometry in cell biology. Methods 20, 383–397 Pearson, W. R. (1990) Rapid and sensitive sequence comparison with FASTP and FASTA. Methods Enzymol 183, 63–98 Gatlin, C. L., Kleemann, G. R., Hays, L. G., Link, A. J., and Yates, J. R. (1998) Protein identification at the low femtomole level from silver-stained gels using a new fruitless electrospray interface for liquid chromatographymicrospray and nanospray mass spectrometry. Anal Biochem 263, 93–101 Frohlich, T., and Arnold, G. J. (2006) Proteome research based on modern liquid chromatography – tandem mass spectrometry: separation, identification and quantification. J Neural Transm 113, 973–994 McDonald, W. H., and Yates, J. R. (2002) Shotgun proteomics and biomarker discovery. Dis Markers 18, 99–105 Washburn, M. P. (2004) Utilisation of proteomics datasets generated via multidimensional protein identification technology (MudPIT). Brief Funct Genomic Proteomic 3, 280–286 Wilkins, M. R., Gasteiger, E., Gooley, A. A., Herbert, B. R., Molloy, M. P., Binz, P. A., et al. (1999) High-throughput mass spectrometric discovery of protein post-translational modifications. J Mol Biol 289, 645–657
Organelle Proteomics 49. Mann, M., and Jensen, O. N. (2003) Proteomic analysis of post-translational modifications. Nat Biotechnol 21, 255–261 50. Gygi, S. P., Rist, B., Gerber, S. A., Turecek, F., Gelb, M. H., and Aebersold, R. (1999) Quantitative analysis of complex protein mixtures using isotope-coded affinity tags. Nat Biotechnol 17, 994–999 51. Ross, P. L., Huang, Y. N., Marchese, J. N., Williamson, B., Parker, K., Hattan, S., et al. (2004) Multiplexed protein quantitation in Saccharomyces cerevisiae using amine-reactive isobaric tagging reagents. Mol Cell Proteomics 3, 1154–1169 52. Aggarwal, K., Choe, L. H., and Lee, K. H. (2006) Shotgun proteomics using the iTRAQ isobaric tags. Brief Funct Genomic Proteomic 5, 112–120 53. Johansson, C., Samskog, J., Sundstrom, L., Wadensten, H., Bjorkesten, L., and Flensburg, J. (2006) Differential expression analysis of Escherichia coli proteins using a novel software for relative quantitation of LC-MS/MS data. Proteomics 6, 4475–4485 54. Silva, J. C., Denny, R., Dorschel, C. A., Gorenstein, M., Kass, I. J., Li, G. Z., et al..(2005) Quantitative proteomic analysis by accurate mass retention time pairs. Anal Chem 77, 2187–2200 55. Silva, J. C., Gorenstein, M. V., Li, G. Z., Vissers, J. P., and Geromanos, S. J. (2006) Absolute quantification of proteins by LCMSE: a virtue of parallel MS acquisition. Mol Cell Proteomics 5, 144–156 56. Schroder, W. P., and Kieselbach, T. (2003) Update on chloroplast proteomics. Photosynth Res 78, 181–193 57. Ephritikhine, G., Ferro, M., and Rolland, N. (2004) Plant membrane proteomics. Plant Physiol Biochem 42, 943–962 58. Small, I., Peeters, N., Legeai, F., and Lurin, C. (2004) Predotar: a tool for rapidly screening proteomes for N-terminal targeting sequences. Proteomics 4, 1581–1590 59. Nielsen, H., Engelbrecht, J., Brunak, S., and von Heijne, G. (1997) Identification of prokaryotic and eukaryotic signal peptides and prediction of their cleavage sites. Protein Eng 10, 1–6 60. Emanuelsson, O., Nielsen, H., Brunak, S., and von Heijne, G. (2000) Predicting subcellular localization of proteins based on their N-terminal amino acid sequence. J Mol Biol 300, 1005–1016 61. Horton, P., Park, K. J., Obayashi, T., Fujita, N., Harada, H., Adams-Collier, C. J., and Nakai, K. (2007) WoLF PSORT: protein localization predictor. Nucleic Acids Res 35, W585–W587
81
62. Pfanner, N. (2000) Protein sorting: recognizing mitochondrial presequences. Curr Biol 10, R412–R415 63. Bruce, B. D. (2001) The paradox of plastid transit peptides: conservation of function despite divergence in primary structure. Biochim Biophys Acta 1541, 2–21 64. Gomez, S. M., Bil’, K. Y., Aguilera, R., Nishio, J. N., Faull, K. F., and Whitelegge, J. P. (2003) Transit peptide cleavage sites of integral thylakoid membrane proteins. Mol. Cell Proteomics 2, 1068–1085 65. Westerlund, I., Von Heijne, G., and Emanuelsson, O. (2003) Lumen P - a neural network predictor for protein localization in the thylakoid lumen. Protein Sci 12, 2360–2366 66. Abdallah, F., Salamini, F., and Leister, D. (2000) A prediction of the size and evolutionary origin of the proteome of chloroplasts of Arabidopsis. Trends Plant Sci 5, 141–142 67. Emanuelsson, O., Nielsen, H., and von Heijne, G. (1999) ChloroP, a neural networkbased method for predicting chloroplast transit peptides and their cleavage sites. Protein Sci 8, 978–984 68. Sun, Q., Emanuelsson, O., and van Wijk, K. J. (2004) Analysis of curated and predicted plastid subproteomes of Arabidopsis. Subcellular compartmentalization leads to distinctive proteome properties. Plant Physiol 135, 723–734 69. Friso, G., Giacomelli, L., Ytterberg, A. J., Peltier, J. B., Rudella, A., Sun, Q., and Wijk, K. J. (2004) In-depth analysis of the thylakoid membrane proteome of Arabidopsis thaliana chloroplasts: new proteins, new functions, and a plastid proteome database. Plant Cell 16, 478–499 70. Kleffmann, T., Hirsch-Hoffmann, M., Gruissem, W., and Baginsky, S. (2006) plprot: a comprehensive proteome database for different plastid types. Plant Cell Physiol 47, 432–436 71. Heazlewood, J. L., Tonti-Filippini, J., Verboom, R. E., and Millar, A. H. (2005) Combining experimental and predicted datasets for determination of the subcellular location of proteins in Arabidopsis. Plant Physiol 139, 598–609 72. Heazlewood, J. L., Verboom, R. E., TontiFilippini, J., Small, I., and Millar, A. H. (2007) SUBA: the Arabidopsis subcellular database. Nucleic Acids Res 35, D213–D218 73. Granvogl, B., Reisinger, V., and Eichacker, L. A. (2006) Mapping the proteome of thylakoid membranes by de novo sequencing of intermembrane peptide domains. Proteomics 6, 3681–3695
82
Plöscher et al.
74. Ytterberg, A. J., Peltier, J. B., and van Wijk, K. J. (2006) Protein profiling of plastoglobules in chloroplasts and chromoplasts. A surprising site for differential accumulation of metabolic enzymes. Plant Physiol 140, 984–997 75. Aro, E. M., Rokka, A., and Vener, A. V. (2004) Determination of phosphoproteins in higher plant thylakoids. Methods Mol Biol 274, 271–285 76. Zhou, W., Eudes, F., and Laroche, A. (2006) Identification of differentially regulated proteins in response to a compatible interaction between the pathogen Fusarium graminearum and its host, Triticum aestivum. Proteomics 6, 4599–4609 77. Rolland, N., Ferro, M., Ephritikhine, G., Marmagne, A., Ramus, C., Brugiere, S., et al. (2006) A versatile method for deciphering plant membrane proteomes. J Exp Bot 57, 1579–1589 78. Steinmetz, L. M., Scharfe, C., Deutschbauer, A. M., Mokranjac, D., Herman, Z. S., Jones, T., et al..(2002) A systematic screen for human disease genes in yeast. Nat Genet 31, 400–404 79. Prokisch, H., Scharfe, C., Camp, D. G., Xiao, W., David, L., Andreoli, C., et al. (2004) Integrative analysis of the mitochondrial proteome in yeast. PLoS Biol 2, e160 80. Gray, M. W., Burger, G., and Lang, B. F. (1999) Mitochondrial evolution. Science 283, 1476–1481 81. Richly, E., Chinnery, P. F., and Leister, D. (2003) Evolutionary diversification of mitochondrial proteomes: implications for human disease. Trends Genet 19, 356–362 82. Foury, F., Roganti, T., Lecrenier, N., and Purnelle, B. (1998) The complete sequence of the mitochondrial genome of Saccharomyces cerevisiae. FEBS Lett 440, 325–331 83. Scharfe, C., Zaccaria, P., Hoertnagel, K., Jaksch, M., Klopstock, T., Dembowski, M., et-al. (2000) MITOP, the mitochondrial proteome database: update. Nucleic Acids Res 28, 155–158 84. Kumar, A., Agarwal, S., Heyman, J. A., Matson, S., Heidtman, M., Piccirillo, S., et al.
85.
86.
87.
88.
89.
90.
91.
92.
93.
94.
(2002) Subcellular localization of the yeast proteome. Genes Dev 16, 707–719 Taylor, S. W., Fahy, E., and Ghosh, S. S. (2003) Global organellar proteomics. Trends Biotechnol 21, 82–88 Nakai, K., and Horton, P. (1999) PSORT: a program for detecting sorting signals in proteins and predicting their subcellular localization. Trends Biochem Sci 24, 34–36 Pflieger, D., Le Caer, J. P., Lemaire, C., Bernard, B. A., Dujardin, G., and Rossier, J. (2002) Systematic identification of mitochondrial proteins by LC-MS/MS. Anal Chem 74, 2400–2406 Sickmann, A., Reinders, J., Wagner, Y., Joppich, C., Zahedi, R., Meyer, H. E., et-al. (2003) The proteome of Saccharomyces cerevisiae mitochondria. Proc Natl Acad Sci U S A 100, 13207–13212 Huh, W. K., Falvo, J. V., Gerke, L. C., Carroll, A. S., Howson, R. W., Weissman, J. S., and O’Shea, E. K. (2003) Global analysis of protein localization in budding yeast. Nature 425, 686–691 Millar, A. H., Heazlewood, J. L., Kristensen, B. K., Braun, H. P., and Moller, I. M. (2005) The plant mitochondrial proteome. Trends Plant Sci 10, 36–43 Millar, A. H., Sweetlove, L. J., Giege, P., and Leaver, C. J. (2001) Analysis of the Arabidopsis mitochondrial proteome. Plant Physiol 127, 1711–1727 Kruft, V., Eubel, H., Jansch, L., Werhahn, W., and Braun, H. P. (2001) Proteomic approach to identify novel mitochondrial proteins in Arabidopsis. Plant Physiol 127, 1694–1710 Taylor, S. W., Fahy, E., Zhang, B., Glenn, G. M., Warnock, D. E., Wiley, S., et al. (2003) Characterization of the human heart mitochondrial proteome. Nat Biotechnol 21, 281–286 Gaucher, S. P., Taylor, S. W., Fahy, E., Zhang, B., Warnock, D. E., Ghosh, S. S., and Gibson, B. W. (2004) Expanded coverage of the human heart mitochondrial proteome using multidimensional liquid chromatography coupled with tandem mass spectrometry. J Proteome Res 3, 495–505
Chapter 6 Applications of Chemical Tagging Approaches in Combination with 2DE and Mass Spectrometry Alexander Leitner and Wolfgang Lindner Summary Chemical modification reactions play an important role in various protocols for mass-spectrometry-based proteome analysis; this applies to both gel-based and gel-free proteomics workflows. In combination with two-dimensional gel electrophoresis (2DE), the addition of “tags” by means of chemical reactions serves several purposes. Potential benefits include increased sensitivity or sequence coverage for peptide mass fingerprinting and improved peptide fragmentation for de novo sequencing studies. Tagging strategies can also be used to obtain complementary quantitative information in addition to densitometry, and they may be employed for the study of post-translational modifications. In combination with the unique advantages of 2DE as a separation technique, such approaches provide a powerful toolbox for proteomic research. In this review, relevant examples from recent literature will be given to illustrate the capabilities of chemical tagging approaches, and methodological requirements will be discussed. Key words: Mass spectrometry, Chemical tagging, Protein identification, Peptide sequencing, Quantification, Post-translational modification, Review.
1. Introduction The combination of two-dimensional gel electrophoresis (2DE) with mass spectrometric (MS) analysis of digested protein spots has fueled the advance of proteomics research in the early 1990s (1–5). Novel ionization techniques for MS, matrix-assisted laser desorption/ionization (MALDI) and electrospray ionization (ESI), made it possible for the first time to identify a considerable number of the hundreds or thousands of spots that can be separated and visualized on a 2DE gel. The routine application of
David Sheehan and Raymond Tyther (eds.), Methods in Molecular Biology, Two-Dimensional Electrophoresis Protocols, vol. 519 © Humana Press, a part of Springer Science + Business Media, LLC 2009 DOI: 10.1007/978-1-59745-281-6_6
83
84
Leitner and Lindner
MS quickly replaced the classic sequencing technique of Edman degradation. Today, both traditional gel-based techniques and the more recently developed “gel-free” (or “shotgun”) approaches are used in countless proteomic studies. Both strategies have pros and cons that appear in the introduction of most technology papers in the field, and both strategies have their advocates and opponents. However, it is frequently shown that both approaches have unique advantages and yield complementary results. In particular, the possible differentiation of isoforms or members of protein families with considerable sequence homology continues to be the strength of 2DE in combination with MS (6). It is clear that when one deals with highly complex protein mixtures of biological origin such as biofluids or cell lysates, efficient multidimensional separation strategies are required, and 2DE continues to offer the most comprehensive separation on the protein level. Its hyphenation to MS for protein identification and sequencing, although well established nowadays, still leaves room for improvement. In recent years, automated spot picking and sample processing (involving handling steps like destaining, in-gel digestion, desalting, and spotting onto MALDI targets) has been made available (7, 8). Other advances include the use of novel digestion technologies involving microwave irradiation (9) or ultrasound (10), potentially enabling higher sample throughput after in-gel separation. At the MS detection stage, many 2DE workflows still rely on peptide mass fingerprinting for protein identification, i.e., the identification of proteins based on the masses of the peptides resulting from (typically) enzymatic cleavage of the proteins. For this purpose, matrix-assisted laser desorption-time-of-flight (MALDI-TOF) MS combines ease of use and relatively low initial investment and running costs. However, peptide sequencing on these instruments can only be performed by post-source decay (PSD), the dissociation of metastable ions in the field-free region of the TOF analyzer. The quality of PSD spectra is usually not comparable to those obtained from the dissociation of peptide ions with gas molecules in the collision-induced dissociation (CID) process. Only more recently have “true” tandem MS-based analyses become more frequently applied in combination with 2DE-based proteomics. One possible approach is the use of TOF/TOF instrumentation that involves high-energy CID of peptide ions, yielding potentially information-rich, but also quite complex fragmentation patterns (11). In contrast to TOF/TOF MS which is usually carried out without any further separation of the protein digests, proteolytically generated peptides may also be separated by liquid chromatography prior to electrospray ionization-MS/ MS analysis. For this strategy, a number of different types of mass
Methods in Molecular Biology: Two-Dimensional Electrophoresis Protocols
85
spectrometers are employed such as ion traps, quadrupole-TOF instruments, etc. Figure 1 summarizes the most frequently used MS approaches. In all cases, the data that are generated need to be analyzed by bioinformatic methods (12, 13; see also Chapter “Database Interrogation Algorithms for Identification of Proteins in Proteomic Separations”). Usually, the list of peptide masses or the set of tandem MS spectra is compared to theoretical data obtained from protein sequence databases. Alternatively, and only if MS/ MS data of sufficiently high quality are available, results may be directly interpreted for de novo sequencing without referring to protein databases. Regardless of the exact MS approach being followed, it is always desirable to obtain optimum results. To achieve this, chemical reactions can be performed on peptides and proteins
Fig. 1. Common approaches for combining protein separation by 2DE and identification by MS.
86
Leitner and Lindner
at various stages of the analysis protocol to enhance the information content that can be derived from a sample. In the following sections, we give an overview of potential benefits of such “tagging” reactions in combination with 2DE and highlight relevant examples from the recent literature. A “tag” in this case refers to any functional group that is introduced (either on the protein or peptide level) to improve analysis by MS, be it by enhancing the ionization or fragmentation of the peptides, for quantification purposes or any other means. As we and others have summarized previously (14–18), the use of chemical tagging reactions is not restricted to either gel-free or gel-based platforms. In the context of this book, emphasis will however be given to those approaches that have been successfully applied together with 2DE as the separation platform.
2. Concepts and Potential Benefits of Chemical Tagging Approaches
Broadly, applications of chemical tagging reactions in combination with 2DE can be separated into three groups based on the type of information that is obtained (see Fig. 2): Protein identification, quantification, and characterization, the latter referring to the profiling of protein modifications in particular. For protein identification by MALDI peptide mass fingerprinting, it is of utmost importance to identify as many peptides from an in-gel digest as possible in order to achieve significant scores from database searches. However, it is well known that many peptides are not observed at all in a MALDI mass spectrum (19–21). This may be due to several reasons. For example, smaller peptides tend to be obscured by matrix-derived signals in the low mass range; however, they commonly do not provide significant information for database searches as short sequence stretches are rarely unique for a given protein sequence. Highly polar peptides may already be lost during sample clean-up, if a desalting step is performed. Very large peptides are recovered less efficiently from the gel matrix. Most importantly, sequence-specific effects result in vastly different signal abundances for peptides in the medium mass range (~800–3,000 Da) that usually provides most signals for protein identification. Frequently, there is a bias against tryptic peptides terminating with Lys compared to those with Arg as the C-terminal residue (19–21). Thus, several tagging strategies have been introduced that enhance the signal of Lys-peptides (see Subheading 4.1 for details). Other groups have observed an increase in the number of peptide-derived signals – and thus improved sequence coverage – upon introduction of additional hydrophobic groups by
Methods in Molecular Biology: Two-Dimensional Electrophoresis Protocols
87
Fig. 2. Overview of potential application fields for chemical tagging reactions.
different tagging chemistries, which is believed to result in a more efficient desorption/ionization process as well. Tags that can be introduced on various amino acid residues with high specificity may also be used to obtain amino acid composition information that improves database searching. Protein identification by post-source decay or CID MS/MS techniques relies on easily interpretable tandem mass spectra with complete fragment ion series. Sufficient data quality is important for automated database searches, but even more for de novo sequencing approaches where sequence stretches have to be read directly without being able to rely on existing database entries. Different modification chemistries (discussed in detail in Subheading 4.2) that enhance peptide fragmentation have been introduced since the 1990s and have demonstrated their usefulness in combination with 2DE. Gel-based quantification is a well-established technology that has developed from Coomassie stains to highly sensitive
88
Leitner and Lindner
fluorescent dyes and the DIGE technology (see Chapter “TwoDimensional Difference Gel Electrophoresis”) (22). However, some limitations still exist for densitometric approaches, particularly when proteins comigrate to the same position on the gel or changes in expression exceed the dynamic range of the staining procedure (see Chapter “Troubleshooting Image Analysis in 2DE”). Therefore, differential stable isotope labeling (SIL) techniques – which are frequently used for gel-free approaches (see 14–18 and references therein) – may also be used in combination with 2DE. The content of Subheading 4.3 highlights developments in this area and cites studies that have compared traditional gel-based quantification and SIL methodologies. Enrichment and depletion techniques are frequently used to enrich proteins of interest prior to 2DE. In general, these strategies allow profiling of less abundant proteins and therefore increase the dynamic range of 2DE-based analyses. For example, abundant plasma proteins such as albumin and immunoglobulins may be depleted by immunoaffinity chromatography (23, 24) and allow a higher loading of the remaining proteins. Affinity tags can also be introduced via chemical reactions so that a subpopulation of proteins is enriched from a biological sample (see Subheading 4.4). This is of particular importance for proteins carrying (substoichiometric) post-translational modifications or to detect low abundance proteins, e.g., in biomarker discovery.
3. Practical Requirements for Chemical Tagging Approaches
Several conditions have to be met so that a chemical reaction can be successfully employed for protein or peptide modifications in proteome analysis. These requirements will depend on the techniques that are utilized for separation and detection, and some may differ for gel-based and gel-free approaches (14, 18). In all cases, reaction conditions as well as the specificity and the quantitative nature of the reaction are of considerable importance. Many of the tagging reactions that are described in the following therefore rely on the modification of amino groups, either at the N-terminus of a peptide or at the side chain of Lys residues. There are a considerable number of reactions with sufficient specificity toward amines that may be performed and – with the exception of an N-terminally modified, Lys-free peptide – all peptides in a sample will be amenable to such a modification reaction. Sample complexity after in-gel digestion of a protein spot is not a problem. Therefore, methods that target other residues are less frequently used compared to gel-free approaches where they are used for affinity tagging purposes, for example.
Methods in Molecular Biology: Two-Dimensional Electrophoresis Protocols
89
In any experimental set-up, a less than quantitative conversion of the analytes will result in a more complex mixture. For gel-free strategies that include a digestion step of the total protein mixture prior to any downstream separation step, this can be considered more detrimental. However, also in 2DE-based workflows, remaining unreacted peptides or side-products from a tagging reaction complicate automated data analysis and may lead to lower or less significant scores. Methods that rely on in-gel modification reactions must be critically examined as diffusion of reagents into the gel typically results in slower reaction speeds. The occurrence of side reactions other than at the targeted residue also complicates the data analysis process and may prevent the use of filtering based on amino acid composition information. Some chemical reactions require quite drastic conditions (pH extremes, high temperatures) under which some functional groups or modifications in peptides or proteins may not be stable. This may lead to sample losses in the process. The same has to be considered when additional clean-up steps are required to remove excess reagents that would otherwise interfere with MS analysis. If the protein amount in a gel spot is very small, potential benefits of a tagging reaction might thus be offset by reduced recovery rates. Tags that are attached to proteins for the purpose of affinity purification steps prior to 2DE sometimes prevent identification of labeled regions in the protein due to unfavorable ionization/ fragmentation properties or shielding of digestion sites resulting in unexpectedly high numbers of missed cleavages.
4. Selected Examples from the Literature 4.1. Tagging to Enhance MS Sensitivity and Increase Sequence Coverage
In recent years, a number of tagging reactions have been proposed to improve sensitivity and protein sequence coverage from MS analyses, especially in the PMF mode. Among those, the guanidination of Lys residues is the most commonly used modification chemistry (see also Chapter “De Novo Sequence Analysis of N-Terminal Sulfonated Peptides After In-Gel Guanidination”). The reaction of the ε-amino group of Lys with O-methyl isourea (Fig. 3a) results in attachment of a strongly basic guanidino group to the side chain of Lys. The newly formed modified amino acid, homoarginine, differs only by an additional –CH2–group from Arg and possesses very similar physicochemical properties. Guanidination attempts to overcome the bias against Lyscontaining peptides compared to Arg-containing peptides when tryptic digests are analyzed by MALDI-MS (19, 20). Following the modification reaction, all peptides with the exception of
90
Leitner and Lindner
Fig. 3. Chemical tags to increase sensitivity/improve ionization for MS analysis: (a) Guanidination of Lys residues [25–31]; (b) imidazole-type Lys tag [34–38]; (c) coumarin tag [39–40], and (d) dansyl tag [42] as examples for hydrophobic tags attached to amino groups.
the C-terminal peptide will carry the guanidine group which – because of its higher basicity – facilitates ionization. Different labeling protocols have been described in the literature (see 25, 26), all of them require a relatively high solution pH and a sample clean-up step after the reaction because of the large reagent excess being used. Side reactions for peptides carrying Gly at the N-terminus have been observed under some conditions (26), and deamidation of Asn residues might also occur to a small extent (27). First applications in combination with 2DE have been described by Brancia et al. (25, 28), connecting the tagging reaction to an improved MS response and sequence coverage. Cockrill et al. have demonstrated on-target guanidination of in-gel and in-solution digests just prior to MALDI analysis (29), while Sergeant et al. introduced an in-gel guanidination protocol (30; see Chapter “De Novo Sequence Analysis of N-Terminal Sulfonated Peptides After In-Gel Guanidination”). Most recently, Gaskell and coworkers have employed guanidination-type tagging for the analysis of gel spots from the separation of cell lysates (31). In this work, samples were processed automatically by a liquid handling system, demonstrating the applicability of this derivatization protocol to high-throughput analysis. From a majority of the spots
Methods in Molecular Biology: Two-Dimensional Electrophoresis Protocols
91
that were analyzed, higher scores from database searches and/ or higher sequence coverage were also obtained in this study. In addition, 15N2-OMIU was used to enable relative quantification by differential isotope coding (see Subheading 4.3). Frequently, guanidination is also used for the specific modification of Lys amino groups so that only the N-terminal amino group in a peptide is available for further reactions. This has, for example, been used in combination with sulfonation chemistries (see Subheading 4.2) or differential isotope coding with aminereactive reagents for quantification purposes (32). Similar derivatization chemistry involves the amidination of amino groups by reagents such as S-methyl thioacetimidate (27, 33), although the application for gel-derived samples has not been reported up to now. Lys residues may also be modified with an imidazole-type tag originally introduced by Peters et al. (34). The reagent, 2-methoxy-4, 5-dihydro-1H-imidazole, attaches an ionizationenhancing moiety to the ε-amino group, similar to the guanidination reaction (Fig. 3b). The reagent was subsequently commercialized by Agilent Technologies, but has not found widespread use yet (35–38). Potential advantages include enhanced sensitivity, improved peptide fragmentation, and applicability for differential quantification approaches. As an alternative to the attachment of a more polar functional group, the introduction of hydrophobic moieties has also been found to increase the MS response. This is believed to be the result of improved cocrystallization with the MALDI matrix, although this effect is dependent on the preparation technique. For example, Karger and coworkers have used fluorescent coumarin tags to derivatize peptides (Fig. 3c; 39, 40) and have reported improved performance in MALDI-MS. Ullmer et al. used a fluorescent quinoline-type reagent for the same purpose (41). Along this line, Kim and coworkers have used dansylation to improve sequence coverage with similar effects (Fig. 3d; 42). For these tagging chemistries, however, efficient peptide fragmentation typically is impaired (40, 43) so that these tags cannot be used for MS/MS approaches in a straightforward manner. 4.2. Tagging to Improve Peptide Fragmentation and de Novo Sequencing
As mentioned previously, MALDI peptide mass fingerprinting is still the most frequently employed MS workflow in combination with 2DE. When carried out on instruments equipped with TOF analyzers, it is possible to perform peptide sequencing by PSD experiments, although the data quality is usually not comparable with collision-induced dissociation MS/MS spectra. Singly charged ions typically formed by MALDI have the additional disadvantage of a less favorable fragmentation when compared to multiply charged ions that are more commonly formed by electrospray ionization (44). Thus, there is a clear need for
92
Leitner and Lindner
methods to improve peptide fragmentation for MALDI-PSD applications, in particular with regard to de novo sequencing, i.e., the elucidation of (partial) sequence information directly from the spectra without referring to database searches. At the end of the 1990s, Keough and coworkers have demonstrated that sulfonation of the N-terminus of a peptide improves its fragmentation behavior, particularly for MALDI-PSD set-ups (45, 46). The attachment of a permanently negatively charged group on the N-terminus results in the observation of C-terminal y-type fragment ions exclusively. Ideally, the peptide sequence can be read from the PSD or CID-MS/MS spectra beginning from the N-terminus. Different sulfonation reagents have been explored for the reaction (45, 46), although 3-sulfopropionic acid N-hydroxysuccinimide (Fig. 4a) was found most promising as it may be applied in aqueous solution. This compound has been made commercially available by Amersham Biosciences (now GE Healthcare) under the name Ettan CAF MALDI Sequencing Kit. The CAF (chemically assisted fragmentation) concept has been applied in a number of recent studies by different groups. As an example for a study in which 2DE was employed, Samyn and coworkers applied sulfonation chemistry to identify proteins from Halorhodospira halophila, an extremophilic eubacterium whose genome has not yet been sequenced, by MALDI-TOF/ TOF-based de novo sequencing (47). The same group also was able to analyze proteins from the banana plant by homologybased searching following sulfonation (48). A potential disadvantage of sulfonation is reduced sensitivity in positive MS mode as a result of the negative charge in the tag. It is therefore useful to block the amino group of Lys residues to avoid the introduction of multiple tags in a single peptide. This is usually done by guanidination (see earlier). Another complementary sulfonation reagent that has been proposed is 4-sulfophenyl isothiocyanate (SPITC, Fig. 4b). Orig-
Fig. 4. Chemical tags to improve peptide de novo sequencing by suppressing C-terminal fragment ions (b-ions): N-terminal sulfonation by (a) 3-sulfopropionic acid N-hydroxysuccinimide (47–48), and (b) 4-sulfophenyl isothiocyanate (49–54).
Methods in Molecular Biology: Two-Dimensional Electrophoresis Protocols
93
inally applied to peptide derivatization by Gevaert and colleagues (49), SPITC has been used successfully to simplify tandem MS spectra of peptides for de novo sequencing purposes. Chen et al. derivatized in-gel digests of proteins following consecutive guanidination and sulfonation steps while tryptic peptides were applied to ZipTips, allowing efficient removal of reagents (50). Recently, Hjernø et al. employed the technique to enhance protein identification in strawberries (51). Cotter and coworkers developed another optimized protocol (52) not requiring organic cosolvents and have used N-terminal sulfonation by SPITC to study protein ubiquitination (see Chapter “Detection of Ubiquitination in 2DE”) although not in combination with 2DE (53). Stable isotope coding using SPITC has also been recently demonstrated (54). Very recently, the group of Novotny has proposed 3-sulfobenzoic acid succinimidyl ester as a novel sulfonation reagent (55). The main advantage for this reagent was that the authors observed a reduced tendency of partial cleavage of the tag during MALDITOF/TOF MS/MS compared to the established reagents discussed earlier. Other chemically introduced tags have been shown to improve the fragmentation of peptides. For example, the imidazole-type Lys tag described earlier (34) was found to simplify the appearance of MS/MS spectra, as y-ions tend to dominate the spectra. 4.3. Tagging for Quantification Purposes
Various staining protocols for the relative quantification of gel spots are currently in use (22, 56), although they may suffer from limited dynamic range and lead to erroneous results for comigrating spots. In alternative gel-free workflows, differential isotope coding is frequently used for quantification purposes (14, 16–18, 57). Apart from the metabolic introduction of isotope labels during cell culture (58), such labels may again be introduced by chemical means (for detailed discussion, see refs.14, 18 and references therein). Peptides originating from two samples that differ only by their isotopic composition ideally behave identically in further separation steps and during ionization prior to MS analysis although they can be differentiated by their different masses. Two of the most widely used tagging concepts are the ICAT (isotope-coded affinity tag; 59–61) and iTRAQ (isobaric tag for relative and absolute quantification, 62; Fig. 5a), both commercialized by Applied Biosystems (Foster City, CA, USA). ICAT combines an affinity tag that allows enrichment of cysteine-containing peptides from complex mixtures (such as those processed by gel-free “shotgun” proteomics) with differential isotope coding. iTRAQ involves attachment of a tag that enables differentiation of isotope coded variants only at the MS/MS level and thus requires MS/MS workflows. For more detailed discussions on the general
94
Leitner and Lindner
Fig. 5. Chemical tags for relative quantification by stable isotope labeling: (a) iTRAQ label (62, 65, 66) and (b) ICPL tag (71–73, 75).
advantages and drawbacks of these methods, see a recent review (18). Although quantification via gel scanning commonly yields sufficiently reliable data, attempts have also been made to combine 2DE with these and other stable isotope labeling protocols to overcome known methodological limitations. For example, ICAT-based relative quantification has recently been used by Froment et al. to study the human proteasome (63), while Jungblut and coworkers (64) applied ICAT labeling to rat liver proteasomes. iTRAQ tags were employed by Sachon et al. (65) as well as Borner et al. (66) for 2DE. Several studies have also compared gel-based quantification with either ICAT or iTRAQ methodology, or both these approaches (67–70). Lottspeich and coworkers have applied their ICPL (isotopecoded protein label; Fig. 5b) strategy in several 2DE studies (71–73). ICPL is based on a nicotinoylation chemistry originally introduced by James et al. (74) and differs from many other SIL strategies in that the tagging reaction is typically performed on the protein rather than the peptide level. In the most recent work, ICPL labeling was applied to the study of phosphorylation events during apoptosis (75). Several SIL tags that target Cys residues have also shown to be applicable to 2DE. For example, tagging of thiol groups in Cys residues with differentially labeled acrylamide (76–78), Nethylmaleimide, or iodoacetamide (79). In addition, guanidination of Lys residues using differentially isotope-coded variants of the reagent was successful (31, 80), as noted earlier. These and other residue-specific tagging chemistries provide the additional benefit of using the amino acid composition of a given peptide as a filter for data analysis (28, 81–83). Gaskell et al., for example, performed a detailed bioinformatic study and demonstrated that information
Methods in Molecular Biology: Two-Dimensional Electrophoresis Protocols
95
about the presence or absence of certain amino acids in a peptide can compensate lower mass accuracies in PMF experiments (81). However, there is still a lack of widely available software that takes such information into account, as we discussed recently (83). 4.4. Tagging for Profiling Post-translational Modifications
As already pointed out in Chapters “Difficult Proteins” and “Organelle Proteomics”, a major issue affecting all separation techniques in proteomic research is the range of protein abundance in a given sample. For example, it is well known that plasma protein concentrations vary by roughly ten orders of magnitude, with albumin contributing approximately half of the total plasma protein amount. To increase the depth of proteome coverage, it is not easy to increase the starting protein amount, as the presence of high abundance proteins as well as overloading effects will lead to the deterioration of the separation quality. To overcome this, various technologies have been introduced that either deplete the major proteins or enrich low abundance ones. As depletion techniques, several manufacturers have recently introduced immunoaffinity columns containing antibodies directed against major plasma proteins, for example. Another approach is the “Equalizer” technology developed by Ciphergen (Fremont, CA, USA) that is based on a solid-phase peptide library (84). Enrichment of certain protein or peptide classes, on the other hand, is particularly relevant for the study of proteins that carry rare post-translational modifications. In this regard, chemical tagging is also being used as part of some preconcentration strategies. For example, Snyder et al. have developed a strategy for enrichment of proteins carrying S-nitrosylated cysteine residues (85). Their “biotin-switch” method is based on the selective reduction and biotinylation of NO-Cys residues after a prior protection of unaffected cysteines. Biotinylated peptides are subsequently enriched by avidin affinity chromatography. The method has been quite frequently applied to study S-nitrosylation (for recent examples, see 86, 87); recently, an improved tagging protocol has been reported by Hao et al. (88). Biotinylation of functional groups of interest is generally a widely employed technique to enrich protein subclasses prior to 2DE, although the avidin affinity chromatography step has its drawbacks such as potential irreversible and/or nonspecific binding. However, a wide range of biotinylation reagents are commercially available and the technique is considered mature. Thus, biotin tags have been used to probe cell surface proteomes (89–90) using a nonmembrane-permeable tag directed toward amino groups, for example. Oxidized proteins carrying carbonyl groups may be labeled by biotin hydrazide-type tags, as demonstrated by Yoo and Regnier (91). Holland et al. recently used
96
Leitner and Lindner
biotinylation of Cys residues to probe isoforms of k-casein (92) in the presence of a- and b-caseins, as these latter are Cys-free proteins. Enrichment of the κ-isoforms allowed their profiling by 2DE, as otherwise all caseins would migrate to similar positions in the gel, thus obscuring minor isoforms. Other, more common PTMs such as phosphorylation (see Chapter “Phosphoproteome Analysis by In-Gel Isoelectric Focusing and Tandem Mass Spectrometry”) or glycosylation are more frequently identified by specific in-gel staining or blotting, or enrichment methods that do not involve chemical tags, e.g., immunoprecipitation or affinity chromatography.
5. Conclusions and Outlook As we have outlined here, chemical tags are in use for a variety of applications in 2DE-based proteomics research. Tags may enhance sensitivity and facilitate data analysis, especially in cases where no annotated databases are available. In addition, novel relative quantification approaches based on stable isotope-coded tags complementary to densitometry are now possible. Finally, various chemically introduced affinity tags help to advance the study of post-translationally modified proteins. In the future, we envisage that more tools will be developed, in part ones that are already used in gel-free approaches; but also new, innovative applications for chemical reactions will continue to find their place in proteomics research. We apologize to all the researchers whose work could not be cited due to space limitations.
References 1. Henzel, W. J., Billeci, T. M., Stults, J. T., Wong, S. C., Grimley, C., and Watanabe, C. (1993) Identifying proteins from two-dimensional gels by molecular mass searching of peptide fragments in protein sequence databases. Proc. Natl. Acad. Sci. U.S.A. 90, 5011– 5015 2.. Yates, III, J. R., Speicher, S., Griffin, P. R., and Hunkapiller, T. (1993) Peptide mass maps: a highly informative approach to protein identification. Anal. Biochem. 214, 397–408 3. James, P., Quadroni, M., Carafoli, E., and Gonnet, G. (1994) Protein identification in DNA databases by peptide mass fingerprinting. Protein Sci. 3, 1347–1350
4. Pappin, D. J. C., Hojrup, P., and Bleasby, A. J. (1993) Rapid identification of proteins by peptide-mass fingerprinting. Curr. Biol. 3, 327–332 5. Mann, M., Hojrup, P., and Roepstorff, P. (1993) Use of mass spectrometric molecular weight information to identify proteins in sequence databases. Biol. Mass Spectrom. 22, 338–345 6. Nesvizhskii, A. I., and Aebersold, R. (2005) Interpretation of shotgun proteomic data – the protein inference problem. Mol. Cell. Proteomics 4, 1419–1440 7. Canelle, L., Pionneau, C., Marie, A., Bousquet, J., Bigeard, J., Lutomski, D., et al. (2004)
Methods in Molecular Biology: Two-Dimensional Electrophoresis Protocols
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
Automating proteome analysis: improvements in throughput, quality and accuracy of protein identification by peptide mass fingerprinting. Rapid Commun. Mass Spectrom. 18, 2785–2794 Jahn, O., Hesse, D., Reinelt, M., and Kratzin, H. D. (2006) Technical innovations for the automated identification of gel-separated proteins by MALDI-TOF mass spectrometry. Anal. Bioanal. Chem. 386, 92–103 Sun, W., Gao, S. J., Wang, L. J., Chen, Y., Wu, S. Z., Wang, X. R., et al. (2006) Microwave-assisted protein preparation and enzymatic digestion in proteomics. Mol. Cell. Proteomics 5, 769–776 Lopez-Ferrer, D., Capelo, J. L., and Vazquez, J. (2006) Ultra fast trypsin digestion of proteins by high intensity focused ultrasound. J. Proteome Res. 4, 1569–1574 Bienvenut, W. V., Deon, C., Pasquarello, C., Campbell, J. M., Sanchez, J. C., Vesta, M. L., et al. (2002) Matrix-assisted laser desorption/ ionization-tandem mass spectrometry with high resolution and sensitivity for identification and characterization of proteins. Proteomics 2, 868–876 Xu, C., and Ma, B. (2006) Software for computational peptide identification from MS-MS data. Drug Disc. Today 11, 595–600 Palagi, P. M., Hernandez, P., Walther, D., and Appel, R. D. (2006) Proteome informatics I: bioinformatics tools for processing experimental data. Proteomics 6, 5435–5444 Leitner, A., and Lindner, W. (2004) Current chemical tagging strategies for proteome analysis by mass spectrometry. J. Chromatogr. B 813, 1–26 Mirzaei, H., and Regnier, F. (2005) Structure specific chromatographic selection in targeted proteomics. J. Chromatogr. B 817, 23–34 Ong, S.-E., and Mann, M. (2005) Mass spectrometry-based proteomics turns quantitative. Nat. Chem. Biol. 1, 252–262 Regnier, F. E., and Julka, S. (2006) Primary amine coding as a path to comparative proteomics. Proteomics 6, 3968–3979 Leitner, A., and Lindner, W. (2006) Chemistry meets proteomics: the use of chemical tagging reactions for mass spectrometry-based proteomics. Proteomics 6, 5418–5434 Krause, E., Wenschuh, H., and Jungblut, P. R. (1999) The dominance of arginine-containing peptides in MALDI-derived tryptic mass fingerprints of proteins. Anal. Chem. 71, 4160– 4165 Baumgart, S., Lindner, Y., Kühne, R., Oberemm, A., Wenschuh, H., and Krause, E. (2004) The
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
97
contributions of specific amino acid side chains to signal intensities of peptides in matrix-assisted laser desorption/ionization mass spectrometry. Rapid Commun. Mass Spectrom. 18, 863–868 Schlosser, G., Pocsfalvi, G., Huszár, E., Malorni, A., and Hudesz, F. (2005) MALDITOF mass spectrometry of a combinatorial peptide library: effect of matrix composition on signal suppression. J. Mass Spectrom. 40, 1590–1594 Miller, I., Crawford, J., and Gianazza, E. (2006) Protein stains for proteomic applications: which, when, why? Proteomics 6, 5385–5408 Jacobs, J. M., Adkins, J. N., Qian, W.-J., Liu, T., Shen, Y. F., et al. (2005) Utilizing human blood plasma for proteomic biomarker discovery. J. Proteome Res. 4, 1073–1085 Gong, Y., Li, X., Yang, B., Ying, W. T., Li, D., Zhang, Y. J., et al. (2006) Different immunoaffinity fractionation strategies to characterize the human plasma proteome. J. Proteome Res. 5, 1379–1387 Brancia, F. L., Oliver, S. G., and Gaskell, S. J. (2000) Improved matrix-assisted laser desorption/ionization mass spectrometric analysis of tryptic hydrolysates of proteins following guanidination of lysine-containing peptides. Rapid Commun. Mass Spectrom. 14, 2070–2073 Beardsley, R. L., and Reilly, J. P. (2002) Optimization of guanidination procedures for MALDI mass mapping. Anal. Chem. 74, 1884–1890 Beardsley, R. L., Sharon, L. A., and Reilly, J. P. (2005) Peptide de novo sequencing facilitated by a dual-labeling strategy. Anal. Chem. 77, 6300–6309 Brancia, F. L., Butt, A., Beynon, R. J., Hubbard, S. J., Gaskell, S. J., and Oliver, S. G. (2001) A combination of chemical derivatisation and improved bioinformatic tools optimises protein identification for proteomics. Electrophoresis 22, 552–559 Cockrill, S. L., Foster, K. L., Wildsmith, J., Goodrich, A. R., Dapron, A. R., Hassell, T. C., et al. (2005) Efficient micro-recovery and guanidination of peptides directly from MALDI target spots. Biotechniques 38, 301–304 Sergeant, K., Samyn, B., Debyser, G., and Van Beeumen, J. (2005) De novo sequence analysis of N-terminal sulfonated peptides after in-gel guanidination. Proteomics 5, 2369–2380 Warwood, S., Mohammed, S., Cristea, I. M., Evans, C., Whetton, A. D., and Gaskell, S. J.
98
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
Leitner and Lindner (2006) Guanidination chemistry for qualitative and quantitative proteomics. Rapid Commun. Mass Spectrom. 20, 3245–3256 Ji, C., Guo, N., and Li, L. (2005) Differential dimethyl labeling of N-termini of peptides after guanidination for proteome analysis. J. Proteome Res. 4, 2099–2108 Beardsley, R. L., and Reilly, J. P. (2003) Quantitation using enhanced signal tags: a technique for comparative proteomics. J. Proteome Res. 2, 15–21 Peters, E. C., Horn, D. M., Tully, D. C., and Brock, A. (2001) A novel multifunctional labeling reagent for enhanced protein characterization with mass spectrometry. Rapid Commun. Mass Spectrom. 15, 2387–2392 Cindric, M., Cepo, T., Skrlin, A., Vuletic, M., and Bindila, L. (2006) Accelerated on-column lysine derivatization and cysteine methylation by imidazole reaction in a deuterated environment for enhanced product ion analysis. Rapid Commun. Mass Spectrom. 20, 694–702 Fierro-Monti, I., Mohammed, S., Matthiesen, R., Santoro, R., Burns, J. S., Williams, D. J., et al. (2006) Quantitative proteomics identifies Gemin5, a scaffolding protein involved in ribonucleoprotein assembly, as a novel partner for eukaryotic initiation factor 4E. J. Proteome Res. 5, 1367–1378 Cindric, M., Bindila, L., Cepo, T., and PeterKatalinic, J. (2006) Mass spectrometry-based glycoproteomic approach involving lysine derivatization for structural characterization of recombinant human erythropoietin. J. Proteome Res. 5, 3066–3076 Joss, J. L., Molloy, M. P., Hinds, L. A., and Deane, A. M. (2006) Evaluation of chemical derivatisation methods for protein identification using MALDI MS/MS. Int. J. Pept. Res. Ther. 12, 225–235 Pashkova, A., Moskovets, E., and Karger, B. L. (2004) Coumarin tags for improved analysis of peptides by MALDI-TOF MS and MS/ MS. 1. Enhancement in MALDI MS signal intensities. Anal. Chem. 76, 4550–4557 Pashkova, A., Chen, H.-S., Rejtar, T., Zang, X., Giese, R., Andreev, V., et al. (2005) Coumarin tags for analysis of peptides by MALDI-TOF MS and MS/MS. 2. Alexa Fluor 350 tag for increased peptide and protein identification by LC-MALDI-TOF/TOF MS. Anal. Chem. 77, 2085–2096 Ullmer, R., Plematl, A., and Rizzi, A. (2006) Derivatization by 6-aminoquinolyl-N-hydroxysuccinimidyl carbamate for enhancing the ionization yield of small peptides and glycopeptides in matrix-assisted laser desorption/ ionization and electrospray ionization mass
42.
43.
44.
45.
46.
47.
48.
49.
50.
spectrometry. Rapid Commun. Mass Spectrom. 20, 1469–1479 Park, S.-J., Song, J.-S., and Kim, H.-J. (2005) Dansylation of tryptic peptides for increased sequence coverage in protein identification by matrix-assisted laser desorption/ionization time-of-flight mass spectrometric peptide mass fingerprinting. Rapid Commun. Mass Spectrom. 19, 3089–3096 Amoresano, A., Monti, G., Cirulli, C., and Marino, G. (2006) Selective detection and identification of phosphopeptides by dansyl MS/MS/MS fragmentation. Rapid Commun. Mass Spectrom. 20, 1400–1404 Cramer, R., and Corless, S. (2001) The nature of collision-induced dissociation processes of doubly protonated peptides: comparative study for the future use of matrix-assisted laser desorption/ionization on a hybrid quadrupole time-of-flight mass spectrometer in proteomics. Rapid Commun. Mass Spectrom. 15, 2058–2066 Keough, T., Youngquist, R. S., and Lacey, M. P. (1999) A method for high-sensitivity peptide sequencing using post-source decay matrix-assisted laser desorption ionisation mass spectrometry. Proc. Natl. Acad. Sci. U.S.A. 96, 7131–7136 Keough, T., Lacey, M. P., and Youngquist, R. S. (2002) Solid-phase derivatization of tryptic peptides for rapid protein identification by matrix-assisted laser desorption/ionization mass spectrometry. Rapid Commun. Mass Spectrom. 16, 1003–1015 Samyn, B., Sergeant, K., Memmi, S., Debyser, G., Devreese, B., and van Beeumen, J., (2006) MALDI-TOF/TOF de novo sequence analysis of 2-D PAGE-separated proteins from Halorhodospira halophila, a bacterium with unsequenced genome. Electrophoresis 27, 2702–2711 Samyn, B., Sergeant, K., Carpentier, S., Debyser, G., Panis, B., Swennen, R., et al. (2007) Functional proteome analysis of the banana plant (Musa spp.) using de novo sequence analysis of derivatized peptides. J. Proteome Res. 6, 70–80 Gevaert, K., Demol, H., Martens, L., Hoorelbeke, B., Pupye, M., Goethals, M, et al. (2001) Protein identification based on matrix assisted laser desorption/ionization-post source decay-mass spectrometry. Electrophoresis 22, 1645–1651 Chen, P., Nie, S., Mi, W., Wang, X.-C., and Liang, S.-P. (2004) De novo sequencing of tryptic peptides sulfonated by 4-sulfophenyl isothiocyanate for unambiguous protein identification using post-source decay matrix-assisted
Methods in Molecular Biology: Two-Dimensional Electrophoresis Protocols
51.
52.
53.
54.
55.
56.
57.
58.
59.
60.
61.
laser desorption/ionization mass spectrometry. Rapid Commun. Mass Spectrom. 18, 191–198 Hjernø, K., Alm, R., Canbäck, B., Matthiesen, R., Trajkovski, K., Bjork, L., et al. (2006) Down-regulation of the strawberry Bet v 1-homologous allergen in concert with the flavonoid biosynthesis pathway in colorless strawberry mutant. Proteomics 6, 1574–1587 Wang, D., Kalb, S. R., and Cotter, R. J. (2004) Improved procedures for N-terminal sulfonation of peptides for matrix-assisted laser desorption/ionization post-source decay peptide sequencing. Rapid Commun. Mass Spectrom. 18, 96–102 Wang, D., and Cotter, R. J. (2005) Approach for determining protein ubiquitination sites by MALDI-TOF mass spectrometry. Anal. Chem. 77, 1458–1466 Guillaume, E., Panchaud, A., Affolter, M., Desvergnes, V., and Kussmann, M. (2006) Differentially isotope-coded N-terminal protein sulphonation: combining protein identification and quantification. Proteomics 6, 2338–2349 Alley Jr., W., Mechref, Y., Klouckova, I., and Novotny, M. V. (2007) Improved collisioninduced dissociation analysis of peptides by matrix-assisted laser desorption/ionization tandem time-of-flight mass spectrometry through 3-sulfobenzoic acid succinimidyl ester labeling. J. Proteome Res. 6, 124–132 Righetti, P. G., Castagna, A., Antonucci, F., Piubelli, C., Cecconi, D., Campostrini, N., et al. (2004) Critical survey of quantitative proteomics in two-dimensional electrophoretic approaches. J. Chromatogr. A 1051, 3–17 Yan, W., and Chen, S. S. (2005) Mass spectrometry-based quantitative proteomic profiling. Brief. Funct. Genomic. Proteomic. 4, 27–38 Beynon, R. J., and Pratt, J. M. (2005) Metabolic labeling of proteins for proteomics. Mol. Cell. Proteomics 4, 857–872 Gygi, S. P., Rist, B., Gerber, S. A., Turecek, F., Gelb, M. H., and Aebersold, R. (1999) Quantitative analysis of complex protein mixtures using isotope-coded affinity tags. Nat. Biotechnol. 17, 994–999 Smolka, M., and Zhou, H. (2002) Quantitative protein profiling using two-dimensional gel electrophoresis, isotope-coded affinity tag labeling, and mass spectrometry. Mol. Cell. Proteomics 1, 19–29 Hansen, K. C., Schmitt-Ulms, G., Chalkley, R. J., Hirsch, J., et al. (2003) Mass spectrometric analysis of protein mixtures at low levels using
62.
63.
64.
65.
66.
67.
68.
69.
70.
99
cleavable 13C-isotope-coded affinity tag and multidimensional chromatography. Mol. Cell. Proteomics 2, 299–314 Ross, P. L., Huang, Y. L. N., Marchese, J. N., Williamson, B., Parker, K., Hattan, S., et al. (2004) Multiplexed protein quantitation on Saccharomyces cerevisiae using amine-reactive isobaric tagging reagents. Mol. Cell. Proteomics 3, 1154–1169 Froment, C., Uttenweiler-Joseph, S., Bousquet-Dubouch, M.-P., Matondo, M., Borges, J. P, Esmenjaud, C., et al. (2005) A quantitative proteomic approach using two-dimensional gel electrophoresis and isotope-coded affinity tag labeling for studying human 20S proteasome heterogeneity. Proteomics 5, 2351–2363 Schmidt, F., Dahlmann, B., Janek, K., Kloß, A., Wacker, M., Ackermann, R., et al. (2006) Comprehensive quantitative proteome analysis of 20S proteasome subtypes from rat liver by isotope coded affinity tag and 2-D gelbased approaches. Proteomics 6, 4622–4632 Sachon, E., Mohammed, S., Bache, N., and Jensen, O. N. (2006) Phosphopeptide quantitation using amine-reactive isobaric tagging reagents and tandem mass spectrometry: applications to proteins isolated by gel electrophoresis. Rapid Commun. Mass Spectrom. 20, 1127–1134 Borner, G. H. H., Harbour, M., Hester, S., Lilley, K. S., and Robinson, M. S. (2006) Comparative proteomics of clathrin-coated vesicles. J. Cell Biol. 175, 571–578 Schmidt, F., Donahoe, S., Hagens, K., Mattow, J., Schaible, V. E., Kaufmann, S. H. E., et al. (2004) Complementary analysis of the Mycobacterium tuberculosis proteome by two-dimensional electrophoresis and isotopecoded affinity tag technology. Mol. Cell. Proteomics 3, 24–42 Choe, L. H., Aggarwal, K., Franck, Z., and Lee, K. H. (2005) A comparison of the consistency of proteome quantitation using twodimensional electrophoresis and shotgun isobaric tagging in Escherichia coli cells. Electrophoresis 26, 2437–2449 Reinders, Y., Schulz, I., Gräf, R., and Sickmann, A. (2006) Identification of novel centrosomal proteins in Dictyostelium discoideum by comparative proteomic approaches. J. Proteome Res. 5, 589–598 Wu, W. W., Wang, G., Baek, S. J., and Shen, R.-F. (2006) Comparative study of three proteomic quantitative methods, DIGE, cICAT, and iTRAQ, using 2D gel- or LC-MALDI TOF/TOF. J. Proteome Res. 5, 651–658
100
Leitner and Lindner
71. Schmidt, A., Kellermann, J., and Lottspeich, F. (2005) A novel strategy for quantitative proteomics using isotope-coded labels. Proteomics 5, 4–15 72. Hochleitner, E. O., Kastner, B., Fröhlich, T., Schmidt, A., Luhrmann, R., Arnold, G., et al. (2005) Protein stoichiometry of a multiprotein complex, the human spliceosomal U1 small nuclear ribonucleoprotein. J. Biol. Chem. 280, 2536–2542 73. Sarioglu, H., Brandner, S., Jacobsen, C., Meindl, T., Schmidt, A., Kellermann, J., et al. (2006) Quantitative analysis of 2,3,7,8tetrachlorodibenzo-p-dioxin-induced proteome alterations in 5L rat hepatoma cells using isotope-coded protein labels. Proteomics 6, 2407–2421 74. Münchbach, M., Quadroni, M., Miotto, G., and James, P. (2000) Quantitation and facilitated de novo sequencing of proteins by isotopic N-terminal labeling of peptides with a fragmentation-directing moiety. Anal. Chem. 72, 4047–4057 75. Tabbert, A., Kappes, F., Knippers, R., Kellermann, J., Lottspeich, F., and Ferrando-May, E. (2006) Hypophosphorylation of the architectural chromatin protein DEK in death-receptor-induced apoptosis revealed by the isotope coded protein label proteomic platform. Proteomics 6, 5758–5772 76. Sechi, S. (2002) A method to identify and simultaneously determine the relative quantities of proteins isolated by gel electrophoresis. Rapid Commun. Mass Spectrom. 16, 1416–1424 77. Gehanne, S., Cecconi, D., Carboni, L., Righetti, P. G., Domenici, E., and Hamdan, M. (2002) Quantitative analysis of two-dimensional gel-separated proteins using isotopically marked alkylated agents and matrix-assisted laser desorption/ionization mass spectrometry. Rapid Commun. Mass Spectrom. 16, 1692–1698 78. Cahill, M. A., Wozny, W., Schwall, G., Schroer, K., Holzer, K., Poznanovic, S., et al. (2003) Analysis of relative isotopologue abundances for quantitative profiling of complex protein mixtures labelled with the acrylamide/D3-acrylamide alkylation tag system. Rapid Commun. Mass Spectrom. 17, 1283–1290 79. Kurono, S., Kurono, T., Komori, N., Niwayama, S., and Matsumoto, H. (2006) Quantitative proteome analysis using D-labeled N-ethylmaleimide and 13C-labeled iodoacetanilide by matrix-assisted laser desorption/ ionization time-of-flight mass spectrometry. Bioorg. Med. Chem. 14, 8197–8209 80. Brancia, F. L., Montgomery, H., Tanaka, K., and Kumashiro, S. (2004) Guanidino labeling
81.
82.
83.
84.
85.
86.
87.
88.
89.
derivatization strategy for global characterization of peptide mixtures by liquid chromatography matrix-assisted laser desorption/ ionization mass spectrometry. Anal. Chem. 76, 2748–2755 Sidhu, K. S., Sangvanich, P., Brancia, F. L., Sullivan, A. G., Gaskell, S. J., Wolkenhauer, O., et al. (2001) Bioinformatic assessment of mass spectrometric chemical derivatisation techniques for proteome database searching. Proteomics 1, 1368–1377 Zou, J., Turner, A. N., and Phelps, R. G. (2003) Mass and composition matrixassisted laser desorption ionization time of flight analysis enabling inference of the sequence of most peptides where the protein of origin is known. Anal. Chem. 75, 2653–2662 Leitner, A., Amon, S., Rizzi, A., and Lindner, W. (2007) Use of the arginine-specific butanedione/phenylboronic acid tag for analysis of peptides and protein digests using matrix-assisted laser desorption/ionization mass spectrometry. Rapid Commun. Mass Spectrom. 21, 1321–1330 Righetti, P. G., Boschetti, E., Lomas, L., and Citterio, A. (2006) Protein equalizer (TM) technology: the quest for a “democratic proteome”. Proteomics 6, 3980–3992 Jaffrey, S. R., Erdjument-Bromage, H., Ferris, C. D., Tempst, P., and Snyder, S. H. (2001) Protein S-nitrosylation: a physiological signal for neuronal nitric oxide. Nat. Cell Biol. 3, 193–197 Dall’Agnol, M., Bernstein, C., Bernstein, H., Garewal, H., and Payne, C. M. (2006) Identification of S-nitrosylated proteins after chronic exposure of colon epithelial cells to deoxycholate. Proteomics 6, 1654–1662 Greco, T. M., Hodara, R., Parastatidis, I., Heijnen, H. F. G., Dennehy, M. K., Liebler, D. C., et al. (2006) Identification of S-nitrosylation motifs by site-specific mapping of the S-nitrocysteine proteome in human vascular smooth muscle cells. Proc. Natl. Acad. Sci. U.S.A. 103, 7420–7425 Hao, G., Derakshan, B., Shi, L., Campagne, F., and Gross, S. S. (2006) SNOSID, a proteomic method for identification of cysteine S-nitrosylation sites in complex protein mixtures. Proc. Natl. Acad. Sci. U.S.A. 103, 1012–1017 Shin, B. K., Wang, H., Yim, A. M., Le Naour, F., Brichory, F., Jang, J. H., et al. (2003) Global profiling of the cell surface proteome of cancer cells uncovers an abundance of proteins with chaperone function. J. Biol. Chem. 278, 7607–7616
Methods in Molecular Biology: Two-Dimensional Electrophoresis Protocols 90. Jang, J. H., and Hanash, S. (2003) Profiling of the cell surface proteome. Proteomics 3, 1947–1954 91. Yoo, Y. S., and Regnier, F. E. (2004) Proteomic analysis of carbonylated proteins in two-dimensional gel electrophoresis using avidin–fluorescein affinity staining. Electrophoresis 25, 1334–1341
101
92. Holland, J. W., Deeth, H. C., and Alewood, P. F. (2006) Resolution and characterisation of multiple isoforms of bovine kappa-casein by 2-DE following a reversible cysteinetagging enrichment strategy. Proteomics 6, 3087–3095
Chapter 7 Immunoblotting 2DE Membranes Brian McDonagh Summary Two-dimensional electrophoresis (2DE) immunoblotting combines the resolving power of 2DE and the selectivity of antibodies, allowing the selection of individual or sets of proteins from the total proteome. It is essential when immunoblotting membranes that reproducible 2DE gels are obtained consistently. Afterwards, it is a matter of applying the same principles as 1DE immunoblotting. However, it should be noted that the most sensitive methods of detection should be used. This chapter outlines some of the considerations that should be taken into account, ranging from the choice and type of antibodies, buffers involved, and methods of detection and visualisation. Key words: Immunoblotting, Two-dimensional electrophoresis, Detection, Staining, Antibody, Review.
1. Introduction Two-dimensional electrophoresis (2DE) allows the separation of thousands of proteins according to their isoelectric point (pI) and relative molecular mass (Mr) (1) (see also Chapters “Two-Dimensional Electrophoresis (2DE): An Overview” and “Solubilization of Proteins in 2DE: An Outline”). If the gels are quantifiably stained, it can also lead to information regarding their relative abundance. The development of electroblotting allows the proteins to be transferred onto an inert membrane such as nitrocellulose or polvinylidene difluoride (PVDF) (2). Immobilisation of proteins (3) onto such inert surfaces allows them to be accessible to a variety of molecular probes including monoclonal and polyclonal antibodies (hence immunoblotting) but also to some sequencing technologies. David Sheehan and Raymond Tyther (eds.), Methods in Molecular Biology, Two-Dimensional Electrophoresis Protocols, vol. 519 © Humana Press, a part of Springer Science + Business Media, LLC 2009 DOI: 10.1007/978-1-59745-281-6_7
103
104
McDonagh
It is essential that reproducible and consistent 2DE gels are produced before proceeding to immunoblotting; afterwards the same principles of immunoblotting apply to 2DE as 1DE, although the most sensitive detection methods should be employed. Classical immunoblotting relies on the ability of a primary antibody to detect an epitope amongst the thousands of proteins on a membrane. A secondary antibody that recognises the primary antibody is often used to build up a protein-antibody-antibody sandwich and so to amplify the initial signal. The secondary antibody usually has an enzyme attached that allows detection by either chromogenic means or by chemiluminescence, when appropriate substrates are applied to the membrane, thus allowing detection on the membrane, X-ray film, or by digital chemiluminescence detector (CCD). Although it is semi-quantitative, the intensity of the spot relative to other spots on the membrane and the Mr relative to the marker can give an indication of how much of the protein of interest is in the sample. The appearance/disappearance of a spot compared to controls can indicate a modification of the protein depending on the primary antibody used. As this is such a broad area and there can be major differences in protocols depending on the protein epitope or antigen being analysed, this chapter will take each step in immunoblotting (Fig. 1) and explore some of the options available.
1. TOTAL PROTEIN STAIN
2. BLOCKING
BLOCK FOR 1H @ RT WASH 3 X 10 MIN INCUBATE WITH PRIMARY ANTIBODY
3. ANTIBODY INCUBATION
WASH 3 X 10 MIN INCUBATE WITH SECONDARY ANTIBODY WASH 3 X 10 MIN ANTIBODY DETECTION
4. ANTIBODY DETECTION
Fig. 1. General steps involved in immunoblotting. Total proteins stain was BLOT-FastStain (Genotech). Antibody used was anti-DNP, used for detection of derivatised carbonyl groups. Antibody detected using goat anti-rabbit HRP and chemiluminescence on X-ray film.
Immunoblotting 2DE Membranes
105
2. Total Protein Staining After electroblotting proteins onto a membrane from a gel, it is essential to stain the total protein on the membrane before incubation with antibodies. This is to ensure that the 2DE separation was reproducible with a good resolution of spots and that protein transfer was complete with no air bubbles. It is necessary to use a reversible total protein stain that is compatible with immunodetection products to be used in later stages (see Note 1). For a comprehensive review on protein staining for proteomic applications see (4).
3. Wash and Blocking Buffer Selection
The washing steps in immunoblotting are very important and can reduce non-specific binding of antibodies. The choice of buffer is usually between phosphate-buffered saline (PBS) and Tris-buffered saline (TBS) depending on the antibodies used. It has been recommended for instance that PBS buffer should not be used when probing for phosphate substrates, as the phosphate ions may interfere with detection. Washing buffer usually includes a small amount (~ 0.05%) of a detergent, e.g. Tween 20. This can help remove any non-specifically bound antibodies or any blocking substrate that may have clumped together. A high concentration of detergent may be too harsh and result in a low signal
4. Antibody Selection In some instances both monoclonal and polyclonal antibodies are available for the same epitope. The advantages of using a monoclonal antibody over a polyclonal include a better signalto-noise ratio and higher specificity. This is of great importance when probing for either a specific epitope or a more general posttranslational modification, e.g. carbonylation of amino acid side chains in oxidative stress. Other important considerations when selecting antibodies arise from the methods of detection. Primary antibodies are now available that have an enzyme linked to them as opposed to using a secondary antibody raised against the primary with an enzyme attached to the secondary antibody. The advantages of this include that it is quicker and will eliminate any cross-reactivity of the secondary antibody. However attaching the
106
McDonagh
enzyme directly may diminish the immunoreactivity of the primary antibody, there is little signal amplification, and there is no flexibility in the choice of primary antibody from one experiment to another. Using a secondary antibody with an enzyme attached does not affect the immunoreactivity of the primary antibody and the sensitivity is often increased. There are also a wide variety of primary and secondary antibodies available commercially, and this system is versatile in that many of the primary antibodies are raised in one species and the same secondary labelled antibody can be used. When using a new antibody it is usually necessary to optimise the dilutions to be used. Primary antibody dilutions are usually diluted with wash buffer containing blocking substrate, while secondary antibodies are diluted in washing buffer. The blocking buffer is important in reducing the amount of non-specific binding and can improve sensitivity by reducing background staining on blots. The blocking substrates most widely used include bovine serum albumin (BSA), non-fat milk powder, and casein. It is important in all cases that the blocker is well dissolved or it will show up as ‘speckles’ in the final blot. Again, the choice of blocker and its concentration may vary according to the antibodies being used. A general working solution that works well for the majority of proteins is 1% BSA in PBS-Tween (0.5%). If there is excess blocking buffer it may mask antibody antigen interactions. But it is important to remember that individual blocking buffers are not compatible with every system, for instance using milk with anti-biotin is not recommended as there may be some biotin naturally present in milk.
5. Antibody Detection The two most widely used methods of antibody detection are chromogenic detection and chemiluminescence. The two most common enzymes attached to antibodies for chromogenic detection are alkaline phosphatase (AP) and horseradish peroxidase (HRP). When the enzyme comes into contact with its chromogenic substrates, they are converted to an insoluble coloured product that precipitate on the membrane and therefore require no specific equipment for visualisation. AP is an enzyme of about 140 kDa and has an optimum pH range of 8–10, while HRP is about 40 kDa with a pH optimum of 5–7. There are a range of substrates that will react with both enzymes. The major advantage of chromogenic detection is that it stays linear for a longer period of time than chemiluminesence. Chemiluminesence detection relies on the release of energy in the form of light when a substrate comes in contact with HRP.
Immunoblotting 2DE Membranes
107
It is generally considered to be more sensitive than chromogenic, and it is also amenable to multiple exposures by stripping the membrane of antibodies and reprobing to visualise another protein or optimise detection. It has a large linear range but, over a period of time, light production decreases and stops, so optimisation of antibody dilutions is crucial. Chemiluminesence can be detected by X-ray film in a darkroom or by a CCD detection camera. With X-ray film the goal is to time exposure of the membrane to the film, so there are clear distinct spots with low background (see Note 2). CCD detection cameras offer the advantages of instant image manipulation, greater resolution, and a large dynamic range. They also eliminate the need for darkroom facilities; however, the exposure time for the camera also needs to be optimised. The development of fluorescence 2-D difference gel electrophoresis (DIGE: see Chapters “High-Resolution 2DE”, “Two-Dimensional Difference Gel Electrophoresis”) (5) allowing multiple samples to be co-separated on one 2DE gel, has also led to developments in immunoblotting. The CyDye™ Fluors (GE Healthcare) enable fluorescent detection of antibodies. They are reported to have a wider linear range than conventional enzymelinked antibodies and are stable for a much longer period of time. It also allows multiplexing on the one membrane, i.e. using two separate antibodies. The different antibodies can then be detected depending on the CyDye conjugated to the secondary antibody on a fluorescent scanner. This can allow the detection of two different proteins together, for instance a protein of interest and a housekeeping protein. It can also facilitate detection of post-translational modifications of the same protein, e.g. phosphorylation, if there are primary antibodies available. Thus, this method can eliminate the requirement for, and variability associated with, stripping and re-probing the membrane. The main disadvantage to this technology is the high cost of necessary equipment for visualisation and quantitative analysis and the CyDye-linked antibodies.
6. Stripping Chemiluminesence offers the advantage of using a stripping reagent and re-probing the membrane several times, thus saving time that would be necessary to electrophorese another sample. PVDF membranes are more amenable to this than nitrocellulose. Care must be taken as over-harsh conditions may damage the antigen or even remove protein from the membrane. If conditions are not harsh enough some of the previous detection reagents will
108
McDonagh
remain, although this can easily be checked by re-incubating with chemiluminescent substrate and checking for spots still visible. Generally stripping buffers consist of a reducing agent, detergent, and either a low or high pH, e.g.: 2% SDS, 62.5 mM Tris–HCl, pH 6.7, 0.5% β-mercaptoethanol The blot is incubated for 20 min at 40°C (see Note 3).
7. Antibody Optimisation Depending on the antigen or epitope it may be possible to optimise the conditions for immunoblotting by dot blotting. If not, conditions should be optimised on 1DE. The gels are blotted as normal and stained with a general Protein stain, e.g. Ponceau. It is helpful at this stage to mark where each lane is. The membrane should be blocked and then cut into its separate lanes. Each lane can then be exposed to a wide range of antibody dilutions, both of primary and secondary. For detection it is important to ensure that all lanes are exposed to substrates for the same length of time and developed or visualised together. The optimal dilutions should then become clear on detection and further optimisation can be achieved after this stage. Some common problems encountered with immunoblotting include a high background signal. This may be the result of crossreactivity with blocking agent and antibodies, including Tween in the washing buffer may help. It may be due to the concentration of antibodies being too high or over-long incubation with the antibody. Also it is important not to let membranes dry out for too long. When there is a very low or no signal it could be that the antibodies are too dilute, washing conditions may be too harsh, or there may be too high concentration of or noncompatible blocking agent.
8. Notes 1. If the total protein stain is quantifiable it can also allow comparisons between blots, i.e. that the actual protein load on the membrane is comparable between blots. For 2DE a sensitive method of staining is required; the BLOT-FastStain™ (Genotech) reports sensitivity in the 0.5-ng range. From experience of using different reversible stains, it is relatively quick, sensitive for 2DE, and reproducible.
Immunoblotting 2DE Membranes
109
2. This may involve timing different exposures and can vary with antibodies and dilutions of antibodies. 3. As every antibody-antigen interaction and target protein is unique, no one stripping condition will work in all cases. It is often necessary to optimise conditions accordingly. For instance changes in the incubation temperature or concentration of reducing agent or detergent can dramatically alter stripping results especially in 2DE immunoblotting.
References 1. O’Farrell, P. H. (1975). High resolution two-dimensional electrophoresis of proteins. J. Biol. Chem. 250, 4007–4021 2. Towbin, H., Staehelin, T., and Gordon, J. (1979). Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc. Natl. Acad. Sci. U.S.A. 76, 4350–4354 3. Burnette, W. N. (1981). “Western blotting”: electrophoretic transfer of proteins from sodium dodecyl sulfate – polyacrylamide gels
to unmodified nitrocellulose and radiographic detection with antibody and radioiodinated protein A. Anal Biochem. 112, 195–203 4. Miller, I., Crawford, J., and Gianazza, E. (2006). Protein stains for proteomic applications: which, when, why? Proteomics 6, 5385–5408 5. Unlu, M., Morgan, M. E., and Minden, J. S. (1997). Difference gel electrophoresis: a single gel method for detecting changes in protein extracts. Electrophoresis, 18, 2071–2077
Protocols Chapters
Chapter 8 Troubleshooting Image Analysis in 2DE Bettina Levänen and Åsa M. Wheelock Summary The image analysis part of gel-based proteome research plays an important role in the overall success of the experiment. The main purpose of software-assisted 2DE gel analysis is to detect the protein spots, match them between gels within an experiment, and identify any differences in protein expression between sets of samples. Efficient analysis of protein expression relies on automated image processing techniques. There are several factors to consider in the choice of software product, as well as in the implementation of the analysis itself. Successful quantification of protein expression levels is largely dependent on the algorithms for spot matching, normalization, and background subtraction provided by the 2DE analysis software. In addition to generic protocols for image acquisition and subsequent 2DE image analysis (using Progenesis PG200), this chapter describes methods for quantitative and qualitative evaluation of the quality of the image analysis. Key words: Image analysis; Proteomics, 2DE analysis software, Progenesis, Variance, Spot quantification, Background subtraction, Image acquisition.
1. Introduction One of the central purposes of the postexperimental image analysis of two-dimensional electrophoresis (2DE) gels is to locate and quantify alterations in protein expression, including downregulation, upregulation, and silencing of expressed proteins. As such, the software programs used for analyzing 2DE gel images are primarily developed to correctly define protein spot boundaries, match corresponding protein spots between gels, quantify them, as well as recognize statistically significant changes in protein expression. However, in addition to the biological variance of
David Sheehan and Raymond Tyther (eds.), Methods in Molecular Biology, Two-Dimensional Electrophoresis Protocols, vol. 519 © Humana Press, a part of Springer Science + Business Media, LLC 2009 DOI: 10.1007/978-1-59745-281-6_8
113
114
Levänen and Wheelock
interest, the 2DE separation method itself introduces considerable experimental variance. Accordingly, the emphasis in the image analysis process lies not only in finding changes in protein expression between different biological states, but also in reducing possible variance arising from the experiment by using a range of available algorithms. Although the order of events may differ between various 2DE software products, the following algorithms are generally included: Spot detection, image warping, spot matching, background subtraction, normalization, and spot quantification. The purpose and theory behind each of these steps, including the proceeding image acquisition step, are reviewed here. In order to perform the image analysis, the data contained in the gel have to be transformed into an appropriate digital format with sufficient dynamic range. The dynamic range depends both on the type of image acquisition instrumentation and on the protein visualization method used. As a result, two types of dynamic ranges should be considered. The first one refers to the ability of the protein stain to provide a linear relationship between the staining intensity and the actual amount of protein available in a 2DE spot. The fact that protein abundances in biological samples may span over as much as 12 orders of magnitude (see also Chapters “Difficult Proteins,” “Organelle Proteomics,” and “Applications of Chemical Tagging Approaches in Combination with 2DE and Mass Spectrometry”) puts high demands both on sensitivity and dynamic range of protein stains used in quantitative 2DE. A comparison of the three most commonly used visualization techniques (silver stain, SYPRO Ruby, and minimal difference gel electrophoresis (DIGE)) is provided in Fig. 1 (for DIGE see also Chapters “High-Resolution 2DE” and “TwoDimensional Difference Gel Electrophoresis”). The range of protein abundances on the x-axis corresponds to abundances from a few molecules up to micromolar concentrations, thus providing a comprehensive view of possible physiological protein abundances. Silver staining (Fig. 1, small/grey box), with its dynamic range of a mere 1–2 orders of magnitude is not well suited for quantitative proteomics studies. As evident by the detection range of minimal DIGE (Fig. 1, large/red box), even state-of-the-art protein detection methods cover a very small portion of the potential physiological range. The constraints provided by available protein detection technology emphasize the importance of transferring the quantitative information in its entirety during image acquisition. As such, the dynamic range of the pixel intensity also has to be considered. A pixel is a single point in a graphic image (“pix” being an abbreviation for picture, and “el” for element). 2DE programs use a grayscale to define the intensity of each spot, and the number of bits used to represent each pixel determines how many shades of gray can be displayed. The two main types of
Troubleshooting Image Analysis in 2DE
115
Fig. 1. The detection range of the most commonly used protein stains in 2DE shown in a perspective of the corresponding physiological range of protein abundances. The large box corresponds to the detection range of minimal DIGE, often considered to be the state of the art for quantitative 2DE. The smaller box corresponds to the detection range of silver stain, which in spite of its poor dynamic range still is commonly utilized in quantitative 2DE analysis. The dark grey/blue line represents the dynamic range of SYPRO Ruby staining.
image acquisition devices used in 2DE, laser scanner-based and CCD (charge-coupled device) camera-based instruments, typically offer a 16-bit range for the pixel intensity, making it possible to display pixel intensities in a range of 216 (65,536). For more information regarding the technical differences between the two types of instruments, please see (1). Resolution refers to the sharpness and clarity of an image, and is commonly quantified as “dots-per-inch” (dpi), i.e., number of pixels/area. However, laser scanners utilized in 2DE image acquisition often define the actual pixels size in μm, e.g., 50, 100, or 200 μm. The required pixel size depends on the gel size, and how small spots one wants to detect. The key point is to have a statistically sufficient number of pixels within the spot to be quantified. For most applications, an image resolution of about 100 μm (256 dpi) should be sufficient, resulting in a spot with a diameter of 1 mm having a 10-pixel diameter and consequently containing about 80 pixels. Several manufacturers provide their own file format output, which normally is a variant of the TIFF format. Ideally, the 2DE analysis software can read the manufacturer-specific file format directly, thus omitting the need to modify the 2DE image. If this is not the case, the scanned image might have to be exported as a standard TIFF file prior to analysis.
116
Levänen and Wheelock
Spot detection is one of the most important steps of 2DE gel analysis, as this is the basis for all the following work of spot matching and quantification. Due to the importance of spot detection, this algorithm is central in any chosen product. Two main classes of spot detection algorithms are used in modern 2DE analysis software: spot-based and pixel-based methods. As the name implies, the latter makes use of the existing pixels in the image when defining spot boundaries. In contrast, spot-based algorithms remodel the detected spot based on the assumption that the spot should be round, often by fitting to a Gaussian function (e.g., PDQuest). Although this model is a relatively good representation of a majority of the spots, the inflexibility of this model can cause problems when large deviations from the ideal round spot shape occur. Most image analysis programs unfortunately do not detect all spots correctly, and extensive manual review of automatic spot detection is necessary. Some common problems associated with inaccurate spot detection are shown in Fig. 2 including technical noise originating from protein visualization and image acquisition (speckles), irregularly shaped protein spots as a result of various distortions in the separation technique, and insufficient separation of spots with resulting overlapping spots or streaks (see Subheading 3.2). However, manual spot editing should be kept to a minimum as it introduces subjectivity into the analysis. As such, it is advisable to optimize the various settings available in the spot detection wizard (see Subheading 3.2, step 2) to achieve the best possible detection of true spots (true positives), and simultaneously exclude the majority of artifacts from being detected as spots (false positives). The performance of the various settings can be evaluated, e.g., through plotting of a free-response operator characteristics (FROC) curve (see Subheading 3.4). FROC curves can be used to attain a quantitative measure of spot detection and/or spot-matching performance (2, 3). It is defined as the relationship between the true positive fraction (TPF; number of correctly detected (or matched) spots/ total number of detected spots in the image) and the false positives. Accordingly, the ideal FROC values are TPF = 1 (all true spots successfully detected) and FP = 0 (no false spots detected) (see example in ref. 3). Matching refers to determining which spots represent the same protein species on different gels. This step is important in order to compare the resulting spot volumes between images. The image analysis program looks for similar features in the different gel images to identify corresponding spots and performs gel matching accordingly. There are two principal approaches also for automatic matching: the spot-based and the pixel-based method. As with spot detection, automatic spot matching generally leaves much to be desired, and extensive manual review is required. In most programs, the matching algorithm is supplemented
Troubleshooting Image Analysis in 2DE
117
Fig. 2. Gel images showing examples of encountered challenges associated with spot detection viewed in the image window (left ) and the 3D view (right ). (a) Spots are seen as round hill like tops, whereas spikes are formed like needles. (b) Horizontal streaking and incompletely focused spots appear as a ridge in the 3D viewer. (c) By using the 3D image window, it is easier to determine if a spot is in fact two separate spots or not.
with a warping algorithm, which deforms the image in order to counteract differences in the spot separation pattern that derive from technical variance. Warping can be performed in association with automatic matching, or manually through insertion of userdefined vectors or “seeds.” All gel images consist, to some extent, of background noise which appears as noisy irregular surfaces on flat areas, often arising from the staining or image acquisition process. The background does affect the quantification of a chosen protein spot and, thus, background subtraction is needed in order to obtain accurate measurements. Background subtraction can be local, considering only pixels in the vicinity of the spot when estimating the background, or global, considering all the pixels not defined as belonging to a spot in the image (4). The main purpose of normalization is to correct for experimental variance. The standard method available in most 2DE analysis programs involves dividing the individual spot values with the sum of all the spot values in the gel image, and thereby acquiring
118
Levänen and Wheelock
a relative quantity. In the case of DIGE experiments (see Chapters “High-Resolution 2DE” and “Two-Dimensional Difference Gel Electrophoresis”), ratiometric normalization is utilized. However, the algorithms required for ratiometric normalization are only available in upper-end 2DE analysis software, sometimes requiring purchase of an additional module. The majority of the variance observed in 2DE gels has been attributed to limitations in protein transfer during the method itself, e.g., incomplete absorption of the protein sample during rehydration and deficient transfer of sample from the first dimension to the second. However, recent studies have shown that the 2DE analysis software used for postexperimental analysis can introduce variance into the study (for review, see ref.4) and has been shown to cause as much as 30% of the total technical variance (5). The choice of algorithms, such as background subtraction or normalization, can affect the overall variance. Therefore, methods for evaluating the effect of the various background subtraction methods, etc. offered by 2DE analysis software have been included in Subheading 3.3. Data generated by 2DE can be analyzed statistically in a variety of ways, including significance thresholds, univariate and multivariate methods (for further reading, see refs.6, 7). The classical univariate tools, such as analysis of variance (ANOVA) and Student’s t test, compare evaluated changes in expression of individual protein spots reiteratively and are included in most 2DE analysis software. Univariate analyses are made under the assumption that the data are normally distributed, and it is important to verify that this is the case (Subheading 3.3). Furthermore, the low number of replicates and the high number of variables typical for proteomics studies introduce a high risk of generating type I error (false positives) in univariate analyses. In contrast, multivariate methods are adapted to the high number of variables in proteomics studies and examine all the available protein spot data simultaneously in order to identify global differences in protein expression between samples. Principal component analysis (PCA) is the most commonly available multivariate method in 2DE analysis software. A combination of both univariate and multivariate methods may aid identification of the truly significant biological alterations in protein expression. In this chapter, we provide a description of the general procedure of 2DE image analysis. Although Progenesis PG200 (Nonlinear Dynamics) has been used for illustrative purposes, the general protocol is applicable to other 2DE analysis software as well. As the quality of gel images used in the analysis is an important aspect in reducing the overall technical variance of the study, a protocol on image acquisition has also been included. The automatic image analysis performed by 2DE software is not fully reliable, and as considerable user intervention is required in order to create reliable results, the analysis becomes an arduous
Troubleshooting Image Analysis in 2DE
119
process. In addition, user intervention results in the process not being completely objective. Toward this end, protocols on how to evaluate the overall quality of the image analysis as well as the resulting quantitative data have been included in this chapter. Subheading 4 provides further advice on how to optimize and evaluate the quality of the analysis.
2. Materials The methods described in this chapter encompass solely the postelectrophoretic analysis of 2DE gel images, and thus assume that the 2DE gels to be analyzed have been separated and stained (see Chapters “Analysis of Bacterial Proteins by 2DE,” “Proteomic Analysis of Caenorhabditis elegans,” “Analysis of Proteins from Marine Molluscs,” “Preparation and Analysis of Plant and Plastid Proteomes by 2DE,” “High-Resolution 2DE,” and “Silver Staining of Proteins in 2DE Gel”). Required materials include an image acquisition device, 2DE analysis software, and a computer with adequate hardware capacity for the analysis (as specified by the respective software manufacturer). In addition to operating system, processor and RAM, it is worthwhile making sure the graphics card is adequate, as use of the 3D viewer is highly recommended. The specific equipment utilized in these protocols includes a Typhoon laser scanner run by Image Quant software (GE Healthcare, Uppsala, Sweden) and Progenesis PG200 2DE analysis software (Nonlinear Dynamics, Newcastle, UK). However, the protocols can easily be adjusted to other 2DE software as the graphical interface and the features provided by the various products by and large resemble each other. In terms of image acquisition equipment, it is recommended to use a 16-bit system with fluorescence capability in conjunction with a fluorescent stain for quantitative purposes. For the methods outlined in Subheadings 3.3 and 3.4, access to Microsoft Excel (Microsoft, Redmond, WA, USA) is required. In addition, the Analyse-it module for Excel (Analyse-it Software Ltd., Leeds, UK) is utilized for the distribution analyses performed in Subheading 3.3. Successful use of the protocols requires that the user has some experience with Excel and has performed the tutorials for PG200 provided by the manufacturer. The choice of software program plays an important role in the overall success of the 2DE proteome research, as the chosen product will have a large effect on the accuracy in the analysis and subsequent interpretation of the results obtained. The time spent by the computer analyzing the data, the amount of user intervention required, as well as the amount and choice of tools available, and appropriateness for one’s needs may also differ between
120
Levänen and Wheelock
Table 1 Commercially available 2D image analysis software Software
Company
Weblink
Progenesis
Nonlinear Dynamics
www.nonlinear.com
SameSpots
Nonlinear Dynamics
www.nonlinear.com
DeCyder
GE Healthcare
www.gelifesciences.com
ImageMaster
GE Healthcare
www.gelifesciences.com
Delta2D
Decodon
www.decodon.com
PDQuest
Bio-Rad
www.bio-rad.com
Proteomweaver
Bio-Rad
www.proteomweaver.com
Melanie
GeneBio
www.genebio.com
Gelfox
Alpha Innotech corporation
www.alphainnotech.com
products. As such, it is advisable to evaluate several programs prior to committing to a purchase. Table 1 provides a list of the most commonly used software products available on the market, as well as the manufacturers’ websites.
3. Methods 3.1. Image Acquisition
1. Fire up the lasers at least 10 min prior to scanning your gels. Insufficient warm-up of the lasers may result in quantitative differences of consecutive scans. 2. Ensure that the glass platen of the scanner is properly cleaned. Any stain residues or dust particles may result in speckling of your gel image. Since many postelectrophoretic stains contain hydrophobic residues, it may be more efficient to use a solvent such as ethanol or propanol rather than water for cleaning purposes. Make sure that the glass platen tolerates the solvent prior to use. 3. Place the gel on the glass platen of the scanner starting in one corner to avoid bubbles being trapped underneath the gel. If the scanner has a sealed glass platen, add some water along the edges of the gel to avoid that it dries out during scanning. 4. Select the excitation wavelength (laser) and emission filters that match the spectra of the fluorescent stain used for protein
Troubleshooting Image Analysis in 2DE
121
visualization. A band pass filter generally results in reduced background, while a long pass emission filter generally results in a stronger overall signal. 5. Perform a prescan at the lowest resolution (400–800 μm) in order to optimize the photomultiplier tube (PMT) setting to gain the maximal dynamic range of the pixel intensity. Evaluate the pixel intensities using the histogram tool by choosing Grey/Color Adjust under the View menu. Ideally, the spot intensities should be spread out over the entire available range (Fig. 3). You can estimate the effective dynamic range (see Subheading 1 for details) by moving the arrow below the histogram to the left edge of the peak, and the arrow above until the darkest spots are colored red. The range is displayed in numbers to the right. Please note that the pixels in the dye front tend to be much more intense than the spots, so the right edge of the histogram does not correspond to the highest abundance spot. 6. Scan the gel at desired resolution using the optimized PMT setting. For large format gels, a resolution of 100 μm is generally sufficient (see Subheading 1 for details). 3.2. Software-Assisted 2DE Image Analysis
1. Ensure that the file formats of your 2DE image analysis software and image acquisition equipment are compatible, i.e., that the 2DE analysis software is capable of reading the file format of your images. If this is not that case, it generally works to export the images as a TIFF file from the scanner software. Please make sure that the entire dynamic range of the pixel intensities are maintained after the export (see Subheading 3.1, step 5). 2. Most analysis programs provide a tool for automated spot detection. It is highly recommended to use the “analysis wizard” when such is available, as it tends to standardize the analysis process and thus minimize subjectivity. The wizard guides you through the stages of creating and analyzing the experiment. It allows you to create a name for your experiment, select your images to include in your experiment, set the area of interest, and to define the settings. As a rule of thumb, it is better to set any user-defined variables toward maximal sensitivity, as it generally introduces less variance to remove spots rather than adding them manually. Following are some pointers to the different steps in the wizard. • By default, the program includes the entire gel area in the analysis. However, the gel area of interest can be defined for each individual image through the wizard. This is of particular help when the gels have unusually streaky edges (see Notes 1 and 2).
122
Levänen and Wheelock
Fig. 3. View of the Grey/Color Adjust tool in the Image Quant software utilized to assess the dynamic range of pixel intensities following scanning with the Typhoon fluorescent scanner. At a first glance, the histogram appears to indicate that the majority of the available dynamic range is utilized (16 bits). However, the region to the right of the upper arrow is reflecting the pixel intensity of the IEF front, and not the spot intensities. Through adjusting the upper arrow until the darkest spot starts turning red, indicating saturation, the true dynamic range of about 42,000 is revealed.
Troubleshooting Image Analysis in 2DE
123
• In order to perform warping and matching, it is usually necessary to create a reference gel prior to start (see Notes 3 and 4). Most software packages build up a reference gel containing all the matched spots. PG200 allows you to choose an image to use as the reference. The image with the most abundant spots should in this case be chosen. This step is also incorporated in the spot detection wizard. • Spot filtering based on spot size, spot volume, etc. can be defined in the wizard in order to permanently remove populations of spots by criteria defined by the user. This can be useful for excluding exceptionally weak or small spots from the experiment. However, in PG200 filtering can be performed after the automatic analysis and can be ignored while setting up the experiment with the wizard. 3. Manually review the automatic spot detection and matching of all the spots in your experiment (see Notes 5–9). Most software products have some sort of tools or views to aid manual editing. In PG200, these include the image window, the 3D view, the comparison window, the measurement window, the histogram window, and the montage window. The visual layout can be set according to your own preference (see Fig. 4 for an example). In order to keep the manual spot editing as objective as possible, one should consider limiting the editing to the semiautomatic tools provided with the software, thereby withdrawing from manually drawing or adding complete spots. Semiautomatic spot editing tools for merge (Fig. 5), split, delete, or add new spots to the gels are available in most software, including PG200. 4. Spot editing (see Notes 10 and 11) usually leads to loss of background subtraction and normalization, so it is recommended to repeat these algorithms even if they were included in the automated analysis. Perform the background subtraction first, as this leads to a loss of normalization (see Note 12). There are several background subtraction methods, and it is up to the user’s preference which one to select. However, the background subtraction algorithms can have a significant effect on the overall variance in the experiment (3, 5), and it is advisable to compare the results of several options for each individual experiment (see Note 13). Moreover, it is important to use the same background subtraction method on all the gels in one experiment. In PG200, this is accomplished by choosing the “apply to all gels in experiment” in the pulldown menu when the background subtraction window is activated. Note that the background subtraction option may be disabled if the reference gel image is highlighted. Most software packages offer both local and global background
124
Levänen and Wheelock
Fig. 4. Example of a user-defined setup of selective windows in PG200 during manual spot editing.
subtraction algorithms (see Subheading 1 for details), and in some cases (e.g., in PG200) the option of omitting background subtraction altogether is also included. Although the latter is a good reference in terms of evaluating the variance introduced by the various algorithms (see Subheading 3.3), it is not recommended to use this option for the final analysis. Which background subtraction method that results in the best result depends greatly on the characteristics of gel image and the staining technique used, and has to be optimized on a case-by-case basis. 5. Normalization should always be calculated after finishing the spot editing and the background subtraction, in order to acquire the correct result. More sophisticated normalization methods, such as direct ratiometric normalization used in DIGE experiments (see Chapters “High-Resolution 2DE” and “Two-Dimensional Difference Gel Electrophoresis”), may not be available in more basic 2DE analysis software. Total spot volume normalization divides each individual spot volume with the sum of all the spot volumes detected in the
Troubleshooting Image Analysis in 2DE
125
Fig. 5. The Montage window in PG200 (similar to the group consensus tool in PDQuest) used for spot editing. In gel image C3, the software has mistakenly divided the protein spot into two different spot regions. This can be corrected through use of the semiautomatic “merge spots” tool.
image. Total spot ratio normalization is essentially the same, but by dividing by the ratios rather than total spot volumes, the magnitude of the spot volumes is maintained. Again, make sure that you are applying the normalization method to all gels in the experiment. 6. Most 2DE analysis software packages provide some sort of statistical tools available in the product itself, but these are often very basic and a complementary statistics option is recommended (Note 15). 3.3. Evaluating Variance and Data Distribution
1. Following completion of all the steps in Subheading 3.2, including careful manual review of spot detection and matching, export the normalized spot volumes of all the
126
Levänen and Wheelock
spots that are matched across all gels. It is important to include the majority of the spots, and not just a select few, “well-behaved” spots (see Note 14). In PG200, this is performed through opening the comparison window and checking the “normalized volume” in the selection guide in the upper left hand corner. Assure that all the gels of interest are selected, and proceed to “copy to excel” under the Edit menu. 2. In Excel, calculate the average spot volume, standard deviation, and finally coefficient of variance (CV) across replicate gels. 3. Sorting the data according to ascending average spot volume followed by plotting the CVs in a bar graph provides a visual tool to reveal any bias in variance related to the spot volume size. 4. Using the Analyse-it module for Excel, analyze the distribution of the resulting CVs as well as the average spot volumes using the “Summary (continuous)” tool under “Descriptive” in the “Analyse” menu. The resulting summary gives you both a visual aid for determining the distribution pattern of the spot volumes and their variance in terms of a histogram and a Q–Q plot, and more quantitative information in terms of a Shapiro– Wilks W test (8). If the distribution patterns are non-normal, transformation of the data should be considered. The most common transformation of 2DE data is logarithmic transformation, although some reports indicate that this may cause increased variance for low abundance spots (9). Some software programs offer automation of log transformation, and in some cases it may be the default setting (e.g., DeCyder). 5. Perform the same distribution analysis for the actual spot volumes of the same spot across replicate gels. This will give you information regarding the normality of the spot abundances in your experiment. 3.4. Evaluation of Spot Detection and Matching
1. Perform automatic spot detection and matching for the different settings or algorithms you want to compare as outlined in Subheading 3.2. 2. Prior to manual editing, count the number of correctly detected/ matched spots (true positives) and the number of incorrectly detected spots such as artifacts, speckles, etc. (false positives) for the different settings (see Note 14). 3. Calculate the total amount of spots in the gel, both detected (TP) and non-detected (false negatives (FN)), through visual inspection of your gels. Calculate the true positive fraction (TPF) = TP/(TP + FN).
Troubleshooting Image Analysis in 2DE
127
4. Plot the TPF against the number of FP for each of the settings to form a FROC curve. The optimal value will be located in the upper left of your graph (for practical example, see ref. 3).
4. Notes 1. Exclude the dye front and edges of the IPG strip from the image when assessing the dynamic range of pixel intensities, as these may give a false sense of a dynamic range. 2. Cropping the gel images to exclude unwanted areas such as dye front and other edges may improve the automatic matching ability of the software, as the alignment algorithms often are biased toward the positioning of the edges of the gel image. Be careful to use an imaging program that does not alter the quantitative aspects of the gel image when performing the cropping. 3. In most 2DE analysis software, the introduction of a few user-defined matches (landmarks) will greatly aid the automatic warping and alignment of the gel images. Although not required, it is thus a good idea to manually match at least one spot in each corner and one in the middle of the gel image prior to using the automatic matching tool. In PG200 it is necessary to check the option of “use manual matches as seeds” prior to matching to avoid that the manual matches are lost. 4. Factors that make spot recognition challenging are streaky gels, weak spots, regions with high spot density, overlapping spots, as well as artifacts such as droplets, dust, and possible breakage of the gel (see Fig. 2). There are no clear rules to the definition of a spot, what to consider as a spot and which ones to exclude. This adds the uncertainty of individual differences of opinion, and it is recommended that one person is responsible for all the analysis work within one experiment. 5. It is not always easy to decide if a spot is a peak or a speckle, a protein or not. The spots of proteins appear as peaks with curving sides, whereas dirt on the gel is seen as spikes more like needles (see Fig. 2a). It is advisable to use the 3D tool to provide clarity in dubious cases. 6. Some products have an option of filtering the images for speckles, termed speckle filters or salt-and-pepper filters. When using a software analysis program without this type of filter, it is worth considering the choice of acrylamide or stain, as these might affect the amount of speckles in the gel (5).
128
Levänen and Wheelock
7. Make sure that the spot outlines of matched spots are somewhat similar across the images in an experiment. The background subtraction algorithms are not perfect, and excessive differences in spot size outline may result in inaccurate spot volumes and increased variance (does not apply to PDQuest, as spot outlines are not visible). 8. The appearance of horizontal streaks and incompletely focused spots is common in 2DE gels (see Fig. 2b). Do not be tempted to include these poorly resolved spots in your quantitative analysis. Chances are that the streaking has resulted in colocation of multiple protein species, and the quantity detected by the 2DE software may not be representative of the true abundance in the protein sample. If proteins in streaky regions are of special interest, the separation technique has to be optimized. The 2DE analysis software cannot solve problems of poor separation. 9. The human eye is extremely good at pattern recognition – if you cannot discern a resemblance in the spot separation pattern by eye, chances are the 2DE analysis software will fail to do so as well. 10. To determine if a spot consists of one spot or two can be challenging during spot editing. In this case, it has proven helpful to take use of the 3D window (see Fig. 2c), as well as the montage window (see Fig. 5). Switching between these panels is a valuable aid in assessing the spot. 11. Excessive manual editing can lead to decreased objectivity. 12. Consider transforming your data if you notice a non-normal distribution pattern. 13. The methods for quantification of the variance described in Subheading 3.3 can also be used to evaluate the performance of various software programs (3). 14. Analyzing copies of the same gel image, cropped separately, as replicate images makes it possible to evaluate the reproducibility of the 2DE analysis software. As all the images are originally the same, you will evaluate only the variance arisen from the software analysis, excluding biological and technical variance (3). 15. Missing data are common in proteomic studies, due to mismatching or actual missing spots. Setting all the missing numbers to the value zero may be erroneous due to inaccuracies in the background subtraction. Copying the spot boundaries from the gel image where the spot exists to the one where the spot is lacking generally gives a more statistically viable solution.
Troubleshooting Image Analysis in 2DE
129
References 1. Miura, K. (2001) Imaging and detection technologies for image analysis in electrophoresis. Electrophoresis 22, 801–813. 2. Rogers, M., Graham, J., and Tonge, R.P. (2003) Using statistical image models for objective evaluation of spot detection in twodimensional gels. Proteomics 3, 879–886. 3. Wheelock, A.M. and Wheelock, C.E. (2008) Bioinformatics in gel-based proteomics. In Plant Proteomics: Technologies, Strategies and Applications. Series: Protocols in Molecular Biology, Chapter 8. G.K. Agrawal and R. Rakwal (Eds), Wiley, Hoboken, NJ, PP. 107– 125. 4. Wheelock, A.M. and Goto, S. (2006) Effects of post-electrophoretic analysis on variance in gel-based proteomics. Expert Rev. Proteomics 3, 129–142. 5. Wheelock, A.M. and Buckpitt, A.R. (2005) Software-induced variance in two-dimensional
6.
7.
8.
9.
gel electrophoresis image analysis. Electrophoresis 26, 4508–4520. Urfer, W., Grzegorczyk, M., and Jung, K. (2006) Statistics for proteomics: a review of tools for analyzing experimental data. Proteomics 6(Suppl 2), 48–55. Chich, J.F., David, O., Villers, F., Schaeffer, B., Lutomski, D., and Huet, S. (2007) Statistics for proteomics: experimental design and 2-DE differential analysis. J. Chromatogr. B Analyt. Technol. Biomed. Life Sci. 849, 261–272. Shapiro, S.S. and Wilk, M.B. (1965) An analysis of variance test for normality. Biometrika, 52, 591, Parts 3–4. Karp N.A., Griffin, J.L., and Lilley, K.S. (2005) Application of partial least squares discriminant analysis to two-dimensional difference gel studies in expression proteomics. Proteomics 5, 81–90.
Chapter 9 Analysis of Bacterial Proteins by 2DE Philip Cash and Evelyn Argo Summary Two-dimensional gel electrophoresis (2DE) is a key analytical method for investigating bacterial proteomes. The relatively simple genomes of many bacteria combined with only limited post-translational modifications of bacterial proteins mean that a significant proportion of the proteome is open to analysis by 2DE. The applications of 2DE in the field of microbiology are diverse and range from analysing physiological responses of the bacteria to environmental stress to investigating bacterial pathogenesis in human bacterial pathogens. The standard approach for 2DE in the analysis of bacterial proteins uses immobilised pH gradient (IPG) gels in the first dimension for charge separation and then an orthogonal separation, in the presence of SDS, to resolve the proteins according to their molecular mass. Protocols are presented in this chapter for small (7-cm-length IPG gel strips)- and medium (11- or 13-cm-length IPG strips)-format 2D gels using IPG gels and SDS-containing polyacrylamide slab gels for the second dimension. The application of the methods are demonstrated for the analysis of cell lysates prepared from Helicobacter pylori, although the same protocols have been used to analyse proteins from a variety of human bacterial pathogens. Key words: Bacteria, Helicobacter pylori, Proteins, Proteome, 2D Electrophoresis.
1. Introduction As in many fields of bioscience research, proteomics is a key approach for functional genomic studies in Microbiology with both gel- and non-gel based proteomic technologies being used for the analysis of bacterial proteomes (1–3). High-resolution 2-Dimensional Gel Electrophoresis (2DE) has been extensively used for the characterisation of bacterial protein synthesis, and the early papers by O’Farrell (4) documenting the development of 2-DE for the analysis of complex protein mixtures describe the proteins
David Sheehan and Raymond Tyther (eds.), Methods in Molecular Biology, Two-Dimensional Electrophoresis Protocols, vol. 519 © Humana Press, a part of Springer Science + Business Media, LLC 2009 DOI: 10.1007/978-1-59745-281-6_9
131
132
Cash and Argo
of the modern-day workhorse of microbiology, Escherichia coli. The relatively small coding capacity of many bacterial genomes and the limited extent to which post-translational events modify the bacterial proteins means that bacterial proteomes are of lower complexity than eukaryotic proteomes. Consequently, a significant proportion of the bacterial proteome is open to characterisation using 2DE provided care is taken in the preparation of the proteins prior to separation. The usefulness and reliability of the method for analysing bacterial proteomes has improved with the continuing technical developments that have benefited all applications of 2DE (5). Over the years 2-DE has been used for a variety of microbiological applications (reviewed in (6, 7)). Early applications used 2DE to compare different bacterial isolates to investigate bacterial epidemiology or to identify protein markers for pathogenesis. At this stage comparisons were confined to pattern matching, and only limited data were available on the identities of the proteins detected on the gel. In addition, the development of detailed proteomic databases was mainly restricted to wellcharacterised bacteria such as E. coli (8, 9). The advent of rapid protein identification methods, using peptide mass mapping and mass spectrometry (MS; see also Chapters “Applications of Chemical Tagging Approaches in Combination With 2DE and Mass Spectrometry”, “Phosphoproteome Analysis by In-Gel Isoelectric Focusing and Tandem Mass Spectrometry”, “Shotgun Protein Analysis by Liquid Chromatography-Tandem Mass Spectrometry”, “de novo Sequence Analysis of N-Terminal Sulfonated Peptides After In-Gel Guanidination”), as well as the increased amounts of genome sequence data for many human pathogens has opened the use of 2DE to a wide range of different bacterial pathogens. A number of proteomic databases (see Chapter “Creating 2DE Databases for the World Wide Web”), some of which are accessible via the Internet, based on the use of 2DE have now been developed for some bacterial pathogens (10). Current applications of 2DE in microbiology are diverse. Numerous investigators have used 2DE to monitor the response of the bacteria to environmental stress at the level of the proteome (e.g. 11–14). In a series of papers Hecker and colleagues reported on a detailed study of physiological responses in Bacillus subtilis at the level of the proteome, studies which have made extensive use of 2DE (for review see (2)). In the area of Medical Microbiology, 2DE has been used for comparative studies of bacterial pathogens in epidemiological studies as well as to identify potential pathogenic determinants (15, 16). More applied applications of 2DE have seen investigations of the mechanisms of antibiotic action and subsequent development of resistance (17–19). The protocols presented in the current chapter relate to the analysis of Helicobacter pylori soluble cellular proteins. H. pylori is a significant human pathogen which causes gastric
Analysis of Bacterial Proteins by 2DE
133
ulcers as well as being a predisposing factor in gastric cancer. A large body of data is available on the proteomics of H. pylori that describes the basic characterisation of total cellular and membrane proteins as well as the potential of proteomics to identify proteins usable for putative vaccines (20–23). Although the protocols presented later are applied to the analysis of H. pylori, they can be readily adapted to the characterisation of other Gram-negative and Gram-positive bacteria. Suggested modifications to the basic protocol are provided to expand the range of starting material open to analysis.
2. Materials 2.1. Bacterial Growth Media
1. Blood agar plates are prepared with Columbia Blood Agar Base (Oxoid, UK) as described by the manufacturer and supplemented with 5% sterile defibrinated horse blood. 2. Brucella broth (Sigma, UK) media is prepared according to the manufacturer’s instructions. The medium is supplemented with 5% foetal calf serum and 0.2% β-cyclodextrin (Sigma, UK). 3. Microaerophilic atmospheres (5% O2, 85% N2, 10% CO2) are generated using ‘Campygen. packs’ (Oxoid, UK).
2.2. Preparation of Bacterial Proteins
1. Phosphate-buffered saline (PBS): 0.17 M NaCl, 3 mM KCl, 2 mM NaH2PO4, 10 mM Na2HPO4, pH 7.4. 2. 2DE lysis buffer: 0.01 M Tris–HCl, pH 6.8, 1 mM EDTA, 8 M urea, 0.05 M DTT, 10% glycerol, 5% NP 40, 6% pH 3.5–10 carrier ampholytes (‘Resolyte’, BDH).
2.3. 1- and 2-Dimensional Gel Electrophoresis
1. 10% Acrylamide resolving gel for 1-dimensional gel electrophoresis (1DE): 6.67 mL acrylamide monomer solution (30% acrylamide:0.8% bis-acrylamide), 3.75 mL 2 M Tris–HCl, pH 8.8, 0.2 mL 10% SDS, 9.38 mL MilliQ water. The gel solution is polymerised with 100 μL 10% ammonium persulphate and 10 μL TEMED immediately before preparing the gels. 2. 2.6% Acrylamide Stacking gel for 1DE: 0.6 mL acrylamide monomer solution (30% acrylamide:0.8% bis-acrylamide), 1.2 mL 0.5 M Tris–HCl, pH 6.8, 0.05 mL 10% SDS, 3.13 mL milliQ water. The gel solution is polymerised with 50 μL 10% Ammonium persulphate and 5 μL TEMED immediately before preparing the gels. 3. 3× concentrate 1DE sample buffer: 2.0 mL 0.5 M Tris–HCl, pH 6.8, 1.7 mL 25% SDS, 1.6 mL 2-mercaptoethanol, 2.0
134
Cash and Argo
mL glycerol, bromophenol blue is added to provide a blue colour to the solution. 4. 1DE gels are cast and run in a Mini-Protean 3 electrophoresis kit (Bio-Rad Laboratories Ltd., Hemel Hempstead, UK). A PowerPac 3,000 power supply (Bio-Rad Ltd.) is used for electrophoresis. 5. Electrophoresis of the first-dimension IPG gels is carried out on a Multiphor II electrophoresis unit cooled using a MultiTemp water bath (GE Healthcare, Little Chalfont, UK). An EPS 3,500XL electrophoresis power supply (GE Healthcare) is used for the electrophoresis. 6. Immobilised pH Gradient (IPG) gel strips are obtained from either Bio-Rad Laboratories or GE Healthcare. 7. IPG re-swell buffer: 7 M urea, 2 M thiourea, 4% CHAPS, 0.3% DTT 1% ampholytes. For pH 5–8 L IPG gel strips the ampholytes are prepared by mixing pH 4–7 and pH 3–10 buffers (GE Healthcare) in a 3:1 ratio (see Note 1). 8. IPG equilibration buffer: 50 mM Tris–HCl, pH 8.8, 6 M urea, 30% glycerol, 2% SDS. 9. Second-dimension electrophoresis was carried out on SE250 slab gel units (GE Healthcare) cooled during electrophoresis with running tap water. 10. Second-dimension gel solutions. 10% acrylamide: 32 mL acrylamide monomer solution (30% acrylamide:0.8% bisacrylamide), 18 mL 2 M Tris–HCL, pH 8.8, 1 mL 10% SDS, 22.8 mL milliQ water, 22.8 mL 75% glycerol. The gel solution is polymerised with 100 μL 10% ammonium persulphate and 10 μL TEMED immediately before preparing the gels. 11. Gel overlay buffer: 0.5 M Tris–HCl, pH 8.8, 0.1% SDS. 12. SDS–PAGE electrophoresis running buffer: 0.4% Glycine, 0.05 M Trizma base, 0.1% SDS. 13. Bromophenol blue tracking dye prepared in 1% sucrose with sufficient dye to provide a blue colour. 2.4. Staining Using Colloidal Coomassie Brilliant Blue G250
1. Gel fixative solution: 50% ethanol, 2% orthophosphoric acid. 2. Gel equilibration solution for staining: 34% methanol, 17% ammonium sulphate, 2% orthophosphoric acid.
3. Methods
3.1. Sample Preparation
Throughout the preparation and analysis of bacterial proteins care must be taken to standardise the protocols. This optimisation starts with defining both the growth conditions (media and stage of
Analysis of Bacterial Proteins by 2DE
135
growth at which the bacterial cells are harvested) and the method of cell disruption. This standardisation minimises technical variation between replicate preparations of the bacterial proteins. 1. Grow H. pylori type strains and clinical isolates on blood agar plates incubated at 37°C for 3–4 days in gas jars (Oxoid Ltd) containing a microaerophilic atmosphere. Typically, two blood agar plates showing a confluent growth of bacteria provide sufficient material for subsequent analysis. 2. For broth-grown bacteria, inoculate 15 mL of media with an H. pylori suspension, prepared from bacteria grown on agar plates, and incubate stationary at 37°C for 3–4 days in gas jars containing a microaerophilic atmosphere. 3. Scrape the colonies from the surface of the agar plates using a sterile plastic plating loop and emulsify in 1 mL PBS. For bacteria grown in broth culture, recover the bacterial cells from the growth medium by centrifugation at 1,500 × g for 5 min on a Juan BR3.11 centrifuge and resuspend in 1 mL of PBS. Subsequent processing is the same irrespective of the method of growing the bacteria. 4. Pellet the bacteria by centrifugation at 6,000 × g for 5 min from the PBS and resuspend in 1 mL of PBS. At this point an estimate can be made on the amount of lysis buffer required to disrupt the bacteria cells. Prepare a 1:50 dilution of the cell suspension in PBS and determine the OD605 of this dilution. The final bacterial pellet generated in step 5 is resuspended using 0.1 mL of 2DE lysis buffer for each 0.1 OD605 unit of the diluted sample. Although this is empirically determined, it is suitable for a number of bacteria including H. pylori, Haemophilus influenzae, E. coli, and Streptococcus pneumoniae. 5. Pellet the bacterial cells from the original suspension and remove all the PBS supernatant. Resuspend the bacterial pellet in the volume of 2DE lysis buffer determined earlier. Leave the suspension on ice for 5 min and remove insoluble debris by centrifugation at 11,500 × g for 5 min. Store the supernatant at −70°C until required (see Note 2). The protocol presented earlier produces a reproducible representation of the abundant soluble cytoplasmic bacterial proteins (see Note 3). 3.2. 2 -Dimensional Gel Electrophoresis
The 2DE protocol described later uses 7 cm IPG strips for the first dimension and 7 × 8 cm slab gels for the second dimension on which to analyse the bacterial proteins. This gel format is suitable for many studies of microbial proteomes using 2DE. A representative 2DE protein profile for H. pylori cellular proteins analysed by this method is shown in Fig. 1.
136
Cash and Argo
Fig. 1. Representative profile of the Helicobacter pylori (strain 26,695) cell proteins prepared and analysed on smallformat 2DE gels as described in the text. The pH range is taken from the manufacturer’s information and the molecular weight scale is determined by the co-electrophoresis of known molecular weight standard proteins with the bacterial cellular proteins on a parallel gel.
3.2.1. Estimating the Protein Load for 2DE
To aid comparison of 2DE profiles of similar complexity, the gels ought to be loaded with equivalent amounts of total protein. Commercial protein assay kits are available to measure protein concentration in the presence of the urea and detergent included in the 2DE lysis buffer. Although requiring confirmation for each sample type, 75–100 μg of protein is loaded on the 7-cm IPG gels to produce a clear profile when stained using colloidal Coomassie blue G250. An alternative empirical approach using 1DE may be used to estimate the required load based on the intensity of the gel staining as follows: 1. Prepare a slab gel in the Bio-Rad Protean 3 electrophoresis unit which is assembled according to the manufacturer’s instructions. A 10% polyacrylamide resolving gel is sufficient for these analyses. During polymerisation (1 h at room temperature) overlay the top of the resolving slab gel with water-saturated butanol to provide an even surface. Wash traces of butanol off before adding the 2.6% polyacrylamide stacking gel. Add a plastic comb to the stacking gel to form the gel tracks. Assemble the slab gel into the lower tank and add the SDS–PAGE electrophoresis running buffer to both the upper and lower electrode reservoirs.
Analysis of Bacterial Proteins by 2DE
137
2. Mix the bacterial protein lysate with the 3× concentrate 1DE sample buffer in a 2:1 ratio and heat at 100°C for 5 min. Load 3 μL of the treated protein sample to the tracks on the prepared slab gel. 3. Carry out the electrophoresis out at 200 V (constant voltage) for 55 min. Place the electrophoresis unit in an ice bath to aid cooling of the gel during the electrophoresis. 4. Stain the gel using colloidal Coomassie brilliant blue G250 as described later. 5. The volume of the original protein lysate required for 2DE is estimated from the staining intensity of the sample track. 6. A simple calibration gel series is generated by loading a range of volumes of the cell lysate to the 2DE gels and determining the optimal loading. Once established this protocol is relatively robust for many different protein samples. 3.2.2. First Dimension
1. Adjust the required volume of the bacterial cell lysate to a final volume of 140 μL with IPG re-swell buffer, leave for 10 min at room temperature, and then clarify at 11,000 × g for 5 min, and use 125 μL of the supernatant to re-hydrate the IPG strip overlaid with Dry Strip Cover Fluid (GE Healthcare). Rehydrate the IPG gel strip for 18 h at room temperature in an IPG Reswelling Cassette (GE Healthcare). For samples prepared using the 2DE lysis buffer a maximum volume of 30 μL the bacterial cell lysate can be mixed with the IPG re-swell buffer and loaded to each IPG gel as described earlier; using larger volumes of the cell lysate can lead to artefacts in the protein separation. There is no such limitation to the volume used when the bacteria are lysed directly in IPG re-swell buffer (see Note 4). 2. Assemble the IPG gel strips in the Multiphor II flat-bed electrophoresis system according to the manufacturer’s instructions. Isoelectric focusing (IEF) is performed in three stages with a ramped voltage change between each step, 200 V for 1 min; 3,500 V for 90 min; 3,500 V for 180 min (all stages at 2 mA and 5 W). The ceramic support plate of the Multiphor II apparatus is cooled to 20°C using a MultiTemp water bath. 3. Following electrophoresis, the IPG gel strips are equilibrated in IPG equilibration buffer containing 1% DTT for 30 min followed by a further 30-min incubation in IPG equilibration buffer containing 2.5% iodoacetamide. Each equilibration is carried out in 7 mL of the appropriate buffer and the gels are agitated throughout the equilibration process.
3.2.3. Second Dimension
The following protocol for the 2nd -dimension electrophoresis uses the SE250 electrophoresis units (GE Healthcare) (see Note 5).
138
Cash and Argo
1. Separate the proteins by molecular weight on 10% single concentration polyacrylamide slab gels prepared as batches of nine gels. Individual gel sandwiches for the SE250 gel apparatus are made with a plain glass and a notched aluminium plate separated by plastic spacers of 1 mm thickness. Assemble the gel sandwiches in a multi-gel-casting unit (GE Healthcare) separated from each other by Parafilm strips. Fill in excess space in the casting unit using a combination of glass plates and plastic sheets. 2. Add the acrylamide solution to the casting unit from the top until its level is approximately 5 mm from the top of the aluminium notched plate. Overlay the surfaces of the individual gels with 1 mL gel overlay buffer and allow the gels to polymerise overnight at room temperature after which separate the individual gel sandwiches (see Note 6). 3. Drain the equilibrated IPG gels of the final IPG equilibration buffer for 10 min before dipping into clean milliQ water to wash the gel surface. This process removes traces of iodoacetamide, which can, on occasions, interfere with the seconddimension electrophoresis. Immediately place the washed IPG strips on the surface of the second-dimension slab gels with the plastic backing of the IPG gel strip against the glass plate of the gel sandwich. 4. Assemble the second-dimension slab gels into the individual SE250 electrophoresis units and fill the upper and lower buffer chambers with the SDS–PAGE electrophoresis running buffer. Add 5 μL of bromophenol blue tracking dye at one end of the IPG gel strip to monitor the progress of the electrophoresis. 5. Carry out the electrophoresis at 75 V (constant voltage) for 60 min followed by 150 V (constant voltage) for a further 160 min. Cool the gels with running tap water during the electrophoresis. 3.2.4. Colloidal Coomassie Brilliant Blue G250 Staining
Gels are stained in plastic boxes in batches of up to ten gels. The gels are agitated throughout the staining procedure. 1. Following electrophoresis transfer the slab gels to gel fixative solution and leave on a shaker overnight. The volume of the fixative solution is sufficient to allow the gels to move freely during this step. 2. Wash the gels three times for 30 min each wash in an excess of MilliQ water. 3. Add gel equilibration solution to the gels using 20 mL of solution for each small-format gel in the batch being stained. Incubate for 1 h at room temperature. For larger gels, the amount of gel equilibration solution used is increased in proportion to the slab gel volume.
Analysis of Bacterial Proteins by 2DE
139
4. Add 0.2 g per 200 mL of Coomassie Brilliant Blue G250 powder directly to the solution containing the gels and stain the gels for up to 4 days. Protein spots are normally visible after a few hours, but optimal staining is achieved after 4 days. 5. Once the required staining time is complete, wash the gels 2–3 times with MilliQ water to remove traces of stain particles and ammonium sulphate and then soak for 1 h in 0.5% glycerol. Dry the gels between cellophane sheets using a heated drier (Easy Breeze, GE Healthcare). 3.3. PostElectrophoretic Analysis of the Separated Proteins
Post-electrophoretic analysis of the proteins resolved using either the small- or medium-format-sized gels described in this chapter follows the majority of the well-documented protocols available in the literature. With bacterial systems, particularly the analysis of type strains with complete genome sequences, a significant proportion of the proteins detected by Coomassie blue staining are amenable to identification by peptide mass mapping combined with either MALDI-ToF MS or LC-MS/MS for analysing the peptide masses (24, 25). For gels stained using colloidal Coomassie blue G250 as described earlier, quantitative data can be taken from the stained gels (see Chapter “Troubleshooting Image Analysis in 2DE”). Stained gels are scanned using a Molecular Dynamics Personal Densitometer at 50 μm resolution to generate 12-bit images, which are transferred to the Progenesis 2D analytical software suite from Nonlinear Dynamics (Newcastle, UK). A representative gel image is selected to serve as the reference image, and the remaining gels within the data set are warped sequentially to the reference image using the in-built routines available in the Progenesis software. The semi-automated routines available in the Progenesis software are used to detect and edit the spot boundaries on the reference image before transferring these boundaries from the reference image down through the other gel images forming the data set. This step utilises the ‘SameSpots’ routine available in the Progenesis software. The transfer of the equivalent spots boundaries simplifies the otherwise labourintensive procedures of spot detection and gel matching associated with previous versions of this software suite. The workflow for gel analysis summarised earlier also opens a route for simplifying the comparison of different bacterial isolates. Even isolates of the same bacterial species can show relatively high levels of protein variation that make reliable comparisons of single stained gels difficult and unreliable. Although difference gel electrophoresis (DIGE; see Chapter “Two-Dimensional Difference Gel Electrophoresis”) with fluorescently labelled proteins can be used for comparisons, many epidemiological studies take place over many months or years when it is not possible to
140
Cash and Argo
produce the pooled reference sample typically required in DIGE experiments. The following comparative strategy can be used for the gel analyses under these conditions. Select a single type isolate for the study against which all other isolates are compared. For each new bacterial isolate analysed process a co-electrophoresis 2DE gel in parallel with the single sample gels for the type isolate and new bacterial isolate. The co-electrophoresis gel contains sufficient sample from each bacterial isolate so that the major proteins for each are detectable (Fig. 2). The Progenesis software is used to warp the single sample gels to the co-electrophoresis gel by matching the respective major proteins present in the profiles.
Fig. 2. Comparison of the H. pylori and H. bilis proteomes by 2DE. Protein lysates for each bacteria were prepared and analysed either singly (a) and (c) or by co-electrophoresis (b) by 2DE as described in the text. Only a small portion of the total 2DE protein profile is shown; this covers the area enclosed by the solid rectangle in Fig. 1.
Analysis of Bacterial Proteins by 2DE
141
Once warping and spot detection is complete then the protein profiles on the single sample gels can be compared directly. As a refinement to this simplistic comparison a representative 2DE protein profile of the standard type isolate can be selected as the basis for the entire study. The co-electrophoresis gel profile is warped to this protein pattern before the single sample gels are warped into the co-electrophoresis gel. The inclusion of this added level of warping means that protein profiles analysed in different gel runs processed weeks or months apart may be compared as they have all been warped into the equivalent geometric space. Although warping can overcome many of the minor technical variations inherent with 2DE, it does not take the place of processing the gels under consistent and reproducible conditions.
4. Notes 1. Unless stated otherwise all electrophoresis solutions use highpurity or electrophoresis-grade reagents and milliQ water is used throughout. 2. The described protocol is suitable for many Gram-negative bacteria where the cell wall is relatively thin and the cells can be lysed directly using non-ionic detergents. In some cases improved solubilisation can be achieved by disrupting the bacterial cells directly in the IPG re-swell buffer rather than the 2DE lysis buffer. Gram-positive bacteria have a thicker cell wall and more rigorous protocols are required to break the cells. A number of investigators have described the use of an enzymatic digestion of the cross-linking molecules in order to weaken the cell wall. As a more general alternative, which is also suitable for small samples, the bacterial cells are broken by sonication with a micro-probe (Labplant Ultrason 250). The bacteria are disrupted directly in the 2DE lysis buffer using 6 × 30 s pulses of sonication. The tube is held on ice during the sonication, and there is a 30-s pause between each pulse to minimise warming the sample. To reduce frothing of the lysis buffer the sonication is carried out in a minimum volume of 250 μL of 2DE lysis buffer. Sonication has been used successfully to prepare cellular proteins from Streptococci, Lactococci, and B. subtilis. 3. Additional modifications to the sample preparation protocol can be used to enrich sub-fractions of the bacterial cell to analyse specific protein classes (for example hydrophobic proteins (26, 27) or low-abundance proteins (24) (see also
142
Cash and Argo
Chapter “Difficult Proteins”). A number of commercial sample preparation kits are available to prepare these subcellular fractions. 4. For each type of sample to be analysed, it is advisable to optimise the pH range covered by the IPG gel strip used for the first-dimension electrophoresis. Preliminary trial analyses can be made on IPG gel strips with the following ranges 3–6 L, 4–7 L, 5–8 L, 3–10 L, and 3–10NL. These standard ranges are available from many commercial suppliers. This approach was used for H. pylori proteins, and the pH5–8 L IPG gel strips presented here were adopted to provide the coverage of a significant proportion of the proteins encoded by the H. pylori genome (25). With the commercial availability of IPG gel strips covering the pH range 6–11 there is increasing interest in the analysis of basic proteins. The protocol presented in this chapter has not been specifically optimised for the analysis of basic proteins. 5. Although the small-format gel system presented here is eminently suitable for many bacterial samples, scaling to large gel formats where sufficient sample is available can achieve improved resolution. A simple scale-up using either 11-cm or 13-cm IPG gel strips (Bio-Rad Ltd and GE Healthcare, respectively) over similar pH ranges described here can be achieved with minimal additional equipment or modification to the protocol. Normally, up to 250 μg of protein is loaded to these IPG strips for analytical gels stained as described in this chapter. For these larger gels 200 μL and 250 μL of the protein suspension are used to rehydrate the 11-cm and 13-cm strips, respectively. Suitable run conditions on the Multiphor II apparatus for the first dimension are 300 V for 1 min; 3,500 V for 90 min; and 3,500 V for 5 h (all stages at 2 mA and 5 W) with ramped voltage changes between each step. The final step in the electrophoresis can be extended to improve the focusing of the proteins under some conditions. The seconddimension slab gels are run using Protean II gel electrophoresis units (Bio-Rad Ltd) at 100 V (constant voltage) for 60 min and then 15 mA per gel for a total of 400 min. The slab gels are prepared using a multi-gel-casting unit (Bio-Rad) as for the small-format gels in order to improve gel-to-gel reproducibility. Comparative 2DE protein profiles for H. pylori proteins analysed on the small- and medium-format 2DE gels are shown in Fig. 3. 6. It is essential for the upper surface of the slab gel to be smooth and straight in order to achieve close contact between the IPG strip and the slab gel itself. Slight imperfections can be overcome by adding a small amount of 1% agarose, prepared in 0.5 M Tris– HCl, pH 8.8, 0.1% SDS, to provide a straight surface.
Analysis of Bacterial Proteins by 2DE
143
Fig. 3. Comparison of the 2DE protein profiles of H. pylori cellular proteins analysed on (a) small-and (b) medium-format gels. H. pylori cellular proteins were prepared and analysed on either small- or medium-format 2DE gels. The first-dimension range for both formats was pH 5–8, whilst the second-dimension gels were either 10% or 12% polyacrylamide slab gels for the small- and medium-format gels, respectively. The panels show the same relative region of the profile (indicated by the dashed box outline in Fig. 1), but the magnification of the images has been adjusted for clarity of the comparison in the figure.
Acknowledgements Research supporting the development of the protocols described earlier was supported, in part, by grants from Grampian University Hospitals Trust and The Cunningham Trust.
References 1.
2.
3.
4.
Giometti, C.S., Khare, T., Tollaksen, S.L., Tsapin, A., Zhu, W., Yates, J.R., III, and Nealson, K.H. (2003) Analysis of the Shewanella oneidensis proteome by two-dimensional gel electrophoresis under nondenaturing conditions. Proteomics 3, 777–785 Hecker, M., and Volker, U. (2004) Towards a comprehensive understanding of Bacillus subtilis cell physiology by physiological proteomics. Proteomics 4, 3727–3750 VerBerkmoes, N.C., Hervey, W.J., Shah, M., Land, M., Hauser, L., Larimer, F.W., et al. (2005) Evaluation of “shotgun” proteomics for identification of biological threat agents in complex environmental matrixes: experimental simulations. Anal. Chem. 77, 923–932 O’Farrell, P.H. (1975) High resolution twodimensional electrophoresis of proteins. J. Biol. Chem. 250, 4007–4021
5.
6.
7.
8.
9.
Gorg, A., Weiss, W., and Dunn, M.J. (2004) Current two-dimensional electrophoresis technology for proteomics. Proteomics 4, 3665–3685 Cash, P. (1998) Characterisation of bacterial proteomes by two-dimensional electrophoresis. Anal. Chim. Acta 372, 121–146 Cash, P. (2003) Proteomics of bacterial pathogens. Adv. Biochem. Eng. Biotechnol. 83, 93– 115 VanBogelen, R.A., Sankar, P., Clark, R.L., Bogan, J.A., and Neidhardt, F.C. (1992) The gene-protein database of Escherichia coli: edition 5. Electrophoresis 13, 1014–1054 VanBogelen, R.A., Abshire, K.Z., Pertsemlidis, A., Clark, R.L., and Neidhardt, F.C. (1996) Gene-protein database of Escherichia coli K-12, edition 6, in Escherichia coli and Salmonella: Cellular and Molecular Biology (Neidhardt, F.C., Curtiss, R., Ingraham, J.L., Lin, E.C.C.,
144
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
Cash and Argo Low, K.B., Magasanik, B., et al. eds.), ASM Press, Washington, D.C. pp. 2067–2117 Pleissner, K.P., Eifert, T., Buettner, S., Schmidt, F., Boehme, M., Meyer, T.F., et al. (2004) Web-accessible proteome databases for microbial research. Proteomics 4, 1305–1313 Cash, P., Argo, E., and Bruce, K.D. (1995) Characterisation of Haemophilus influenzae proteins by 2-dimensional gel electrophoresis. Electrophoresis 16, 135–148 Giard, J.C., Laplace, J.M., Rince, A., Pichereau, V., Benachour, A., Leboeuf, C., et al. (2001) The stress proteome of Enterococcus faecalis. Electrophoresis 22, 2947–2954 Kan, B., Habibi, H., Schmid, M., Liang, W., Wang, R., Wang, D., and Jungblut, P.R. (2004) Proteome comparison of Vibrio cholerae cultured in aerobic and anaerobic conditions. Proteomics 4, 3061–3067 Benndorf, D., Thiersch, M., Loffhagen, N., Kunath, C., and Harms, H. (2006) Pseudomonas putida KT2440 responds specifically to chlorophenoxy herbicides and their initial metabolites. Proteomics 6, 3319–3329 Shaw, A.C., Gevaert, K., Demol, H., Hoorelbeke, B., Vandekerckhove, J., Larsen, M.R., et al. (2002) Comparative proteome analysis of Chlamydia trachomatis serovar A, D and L2. Proteomics 2, 164–186 Lee, A.Y., Park, S.G., Jang, M., Cho, S., Myung, P.K., Kim, Y.R., et al. (2006) Proteomic analysis of pathogenic bacterium Vibrio vulnificus. Proteomics 6, 1283–1289 Cash, P., Argo, E., Ford, L., Lawrie, L., and McKenzie, H. (1999) A proteomic analysis of erythromycin resistance in Streptococcus pneumoniae. Electrophoresis 20, 2259–2268 McAtee, C.P., Hoffman, P.S., and Berg, D.E. (2001) Identification of differentially regulated proteins in metronidozole resistant Helicobacter pylori by proteome techniques. Proteomics 1, 516–521 Peng, X., Xu, C., Ren, H., Lin, X., Wu, L., and Wang, S. (2005) Proteomic analysis of
20.
21.
22.
23.
24.
25.
26.
27.
the sarcosine-insoluble outer membrane fraction of Pseudomonas aeruginosa responding to ampicillin, kanamycin, and tetracycline resistance. J. Proteome Res. 4, 2257–2265 McAtee, C.P., Lim, M.Y., Fung, K., Velligan, M., Fry, K., Chow, T.P., and Berg, D.E. (1998) Characterization of a Helicobacter pylori vaccine candidate by proteome techniques. J. Chromatogr. B. Biomed. Sci. Appl. 714, 325–333 Bumann, D., Meyer, T.F., and Jungblut, P.R. (2001) Proteome analysis of the common human pathogen Helicobacter pylori. Proteomics 1, 473–479 Bumann, D., Aksu, S., Wendland, M., Janek, K., Zimny-Arndt, U., Sabarth, N., et al. (2002) Proteome analysis of secreted proteins of the gastric pathogen Helicobacter pylori. Infection & Immunity 70, 3396–3403 Bae, S.H., Harris, A.G., Hains, P.G., Chen, H., Garfin, D.E., Hazell, S.L., et al. (2003) Strategies for the enrichment and identification of basic proteins in proteome projects. Proteomics 3, 569–579 Birch, R.M., O’Byrne, C., Booth, I.R., and Cash, P. (2003) Enrichment of Escherichia coli proteins by column chromatography on reactive dye columns. Proteomics 3, 764–776 Uwins, C., Deitrich, C., Argo, E., Stewart, E., Davidson, I., Cash, P., et al. (2006) Growth-induced changes in the proteome of Helicobacter pylori. Electrophoresis 27, 1136–1146 Molloy, M.P., Herbert, B.R., Slade, M.B., Rabilloud, T., Nouwens, A.S., Williams, K.L., and Gooley, A.A. (2000) Proteomic analysis of the Escherichia coli outer membrane. Eur. J. Biochem. 267, 2871–2881 Nouwens, A.S., Cordwell, S.J., Larsen, M.R., Molloy, M.P., Gillings, M., Willcox, M.D., and Walsh, B.J. (2000) Complementing genomics with proteomics: the membrane subproteome of Pseudomonas aeruginosa PAO1. Electrophoresis 21, 3797–3809
Chapter 10 Proteomic Analysis of Caenorhabditis elegans Pan-Young Jeong, Keun Na, Mi-Jeong Jeong, David Chitwood, Yhong-Hee Shim, and Young-Ki Paik Summary Proteomic studies of the free-living nematode Caenorhabditis elegans have recently received great attention because this animal model is a useful platform for the in vivo study of various biological problems relevant to human disease. In general, proteomic analysis is carried out in order to address a specific question with respect to differential changes in proteome expression under certain perturbed conditions. In this chapter, we focus on gel-based proteomic analysis of C. elegans subjected to two specific stress conditions during development: induction of the dauer state for whole body protein expression and a temperature shift for egg protein expression. Utilizing these differently perturbed C. elegans protein samples, two-dimensional electrophoresis and differential in-gel electrophoresis methods have led to the discovery of remarkable aspects of the worm’s biology. We also provide numerous details about the technical points and protocols necessary for successful experimentation. Key words: 2DE, Caenorhabditis elegans, Dauer larva, Azacoprostane, Egg proteome, MALDITOF, DIGE.
1. Introduction Since C. elegans was established as an animal model organism in 1965 for the study of various biological problems, genomic and proteomic analyses have been used with increasing frequency to discover major aspects of its biology relevant to human disease. This is because C. elegans has many unique features that make it an ideal model organism, such as a convenient inbreeding reproductive mode (i.e., hermaphroditic), a short life span (2~3 weeks), a large brood size (>300 progeny), a variety of readily produced genetic mutants, and ease of cultivation in the laboratory (1). David Sheehan and Raymond Tyther (eds.), Methods in Molecular Biology, Two-Dimensional Electrophoresis Protocols, vol. 519 © Humana Press, a part of Springer Science + Business Media, LLC 2009 DOI: 10.1007/978-1-59745-281-6_10
145
146
Jeong et al.
After the genome of C. elegans had been completely sequenced, proteomics became an essential experimental strategy in analyzing global gene expression in C. elegans under particular physiological conditions (2). Large-scale analysis of all possible proteins expressed under specific physiological conditions, termed expression proteomics, is usually a good starting point for the study of the C. elegans proteome, through which one can profile those proteins of interest in a high-throughput manner. Among several proteomics platforms, we describe the most common gel-based methods such as two-dimensional electrophoresis (2DE) and differential gel electrophoresis (DIGE) for proteomic analysis of C. elegans obtained under two different physiological conditions (Fig. 1). Firstly, we describe analysis of whole body proteins isolated from normal mixed-stage worms and dauer larvae. When nutritional and environmental conditions are adequate for growth, C. elegans develops rapidly from the embryo, through four larval stages
Proteome Analysis of C. elegans (N2 wild type)
Whole Body Proteome
Egg Proteome
Dauer Effect Control ( ) vs. Dauer Larva ( )
Temperature Effect 20 °C N2 Egg vs. 25 °C N2 Egg
2-DE
2-D DIGE
Spot Analysis
Spot Analysis
MALDI-TOF-MS
MALDI-TOF-MS
Mapping
Mapping
Analysis (e.g. Actin protein)
Analysis (e.g. Stress protein)
Fig. 1. Overall strategy for 2DE analysis of C. elegans protein.
Proteomic Analysis of Caenorhabditis elegans
147
(L1–L4), to reproductive adults. In contrast, during unfavorable conditions, C. elegans arrests prior to reproductive maturity, either as an L1 larva or as a specialized L3 diapause form termed the dauer larva (3). In this chapter, using 2DE, we describe identification of proteins differentially expressed in dauer larvae compared to normal, mixed-stage worms. Secondly, we describe a DIGE method for the proteomic analysis of egg proteins prepared from adult C. elegans grown at either 20°C or 25°C. The DIGE methods were highly useful in quantifying egg proteins differentially expressed upon temperature shift, an essential part of studies of the effect of azacoprostane on C. elegans development. Thus, this chapter is intended to provide the reader with the necessary information for systematic analysis of the C. elegans proteome using 2DE or DIGE. We highlight our use of these methods to discover biomarkers involved in dauer formation and the worm’s adaptation to stress caused by temperature shift. 1.1. Analysis of Whole Body Proteome of C. elegans: Mixed-Stage Worms Versus Dauer Larvae
To explore if there is a substantial change in protein expression between dauer larvae and mixed-stage worms, 2DE proteomic analysis was performed. Figure 2a is a typical 2DE gel image showing separation of proteins from mixed-stage normal worms at pH 3–10. Because protein spots in the acidic and alkaline pH areas a pH 3
pH 10 kDa
97 66 45 30
20
Fig. 2. 2DE gel pattern of protein extracted from mixed-stage worms and dauer larvae of C. elegans. 2DE gel image of proteins from mixed-stage worms at (a) pH 3–10, (b) from dauer larvae at pH 3–6, and (c) from dauer larva at pH 5–8. Proteins extracted from wild-type worms were separated on a nonlinear IPG strip, followed by a 9–16% SDS–polyacrylamide gel. The gel was stained with CBB.
148
Jeong et al.
b pH 3
pH 6 kDa
97 66 45
30
20
c pH 5
pH 8 kDa
97 66 45 30
20
Fig. 2. (continued)
were not well resolved, whole extracts of mixed-stage worms and dauer larvae were separated on two different immobilized pH gradient (IPG) strips (i.e., pH 3–6 and pH 5–8) followed by 9–17% SDS–PAGE. Aligning spots by image analysis, we directly
Proteomic Analysis of Caenorhabditis elegans
149
compared gels representing mixed-stage worms versus dauer larvae (n = 3) and analyzed the differentially expressed protein spots in the dauers. The 2DE gel patterns of dauer larvae at pH 3–6 (Fig. 2b) and pH 5–8 (Fig. 2c) are shown. Spots in the 2DE gel were then isolated, digested with trypsin, treated with mixed POROS R2 and R3 resins, and analyzed by MALDI-TOF-MS or MALDITOF-MS/MS (Subheadings 2 and 3). Proteins were identified by mass fingerprinting of the selected peptide peaks by applying low tolerance (<20 ppm) with recalibration. The location of the corresponding peak for each protein and its mass and pI were also confirmed (Table 1). We focused specifically on protein spots that showed >2.0-fold (n = 3) intensity in dauers compared to mixedstage worms, or proteins that were detected only in dauer larvae. These proteins were grouped into three functional categories: (1) oxidative stress-defense related proteins, (2) muscle proteins, and (3) energy generation and other proteins (Table 1).
Table 1 Proteins that were more abundant in dauer larvae of C. elegans than in mixed-stage worms
ID numbera Name of protein
Increase fold (n = 3) in Accession Molecular Coverage Matched dauer larvab number mass (Da)/pI (%) peak
Stress-defense related proteins 1
Heat shock protein HSP-12.6
M
17541102
12621/5.6
64
5
2
Thioredoxin family member
1.387
17539056
16965/5.0
31
4
3
Superoxide dismutase [Cu–Zn]
1.302
464769
16237/6.1
52
6
4
Glutathione S-Transferase, P subunit (gst-1)
2.076
1753834
23902/5.9
29
6
5
Glutathione S-transferase GST-7
1.155
17534685
23086/6.3
25
8
Muscle structural proteins 6
Myosin light chain (mic-3)
1.601
25151365
17145/4.6
70
12
7
Myosin light chain (mLc-2)
1.724
17569077
18603/5.1
67
9 (continued)
150
Jeong et al.
Table 1 (continued)
ID numbera Name of protein
Increase fold (n = 3) in Accession Molecular Coverage Matched dauer larvab number mass (Da)/pI (%) peak
8
Levamisole resistant-11, 2.07 tropomyosin isoform (lev-11)
32563577
32937/4.7
37
10
9
Levamisole resistant-11, 1.439 tropomyosin isoform (lev-11)
25151024
29632/4.7
47
15
10
Levamisole resistant-11, 3.629 tropomyosin isoform (lev-11)
32563577
32937/4.7
44
14
11
CeTMI
M
1208409
32952/4.7
42
12
12
CeTMI
2.471
1208409
32952/4.7
33
10
13
Troponin C (pat-10)
2.44
17507581
18517/4.2
44
5
14
Troponin T TNT-2
4.987
17568063
53810/4.2
68
9
15
Actin (act-3)
4.492
6628
41709/5.3
28
8
16
Actin (act-3)
2.889
6628
41709/5.3
35
10
17
Actin (act-3)
9.424
6628
41709/5.3
27
7
18
Actin (act-3)
2.39
6628
41709/5.3
20
6
19
Actin (act-4)
M
17568987
37279/5.4
34
8
20
Actin (act-4)
M
17568987
37279/5.4
34
8
21
Actin (act-4)
9.427
17568985
41778/5.3
44
12
22
Actin (act-4)
M
17568987
37279/5.4
36
9
23
Actin (act-4)
8.462
17568987
37279/5.4
33
8
24
Actin (act-4)
M
17568987
37279/5.4
21
6
25
Actin (act-4)
1.702
17568987
37279/5.4
19
4
26
Actin (act-5)
8.517
17551718
41873/5.4
16
4
27
Annexin (nex-1)
3.213
17554342
35696/6.1
37
11
28
Annexin (nex-1)
1.138
17554342
35696/6.1
40
13
29
Disorganized muscle protein DIM-1 short isoform (dim-1)
1.507
25148955
35539/5.2
48
13
30
Disorganized muscle protein DIM-1 short isoform (dim-1)
M
25148955
35539/5.2
24
6
(continued)
Proteomic Analysis of Caenorhabditis elegans
151
Table 1 (continued)
ID numbera Name of protein
Increase fold (n = 3) in Accession Molecular Coverage Matched dauer larvab number mass (Da)/pI (%) peak
31
1.867
Disorganized muscle protein DIM-1 short isoform (dim-1)
25148955
35539/5.2
27
7
Electron transport system/energy generation proteins 32
Cytochrome c oxidase subunit Va
1.13
17555666
20111/5.8
38
7
33
probable cytochrome P450 E03E2.1 [similarity]
M
7498337
55841/8.6
21
4
34
NADH-ubiquinone oxidoreductase, PDSW subunit (1L97)
M
17507827
31242/6.1
15
4
35
ATP synthase subunit (atp-2)
2.403
25144756
57527/5.5
24
9
36
ATP synthase subunit ATP-2
1.297
25144756
57527/5.5
48
11
Other proteins 37
Transthyretin-like family member (4L828)
2.997
17542886
14693/5.2
34
4
38
Transthyretin-like protein precursor family member (5O10)
1.698
17559006
17777/5.8
14
3
39
Arginine kinase family member
M
32566409
39991/6.2
14
5
40
Ump-cmp kinase (2L419)
3.248
17533833
21240/5.9
37
6
41
Aspartic protease ASP-4 1.829
17549909
49278/6.1
25
6
42
4-Hydroxyphenylpyruvate dioxygenase (hpd-1)
1.885
17555220
44383/5.4
40
13
43
Homogentisate oxidase HGO-1
1.822
17507969
49239/5.9
28
7
44
Initiation factor five eIF- 2.017 5A homolog IFF-2
17534327
17954/5.4
42
8
45
Allergen V5/Tpx-1 related family member (5E293)
17561866
22470/4.5
20
3
M
(continued)
152
Jeong et al.
Table 1 (continued)
ID numbera Name of protein
Increase fold (n = 3) in Accession Molecular Coverage Matched dauer larvab number mass (Da)/pI (%) peak
46
Allergen V5/Tpx-1 related family member (5E290)
5.484
17561870
22455/4.50
26
5
47
Ribosomal protein, small subunit (30.7 kD) (rps-0)
1.599
17554768
30703/5.5
23
6
48
Enoyl-coA hydratase
1.741
17560910
28438/6.5
26
5
49
Galectin (lec-1)
1.874
25153023
31810/6.1
24
7
50
Galectin (lec-1)
1.709
25153023
31810/6.1
36
8
51
Galectin (lec-1)
1.678
25153023
31810/6.1
28
6
52
Galectin (lec-2)
1.35
25154078
31296/6.2
29
7
53
Galectin (lec-2)
1.138
25154078
31296/6.2
34
8
54
Galectin (lec-4)
3.217
9857647
32392/6.0 27
7
M observed in the three different gels only in dauer larvae a ID number indicates the protein spot in the 2DE master reference gel b The mean (n= 3) factor of increase in dauer larvae compared to mixed-stage worms, obtained from the three different gels
1.2. Differentially Expressed Proteins in the Dauer Larva 1.2.1. Oxidative StressDefense Related Proteins
1.2.2. Muscle Proteins
Two proteins thought to be involved in stress resistance were more abundant in dauer larvae than in mixed-stage worms: Heat shock protein-12.6 (HSP-12.6); newly detected) and glutathione transferase-1 (GST-1; 2.10-fold greater in dauers) (Fig. 2b, c, Table 1). Eukaryotic cells respond to heat shock by inducing a conserved set of HSPs, which act as molecular chaperones. They have the ability to prevent protein aggregation and in some cases actually promote the renaturation of unfolded polypeptides in vitro (4, 5). The glutathione and thioredoxin systems represent two major antioxidant defense lines in most eukaryotes and prokaryotes. Overall, a few of these defense-related proteins are upregulated (e.g., 1.3–2.1-fold) in dauer larvae, an observation consistent with previous genomics work (6–7). Several muscle proteins were detected with >2-fold upregulation or newly detected in dauer larvae: two levamisole resistant-11 proteins (2.07–3.63-fold), troponin C and T (2.44–4.98-fold), 5 ACT-3 (4.49–9.42-fold), annexin (nex-1) (3.23-fold), and DIM-1 (newly appeared) (Table 1, Fig. 2b, c). The increase in these
Proteomic Analysis of Caenorhabditis elegans
153
rather diverse muscle proteins might be caused by rearrangement of muscle proteins as part of a defense mechanism against oxidative stress and could indicate the presence of proteolysis in the dauer state. The slender body and specialized cuticle might also be attributed to changes in structural, muscle, and other proteins in dauers. 1.2.3. Energy Generation and Other Proteins
Dauer larvae are known to exhibit reduced energy generation and consumption (8). In our experiments, ATP synthase subunit (atp-2) was increased 2.4-fold (n = 3) in dauers, perhaps due to decreased ATP levels. The NADH-ubiquinone oxidoreductase was newly detected in dauer larvae; this enzyme not only provides the input to the respiratory chain from the NAD-linked dehydrogenases of the citric acid cycle but also couples the oxidation of NADH and the reduction of ubiquinone to the generation of a proton gradient then used for ATP synthesis. The energy supply for eukaryotic ciliary and flagellar movement is thought to be maintained by ATPregenerating enzymes such as adenylate kinase, creatine kinase, and arginine kinase. Dauer larvae contained newly detected spots of arginine kinase, which catalyzes the reversible transfer of a phosphoryl group between a phosphorylated guanidine phosphagen and adenosine diphosphate (ADP), and allergen V5/Tpx-1 related proteins (5.48-fold). This result suggests that proteins involved in energy generation are decreased while mitochondrial oxidation increases. Dauers also possessed an increase of galectin (3.21-fold), which may lead to strengthening defense mechanisms and stabilizing cellular structure, as galectins are believed to mediate cell– cell and cell–extracellular matrix interactions during development, inflammation, apoptosis, and tumor metastasis (9, 10). We also found increased UMP-CMP kinase (3.248 fold; n = 3) in dauer larva; this enzyme catalyzes an important step in nucleic acid synthesis: the phosphorylation of UTP, CTP, and dCTP.
1.3. Analysis of Egg Proteome by DIGE Upon Temperature Shift
C. elegans develops in a temperature-dependent manner: ca. 14 h are required from egg to L1 development (embryogenesis) at 20°C, but only 10 h at 25°C. Embryonic lethality at 25°C is almost three times that at 20°C (data not shown). During the course of studies on the effects of disruption of sterol biosyntheses in C. elegans by azacoprostane, we wondered if the expression of egg proteins would be changed when adults were perturbed by a shift in growth temperature (20–25°C). Therefore, to understand the causes of these temperature-associated differences in developmental speed and embryonic lethality, we performed proteomic analysis of eggs, which are free of E. coli proteins and also accurately staged from a developmental perspective. For the proteomic analysis, we searched for the most sensitive dye to detect the low-abundance proteins present in eggs; the detection limit of the CyDye used in DIGE (30–100 pg; see also Chapters “HighResolution 2DE”, “Two-Dimensional Difference Gel Electro-
154
Jeong et al.
phoresis”) is much lower than conventional dyes used in 2DE such as Coomassie brilliant blue (CBB; 100 ng) and silver stain (200 pg) (see also Chapters “Two-Dimensional Electrophoresis (2DE): An Overview” and “Silver Staining of Proteins in 2DE Gels”). In our preliminary data, the number of proteins detected by DIGE was fourfold higher than CBB-visualized proteins in the azacoprostane-treated sample (41 vs. 168 proteins, data not shown). DIGE has other advantages over general 2DE because in 2-D DIGE the labeled samples and internal standard are then mixed and coseparated on the same 2-DE gel. Coseparation of different samples on the same gel suppresses experimental variations intrinsic to 2-DE conditions, thus enabling accurate spot detection and matching, resulting in reduction of variation between experiments (11). Furthermore, the amount of proteins required for DIGE is only 1/20 (50 mg vs. 1 mg) of that needed for 2DE. Thus, we employed the DIGE system for detecting differentially expressed proteins in eggs obtained at two different temperatures. We anticipated that this experiment would provide guidance on selection of optimal experimental conditions egg proteome analysis in the presence of azacoprostane to maximize the detection of drug effects. 1.4. Egg Protein Analysis by DIGE
DIGE analysis was performed on eggs collected from F1 worms grown at either 20 or 25°C. To optimize Isoelectric Focusing (IEF), three different running conditions were evaluated (i.e., 95,000 V; 100,000 V; and 105,000 V), and then the pattern of protein spots was examined. Because 100,000 V appeared to provide the best spot resolution (Fig. 3), we used this condition throughout the work. After eggs were prepared from adult N2 worms grown at 20 or 25°C, cell lysates of eggs were first labeled with Cy3 (green, for egg samples obtained from F2 grown at 20°C) or Cy5 (red, analogous samples at 25°C). An aliquot of internal pooled standard (Subheading 3.6) was labeled with Cy2. Each labeled sample (50 mg) was mixed and run on the 2DE gel (20 × 24 cm). The image shown in Fig. 4a is an overlay of Cy3- and Cy5-labeled proteins in one gel. In a separate gel (20 × 24 cm), an unlabeled preparation of the same sample (1 mg) was also run on 1DE and 2DE. This gel displayed the same pattern of spot distribution when stained with CBB. Quantitative differences between the two differentially labeled samples were assessed with an image analysis system (see also Chapter “Troubleshooting Image Analysis in 2DE”). Corresponding spots differentially detected in DIGE gels were identified by overlaying CBB stained gels, then excising these spots from the CBB-stained gel for MALDI-TOF analysis. From these series of protein identifications by MALDITOF, about 55 differentially expressed spots were found, among which 26 proteins increased whereas 29 proteins decreased at
Proteomic Analysis of Caenorhabditis elegans
155
a
b
c
Fig. 3. 2DE gel image of samples from IEF subsequently run under different running voltages. Multiphor II system (GE Healthcare) was used for IPG focusing; total voltage applied to each IPG strip was (a) 95,000 V, (b) 100,000 V, or (c) 105,000 V. The optimal condition was found to be 100,000 V.
156
Jeong et al. pH 3
pH 10
a
pH 3
pH 10
b
19.
6.
4. 22.
39.
22. 6. 34 37.
1. 5. 1. 8. 7. 4. 11. 2. 2. 34. 10. 9. 8. 14. 13. 12. 14. 15. 18. 20. 16. 15. 21. 24. 17. 25. 28. 3.
23. 6.
27. 26. 30.
29. 36.
39.
42.
31. 40.
32. 35. 38.
33.
41. 44.
43.
49. 20.
31. 53. 34.
51.
48.
47. 54.
45.
46.
49. 47. 55. 52.
47 50. 55.
49.
Proteomic Analysis of Caenorhabditis elegans
157
25°C (Fig. 4a). For example, most of the proteins involved in oxidative stress response were decreased at 25°C. They were Y71H2AM5 (cytochrome oxidase c subunit VIb), GST-7, T02H6.11 (ubiquinol–cytochrome C reductase complex 14 kD subunit), PRDX-2 (peroxiredoxin family), LET-721 (electron transfer flavoprotein-ubiquinone oxidoreductase), and UCP-4 (mitochondrial carrier protein). Chaperone proteins (e.g., DAF-21, T21B10.7, SIP-1) were also less abundant at 25°C. The consequence of this decrease in proteins involved in oxidative stress responses appears to be an observed threefold increase in embryonic lethality at 25°C as compared to 1.0% at 20°C (data not shown). Similar results in lethality and oxidative stress response protein decrease were seen in egg proteins obtained from azacoprostane-treated worms (data not shown). We also found that there is a decrease in proteins associated with protein processing (e.g., CYN-5 and -6) and protein folding, such as ribosome protein (F25H2.10, RPA-2, RPS-21). The 25°C eggs contained higher levels of proteins involved in the cell cycle, such as MCM-7, and RUVB-1, amino acid metabolism-related proteins such as K02F2.2, F01G10.1, TBH-1, WRS-1, and GRS-1, and transcription regulation protein (EFT-2). These results suggest that a temperature change from 20 to 25°C during embryonic development might cause significant perturbation in C. elegans metabolic pathways and cell cycle regulation. Therefore, these proteins differentially expressed upon temperature change may be used as good indicators for monitoring disturbance in embryonic development under heat stress. Fig. 4. Effect of growth temperatures on egg protein expression. (a) DIGE image of egg protein sample labeled with CyDye. To identify temperature-sensitive proteins during embryogenesis, worms grown at 20°C or 25°C and egg total proteins were extracted for proteomic and DIGE analysis. 20°C-cultured N2 egg lysates were labeled with Cy3 (green), 25°C-cultured N2 egg lysates were labeled with Cy5 (red), and an aliquot of internal pooled standard was labeled with Cy2. The image is an overlay of Cy3- and Cy5-labeled proteins. (b) 2DE gel image of egg protein sample (used for (a)) stained with CBB. Differentially expressed proteins identified from twice-performed DIGE of proteins from eggs of C. elegans cultured at 20°C and 25°C are labeled as follows. (1) EFT (elongation factor family member), (2) MCM-7 (yeast MCM related family member), (3) hypothetical protein ZK1151.1, (4) PDI-2 (protein disulfide-isomerase (EC 5.3.4.1)) precursor (PDI-2), (5) TBB-1 (tubulin, beta family member), (6) DAF-21 (abnormal dauer formation family member), (7) GRS-1 (glycyl tRNA synthetase family member), (8) VIT-2 (vitellogenin-2 precursor), (9) hypothetical protein C01B10.8, (10) TBH-1 (tyramine beta-hydroxylase family member), (11) HSP-6 (heat shock protein family member), (12) F01G10.1, (13) LET-721, (14) T21B10.7, (15) VIT-6 (vitellogenin structural genes (yolk protein genes) family member), (16) CCT-4 (chaperonin containing TCP-1 family member), (17) ZK829.4, (18) RUVB-1 (RUVB (recombination protein) homolog family member), (19) TAG-194 (temporarily assigned gene name family member), (20) Y4C6B.1, (21) hypothetical protein C34E10.6, (22) C10G11.7, (23) T05H4.6, (24) FKB-6 (FK506-binding protein family member), (25) F57B10.3a, (26) putative RNA helicase, (27) Y71F9AL.16, (28) K02F2.2, (29) WRS-1 (tryptophanyl tRNA synthetase family member), (30) R07E5.3, (31) F43G9.1, (32) GPD-1 (GPD (glyceraldehyde 3-phosphate dehydrogenase) family member), (33) F56H6.7, (34) DYS-1 protein, (35) R09E12.3, (36) F25H2.10, (37) HIM-3 (high incidence of males (increased X chromosome loss) family member, (38) RACK-1 (RACK1 (mammalian receptor of activated C kinase) homolog family member), (39) W03D2.4, (40) NEX-1 (annexin family member), (41) C17C3.1a, (42) ZK858.3, (43) F58B3.9, (44) TCT-1 (TCTP (translationally controlled tumor protein) homolog family member), (45) PRDX-3 (peroxiredoxin family member), (46) GST-7, (47) SIP-1 (stress induced protein family member), (48) CYN-6 (cyclophylin family member), (49) CYN-5 (chain A, cyclophilin-5), (50) Y71H2AM.5, (51) F09F7.4a, (52) T02H6.11, (53) RPA-2 (ribosomal protein, acidic family member), (54) RPS-21 (ribosomal protein, small subunit family member), (55) UCP-4 (uncoupling protein (mitochondrial substrate carrier) family member).
158
Jeong et al.
2. Materials 2.1. Strain and Mixed N2 Culture
1. The wild-type C. elegans variety Bristol, N2 strain, was obtained from the Caenorhabditis Genetics Center (University of Minnesota, USA). 2. Wild-type C. elegans N2 was grown on nematode growth medium (NGM) plates and S-basal broth medium, under standard uncrowded and well-fed conditions at 20°C unless otherwise noted (12). 3. Mixed N2 were isolated by washing with 0.1 M NaCl and 35% sucrose floatation and then were stored at −70°C until use.
2.2. Solutions and Buffers for C. elegans Culture (13)
1. Grow E. coli (OP50) in liquid culture as a food source for C. elegans. 2. Nematode growth medium (NGM) plate: Combine 1 L of medium containing 3 g/L NaCl, 2.5 g/L bacto-peptone, 17 g/L agar, 1 mL cholesterol stock solution (5 mg/mL in ethanol), 1 mL of 1 M CaCl2, 1 mL of 1 M MgSO4, and 25 mL of 1 M KH2PO4 (pH 6.0). 3. M9 buffer: 3 g KH2PO4, 6 g Na2HPO4·2H2O, 5 g NaCl, and distilled water to a final volume of 1 L. Adjust the buffer to pH 7.0 and autoclave. After autoclaving, add 1 mL of sterile 1 M MgSO4. 4. S-basal buffer: 5.85 g NaCl, 1 g K2HPO4, 6 g KH2PO4, 1 mL cholesterol (5 mg/mL in ethanol), and distilled water to a final volume of 1 L. Sterilize by autoclaving. 5. 1 M potassium citrate (pH 6.0) solution: 20 g citric acid monohydrate, 293.5 g tri-potassium citrate monohydrate, and distilled water to a final volume of 1 L. Sterilize by autoclaving. 6. Trace-metal solution: 1.86 g disodium EDTA, 0.69 g FeSO4·7H2O, 0.2 g MnCl2·4H2O, 0.29 g ZnSO4·7H2O, 0.025 g CuSO4·5H2O, and distilled water to a final volume of 1 L. Sterilize by autoclaving and store in the dark. 7. 1 M CaCl2 solution: 111 g CaCl2 in 1 L of distilled water. Sterilize by autoclaving. 8. S Medium: 1 L S-basal buffer, 10 mL of 1 M potassium citrate (pH 6.0), 10 mL trace-metal solution, 3 mL of 1 M CaCl2, and 3 mL of 1 M MgSO4. Add components using sterile technique and do not autoclave.
2.3. Microscopy and Photography (13)
1. Microscope: Axiovert 135 (Zeiss). 2. Microcoverglass: 48 × 60 mm No. 1. (Thomas Red Label). 3. Image capture: AxioCam (Zeiss).
Proteomic Analysis of Caenorhabditis elegans
2.4. Isoelectric Focusing with Immobilized pH Gradient Strip (14)
159
1. MultiPhor™ (GE Healthcare) or Protean IEF cell (Bio-Rad): Numerous commercially available IEF units exist. 2. Reswelling tray. 3. Mineral oil: Immobiline Dry strip cover fluid (GE Healthcare). 4. Power supply, such as the EPS 3501 XL power supply (GE Healthcare). 5. Thermostatic circulator: Multitemp III thermostatic circulator (GE Healthcare). 6. IPG strips: Immobiline Dry Strip, pH 3–10 nonlinear (NL) or pH 4.0–5.0, and pH 5.5–6.7, 18 cm long, 0.5 mm thick (GE Healthcare) or with the same pH ranges for the ReadyStrip IPG strip (Bio-Rad). 7. Carrier ampholyte mixtures: IPG buffer or Pharmalyte, same range as the selected IPG strip. 8. Sample buffer: 7 M urea, 2 M thiourea, 4% CHAPS, 0.5% ampholyte, 100 mM DTT, 40 mM Tris, and a trace of bromophenol blue. 9. IPG equilibration buffer: 375 mM Tris–HCl, pH 8.8, containing 6 M urea, 2% sodium dodecyl sulfate (SDS), 5 mM tributyl phosphine (TBP), 2.5% acrylamide solution, and 20% glycerol. 10. Agarose overlay solution: 0.5% agarose, 24.8 mM Tris, pH 8.3, 192 mM glycine, 0.1% SDS, and a trace of bromophenol blue.
2.5. Chemicals for DIGE
1. DIGE CyDye: Cy2, Cy3, Cy5 (GE Healthcare). 2. IPG strip: Immobiline Dry Strip, pH 3–10 NL, 24 cm long, 0.5 mm thick (GE Healthcare). 3. Lysis buffer: 7 M urea, 2 M thiourea, 4% CHAPS, 30 mM Tris–HCl, pH 8.5. 4. 2× sample buffer: 7 M urea, 2 M thiourea, 4% CHAPS, 130 mM DTT, 30 mM Tris–HCl, pH 8.5. 5. Quenching solution: 10 mM lysine.
2.6. Preparation of 2DE Gels
1. Gradient former: One of the two Bio-Rad models can be used for this step: Model 385 (30–100 mL capacity) or Model 395 (100–750 mL capacity). 2. Orbital shaker with speed controller: Take care not to produce bubbles. 3. SDS–PAGE: Protean II xi multicell and multicasting chamber (Bio-Rad) or Ettan Dalttwelve electrophoresis system (GE Healthcare).
160
Jeong et al.
4. 5× Tris–HCl buffer: Dissolve 227 g Tris into 800 mL distilled water and adjust the buffer to pH 8.8 with HCl (~30 mL). Add distilled water to a final volume of 1 L. 5. 5× Gel buffer: Dissolve 15 g Tris, 72 g glycine, and 5 g SDS into 800 mL distilled water and add distilled water to a final volume of 1 L. 6. SDS equilibration buffer: Dissolve 36 g urea, 2 g SDS, 20 mL 5× Tris–HCl buffer (pH 8.8), 40 mL 50% glycerol, and 31.25 mL acrylamide monomer. 7. Acrylamide stock solution: Acrylamide/bis-acrylamide 37:5.1, 40% solution (Amresco, M157, 500 mL). 8. Fixing solution: 40% methanol and 5% phosphoric acid in distilled water. 9. CBB G-250 staining solution: 17% ammonium sulfate, 3% phosphoric acid, 34% methanol, and 0.1% CBB G-250 in distilled water. 2.7. 2DE Gel Image Analysis
1. Scanner with transparency unit, such as Bio-Rad GS710 or GS800. 2. 2DE gel image analysis program: Image Master Platinum 5 (GE Healthcare), PDQuest 7.3.0 (Bio-Rad), or Progenesis Discovery (NonLinear Dynamics, Ltd.).
2.8. Destaining and In-gel Tryptic Digestion
1. Speed Vac (Heto). 2. Sequencing-grade modified trypsin (Promega, V5111, 100 mg, 18,100 U/mg): 10 mg/mL in 25 mM ammonium bicarbonate, pH 8.0. 3. 50 mM Ammonium bicarbonate.
2.9. Desalting of Peptides and MALDI Plating
1. GELoader tips (Eppendorf, No. 0030 048.083, 20-mL capacity). 2. Poros 10 R2 resin (PerSeptive Biosystems, 1–1118–02, 0.8 g). 3. Oligo R3 resins (PerSeptive Biosystems, 1–1339–03, 6.3 g). 4. 2% formic acid in 70% acetonitrile (ACN). 5. 0.1% trifluoroacetic acid in 70% ACN. 6. 1-mL syringe. 7. Matrix: a-cyano-4-hydroxycinnamic acid (CHCA). 8. Opti-TOF™ 384-well insert (123 × 81 mm, 1016491, Applied Biosystems).
2.10. MALDI-TOF and Peptide Mass Fingerprinting
1. MALDI-TOF and MALDI-TOF/TOF: Voyager DE-Pro and 4800 MALDI TOF/TOF™ Analyzer (Applied Biosystems) equipped with a 355-nm Nd:YAG laser. 2. The pressure in the TOF analyzer is approximately 7.6e-07 Torr.
Proteomic Analysis of Caenorhabditis elegans
161
3. Methods 3.1. Nematode Culture and Sample Preparation for 2DE
1. Grow wild-type C. elegans (N2) at 20°C in S medium with 5 mg/mL cholesterol (Steraloids Inc, Newport, USA) using the OP50 strain of E. coli as a food source. 2. To prepare 2DE sample from either mixed stages or dauer larvae, grind worms in liquid nitrogen and then resuspend in an appropriate volume of sample buffer A. 3. Sonicate suspensions for approximately 240 s on ice (see Note 1), and collect the soluble fractions by centrifugation at 105,000 × g for 40 min at 4°C. 4. Treat with 50% trichloroacetic acid (TCA) solution. 5. Determine protein concentration in the soluble fraction by the Bradford method (15) using bovine serum albumin as a standard. 6. Store aliquots at −70°C until use.
3.2. Preparation of Dauer Larvae
1. Isolate dauer larvae as previously described with minor modifications (16) (see Note 2). 2. Grow wild-type animals in liquid culture at 20°C until food is exhausted. Then add fresh food to the medium. 3. Incubate the worms at 20°C until food is exhausted, and many dauer larvae are produced (see Note 3). 4. First isolate the worms by washing with 0.1 M NaCl and 35% sucrose floatation. Next, purify dauers by treating the worms with 1% SDS for 30 min, filtering and washing through gauze with distilled water. Store the dauers at −70°C until use (see Note 4).
3.3. Large-Volume Culture of C. elegans in NGM Broth1
1. Start the broth culture of C. elegans crowdedly cultured in the large NGM plates of N2 strain containing a mixture of adults and larvae (see Note 5). 2. Inoculate the worms and bacteria (OP50) into a fermenter containing S-basal medium. 3. Maintain the temperature at 20°C. The air flow rate is zero at the time of inoculation but after a short period of adaptation is increased to 3.0 L/min. 4. Set agitation speed at 100 rpm. When bacteria are exhausted, add OP50 again. Grow the worms continuously under this condition to produce lots of dauer larvae, even though the
1 This Method is useful for Preparation of large quantities of protein extracts from dauer larvae
162
Jeong et al.
culture is not starved; dauers are produced in abundance because, when a culture is dense enough, daumone becomes a dominating influence (see Note 6). 5. When over 80% of the nematodes are dauers (an indicator of dauer pheromone), remove the worms by centrifuging in a Sorvall RC26plus centrifuge with SLA-3000 rotor for 15 min at 4,000 × g at 4°C. 3.4. Egg Preparation
1. Collect worms grown in S-basal medium and centrifuge them into pellets (see Note 7). 2. Suspend the worms in sterile S-basal buffer. Mix the suspension with NaOCl and 5 M NaOH in a ratio of 15: 4: 1, and then vortex the tube for a few minutes. For example, if the final solution is 1.0 mL, the ratio of volumes of each component is: S-basal with worms: NaOCl: 5 M NaOH = 750 mL: 200 mL: 50 mL. Note that worm pellets should not be too large and that the entire process should be completed within 4 min (see Note 8). 3. When about half of the adults are broken down, wash the eggadult suspension with S-basal medium three times and then filter on a 40-mm filter. We harvest the filtered egg. 4. If it is required to synchronize to L1, add S-basal medium and continually shake the suspension. Remove salts with doubledistilled water before protein extraction (see Note 9).
3.5. Egg Protein Extraction
1. Extract egg lysates directly with lysis buffer containing protease inhibitor (Complete Protease Inhibitor Cocktail, Roche). 2. Centrifuge the extracts at 35,300 × g for 45 min and collect supernatants. 3. Treat supernatants with 50% TCA overnight at −20°C and centrifuge at 20,000 × g for 20 min. 4. Wash pellets twice with 100% iced acetone and centrifuge at 20,000 × g for 10 min. 5. Dissolve the pellets in lysis buffer (see Note 10). 6. Adjust the pH of the protein extract to pH 8.5 by adding 50 mM NaOH (monitor with a pH indicator strip), and determine protein concentration using a 2D Quant kit (GE Healthcare).
3.6. CyDye Minimal Labeling of Egg Proteins
1. Perform CyDye labeling according to the manufacturer’s (GE Healthcare) protocol (see Note 11). 2. Mix 50 mg protein sample with 400 pmole CyDye fluor by vigorous vortexing and incubate on ice in the dark for 30 min (see Note 12). 3. Label protein from purified N2 eggs at 20°C and 25°C with Cy3 and Cy5, respectively, and then mix with Cy2-labeled internal pooled standard.
Proteomic Analysis of Caenorhabditis elegans
163
4. Prepare the internal pooled standard sample by pooling 50 mg of protein from each of the two samples. Quench all labeled samples by adding 10 mM lysine for 10 min (see Note 13). 5. Suspend the total protein (150 mg) in the same volume of 2× sample buffer, and then analyze with DIGE. 3.7. 2DE of C. elegans Whole Body Proteins
1. Suspend protein samples (1.5 mg for preparative gels) in sample buffer B to obtain a final volume of 400 mL. 2. Apply aliquots of C. elegans proteins onto an IPG, 3–10NL (GE Healthcare), 3–6L, 5–8L (Bio-Rad) that had been rehydrated with sample protein solution at 20°C for 14 h. 3. Perform IEF at 20°C under a current limit of 50 mA/strip as follows: 100 V for 3 h; 300 V for 2 h; 1,000 V for 1 h; 2,000 V for 1 h; and then continuously at 3,500 V until reaching optimal voltage hour (Vh). 4. Perform focusing for a total of 12,000 V. Equilibrate IPG strips by gently shaking for 20 min in IPG equilibration buffer (see Note 14). 5. In the second dimension of electrophoresis, use vertical SDS gradient slab gels (9–17%; 180 × 200 × 1.5 mm). 6. Cut equilibrated IPG strips to size, and then overlay on the second-dimension gel with agarose overlay solution. Perform electrophoresis at a constant 15 mA per gel. 7. After protein fixation in 40% methanol and 5% phosphoric acid for 12 h, stain the gel with CBB G-250 overnight (see Note 15). 8. After destaining, obtain the gel image using a GS-710 image scanner (Bio-Rad). 9. Process the gel images with Melanie 4 software (GeneBio).
3.8. DIGE of C. elegans Egg Proteins
1. Rehydrate the 450 mL protein solutions (CyDye-labeled 150 mg protein for DIGE and each 1 mg protein for preparative 2DE gels) with a 24 cm Immobiline DryStrip (pH 3–10NL) in a strip holder at room temperature for 16 h, and then focus (IEF) using a MultiPhor II electrophoresis system (GE Healthcare) (see Note 16). 2. Set electrophoresis conditions at 20°C as follows – step 1: 100 V for 4 h; step 2: 300 V for 2 h; step 3: 600 V for 1 h; step 4: 1,000 V for 1 h; step 5: 2,000 V for 1 h; step 6: 3,500 V for 27 h. 3. After IEF, subject IPG strips to a one-step reduction and alkylation by TBP-equilibration buffer for exactly 25 min. 4. Apply the strips to second-dimension 9–16% SDS–PAGE gels using an Ettan Dalttwelve electrophoresis system (GE Healthcare). 5. Electrophorese at 20°C as follows: 2.5 W/gel for 30 min, then 16 W/gel for 6 h. For 2DE analysis, IEF/SDS–PAGE is
164
Jeong et al.
performed using 1 mg of each sample; the gel is stained with CBB and then scanned. 6. Perform all procedures in the dark (see Note 17). 3.9. DIGE Image Analysis (CyDye Labeling)
1. Perform spot matching and statistical analysis using DeCyder BVA (Biological Variance Analysis) v6.5.11 (GE Healthcare) (see also Chapter “Troubleshooting Image Analysis in 2DE”). 2. Set the estimated number of spots to 3,500. 3. Group each gel into “standard,” “20dgr N2,” or “25dgr N2” to allow comparison between the different gels. 4. Accept only statistically significant spots (p <0.05). 5. Filter the accepted spots based on an average volume ratio of 1.5-fold.
3.10. Image Scan and Acquisition
1. Visualize the CyDye-labeled gels using a Typhoon 9400 imager (GE Healthcare) and scan according to excitation/ emission wavelengths for each DIGE fluor: Cy2 (488/520 nm), Cy3 (532/580 nm), and Cy5 (633/670 nm) (see Note 18). 2. Capture images and select areas of interest with ImageQuant v2005 (GE Healthcare). 3. In the case of preparative gels for protein identification, stain with CBB.
3.11. In-Gel Digestion of Protein Spots from 2DE and Protein Identification by MALDI-TOF-MS
1. Excise gel spots with end-removed pipette tips to accommodate various spot diameters. 2. Destain the gel slice in the microtube and dehydrate with 50 mL acetonitrile for 5 min at room temperature. 3. Rehydrate dried gels with 10 mL trypsin solution for 45 min on ice. 4. After removing the excess solution, digest proteins in the gels at 37°C for 24 h. 5. Treat the peptide mixtures thus obtained with the resin solution (POROS R2:Oligo R3 = 1:1 in 70% acetonitrile) (17). 6. Analyze the peptides with a 4800 MALDI-TOF mass spectrometer (Applied Biosystems) operated in the reflectron/ delayed extraction mode with an accelerating voltage of 20 kV and sum data from 1,000 laser pulses. 7. Mix approx. 0.5 mL of a-cyano-4-hydroxycinnamic acid with the same volume of sample. Use the following parameters for time-of-flight measurement: 20 kV accelerating voltage, 75% grid voltage, 0% guide wire voltage, 120 ns delay, and low mass gate of 500 Da. Perform internal calibration also using
Proteomic Analysis of Caenorhabditis elegans
165
autodigestion peaks of porcine trypsin ([M + H]+, 842.5090, and 2211.1064). 8. Analyze the peptide mass profiles produced by MALDI-TOFMS using search programs such as MS-Fit 3.2 provided by UCSF (http://prospector.ucsf.edu) and ProFound (Version 4.7.0) provided by The Rockefeller University (http://129.85.19.192/ profound_bin/WebProFound.exe) with the NCBI database. Use a mass tolerance of 20 ppm for masses measured in reflectron mode. 3.13. MALDI-TOF/TOFMass Spectrometry
1. Use an Applied Biosystems 4800 Proteomics Analyzer (Applied Biosystems, Framingham. MA). 2. This TOF/TOF instrument is equipped with an Nd:YAG laser with 355-nm wavelength of <500-ps pulse and 200-Hz firing rate. 3. In both MS and MS/MS mode, the TOF/TOF operates at a 200 Hz repetition rate. 4. To attain such a high repetition rate, the high-voltage source 1 pulser circuit is modified. 5. The instrument uses a two-stage mirror for better focusing across the broad energy range of fragments in the MS/ MS mode, without sacrificing performance in the MS-only mode. 6. Collision energy in the TOF/TOF instrument is defined by the potential difference between the source acceleration voltage and the floating collision cell, while this voltage difference has been set as 1 or 2 kV. Air is used as a collision gas at pressures of 1.5 × 10–6 Torr and 5 × 10–6 Torr. 7. All fragments are formed from the ion of interest in this region. MS/MS data are further processed using DataExplorer 4.0 (Applied Biosystems, Framingham, MA). 8. Calibrate the spectrum with the tryptic autodigested peaks (m/z 842.5090 and 2211.1046), and monoisotopic peptide masses obtained with Data Explorer 3.5 (PerSeptive Biosystems). 9. Search the Swiss-Prot and NCBInr databases with the Matrix Science (http://www.matrixscience.com) search engine.
3.13. 2D Gel Image Analysis
1. Import the gel image (a 12–16 bit tiff file format is recommended) and convert to an ImageMaster file (*.mel). 2. Detect the protein spots and determine the volume and the % volume of each spot. The % volume is the normalized value that remains relatively independent of any irrelevant variations between gels, particularly those caused by varying experimental conditions. 3. Select differentially displayed protein spots.
166
Jeong et al.
4. Notes 1. Prevention of heat generation: Sonicate for 30 s, cool for 60 s, and pulse (50% reduced) to prevent heat generation. 2. Prevention of contamination: Prevent any contamination from yeast or fungi. Once contaminated, dauer formation frequency is drastically reduced or worms are dead. 3. Dauer larvae preparation: It would be better to add excessive amounts of food to increase the population. Additional food can be supplied when food is exhausted, which leads to an increase in the population that results in dauer formation (>80%). If food is not adequate, arrested worms at the L1 stage can be seen. 4. Filtration: Purified dauer populations were isolated by filtering through 40-mm Falcon cell strainers after 1% SDS treatment. This procedure should be done at room temperature to prevent SDS precipitation. 5. Inoculation of C. elegans to the large broth culture: About 2 mL of a suspension of worms collected in large NGM plate are inoculated into fermenter containing 6 L culture broth. 6. Long-term dauer population: When >80% of worm population reach the dauer state, worms are cultured 5 more days to obtain homogeneous long-term dauers. 7. Preparation of C. elegans eggs: Newly collected eggs are washed with double distilled water to remove any residual salts. This is very important because S-basal medium contains usually high salt levels that might cause streaking problems in 2DE gel analysis. To separate eggs from worm debris, filter the pellets through a 40-mm Falcon cell strainer, which allows eggs to pass through (see also Note 4). When C. elegans is prepared at different stages, separation on a sucrose gradient removes L4 or adults. Also, L1–L3 worms can pass through a10-mm-pore diameter sieve, which also removes any worms larger than these stages. 8. Problem of bacterial contamination: If worms are contaminated with bacteria, the effect of NaOCl is reduced. In this case, centrifuge at 800 × g for 1 min to remove residual bacteria. If worms are exposed to NaOCl for a longer time, egg hatching will be substantially decreased. 9. Sample preparation: It is also important to remove salts for many good reasons in the process of 2DE and DIGE. Residual salts may cause a shift in protein migration in IEF or horizontal streaking phenomena in the gel image (see also Chapters “Two-Dimensional Electrophoresis (2DE): An Overview and “Solubilization of Proteins in 2DE: An
Proteomic Analysis of Caenorhabditis elegans
167
Outline”). Among the many methods of salt removal (e.g., ultrafiltration, gel filtration, solid-phase extractions, dialysis, etc.), TCA precipitation was found to be the most efficient way to remove salts. This method is quick and removes other debris (e.g., lipids, polysaccharides, nucleic acids) that might adversely affect gel separation. When TCA precipitation is performed, the resulting pellets should be washed with 100% iced acetone. 10. TCA removal: Any residual TCA remaining in the sample may cause an acidic change in pH of the sample, thereby necessitating use of greater amounts of the NaOH solution to adjust pH to 8.5 for CyDye labeling. After washing the sample with acetone, samples can be dried no longer than 5 min. Additional drying may affect the solubility of sample pellets. It is desirable to centrifuge the sample at 20,000 × g for 20 min in order to remove any insoluble proteins in the sample. After immediately mixing the protein extracts, they should be divided into aliquots and frozen at −80°C until use, when they should be thawed on ice. HPLC-grade water should be used to make the lysis buffer. 11. Sample labeling using CyDye fluors: Each CyDye should be dissolved in dimethylformamide (DMF) to create a stock solution of 400 pmole/mL. It is also important to maintain the ratio between protein (50 mg) and CyDye fluor (400 pmol). For the best quality of CyDye, use fresh DMF (<2 month after opening cap). Because CyDye is very sensitive to light, perform all sample labeling procedures in the dark. 12. Addition of dye: Add the dye to the walls of reaction tubes to make sure that the two labeling reactions initiate at the same time. When adding the dye, mix completely by repeating vortexing and centrifuging several times. 13. Quenching: Quenching solutions also should be added to the walls of reaction tubes. After quenching, labeled samples should be loaded immediately onto the gels. If this is not possible, stop the reaction before mixing the two-dye reaction mixture and freeze the samples immediately after quenching. 14. Storage of equilibrated IPG strip: After IEF, the strip should be equilibrated immediately to prevent diffusion and loss of protein. An equilibrated strip can be stored for several weeks at −80°C until use. 15. Coomassie Brilliant Blue G-250 Staining (13): Fix the separated proteins into the gel in 200 mL fixing solution for 1 h. Decant the fixing solution and stain the gel in CBB G-250 overnight. Decant the staining solution. Wash several times (>3 times) in distilled water for more than 4 h. Scan the gel, then wrap the gel in plastic, and store at 4°C.
168
Jeong et al.
16. pH range of strip and gel gradient: We used 3–10 nonlinear IPG strips to obtain clearly displayed spot images when proteins were separated on the gel on 9–17% gradient SDS–PAGE. 17. Rehydration and IPG strip focusing: There are two methods for sample rehydration: one is a cup-loading method in which the strip is soaked in sample buffer only and rehydrated after loading the sample in cup. The other method is an in-gel based rehydration method, in which the strip is soaked with the sample buffer and labeled sample. We use the in-gel based rehydration method because the cup-loading method often creates vertical smears around the cup-loading portion in the gel image. In the case of C. elegans egg proteins, optimal voltage was 100,000 V (Fig. 2). 18. DIGE Image acquisition: After the second-dimensional separation, gel plates can be directly scanned by Typhoon 9400 (no removal of gels from the plate is necessary). Images scanned by Typhoon 9400 can then be edited by the ImageQuant program, in which we usually crop the size of the gel image and spot position in a parallel manner because well-cropped images facilitate spot matching and analysis in DyCyder software (see Chapter “Immunoblotting 2DE Membranes”).
Acknowledgments This study was supported by a grant from the Korean Health 21 R&D project, Ministry of Health & Welfare, Republic of Korea (A030003 to YKP), by the Technology Development Program of the Ministry of Agriculture and Forestry, Republic of Korea (606001–53–1-SB010 to YKP), and by a grant from the Basic Research Program that is supported by KOSEF (RO1–2005–000–11021–0 to YHS).
References 1. Riddle, D., and Albert, P. (1997) C. elegans II (Riddle, D., Blumenthal, T., Meyer, B., and Priess, J., eds.), Cold Spring Harbor Laboratory Press, pp. 739–768 2. Paik, Y. K., Jeong, S. K., Lee, E. Y., Jeong, P. Y., and Shim, Y. H. (2006) C. elegans: an invaluable model organism for the proteomics studies of the cholesterol-mediated signaling pathway. Expert Rev Proteomics 3, 439–453 3. Cassada, R. C., and Russell, R. L. (1975) The dauer larva, a post-embryonic developmen-
tal variant of the nematode Caenorhabditis elegans. Dev Biol 46, 326–342 4. Horwitz, J. (1992) a-Crystallin can function as a molecular chaperone. Proc Natl Acad Sci U S A 89, 10449–10453 5. Veinger, L., Diamant, S., Buchner, J., and Goloubinoff, P. (1998) The small heat-shock protein IbpB from Escherichia coli stabilizes stress-denatured proteins for subsequent refolding by a multichaperone network. J Biol Chem 273, 11032–11037
Proteomic Analysis of Caenorhabditis elegans 6. Holt, S. J., and Riddle, D. L. (2003) SAGE surveys C. elegans carbohydrate metabolism: evidence for an anaerobic shift in the longlived dauer larva. Mech Ageing Dev 124, 779–800 7. Wang, J., and Kim, S. K. (2003) Global analysis of dauer gene expression in Caenorhabditis elegans. Development 130, 1621–1634 8. Wadsworth, W. G., and Riddle, D. L. (1989) Developmental regulation of energy metabolism in Caenorhabditis elegans. Dev Biol 132, 167–173 9. Raz, A., Zhu, D. G., Hogan, V., Shah, N., Raz, T., Karkash, R., et al. (1990) Evidence for the role of 34-kDa galactoside-binding lectin in transformation and metastasis. Int J Cancer 46, 871–877 10. Perillo, N. L., Pace, K. E., Seilhamer, J. J., and Baum, L. G. (1995) Apoptosis of T cells mediated by galectin-1. Nature 378, 736–739 11. Alban, A., David, S. O., Bjorkesten, L., Andersson, C., Sloge, E., Lewis, S., et al. (2003) A novel experimental design for comparative two-dimensional gel analysis: Two-dimensional difference gel electrophoresis incorporating a pooled internal standard. Proteomics 3, 36–44
169
12. Brenner, S. (1974) The genetics of Caenorhabditis elegans. Genetics 77, 71–94 13. Choi, B. K., Shin, Y. K., Lee, E. Y., Jeong, P. Y., Shim, Y. H., Chitwood, D., and Paik, Y. K. (2008) Proteomic analysis of the sterol-mediated signaling pathways in C. elegans. Methods Mol Biol, Humana Press, 462 (11), 167–179 14. Cho, S. Y., Lee, E. Y., Kim, H. Y., Kang, M. J., Lee, H. J., Kim, H., et al. (2008) Protein profiling of human plasma samples by twodimensional electrophoresis. Methods Mol Biol, Humana Press, 428, 57–75 15. Bradford, M. M. (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72, 248–254 16. Epstein, H. F., and Shakes, D. C. (1995) Caenorhabditis elegans: modern biological analysis of an organism. In: Cell Biology, Academic Press, London 17. Larsen, M. R., Sorensen, G. L., Fey, S. J., Larsen, P. M., and Roepstorff, P. (2001) Phosphoproteomics: evaluation of the use of enzymatic de-phosphorylation and differential mass spectrometric peptide mass mapping for site specific phosphorylation assignment in proteins separated by gel electrophoresis. Proteomics 1, 223–238
Chapter 11 Protein Extraction for 2DE Claus Zabel and Joachim Klose Summary Our protein extraction protocol for two-dimensional gel electrophoresis (2DE) was updated to meet current needs in the field of proteomics. This protocol summarizes our experience using this method since its introduction over 30 years ago. We provide a total as well as fractionated extraction protocol. The former is easy and fast to use, suitable for most standard 2DE applications, whereas the latter is used for special applications such as the extraction of membrane or nuclear proteins. Both extraction protocols stress the need that protease inhibitors are added early to still deep frozen tissue to preclude an activation of proteases which destroy proteins and make them inaccessible to analysis. We also emphasize that, to remain soluble, proteins need to stay in an environment resembling a living cell as closely as possible. Sample dilution is therefore kept to a minimum and the pH of the extract is close to in vivo conditions at pH 7.1. In addition there are no precipitation/resolubilization steps which could irreversibly remove proteins from the extract. Furthermore, the total extraction does not even require centrifugation. Our extraction protocol is compatible with recent advances in 2DE-staining techniques such as differential in gel electrophoresis and fluorescence staining as well as mass spectrometry. Key words: Protein extraction, Two-dimensional gel electrophoresis, Fractionation, Protease inhibition.
1. Introduction The protocol for extracting proteins from mouse and human tissues (organs) described in this chapter adheres to a strategy that is based on a rationale to include all protein species of a particular tissue in a set of samples which are suitable for two-dimensional electrophoresis (2DE), particularly for large gel 2DE described in Chapter “High Resolution Large Gel 2DE” (1). The extraction procedure is one of the most important steps to maintain reproducibility in gel-based proteomics (2). The ultimate goal is resolution and visualization of all protein species in 2DE gels. David Sheehan and Raymond Tyther (eds.), Methods in Molecular Biology, Two-Dimensional Electrophoresis Protocols, vol. 519 © Humana Press, a part of Springer Science + Business Media, LLC 2009 DOI: 10.1007/978-1-59745-281-6_11
171
172
Zabel and Klose
This aim determines some features of our tissue extraction procedure for retrieving proteins (see also Chapters “Solubilization of Proteins in 2DE: An Outline” and “Difficult Proteins”). Since the last publication of our comprehensive protein extraction protocol (3) we found that an extraction protocol for total cell or tissue proteins may be sufficient for many more users than a protocol for fractionated extraction (4). Still, fractionated extraction has also a number of applications, especially when only certain classes of proteins such as cytoplasmic, membrane, or nuclear proteins are of interest. 1.1. Total Extraction of all Cellular Proteins
Extraction of total proteins at once ensures that no distribution artifacts occur as may be seen when proteins are fractionated into different subcellular fractions. Therefore, the total amount of a given protein is determined within each protein sample investigated. In addition, the total extraction procedure is much simpler, takes less time, and is highly compatible with new protein visualization techniques such as differential in gel electrophoresis (DIGE). This extraction procedure can be applied to almost any tissue, cell culture, or subcellular components. No ultracentrifugation steps are involved which avoids protein precipitation and loss during sample preparation. The total extraction procedure is based on adding protease inhibitors, detergent (CHAPS), urea/thiourea in very close succession to solubilize proteins in a tissue/cell as much as possible and keep them from degradation. Grinding the deep frozen tissue and sonication are the second instrumental part in the solubilization process. DNA-bound proteins are recovered by DNAse treatment. In this way all proteins are solubilized in one fraction without precipitation and fractionation steps.
1.2. Fractionated Extraction of Cellular Proteins
On the other hand, using a fractionation procedure, many different protein species of a tissue can be distributed over several 2DE gels and this certainly increases resolution. However, a postulate is that fractionation of tissue proteins results in fraction-specific proteins. Usually, cell fractionation is performed with the aim of isolating special cell organelles (nuclei, mitochondria) or cell structures (membranes) (see Chapters “Organelle Proteomics” and “Blue Native Gel Electrophoresis (BN–PAGE) Proteomics”, “2DE for Proteome Analysis of Human Metaphase Chromosomes”, and “Microsomal Proteomics”). Proteins are then extracted from these subcellular fractions. This procedure, however, includes washing steps to purify cell fractions and eliminate cell components which are not of interest and cell debris which are rejected. Using fractionation in this way, an uncontrolled loss of proteins is unavoidable. The tissue fractionation procedure described in this chapter renounces the isolation of defined cell components using washing and precipitation steps. This allows us to avoid any selective loss of proteins which is of utmost importance for all protein extraction
Protein Extraction for 2DE
173
protocols. Mouse (human) tissues (liver, brain, heart) are fractionated into three fractions: (1) The “supernatant I + II” (SI + II) containing the proteins soluble in buffer (cytoplasmic and nucleoplasmic proteins), (2) the “pellet extract” (PE) containing proteins soluble in the presence of urea and CHAPS (proteins from membranes and other structures of the cells and cell organelles), and (3) the “pellet suspension” (PS) containing proteins released by DNA digestion (histones and other chromosomal proteins). The SI + II fraction is obtained by homogenization, sonication, and centrifugation of tissue and re-extraction of pellet (I) and the combination of the two supernatants gained in this way. The solution thus obtained is the first protein sample. The pellet (II) that remained is extracted with urea and CHAPS and the homogenate is centrifuged. The supernatant is the PE fraction and constitutes the second protein sample. The final pellet (III) is suspended into buffer containing benzonase, a DNA digesting enzyme. The pellet suspension (PS fraction) is the third protein sample. It is applied to 2DE without further centrifugation. Care is taken during the whole procedure to avoid loss of material. In spite of this caution, some material may be lost, for example by transfer of the pulverized, frozen tissue from the mortar to a tube or by removing glass beads from the sonicated homogenate. However, this does not lead to a preferential loss of certain protein species or protein classes. 1.3. Important Considerations When Using Protein Extraction Procedures
Generally, it is evident that the best conditions for keeping proteins stable and soluble are present in living cells (1, 2). Therefore, an important principle of our tissue extraction procedure is to extract proteins, using conditions as close to the natural environment in the cell as possible. That means keeping ionic strength of the tissue homogenate at 150–200 mM, the pH in a range of 7.0–7.5, protein concentration high, and protecting the proteins against water by adding glycerol to the buffer. Generally, the best conditions for the first tissue extraction step would be if an addition of diluent that disturbs the natural protein concentrations and milieu of the cell were avoided. We prepared a pure cell sap from a tissue by homogenizing it without additives (except for protease inhibitor solutions added in small volumes) followed by high-speed centrifugation and extracted the pellet that resulted successively in increasing amounts (0.5, 1 or 2 parts) of buffer. The series of protein samples obtained were separated by 2DE and the patterns compared. The results showed that by increasing the dilution of proteins the number of spots and their intensities decreased in the lower part of the 2DE patterns and increased in the upper part. The same phenomenon, but less pronounced, was observed even when the cell sap was diluted successively. Apparently, low molecular weight proteins are best dissolved in the pure cell sap and, presumably, tend to precipitate
174
Zabel and Klose
in more diluted extracts. In contrast, high molecular weight proteins dissolve better in more diluted samples. This effect was most pronounced in protein patterns from liver and not obvious in patterns from heart muscle. This is probably due to the high protein concentration in the liver cell sap that is not reached in extracts of other organs. The dependence of protein solubility on the molecular weight of proteins is obscured in 2DE patterns by another effect that causes a similar phenomenon. An increasing protein concentration of the first tissue extract leads to a higher activity of proteases released by breaking cellular structure by homogenization. Protein patterns from pure liver cell sap, extracted without protease inhibitors (and even with inhibitors), showed an enormous number of spots in the lower part of the gel and a rather depleted pattern in the upper part. At a pH surrounding 6, protein spots disappeared almost completely, in the upper as well as in the lower part, suggesting that these proteins are most sensitive to degradation by proteases. By extracting tissue or the first pellet with increasing amounts of buffer, the 2DE pattern (spot number and intensity) shifted from the lower part to the upper part of the gel. Again, this observation was made particularly in liver. The consequence of these observations for our total as well as our fractionated protein extraction procedure was to maintain the total extract or to obtain supernatants I and II at concentrations that keep all the soluble proteins in solution but do not reach a level where proteases cannot be inhibited effectively anymore. As a consequence, we introduced buffer factors which adjust the concentration of the different extracts of each organ. The optimum protein concentrations in the extracts and therefore the buffer factors were determined experimentally. An optimum was reached when a maximum number of spots were present in the upper as well as in the lower part of a 2DE protein pattern. In addition, the region close to pH 6 should not be depleted of spots, a process starting at high molecular weight. The methods described in the following sections were developed with mouse tissues but were found to be applicable to corresponding human tissues, cell culture cells, and subcellular fractions as well.
2. Materials 2.1. Equipment
1. A small apparatus is preferred for performing sonication in a water bath (Transsonic 310, FAUST, Singen, Germany). 2. Glass beads added to tissue samples for sonication: Size (diameter) of glass beads should be 2.0–2.5 mm. A factor 0.034 was calculated for this bead size (see Note 1).
Protein Extraction for 2DE
175
3. Mortar and pestle: Build and size of this equipment are shown in Fig. 1. Mortar and pestle are manufactured of agate or glass (WITA GmbH, Teltow, Germany). Glass was found to be more stable in liquid nitrogen. Alternatively an electronic mortar grinder can be used (Mortar Grinder RM 200; RETSCH, Haan, Germany). 4. A small spatula is formed into a shovel by wrought-iron work (Fig. 1) and used to transfer tissue powder from a mortar to tubes. 2.2. Reagents
1. Buffer A: 20% glycerol, 100 mM KCl, 50 mM Tris–HCl, pH 7.1, filtered and aliquoted into 1 mL portions, and stored at −70°C. 2. Phosphate buffer, pH 7.1: Add 67 mL 200 mM Na2HPO4·2H2O to 33 mL 200 mM NaH2PO4. 3. Buffer B: 20% glycerol, 200 mM KCl, 100 mM phosphate buffer, pH 7.1, filtered, aliquoted precisely into 0.9 mL portions and stored at −70°C. When used, 100 μL of aqueous CHAPS solution is added. The CHAPS concentration in this aqueous solution is calculated (see Table 1) so that the pellet II/buffer homogenate (see Table 1) contains 4.5% CHAPS. This concentration was found to be best when 2DE patterns are compared. The concentration was determined empirically using protein samples containing different amounts of CHAPS.
Fig. 1. Special equipment for pulverizing frozen tissue. Glass mortar and plastic pestle (left and middle section). Spatula used to transfer frozen tissue from mortar to the test tube A regular spatula was molded into a small shovel (right section).
P-MgCl2 (Σ3 × 0.021) Benzonase (Σ3 × 0.025) 4 °C; 15 min
P-MgCl2 (Σ3 × 0.021) Benzonase (Σ3 × 0.025) 4 °C; 15 min
P-MgCl2 (Σ3 × 0.021)
Benzonase (Σ3 × 0.025) 4 °C; 15 min
Benzonase treatment
4°C; 15 min
4°C; 30 min
Σ3 (weight of sample determined after sonication)
Σ3 (weight of sample determined after sonication)
4°C; 15 min
Σ3 (weight of sample determined after sonication)
Sonication, 6 × 10 s
Sonication, 6 × 10 s
Stirring
Sonication, 12 × 10 s
Number of glass beads (Σ2 × 0.034)
Number of glass beads (Σ2 × 0.034)
Σ2 (mg Σ1 + mg Inhibitor 1C + mg Inhibitor 2)
Number of glass beads (Σ2 × 0.034)
Sonication
Σ2 (mg Σ1 + mg Inhibitor 1C + mg Inhibitor 2)
Inhibitor 2 ((Σ1 × 0.01)
Inhibitor 2 ((Σ1 × 0.01)
Inhibitor 2 ((Σ1 × 0.01)
Σ2 (mg Σ1 + mg Inhibitor 1C + mg Inhibitor 2)
Inhibitor 1C (Σ1 × 0.08)
Inhibitor 1C (Σ1 × 0.08)
Inhibitor 1C (Σ1 × 0.08)
Σ1 (mg heart tissue + mg P-CHAPS)
Σ1 (mg brain tissue + mg P-CHAPS)
38 mg bidistilled water
Σ1 (mg liver tissue + mg P-CHAPS)
38 mg bidistilled water
47 mg bidistilled water + 900 μL buffer P
65 mg CHAPS
P-CHAPS (1.6 mg heart tissue)
65 mg CHAPS
56 mg CHAPS
Buffer P-CHAPS
P-CHAPS (1.6 × mg brain tissue)
Buffer P-CHAPS
Buffer P-CHAPS
Heart tissue (25–150 mg)
Heart
P-CHAPS (2.5 × mg liver tissue)
Brain tissue (25–150 mg)
Brain
Liver tissue (25–150 mg)
Homogenization
Liver
Table 1 Total tissue protein extraction protocol
176 Zabel and Klose
Σ5 (weight of sample determined after aliquot)
Σ5 (weight of sample determined after aliquot)
Σ5 (weight of sample determined after aliquot)
Σ6 (weight of sample determined after aliquot DIGE) DTT (Σ6 × 0.1) RT, 5 min Ampholine (Σ6 × 0.1) Σ7 (weight of sample after DTT + Ampholine addition)
Aliquot DIGE (min. 20 μL) Σ6 (weight of sample determined after aliquot DIGE) DTT (Σ6 × 0.1) RT, 5 min Ampholine (Σ6 × 0.1) Σ7 (weight of sample after DTT + Ampholine addition)
Aliquot DIGE (min. 20 μL)
Σ6 (weight of sample determined after aliquot DIGE)
DTT (Σ6 × 0.1) RT, 5 min
Ampholine (Σ6 × 0.1)
Σ7 (weight of sample after DTT + Ampholine addition)
5 μL/gel
5 μL/gel
Aliquot DIGE (min. 20 μL)
Stirring RT, 30 min
Stirring RT, 30 min
IEF (1-D)
Thiourea 2 M (Σ5 × 0.3)
Thiourea 2 M (Σ5 × 0.3)
Thiourea 2 M (Σ5 × 0.3)
5 μL/gel
Stirring RT, 30 min
Urea 6 M (Σ5 × 0.78)
Urea 6 M (Σ5 × 0.78)
Urea 6 M (Σ5 × 0.78)
Urea/thiourea
Aliquot 5 μL
Heart tissue/Σ4
Σ4 (weight of sample determined after benzonase treatment)
Aliquot 5 μL
Brain tissue/Σ4
Σ4 (weight of sample determined after benzonase treatment)
Aliquot 5 μL
Protein concentration
Liver tissue/Σ4
Control factor
Σ4 (weight of sample determined after benzonase treatment)
Protein Extraction for 2DE 177
178
Zabel and Klose
4. Buffer C: 1 mM MgSO4·7H2O, 50 mM Tris–HCl, pH 8.0 (1 mg of pellet III homogenized in 1 mL of this buffer yields a concentration of 1 mM MgSO4). The final solution is filtered, aliquoted into 1 mL portions, and stored at −70°C. 5. Buffer P: 7.7% glycerol (final concentration in sample if buffer factor is 1.6), 50 mM KCl, 50 mM Tris–HCl, pH 7.5. The final solution is filtered and aliquoted into 0.9 mL units and stored at −70°C. 6. Buffer P-MgCl2: 5 mM MgCl2 in buffer P. Final solution is aliquoted into 20 μL units and stored at −70°C. 7. Protease inhibitor 1A: One tablet of CompleteTM (ROCHE, Mannheim, Germany) is dissolved in 2 mL buffer A (according to the manufacturer’s instructions) and the resulting solution aliquoted into 50, 80, and 100 μL units. Inhibitor 1B was prepared in the same way but using buffer B (0.9 mL buffer + 100 μL·H2O) and aliquoted into 30 and 50 μL units. Inhibitor 1C was prepared in the same way as inhibitor 1A but using buffer P. Inhibitor 2 (pepstatin A): Prepared as a stock solution (9.603 mg/100 mL ethanol) and aliquoted into 100 μL portions. All inhibitor solutions are stored at −70°C. 8. DTT solution: 2.16 g DTT is dissolved in 10 mL bidistilled water. The solution is aliquoted into 100 μL portions and stored at −70°C. 9. Sample diluent: Buffer P-MgCl2. The solution is aliquoted into 250 μL units and stored at −70°C.
3. Methods 3.1. Extraction of Total Proteins
To obtain good results with protein extraction harvest organs rapidly and remove all contaminating material such as hair and blood.
3.1.1. Dissection of Mouse Tissue
1. Kill the mouse by decapitation. Thereby, the body is allowed to bleed. The following steps are performed in a cold room:
Dissection of Mouse Liver
2. Cut open the abdomen and perfuse the liver by the vena femoralis on both sides with 5 mL saline (0.9% NaCl solution). 3. Dissect the complete liver from the body, remove the gall bladder without injuring it, and separate the liver into its different lobes. The central part of each lobe, i.e., the region where blood vessels enter the liver lobe, is cut off as are the remainders of other tissues (diaphragm, fascia).
Protein Extraction for 2DE
179
4. Cut the liver lobes into 2–4 pieces, rinse in ice-cold saline, and briefly immerse it there. Immediately after this step the next organ (e.g., heart) is prepared from the same animal, if desirable, and processed to the same stage of preparation as liver. 5. Cut the liver pieces into smaller pieces (about 5 × 5 mm) and transfer each piece to filter paper, immerse into liquid nitrogen, and put into a screw-capped tube where all pieces are collected. During preparation time, tubes are kept in a liquid nitrogen containing box and afterward stored at −70°C. Dissection of Mouse Brain
1. Kill mouse by decapitation. The following steps are performed in a cold room. If several organs have to be taken from the same animal, start with the brain. 2. Cut off the skin of the head and open the cranium starting from the spinal canal proceeding in a frontal direction. Break the cranial bones apart to expose the brain. Remove the brain including both bulbi olfactorii and a small part of the spinal cord. Transfer the brain into a petri dish containing ice-cold saline. Remove any blood vessels and blood from the outside of the brain. 3. As desired, cut the brain to retrieve the desired brain regions or in half along the corpus callosum, place the pieces on filter paper, and then individually immerse into liquid nitrogen and collect in a screw-capped tube. Keep the tubes in liquid nitrogen during tissue harvest and finally store at −70°C.
Dissection of Mouse Heart
1. Kill mouse by decapitation. The following steps are performed in a cold room. 2. Open the thorax and remove the heart. Place the heart into a petri dish containing ice-cold saline. Cut off both atria and open the ventriculi to remove any blood and blood clots. 3. Dry the heart on filter paper, freeze in liquid nitrogen and store in a screw-capped tube at −70°C.
3.1.2. Total Extraction of Liver Proteins
1. Fill frozen liver pieces into a small plastic tube of known weight and weigh quickly without thawing. The amount of liver tissues used for extraction should be between 50 and 150 mg. 2. Place a mortar, pestle, and a small metal spoon into a styrofoam box that contains liquid nitrogen to a level not exceeding the height of the mortar. The mortar should be precooled for at least 3–4 min. Precool pestle and spatula. 3. Put frozen liver pieces into the mortar, and add buffer P, inhibitors 1C and 2. The required volumes of each of the solutions are calculated as indicated in Table 1. The precise amount of each solution is pipetted as a droplet onto a small spoon-like spatula that was kept in the N2-box before use. The
180
Zabel and Klose
solution is immediately frozen into an ice bead that can easily be transferred into the mortar. 4. Slowly grind all frozen components in the mortar to powder. Care should be taken that small pieces of material are not ejected from the mortar when solid frozen material is broken by grinding into smaller parts. 5. Transfer the powder into a prefrozen 2-mL Eppendorf tube using a special spatula (Fig.1). Forceps are used to freeze the tube briefly in N2 and then to hold the tube close to the mortar in the N2 box. Care is taken that no powder is left in the mortar or on the pestle. For collection of this powder always use the same type of plastic tube. This contributes to reproducibility of the subsequent sonication step. Compress the powder collected in the tube by gently knocking the tube against the mortar. The powder can be stored at −70°C or immediately subjected to sonication. 6. For sonication, a predetermined number of glass beads (see Table 1 and Note 1) are added to the sample and the powder is then thawed and kept immersed in ice. Sonication is performed in an ice-cold water bath. The fill height of the water is critical for the sonication effect and should always be at the level indicated in the manufacturer’s instruction manual. Furthermore, when dipping the sample tube into the water, it is important to do this at the “sonication center” visible by a concentric water surface motion which appear when holding the tube into the water. We prefer a small sonication apparatus that forms only one sonication center (see Subheading 2.1). Sonication is performed for 10 s. Immediately afterward, the sample is stirred with a thin wire for 50 s with the tube still remaining in the ice water. The tube is then kept immersed in ice for 1 min. Now the next sonication circle is started, until a total of six 2-min steps (see Note 1). After sonication, the tubes are turned upside down and punctured with a steel needle to generate a 2-mm diameter opening. The holes should not be larger than that to keep the glass beads from crossing over. A second tube is attached to the bottom of the first. Both tubes are transferred into a centrifuge using other samples or dummy tubes for balancing. Tubes are quickly centrifuged at 2,000 × g for 1 min. The sample is thereby transferred in to the lower tube, whereas the glass beads remain in the upper. 7. Add a small bar magnet and slowly stir homogenate at 4°C for 15 min. 8. Add benzonase (Merck, Darmstadt, Germany) and P-MgCl2. Slowly stir the homogenate for 15 min at 4°C (DNA digestion). 9. Determine control factor (Table 1).
Protein Extraction for 2DE
181
10. Remove aliquot for determination of protein concentration (Note: When sample volume is very low, remove aliquot only after addition of urea/thiourea). 11. Add urea and thiourea to the homogenate. The amounts indicated in Table 1 are transferred into a capped 1.5-mL tube. The tube is sealed and turned upside down. Now the bottom of the tube is cut away by scissors and the tube is put on top of the 2-mL tube containing the homogenate. A quick spin transfers the urea/thiourea completely into the homogenate (1 min, 1,000 × g). The homogenate is stirred for about 30 min until the urea/thiourea is dissolved. Now the aliquot for DIGE can be sampled (Table 1 and Note 8). 12. Add DTT and stir homogenate for at least 5 min; then add Ampholine 2–4. Now, carefully remove the small bar magnet. Now aliquot samples to omit freeze and rethaw cycles. Usually samples need to be diluted (most commonly 1:2) by sample dilution solution (Buffer P-MgCl2) before application to isoelectric focusing (IEF). 13. Samples are quick-frozen and stored at −70°C. 3.1.3. Extraction of Total Proteins From Other Tissues
3.2. Fractionated Extraction of Proteins 3.2.1. Fractionated Extraction of Liver Proteins Extraction of Liver Proteins Soluble in Buffer (Supernatant I + II)
In general, the total extraction protocol differs only in buffer factors and the repeats of sonication (Table 2). Only very special tissue such as lens would show deviations. In our experience lens does not require DNAse treatment. 1. Fill frozen liver pieces into a small plastic tube of known weight and weigh quickly without thawing. The weight of the liver tissues should be between 25 and 260 mg (for lower amounts see Note 4). 2. Place a mortar, pestle, and a small metal spoon into a styrofoam box that contains liquid nitrogen to a level not exceeding the height of the mortar. 3. Put frozen liver pieces into the mortar, and add buffer A, inhibitors 1A and 2. The required volumes of each of the solutions are calculated as indicated in Table 2. The precise amount of each solution is pipetted as a droplet onto the spoon that was kept in the N2 box before use. The solution is immediately frozen into an ice bead that can easily be transferred into the mortar. 4. Grind all frozen components in the mortar to powder. Care should be taken that small pieces of material do not jump out of the mortar when starting to break up the hard frozen material with the pestle. 5. Transfer the powder into a 2-mL Eppendorf tube using a special spatula (Fig.1). Forceps are used to freeze the tube briefly in N2 and then to hold the tube close to the mortar in
d
– Pellet powder – Sonication, 6 × 10 s Number of glass beads [Σ3 + (Σ3 × 0.08) + (Σ3 × 0.01)] × 0.034
Inhibitor 2 (Σ3 × 0.01)
6.6 μL
Inhibitor 2 (Σ3 × 0.01)
Pellet powder
Inhibitor 1A (Σ3 × 0.08)
Buffer A (pellet I mg × 0.5)
Pellet I weight
26.4 μL
330
220 μL
110 mg
Supernatant I store frozen
Centrifugation
Inhibitor 1A (Σ3 × 0.08)
Σ3
Buffer A (pellet I mg × 2)
Pellet I weight
Supernatant I store frozen
Centrifugation
23
No sonication
Sonication, 6 × 10 s
Number of glass beads (Σ2 × 0.034)
Brain powder
688
Liver powder
Σ2
Inhibitor 2 (Σ1 × 0.01)
12.5 μL
Inhibitor 2 (Σ1 × 0.01)
No buffer
Inhibitor 1A (Σ1 × 0.08)
625
375 μL
25–250 mg fine pieces
50 μL
a
250 mgb (A)c
Brain
Inhibitor 1A (Σ1 × 0.08)
Σ1
Buffer A (liver mg × 1.5)
Liver pieces
Sl + II fraction: Liver
Table 2 Fractionated tissue protein extraction protocol
Pellet powder
Inhibitor 2 (Σ3 × 0.01)
Inhibitor 1A (Σ3 × 0.08)
Buffer A (pellet I mg × 1.0)
Pellet I weight
Supernatant I store frozen
Centrifugation
Sonication, 12 × 10 s
Heart powder
Inhibitor 2 (Σ1 × 0.01)
Inhibitor 1A (Σ1 × 0.08)
Buffer A (liver mg × 1.0)
100–130 mg (total heart)
Heart
182 Zabel and Klose
Supernatant I + II, ready for use, store frozen in aliquots
100 μL 200 μL
Diluent
Supernatant I + II, ready for use, store frozen in aliquots
77 mg CHAPS 27 μL bidistilled water
73 mg CHAPS (displace 69 μL)
31 μL bidistilled water
Buffer B/CHAPS (pellet II mg × 1.4)
Pellet II weight
900 μL buffer B
147 μL
92 mgd (C)c
Brain
9 μL/gel
Ampholyte pH 2–4 (50 μL × 0.1)
900 μL buffer B
Buffer B/CHAPS (pellet II mg × 1.6)
Pellet II weight
PE fraction: Liver
8 μL/gel
No diluent
100 μL
Final volume of 50 μL supernatant plus additives
2-D electrophoresis
Final volume of 50 μL supernatant plus additives
5 μL
Ampholyte pH 2–4 (50 μL × 0.1)
DTT solution (50 μL × 0.1)
5 μL
DTT solution (50 μL × 0.1)
Urea (50 μL × 1.08)
Aliquot of supernatant I + II (store rest of supernatant I + II frozen)
Supernatant I + II weight
Supernatant II add to I
54 mg
50 μL
628 mg (B)
c
Urea (50 μL × 1.08)
Aliquot of supernatant I + II (store rest of supernatant I + II frozen)
Supernatant I + II weight
e
Centrifugation
Centrifugation
Supernatant II add to I
No stirring
Stirring
38 μL bidistilled water
65 mg CHAPS
900 μL buffer B
(continued)
Buffer B/CHAPS (pellet II mg × 2.2)
Pellet II weight
Heart
6 μL/gel
Supernatant I + II, ready for use, store frozen in aliquots
Diluent
Final volume of 50 μL supernatant plus additives
Ampholyte pH 2–4 (50 μL × 0.1)
DTT solution (50 μL × 0.1)
Urea (50 μL × 1.08)
Aliquot of supernatant I + II (store rest of supernatant I + II frozen)
Supernatant I + II weight
Supernatant II add to I
Centrifugation
Stirring
Protein Extraction for 2DE 183
2DE
e
8 μL/gel
20 μL
Ampholyte pH 2–4 (Serva) (supernatant III mg × 0.0526)
Pellet extract, ready for use, store frozen in aliquots
375 mg (D)c
19 μL
207 mg
19 + 147f = 166
19 μL
239
Supernatant III weigh
Centrifugation
Stirring
DTT solution [(pellet II mg × 0.3) + Σ5] × 0.1
Urea [(pellet II mg × 0.3) + Σ5] × 1.08
Stirring
Pellet powder
Σ5
Inhibitor 1B (Σ4 × 0.08)
Σ4
1,000 μL buffer B/CHAPS
PE fraction: Liver
Table 2 (continued)
8 μL/gel
Pellet extract, ready for use, store frozen in aliquots
Ampholyte pH 2–4 (Serva) (supernatant III mg × 0.0526)
Supernatant III weigh
Centrifugation
Stirring
DTT solution [(pellet II mg × 0.56) + Σ5] × 0.1
Urea [(pellet II mg × 0.56) + Σ5] × 1.08
Stirring
Pellet powder
Inhibitor 1B (Σ4 × 0.08)
1,000 μL buffer B/CHAPS
Brain
7 μL/gel
Pellet extract, ready for use, store frozen in aliquots
Ampholyte pH 2–4 (Serva) (supernatant III mg × 0.0526)
Supernatant III weigh
Centrifugation
Stirring
DTT solution [(pellet II mg × 0.25) + Σ5] × 0.1
Urea [(pellet II mg × 0.25) + Σ5] × 1.08
Stirring
Pellet powder
Inhibitor 1B (Σ4 × 0.08)
1,000 μL buffer B/CHAPS
Liver
184 Zabel and Klose
9 μL/gel
9.4 μL
8 μL/gel
Ampholyte pH 2–4 [(pellet III mg × 0.56) + Σ7] × 0.1
Stirring
DTT solution (Σ7 × 0.1)
Urea (Σ7 × 1.08)
Stirring
Benzonase (Merck) (Σ6 × 0.025)
Pellet powder
Buffer C (pellet III mg × 1.0)
Pellet III
8 μL/gel
Ampholyte pH 2–4 [(pellet III mg × 0.25) + Σ7] × 0.1
Stirring
DTT solution (Σ7 × 0.1)
Urea (Σ7 × 1.08)
Stirring
Benzonase (Merck) (Σ6 × 0.025)
Pellet powder
Buffer C (pellet III mg × 1.0)
Pellet III
b
All factors used in this table are explained in Note 5 This figure is given as an example. The amount of starting material may vary from 25 to 250 mg (see Note 5) c B ?sp;A = control value; D ?sp;C = A,B,C and D are control values (see Note 2) d This figure is provided as an example and is not the result of a calculation. The pellet weight varies because slight losses of material are unavoidable, even during very precise work e See Footnote d for pellets; this also holds true for supernatants f Buffer B/CHAPS volume g Buffer C
a
2DE
Pellet suspension, ready for use, store frozen in aliquots
Ampholyte pH 2–4 [(pellet III mg × 0.3) + Σ7] × 0.1
Stirring
7.3 μL
79 mg
Urea (Σ7 × 1.08)
DTT solution (Σ7 × 0.1)
3.5 + 69 g = 73
3.5 μL
138
69 μL
69 mgd
Σ7
Stirring
Benzonase (Merck) (Σ6 × 0.025)
Pellet powder
Σ6
Buffer C (pellet III mg × 1.0)
Pellet III weight
Protein Extraction for 2DE 185
186
Zabel and Klose
the N2 box. Care is taken that no powder is left in the mortar or on the pestle. For collection of this powder always use the same type of plastic tube. This contributes to reproducibility of the subsequent sonication step. Compress the powder collected in the tube by gently knocking the tube against the mortar. The powder is stored at −70°C or immediately subjected to sonication. 6. For sonication a predetermined number of glass beads (see Table 2 and Note 1) are added to the sample, and the powder is then thawed and kept immersed in ice. Sonication is performed in an ice-cold water bath. The fill height of the water is critical for effective sonication and should always be at the level indicated in the instruction manual of the manufacturer. Furthermore, when dipping the sample tube into the water, it is important to find the “sonication center” visible by concentric water surface motion which appear when inserting the tube. We prefer a small sonication apparatus that forms only one sonication center (see Subheading 2.1). Water must be kept ice cold. Sonication is performed for 10 s. Immediately afterward the sample is stirred with a thin wire for 50 s with the tube still remaining in the ice water. The tube is then kept immersed in ice for 1 min. Now the next sonication circle is started, until a total of six 2-min steps (see Note 1). After sonication, the tubes are turned upside down and punctured with a steel needle to generate a 2-mm diameter opening. The holes should not be larger than that to keep the glass beads from crossing. A second tube is attached to the bottom of the first. Both tubes are transferred into a centrifuge using other samples or dummy tubes for balancing. Tubes are quickly centrifuged at 2,000 × g for 1 min. The sample is thereby transferred into the lower tube, whereas the glass beads remain in the upper. The homogenate is then frozen in liquid nitrogen and is stored at −70°C. Be careful to hold the tube at an angle of about 45° while freezing to decrease the difficulty of retrieving the homogenate. 7. Detach the frozen homogenate in the tube from the wall by quickly knocking at the tube with a large screw driver. Transfer the frozen piece of sample into a preweighed centrifuge tube and briefly thaw the homogenate. Centrifuge for a maximum of 226,000 × g for 30 min at 4°C. 8. Completely withdraw the supernatant (I) with a Pasteur pipette and fill into a preweighed small test tube. The centrifuge tube is kept on ice and the pipette is put back into the tube again with the tip at the center of its bottom (Note: the pellet sticks to the wall if a fixed-angle rotor was used). In this position, remainders of the supernatant accumulate at the bottom of the tube and inside the pipette. Withdraw residual supernatant with the pipette and transfer it to the test tube. Now the supernatant is frozen in liquid nitrogen and stored at −70°C.
Protein Extraction for 2DE
187
9. Weigh the pellet (I) left in the centrifuge tube on ice and add buffer A at an amount calculated as indicated in Table 2. Mix the pellet and buffer by vortexing, collect the homogenate at the bottom by a short spin, freeze in liquid nitrogen and store at −70°C, or process further immediately. 10. Transfer the homogenate from the centrifuge tube to the mortar after detaching the frozen homogenate from the wall by knocking onto the bottom of the tube. Grind the homogenate together with inhibitors 1A and 2 to powder as described earlier for the liver pieces. Transfer the powder back into the used centrifuge tube, taking care that no powder remains in the mortar. 11. Thaw the powder and slowly stir the homogenate for 45 min in a cold room. 12. Centrifuge the homogenate as described in step 7. 13. Completely withdraw the supernatant (II) from the pellet. This is done in such a way that a white layer which partially covers the surface of the supernatant sinks unaffected onto the pellet. Collect the remainder of the supernatant as mentioned in step 8. Add supernatant II to supernatant I and thoroughly mix both solutions. Determine the weight of the total supernatant. 14. Take a 50-μL aliquot from the total supernatant and mix with urea, DTT solution, and ampholyte pH 2–4, as indicated in Table 2. The final concentrations of these components are: 9 M urea, 70 mM DTT, and 2% ampholytes. These three components should be added to the supernatant in the order indicated here and each component should be mixed and dissolved in the supernatant before adding the next. The final volume of this supernatant mixture is 100 μL. Add 100 μL sample diluent (see Table 3) and mix before IEF run.
Table 3 Sample diluent Component
Mixture
Final concentration
Urea
1.08 g (=0.80 mL)
9.000 M
DTT-solutiona
0.10 mL
0.070 M
Servalyt pH 2–4 (Serva)
0.10 mL
2.000%
Bidistilled water
1.00 mL
50.000%
Sample diluent
2.00 mL
a
See text Subheading 2.2, item 5
188
Zabel and Klose
15. The resulting solution is the final sample (“supernatant I + II”, SI + II). Divide the sample into several portions, freeze each portion in liquid nitrogen, and store at −70°C. Usually 8 μL of sample is applied to an IEF gel, if the large gel 2DE technique described in Chapter “High Resolution Large Gel 2DE” is used (see Note 6). Freeze the remaining portion of the pure supernatant and store at −70°C. 16. Determine the weight of the pellet (II). Collect the pellet at the bottom of the tube by a short spin, then freeze in liquid nitrogen and store at −70°C. Extraction of the Pellet Proteins Soluble in the Presence of Urea and CHAPS (Pellet Extract)
1. Grind pellet II, buffer B/CHAPS, and inhibitor 1B to powder in a mortar placed in liquid nitrogen (see Subheading “Extraction of Liver Proteins Soluble in Buffer (Supernatant I + II)”, steps 2–4). The calculations for buffer and inhibitor volumes are provided in Table 2. Transfer the powder back to the centrifuge tube; avoid leaving any remainders in the mortar or on the pestle. 2. Thaw the powder mix and stir slowly for 60 min in a cold room (CHAPS solubilizes structural proteins). 3. Add urea (for the amount see Table 2) to the homogenate and stir the mixture for 45 min at room temperature (urea reaction). Some minutes after adding urea a large part of it is already dissolved. At this point add DTT solution (for amount see Table 2). 4. Remove the magnet rod from the homogenate. At this step, also avoid any loss of homogenate. Centrifuge the homogenate at 17°C for 30 min at 226,000 × g. 5. Completely withdraw the supernatant (III) with a Pasteur pipette and fill it into a small preweighed test tube. Collect the remainders of the supernatant as mentioned in Subheading “Extraction of Liver Proteins Soluble in Buffer (Supernatant I + II)”, step 8. Determine the weight of the supernatant. 6. Add ampholytes of pH 2–4 (for amount see Table 2) to the supernatant and immediately mix solution. 7. The resulting solution is the final sample (“pellet extract”, PE). Divide the sample into several portions, freeze each portion in liquid nitrogen, and store at −70°C. Usually the volume of sample applied per IEF gel is 8 μL, if large gel 2DE (see Chapter “High Resolution Large Gel 2DE”) is used (see Note 6). 8. Determine the weight of the pellet (III). Collect the pellet on the bottom of the tube by a short spin; then freeze in liquid nitrogen and store at −70°C.
Protein Extraction for 2DE Suspension of the Remaining Pellet (Pellet Suspension)
189
1. Grind pellet III and buffer C to powder in a mortar as described in Subheading “Extraction of Liver Proteins Soluble in Buffer (Supernatant I + II)”, steps 2–4. The buffer volume is calculated as indicated in Table 2. The powder is transferred into a test tube, avoiding any loss of material. 2. Thaw powder and add benzonase (Merck, Darmstadt, Germany; for the amount see Table 2). Slowly stir the homogenate for 30 min in a cold room (DNA digestion). 3. Add urea (for amount see Table 2) and stir the homogenate at room temperature for another 30 min. During this time add DTT solution (for amount see Table 2) once the largest part of urea is dissolved. Finally add ampholytes of pH 2–4 (for amount see Table 2) and quickly mix with the homogenate. 4. The resulting solution is the final sample (“pellet suspension”, PS). It is frozen in liquid nitrogen and stored at −70°C. Usually 9 μL of sample is applied per IEF gel (if large gel 2DE is used; see Chapter “High Resolution Large Gel 2DE” and Note 6). The sample still contains some fine undissolved material and is therefore transferred to the gel with a thin Pasteur pipette instead of a microliter syringe.
3.2.2. Fractionated Extraction of Brain Proteins
Frozen brain tissue is transferred into a mortar that was placed into a box containing liquid nitrogen and crushed with the pestle to fine pieces. The crushed material is transferred completely back into test tubes so that one tube contains 25–250 mg of frozen tissue (weigh the tube without thawing the tissue). This material is used to prepare the supernatant I + II, the pellet extract, and the pellet suspension. The procedure follows that of liver extraction with some differences indicated in Table 2. One exception is that the tissue powder is produced without buffer and subjected to centrifugation without sonication so that a rather small amount of supernatant I results. Sonication is only performed with pellet I homogenate.
3.2.3. Fractionated Extraction of Heart Proteins
The supernatant I + II, the pellet extract, and the pellet suspension are prepared from a single heart. The procedure is as described for liver with some modifications which are indicated in Table 2.
4. Notes 1. Sonication: Conditions for sonicating mouse tissue homogenates (liver, brain, heart) were optimized to break membranes of all cells and cell nuclei of a tissue. Three parameters were
190
Zabel and Klose
varied in the experiments: the time of sonication, the number of sonication repeats, and the number of glass beads added per volume of homogenate. The effect of the various parameters was determined by inspection of sonicated material under the microscope. Sonication for 10 s increases the temperature of the homogenate from 0 to 11–12°C. Therefore, sonication was not performed for more than 10 s. Glass beads are essential for breaking cellular structures (membranes). Since most of the homogenates are rather viscous fluids, the beads cannot flow freely and their addition is limited. For a given homogenate volume a certain number of glass beads are necessary to expose it evenly to the disrupting power of sonication. Due to viscosity, an increase in this number has not much effect. This number can be calculated. By standardization experiments we determined that a factor 0.034 is a good approximation for the optimal number of glass beads for a given volume of homogenate. It can be concluded from the aforementioned that the only parameter which can be varied without harming the sample while still increasing the effect of sonication was the number of repeats of the 10-s sonication period. Under the conditions described in the Subheading 3 (Step 6) the membranes of all cells were broken and no longer visible under the microscope. However, a certain number of intact nuclei were still detectable. Still, this shortcoming was not compensated for by increasing the number of sonication repeats because we assume that homogenization, stirring, and high-speed centrifugation may eventually extract all nucleoplasm proteins. A more aggressive sonication procedure using a metal tip cannot be recommended. We observed heavily disturbed 2DE patterns as a result of employing this technique: many protein spots disappeared depending on the extent of sonication and new spot series occurred in the upper part of the 2D gel, apparently as a result of aggregation of protein fragments. 2. Control values: Control values were calculated for each sample prepared by the fractionated extraction procedure to monitor correctness and reproducibility of the preparation. The calculation of control values is shown in Tables 1 and 2. To visualize the calculation procedure an example from a real experiment is provided here: From a series of 73 individual mouse hearts the SI + II fractions were prepared and the control values B , A (see Table 2) were calculated: 60 samples yielded values between 1.97 and 2.10, three samples between 1.94 and 1.96, and six samples between 2.11 and2.13. Four samples with largest deviations (1.90, 1.91, 2.16, and 2.24) were excluded from the investigation. The range 1.97–2.10 (mean: 2.04 ± 0.04) was used as reference value for preparation of mouse heart SI + II extracts.
Protein Extraction for 2DE
191
3. The extraction procedure was adapted for labeling with CyDyes. It is important that only urea and thiourea should be present in the sample for labeling. DTT and ampholines are added only afterward since both bind CyDyes. In addition, determination of the protein concentration is now mandatory since the amount of CyDyes for labeling is calculated based on protein concentration. Therefore a 5-μL aliquot of the sample is specifically used for this purpose (Table 1, protein concentration). 50 μg protein labeled by 400 pMol CyDye works well for brain samples. 4. The tissue amount indicated (25–150 mg) yields enough sample to run a large number of 2DE gels so that less rather than more material may be used. In cases in which only a very small amount of tissue is available (e.g., 2–5 mg heart biopsy samples, 10–12 mg of two mouse eye lenses, early mouse embryos) a total protein extract should be prepared instead of SI + II, PE, and PS fractions. Small plastic tubes and a glass rod with a rough surface at the well-fitting tip may serve as mortar and pestle. 5. Explanation of correction factors used in Table 2: Factors were calculated to determine the amounts of urea, DTT solution, and ampholytes necessary to transmute any volume of a solution (theoretically water) into a mixture containing 9 M urea, 70 mM DTT, and 2% ampholytes. Calculations of factors: 500 μL water (aqueous protein extract) + 540 mg urea (displaces 400 μL) + 50 μL DTT solution (see Subheading 2.2, item 5) + 50 μL ampholyte solution (commercial solutions which usually contain 40% ampholytes) = 1 mL. If the volume of a protein solution to be mixed with urea, DTT, and ampholytes is n μL, the amounts of the components to be added are: (540 500) × n = 1.08 × n mg urea, (500 50) × n = 0.1 × n μL DTT solution, and 0.1 × n μL ampholyte solution. If the protein solution already contains urea and DTT (see Table 2: PE preparation), the factor 0.0526 is used to calculate the ampholyte volume for this solution. Calculation of this factor: (500 μL extract + 400 μL urea + 50 μL DTT solution) ÷ 50 μL ampholyte solution = 0.0526. If a protein solution of n μL includes a cell pellet, i.e., insoluble material, n μL volume should be reduced by the volume of insoluble material (theoretically by the volume of the dry mass of this material). For this reason a pellet factor (e.g., 0.3) was introduced. This factor was determined experimentally using urea as an indicator. The factor reduces the volume n μL of a protein solution (containing a pellet) to volume n´ μL; n´ μL × 1.08 yields an amount of urea that is added to the n μL of the solution, at the border of solubility, i.e., about 9 M. Note that pellet III contains urea
192
Zabel and Klose
and DTT by the previous steps. Therefore, in this case the pellet volume was not taken into account when calculating the amounts for urea and DTT to be added to the final pellet suspension. In all these calculations no distinction was made between volumes (μL) and weights (mg). A density of 1 was assumed for all tissues and solutions. This makes the calculation somewhat incorrect but more practicable and reproducible. The inhibitor 2 solution was prepared as concentrated as possible to keep the volume of this solution small (1/50th of the homogenate volume, i.e., factor 0.02). This allows for ignoring the volume determination error that results when this inhibitor solution is added instead of including it to the total volume of the homogenate. The factor 0.08 for all inhibitor 1 (A, B, and C) solutions was derived from the dilution requirements according to manufacturer’s instruction (Complete™ tablets, ROCHE, Mannheim, Germany): 1 tablet was dissolved in 2 mL buffer and this volume added to 25 mL of the homogenate, i.e., the volume of the inhibitor 1 solution to be added to n μL homogenate is (2 25) × n = 0.08 n μL. The volume of benzonase solution (ready-made solution from MERCK, Darmstadt, Germany) necessary to digest DNA in the pellet III suspension was determined experimentally: a chromatin pellet was prepared from isolated liver cell nuclei and found to change from a gelatinous clot to a fluid if treated as follows: 1 g chromatin pellet + 1 mL buffer + 0.050 mL benzonase solution (= 0.025 mL Benzonase/mL homogenate), stirred for 30 min at 4°C. If benzonase from other sources is used the amount added has to be adjusted. Note that tissue to be homogenized must be free of any wash solutions; otherwise, the calculated buffer volumes will be too high. 6. Amount of protein applied per gel: The protein amount applied to IEF gels (see Tables 1 and 2) contains about 100 μg protein. There is, however, no need to determine the protein concentration of each sample prepared in order to get protein patterns of reproducible intensities (silver staining only). The concept of the procedure described here for extracting tissues was to keep the volume strongly correlation to the amount of the starting material (tissue or pellet) which was extracted. Therefore, by working precisely, the final sample should always contain nearly the same protein concentration. This was confirmed by determining the protein concentrations of a large number of protein samples. Accordingly, the reproducibility of the pattern intensity depends on the precise sample volume applied to the gel – and, of course, on the protein-staining procedure. The sample volumes per gel given in Table 2 are adapted for silver-staining protocols (see Chapter “Silver Staining of Proteins in 2DE Gels”). As a general guideline one
Protein Extraction for 2DE
193
should take into account: decreasing the protein amount per gel and increasing the staining period is better than vice versa; diluted samples using reasonable volumes are better than concentrated protein samples at a small volume since in the first case clogging is avoided. 7. Maximum resolution of tissue protein fractions by 2DE: Fig. 2 shows 2DE patterns of three protein fractions SI + II, PE, and PS of mouse liver. The SI + II fraction reveals the highest number of protein spots. When spots were counted visually, i.e., by placing a 2DE gel on a light box and dotting each spot with a pencil (3), about 9,200 proteins were detected in this fraction. The SI + II pattern of the brain revealed about 7,700 proteins, that of the heart about 4,800 protein spots. The high spot numbers reflect the high resolution of the large gel 2DE (see Chapter “High Resolution Large Gel 2DE”), which reveals many weak spots between major spots. All these spots were counted with high precision and great care. 8. Effect of fractionated extraction of tissue proteins: The purpose of fractionating the proteins of a tissue was to increase the number of proteins detectable by 2DE. This goal, however, would only be achievable if each fraction contained a notable number of proteins which are strongly fraction specific, so that the total tissue proteins can be distributed over several gels. Comparison of the 2DE patterns from the SI + II and PE fractions of the liver (Fig. 2) showed that the PE pattern revealed about 2,000 protein spots not detectable among the 9,200 spots of the SI + II pattern. The PS pattern revealed only about 70 additional spots. The PS protein spots represent classes of the most basic proteins (Fig. 2) and belong mainly to the chromosomal proteins (e.g., histones). Therefore, the PS fraction is only of interest when a class of very basic proteins is subject of the investigation. Considering the 2DE patterns of the SI + II and PE fraction in more detail (Fig. 3), quite a number of very prominent spots can be observed which are visible in one pattern but do not occur, not even in trace amounts, in the others. At the same time, other spots revealing only low intensities are present in both patterns. This suggests that the phenomenon where many protein spots of the SN I + II pattern also occur in the PE pattern is not due to impurities of the pellet fraction by supernatant proteins. We found, using centrifugation at different speeds and samples prepared with different degrees of viscosity, that overlap in 2DE patterns is due to the fact that most proteins exist solitarily as well as in protein complexes. Apparently, protein complexes sink to the pellet at high speed and are later resolved again by detergent and urea extraction in the PE fraction.
194
Zabel and Klose
Fig. 2. 2-DE protein patterns from mouse liver. Tissue was fractionated into (a) supernatant I + II (b) pellet extract and (c) pellet suspension as described in Subheading 3. Three fractions were subjected to large gel 2DE (see Chapter “High Resolution Large Gel 2DE”). In the pellet extract pattern many protein spots were visible which are not present in the supernatant pattern and vice versa. Pellet-specific spots occur predominantly in the basic half, supernatant-specific spots more on the acid half of the pattern. The pellet suspension pattern reveals very basic proteins of a tissue extracts. Since IEF gels do not cover the entire basic pH range, these proteins cannot reach their isoelectric points. To prevent these proteins from accumulating at the basic end of a gel, the IEF run was shortened by 2 h at the 1,000 V level. Consequently, the very basic proteins form streaks instead of focused spots. For evaluation of patterns in terms of spot number see Notes 7 and 8.
Protein Extraction for 2DE
195
Fig. 3. Sections from 2DE patterns shown in Fig.2. (a) The supernatant I + II and (b) the pellet extract of liver are compared. (i) Some of the prominent protein spots present in the supernatant pattern (northeast arrow) but completely absent in the pellet extract pattern (northeast arrow), (ii) the reverse situation (southwest arrow, and (iii) some spots of low intensity present in both patterns (southeast arrow) are indicated. Other spots show a high intensity in one pattern but low intensity in the other. These spots may reflect naturally occurring unequal distributions of proteins between the two different fractions due to a formation of protein complexes rather than contaminations of one fraction by the other (see Note 8).
Taking into account all three fractions, the total liver proteins could be resolved to about 11,270 different proteins (polypeptide spots). This, however, does not mean that the protein sample preparation procedure described here – in combination with the 2DE technique in Chapter “High Resolution Large Gel 2DE” – revealed all proteins present in liver. Many proteins may exist in undetectable amounts, and special fractionation procedures with subsequent protein concentration steps would
196
Zabel and Klose
be required to detect these proteins. We isolated liver and brain cell nuclei and extracted the nuclear pellet in a similar way as was done for tissues. The 2DE patterns showed that the nuclear extracts add a large number of new proteins to those already known from the tissue extract patterns. However, protein spots present in both the nuclear and the tissue extracts occur as well, particularly in the acidic halves of the supernatant patterns. In general, the proteins represented by the SI + II, PE, and PS patterns can be considered as the main population of protein tissue to which further proteins may be added by analyzing purified and concentrated subfractions. Therefore, the amount of spots (protein isoforms) detectable with fractionated extraction is larger than by total extracts, but the effort involved is also exceptionally higher.
Acknowledgment The authors appreciate critical comments and support from Marion Herrmann and Yvonne Kläre.
References 1. Klose, J., and Kobalz, U. (1995) Twodimensional electrophoresis of proteins: an updated protocol and implications for a functional analysis of the genome. Electrophoresis 16, 1034–1059 2. Challapalli, K. K., Zabel, C., Schuchhardt, J., Kaindl, A. M., Klose, J., and Herzel, H. (2004) High reproducibility of large-gel twodimensional electrophoresis. Electrophoresis 25, 3040–3047
3. Klose, J. (1999) Fractionated extraction of total tissue proteins from mouse and human for 2-D electrophoresis. Methods Mol Biol 112, 67–85 4. Zabel, C., Sagi, D., Kaindl, A. M., Steireif, N., Klare , Y. , Mao, L. , (2006) Comparative proteomics in neurodegenerative and nonneurodegenerative diseases suggest nodal point proteins in regulatory networking. J Proteome Res 5, 1948–1958
Chapter 12 Analysis of Proteins from Marine Molluscs Suze Chora, Maria João Bebianno, and Michèle Roméo Summary Application of the two-dimensional gel electrophoresis (2DE) protocols which were developed for samples of mammalian origin gives unsatisfactory results when used in samples from marine molluscs. This chapter describes a detailed protocol of 2DE that can be applied to these organisms, especially for Ruditapes decussatus and Bathymodiolus azoricus. Key words: “Two-dimensional electrophoresis (2DE)”, Marine bivalves, Ruditapes decussatus, Bathymodiolus azoricus.
1. Introduction Marine molluscs, particularly mussels, oysters, and clams, have been used worldwide as bioindicators to assess the impact of pollutants in coastal marine ecosystems and their health status. More recently, the discovery in 1997 of large mussel beds in deep sea hydrothermal vents (1), particularly in the Mid-Atlantic Ridge, has attracted much scientific attention due to their capacity to live in one of the most extreme environments on Earth characterized by high temperature and pressure, low pH, enriched in toxic sulphide species (2), radionuclides, and naturally high bioavailable metal concentrations that would be toxic or even lethal to coastal marine species (3, 4). Therefore these species have been considered as models to assess pollution effects in natural contaminated environments (5). The isolation of proteins from marine bivalves from coastal or extreme environments, like deep sea hydrothermal vents, using two-dimensional electrophoresis (2DE) was proved not to be straightforward when using standard protocols developed for David Sheehan and Raymond Tyther (eds.), Methods in Molecular Biology, Two-Dimensional Electrophoresis Protocols, vol. 519 © Humana Press, a part of Springer Science + Business Media, LLC 2009 DOI: 10.1007/978-1-59745-281-6_12
197
198
Chora, Bebianno, and Roméo
mammalian tissues mainly due to the excess of salt water and the lipid content of some of particular tissues studied. Recently, 2DE protocols were developed for bivalve species in particularly for the coastal clam Ruditapes decussatus and the deep sea hydrothermal vent mussel Bathymodiolus azoricus. Two tissues were used: gills and digestive gland. 2DE gels (18 × 18 cm) were able to separate more than 2,000 proteins. Although the genome of these species is unknown, a number of proteins have already been sequenced. Some relevant proteins were separated from the spots and some of them sequenced. In bivalves, gill and digestive gland tissues usually do not provide good resolution 2DE gels because they may contain exogenous proteins. To overcome this problem in this protocol we describe the best technique for protein separation using 2DE.
2. Materials 2.1. Equipment
IKA, model Ultra-Turrax TD 25, Ettan IPGphor II (GE Healthcare), Protean II xi cell (Bio-Rad, Labs). Vertical system.
2.2. Solutions and Reagents
1. Homogenization buffer: 10 mM HEPES and 250 mM Sucrose, 1 mM dithiothreitol (DTT), 1 mM EDTA, 1 mM PMSF, 10% protease inhibitors (protease inhibitor tablets, Sigma, Aldrich). DTT, PMSF, and protease inhibitors are added just prior to use. This can be stored at 4°C for 1 week. 2. Precipitation solution: 10% trichloroacetic acid (TCA) in acetone containing 20 mM DTT. DTT is added just prior to use. This can be stored at 4°C for 1 week. 3. Sample dilution buffer: 7 M urea (GE Healthcare), 2 M thiourea (GE Healthcare), amberlite MB-150 1% (Sigma, Aldrich) (6). Filter solution through 0.20-mm filter. Add 4% CHAPS (GE Healthcare), 0.8% Pharmalyte (GE Healthcare), 65 mM DTT (GE Healthcare), 10% isopropanol (6), and a few grains of bromophenol blue (BPB). Store in 2.5 mL aliquots at −70°C for 2 months. 4. SDS equilibration buffer: urea 6 M, Tris–HCl, pH 8.8, 75 mM, glycerol 3%, SDS 4%, and a few grains of bromophenol blue (BPB). Store aliquoted. Can be stored at −20°C for 2 months (see Note 1). 5. Electrophoresis buffer: SDS: 0.1%, Tris base: 25 mM, Glycine: 192 mM. Store at room temperature. 6. 10% SDS solution: Filter solution through a 0.45-mm filter and store at room temperature.
Analysis of Proteins from Marine Molluscs
199
7. 10% ammonium persulphate solution: Prepare just prior to use. 8. Agarose sealing solution: 0.5% agarose. Add all ingredients into a 500 Erlenmeyer flask. Swirl to disperse. Heat on a heating stirrer until the agarose is completely dissolved (see Note 2). Dispense 2 mL aliquots and store at room temperature. 9. EDTA 500 mM: store at room temperature. 10. Immobiline DryStrip, pH 3–10 NL, 18 cm (GE Healthcare), immobilized pH-gradient isoelectric focusing gels for the first-dimension separation step. 2.3. Software
PdQuest (Bio-Rad, Laboratories, Hercules, CA).
3. Methods 3.1. Preparation of Animals
3.2. Protein Sample Preparation
In contamination experiments animals should be acclimated in clean sea water for 7 days prior to the beginning of the experiment (7) (see Note 3). 1. Dissect tissues at 4°C, using a scalpel cooled by dipping periodically into liquid nitrogen. It is convenient to use a pair of thin cotton gloves during these procedures to avoid transferring heat to the samples. 2. Collect tissues in 5-mL screw-capped plastic tubes on liquid nitrogen. 3. When exposed to the cold, the tissues immediately freeze and can subsequently be stored at −70°C or below (see Note 4).
3.3. Total Protein Extraction
Having obtained clean samples, it is very important to mince them effectively. Failure to do so will result in selective extraction of protein, which will distort the results of the experiment (see Note 5). 1. Suspend each sample in three volumes of homogenization buffer and homogenize in 15-mL tubes, using an Ultra-Turrax IKA-Werke homogenizer at 4°C in a cold room, on ice (see Note 6 and Chapter “Difficult Proteins”). 2. Centrifuge the homogenate at 15,000 × g for 2 h at 4°C. 3. Transfer the supernatant into 2.5-mL Eppendorf tubes for subsequent quantification of sample protein (see Subheading 3.6). 4. Prepare aliquots of sample so that the final quantity of protein is 300 mg.
200
Chora, Bebianno, and Roméo
3.4. Precipitation Procedure
Protein precipitation is an optional step in sample preparation for 2DE. Precipitation is generally employed selectively to separate proteins in the sample from contaminating species such as salts, detergents, nucleic acids, lipids, etc. that might otherwise interfere with the 2DE result (8). 1. Suspend each sample in nine volumes of precipitation solution (see Note 7). 2. Precipitate proteins for at least 2 h at −20°C. 3. Pellet proteins by centrifugation at 10,000 × g for 30 min. 4. Wash pellet with cold acetone (see Note 8). 5. Remove residual acetone by air drying.
3.5. Protein Solubilization
1. Resuspend the precipitate in the sample dilution buffer. Use 300 mg in 300 mL, vortex, and leave for 30 min. 2. Centrifuge at 14,000 × g for 10 min (see Note 9).
3.6. Quantification of Sample Protein
Quantification of the proteins concentration was carried out using ovalbumin as a protein standard using the Bradford method (9).
3.7. Isoelectric Focusing
Isoelectric focusing is an electrophoretic method that separates proteins according to their isoelectric points (pI). For this separation wide-range immobilized pH-gradient (IPG) gels, with pH values ranging from 3 to 10, in 18-cm length strips, were used (see Note 10). About 300 mg of total extractable protein can be resolved by IEF on immobiline ® DryStrip, pH 3–10, NL, 18 cm (Pharmacia Biotech).: 1. Pipette the samples (prepared as described in the Subheading3.5) into each strip holder. Distribute the solution evenly over the channel length and remove any large bubbles. 2. Carefully remove the cover foil from the Immobiline DryStrip, starting from the anodic end (+ end). 3. Carefully place the Immobiline DryStrip into the holder channel, gel-side down (see Note 11). 4. Overlay the strip with Immobiline DryStrip Cover Fluid to minimize evaporation and prevent urea crystallization. 5. Apply the pressure blocks on the underside of the cover to ensure that the Immobiline DryStrip gel maintains good contact with the electrodes as the gel swells. 6. Ensure that the strip holders are properly positioned on the Ettan IPGphor II platform. Use the guide marks along the sides of the platform to position each strip holder and check that the pointed end is over the anode (pointing to the back of the unit) and the blunt end is over the cathode. (Please refer
Analysis of Proteins from Marine Molluscs
201
to the Ettan IPGphor II user manual for complete details). Check that both external electrode contacts on the underside of each Strip Holder make metal-to-metal contact with the platform. 7. Close the Ettan IPGphor II lid. Program the first step of Ettan IPGphor II to make an active rehydration during 12 h under low voltage (20 V) at low current (50 mA/IPG strip), at 20°C (see Note 12). 8. Program the other steps of the IEF: 1,000 V, 1 h; 4,000 V, 1 h; 8,000 V, 1 h, and 8,000 V, until 50,000 V (~5 h) (see Note 13). 9. After IEF is complete, proceed to the second-dimension separation immediately or store the Immobiline DryStrip gels at −60°C or below in screw-capped tubes. 3.8. Second-Dimension SDS–PAGE 3.8.1. Equilibrating Immobiline DryStrip Gels
In the second-dimension separation, where proteins are separated by molecular weight, traditional SDS–PAGE is employed. 1. Place the IPG strips in individual tubes, with the support film towards the tube wall. 2. Add 5 mL of SDS equilibration buffer, 100 mg DTT per strip, 0.5 mL EDTA. 3. Cap or seal the tubes with flexible paraffin film and place them on their sides on a rocker for the equilibration process. Equilibrate for 15 min. 4. Pour off buffer from earlier step and add the appropriate volume of SDS equilibration buffer, 250 mg iodoacetamide, 0.5 mL EDTA to each strip. Again cap or seal the tubes with flexible paraffin film and place them on their sides on a rocker for the equilibration process. Equilibrate for an additional 15 min (see Note 14).
3.8.2. Electrophoresis Using a Vertical Electrophoresis System
Cast the 10% polyacrylamide gels using 30% Acrylamide/Bis solution (Bio-Rad Laboratories, Hercules, CA) in 1-mm cassettes (see Note 15). 1. Place the strip with the acidic end to the left, gel surface up onto the protruding edge of the longer glass plate. 2. With a thin plastic ruler, gently push the Immobiline dry strip gel down so that the entire lower edge of the Immobiline DryStrip gel is in contact with the top surface of the gel (see Notes 16 and 17). 3. Seal the Immobiline DryStrip gel in place with the agarose sealing solution (see Note 18). 4. Insert the cassettes into the tank and pour the electrophoresis buffer to the fill line. 5. Close the lid and connect the power leads to the power supply.
202
Chora, Bebianno, and Roméo
6. Electrophoresis is performed at constant current (90 A) in two steps. During the initial migration and stacking period (15 min), the current is 80 V, and during the separation we apply 200 V (see Note 19). 7. Stop electrophoresis when the dye front is approximately 1 mm from the bottom of the gel (see Note 20).
4. Notes 1. This is a stock solution. Just prior to use, add DTT or iodoacetamide, for first and second equilibration, respectively, and the EDTA 0.1 mM. 2. Do not allow the solution to boil over. 3. The animals should have the same size to ensure an equal sample. 4. It is very important that the tissues never defrost between collection and solubilization in electrophoresis buffer in order to avoid proteolysis. 5. Direct grinding of frozen tissues in liquid nitrogen does not result in a sufficiently fine mincing for efficient extraction. 6. Start 30 s with a low speed and two times 15 s in a high speed. In this step the homogenate can be frozen at −70°C. 7. This approach limits proteolysis and other protein modifications. Protease inhibitors were found to be efficient enough to stop the protease activities in the sample preparation process. A TCA–acetone treatment was employed in addition to PMSF where acid and organic solvent denature almost all proteins including proteases. 8. Residual TCA must be removed by extensive washing with acetone because extended exposure to this low pH solution may cause some protein degradation or modification. 9. This step helps to precipitate remaining salts. 10. These varied pH intervals allow fine-tuning of each separation strategy to increase first-dimension loading and resolve a greater number of spots from crowded areas, and they are available with strip lengths of 7, 11, 13, 18, and 24 cm. Choose shorter strips, i.e. up to 13 cm, for fast, cost-effective screening, or when the most abundant proteins are of highest interest. The shortest IPG strips give the fastest results, but their sample load is limited. Use longer strips, i.e. 18- and 24-cm strips, for maximal resolution and loading capacity. Longer strips allow detection of more spots and make it easier
Analysis of Proteins from Marine Molluscs
203
to select and identify the proteins in the map, but require longer times in both the first- and the second-dimension separations. 11. To help coat the entire gel, gently lift and lower the strip and slide it back and forth along the surface of the solution. Be careful not to trap bubbles under the Immobiline dry strip gel. 12. The active rehydration facilitates the entry of high-molecularweight proteins into the strips. 13. Many factors affect the amount of time required for complete focusing, and each specific set of conditions, e.g. sample and rehydration solution composition, Immobiline dry strip gel length, and pH gradient. Ramping the voltage slowly while the sample is entering the IPG strip improve results. 14. Be consistent with the timing of the equilibration steps. 15. The composition of this gel should be selected to resolve proteins in the MW range of interest. Thinner gels stain and destain more quickly and generally give less background staining. Thicker gels have a higher protein capacity. Thicker gels are also less fragile and easier to handle. 16. Ensure that no air bubbles are trapped between the Immobiline drystrip gel and the slab gel surface. 17. The MW marker proteins can be applied to a paper piece; then pick up the application piece with forceps and apply to the top surface of the gel next to one end of the Immobiline DryStrip gel. 18. The agarose sealing solution prevents the Immobiline dry strip gel from moving or floating in the electrophoresis buffer. 19. For these vertical systems, cooling is optional. However, temperature control improves gel-to-gel reproducibility, especially if the ambient temperature of the laboratory fluctuates significantly. For best results, gels should be run at 25°C. 20. After this step silver staining is the most sensitive staining technique (see Chapter “Silver Staining of Proteins in 2DE Gels”). For protein identification we use PDQuest, Bio-Rad, Laboratories, Hercules, CA.
References 1. Edmond, J.M., Measures, C., Magum, B.M., Grant, B., Sclater, F.R., et al. (1979) On the formation of metal-rich deposits at ridge crests. Earth and Planetary Science Letter 46, 19–30 2. Blum, J., and Fridovich, I. (1984) Enzymatic defenses against oxygen toxicity in the hydrothermal vent animals Riftia pachyptila and
Calyptogena magnifica. Archives of Biochemistry and Biophysics 228, 617–620 3. Desbruyères, D., Biscoito, M., Caprais, J.C., Colaço, A., Comtet, T., Crassous, P., et al. (2001) Variations in deep-sea hydrothermal vent communities on the mid-atlantic ridge near the Azores plateau. Deep Sea Research Part I: Oceanographic Research Paper 48, 1325–1346
204
Chora, Bebianno, and Roméo
4. Géret, F., Rainglet, F., and Cosson, R.P. (1998) Comparison between isolation protocols commonly used for the purification of mollusc metallothioneins. Marine Environmental Research 46, 545–550 5. Bebianno, M.J., Company, R., Serafim, A., Camus, L., Cosson, R., and Fiala-Médoni, A. (2005) Antioxidant systems and lipid peroxidation in Bathymodiolus Azoricus from mid Atlantic ridge hydrothermal vent fields. Aquatic Toxicology 75, 354–373 6. Starita-Geribaldi, M., Roux, F., Garin, J., Chevallier, D., Fénichel, P., and Pointis, G., (2003) Development of narrow immobilized pH gradients covering one pH unit for human
seminal plasma proteomic analysis. Proteomics 3, 1611–1619 7. Coelho, M. R., Bebianno, M.J., and Langston, W.J. (2002). Routes of TBT uptake in the clam Ruditapes decussates – Food as vector of TBT uptake. Marine Environmental Research 54, 193–207 8. Bjellqvist, B., Laird, N., McDowell, M., Olsson, I., and Westermeier, R. (1998). 2–D Electrophoresis, Principles and Methods. Amersham-Biosciences 100 pp 9. Bradford, M. (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Analytical Biochemistry 72, 248–254
Chapter 13 Preparation and Analysis of Plastid Proteomes by 2DE Anne von Zychlinski and Wilhelm Gruissem Summary Most proteomic analyses require prefractionation and protein purification strategies to achieve maximal proteome coverage, especially in plants in which cells often have a few highly abundant proteins and substances like polyphenols or secondary metabolites that can have significant impact on proteome coverage. Several methods have been developed to reduce cellular complexity and increase protein dynamic range. One approach is the display of the plant cell proteome on a single two-dimensional gel. Other approaches use fractionation strategies to reduce sample complexity to a subset of functionally related proteins or pathway modules. Here we describe a strategy to separate the proteome of a purified cell organelle using two-dimensional gel electrophoresis (2DE). The proteome of plant chloroplasts and nonphotosynthetic plastids was further fractionated by a differential protein solubilization method that is fully compatible with 2DE. The final protein complement of individual fractions comprised approximately 1,000 different protein species that can be fully resolved and visualized in a single 2DE gel. Key words: Plastid proteomics, Rice, Arabidopsis, Differential solubilization.
1. Introduction A typical plant cell is filled with compounds that make any proteome study extremely challenging. Polyphenols are major components of the cell wall, the vacuole is filled with proteases and secondary metabolites, and specialized organelles such as the chloroplast contain highly abundant proteins involved in photosynthesis (e.g., ribulose 1,5-bisphosphate carboxylase/oxygenase, chlorophyll a/b-binding proteins, ATP synthase). In addition, storage tissues that are often found in roots or seeds contain high amounts of starch or storage proteins such as glutelins (1).
David Sheehan and Raymond Tyther (eds.), Methods in Molecular Biology, Two-Dimensional Electrophoresis Protocols, vol. 519 © Humana Press, a part of Springer Science + Business Media, LLC 2009 DOI: 10.1007/978-1-59745-281-6_13
205
206
Zychlinski and Gruissem
A survey of methods shows that it is not possible to extract, solubilize, purify, and display all plant proteins in one experiment on a single 2DE gel, because of their differences in hydrophobicity, molecular weight, charge, and abundance (see also Chapters “Solubilization of Proteins in 2DE: An Outline,” “Difficult Proteins,” and “Protein Extraction for 2DE”). Complete analysis of the plant proteome requires sample preparation procedures that combine efficient protein solubilization together with reduction of sample complexity by prefractionation of proteins prior to 2DE separation. Protein prefractionation strategies compatible with 2DE have been reported (for review see ref 2), including serial extraction protocols for plant tissues such as phenol extraction (3, summarized in ref. 4), polyethylene glycol extraction (3), and serial detergent extractions (5) that combine protein solubilization and prefractionation in one procedure. However, these prefractionation strategies make use of different physicochemical protein properties, and therefore proteins present in a particular fraction do not necessarily reflect their functional context, which may complicate a quantitative comparison of complete pathway modules. Alternatively, proteome analysis of isolated and purified cell organelles reduces sample complexity to a subset of functionally related proteins or pathway modules. However, even the organelle proteome is highly complex and contains a heterogenous composition of proteins (see also Chapters “Two-Dimensional Electrophoresis (2DE): An Overview,” “Organelle Proteomics,” “Proteomic Analysis of Caenorhabditis elegans,” “Blue Native Gel Electrophoresis (BN-PAGE) Proteomics,” “2DE for Proteome Analysis of Human Metaphase Chromosomes,” and “Microsomal Proteomics”). It is therefore of advantage to combine functional prefractionation methods with fractionation using physicochemical properties of the proteins. Here we describe the strategy we developed for analysis of the proteome of purified plastids using 2DE. Arabidopsis chloroplasts as well as rice chloroplasts and etioplasts were purified to high homogeneity by different density gradient centrifugations. Purified plastids were then subjected to a serial protein extraction procedure that is fully compatible with 2DE. Using this strategy, we reduced the complexity from approximately 15,000 to 30,000 proteins present in different types of plant cells to approximately 3,000 proteins that are predicted to be present in plastids. The final serial extraction separates these 3,000 proteins into three fractions of water-soluble proteins, peripherally associated proteins, and integral membrane proteins. Most of the approximately 1,000 proteins present in each fraction can be resolved and visualized in a single 2DE gel.
Preparation and Analysis of Plastid Proteomes by 2DE
207
2. Materials All chemicals are of highest analytical grade. 2.1. Protein Precipitation
– 100% acetone containing 20mM DTT, store at –20°C. – 10% TCA in acetone containing 20mM DTT, store at –20°C (see Note 1). – 1M DTT stock solution. 1.54g/10mL ddH2O, filter sterilize through a 0.2μm filter unit (Millipore), store 1mL aliquots at –20°C.
2.2. Organelle Isolation 2.2.1. Isolation of Chloroplasts from Arabidopsis
1. Grinding buffer (GB) 5×. 2.25M sorbitol, 250mM HEPES/ KOH, pH 7.8, 12.5mM MgCl2, filter sterilize through 0.2-μm filter unit (Millipore). 2. Isolation buffer (IB) 5×. 1.65M sorbitol, 250mM HEPES/ KOH, pH 7.8, 12.5mM MgCl2, filter sterilize through 0.2μm filter unit (Millipore). • Store GB and IB at 4°C (see Note 2). – PCBF. Dissolve 3g polyethylene glycol (MW8,000) and 1g Ficoll in 100mL Percoll, store at 4°C. – 1M DTT stock solution (see Subheading 2.1). – Percoll step gradient. 10% Percoll: 10mL PCBF, 20mL 5× IB, 0.5mL DTT stock, 69.5mL ddH2O. 30% Percoll: 30mL PCBF, 20mL 5× IB, 0.5mL DTT stock, 49.5mL ddH2O. 60% Percoll: 60mL PCBF, 20mL 5× IB, 0.5mL DTT stock, 19.5mL ddH2O.
2.2.2. Isolation of Etioplasts from Rice
1. 10× salts. 100mM HEPES/KOH, pH 7.8, 20mM EDTA, 20mM MgCl2, 10mM tetrasodiumpyrophosphate, filter sterilize through 0.2μm filter unit (Millipore). Prepare 500mM EDTA and 1M MgCl2 stock solutions and store them at room temperature. Prepare HEPES/KOH, pH 7.8, fresh. Bring 50mL ddH2O to boil in a microwave and dissolve Na4P2O5 in the boiling water. Add the reagents for 10× salts in the following order: HEPES, Na4P2O5, EDTA, and then add MgCl2 drop wise. Otherwise the solution turns cloudy. 2. Etioplast isolation solution (E.I.S.). 1× salts, 600mM sorbitol, filter sterilize through 0.2μm filter unit (Millipore) and store at 4°C. 3. E.I.S. supplemented with 0.2% BSA for the initial grinding step (add 2g BSA fraction V per 1L E.I.S.). 4. Nycodenz stock. 50% Nycodenz in 1× salts, 5mM DTT. Prepare the stock by weighing 25g Nycodenz into a 50mL Falcon tube, add 5mL 10× salts, adjust volume to 50mL, wrap in aluminum foil, and dissolve overnight on a rotating wheel. DTT
208
Zychlinski and Gruissem
is added just before use and the solution is kept at 4°C. The 50% Nycodenz stock solution is diluted with E.I.S. plus 5mM DTT to the required 10, 15, 20, and 25% Nycodenz solutions for the step gradient (i.e., take equal volumes of E.I.S. and 50% Nycodenz stock to get 25% Nycodenz). Prepare all buffers and solutions fresh. 2.3. Differential Protein Solubilization
1. Low-salt buffer. 20mM Tris–HCl, pH 8, 5mM MgCl2, 1mM DTT, 2× EDTA-free protease inhibitor cocktail (Roche). Add one tablet of complete EDTA-free protease inhibitor cocktail per 25mL of low-salt buffer prior to use and keep on ice. Prepare 1M Tris base and 1M Tris–HCl, pH 8, stocks and store at room temperature. 2. Urea buffer. 8M urea, 20mM Tris–HCl, pH 8, 5mM MgCl2, 20mM DTT, 2× protease inhibitor cocktail (Roche), keep at room temperature. 3. Detergent buffer. 2M thiourea, 7M urea, 20mM Tris base, 40mM DTT, 2× complete EDTA-free protease inhibitor cocktail (Roche), keep at room temperature. Make up 100–500mL 9M urea and 7M urea/2M thiourea solutions, stir both solutions for 1h at room temperature with one tablespoon mixed bed ion exchange resin, filter through 3mm Whatman filter paper and store aliquots at –20°C. Only thaw once and never heat above 25°C. Prepare all buffers for differential protein solubilization fresh.
2.4. Determination of Chloroplast Chlorophyll Content 2.5. 2DE (Fig. 1)
1. 80% acetone. 2. Low-salt buffer (see Subheading 2.3). 1. First dimension 2DE solubilization buffer: 40mM Tris base, 7M urea, 2M thiourea (thawed from frozen aliquots; see Subheading 2.3), 2% CHAPS, 0.5% Brij 35, 0.4% carrier ampholytes, 20mM DTT, 2mM TBP, 1× complete EDTA-free protease inhibitor cocktail (dissolve Brij 35 avoiding formation of air bubbles, add TBP last). 2DE rehydration buffer: same as solubilization buffer without Tris base. Equilibration buffer: 6M urea, 2% SDS, 50mM Tris–HCl, pH 8.8, 20% glycerol; step 1: plus 2% DTT; step 2: plus 2.5% iodoacetamide and traces of bromophenol blue (see Note 3). 2. Second dimension • Stock solutions: – Acrylamide stock. 30% acrylamide/bis solution (37.5:1 with 2.6% C). – SDS stock. 10% SDS in ddH2O. – Tris stock. 1.5M Tris–HCl, pH 8.8.
Preparation and Analysis of Plastid Proteomes by 2DE
209
pH 3-10
pH 3-10
b
a transketolase
-ATPase
chaperonine 60 kDa
aminomethyl RBCL transferase (T-protein) malate dehydrogenase
-ATPase
PS II OEC protein ferredoxin-NADP reductase PS II OEC protein
pH 3-10
pH 3-10
d
c conserved hypothetical protein
chloroplast protease
-ATPase
PS II OEC protein chlorophyll a/b-binding protein type III chlorophyll a/bbinding protein CP26
chlorophyll a/b-binding protein
Fig. 1. 2DE maps of serially extracted rice proteins. (a) Soluble proteins. (b) Peripheral proteins. (c) Integral membrane proteins. (d) Crude protein complement of the chloroplast. 2DE was performed in an IPG from pH 3 to 10 followed by SDS PAGE using a linear pore gradient from 10 to 18% T. proteins were detected by in-gel silver staining. For protein identification, excised and tryptically digested spots were analyzed by electrospray ionization ion trap tandem mass spectrometry. The raw data were searched against the rice amino acid sequence database using the SEQUEST algorithm.
– APS stock. 10% ammonium persulfate (prepare fresh by dissolving 0.1g APS in 1mL ddH2O). – Displacing solution. 50mL glycerol, 25mL Tris stock, 25mL H2O, traces of bromophenol blue. – Overlay solution. 25mL Tris Stock, 1mL SDS stock, 74mL H2O. – Agarose solution. 0.5% (w/v) agarose in 1× electrophoresis buffer.
210
Zychlinski and Gruissem
• Gel solutions and electrophoresis buffer: – For 12% homogenous gels (1L). 200mL Tris stock, 400mL acrylamide stock, 187.5mL glycerol, 191mL ddH2O, 20mL SDS stock, 400μL APS stock, 422μL TEMED. Degas Tris/acrylamide/glycerol mixture for 20min in a sonicator bath; then add SDS, TEMED, and APS. – For 10–18% gradient gels (450mL each): – Light solution. 150mL acrylamide stock, 113mL Tris stock, 4.5mL SDS stock, 182mL H2O, 200μL APS, 211μL TEMED. – Heavy solution. 270mL acrylamide stock, 113mL Tris stock, 62mL glycerol, 4.5mL SDS stock, 200μL APS, 211μL TEMED. – 10× electrophoresis buffer (1L). 30.28g Tris base, 144g glycine, 10g SDS. Prepare at least 1L of 10× buffer for one electrophoresis run.
3. Methods 3.1. Plant Growth Conditions
Arabidopsis. Sow A. thaliana seeds directly on soil and grow plants in individual pots under short-day conditions (8h light and 16h dark) in a growth cabinet at 22°C for 6 weeks. Water plants (without fertilizer) as needed and harvest leaves before flowering. Rice. Wash rice (Oryza sativa, japonica cultivar) seeds in 5% sodium hydrochloride solution for 10min, rinse four times with deionized water, and imbibe overnight at 29°C. Use two hundred grams of rice seeds for a single plastid isolation experiment. Transfer the imbibed seeds to trays containing wet Vermiculite supplemented with half-concentrated Murashige & Skoog medium (3:2, vermiculite:medium). Cover the trays with a plastic lid and seal with tape (see Note 4). Grow seedlings at 29°C for 10 days either in the dark, under long-day conditions (16h light and 8h dark), or illuminate seedlings for 2, 4, and 8h prior to isolation of plastids. Cell suspension. Subculture Arabidopsis cell suspension once a week by transferring 5mL of saturated culture into 100mL of fresh medium (Murashi & Skoog, including vitamins supplemented with 3% sucrose and 0.5mg/L naphthalene acetic acid and 0.05mg/L kinetin). Grow cell suspension cells at 21°C either in the dark or in the light.
3.2. Complete Protein Precipitation
1. Harvest 10-day-old rice plants and grind them to a fine powder using a mortar and pestle in liquid nitrogen. Alternatively harvest Arabidopsis cell suspension cells using a suction filter
Preparation and Analysis of Plastid Proteomes by 2DE
211
device and rinse with cold ddH2O before grinding them to a fine powder using a mortar and pestle in liquid nitrogen. 2. Add low-salt buffer and urea buffer during the grinding process. Add about 1mL of buffer per 100μg of plant powder. 3. Transfer samples to 2mL Eppendorf tubes and centrifuge for 30min at 16,000×g to pellet plant debris. Depending on the volume, transfer the supernatant to new 2mL Eppendorf tubes or 15/50mL Falcon tubes. 4. Add four volumes 10% TCA in acetone containing 20mM DTT to the protein samples. Vortex well and precipitate at least for 2h (or overnight) at –20°C. 5. Centrifuge at 16,000×g for 20min at 4°C. Carefully pour off the supernatant. 6. Wash the protein pellet twice with 100% acetone plus 20mM DTT. Vortex during each washing to make sure the pellet is well broken up. Centrifuge each washing at 16,000×g, 4°C for 20min. 7. Discard final washing supernatant. Either air dry or SpeedVac samples until remaining acetone is evaporated. 8. Directly solubilize the protein pellet in 2DE rehydration buffer. Alternatively, directly precipitate the proteins from the ground plant powder by adding four volumes 10% TCA in acetone containing 20mM DTT. Process samples further as described in steps 4–8. 3.3. Subcellular Prefractionation
Isolation of chloroplasts from Arabidopsis using a combination of differential and density gradient centrifugation (6). 1. Cast eight discontinuous Percoll gradients in 30mL Corex tubes. 4mL of 10% Percoll, 15mL of 30% Percoll, 7mL of 60% Percoll. Prepare the gradients very carefully and avoid mixing of the different solutions; the boundary between the different concentrations has to be clearly visible. Move gradients to cold room. 2. Homogenization of plant material. –
Harvest the leaves without roots and weigh the material.
–
Cut the plant material into 1–2cm small pieces.
–
Use ca. 75mL 1× GB per 75g of plant material and homogenize in a mortar using a pestle.
–
Filter the homogenate through two layers of miracloth and repeat the homogenization and filtration at least twice.
–
Refilter the combined filtrate again through two new layers of miracloth.
212
Zychlinski and Gruissem
3. Centrifuge the filtrate at 4,100×g for 20min at 4°C in a Sorvall RC5C centrifuge using a GSA rotor. Pour off supernatant and carefully resuspend the pellets in 2–5mL of 1× IB using a small paint brush. Pipette the suspension 10–15 times through a 10mL plastic pipette to break up aggregates, but avoid air bubbles. 4. Load ∼4mL chloroplast suspension onto one Percoll step gradient and centrifuge the gradients in a Sorvall HB-4 rotor at 8,000×g for 20min at 4°C without breaking to stop the centrifuge. Intact chloroplasts sediment at the interface of the 30–60% Percoll step. The upper band contains mostly mitochondria and broken chloroplasts. 5. Collect the chloroplasts from the 30–60% Percoll step interface into an SS34 centrifuge tube and dilute the chloroplast suspension with at least five volumes of 1× IB. Pellet chloroplasts for 2min at 4,300×g and 4°C. Pour off the supernatant. 6. Resuspend the chloroplast pellet in 1–3mL 1× IB. 7. Load a maximum of 4mL chloroplast suspension onto a new Percoll gradient and proceed exactly as for the first gradient. After the second gradient, wash the chloroplasts twice with a centrifugation for 2min at 4,300×g and 4°C. Pour off the supernatant and store the final pellet at –80°C. Isolation of Plastids from Rice Leaf Tissue. This protocol uses a combination of two consecutive density gradient centrifugation steps with Nycodenz as the density gradient medium and several differential centrifugation steps. Carry out the isolation of rice plastids at 4°C (see Note 5). 1. Harvest rice shoots excluding roots and cut with scissors into 5–10mm length. Homogenize batches of 50g plant material in 500mL of etioplast isolation solution (E.I.S.) supplemented with 0.2% BSA using a Waring blender. Blend 3×5s at low speed followed by 3×5s at maximum speed. The plant material should be well homogenized. Filter the homogenate through two layers of Miracloth. The homogenization step and the filtration are repeated once. Pool the two filtrates and refilter them through four layers of Miracloth. 2. Subsequently, centrifuge the combined filtrate for 4min at 200×g to remove cellular debris. Decant the supernatant carefully into new centrifuge tubes. 3. Recentrifuge the supernatant for 10min at 8,000×g. Pour off supernatant and carefully resuspend the pellets in 2–5mL E.I.S. supplemented with 0.2% BSA using a small paint brush. Suck the suspension 5–10 times through a 10mL plastic pipette to break up aggregates, but avoid air bubbles. The resuspended pellets are now subjected to Nycodenz density gradient centrifugation.
Preparation and Analysis of Plastid Proteomes by 2DE
213
4. Adjust the plastid suspension with the 50% Nycodenz stock solution to a final Nycodenz concentration of 30% and a maximal volume of 20mL (e.g., to 8mL of plastid suspension add 12mL 50% Nycodenz). 5. Transfer 5mL of plastid suspension in 30% Nycodenz into a 30mL Corex tube (No. 8,445) and carefully overlay the suspension with a Nycodenz step gradient of 6mL 25%, 8mL 20%, 6mL 15%, and 3mL 10% Nycodenz in E.S.I. (see Note 6). The interfaces between the different Nycodenz concentrations should be very clear and sharp. Centrifuge the gradient for 45min at 4°C and 8,000×g without a break. 6. After centrifugation two yellow (etioplast) bands (bands 2 and 3) should be visible at the interfaces of the 20–15% and 25–20% Nycodenz steps. Collect bands 2 and 3 from the gradient and pool the two. Dilute the suspension threefold with E.I.S. plus 5mM DTT. 7. Centrifuge the plastid suspension for 5min at 8,000×g to remove remaining Nycodenz. 8. Resuspend the pellet in approximately 20mL E.I.S. per centrifuge tube and centrifuge at 500g for 10min. Pour off the supernatant (∼90% of the pellet will be lost during this and the following step). Resuspend the remaining pellet as before and centrifuge again for 15min at 500×g. 9. Subject the pellet to a second Nycodenz gradient centrifugation using only one gradient. Resuspend the pellet in 1–2mL E.I.S. and add 2–3mL 50% Nycodenz stock, respectively, and perform the second gradient as described in step 5. 10. Keep bands 2 and 3 separate this time. Dilute threefold with E.I.S. and centrifuge at 500×g for 15min. Pour off the supernatant and resuspend the pellets in 1mL E.I.S., transfer them to 2mL Eppendorf tubes, and centrifuge for 5min at 5,000×g. 11. Drain the supernatant and store the plastid pellets at –80°C. Although bands 2 and 3 contain intact etioplasts that do not differ in their protein pattern as judged by 2DE (7), they were further analyzed separately because only band 2 plastids were visible after illumination for 2, 4, and 8h. The isolation of rice chloroplasts follows the same procedure with modifications at steps 6, 8, and 10 as follows: 12. After the gradient centrifugation two major green bands are visible at the interface of 20–25% and 25–30% Nycodenz. Harvest and combine both bands. 13. Centrifuge at 3,000×g for 5min. Pour off supernatant and repeat centrifugation step once.
214
Zychlinski and Gruissem
14. Keep bands 3 and 4 separate at this time. Dilute threefold with E.I.S., centrifuge for 5min at 3,000×g, and pour off supernatant. 3.4. Organelle Lysis and Serial Protein Extraction
Chloroplasts and etioplasts are enclosed by a double membrane that can be disrupted by osmotic shock treatment and shear forces (see Note 7).
3.4.1. Lysis of Chloroplasts in Low-Salt Buffer
Perform all steps for the first extraction of soluble proteins at 4°C. Thaw isolated chloroplasts on ice and resuspend in ∼0.5mL of ice-cold low-salt buffer, which osmotically disrupts the plastid membranes. The 0.5mL resuspension volume corresponds to the amount of plant material used in Subheading 3.3 and has to be adjusted for other amounts of starting material. For less material the volume of low-salt buffer has to be reduced accordingly. The volume of the suspension is normalized to the chlorophyll concentration.
3.4.2. Determination of Chlorophyll Concentration
1. Pipette 1, 2, and 10μL of the chloroplast suspension into 999, 998, and 990μL of 80% acetone in water, respectively, and vortex. 2. Measure the OD of the three solutions at wavelengths of 645nm and 663nm against 1mL 80% acetone. Calculate the chlorophyll concentration by the following formula: μg Chlorophyll/mL=(A633×8.02+A645×20.2)×dilution factor (i.e., 1,000, 500, or 100). 3. Adjust the final volume of the organelle suspension to a chlorophyll concentration of maximum 1mg/mL with ice-cold low-salt buffer.
3.4.3. Extraction of WaterSoluble Proteins
1. To completely disrupt chloroplasts and extract soluble proteins, pass the suspension at least ten times through a 0.7μm needle using a 10mL syringe and incubate on ice for 10min. 2. Repeat the procedure at least twice, resulting in a total incubation time of 30min. 3. Monitor chloroplast lysis by microscopic inspection of the suspension. If a high amount of chloroplasts appears to be still intact, repeat the lysis procedure until most of the organelles are disrupted. Alternatively, increase the volume of the lowsalt extraction buffer. However, it should be considered that the final protein concentration will be reduced by a larger volume of extraction buffer. We observe that alternate freezing and thawing cycles as an alternative method for the disruption of the chloroplasts result in a different protein pattern during the first two solubilization steps. A higher amount of
Preparation and Analysis of Plastid Proteomes by 2DE
215
peripheral membrane proteins overlapped with the fraction of soluble proteins. 4. Centrifuge the final suspension for 20 min at 16,000×g and 4°C. Further centrifuge the supernatant for 30min at 50,000×g and 4°C to remove all insoluble material. 5. Store the final supernatant (sample 1) containing water-soluble proteins at –80°C. 3.4.4. Extraction of Peripheral Membrane Proteins
1. Wash the pellet of the first centrifugation step containing the water-insoluble proteins with 1mL low-salt buffer. 2. Resuspend the pellet completely and vortex several times. 3. Remove the buffer by centrifugation for 20min at 16,000×g and 4°C. 4. Repeat the washing procedure twice to remove all soluble protein remnants. Perform the following extractions of the peripheral and integral membrane proteins at room temperature to improve the solubilization and to avoid the precipitation of urea. 5. Resuspend the washed pellet in urea buffer, which solubilizes peripheral membrane proteins. The volume of urea buffer should be the same as the final volume of low-salt buffer used in the first extraction step after the adjustment to 1mg chlorophyll per mL buffer. 6. Vortex the suspension several times. It can also be passed through a 0.7μm needle to resuspend the membranes more easily. Avoid loss of material due to adhesion of membrane material to the wall of tubes, pipette tips, and the syringe. Incubate the suspension for 5min at room temperature. 7. Repeat resuspension of the membrane material twice. The final suspension is centrifuged for 20min at 16,000×g and 15°C. 8. Further centrifuge the supernatant for 30min at 50,000×g and 15°C to remove all insoluble material. 9. Store the final supernatant (sample 2), containing mostly peripheral membrane proteins at –80°C.
3.4.5. Extraction of Integral Membrane Proteins
1. Wash the pellet of the first centrifugation step (which contains the water-insoluble proteins) three times in urea buffer according to the procedure described in Subheading 3.4.4. 2. Solubilize integral membrane proteins in detergent buffer according to the procedure described for the solubilization of peripheral membrane proteins in Subheading 3.4.4. 3. Store the final sample containing the integral membrane proteins at –80°C.
216
Zychlinski and Gruissem
3.4.6. Further Extraction of Membrane Proteins
If the analysis is focused on the separation of membrane proteins it is recommended to continue with a serial detergent extraction to further fractionate the membrane proteins. Therefore the pellet remaining after the extraction of peripheral membrane proteins (see Subheading 3.4.4) can be solubilized in a weak detergent buffer containing max 1% CHAPS and no TBP. In a following extraction step, either a higher detergent concentration or stronger detergents or a combination of two compatible detergents can be used in combination with TBP to extract the remaining integral membrane proteins. Although we observed a strong overlap between the two fractions of integral membrane proteins some protein spots were better resolved and a higher number of protein spots could be detected. Combination of maximum 2% ASB-14 (or maximum 3% CHAPS) with 0.5% Brij 35 in 2DE rehydration buffer revealed an improved solubilization of membrane proteins and a reasonable resolution in 2DE (see Note 8). Brij 35 showed a much better protein spot resolution and less detergent smear than other nonionic detergents like Triton X-100 and NP40.
3.5. 2-Dimensional Electrophoresis (2DE) (Fig. 1)
The following procedure was specifically developed for the Ettan IPGphor and Ettan DALT 12 system (GE Healthcare) and has to be adapted to other systems. It is critical that all equipment is carefully cleaned with nonionic detergents (e.g., IPG strip holder cleaning solution, GE Healthcare) and rinsed well with deionized water.
3.5.1. First-Dimension Electrophoresis
1. Use 24cm Biorad IPG strips (pH gradient 3–10 or 4–7). 2. Resuspend the protein pellets in ∼200μL 2D solubilization buffer and determine the protein concentration. 3. Use a total volume of 420μL to rehydrate the immobilized pH gradient (IPG) strips and 150μg of protein sample. The protein concentration of the samples should be between 1 and 5μg/μL. For example, if the protein sample has a concentration of 1μg/μL, mix 150μL sample with 270μL rehydration buffer. 4. Prewet the rehydration tray with rehydration buffer to coat the surface and carefully drain the buffer. This helps to apply the sample without air bubbles evenly into the tray. Apply the 420μL protein sample to the rehydration tray. Remove the coating foil from the IPG strip only now and slowly place the strip on top of the sample. 5. Add 500μL paraffin oil on each end of the IPG strip. Make sure the surface of the IPG strip is completely covered with oil (otherwise add more). 6. Rehydrate the IPG strips overnight for a maximum of 16h at room temperature. Shield the rehydration unit from direct light by placing carton paper on top of it.
Preparation and Analysis of Plastid Proteomes by 2DE
217
7. Next morning: Place two layers of white printer paper in front of you. Wet 3-MM Whatman paper and place on the printer paper. 8. Remove the rehydrated IPG strip from the rehydration tray and lay it with the foil side on the prewetted Whatman paper. Drain off the excess oil and blot the gel side gently with wet Whatman paper. 9. Cut 3-MM Whatman paper into 3×5mm electrode pads. Prewet them and remove access water by laying the pads onto dry Whatman paper; electrode pads must be damp, but not really wet. 10. Deliver 40μL kerosene into each IPGphor cup-loading strip holder (the surface has to be just covered). 11. Place rehydrated IPG strips plastic-backing down into the IPGphor cup-loading strip holders and place one prewetted electrode pad on the cathodic and anodic ends of the IPG strip. Mount the electrodes and gently press them down onto the electrode pads. Overlay the IPG strips with ca. 3mL silicone oil (or cover fluid). 12. Close the lid and start the isoelectric focusing (IEF) run. 13. IEF running conditions: – Limit maximum 50μA per strip. – Set temperature: 20°C. – 500V gradient 1h; 500V step-n-hold 1h; 1,000V gradient 1h; 2,000V gradient 1:30h; 2,000V step-n-hold 1h; 4,000V gradient 30min; 4,000V step-n-hold 6h; 8,000V gradient 30min; 8,000V step-n-hold until 80–100kVh. – Change electrode pads after second, fourth, and sixth steps. 14. After the IEF run is completed store focused strips between transparent sheet protectors at –80°C. 15. Equilibration of focused IPG strips prior to SDS-PAGE. Incubate the IPG strips for 10min each in equilibration buffer steps 1 and 2. 3.5.2. Second-Dimension Electrophoresis
1. Cast gels 1 day before use (either 12% homogenous or 10–18% gradient gels). Gels are poured using a gradient pump. When pouring homogenous gels, only fill the front chamber of the gradient maker with gel solution. All other steps remain the same. 2. Assemble 14 gel cassettes (1mm thick) into the Ettan Dalt gel caster. Separate each cassette with a separator sheet and tight the gel caster firmly. Place a blue pipette tip with cut off end into the side chamber. Make sure the tubing of the gradient mixer fits into the pipette tip.
218
Zychlinski and Gruissem
3. Immediately before casting the gels, add TEMED and fresh APS to the gel solution. 4. Pour the heavy solution into the gradient mixer, making sure that the connection tube between the heavy and the light chamber is closed. Start the gradient pump and hold the pump tubing into the heavy solution to remove any air from the tubing. Turn the pump off while the tubing is still in contact with the gel solution. Now fill the light solution into the chamber and open the connection tube between the chambers and place the pump tubing into the blue tip in the side chamber. Turn the pump back on. 5. While the gels are poured fill the side chamber with 80mL displacing solution. 6. Stop the pump when the gel cassettes are filled to approximately 5–6cm from the top end. Remove the blue tip and allow the displacing solution to enter the gel caster. Stop the displacing solution when the gels are filled to 0.5cm from the top end. (Ideally this step is not necessary and the displacing solution is used up when the gels are at the correct level.) Carefully pipette 1mL overlay solution onto each gel. 7. Allow the gels to polymerize for at least 4h at room temperature. We prefer to keep the gels overnight in the cold room before use. Therefore generously pour overlay solution onto the gels after polymerization and cover the gel caster tightly with Saran wrap. 8. Pour 7.5L of 1× electrophoresis buffer into the Ettan Dalt II electrophoresis tank, turn on the pump and cooling to 20°C. Disassemble gel caster and rinse the glass plates with deionized water. Drain overlay solution. 9. Remove excess equilibration solution from the IPG strips by briefly rinsing them in electrophoresis buffer and mount onto gel. Slide them onto the gel surface using a small plastic ruler and secure the strips with 2mL hot (50–60°C) agarose solution. Let the agarose settle for at least 10min before inserting the gels into the gel tank. 10. Dip the gel cassettes into electrophoresis buffer and slide them into the gel tank. If necessary, fill unoccupied slots with blank cassettes. Pour 2.5L 1× electrode buffer into the upper buffer chamber onto the gels. 11. Run the gels at 1W per gel for 45min, and then increase up to 17W per gel until dye front is 1mm from the bottom of the gels. 12. After electrophoresis, remove gel cassettes from the tank and carefully open them using a plastic spatula. Cut off the agarose overlay with a plastic ruler or separator sheet; peel the gels off
Preparation and Analysis of Plastid Proteomes by 2DE
219
the glass plates. When handling gradient gels try to handle them at the bottom end as this reduces the risk of damage. Then place them into the appropriate fixing solution. 13. Stain gels either with SyproRuby or silver stain (Chapter “Silver Staining of Proteins in 2DE Gels”). For both staining methods use 250mL solution per gel and keep gels in separate plastic containers during the staining procedure. After staining, spots are visualized with a white-light flat bed scanner for silver-stained gels or a Typhoon 9,400 scanner at 488nm for SyproRuby-stained gels.
4. Notes 1. Acetone solutions should be stored in glass bottles. 2. Filter IB first so you can use the same filter unit for GB. Best to prepare IB and GB a day before use and store at 4°C. 3. Make up 500–1,000mL equilibration buffer without DTT or iodacetamide and store in 50mL Falcon tubes at –20°C. 4. Seeds are “sprinkled” on top of the vermiculite/water mix; try to have seeds individually as this will result in healthier plant growth and less contamination. 5. Rice plastids have to be isolated using Nycodenz as a gradient medium because they highly aggregate in Percoll gradients. 6. It does not work if you lay the plastid solution on top of precast Nycodenz gradients. 7. Start the serial extraction early; it takes a whole day. 8. Do not mix CHAPS with ASB-14. In our hands this combination resulted in detergent smears and was not usable for subsequent 2DE. References 1. Chen, S., and Harmon, A.C. (2006) Advances in plant proteomics. Proteomics 6, 5504–5516. 2. Agrawal, G.K., Jwa, N.-S., Iwahashi, Y., Yonekura, M., Iwahashi, H., and Rakwal, R. (2006) Rejuvenating rice proteomics: Facts, challenges and visions. Proteomics 6, 5549–5576. 3. Kim, S.T., Cho, K.S., Jang, Y.S., and Kang, K.Y. (2001) Two-dimensional electrophoretic
analysis of rice proteins by polyethylene glycol fractionation for protein arrays. Electrophoresis 22, 2103–2109. 4. Natarajan, S., Xu, C.P., Caperna, T.J., and Garrett, W.A. (2005) Comparison of protein solubilization methods suitable for proteomic analysis of soybean seed proteins. Anal. Biochem. 342, 214–220. 5. Luche, S., Santoni, V., and Rabilloud, T. (2003) Evaluation of nonionic and zwitterionic
220
Zychlinski and Gruissem
detergents as membrane protein solubilizers in two-dimensional electrophoresis. Proteomics 3, 249–253. 6. Kleffmann , T. , Russenberger, D. , von Zychlinsky, A. , Christopher, W. , Sjolander, K. , Gruissem , W. , and Baginsky, S. (2004) The Arabidopsis thaliana chloroplast proteome reveals pathway abundance and
novel protein functions . Curr. Biol . 14 , 354 – 7 . 7. Von Zychlinski, A., Kleffman, T., Krishnamurty, N., Sjölander, K., Baginsky, and Gruissem, W. (2005) Proteome analysis of the rice etioplast: metabolic and regulatory networks and novel protein functions. Mol. Cell. Proteomics 4, 1072–1084.
Chapter 14 High-Resolution 2DE Katrin Marcus, Cornelia Joppich, Caroline May, Kathy Pfeiffer, Barbara Sitek, Helmut Meyer, and Kai Stuehler Summary About 30 years ago two-dimensional gel electrophoresis (2DE) was developed independently by Klose and O’Farrell representing the combination of two orthogonal separation techniques. In the first dimension the proteins are separated by isoelectric focusing (IEF) according to their isoelectric point. In the second dimension proteins are separated according to their electrophoretic mobility by conventional SDS-PAGE. For IEF two different techniques, immobilized pH gradient (IPG) and carrier-ampholytebased IEF (CA-based IEF), respectively, are currently applied. With a resolution of up to 10,000 protein spots in one gel, 2DE offers a huge potential to give a comprehensive overview of the proteins present in the examined system. In combination with image analysis and mass spectrometry 2DE is still the method of choice to analyse complex protein samples. In this chapter we provide detailed protocols for both 2DE systems and give an overview about the latest developments including the two-dimensional difference gel electrophoresis (DIGE) system. Key words: 2DE, Carrier-ampholyte system, Immobilized pH gradient, IPG-system, DIGE.
1. Introduction In the mid-1990s Mark Wilkins and Keith Williams first introduced the term proteome (1) as the protein complement of a genome in a given cell, tissue, or organism at specific time points and under certain conditions. The understanding of protein networks and their quantitative changes in a global context strongly depends on the capability simultaneously to analyse proteins and to quantify variations in protein expression. A number of different methods have been established over the years allowing the analysis of complex protein mixtures including two-dimensional electrophoresis (2DE), multi-dimensional protein identification David Sheehan and Raymond Tyther (eds.), Methods in Molecular Biology, Two-Dimensional Electrophoresis Protocols, vol. 519 © Humana Press, a part of Springer Science + Business Media, LLC 2009 DOI: 10.1007/978-1-59745-281-6_14
221
222
Marcus et al.
technologies, stable isotope labelling, and protein arrays (2–6). 2DE – a combination of isoelectric focusing (IEF) in the first dimension and SDS-PAGE in the second dimension – represents the only technique with a satisfactory separation of complex protein mixtures available at the moment. There are two different methods of separation techniques in the first dimension: the method of Klose (7) and O’Farrell (8) where the pH gradient is formed via carrier ampholytes (CA) (amphoteric, oligoaminooligocarbonic acids with high buffer capacity at their pI) during the focusing process and the method described by Bjellqvist and Görg (9–11) using IPG in the first dimension. Due to its simple handling and commercialization IPG-based IEF is typically used for 2DE-based proteome analysis allowing widespread applications. Meanwhile a number of different IPG strips with variations in length (7–24 cm) and pH range (narrow or broad, e.g. pH 4–5 or 3–11; linear or non-linear) are provided by different manufacturers (see also Chapter “Selection of pH Ranges in 2DE”). Additionally, IPG-based 2DE facilitates the application of high sample amounts (up to 2 mg of protein on a 24-cm strip) for sensitive protein detection and quantification. In contrast, CA-based IEF as a labour-intensive technique failed to achieve widespread application, but is still used in more specialized laboratories. CAbased IEF provides the advantage of higher resolution as it is possible to use setups with gel dimension of up to 30 × 40 cm (7). For protein visualization several staining methods differing in sensitivity and linear dynamic detection ranges exist (see Table 1; (12–14); also Chapter “Silver Staining of Proteins in 2DE Gels”). Most popular staining methods are summarized in Table 1. In differential proteome analysis the protein pattern of two different samples (e.g. different cell states, transgenic vs. wild type mice, diseased human tissue vs. control tissue) is compared with the help of automated image analysis tools, e.g. Image Master, PDQuest®, ProteomWeaver®, Delta 2D® (Chapter “Troubleshooting Image Analysis in 2DE”). Protein spots with altered expression levels are excised from the gel, proteolytically digested, analysed by mass spectrometry (MS) followed by automated protein identification (Chapters “Shotgun Protein Analysis by Liquid Chromatography-Tandem Mass Spectrometry,” “De Novo Sequence Analysis of N-Terminal Sulfonated Peptides After inGel Guanidination,” “Tryptic Digestion of in-Gel Proteins for Mass Spectrometry Analysis,” and “Database Interrogation Algorithms for Identification of Proteins in Proteomic Separations”). In the original setup the lysate of each sample is separated in an extra gel followed by the aforementioned workflow (Fig. 1a). Small alterations in gel casting and/or running conditions result in reproducibility alterations complicating subsequent image analysis. An important improvement for 2DE-based proteome analysis was the introduction of two-dimensional difference in-gel
3–5 ng <1 ng 1 ng
Silver (MS-compatible)
Silver
SyproRuby « 1 ng 0.1–0.2 ng 0.005–0.01 ng
Phosphoimaging ( P, C, S)
DIGE, minimal
DIGE, saturation
35
8–10 ng
Colloidal Coomassie
14
10 ng
Imidazole-zinc
32
Detection limit
Staining method
3–5 orders of magnitude
3–5 orders of magnitude
Five orders of magnitude
Three orders of magnitude
Two orders of magnitude
Two orders of magnitude
Three orders of magnitude
No quantitation, negative stain
Linear dynamic range
(21)
(20)
(19, 20)
(17–19)
Chapter “Silver Staining of Proteins in 2DE Gels” (16)
(15)
Chapters “Difficult Proteins,” “Analysis of Bacterial Proteins by 2DE,” and “Proteomic Analysis of Caenorhabditis elegans” (13, 14)
(12)
Reference
Table 1 Most commonly used protein staining methods for proteome analysis. Sensitivity as well as the linear dynamic range are important parameters in protein quantification and differential description of a proteome
High-Resolution 2DE 223
224
Marcus et al.
a
2-DE based Differential proteome analysis Sample A
Sample B
b
Three colour labelling: CyDye DIGE minimal fluors
Sample A Sample B Int. Standard Cy TM3
2-DE, Visualization, Image acquisition
Cy TM5
Cy TM2
Two colour labelling: CyDye DIGE saturation fluors Sample A
Int. Standard
Cy TM3
Cy TM5
2D DIGE
2D DIGE
Image Acquisition
Image Acquisition
Differential gel analysis
Differential In-gel analysis
Fig. 1. 2DE-based differential proteome analysis: In differential proteome analysis the protein pattern of two different samples (e.g. different cell states) are compared with the help of automated image analysis tools (see also Chapter “Troubleshooting Image Analysis in 2DE”). (a) In the original setup the lysate of each sample is separated in an extra gel followed by the earlier-described workflow. Small alterations in the gel casting and/or running conditions result in reproducibility alterations complicating subsequent image analysis. (b) By the introduction of the DIGE system in 1997 (22) technical variances are minimized and accurate quantification of difference in protein amounts between samples with high statistical significance is possible. In this technique, proteins of different samples are covalently labelled with spectrally resolvable fluorescent dyes (CyDyes™, Cy™2, Cy™3, Cy™5) prior to their simultaneous separation in one gel. Different forms of CyDyes™ are commercially available (GE Healthcare, Munich, Germany): the CyDye™ minimal dyes for general differential analysis allowing consideration of three different CyDyes™ in a multiplexing experiment and the CyDye™ saturation dyes for differential analysis of scarce samples applying two different fluorescent dyes.
electrophoresis (DIGE) system in 1997 (22; Chapter “TwoDimensional Difference Gel Electrophoresis”). In this technique proteins are covalently labelled with spectrally resolvable fluorescent dyes (CyDyes™, Cy™2, Cy™3, Cy™5) prior to separation (Fig. 1b). Two different samples can be labelled with individual CyDyes™ and separated in one gel resulting in minimal technical variances that may develop otherwise due to gel-to-gel variation. The different dyes do not affect the relative mobility of proteins during electrophoresis. Thus, a given protein labelled with any of the CyDye™ fluors will migrate to the same position in the 2DE gel. Visualization and digitalization of the gel images are achieved using a fluorescence laser scanner. The key benefit of DIGE technology is the introduction of an internal standard (preferably a mixture of all samples used in the differential study). The internal standard is used to match the protein patterns across gels minimizing quantification errors. In consequence, accurate quantification of differences in protein amounts between samples
High-Resolution 2DE
225
with a high statistical significance is possible. After separation and visualization, proteins of interest (proteins detected to be differentially expressed) are excised from the gels, proteolytically digested, and analysed as described earlier. Different forms of CyDyes™ are commercially available (GE Healthcare, Munich, Germany): • The CyDye™ minimal dyes for general differential analysis. • The CyDye™ saturation dyes for differential analysis of scarce samples. CyDye™ minimal dyes react with the ε-amino groups of lysine residues of the proteins via an NHS-ester bond adding about 450 Da to the protein mass. In this approach approximately 3% of all proteins present are labelled on only a single lysine residue (‘minimal labelling’). The rest remains unlabelled. This makes this technique very robust and further tedious optimization and adaptation of the labelling procedure usually is not necessary. Three different dyes are available enabling the co-separation of three samples in one gel (generally two samples plus internal standard). The pooled internal standard is a mixture of equal quantities of all samples analysed within one study. The simultaneous separation of the sample together with internal standard in each gel allows an exact calculation of ratios for the same protein spot within one gel as well as between different gels. Additionally, the number of gels run within one study is reduced by half (Fig. 2).
Fig. 2. Combination of IPG- and large-gel system – DIGE gel of mouse hippocampus protein lysate. The protein lysate of wild-type mice was labelled with Cy™3 and the protein lysate of the transgenic mice was labelled with Cy™5. The internal standard was a mixture of all samples included in the study (transgenic and wild type) and was labelled with Cy™2. All three samples were separated in one gel. Proteins were separated using IPG strips (24 cm, pH 3–11, non-linear) in the first dimension followed by large-gel SDS PAGE according to Klose (7) (Chapter “High-Resolution Large Gel 2DE”) in order to improve resolution.
226
Marcus et al.
CyDye™ saturation dyes enable a comprehensive differential analysis of scarce sample amounts down to 2 μg total protein/ gel (23, 24). They react with thiol groups of cysteine residues in the protein sample via a maleimide group. All proteins are labelled completely resulting in a high labelling concentration (‘saturation labelling’). The sensitivity of this method is about 20–50 fold higher than with normal silver staining (Fig. 3). Indeed, as there are only two different CyDye™ saturation dyes available, every sample has to be separated in an individual gel, but the simultaneous separation of the internal standard results in accurate quantification. In contrast to the minimal labelling procedure, a labelling optimization is essential for every experiment/study in order to determine the appropriate amount of dye (Fig. 4). After separation the protein patterns are digitalized using confocal fluorescent imagers. For each dye a gel image is generated at a specific wavelength without any crosstalk. Appropriate analysis software then applies co-detection algorithms permitting automated detection, background subtraction, quantification, normalization, as well as inter-gel matching.
a
b
MW
MW
116 97
116 97
55
55
36
36
21 14
21 14
4
pI
9
4
pI
9
Fig. 3. Sensitivity of CyDye™ saturation labelling and silver staining. 2 μg of a protein lysate was separated by CAbased 2DE and visualized by either (a) silver staining according to Heukeshoven & Dernick (16) or (b) fluorescence detection after CyDye™ saturation labelling using the Typhoon fluorescence imager (GE Healthcare). Only a few spots were detected after classical silver staining, whereas 50–fold increase of sensitivity could be observed by application of the CyDye™ saturation labelling followed by fluorescence detection. First the introduction of the CyDye™ saturation labelling enables the differential display of scarce samples. For more details and applications please refer to (21, 23–28).
High-Resolution 2DE
227
a Lack of dye => vertical shift (shift in molecular weight), insufficient labelling of cystein-residues
b Excess of dye => horizontal shift (pI shift), besides cystein-residues additionally lysine-residues are labelled
c Optimal ratio between protein amount and dye concentration => exact overlap of both dyes, exclusive labelling of cystein-residues
Fig. 4. Optimization of CyDye™ saturation labelling. The CyDye™ saturation labelling procedure cannot be applied according to a generalized protocol making a labelling optimization in a 2DE titration experiment obligatory to determine the optimal TCEP/dye: protein ratio (25). Assuming an average cysteine content of 2%, 2.5 μg protein lysate requires 1 nmol TCEP and 2 nmol dye for efficient labelling. (a) If the amount of dye is too low, some cysteine thiol groups may not be labelled resulting in vertical (molecular weight) shifts in the gel. (b) Excess of dye may cause non-specific labelling of for instance the ε-amino groups of lysine-residues and a pI charge shift can be observed. (c) Only if the ratio of dye: protein is optimal an exact overlap of both dyes is obtained.
2. Materials 2.1. Equipment (see Notes 1 and 2)
1. Horizontal electrophoresis apparatus.
2.1.1. IPG-Based 2DE
3. Thermostatic circulator.
2. Power supply. 4. Immobiline DryStrips (see Note 3). 5. Reswelling cassette. 6. Electrode strips. 7. Sample cups and sample cup holders. 8. Glass plates and spacer. 9. Grease (e.g. Roth, Life Sciences, Karlsruhe, Germany). 10. Multi-casting chamber. 11. Vertical SDS-PAGE unit. 12. Gel dryer.
2.1.2. CA-Based 2DE
1. Vertical electrophoresis apparatus for first dimension (handmade, essentially as described in ref. 29). 2. Power supply.
228
Marcus et al.
3. Thermostatic circulator. 4. Glass plates and spacer. 5. Apparatus for SDS-PAGE: Desaphor VA300 (Desaga, Heidelberg, Germany). 6. Gel carrier grooves (handmade essentially as described in ref 29). 7. Gel dryer.(see Note 4) 2.1.3. DIGE: CyDye™ Minimal Labelling/CyDye™ Saturation Labelling
Both DIGE technologies can be easily combined with both earlier-described 2DE technologies. The following specialized equipment is additionally needed: 1. Glass plates compatible with fluorescence imaging. 2. Confocal fluorescence scanner with appropriate excitation wavelength length (488 nm, 532 nm, 633 nm) and emission filters.
2.2. Buffers and Solutions 2.2.1. IPG-Based 2DE (See Note 5) Isoelectric Focusing SDS-PAGE
Rehydration solution. 8 M urea, 2 M Thiourea, 2% CHAPS, 2% Servalyte, 100 mM dithiothreitol (DTT; freshly added, store at −20°C as a stock solution of e.g. 2 M) (see Notes 6 and 7). 1. Equilibration solution. 6 M urea, 2% sodium dodecyl sulphate (SDS), 30% glycerol, 3.3% separation gel buffer, 65 mM DTT, 280 mM iodoacetamide. 2. Running buffer (pH 8.8). 24 mM Tris–HCl, 0.1% SDS, 0.2 M glycine. 3. Separation gel buffer (pH 8.8). 1.5 M Tris–HCl, 0.4% SDS. 4. Agarose solution. 0.4% agarose (in running buffer), 0.01% bromophenol blue. 5. Gel solution (for one 11%-gel 25 × 20 cm). 25 mL separation buffer, 29 mL deionized water, 31 mL Acrylamide (24:1, 30% T, 2.6% C), 55 μL ammonium persulphate solution (APS; 40%), 46 μL TEMED. 6. Cover solution. Water-saturated isopropanol or 0.1% SDS.
2.2.2. CA-Based 2DE Isoelectric Focusing
1. Separation gel. 3.5% acrylamide, 0.3% piperazindiacrylamide, 4% CA mixture (pH 2–11), 9 M urea, 5% glycerol, 0.06% TEMED. 2. Cap gel. 12.3% acrylamide, 0.13% piperazindiacrylamide, 4% CA mixture, 9 M urea, 5% glycerol, 0.06% TEMED. 3. Anodic solution. 3 M urea, 4.3% phosphoric acid (0.87 M). 4. Cathodic solution. 9 M urea, 5% glycerol, 0.75 M ethylenediamine 5. Sephadex solution. 270 mg Sephadex suspension (20 g Sephadex is swollen in 500 mL water, resuspended into 1 L of 25% glycerol solution, and filtered), 233 mg urea, 98 mg thiourea (final concentration 7 M urea, 2 M thiourea), 25 μL ampholine mixture, 25 μL DTT (1.08 g/5 mL).
High-Resolution 2DE
229
6. Protection solution. 30% urea, 5% glycerol, 2% Servalyte 2–4. (4.75 M urea) 7. Equilibration solution. Tris-Base pH 6.8, 40% glycerol, 65 mM DTT, 3% SDS. SDS-PAGE
1. Gel solution. 570 mM Tris-Base, 180 mM Tris–HCl (pH not adjusted), 0.03% TEMED, 0.1% SDS, 15% acrylamide, 0.2% bisacrylamide (approximately 77 mL per gel), 144 μL 40% APS. 2. Running buffer. 25 mM Tris-Base (pH not adjusted), 190 mM glycine, 0.01% SDS. 3. Protection solution. 570 mM Tris-Base, 180 mM Tris–HCl (pH not adjusted), 0.1% SDS. 4. Agarose solution. 1% agarose (dissolved in running buffer), 0.01% bromophenol blue.
2.2.3. DIGE: CyDye™ Minimal Labelling
1. CyDye™ DIGE minimal fluors (Cy™2, Cy™3, Cy™5). 2. Lysis buffer (pH 8.5). 30 mM Tris–HCl, 7 M urea, 2 M Thiourea, 4% CHAPS. 3. Anhydrous dimethylformamide (DMF). 4. Stopping solution. 10 mM lysine. 5. Optional. Lysis buffer (pH 8.5) (2×): 8 M urea, 4% CHAPS, 2% CA, 2% DTT.
2.2.4. CyDye™ Saturation Labelling
1. CyDye™ DIGE saturation fluors (Cy™3, Cy™5). 2. Lysis buffer (pH 8.0). 30 mM Tris–HCl, 7 M urea, 2 M Thiourea, 4% CHAPS. 3. Reduction solution. 2 mM TCEP (Tris-(carboxyethyl) phosphine hydrochloride) for analytical gels and 20 mM TCEP solution for preparative gels. 4. Anhydrous dimethylformamide (DMF). 5. Lysis buffer (pH 8.5) (2×). 8 M urea, 4% CHAPS, 2% CA, 2% DTT.
3. Methods 3.1. IPG-Based 2DE
These instructions are modified from the 2DE technique as described by Görg et al. (30–32). It can be easily adapted to the DIGE technique (see later) and other formats, including analytical and preparative gels.
3.1.1. Isoelectric Focusing
Sample application is achieved either via sample cups or via in-gel rehydration (see Note 8). Here we describe the sample cup application procedure:
230
Marcus et al.
1. Prior to IEF rehydrate the IPG gel strips overnight (at least 8 h) at RT in the rehydration solution (for 18 cm strips use a volume of 350 μL; for 24 cm strips use 550 μL). 2. Rinse the rehydrated strips with deionized water for a few seconds and carefully remove excess water with filter paper. 3. Place the strips, gel-side up, in the aligned tray. The basic ends of the strip must be faced towards the cathode. 4. Place two electrode strips (or filter paper, respectively) soaked with water at the ends of the strips, one at the basic and one at the acidic end. 5. Position the electrodes and press them gently on the top of the electrode strips. 6. Apply sample cup holder on top of the strips (basic or acidic end; see Note 9) depending on the pH gradient used. 7. Fix sample cups at the sample cup holders and pipette the sample into the sample cups (maximum 150 μL sample) (see Note 10). 8. Place the tray on the cooling plate. 9. Cover the strips and the samples with silicone oil. 10. Carry out IEF. For improved sample entry, the voltage at the beginning of the gradient is limited to 150 V. Running conditions depend on the used IPG strip. The maximum current should be restricted to 50–75 μA/strip and the optimum focusing temperature is 20°C. 11. Strips of different pH gradients can be run in parallel keeping in mind the final kVh. 12. Strips can be stored at −80°C. 3.1.2. SDS-PAGE
Preparing the Multi-Casting Chamber
In the second dimension the proteins are separated by exploiting their molecular mass and electrophoretic mobility. Therefore, proteins must be loaded with SDS after IEF. Additionally, disulphide bonds are reduced and alkylated by the addition of DTT and iodoacetamide to ensure complete denaturation of protein. SDS-PAGE can be run horizontally or vertically (33). In most modern applications, vertical electrophoresis is performed as it allows for running of several gels (up to 12) in parallel. 1. Prior to the assembly of the casting chamber, carefully clean the glass plates with deionized water and ethanol/water. 2. Dry the glass plates with lint-free tissue (Kim Wipe). 3. Stick two spacers (one on each side) on the plate using small amounts of silicone. 4. Place the second glass plate on the spacers.
High-Resolution 2DE
231
5. Stack the glass plates in the casting chamber. Each set of glass plates should be separated from the next by thin plastic sheets and after each second set of sheets a thicker plastic plate should be introduced. 6. Place the sealing tape in the groove of the casting chamber and screw down the chamber with the screws. Preparation of the Gels
Before pouring the gels the exact gel volume for one gel should be estimated (e.g. in experiments preparing 20 × 25 × 1 cm gels with 11% T in the ETTAN twelve gel caster multicasting chamber a volume of 80 mL gel solution is required). 1. Mix gel solution thoroughly. 2. Pour the gel solution bottom up into the chamber using a gradient mixer or separating funnel in order to prevent the creation of air bubbles. 3. Allow the solution to fill the chamber until there is a gap of about 0.5 cm at the top of the glass plates. 4. Do not allow air bubbles to enter the chamber. 5. Seal the tube using a clamp and remove the tube from the gradient mixer. 6. Overlay the gel surface with cover solution to exclude air and ensure a level surface on the top of the gel. 7. Allow polymerization for at least 2 h at RT. 8. The gels can be stored up to 1 week at 4°C (see Note 11). 9. Prior to electrophoresis wash the gel surface carefully with deionized water.
Equilibration of IPG Strips and Electrophoresis
1. After IEF or storage at −80°C equilibrate the strips in two consecutive steps (15 min each) (see Note 12). In the first step, 65 mM DTT is added to the equilibration buffer (reduction of proteins). Remove the solution after 15 min and replace it by equilibration buffer containing 280 mM iodoacetamide (alkylation of proteins). After an additional 15 min, remove solution. Equilibration should be carried out in suitable vessels (e.g. special plastics tubes or dishes). 2. Rinse the strip with running buffer and gently dry with Whatman paper to remove excess solution. 3. Fill the gap on the top of the gel with the molten agarose solution (see Note 13). 4. Immediately transfer the equilibrated strip onto the top of the gel using clean tweezers and spatulas. 5. Insert the gel cassette into the electrophoresis chamber and run the gels for 45 min at 2 W/gel and 3–4 h (depending on the number of gels) at 20 W/gel (see Note 14).
232
Marcus et al.
6. The proteins may be detected by any of the staining techniques applicable to SDS-PAGE analysis (see Table 1). 7. After staining the gels should be dried to ensure an improved stability of the gels and to avoid protein loss. 3.2. CA-Based 2DE
In-depth instructions including detailed figures of the equipment used as well as applied workflow are available (29, 34). CA-based 2DE can be easily adapted to the DIGE technique (see later) and other formats, including analytical as well as preparative gels.
3.2.1. Isoelectric Focusing
1. Prepare IEF tube gels (1.5 × 20 cm) 2 days before running. 2. Add 45 μL of APS solution to 2 mL of separations gel solution and fill tube to first mark (20 cm) using a syringe. 3. Now, add 14 μL APS solution to 0.7 ml cap gel solution and cast cap gel to second mark (20.5 cm) behind the separation gel. 4. To prevent urea crystallization, place an air cushion under the cap gel to third mark (21 cm). 5. Before starting IEF apply 2 mm Sephadex solution to prevent protein precipitation on surface of the separation gel. 6. Then load and overlay the sample with protection solution (ca. 5 mm) to prevent direct contact with the cathodic buffer. 7. Fill anodic (bottom) and cathodic buffers (top) into the IEF chambers. Ensure that no air bubbles hamper IEF (see Note 15). 8. Start IEF applying a step-wise voltage program (100 V for 1 h; 200 V/1 h; 400 V/17.5 h; 650 V/1 h; 1,000 V/0.5 h; 1,500 V/10 min; 2,000 V/5 min).
3.2.2. SDS-PAGE Preparing the Multi-Casting Chamber
Preparation of the Gels
1. Whilst the 21.5 h IEF is running, prepare gels for the second dimension. 2. Clean glass plates intensively using a lint-free paper towel – for the first wash use deionized water followed by 100% ethanol and finally 70% ethanol. To ensure correct gel dimension place two 1.5-mm plastic spacers between two plates sealed by silicone. 1. To start polymerization, add 288 μL APS solution to 144 mL gel solution, mix the solution by inverting, and cast the gel between the two cleaned glass plates fixed onto a polymerization station. Mix gel solution thoroughly. 2. Pour the gel solution top down into the chamber preventing the creation of air bubbles. 3. Allow the solution to fill the chamber until there is a gap of about 0.5 cm at the top of the glass plates. 4. Overlay the gel surface with isopropanol to exclude air and ensure a level surface on the top of the gel. 5. Allow polymerization for at least 45 min at room temperature.
High-Resolution 2DE
233
6. Remove isopropanol and wash the surface with protection solution to protect gel drying. 7. The gels can be stored up to 1 week at 4°C (see Note 11). 8. Prior to electrophoresis wash the gel surface carefully with deionized water. Equilibration of IEF Gels
1. After IEF, extrude the gel by means of inserting a nylon fibre (see Note 16) into the gel groove of the IEF gel carrier and incubate with equilibration solution for 15 min to load proteins with SDS. 2. Rinse the gel three times with running buffer before applying to the second dimension 3. For the transfer of the IEF gel into SDS-PAGE gel hold the groove with gel in contact with the edge of the glass plate and slide the gel between the glass plates using a wire suitably formed. 4. Overlay the IEF gels with molten agarose solution. 5. Add running buffer to the upper and lower (15°C) chambers and start electrophoresis. For the entrance of the proteins into the SDS-PAGE gel apply low current (75 mA) for 15 min. When the proteins have entered increase the current to 200 mA for approximately 5–7 h. 6. The proteins may be detected by any of the staining techniques applicable to SDS-PAGE analysis (see Table 1). 7. After staining, gels should be dried to ensure an improved stability of the gels and to avoid protein loss (see Note 17).
3.3. DIGE: CyDye™ Minimal Labelling (Fig. 2)
1. Solubilize proteins in DIGE lysis buffer (see Notes 18 and 19). 2. Add CyDye™ probes to the protein lysate and incubated on ice in the dark for 30 min so that 50 μg of protein is labelled with 400 pmol of CyDye™ (prepared in a working solution containing 400 pmol CyDye™ per μL anhydrous dimethylformamide) (see Note 20). 3. Add 1 μL of 10 mM lysine to stop the reaction and leave for 10 min on ice in the dark. 4. Optional. After labelling add 2× lysis buffer in a 1 + 1 dilution for IEF. 5. Combine samples (50 μg Cy™3 + 50 μg Cy™5 + 50 μg Cy™2 labelled = 150 μg per gel) and separate by IPG-based or CAbased 2DE. 6. Digitalize gel images using a fluorescence scanner (see Note 21).
3.4. DIGE: CyDye Saturation Labelling (Figs. 3 and 5)
1. Solubilise proteins in DIGE lysis buffer. 2. Optimize the labelling protocol for every individual sample (see Note 22).
234
Marcus et al.
3. Adjust pH to 8 and protein concentration to 0.55–10 mg/mL. 4. Reduce cysteine residues at 37°C for 1 h by incubating with TCEP (2 mM solution for analytical or 20 mM for preparative gels) in the dark with a volume determined in the labelling optimization experiment. 5. Add CyDye™ DIGE Fluor saturation dye (2 mM working solution working solution for analytical or 20 mM for preparative gels in anhydrous dimethylformamide) for reaction at 37°C in the dark for 30 min (see Note 23). 6. Stop reaction with excess DTT using 2× lysis buffer. 7. Combine samples and separate by IPG-based or CA-based 2DE. 8. Digitalize gel images using a fluorescence scanner (see Note 21). a CA-based 2D-PAGE MW 116 97
C
b
B
55
MW 36
D
IPG-based2D-PAGE
116 C 97
A
B 55
A
36
21 14
D
E 21
E
14 4
pI
9
3
pI
11
Fig. 5. Comparison of CA-based and IPG-based 2DE. In the pilot phase of the HUPO (Human Brain Proteome Project) we evaluated and compared the performance of both 2DE methods – CA-based 2DE (a) and IPG-based 2DE (b) (35) considering the resolving power and reproducibility of both systems. Mouse brain tissue was prepared according to our optimized protocol; the protein lysate was divided and separated by both methods in parallel. Both techniques were proven to be suitable for comprehensive differential proteome analysis and respective advantages and disadvantages were worked out (displayed in A & B): The advantages of the protein separation by CA-based 2DE with respect to differential quantitative gel-based proteome analysis proved to be a good separation in the central pH area (first dimension) (frame A), no precipitation effects at the application point (frame C), and a good resolution in the low molecular weight area (second dimension) (frame E). In contrast, disadvantageous points, such as streaking in the basic pH area (first dimension) (frame D) and a sub-optimal separation in the second dimension (high molecular weight area) (frame B), make differential analysis more difficult. Protein separation by IPG-based 2DE shows less streaking in the upper molecular weight area (frame B) and a good separation in the central pH area (frame A) making it a suitable method for differential quantitative gel-based proteome analysis. On the other hand precipitation effects at the application point (frame C), streaking in the basic pH area (frame D), and sub-optimal separation in the low molecular weight area (frame E) hamper the differential analysis (Reproduced with permission from ref. 35).
High-Resolution 2DE
235
4. Notes 1. The sample preparation and the equipment applied strongly depend on the characteristics of the respective sample. For example, special cells can be disrupted by sonication. The preparation of protein extracts from tissues usually needs more drastic treatments using a mortar, etc. Additionally, a sample preparation protocol depends on the class of proteins that should be analysed, e.g. if nuclear proteins are in the focus of the differential study, a specialized sample preparation protocol has to be tailored. Thus, the sample preparation must be individually optimized and adapted to the particular study to be undertaken. As a comprehensive description of sample preparation protocols would go beyond the scope of this chapter we will just make some general statements and would like to refer the reader to some excellent and comprehensive reviews ((36–40); see also Chapters “Solubilization of Proteins in 2DE: An Outline,” “Difficult Proteins,” and “Protein Extraction for 2DE”). 2. Adequate equipment is available from companies such as GE Healthcare Bio Sciences or BioRad, both Munich, Germany. 3. Different pH gradients can be used, either self-made as described in (32, 37, 41) or commercially available from GE Healthcare, BioRad, Serva Electrophoresis, Heidelberg, Germany. 4. For CA-based IEF all necessary equipment is pictured and described in detail in (29). 5. For IPG-based 2DE we indicate all buffers and solutions according to our general protocol for the analysis of protein lysates from brain samples. An adaptation of the composition of buffers and solutions as well as the running conditions will be necessary for every individual sample. Further detailed protocols for IPG-based 2DE are available (30–32, 35, 37, 42). 6. The pH range of the Servalytes depends on the pH range of the IPG strip used (e.g. IPG 3–11 use Servalyte 3–11). Alternative compositions of rehydration solution are described in the literature. Generally it should contain chaotropes (8 M urea or 6–8 M urea + 2 M thiourea), 0.5–4% non-ionic or zwitterionic detergents (e.g. CHAPS), a reductant (typically 0.4% DTT), and 0.5–2% CA (e.g. IPG buffer, Servalyte, Pharmalyte). 7. Never heat urea solutions above 37°C to reduce the risk of protein carbamylation. Additionally, it is important to deionize urea with an ion exchange resin prior to adding the other components, because urea in aqueous solutions exists in equilibrium with ammonium cyanate which might react with protein amino groups introducing additional charge and giving rise to artefactual spots in 2DE.
236
Marcus et al.
8. Another option for sample application is in-gel sample rehydration (43, 44). The sample is introduced in the rehydration solution and either actively (by the application of a certain voltage) or passively soaked into the strip during the overnight rehydration step. This method might be useful in case of highly diluted samples and avoids sample precipitation at the application point. Indeed, it holds the risk of protein degradation due to the long rehydration times as well as the appearance of horizontal smear caused by precipitation effects of individual proteins in the sample. The best parameters for sample application should be evaluated for every individual sample. 9. Anodic sample application is recommended for the more acidic (including the broad pH 3–11) IPGs, whereas cathodic sample application is advantageous when using the more basic IPG strips (32, 45). 10. Diluting sample with rehydration solution and covering it with silicone oil minimizes protein precipitation at the application point. 11. The quality of gels decreases with storage time. In consequence to ensure highest quality and reproducibility the gels should be used directly after casting. 12. In some cases it might be advantageous to extend the equilibration time to 20 min (e.g. 45) especially when equilibrating basic IPG strips. Drop-shaped protein spots point to insufficient SDS load. 13. The agarose solution has to be hot when loading; otherwise, it will set quickly. 14. The protein transfer from the IPG strip to the SDS-PAGE gel is enhanced when starting the electrophoresis with low voltage. Second-dimension gels of different dimensions will require the application of a different voltage and running times. 15. Air bubbles should be avoided by applying the solution slowly under the surface (1 mm) of the solution which has been applied before. For Sephadex application the small volume of separation gel buffer generated during polymerization is sufficient. 16. To prevent destruction of the IEF gel by the nylon fibre (a) the thermoplastic nylon fibre should be fitted to the tube inner diameter by melting one end into the tube and (b) the gel should be extruded using the cap gel (acrylamide concentration 12.3%) as a cushion, or (c) another possibility is to polymerize a highly concentrated acrylamide solution (15%) above the extruding side and use this gel piece as a cushion.
High-Resolution 2DE
237
17. Both methods can be combined with each other. For example, for an increase of resolution in the second-dimension large-gel SDS-PAGE (as described for the CA-based system) can be applied after IPG-based IEF (Fig. 3). 18. Samples produced by acidic precipitation should be lysed with DIGE lysis buffer adjusted to an appropriate pH between 8 and 9. Lower pH during the labelling reaction negatively affects labelling efficiency as well as kinetics in a negative way. The protein concentration should then be determined using a standard protein quantification method. Protein concentrations in a range from 1 to 20 mg/mL have been successfully labelled using the standard protocol. 19. Primary amines or thiols should not be present, because they will compete with the protein for the dye. 20. Samples can now be stored for at least three months at −70°C. Lysis buffer in classical 2DE traditionally contains primary amines (e.g. CA, Pharmalytes) and excess of thiols (e.g. DTT, DTE) which should not be present during the labelling reaction. At neutral or acidic pH lysine residues in proteins are +1 positively charged. CyDye™ minimal dyes also carry an intrinsic +1 charge which, when coupled to the lysine, replaces the lysine’s +1 charge with its own, ensuring that the pI of the protein does not significantly alter. 21. Choose excitation wavelengths and emission filters specific for each of the CyDyes™ according to the scanner user guide. 22. CyDye™ saturation labelling aims to label all available cysteines on each protein. In some proteins cysteine residues are buried and the extent of labelling will depend on the accessibility of cysteine within the protein under the reaction conditions used. Thus, in a first step optimal conditions for complete unfolding and breakage of disulphide bonds must be assessed. This can be achieved under denaturing conditions with a reducing agent such as Tris-(carboxyethyl) phosphine hydrochloride (TCEP). Additionally protein quantification of scarce samples is a challenging task because on the one hand the whole available protein moiety must be applied for differential analysis. On the other hand existing quantification methods are not sensitive enough to reliably detect protein amounts in the range of 1–3 μg of total protein. Therefore, labelling optimization in a 2DE titration experiment is obligatory to determine the optimum amount of TCEP and dye required for the protein extract being used. The molar ratio of TCEP: dye should always be kept at 1:2 to ensure efficient labelling. Typically, 5 μg of protein lysate requires 2 nmol TCEP and 4 nmol dye for the labelling reaction (assuming an average cysteine content of 2%).
238
Marcus et al.
If the amount of TCEP/dye is too low, available thiol groups on some proteins will not be labelled and show molecular weight trains in the 2DE image (Fig. 4). If the amount of TCEP/dye is too high, non-specific labelling of the amine groups on lysine residues can occur and will be observed as isoelectric point charge trains on the gel. 23. Generally Cy™3 is used for the pooled standard and Cy™5 for the individual samples.
Acknowledgment This work was supported by the Bundesministerium für Bildung und Forschung (NGFN, FZ 031U119 and 01GR0440) and the Ministerium für Wissenschaft und Forschung Nordrhein Westfalen. References 1. Wilkins, M. R., Sanchez, J. C., Gooley, A. A., Appel, R. D., Humphery-Smith, I., Hochstrasser, D. F., et al. (1996) Progress with proteome projects: Why all proteins expressed by a genome should be identified and how to do it. Biotechnol Genet Eng Rev 13, 19–50. 2. Gygi, S. P., Rist, B., Gerber, S. A., Turecek, F., Gelb, M. H., and Aebersold, R. (1999) Quantitative analysis of complex protein mixtures using isotope-coded affinity tags. Nat Biotechnol 17, 994–999. 3. Lohaus, C., Nolte, A., Bluggel, M., Scheer, C., Klose, J., Gobom, J., et al. (2007) Multidimensional chromatography: A powerful tool for the analysis of membrane proteins in mouse brain. J Proteome Res 6, 105–113. 4. Washburn, M. P., Wolters, D., and Yates, J. R., III (2001) Large-scale analysis of the yeast proteome by multidimensional protein identification technology. Nat Biotechnol 19, 242–247. 5. Schmidt, A., Kellermann, J., and Lottspeich, F. (2005) A novel strategy for quantitative proteomics using isotope-coded protein labels. Proteomics 5, 4–15. 6. Lücking, A. and Cahill, D. J. (2006) Protein biochips in the proteomics field. In Proteomics in Drug Research. Methods and Principles in Medicinal Chemistry, (Hamacher, M., Marcus, K., Stühler, K., van Hall, A., Warscheid, B., and Meyer, H.E., eds.), Wiley-VCH, Weinheim, pp. 137–158. 7. Klose, J. (1975) Protein mapping by combined isoelectric focusing and electrophoresis of
mouse tissues. A novel approach to testing for induced point mutations in mammals. Humangenetik 26, 231–243. 8. O’Farrell, P. H. (1975) High resolution twodimensional electrophoresis of proteins. J Biol Chem 250, 4007–4021. 9. Bjellqvist, B., Ek, K., Righetti, P. G., Gianazza, E., Gorg, A., Westermeier, R., et al. (1982) Isoelectric focusing in immobilized pH gradients: principle, methodology and some applications. J Biochem Biophys Methods 6, 317–339. 10. Görg, A., Postel W., Günther S., and Weser J., (1985) Improved horizontal two-dimensional electrophoresis with hybrid isoelectric focusing in immobilized pH gradients in the first dimension and laying-on transfer to the second dimension. Electrophoresis 6, 599–604. 11. Gorg, A., Postel, W., and Gunther, S. (1988) The current state of two-dimensional electrophoresis with immobilized pH gradients. Electrophoresis 9, 531–546. 12. Castellanos-Serra, L., and Hardy, E. (2001) Detection of biomolecules in electrophoresis gels with salts of imidazole and zinc II: a decade of research. Electrophoresis 22, 864–873. 13. Neuhoff, V., Arold, N., Taube, D., and Ehrhardt, W. (1988) Improved staining of proteins in polyacrylamide gels including isoelectric focusing gels with clear background at nanogram sensitivity using Coomassie Brilliant Blue G-250 and R-250. Electrophoresis 9, 255–262.
High-Resolution 2DE 14. Neuhoff, V., Stamm, R., Pardowitz, I., Arold, N., Ehrhardt, W., and Taube, D. (1990) Essential problems in quantification of proteins following colloidal staining with coomassie brilliant blue dyes in polyacrylamide gels, and their solution. Electrophoresis 11, 101–117. 15. Shevchenko, A., Wilm, M., Vorm, O., and Mann, M. (1996) Mass spectrometric sequencing of proteins silver-stained polyacrylamide gels. Anal Chem 68, 850–858. 16. Heukeshoven, J., and Dernick, R. (1988) Improved silver staining procedure for fast staining in PhastSystem Development Unit. I. Staining of sodium dodecyl sulfate gels. Electrophoresis 9, 28–32. 17. Lamanda, A., Zahn, A., Roder, D., and Langen, H. (2004) Improved Ruthenium II tris (bathophenantroline disulfonate) staining and destaining protocol for a better signal-tobackground ratio and improved baseline resolution. Proteomics 4, 599–608. 18. Rabilloud, T., Strub, J. M., Luche, S., van Dorsselaer, A., and Lunardi, J. (2001) A comparison between Sypro Ruby and ruthenium II tris (bathophenanthroline disulfonate) as fluorescent stains for protein detection in gels. Proteomics 1, 699–704. 19. Johnston, R. F., Pickett, S. C., and Barker, D. L. (1990) Autoradiography using storage phosphor technology. Electrophoresis 11, 355–360. 20. Tonge, R., Shaw, J., Middleton, B., Rowlinson, R., Rayner, S., Young, J., et al. (2001) Validation and development of fluorescence two-dimensional differential gel electrophoresis proteomics technology. Proteomics 1, 377–396. 21. Shaw, J., Rowlinson, R., Nickson, J., Stone, T., Sweet, A., Williams, K., et al. (2003) Evaluation of saturation labelling two-dimensional difference gel electrophoresis fluorescent dyes. Proteomics 3, 1181–1195. 22. Unlu, M., Morgan, M. E., and Minden, J. S. (1997) Difference gel electrophoresis: A single gel method for detecting changes in protein extracts. Electrophoresis 18, 2071–2077. 23. Sitek, B., Luttges, J., Marcus, K., Kloppel, G., Schmiegel, W., Meyer, H. E., et al. (2005) Application of fluorescence difference gel electrophoresis saturation labelling for the analysis of microdissected precursor lesions of pancreatic ductal adenocarcinoma. Proteomics 5, 2665–2679. 24. Helling, S., Schmitt, E., Joppich, C., Schulenborg, T., Mullner, S., Felske-Muller, S., et al. (2006) 2-D differential membrane proteome analysis of scarce protein samples. Proteomics 6, 4506–4513. 25. Sitek, B., Scheibe, B., Jung, K., Schramm, A., and Stühler, K. (2006), Difference gel elec-
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
239
trophoresis (DIGE): The next generation of two-dimensional gel electrophoresis for clinical research. In Proteomics in Drug Research. Methods and Principles in Medicinal Chemistry, (Hamacher, M., Marcus, K., Stühler, K., van Hall, A., Warscheid, B., and Meyer, H.E., eds.), Wiley-VCH, Weinheim, pp. 33–55. Wilson, K. E., Marouga, R., Prime, J. E., Pashby, D. P., Orange, P. R., Crosier, S., et al. (2005) Comparative proteomic analysis using samples obtained with laser microdissection and saturation dye labelling. Proteomics 5, 3851–3858. Greengauz-Roberts, O., Stoppler, H., Nomura, S., Yamaguchi, H., Goldenring, J. R., Podolsky, R. H., et al. (2005) Saturation labeling with cysteine-reactive cyanine fluorescent dyes provides increased sensitivity for protein expression profiling of laser-microdissected clinical specimens. Proteomics 5, 1746–1757. Fujii, K., Kondo, T., Yokoo, H., Okano, T., Yamada, M., Yamada, T., et al. (2006) Database of two-dimensional polyacrylamide gel electrophoresis of proteins labeled with CyDye DIGE Fluor saturation dye. Proteomics 6, 1640–1653. Klose, J. (1999) Large-gel 2-D electrophoresis. In 2-D Proteome Analysis Protocols (A.J. Link ed.), Humana Press, NJ, pp. 147–172 Görg, A. (1999) IPG-Dalt of very alkaline proteins. In 2-D Proteome Analysis Protocols (A.J. Link ed.), Humana Press, Totowa New Jersey, 197–209 Görg, A., and Weiss, W. (1999) Analytical IPG-Dalt. In 2-D Proteome Analysis Protocols (A.J. Link ed.), Humana Press, NJ, pp. 189– 195 Görg, A., Obermaier, C., Boguth, G., Harder, A., Scheibe, B., Wildgruber, R., et al. (2000) The current state of two-dimensional electrophoresis with immobilized pH gradients. Electrophoresis 21, 1037–1053. Görg, A., Boguth, G., Obermaier, C., Posch, A., and Weiss, W. (1995) Two-dimensional polyacrylamide gel electrophoresis with immobilized pH gradients in the first dimension (IPG-Dalt): the state of the art and the controversy of vertical versus horizontal systems. Electrophoresis 16, 1079–1086. Klose, J., and Kobalz, U. (1995) Two-dimensional electrophoresis of proteins: An updated protocol and implications for a functional analysis of the genome. Electrophoresis 16, 1034–1059. Stuhler, K., Pfeiffer, K., Joppich, C., Stephan, C., Jung, K., Muller, M., et al. (2006) Pilot study of the Human Proteome Organisation Brain Proteome Project: applying different 2-DE
240
36.
37.
38.
39.
40.
Marcus et al. techniques to monitor proteomic changes during murine brain development. Proteomics 6, 4899–4913. Lilley, K. S., Razzaq, A., and Dupree, P. (2002) Two-dimensional gel electrophoresis: recent advances in sample preparation, detection and quantitation. Curr Opin Chem Biol 6, 46–50. Gorg, A., Weiss, W., and Dunn, M. J. (2004) Current two-dimensional electrophoresis technology for proteomics. Proteomics 4, 3665–3685. Stasyk, T., and Huber, L. A. (2004) Zooming in: fractionation strategies in proteomics. Proteomics 4, 3704–3716. Wang, H., and Hanash, S. (2005) Intact-protein based sample preparation strategies for proteome analysis in combination with mass spectrometry. Mass Spectrom Rev 24, 413–426. Bodzon-Kulakowska, A., Bierczynska-Krzysik, A., Dylag, T., Drabik, A., Suder, P., Noga, M., et al. (2007) Methods for samples preparation in proteomic research. J Chromatogr B Analyt Technol Biomed Life Sci 849, 1–31.
41. Gianazza, E. (1999) Casting immobilized pH gradients (IPGs). In 2-D Proteome Analysis Protocols (A.J. Link ed.), Humana Press, NJ, pp. 175–188. 42. Marcus, K., and Meyer, H. E. (2004) Twodimensional polyacrylamide gel electrophoresis for platelet proteomics. Methods Mol Biol 273, 421–434. 43. Sanchez, J. C., Hochstrasser, D., and Rabilloud, T. (1999) In-gel sample rehydration of immobilized pH gradient. In 2-D Proteome Analysis Protocols (A.J. Link ed.), Humana Press, NJ, pp. 221–225. 44. Sanchez, J. C., Rouge, V., Pisteur, M., Ravier, F., Tonella, L., Moosmayer, M., et al. (1997) Improved and simplified in-gel sample application using reswelling of dry immobilized pH gradients. Electrophoresis 18, 324–327. 45. Marcus, K., Moebius, J., and Meyer, H. E. (2003) Differential analysis of phosphorylated proteins in resting and thrombin-stimulated human platelets. Anal Bioanal Chem 376, 973–993.
Chapter 15 Blue Native-Gel Electrophoresis Proteomics Kelly Andringa, Adrienne King, and Shannon Bailey Summary The importance of the mitochondrion in maintaining normal cellular physiology has long been appreciated. Recently there has been an upsurge in mitochondrial research due to increased recognition that a number of diseases are caused by defective functioning of this key intracellular organelle. Given this, along with advances made in proteomics technologies, the mitochondrion is clearly recognized as a top candidate for proteomics analysis. However, mitochondrial proteomics is not a trivial undertaking due to physicochemical properties that impair the resolution of inner mitochondrial membrane proteins when using conventional proteomic gel electrophoresis procedures. To circumvent such problems, many laboratories have adapted blue native-gel electrophoresis (BN-PAGE), a specialized type of native-gel electrophoresis, to generate high-resolution proteomic maps of the oxidative phosphorylation system. In this short overview the concepts and methods of BN-PAGE are presented, which demonstrate the power of using this complementary proteomics approach to identify alterations in the mitochondrial proteome that contribute to disease. Key words: Mitochondria, Oxidative phosphorylation system, Proteomics, Blue native-gel electrophoresis, Post-translational modification.
1. Introduction The fundamental role of the mitochondrion is to maintain ATP levels and energy homeostasis in the cell via proper functioning of the oxidative phosphorylation system, Kreb’s cycle, and β-oxidation. Recently though, interest in mitochondrial biology has increased due to evidence implicating mitochondrial involvement in cellular redox and apoptosis signaling through modulation of intracellular calcium levels, as well as in the production of reactive
David Sheehan and Raymond Tyther (eds.), Methods in Molecular Biology, Two-Dimensional Electrophoresis Protocols, vol. 519 © Humana Press, a part of Springer Science + Business Media, LLC 2009 DOI: 10.1007/978-1-59745-281-6_15
241
242
Andringa, King, and Bailey
oxygen and nitrogen species (1, 2). Given these critical roles, it is not surprising that mitochondrial dysfunction is implicated as a causative factor in human disease. However, the impact of disease on the overall content of mitochondrial proteins, i.e., the “mitochondrial proteome,” has only been investigated in recent years. It should also be mentioned that the mitochondrial proteome is not only defined by protein abundances, but also characterized by post-translational modifications to proteins, which if altered, may negatively impact the structural stability or functionality of proteins. Luckily, advancements have been made in the field of proteomics such that molecular alterations to the mitochondrial proteome can now be linked mechanistically to the development of disease. Since the resurgence of proteomics research, conventional two-dimensional (2D) gel electrophoresis (2DE) has been the principal tool used in this field. Briefly, this technique combines isoelectric focusing (IEF) of proteins in the first dimension where proteins are separated according to differences in net charge, followed by the separation of proteins based on molecular weight in the second dimension using standard SDS-PAGE (see Chapters “Two-Dimensional Electrophoresis (2DE): An Overview,” “Solubilization of Proteins in 2DE: An Outline,” and “Selection of pH Ranges in 2DE”). This technique is capable of resolving hundreds to thousands of proteins in a complex biological sample on a single 2DE gel. Identification of 2DE-resolved proteins is typically done using a variety of mass spectrometry (MS) techniques including matrix assisted laser desorption ionization timeof-flight (MALDI-TOF) MS. While 2DE is compatible for the separation of the more hydrophilic proteins of the mitochondrion, e.g., matrix proteins, analysis of the hydrophobic inner membrane proteins is hindered because these proteins precipitate at the basic end of IEF gels and thereby cannot be resolved (3–5). Indeed, only 5 of the 13 mitochondrial encoded polypeptides were found when mitochondria were subjected to 2DE (6). Since it is known that alterations in the content of respiratory chain proteins contribute to mitochondrial dysfunction and disease, novel separation techniques are required to look for changes in these proteins. To address this, our laboratory (7, 8) and others (9–11) have adapted the blue native-gel electrophoresis (BN-PAGE) technique developed by Schagger and von Jagow (12–14) to reveal alterations to the oxidative phosphorylation system. In this technique, the five oxidative phosphorylation complexes and other functional complexes associated with the inner membrane are separated intact in the first dimension under nondenaturing conditions (i.e., 1DE BN-PAGE). The protein complexes can then be transferred to a denaturing second-dimension gel to separate the complexes into
Blue Native-Gel Electrophoresis Proteomics
243
their individual polypeptide components (i.e., 2DE BN-PAGE). A simplified scheme depicting this procedure is shown in Fig. 1. Analysis of 1DE BN-PAGE gels yields information on the amount of each intact respiratory complex in the inner membrane, which can be verified using a large number of antibodies that are available to many of the individual respiratory chain polypeptides and by enzymatic activity assays (6). Similarly, analysis of 2DE BN-PAGE gels will yield information on the amount of the individual respiratory chain proteins within each complex. By using this approach changes in both nuclear and mitochondrial encoded subunits of the respiratory complexes during disease can be determined simultaneously. Furthermore, because 1DE BN-PAGE is done under nondenaturing conditions, information regarding protein–protein interactions and assembly of the complexes is retained. Thus, the combination of 2DE, BNPAGE, immunoblotting, and MS has the potential to resolve the ETC Complexes Dissociated from Inner Membrane Step 1 Laurylmaltoside Aminocaproic acid
I
II
IV
III
I
II
IV Step 2 Coomassie Blue G-250
III V
V
1st dimension native gel
I V III IV
kDa
II
Complex I (880)
kDa 160 105 75 50 35 30
Complex V (600) Complex III (460)
2nd dimension denaturing gel
Complex IV (200)
Step 4
Step 3
Complex II (140)
25 15 10
2nd dimension gel with complex subunit resolution
Fig. 1. Scheme illustrating analysis of the mitochondrial proteome by BN-PAGE proteomics. This simplified scheme illustrates the four major steps used to generate high-resolution “maps” of the proteins that comprise, and are associated with, the oxidative phosphorylation system. In this technique, protein complexes are extracted from the inner mitochondrial membrane (step 1) and separated under nondenaturing (i.e., native) conditions. During 1DE BN-PAGE the five complexes remain intact and enzymatically active (step 2), whereas under the denaturing conditions used for 2DE BN-PAGE the individual proteins that comprise each complex are separated by molecular weight (step 3). The proteins that make up each complex are aligned vertically within the gel based on size (i.e., high and low molecular weight subunits at the top and bottom of the gel, respectively) (step 4). Both 1DE and 2DE BN-PAGE gels can be subjected to image analysis to determine protein abundances in mitochondria from tissue of interest.
244
Andringa, King, and Bailey
majority of the proteins of the mitochondrial proteome and reveal changes in individual proteins with respect to identity, quantity, and post-translational modification in normal and diseased tissue. A detailed step-by-step discussion of the BN-PAGE technique is provided in the following sections.
2. Materials 2.1. Blue Native-Gel Electrophoresis 2.1.1. First-Dimension BNPAGE (1DE BN-PAGE)
1. “High-Blue” cathode buffer. 50 mM Tricine, 15 mM BisTris, and 0.02% Coomassie brilliant blue G-250, pH 7.0. Store at 4°C (see Note 1). 2. “Low-Blue” cathode buffer. 50 mM Tricine, 15 mM BisTris, and 0.002% Coomassie brilliant blue G-250, pH 7.0. Store at 4°C. 3. Anode buffer. 50 mM BisTris, pH 7.0. Store at 4°C. 4. Mitochondria extraction buffer. 0.75 M Aminocaproic acid, 50 mM BisTris, pH 7.0. Store at 4°C. 5. Coomassie brilliant blue G-250 suspension. 0.5 M Aminocaproic acid and 5% Coomassie brilliant blue G-250, pH 7.0. Store at 4°C (see Note 2). 6. Lauryl maltoside solution. 10% N-dodecyl-β-d-maltoside in water. Store at −20°C. 7. 3× Concentrated Gel Buffer. 1.5 M Aminocaproic acid, 150 mM BisTris, pH 7.0. Store at room temperature (see Note 3). 8. Molecular weight standards for 1DE BN-PAGE. High molecular weight native marker kit (Amersham Biosciences, catalog# 17-0445-01). Dissolve contents of one vial of molecular weight markers into 200 μL of mitochondria extraction buffer (item 4, Subheading 2.1.1), 25 μL lauryl maltoside (item 6, Subheading 2.1.1), and 12 μL Coomassie brilliant blue suspension (item 5, Subheading 2.1.1). Aliquot and store at −20°C.
2.1.2. Second-Dimension BN-PAGE (2DE BN-PAGE)
1. Cathode buffer. 100 mM Tris Base, 100 mM Tricine, and 0.1% SDS, pH 8.25. Store at room temperature. 2. Anode buffer. 200 mM Tris Base, pH 8.9. Store at room temperature. 3. 2D BN-PAGE gel buffer. 3 M Tris Base and 0.3% SDS, pH 8.45. Store at room temperature. 4. SDS/β-Mercaptoethanol solution. 20 μL β-Mercaptoethanol, 200 μL 10% SDS into 1,780 μL ultrapure H2O. Make fresh on day of experiment.
Blue Native-Gel Electrophoresis Proteomics
245
5. Agarose solution for sealing 1DE BN-PAGE gel strips onto second-dimension SDS-PAGE gels: 100 mg low melting temperature agarose, 1 mL 10% SDS, 100 μL β-mercaptoethanol into 9 mL ultrapure H2O. Make fresh on day of experiment. Heat to boiling right before use to dissolve agarose into solution. 2.2. Western Blotting for 1DE and 2DE BNPAGE
1. Standard reagents are used for the western blotting of BNPAGE gels and include the following listed as follows. 2. Western blot transfer buffer. 24 mM Tris base, 194 mM glycine, and 20% methanol in ultrapure H2O. 17.4 g Tris base, 87 g glycine, and 1.2 L of methanol are mixed with 4.8 L of ultrapure H2O. The pH of this solution should be between 8 and 8.5. Store at 4°C. 3. 10× Tris-buffered saline (10×TBS). 0.2 M Tris base, 9% NaCl, pH 7.4. 400 mL of 1 M Tris base stock solution, pH 7.4, is mixed with 180 g NaCl and the final volume is brought to 2.0 L with ultrapure H2O. 4. 10× Tris-buffered saline + Tween 20 (10×TBST). 5.0 mL of Tween 20 (MP Biochemicals, catalog # 103168) is added to 1.0 L of 10×TBS (item 3, Subheading 2.2) and stirred for 1 h at room temperature. 5. 1× Tris-buffered saline + Tween 20 (1×TBST). Dilute 10×TBST (item 4, Subheading 2.2) 1:10 with ultrapure H2O and store at 4°C. 6. Blocking buffers for membranes. The following two solutions can be used for blocking membranes. 5% milk solution, 7.5 g nonfat dry milk in 150 mL of 1×TBTS (item 5, Subheading 2.2), or 1% bovine serum albumin (BSA) solution, 10 g BSA in 1 L 1×TBST (item 5, Subheading 2.2), filter sterilize, and store at 4°C. 7. SuperSignal West Pico Chemiluminescent substrate (Pierce Biotechnology, catalog # 34078) is typically used for the visualization of proteins. Other chemiluminescent reagents may be used for detection of proteins.
2.3. Coomassie Blue – Total Protein Stain for 1DE and 2DE BN-PAGE Gels
1. Coomassie blue stain. 0.3 g Coomassie blue R-250, 100 mL glacial acetic acid, 250 mL isopropanol, and 650 mL ultrapure H2O. Use 100 mL per minigel. Stain gel overnight and do not reuse stain. Destain gels using a 10% glacial acetic acid solution. A small piece of a paper towel is added to the gel container to absorb the Coomassie blue as it “leaches” from the gel. Change the blue-stained paper towels every 2–4 h until a clear background on gels is achieved.
246
Andringa, King, and Bailey
3. Methods 3.1. Mitochondrial Sample Preparation
1. Regardless of tissue used, it is recommended that mitochondria are prepared from fresh tissue samples by validated experimental methodologies. Mitochondria should exhibit tightly coupled respiration with a high respiratory control ratio. After isolation, determination of the protein concentration, and respiratory control ratio, mitochondria can be stored at −80°C before beginning proteomic analyses. For storage, the volume equivalent to 1.0 mg mitochondrial protein is pipetted into a 1.5-mL microcentrifuge tube and mitochondria are centrifuged at high speed (∼21,000 × g) for 10 min at 4°C to form a miniature mitochondrial pellet at the bottom of the tube. The supernatant is carefully removed and pellets are stored, dry at −80°C.
3.2. Experimental Conditions for 1DE BN-PAGE Gels
1. The protocols described in (9) are used to prepare the specialized 5–12% or 5–16.5% gradient gels required for the separation of the oxidative phosphorylation complexes and associated enzyme complexes in the first dimension. These gels can be cast using any standard gradient mixer per manufacturer’s directions. Recipes for these gels are given in Table 1 (see Note 4). 2. For experiments, add 100 μL mitochondria extraction buffer and 12.5 μL of lauryl maltoside solution to each 1.0 mg mitochondrial pellet. Gently resuspend the mitochondrial pellets by pipetting and then place samples on ice for 30–60 min with gentle vortexing every 5–10 min to extract the enzyme complexes from the inner mitochondrial membrane. After this extraction step, centrifuge samples at 21,000 × g for 5–10
Table 1 Components for preparation of gradient gels used for 1DE BN-PAGE (see Note 4) Resolving gel solutions
Stacking gel solution
Light-5%
Heavy-12%
Or heavy-16.5%
4%
Protogel
0.66 mL
1.60 mL
2.20 mL
0.60 mL
Water
1.97 mL
0.56 mL
–
2.40 mL
3× gel buffer
1.34 mL
1.34 mL
1.34 mL
1.50 mL
Glycerol
–
0.47 mL
0.43 mL
–
10% AMPS
26.0 μL
26.0 μL
23.0 μL
70.0 μL
TEMED
4.0 μL
4.0 μL
3.50 μL
9.0 μL
Blue Native-Gel Electrophoresis Proteomics
247
min at 4°C to pellet any nondissolved material, remove the clarified supernatant to a fresh tube, and determine the protein concentration of the extract using standard protein assay procedures (i.e., Bradford protein assay). Keep samples on ice when not manipulating. 3. Immediately before loading samples in wells for electrophoresis, add 6.3 μL of Coomassie Brilliant Blue G-250 suspension to each tube containing approximately 100–120 μL of mitochondrial extract. The amount of protein that is typically loaded onto gels for electrophoresis is 75–250 μg (see Note 5). 4. After assembly of the gel apparatus, fill stacking gel wells with cold “High-Blue” cathode buffer and load sample. Fill the upper (i.e., inner) buffer chamber with cold “High-Blue” cathode buffer and the outer chamber with cold Anode buffer. Electrophoresis is performed at 4°C until samples have migrated into the resolving gel. At this time, the “High-Blue” cathode buffer is poured off gels and replaced with “LowBlue” cathode buffer and electrophoresis is continued for 3–4 h or until the dye front reaches the bottom of the gel (see Note 6). 5. After electrophoresis, gels can be stained with Coomassie Blue to determine the amount of each intact respiratory complex, processed for 2DE BN-PAGE, or subjected to western blotting by the procedures described herein (see Note 7). 6. A representative 1DE BN-PAGE gel from mouse liver mitochondria is shown in Figs. 1 and 2. 3.3. Experimental Conditions for 2DE BN-PAGE Gels
1. To separate the individual proteins that make up each respiratory complex, the intact gel lane containing all five complexes is cut from the 1DE gel and laid across the top of a denaturing Tris-Tricine/SDS-PAGE gel (Fig. 1). Alternatively, the individual complex bands themselves can be cut from the gel and applied to the top of an SDS-PAGE gel, if one is interested in the protein composition of only one specific complex (Fig. 2a). The recipe for preparing one, 1.5-mm thick denaturing gel is provided in Table 2. Leave a 0.5–1.0 cm gap at the top of the gel plate sandwich to accommodate gel strip or gel pieces. 2. To place the entire 1DE BN-PAGE gel on to the top of the SDS-PAGE gel, raise the gel plates in the clamps and place the upper part of the “glass plate sandwich” on to a hot plate or a heating block that has been placed at an angle. This will keep the agarose warm while inserting the 1DE gel strip in between the glass plates. Pour approximately 4–5 mL of the hot agarose solution between the glass plates, and using the back plate as a “staging area” and the agarose as a “lubricant” gently
248
Andringa, King, and Bailey
a
ATP Synthase Protein Gel
1DBN-PAGE Gel
ATP Synthase Western Blot
kDa 250 160 105 75
α ATP Synthase
50
β
β subunit, F1 portion
35 30
25 15
b
ATP Synthase
2D-BN-PAGE – Protein Gel
2D-BN-PAGE – Western Blot
Cytochrome c
2D-BN-PAGE – Western Blot
β subunit, F1 portion cytochrome c
Fig. 2. Identification of proteins in 2DE BN-PAGE gels via western blotting. (a) The intact band for the ATP Synthase (Complex V) was cut from a 1DE BN-PAGE gel and placed on top of a denaturing gel to separate the individual proteins that comprise this complex. Proteins were transferred to nitrocellulose membrane and probed using an antibody directed against the β-subunit of the F1 portion of the ATP Synthase. (b) Similarly, 2DE BN-PAGE gels, containing all complex proteins, can be subjected to western blotting techniques and probed for specific proteins. An example of this is provided again for the β-subunit of the ATP Synthase, as well as cytochrome c.
Table 2 Components of a single 1.5-mm thick denaturing gel used for 2DE BN-PAGE 2DE BN-PAGE gel buffer
2.98 mL
Protogel
2.98 mL
H2O
2.31 mL
Glycerol
0.72 mL
10% AMPS
60.0 μL
TEMED
6.00 μL
Blue Native-Gel Electrophoresis Proteomics
249
slip the entire 1DE gel strip down between the plates until the strip lies on top of the SDS-PAGE gel. Once the gel is in place, a single “tooth” of a gel comb can be inserted into the hot agarose on one end to serve as a well for molecular weight markers. Alternatively, if one is interested in the composition of only one of the respiratory complexes, the individual complex band can be cut from the entire 1DE BN-PAGE gel strip and placed on to the top of the SDS-PAGE resolving gel by the procedure described earlier using hot agarose to “seal” the band in between the gel plates. 3. After inserting the entire 1DE gel strip or the individual protein complex bands, remove the gel plate sandwich from the hot plate and allow the agarose to set for 30 min. Remove excess agarose from the top of the gel and then overlay the gel with the SDS/β-mercaptoethanol solution every 5–10 min for 45 min to ensure proteins are denatured before electrophoresis. 4. For electrophoresis, fill the inner chamber with 2DE cathode buffer and the outer chamber with 2DE anode buffer, load the molecular weight markers, and run gels at 30 V for 45 min followed by 110 V for 1–2 h. After electrophoresis, gels can be stained for total protein using the conditions described earlier or subjected to western blotting. 5. Examples of mouse liver mitochondrial proteins separated by the 2DE BN-PAGE approaches discussed in this section are provided in Figs. 1 – 4. 3.4. Western Blotting Conditions for BNPAGE Gels
1. In general, most standard laboratory reagents, conditions, and protocols can be used for western blotting 2DE BN-PAGE gels. A brief experimental protocol is provided for “wet/tank” transfer of gels to nitrocellulose membranes. 2. Soak 2DE BN-PAGE gels in western blot transfer buffer for at least 20 min to remove gel electrophoresis components as they will interfere with transfer of proteins. 3. Assemble western blotting apparatus as per manufacturer’s directions. Transfer can be accomplished by either a high current, short duration (300 mA, 1 h) or low current, long duration (90 mA, 16 h) transfer. If transfer is conducted overnight please place apparatus in cold room. 4. Following western blot, membranes can be “blocked” with either 5% nonfat milk or 1% BSA solution prepared in 1×TBST. Choice of blocking buffers must be determined for each individual antibody used to minimize background and optimize visualization of protein. 5. Standard procedures and reagents are used for primary and secondary antibody incubations, as well as for visualization of
250
Andringa, King, and Bailey
a. Total Protein I
V III
c. Thiols-IBTP
b. 4-HNE adducts IV II
I
V
III
IV II
I
V
III
IV II
kDa 160 105 75 50 35 30 25 15 10
Fig. 3. Detection of post-translational modifications of liver mitochondrial proteins using 2DE BN-PAGE proteomics and western blotting. Liver mitochondria were subjected to BN-PAGE proteomics using the techniques described in the Subheading 3 and illustrated in Fig. 1(a). After 2DE BN-PAGE, gels were subjected to immunoblotting and probed for (b) 4-hydroxynonenal (4-HNE) adducted proteins and protein cysteinyl groups reactive for 4-iodobutyl triphenylphosphonium (IBTP), a thiol-alkylating reagent that labels reduced thiols but not thiol groups that have been oxidized or modified as a consequence of ROS/RNS (15, 16) (see also Chapters “Detection of 4-Hydroxy-2-Nonenal- and 3-NitrotyrosineModified Proteins Using a Proteomics Approach” and “Proteomic Detection of Oxidized and Reduced Thiol Proteins in Cultured Cells”).
proteins using chemiluminescence reagents. Protein patterns can be captured using either film or imaging instruments. 6. Representative examples of a number of western blots from 2DE BN-PAGE are included, which demonstrate the power of this technique to determine alterations in protein composition (Fig. 2) and post-translational modification (Fig. 3). 3.5. Image Analysis of BN-PAGE Gels
1. Instructions for image analysis of gels are detailed in publications from our laboratory (7, 8, 15); see also Chapter “Troubleshooting Image Analysis in 2DE”). Briefly, scanned TIFF images for 1DE and 2DE BN-PAGE gels can be analyzed using either Scion Image Beta 4.02 (Scion Corp) or Quantity One (BioRad Laboratories) using the directions provided by the manufacturer for each software program.
3.6. Experimental Conditions for Mass Spectrometry Identification of Proteins
1. After excision of protein “spots” from the gel, the gel plugs are processed for MS using standard methods described in (8, 9, 15) (see also Chapters “Methods in Molecular Biology: Two-dimensional Electrophoresis Protocols, “Shotgun Protein Analysis by Liquid Chromatography-Tandem Mass
Blue Native-Gel Electrophoresis Proteomics OxPhos complexes I V III
251
OxPhos complexes
IV II
I V III
IV II
kDa 160 105 75
3 4 5 6 7
50
35
8
30
9
15
34
53 33 45 54 66 46 47 55
63
17 26 27
69
67 68
56 48 49 35 36
18 19
28 29
10 11 12 13 14 15
64
25
2
25
65
23 24
1
20 21
37
50 57 51
38 39 40
30 31
52 41 42 43
32
44
16 22
58 59 60 61 62
10
Fig. 4. Identification of proteins present in 2DE BN-PAGE proteomic map of mouse liver mitochondria using MS. (a) Mouse liver mitochondrial proteins were separated by 2DE BN-PAGE. The proteins that comprise, and are associated with, each of the five respiratory complexes (I, V, III, IV, and II) are aligned vertically under their respective Roman numeral designation. For identification, proteins were excised from gels and analyzed by either MALDI-TOF or *Q-TOF MS. The identified proteins (i.e., circled proteins) are listed by number in Table 3.
Spectrometry,” De Novo Sequence Analysis of N-Terminal Sulfonated Peptides After in-Gel Guanidination,” “Tryptic Digestion of In-Gel Proteins for Mass Spectrometry Analysis,” and “Database Interrogation Algorithms for Identification of Proteins in Proteomic Separations”) and those used by the UAB mass spectrometry shared facility (http://www.uab.edu/proteomics). Destain samples by three 30 min washes with a 50%, 50 mM NH4HCO3/50% acetonitrile solution. Then treat samples with 10 mM dithiothreitol in 50 mM NH4HCO3 for 60 min at 60°C to reduce cysteine residues, followed by alkylation of free cysteines with 55 mM iodoacetamide in 50 mM NH4HCO3 for 60 min at room temperature. Carry out digestion of proteins by 16 h incubation at 37°C with trypsin (12.5 ng/μL, Promega Gold trypsin). Extract the resulting peptide solution with two, 30 min washes of a 50/50 solution of 5% formic acid and acetonitrile, collect supernatants, and dry
252
Andringa, King, and Bailey
using a Savant SpeedVac. Resuspend dried peptide samples in 0.1% formic acid, desalt (C18 ZipTips, Millipore), and dilute1:10 with a saturated solution of α-cyano-4-hydroxycinnamic acid matrix before application to MALDI-TOF target plates. After plating, dry samples before analysis with a Voyager De-Pro mass spectrometer in the positive mode. Analyze spectra using Voyager Explorer software and submit peptide masses identified by mass spectrometry to the MASCOT database (see http://www.matrixscience.com) for protein identification. 2. Proteins that cannot be identified by MALDI-TOF are typically confirmed by tandem MS analysis using a Q-TOF2 mass spectrometer (Micromass, Manchester, UK) using electrospray ionization. Liquid chromatography is done using an LC Packings Ultimate LC, Switchos microcolumn switching unit, and Famos autosampler (LC Packings, San Francisco, CA). Concentrate the samples on a 300-μm i.d. C-18 precolumn at a flow rate of 10 μL/min with 0.1% formic acid and then flush onto a 75-μm i.d. C-18 column at 200 nL/min with a gradient of 5–100% acetonitrile (0.1% formic acid) in 30 min. The nano-LC interface is used to transfer the LC effluent into the MS. The Q-TOF is operated in the automatic switching mode whereby multiply charged ions are subjected to MS/ MS if intensities rise above six counts. Process the tandem MS with the MassLynx MaxEnt 3 software. 3. A list of the proteins identified from a representative 2DE BNPAGE gel from mouse liver mitochondria (Fig. 4) is provided in Table 3 (see Note 8).
Table 3 Mouse mitochondrial liver proteins identified using 2-D BN-PAGE and MALDI-TOF mass spectrometry Mass (kDa)
MOWSE score Accession no.
Propionyl-CoA carboxylase α chain, mitochondrial precursor
79.8
272
gi66773933
NADH dehydrogenase (ubiquinone) Fe-S Protein 1
79.7
272
gi74147040
2
Catalase
59.7
157
gi74228849
3
NADH dehydrogenase (ubiquinone) flavoprotein 1
50.8
176
gi19526814
176
gi89574015
171
gi23346461
Spot no. Protein identification 1
+
Mitochondrial ATP synthase, H transporting F1 complex β subunit 4
NADH dehydrogenase (ubiquinone) Fe-S protein 2
53.3
(continued)
Blue Native-Gel Electrophoresis Proteomics
253
Table 3 (continued) Spot no. Protein identification
Mass (kDa)
MOWSE score Accession no.
5
NADH dehydrogenase (ubiquinone) 1 α subcomplex 10 40.6
160
gi26337977
6
NADH dehydrogenase (ubiquinone) 1 α subcomplex 10
40.6
101
gi26337977
7
NADH dehydrogenase (ubiquinone) 1 α subcomplex 9 42.5
215
gi13543186
8
Nondetectable
–
–
–
9
NADH dehydrogenase (ubiquinone) Fe-S protein 3
30.1
262
gi58037117
10
NADH dehydrogenase (ubiquinone) flavoprotein 2
27.3
80
gi110625954
11
NADH dehydrogenase (ubiquinone) 1 α subcomplex 8 20.0
102
gi21312012
12
NADH dehydrogenase (ubiquinone) 1 α subcomplex subunit 12
17.1
108
gi47117166
13
NADH dehydrogenase 1 β subcomplex 4
15.1
94
gi21314826
14
NADH dehydrogenase (ubiquinone) Fe-S protein 5
12.7
82
gi72004262
15
NADH dehydrogenase (ubiquinone) Fe-S protein 6
13.0
105
gi56711244
16
NADH dehydrogenase (ubiquinone) 1 α subcomplex 2 10.9
96
gi31981600
17
ATP synthase, H+ transporting mitochondrial F1 complex α subunit
59.7
167
gi6680748
Mitochondrial ATP synthase H+ transporting F1 complex β subunit
48.1
167
gi74198645
18
ATP synthase γ chain, mitochondrial precursor
32.9
106
gi21263432
19
ATP synthase H+ transporting mitochondrial F0 complex subunit b isoform1
28.9
189
gi78214312
20
ATP synthase H+ transporting mitochondrial F0 complex subunit d
18.7
102
gi21313679
21
Cytochrome c oxidase subunit IV isoform1
19.5
74
gi6753498
+
22
ATP synthase H transporting mitochondrial F0 complex subunit G
11.4
81
gi31980744
23
Dimethylglycine dehydrogenase
97.2
197
gi21311901
24
Mitochondrial trifunctional protein α subunit
82.6
99
gi129378
25
60-kDa heat shock protein mitochondrial precursor
61.0
99
gi129378
26
Ubiquinol-cytochrome c reductase complex core protein 1
52.7
94
gi14548301
27
Ubiquinol-cytochrome c reductase core protein 2
48.2
98
gi22267442
28
Similar to cytochrome c1
29.5
114
gi109480910 (continued)
254
Andringa, King, and Bailey
Table 3 (continued) Spot no. Protein identification
Mass (kDa)
MOWSE score Accession no.
29
Ubiquinol-cytochrome c reductase Rieske Fe-S polypeptide 1
29.3
86
gi13385168
30
Ubiquinol-cytochrome c reductase complex 14 kDa protein
13.5
104
gi17382333
31
Low molecular mass ubiquinone-binding protein
9.8
84
gi21539585
32
Ubiquinol-cytochrome c reductase complex protein isoform 1
7.4
89
gi37574048
33
Acox 1 protein
60.0
220
gi68262425
Sdha protein
72.2
220
gi15030102
Slc25a13 protein
44.3
220
gi16741519
34
2-Hydroxyphytanoyl-CoA lyase
63.6
96
gi31560355
35
Mitochondrial malate dehydrogenase 2 NAD
30.7
115
gi895741221
36
Electron transfer flavoprotein subunit α mitochondrial precursor
35.0
99
gi21759113
37
Electron-transferring flavoprotein β polypeptide
27.6
154
gi12832367
38
Cytochrome c oxidase subunit 2
25.8
90
gi1706015
39
Peroxisomal acyl-CoA oxidase
74.6
69
gi26324826
40
Cytochrome c oxidase subunit IV isoform 1
19.5
140
gi6753498
41
Cytochrome c oxidase subunit 5A mitochondrial precursor
16.0
95
gi117099
42
Cytochrome c oxidase polypeptide Vla-liver mitochondrial precursor
12.3
88
gi1352173
43
Cytochrome c oxidase subunit VIb polypeptide 1
10.1
97
gi13385090
44
Calcium signal-modulating ligand
6.0
68
gi3513723
45
Succinate dehydrogenase Fp subunit
72.3
200
gi15030102
46
Epoxide hydrolase 2
62.5
NAa
–
Liver carboxylesterase 31 precursor
61.5
Acyl CoA synthetase long chain family member 1
78.1
47
UDP glucuronosyltransferase 2 family polypeptide B5
60.8
91
gi20381430
48
Acyl-coenzyme A dehydrogeanse short chain
44.9
134
gi31982522
49
Ornithine transcarbamylase
39.3
162
gi56789151
50
2,4-dienoyl CoA reductase 1 mitochondrial
36.2
89
gi13385680 (continued)
Blue Native-Gel Electrophoresis Proteomics
255
Table 3 (continued) Spot no. Protein identification
Mass (kDa)
MOWSE score Accession no.
51
Succinate dehydrogenase Ip subunit
31.8
93
gi34328286
52
Succinate dehydrogenase cytochrome b560 subunit mitochondrial precursor
18.4
96
gi21362402
53
Stress-70 protein mitochondrial precursor
73.4
92
gi74205924
54
Electron transfer flavoprotein-ubiquinone oxidoreduct- 68.0 ase mitochondrial precursor
98
gi52000730
55
Cytochrome P450 family 27 subfamily a polypeptide 1
60.7
111
gi30578401
56
Ornithine aminotransferase
48.3
131
gi8393866
57
Solute carrier family 25 member 5
32.9
93
gi22894075
58
Glutathione S Transferase
23.5
NAa
–
Peroxiredoxin 4
31.0
Peroxiredoxin 3
28.3
59
Cytochrome b5 outer mitochondrial membrane precur- 16.3 sor
88
gi31542438
60
Cytochrome b5
10.5
81
gi224985
61
Cytochrome c
11.5
84
gi483111
62
Heat shock protein 1 (chaperonin 10)
11.0
96
gi6680309
63
Similar to glutamate dehydrogenase 1 mitochondrial precursor
54.2
188
gi94396788
64
Peroxisomal acyl-CoA oxidase
74.6
162
gi6429156
162
gi6754256
Heat shock protein 9A 65
Sarcosine dehydrogenase
101.6
205
gi74201196
66
Carbolylesterase 6
61.9
101
gi74227229
67
Prolyl 4-hydroxylase β peptide
57.0
214
gi74198706
Carbolylesterase 2
62.4
214
gi21704206
68
Calreticulin
47.9
112
gi74200069
69
3-hydroxy-3-methylglutaryl-coenzymeA synthase 2
56.9
111
gi21758044
MOWSE score was determined from the MS-Fit software at www.matrixscience.com *Proteins identified by sequencing using Q-TOF MS
256
Andringa, King, and Bailey
4. Notes 1. The pH of solutions 1–5 (Subheading 2.1.1) should be adjusted at 4°C because 1DE BN-PAGE gels are run in the cold room. 2. Mix Coomassie Blue suspension before each use. 3. While the 3× Concentrated Gel Buffer used for 1DE BNPAGE gels (see Subheading 2.1.1, item 6) can be stored at room temperature, the pH must be adjusted at 4°C because gels are run in the cold room. 4. Gel volumes can be increased to prepare a native gel that has an extended length. This facilitates the separation of the five respiratory complexes for image analysis. 5. To facilitate the separation of the large respiratory chain complexes it is recommended that the concentration of the Coomassie Blue G-250 in the sample be one-fourth of the detergent concentration (13, 14). Thus, based on the given recipes for the stock solutions listed and the protocol described herein for the extraction of respiratory complexes from 1.0 mg of mitochondrial protein this works out to be about 6.3 μL of the 5% Coomassie Blue G-250 solution. The addition of a molar excess of the Coomassie Blue G-250 to the sample prevents aggregation of the membrane proteins in the presence of the detergents. The Coomassie Blue G-250 dye binds to the hydrophobic regions on the proteins surface, which induces a negative surface charge thus minimizing protein aggregation. Similarly, the anionic nature of the dye results in the required “charge-shift” on proteins such that they will migrate to the anode at physiological pH (13, 14). 6. Most commercially available mini and large-scale electrophoresis systems can be adapted and used for BN-PAGE. It is also important to note that the running conditions given for 1DE BN-PAGE result in low current output; thus, power supplies equipped to run at very low current are required. 7. Gel strips can be stored at −80°C before performing 2DE BNPAGE. 8. While many of the proteins identified from the 2DE BN-PAGE gel shown in Fig. 4 are subunits of the respiratory complexes, others include mitochondrial matrix proteins, as well as endoplasmic reticulum (ER) proteins. This demonstrates the ability of BN-PAGE to identify functional associations of matrix proteins with the respiratory complexes. Similarly, while it might be argued that the presence of ER proteins in gels simply indicates contamination of mitochondrial preparations with microsomal proteins, this finding again highlights the utility
Blue Native-Gel Electrophoresis Proteomics
257
of this technique to identify unique protein:protein interactions between mitochondrial and nonmitochondrial proteins. This is particularly relevant due to the known interaction of ER and mitochondria in the context of calcium handling and apoptosis signaling mechanisms (2, 17, 18).
Acknowledgments This work was funded by NIH/NIAAA grants AA15172 and DK73775 awarded to SMB. Ms. Adrienne L. King was supported by a Research Supplement to Promote Diversity in HealthRelated Research linked to parent grant AA15172. The authors would like to thank Dr. Paul S. Brookes, University of Rochester, for advice in establishing the BN-PAGE techniques. The authors also would like to acknowledge the assistance from Mr. Landon Wilson and Mr. Marion Kirk in the UAB mass spectrometry (MS) core facility, as well as the invaluable technical contributions of Ms. Anita Pinner and Ms. Laura Chamlee to these studies. The MALDI-TOF in the UAB MS core was purchased with funds provided by an NCRR grant S10 RR-11329. The Q-TOF2 was purchased with funds provided by NCRR grant S10 RR-13795 and UAB Health Services Foundation General Endowment Fund and operation of the MS core is supported, in part, from the UAB Comprehensive Cancer Center Core Support Grant P30 CA13148.
References 1. Camello-Almaraz, C., Gomez-Pinilla, P. J., Pozo, M. J., and Camello, P. J. (2006) Mitochondrial reactive oxygen species and Ca2+ signaling. Am J Physiol Cell Physiol 291, C1082–8. 2. Hajnoczky, G., Csordas, G., Das, S., GarciaPerez, C., Saotome, M., Sinha Roy, S., and Yi, M. (2006) Mitochondrial calcium signalling and cell death: approaches for assessing the role of mitochondrial Ca2+ uptake in apoptosis. Cell Calcium 40, 553–60. 3. Hanson, B. J., Schulenberg, B., Patton, W. F., and Capaldi, R. A. (2001) A novel subfractionation approach for mitochondrial proteins: a three-dimensional mitochondrial proteome map. Electrophoresis 22, 950–9. 4. Lopez, M. F., Kristal, B. S., Chernokalskaya, E., Lazarev, A., Shestopalov, A. I., Bogdanova,
A., and Robinson, M. (2000) High-throughput profiling of the mitochondrial proteome using affinity fractionation and automation. Electrophoresis 21, 3427–40. 5. Santoni, V., Molloy, M., and Rabilloud, T. (2000) Membrane proteins and proteomics: un amour impossible? Electrophoresis 21, 1054–70. 6. Bailey, S. M., Landar, A., and Darley-Usmar, V. (2005) Mitochondrial proteomics in free radical research. Free Radic Biol Med 38, 175– 88. 7. Bailey, S. M., Robinson, G., Pinner, A., Chamlee, L., Ulasova, E., Pompilius, M., et al. (2006) S-adenosylmethionine prevents chronic alcohol-induced mitochondrial dysfunction in the rat liver. Am J Physiol Gastrointest Liver Physiol 291, G857–67.
258
Andringa, King, and Bailey
8. Venkatraman, A., Landar, A., Davis, A. J., Chamlee, L., Sanderson, T., Kim, H., et al. (2004) Modification of the mitochondrial proteome in response to the stress of ethanoldependent hepatotoxicity. J Biol Chem 279, 22092–22101. 9. Brookes, P. S., Pinner, A., Ramachandran, A., Coward, L., Barnes, S., Kim, H., and DarleyUsmar, V. M. (2002) High throughput twodimensional blue-native electrophoresis: a tool for functional proteomics of mitochondria and signaling complexes. Proteomics 2, 969–77. 10. Reifschneider, N. H., Goto, S., Nakamoto, H., Takahashi, R., Sugawa, M., Dencher, N. A., and Krause, F. (2006) Defining the mitochondrial proteomes from five rat organs in a physiologically significant context using 2D blue-native/SDS-PAGE. J Proteome Res 5, 1117–32. 11. Reisinger, V., and Eichacker, L. A. (2006) Analysis of membrane protein complexes by blue native PAGE. Proteomics 6 Suppl 2, 6–15. 12. Schagger, H. (1996) Electrophoretic techniques for isolation and quantification of oxidative phosphorylation complexes from human tissues. Methods Enzymol 264, 555–66. 13. Schagger, H., Cramer, W. A., and von Jagow, G. (1994) Analysis of molecular masses and
14.
15.
16.
17.
18.
oligomeric states of protein complexes by blue native electrophoresis and isolation of membrane protein complexes by two-dimensional native electrophoresis. Anal Biochem 217, 220–30. Schagger, H., and von Jagow, G. (1991) Blue native electrophoresis for isolation of membrane protein complexes in enzymatically active form. Anal Biochem 199, 223–31. Venkatraman, A., Landar, A., Davis, A. J., Ulasova, E., Page, G., Murphy, M. P., et al. (2004) Oxidative modification of hepatic mitochondria protein thiols: effect of chronic alcohol consumption. Am J Physiol Gastrointest Liver Physiol 286, G521–G527. Lin, T. K., Hughes, G., Muratovska, A., Blaikie, F. H., Brookes, P. S., Darley-Usmar, V., et al. (2002) Specific modification of mitochondrial protein thiols in response to oxidative stress: a proteomics approach. J Biol Chem 277, 17048–56. Goetz, J. G., and Nabi, I. R. (2006) Interaction of the smooth endoplasmic reticulum and mitochondria. Biochem Soc Trans 34, 370–3. Walter, L., and Hajnoczky, G. (2005) Mitochondria and endoplasmic reticulum: the lethal interorganelle cross-talk. J Bioenerg Biomembr 37, 191–206.
Chapter 16 2DE for Proteome Analysis of Human Metaphase Chromosomes Kiichi Fukui and Susumu Uchiyama Summary Two-dimensional electrophoresis (2DE) is one of the most effective methods for the reliable separation of proteins in a single gel. In our proteome analyses of human chromosomes, we used two types of 2DE: two-dimensional isoelectric focusing SDS–polyacrylamide gel electrophoresis (2D IEF/ SDS–PAGE) and radical-free and highly reduced two-dimensional polyacrylamide gel electrophoresis (RFHR 2D PAGE) together with one-dimensional SDS–polyacrylamide gel electrophoresis (1DE). Experimental details of these gel electrophoresis procedures that have been shown to be effective for human proteome analyses are described in detail. Key words: Human metaphase chromosomes, Preparation of chromosome proteins, Onedimensional SDS–polyacrylamide gel electrophoresis, Two-dimensional isoelectric focusing SDS–polyacrylamide gel electrophoresis, Radical-free and highly reduced two-dimensional polyacrylamide gel electrophoresis.
1. Introduction The chromosome occupies a unique position among the organelles of plant and animal cells. This is because chromosome formation accompanies the drastic event of nuclear breakdown within a cell in addition to its significant stainability with dyes. Although the chromosome has been studied for more than a hundred years, its higher structure has been an enigma to date. The chromosome contains basically two types of molecules: DNA and the small basic histone proteins. As for other chromosomal proteins, they have not been systematically identified and their structure and function have not been fully identified. As a result, its higher order structure from the David Sheehan and Raymond Tyther (eds.), Methods in Molecular Biology, Two-Dimensional Electrophoresis Protocols, vol. 519 © Humana Press, a part of Springer Science + Business Media, LLC 2009 DOI: 10.1007/978-1-59745-281-6_16
259
260
Fukui and Uchiyama
protein view point has not been established yet, although several models have been proposed by a number of researchers (1). Two-dimensional electrophoresis (2DE) is one of the most effective methods for the reliable separation of proteins in a single gel. In our proteome analyses of human chromosomes we use two types of 2DE, two-dimensional isoelectric focusing SDS–polyacrylamide gel electrophoresis (2D IEF/SDS–PAGE) and radical-free and highly reduced two-dimensional polyacrylamide gel electrophoresis (RFHR 2D PAGE) together with onedimensional SDS–polyacrylamide gel electrophoresis (1DE). Proteome analysis of human chromosomes only allows us to present a comprehensive list of chromosomal proteins, which may contribute to the elucidation of chromosome higher order structure (2). Development of a specific 2DE method for separation of basic proteins combined with adequate chromosome isolation and protein extraction methods has enabled good separation of most of the chromosomal proteins from the metaphase chromosomes, in amounts suitable for their analysis by mass spectrometry (MS). Using these techniques, most of the human metaphase chromosome proteins that most likely have structural roles have been identified (3–6). In this manuscript, we describe the essentials of three types of gel electrophoresis for chromosomal proteins by which almost all chromosomal proteins are adequately separated.
2. Materials 2.1. Cell Lines, Cell Culture, Lysis, and Chromosome Harvesting
1. Cultured cell lines: BALL-1 and HeLa S3. 2. Culture medium: RPMI1640 with fetal bovine serum (FBS, 10% for BALL-1 cells and 5% for HeLa S3 cells). 3. Colcemid: 20 ng/mL final concentration for BALL-1 cells and 100 ng/mL for HeLa S3 cells. 4. Hypotonic solution: 75 mM KCl. 5. Ohnuki’s buffer: 55 mM KCl, 55 mM NaNO3, 55 mM CH3COONa, mixed in proportions 4: 2: 0.8 (7). 6. PA (polyamine) buffer: 15 mM Tris–HCl, pH 7.2, 2 mM EDTA, 80 mM KCl, 20 mM NaCl, 0.5 mM EGTA, 0.2 mM spermine, and 0.5 mM spermidine. 7. Modified PA buffer: PA buffer containing 1 mg/mL of digitonin, 14 mM 2-mercaptoethanol, and 0.1 mM phenylmethylsulfonyl fluoride (PMSF) (8, 9). 8. 1 μg/mL 4,6-diamidino-2-phenylindole (DAPI). 9. Sucrose solution for SG (sucrose gradient) chromosomes: 20% and 60% sucrose in PA buffer.
2DE for Proteome Analysis of Human Metaphase Chromosomes
261
10. Gradient fractionator (BioComp instrument Inc.). 11. Glycerol gradient solutions for PG (Percoll gradient) chromosomes: 5 or 70% glycerol in isolation buffer. 12. Wash buffer for PG chromosomes: 7.5 mM Tris–HCl, pH 7.4, 40 mM KCl, 1 mM EDTA, 0.1 mM spermine, 0.25 mM spermidine, 1% thiodiglycol, and 0.1 mM PMSF. 13. Lysis buffer for PG chromosomes: 15 mM Tris–HCl, pH 7.4, 80 mM KCl, 2 mM EDTA, 0.2 mM spermine, 0.5 mM spermidine, 1% thiodiglycol, 0.1 mM PMSF, and 0.1% Empigen. 14. Dounce homogenizer with B pestle. 15. Percoll solution for PG chromosomes: 89% Percoll, 5 mM Tris–HCl, pH 7.4, 20 mM KCl, 20 mM EDTA, 0.8 mM spermine, 2.25 mM spermidine, 1% thiodiglycol, 0.1 mM PMSF, and 0.1% Empigen. 16. Phosphate-buffered saline (10× PBS): 40 g NaCl, 1.0 g KCl, 14.5 g Na2HPO4•12H2O, 1.0 g KH2PO4 dissolved in 500 mL water. 17. JS-24.38 swinging rotor and JA-20 rotor (Beckman Instruments). 18. Gradient maker (Bio-Rad). 19. Isolation buffer for Percoll gradient (PG) chromosomes: 5 mM Tris–HCl, pH 7.4, 20 mM KCl, 20 mM EDTA, 0.25 mM spermidine, 1% thiodiglycol, 0.1 mM PMSF and 0.1% Empigen (Calbiochem). 2.2. Isolation of Chromosome Proteins
1. MgCl2. 2. Acetic acid. 3. Protein extraction solution: 100 mM MgCl2 in acetic acid/ water, 66/34. 4. Dialysis solution: 2% acetic acid in water. 5. Spectra/P membrane (MWCO = 1,000, Spectrum Medical Industries Inc.).
2.3. One-Dimensional SDS–Polyacrylamide Gel Electrophoresis
1. SDS sample buffer (2×): 62.5 mM Tris–HCl, pH 6.8, 50 mM dithiothreitol, 5% 2-mercaptoethanol, 2% SDS, 0.05% bromophenol blue (BPB). 2. 40% acrylamide stock solution: 193.3 g acrylamide, 6.7 g N, N ¢ -methylene-bis(acrylamide) (abbreviated BIS) in 500 mL of water. 3. 10% ammonium persulfate. 4. Stacking gel buffer (4×, for 100 mL): 0.4 g SDS and 6.04 g Tris dissolved in water. pH is adjusted to 6.8 with concentrated HCl solution.
262
Fukui and Uchiyama
5. Stacking gel solution for 1DE (to obtain a 4% polyacrylamide gel). Add 40 μL of a 10% ammonium persulfate stock solution and 5 μL of tetramethylenediamine (TEMED) in 4 mL of 1× stacking gel buffer. 6. Separating gel buffer stock solution (4×, for 100 mL): 0.4 g SDS, 18.2 g Tris. pH is adjusted to 8.8 with concentrated HCl solution. 7. Separation gel solution for 1DE: 6%, 8%, 12%, or 15% polyacrylamide gel (see Note 1). Add 80 μL of a 10% ammonium persulfate stock solution and 10 μL of TEMED to 8 mL of 1× separating gel buffer. 8. Running buffer: 14.4 g glycine, 3.0 g Tris, 1.0 g SDS dissolved in 1 L of water. 9. Staining solution: 1 tablet of PhastGel Blue R (GE Healthcare), in methanol/acetic acid/water, 25/8/67. 10. Destaining solution for PhastGel Blue R: methanol/acetic acid/water, 10/10/80. 2.4. Two-Dimensional Isoelectric Focusing SDS–Polyacrylamide Gel Electrophoresis
1. MicroSpin G-25 columns (GE Healthcare). 2. Sample buffer for 2D SDS–PAGE: 7 M urea, 2 M thiourea, 65 mM dithiothreitol, 2% CHAPS. 3. Ampholine pH 3.5–9.5 (GE Healthcare). 4. First dimension of separation: Immobiline DryStrip pH 4–7, Immobiline DryStrip pH 6–11 (GE Healthcare). 5. Equilibration Buffer stock solution: 50 mM Tris–HCl, pH 6.8, 6 M urea, 30% glycerol, 2% SDS. 6. Equilibration buffer 1: 10 mL of equilibration buffer stock solution, 25 mg of dithiothreitol. Add water to 100 mL. 7. Equilibration buffer 2: 10 mL equilibration buffer stock solution, 0.45 g iodoacetamide, 1 mg BPB. Add water to 100 mL. 8. 0.5% agarose solution: 0.5% agarose, 125 mM Tris–HCl, pH 6.8, 0.1% SDS. 9. Fixative solution: methanol/acetic acid/water, 20/8/72. 10. Gel electrophoresis system: Ettan IPGphor (GE Healthcare). 11. Imaging system: ImageScanner (GE Healthcare) with ImageMaster 2-D Elite version 4.01 (GE Healthcare).
2.5. Radical-Free and Highly Reduced Two-Dimensional Polyacrylamide Gel Electrophoresis (4–6, 10, 11)
1. 0-D buffer (50×): 60 mL 5 N KOH and 18.5 mL acetic acid. Add water to 500 mL (pH 5.4). 2. 0-D Prebuffer: 20 mL 0-D buffer (50×); add water to 1 L followed by autoclaving. 3. 1-D buffer (4×): 389 g Tris, 256 g boric acid, 64 g EDTA2Na. Add water to 2 L (pH 8.1). 4. 2-D Gel buffer (10×): 96 mL 5 N KOH, 523 mL acetic acid. Add water to 1 L (pH 3.1).
2DE for Proteome Analysis of Human Metaphase Chromosomes
263
5. 2-D EL buffer (10×): 300 g glycine, 32 mL acetic acid. Add water to2 L (pH 3.9). 6. 2-D Prebuffer: 200 mL 2-D Gel buffer (10×). Add water to 2 L. 7. 2-D EP buffer: 200 mL 2-D EL buffer (10×), 5 mL concentrated HCl, add water to 2 L. 8. 8 M Urea: 480 g urea. Add water to 1 L. 9. 0-D Gel solution: 16 g acrylamide (AA), 268 mg BIS, 72 g urea, 0.6 mL tetramethylethylenediamine (TEMED), 4 mL 0-D buffer (50×). Add water, dissolve all reagents with warming, and make up to 200 mL with water. Finally, add 24 g urea. 10. 1-D Gel solution: 16 g AA, 268 mg BIS, 72 g urea, 0.6 mL TEMED, 50 mL 1-D buffer (4×). Add water, dissolve all reagents with warming, and make up to 200 mL with water. Finally, add 24 g urea. 11. 2-D Gel solution: 180 g AA, 5 g BIS, 360 g urea, 5.8 mL TEMED, 100 mL 2-D Gel buffer (10×). Add water, dissolve all reagents with warming, and make up to 1 L with water. Finally, add 120 g urea. 12. APS: 5 g ammonium persulfate (APS), add water to 50 mL. Preparation just prior to use is preferable. 13. RFHR sample buffer: 8 M urea, 1.5% 2-mercaptoethanol. 14. Dye mixture: mix 0-D buffer (50×) and adequate amounts of pyronine G and acridine orange. 15. Gel electrophoresis system: RFHR 2DE system (Nihon Eido, Tokyo, http://www.nihon-eido.jp/,
[email protected]). 16. Imaging system: ImageScanner (GE Healthcare) with ImageMaster 2-D Elite version 4.01 (GE Healthcare).
3. Methods 3.1. Cell Culture, Lysis, and Chromosome Harvesting 3.1.1. Harvesting Chromosomes by Sucrose Gradient
1. Human cell lines, BALL-1 and HeLa S3, are subcultured once in every 3 days in culture medium at 37°C under a 5% CO2-containing atmosphere. At a cell concentration of 5 × 105 cells/mL, treat cells with colcemid for 12 h to arrest cell cycle at the M phase (see Note 2). 2. Collect synchronized cells from 400 mL of cell culture at 20°C by centrifugation at 440 × g for 5 min and resuspend in 40 mL of hypotonic solution (75 mM KCl) or Ohnuki’s buffer at 20°C. This treatment is required to swell the cells in order to lyse them efficiently in step 5.
264
Fukui and Uchiyama
3. After 30 min or 2 h of hypotonic treatment with KCl or Ohnuki’s buffer, respectively, collect the cells at 20°C by centrifugation at 780 × g for 10 min. 4. Hereafter, unless otherwise stated, perform all procedures for chromosome preparation at 4°C. Resuspend the cell pellet in 15 mL of PA buffer for chromosome isolation. Polyamine molecules included in PA buffer are required to maintain chromosome structure stable in vitro. 5. After lysis of cell membranes by vortexing for 30 s, remove cell debris and intact nuclei by centrifugation at 190 × g for 3 min. 6. Dilute the chromosome-containing supernatant in 15 mL of PA buffer and centrifuge at 190 × g for 3 min twice. 7. Carefully recover the final supernatant, which is the chromosome-rich fraction (C-fraction). The precipitate is the nuclei-rich fraction (N fraction). 8. Centrifuge C-fraction again at 1,750 × g for 10 min. 9. Resuspend the precipitated chromosomes (PA chromosomes) in 1 mL of fresh PA buffer (see Note 3). 10. Confirm the nature of the isolated chromosomal sample by both optical and electron microscopy (see Note 4). 11. PA chromosomes are then subjected to sucrose density gradient centrifugation. Layer PA chromosomes onto 35 mL of a linear 20–60% sucrose gradient in PA buffer, prepared using a gradient fractionator (BioComp instrument Inc.), and centrifuge at 2,500 × g for 15 min in a JS-24.38 swinging rotor. Fractions containing chromosomes should be assessed by optical microscopy after DAPI staining and combined together. In our case, chromosomes were obtained in the 40–50% sucrose fractions. 12. Collect SG chromosomes by centrifugation at 1,000 × g for 10 min after dilution of sucrose with five volumes of PA buffer (see Note 5). 3.1.2. Harvesting Chromosomes by Percoll Gradient (PG Chromosomes)
1. Prepare a 5–70% exponential glycerol gradient made from 7 mL of 70% glycerol and 25 mL of 5% glycerol in 50-mL polycarbonate tubes. Use a gradient maker (Bio-Rad) (5, 12). Store the gradient on ice. 2. Harvest the cells at a cell concentration of 5 × 105 cells/mL from 400 mL of cell culture at 20°C by centrifugation at 1,500 × g for 10 min. Suspend them in wash buffer at 20°C and collect them at 1,200 × g for 5 min. Perform three such washes. The cells are thereby hypotonically swollen by wash buffer. 3. Lyse the cells on ice in 5 mL of lysis buffer with 10–12 pestle strokes (counting up and down as one stroke) using a Dounce homogenizer B pestle (see Note 6).
2DE for Proteome Analysis of Human Metaphase Chromosomes
265
4. Layer the lysate over the exponential glycerol gradient, and centrifuge in a JS-24.38 rotor for 5 min at 200 × g, followed by 15 min at 700 × g. Chromosomes form a broad diffuse band in the glycerol gradient, between 10 and 25 mL from the top of the glycerol gradient. 5. Collect a 15 mL fraction containing chromosomes which form a broad band in step 4, and place it on a 5 mL 70% glycerol cushion, and then centrifuge at 3,000 × g for 15 min to recover chromosomes in the glycerol cushion. 6. Combine 10 mL of 70% glycerol cushion-containing chromosomes with 10 mL of 89% Percoll solution, and homogenize with 10 pestle strokes using a Dounce homogenizer B pestle, followed by 30 pestle strokes after further addition of 15 mL of 89% Percoll solution. This step aims to purify further the chromosomes: The chromosomes isolated by glycerol gradient centrifugation indeed still contain many kinds of contaminants, such as mitochondrial and cytoskeletal proteins. To remove these contaminants, chromosomes are mixed with Percoll solution and homogenized. 7. Centrifuge at 45,440 × g for 30 min in a JA-20 rotor. Collect a band (a volume of about 5 mL) containing chromosomes located at a level that is one fifth of the length of the tube measured from the bottom, and dilute threefold in isolation buffer. Subsequently, centrifuge at 3,000 × g for 15 min to pellet PG chromosomes. For PA chromosomes (obtained in Subheading 3.1.1, step 9) and PG chromosomes (obtained in Subheading 3.1.2, step 7) prepared from 2.8 × 108 cells, 500 μg and 44 μg of chromosome proteins were collected, respectively. PG chromosomes are much cleaner than PA chromosomes. By Percoll gradient centrifugation, mitochondrion, cytoplasmic, and cytoskeletal proteins are efficiently removed from chromosomes. Note that PA chromosomes are cruder than the chromosomes obtained after glycerol gradient centrifugation. 3.2. Isolation of Chromosome Proteins and SDS–PAGE
1. Transfer the chromosomes isolated in PA buffer (PA or SG chromosomes) into microtubes and collect by centrifugation at 7,700 × g for 10 min at 4°C followed by resuspension in 1× PBS. 100 μL of PBS is added into the chromosomes isolated from 8.0 × 108 cells. For the already pelleted PG chromosomes, directly add PBS onto the pellet. 2. Chromosome proteins are extracted by the acetic acid method (11, 13) with minor modifications. 1/10th volume of 1 mM MgCl2 and two volumes of acetic acid are added to the chromosome suspension, and the mixture is stirred for 1 h at 4°C with a microtube stirrer. Chromosome proteins are dissociated from chromosomes by decreasing pH with acetic acid.
266
Fukui and Uchiyama
3. After centrifugation at 17,400 × g for 10 min, collect supernatants and add an equal volume of protein extraction solution to the precipitate, followed by stirring for 20 min at 4°C. 4. After centrifugation at 17,400 g for 10 min, collect this second supernatant and pool both supernatants obtained in steps 3 and 4. 5. Dialyze the supernatants against 2% acetic acid in water with Spectra/P membrane for 2 h at 4°C to decrease the concentration of acetic acid. At least three buffer changes are recommended. 6. Freeze the extracted protein solution with liquid nitrogen and lyophilize (see Note 7). 7. Clean glass plates, plastic combs, and spacers using 100% ethanol. Set up gel plates with plastic seal and paper clamps. Stand the plates on top of a sheet of aluminum foil. Pour ethanol to check the gel assembly does not leak. Discard the ethanol. 8. Prepare the separating gel solution and pour the gel solution into the gel assembly until it reaches about 2–3 cm below the top of the glass. Layer ethanol on top of the gel (squirt bottle). Wait for 20 min for the gel to polymerize. 9 . Discard ethanol layer. Prepare stacking gel solution and pour the stacking gel solution on top of the polymerized separating gel. 10. Insert a dry and clean comb without making any air bubbles before the stacking gel polymerizes and wait for 15 min. 11. Remove the comb and spacer slowly, assemble the gel plate onto the electrophoresis tank. Fill the tank with running buffer. 12. Solubilize the lyophilized proteins into 1× SDS sample buffer and boil for 5 min. Load 5–30 μL (5–30 μg) of samples in different wells (the volume depends on the size of the well). 13. Carry out electrophoresis at steady current (30–40 μA) for 1–1.5 h. 14. After electrophoresis, stain the gel with staining solution, followed by destaining with destaining solution. 3.3. 2D IEF/SDS–PAGE
1. Solubilize 500 μg of lyophilized proteins into 50 μL of sample buffer. After centrifugation at 17,400 × g for 10 min at 4°C, desalt supernatant using MicroSpin G-25 columns, because the salts present in the sample would perturb protein isoelectric focusing (IEF) during the following electrophoresis step. Add 1 μL of ampholine, pH 3.5–9.5, into sample solutions for Immobiline DryStrip pH 4–7 or pH 3–10 (see Note 8).
2DE for Proteome Analysis of Human Metaphase Chromosomes
267
2. Set the Immobiline DryStrip on Ettan IPGphor and add sample solution. Carry out electrophoresis. An example of electrophoresis conditions is: 0 V for 2 h; 50 V for 8 h; 1,000 V for 2 h; 1,000 V for 1 h; 2,000 V for 2 h; 5,000 V for 2 h; 8,000 V for over 8 h. 3. After the first dimension of electrophoresis, shake the strip in equilibration buffer 1 for 15 min at room temperature, followed by in equilibration buffer 2 for 15 min at room temperature in the dark to reduce and alkylate cysteines, respectively. 4. Prepare a 12% polyacrylamide gel as described in Subheading 3.2. Put the gel strip on the stacking gel and immobilize with 0.5% agarose solution. 5. Assemble the gel plate onto the electrophoresis tank. Fill the tank with running buffer. Carry out electrophoresis at steady current (30 mA) for 4 h. 6. After electrophoresis, shake the gel in fixative solution for 1 h. Stain the gel in staining solution, followed by destaining in destaining solution. 7. Scan gel images using ImageScanner. Analyze the images by ImageMaster 2-D Elite version 4.01, and calculate isoelectric point, molecular weight, and amount of each spot (see also Chapter “Immunoblotting 2DE Membranes”). 3.4. RFHR 2D PAGE (4–6, 10, 11)
1. Preparation of 0-D gel: Add small amount of APS to 0-D gel solution, pour it into bottom plate, and place 0-D apparatus to seal its bottom. After gelation, add small amount of APS to 0-D gel solution, pour it into half level of 0-D apparatus. 2. Preparation of 1-D gel: Set the sample gel cover to gel container using clip. Seal the sample gel cover with 1-D gel. Attach the anode buffer container and seal the connecting portion with gel container. Pour several mL of 1-D gel solution mixed with APS into gel bottom plate; place gel container to seal its bottom. Fill gel space with gel solution mixed with APS. Finally add gel solution mixed with APS to the pit of the gel top. 3. Preparation of 2-D gel: Connect gel container with anode buffer container. Seal space existing at each end of gel container and around the container using 0-D gel solution. Wipe the extra gel, pour 30 mL of 2-D gel solution mixed with 3 mL of AP into gel making dish. Place gel container to seal its bottom. Pour 300 mL of 2-D gel solution mixed with 10 mL of APS into gel container, immediately followed by the placing gel spacer into container. Place a heavy stone on gel spacer. 4. Prerun, 0-D at 100 V for 1 h: 0-D (−) vessel; 100 mL 0-D buffer. 0-D (+) vessel; dissolve 1 g mercapto-ethyl-amine (MEA) into 100 mL of a solution prepared from 2 mL 0-D buffer (50×), 75 mL 8 M urea, and 23 mL water.
268
Fukui and Uchiyama
5. Prerun, 1-D at 100 V for O/N:1-D (+) vessel, 100 mL of solution composed of 50 mL 1-D buffer (4×) and 150 mL 8 M urea. 1-D (−) vessel. Add 0.85 mL of mercapropropionate acid to 100 mL from the mixed solution (the final pH should be 7.8). 6. Prerun, 2-D at 100 V for O/N:2-D(−) vessel: 400 mL 2-D Prebuffer. 2-D (+) vessel; dissolve 2 g MEA in 400 mL of a solution prepared from 40 mL of 2-D Gel buffer (10×), 300 mL of 8 M urea, and 60 mL of water. Air-fan cooling is required during the electrophoresis. 7. Sample preparation: Solubilize 500–2,000 μg of lyophilized proteins into 50–200 μL of RFHR sample buffer, followed by reduction treatment at 40°C for 30 min. Add 1/50 amount of dye mixture. 8. 0-D electrophoresis (sample concentration): Place sample solution on 0-D gel. Pour anode electrode solution (50 mL 8 M urea, 10 mL 2-D EL buffer (10×), 3.5 g of Cysteine. Add water (100 mL) onto sample solution. Fill the (+) vessel with anode electrode solution. Fill the (−) vessel with 100 mL of cathode electrode solution (0-D buffer). Electrophorese at 100 V until the dye enters the gel for 2–3 cm. It takes 10–15 min. for 100 mL of sample solution. Cut lower part of the gel and place into 1-D cathode electrode solution for the buffer exchange. 9. 1DE: prepare 1-D electrode solution by mixing 150 mL of 8 M urea and 50 mL of 1-D buffer (4×). Place 0-D gel into 1-Dgel. Pour 100 mL of 1-D electrode solution into (+) vessel. Add 0.85 mL of mercaptopropionate acid to the 100 mL of 1-D electrode solution and pour it into (−) vessel. Electrophorese at 150–170 V with air-fan cooling until the acridine orange dye reaches the bottom of the gel. This takes approximately 6 h. Cut cathode electrode side 1-D gel with 10.6 cm and anode electrode side 1-D gel with 5.3 cm. 10. 2DE: Place 1DE gel onto 2DE gel with no air left between gels. Add a small amount of pyronine G solution as a marker. Pour 400 mL of 2-D EP buffer as cathode electrode buffer into (−) vessel. Prepare anode electrode buffer by mixing 300 mL of 8 M urea, 40 mL of 2-D EL buffer (10×), 14 g of cysteine and add water to 400 mL. Pour anode electrode buffer into (+) vessel. Electrophorese at 100 V with air-fan cooling until pyronine G reaches the bottom of the gel. This takes approximately 15–20 h. 11. After electrophoresis, shake the gel in fixative solution for 1 h. Stain in staining solution, followed by destaining in destaining solution. 12. Scan the gel images using ImageScanner. Analyze the images by ImageMaster 2-D Elite version 4.01, and
2DE for Proteome Analysis of Human Metaphase Chromosomes
269
calculate isoelectric point, molecular weight, and amount of each spot. From PA and PG chromosomes isolated from 2.8 × 108 cells, about 200 and 100 proteins were identified, respectively (5, 6). Figure 1 shows the two electrophoresis patterns of the chromosome proteins prepared by Percoll gradient procedures (PG chromosomes). In a 1DE gel, about 50 bands were detected by Coomassie brilliant blue (CBB) staining, and three or four proteins were identified from several single bands. However, in general, it is difficult to identify more than three proteins from single bands by MALDI-TOF MS. If a single band contains several proteins, tandem MS (MS/MS) analysis is recommended (see Chapter “Shotgun Protein Analysis by Liquid ChromatographyTandem Mass Spectrometry”). Alternatively, although the separation of high-molecular-weight (>100 kDa) proteins is difficult, 2DE is also useful for protein identification, because almost all spots contain one protein. In the case of chromosome proteins, more than 100 spots were detected by CBB staining. Furthermore, using 2DE, we could estimate the amount of each protein after CBB staining.
Fig. 1. Electrophoresis patterns of 1-DE PAGE (a) and RFHR 2D PAGE for the chromosomal proteins isolated from PG chromosomes. See text for details.
270
Fukui and Uchiyama
4. Notes 1. The acrylamide percentage is adjusted according to the molecular weight range of the proteins to be detected: 6 or 8% PAGE is used for proteins above 100 kDa, 12% PAGE for proteins of MW 20–100 kDa, and 15% PAGE for proteins of MW 10–60 kDa. 2. For the isolation of chromosomes, 2 L of cell culture (at 7 × 105 cells/mL) can be handled easily in a single preparation. By this synchronization method, a mitotic index above 50% is routinely obtained. 3. The separation profiles between chromosomes and nuclei can be confirmed by flow cytometry (EPICS Elite, Beckman Coulter) after staining of DNA with 35 μg/mL Propidium iodide (PI). 4. For observation by optical microscopy, isolated chromosomes are placed on glass slides and observed using a fluorescence microscope, after staining with 1 μg/mL DAPI. For electron microscopy observation, isolated chromosomes are mounted on a plastic sheet which has been previously coated with poly-L-lysine and are fixed with 4% (p)-formaldehyde in PA buffer. After washing with the same solution, the specimens are dehydrated in an ethanol series and freeze-dried using the t-butyl alcohol drying method (14). Finally, the specimens are osmium coated with an osmium plasma coater and examined under the scanning electron microscope at 10 kV or 15 kV (14). 5. Isolated PA and SG chromosomes can be stored at −80°C in PA buffer containing 70% glycerol until use. Glycerol is used for cryoprotection of chromosomes. 6. Homogenization should be performed gently to avoid damage to chromosomes. After cell lysis, all the procedures for PG chromosome isolation should be performed at 4°C. 7. Lyophilized proteins can be stored as powder material at −80°C. 8. We use 11-cm Immobiline DryStrip for 2DE. Add 0.5 μL of IPG Buffer pH 6–11 (GE Healthcare) into sample solution if you wish to use Immobiline Drystrip pH 6–11 (GE Healthcare).
Acknowledgments The authors would like to thank Drs. Sachihiro Matsunaga, Shin’ichiro Kajiyama, and Hideki Takata for their collaboration during the research on chromosome proteomics. We also thank
2DE for Proteome Analysis of Human Metaphase Chromosomes
271
Drs. Hideji Yoshida and Akira Wada for the technical advice of RFHR 2D PAGE. This work was supported by the Special Coordination Funds from the Ministry of Education, Culture, Sports, Science and Technology (MEXT), Japan, to K. F. and S.U. References 1. Fukui, K., and Uchiyama, S. (2007) Chromosome protein framework from proteome analysis of isolated human metaphase chromosomes. Chem. Rec. 7, 230–237 2. Gassmann, R., Henzing, A. J., and Earnshaw, W. C. (2004) Novel components of human mitotic chromosomes identified by proteomic analysis of the chromosome scaffold fraction. Chromosoma 113, 358–397 3. Sone, T., Iwano, M., Kobayashi, S., Ishihara, T., Hori, N., Takata, H. et al. (2002) Changes in chromosomal surface structure by different isolation conditions. Arch. Histol. Cytol. 65, 445–455 4. Uchiyama, S., Kobayashi, S., Takata, H., Ishihara, T., Sone, T., Matsunaga, S., and Fukui, K. (2004) Protein composition of human metaphase chromosomes analyzed by two-dimensional electrophoreses. Cytogenet. Genome Res. 107, 49–54 5. Uchiyama, S., Kobayashi, S., Takata, H., Ishihara, T., Hori, N., Higashi, T. et al. (2005) Proteome analysis of human metaphase chromosomes. J. Biol. Chem. 289, 16994–17004 6. Takata, H., Uchiyama, S., Nakamura, N., Nakashima, S., Kobayashi, S., Sone, T. et al. (2007) A comparative proteome analysis of human metaphase chromosomes isolated from two different cell lines reveals a set of conserved chromosome-associated proteins. Genes Cells 12, 265–284 7. Ohnuki, Y. (1968) Structure of chromosomes. I. Morphological studies of the spiral
8.
9.
10.
11.
12.
13.
14.
structure of human somatic chromosomes. Chromosoma 25, 402–428 Kuriki, H., and Haruo, T. (1997) Standardization of bivariate flow karyotypes of human chromosomes for clinical applications. J. Clin. Lab. Anal. 11, 169–174 Valdivia, M. M. (1998) Chromosome isolation for biochemical and morphological analysis. In, Spector, D. L., Goldman, R. D., and Leinwand, L. A. (eds.). In Cells: A Laboratory Manual (Spector, D.L., Goldman, R.D. and Leinwald, L.A. eds.) Cold Spring Harbor Laboratory Press, New York, pp. 49.1–49.12 Wada, A. (1986) Analysis of Escherichia coli ribosomal proteins by an improved two dimensional gel electrophoresis. I. Detection of four new proteins. J. Biochem. 100, 1583–1594 Izutsu, K., Wada, A., and Wada, C. (2001) Expression of ribosome modulation factor (RMF) in Escherichia coli requires ppGpp. Genes Cells 6, 665–676 Gasser, S., and Laemmli, U. (1987) Improved methods for the isolation of individual and clustered mitotic chromosomes. Exp. Cell Res. 173, 85–98 Hardy, S. J., Kurland, C. G., Voynow, P., and Mora, G. (1969) The ribosomal proteins of Escherichia coli. I. Purification of the 30S ribosomal proteins. Biochemistry 8, 2897–2905 Inoue, T., and Osatake, H. (1988) A new drying method of biological specimens for scanning electron microscopy: the t-butyl alcohol freezedrying method. Arch. Histol. Cytol. 51, 53–61
Chapter 17 Microsomal Proteomics Diana M. Wong and Khosrow Adeli Summary Proteomic profiling of subcellular compartments has many advantages over traditional proteomic approaches using whole cell lysates as it allows for detailed proteome analysis of a specific organelle and corresponding functional characteristics. The microsome is a critical, membranous compartment involved in the synthesis, sorting, and secretion of proteins as well as other metabolic functions. This chapter will describe detailed methods for the isolation of microsomal organelles including the ER, Golgi, and prechylomicron transport vesicle (PCTV), a recently identified vesicular system involved in intestinal lipoprotein assembly and secretion. Particular focus is given to the isolation of microsomes from primary hepatocytes and enterocytes freshly isolated from rodent liver and intestinal tissue, and their proteomic profiling using a combination of two-dimensional gel electrophoresis and mass spectrometry. Key words: Microsome, Endoplasmic reticulum, Golgi apparatus, Prechylomicron transport vesicle (PCTV), Proteomics, Isoelectric focusing, 2D gel electrophoresis, Mass spectrometry.
1. Introduction Proteomics involves the integration of technologies to analyse the complete complement of proteins expressed (referred to as the proteome) by a biological system in response to stimuli under different physiological/pathological conditions. Examining changes in the proteome offers insight into understanding cellular and molecular mechanisms that cannot be obtained through genomic analysis. The information gap between a genome and associated gene products is largely attributed to post-translational modifications. These modifications have been shown to modulate pivotal regulatory processes such as protein turnover, protein activity, and protein localization David Sheehan and Raymond Tyther (eds.), Methods in Molecular Biology, Two-Dimensional Electrophoresis Protocols, vol. 519 © Humana Press, a part of Springer Science + Business Media, LLC 2009 DOI: 10.1007/978-1-59745-281-6_17
273
274
Wong and Adeli
within a cell. In addition, virtually all known cellular signalling pathways are mediated through a complex cascade of reversible protein phosphorylation. The proteome is a dynamic feature, subject to changes due to developmental stage, disease state, or environmental conditions. Currently, proteomic protocols commonly incorporate two-dimensional electrophoresis (2DE) for protein separation, where proteins are detected with various stains, and the protein profile is analysed using 2DE gel imaging software. Proteins of interest are identified by matrix-assisted laser desorption/ionization-time-of-flight (MALDI-TOF) or tandem mass spectrometry (MS/MS). An important goal of proteomics research is to deposit the results into the proteome database, which contains both protein identification and corresponding functional characteristics. However, 2DE is not an effective technique for profiling membrane proteins because they are not easily solubilized for isoelectric focusing (IEF) (1), and they have difficulty entering the gel due to their size and hydrophobicity (2, 3). (see also Chapter “Difficult Proteins”). An alternative approach to identify membrane proteins is to digest in-solution and then apply to liquid chromatographytandem mass spectrometry (LC-MS/MS). Tandem mass spectra of the peptides can be searched in databases to identify amino acid sequences (4, 5). In addition, 1-dimensional blue native polyacrylamide gel electrophoresis (1-D BN-PAGE; see Chapter 15) is an in-gel technique used to study membrane proteins. Coomassie dye in the sample buffer induces a charge shift in the membrane proteins. Aminocaproic acid also helps in their solubilization (6). Current research using proteomics is focused on two main areas: expression proteomics, which measures the fluctuation of a protein level under certain conditions; and functional proteomics, which characterizes proteins in organelles and complexes. Expression proteomics involves investigating proteins altered in a disease or drug-treated state, compared to normal, to discover diagnostic markers or therapeutic targets (7). However, studies of whole proteomes are not representative of all the proteins contained in the cell since low-abundance proteins are not easily identified (8). A common approach used to enrich low-abundance proteins is to isolate individual subcellular compartments (see Chapters “Organelle Proteomics,” “Preparation and Analysis of Plastid Proteomes by 2DE” and “2DE for Proteome Analysis of Human Metaphase Chromosomes”). Using subcellular fractionation methods, individual organelles can be isolated and their protein complements resolved by proteomics. This method is particularly useful in the investigation of specific subcellular organelle(s) thought to be affected in a disease state. Morand et al. used this comparative approach to identify proteins in hepatic
Microsomal Proteomics
275
endoplasmic reticulum (ER) that were altered in nutritionally induced insulin resistance (9). Microsomes are among the cell’s most active membranous structures involved in the synthesis, sorting, and secretion of proteins as well as other metabolic functions. The microsome consists of a complex network of continuous membranes including ER, ER-Golgi intermediate complex (ERGIC) – also referred to as the vesiculotubular clusters or pre-Golgi intermediates – and the Golgi apparatus. These organelles are membrane-bound compartments that have distinct functions. The trafficking of proteins and lipids in the microsome is mediated by vesicles that bud from the ER and fuse with the cis-Golgi compartment. COPII-coated and COPI-coated vesicles transport proteins from the ER to the Golgi, where the COPI-coated vesicles are also capable of retrograde movement (10). The ERGIC, a membrane system between rough ER and Golgi, is a transient cargo holder of vesicles (11, 12). COPIIcoated vesicles also play an important role in anterograde movement from the ER to the ERGIC, and COPI-coated vesicles play a well-established role in retrograde traffic from the Golgi to ERGIC (13, 14). Studies in the small intestine have recently identified a unique intermediate vesicle compartment, prechylomicron transport vesicles (PCTVs), involved in intestinal packaging of absorbed lipids and formation of chylomicrons (15, 16). Recent proteomic analysis of the microsomal preparations has helped identify many unique microsomeassociated proteins. Specifically, 141 proteins have been identified in the ER (17), 24 proteins have been identified in the ER-GIC (18), and over 400 proteins have been identified in the Golgi (19, 20). The ER and Golgi compartments are in a dynamic equilibrium involving continuous formation and fusion of secretory vesicles in both antegrade and retrograde directions. This results in a continuous change of molecular composition and makes pure isolation of distinct microsomal fractions very challenging. There are several approaches to examine the purity of organelle preparations. A simple way is to immunoblot for proteins which could easily contaminate. For example, when working with ER, is it common to immunoblot for Golgi and lysosomal protein markers. ‘Subtractive proteomics’ is a more advanced technique, which compares a specific cellular compartment of interest to a ‘background’ complex (21). In 2003, Schirmer et al. verified proteins in the nuclear envelope by subtracting proteins also found in the microsomal membrane (22). The following chapter will describe proteomic techniques used to study microsomal fractions. We will focus on microsomal proteomics in primary cultured hepatocytes and enterocytes. Techniques for primary cell isolation, subcellular fractionation,
276
Wong and Adeli
2DE, and MS will be described. In addition, we describe the methodology for isolation and proteomic profiling of the recently identified microsomal compartment, PCTV.
2. Materials Microsomal proteomic analysis can be performed on primary cells from any animal model. Here, male Syrian golden hamsters (Mesocricetus auratus) (Charles River, Montreal, QC, Canada) maintained on a chow diet (Dyets Inc, Bethlehem, PA) are used. An overview of the entire procedure is given in Fig.1.
Intestine
Liver
Preparation of microsomal fraction from liver/intestinal primary cells (microsome, ER, Golgi, or PCTV)
IEF, SDS-PAGE (2 gels)
Fixation and visualization of protein spots
In-solution trypsin digestion
Direct injection of sample into LC-MS/MS
HPLC Image acquisition
High voltage
C18 column
Mass Spectrometer
Software analysis Spot excision
In-gel trypsin digestion mass spectrometry identification
Protein sequencing via tandem mass spectrometry
Database searching
Fig. 1. A schematic overview of the protocols used to perform proteomic analysis of microsomal fractions. Following isolation of primary cells from rodent tissue, the microsomal fraction can be analysed by 2DE followed by MS, or organellar extract subjected to direct trypsin digestion followed by direct injection of the peptide mixture into the LC-MS/MS. Database searching can identify the proteins of interest.
Microsomal Proteomics
2.1. Animal Surgery
277
1. Anaesthesia (isofluorane, nitrous oxide, and oxygen, Baxter, Toronto, ON) and chamber. 2. Surgical tools.
2.1.1. Primary Cell Isolation: Hepatocytes
1. Liver perfusion, digest, wash media (Life Technologies, Gaithersburg, MO). 2. Peristaltic pump (Amersham Biosciences, Piscataway, NJ). 3. Scissors, cell strainer. 4. 1× PBS
2.1.2. Primary Cell Isolation: Enterocytes
1. 100-mm Petridish, 18-gauge needle and syringe. 2. 1× PBS. 3. Cell Recovery Solution (BD Biosciences) (see Note 1). 4. Orbital shaker.
2.2. Microsome Isolation
1. Primary cells. 2. Buffer A (10 mM HEPES, pH 7.2, 0.25 M sucrose 2 mM EDTA) and protease inhibitor cocktail (PI) (Complete-Mini, EDTA-free, Roche, Mississauga, ON, Canada). 3. Homogenizer (Parr bomb + N2 pressure) (Parr Instruments, Moline IL). 4. Ultracentrifuge (Beckman Optima LE-80, SW41Ti rotor, Mississauga, ON, Canada). 5. 12-mL ultracentrifuge tubes (Beckman).
2.2.1. Microsome Isolation for Prechylomicron Transport Vesicle Budding Assay
1. Buffer B (137 mM NaCl, 1.5 mM EDTA, 11.5 mM KH2PO4, 8 mM Na2HPO4, 2.2 mM KCl, 0.5 mM dithiothreitol (DTT), pH 7.2) supplemented with 10 mM glutamine. 2. 50 μCi [3H]-oleic acid (PerkinElmer, Boston, MA). 3. Bovine serum albumin (BSA) (Sigma Aldrich, Oakville, ON, Canada). 4. Sodium oleate (Sigma Aldrich). 5. 1× PBS.
Cytosol Preparation
1. Buffer C (25 mM HEPES, pH 7.2, 125 mM KCl, 2.5 mM MgCl2, 2 mM DTT, 0.5 mM EDTA) and PI. 2. Amicon Ultra 15 Centrifugal Filter Devices, 10,000 Mr cutoff (Millipore, Billerica, MA). 3. Anti-enolase antibody (Santa Cruz Biotechnology).
2.2.2. Endoplasmic Reticulum and Golgi Apparatus Isolations
1. 0.25 M, 0.86 M, 1.15 M, 1.22 M sucrose dissolved in 10 mM HEPES, pH 7.2. 2. Anti-calnexin antibody (Santa Cruz Biotechnology, Santa Cruz, CA). 3. Anti-GS28 antibody (Stressgen Bioreagents, Ann Arbor, MI).
278
Wong and Adeli
PCTV Budding Assay
1. Cytosol buffer (25 mM HEPES, 125 mM KCl, 2.5 mM MgCl2, 2.5 mM DTT, pH 7.2). 2. ATP mixture (5 mM ATP, 25 mM phosphocreatine, 25 units phosphokinase) in 10 mM HEPES, pH 7.2 (see Note 2). 3. 2.5 mM Mg2+, 2.5 mM Ca2+, 5 mM DTT. 4. E600 prep (200 μL E600 stock in 5 mL 10 mM HEPES, pH 7.2) (Sigma Aldrich). 5. Transport Buffer (30 mM HEPES, 0.25 M sucrose, 2.5 mM MgCl2, 30 mM KCl, pH 7.2). 6. 0.10 M and 1.15 M sucrose in 10 mM HEPES, pH 7.2. 7. Gradient maker and stir bar.
2.3. Isoelectric Focusing
1. Lysis buffer (0.25 M sucrose, 10 mM Tris–HCl, pH 7.4) and PI. 2. Acetone. 3. Reagent 3 from ReadyPrep Sequential Extraction Kit (BioRad, Hercules, CA). 4. IPGphore IEF Unit (Amersham Biosciences, Piscataway, NJ). 5. 24-cm pH 3–10 NL Immobiline Drystrips (Amersham Biosciences). 6. Rehydration buffer (8 M urea, 0.5% CHAPS, 15 mM DTT, 0.2% Pharmalyte 3–10, 4–7, or 6–11) and bromophenol blue. 7. Immobiline Drystrip holder (Amersham Biosciences). 8. Drystrip Cover Fluid (Amersham Biosciences). 9. Equilibration buffer (6 M urea, 30% glycerol and 2% SDS in 0.05 M Tris–HCl buffer). 10. DTT, iodoacetamide.
2.4. 2-Dimensional Polyacrylamide Gel Electrophoresis
1. Ettan DALTsix large format PAGE tank (Amersham Biosciences). 2. Stacking gel. 0.5% Agarose made up in 1× running buffer. 3. Resolving gel. 8% sodium dodecyl sulphate (SDS) polyacrylamide gel. (a) M Tris–HCl, pH 8.8, 0.4% SDS. (b) 40% acrylamide/bis solution (29:1 with 3.3%C). (c) N,N,N,N¢-Tetramethyl-ethylenediamine (TEMED) (BioRad). (d) Ammonium persulphate (APS): prepare 1% solution in water. 4. 1× Running buffer (25 mM Tris–HCl, 192 mM Glycine, 0.1% SDS, pH 8.5). 5. Prestained molecular weight markers (Fermentas, Burlington, ON, Canada).
Microsomal Proteomics
279
6. Deep Purple Total Protein Stain (Amersham Biosciences). 7. Typhoon™ 9400 Imager (Amersham Biosciences). 2.5. Mass Spectrometry Sample Preparation
1. Pipette tips, scalpel, syringe. 2. 30 mM ammonium bicarbonate/40% acetonitrile. 3. 50 mM ammonium bicarbonate/1 mM CaCl2 solution. 4. 50% acetonitrile/1% trifluoroacetic acid (TFA). 5. 20% formic acid/15% 2-propanol/25% acetonitrile. 6. 80% acetonitrile. 7. 0.1% trifluoroacetic acid. 8. 0.8 M guanidine chloride/2.5% TFA. 9. SpeedVac centrifuge (Savant, Tamsey, MA). 10. Sequencing-grade trypsin (Promega, Madison, WI) in 50 mM acetic acid. 11. C18 ZipTip™ (μZT) pipette tips (Millipore). 12. α-cyano-4-hydroxycinnamic acid (CHCA), 10 mg/mL. 13. Micromass MALDI-Q-ToF (Waters, Milford, MS).
2.6. Liquid Chromatography Tandem Mass Spectrometry Sample Preparation
1. Digestion buffer. 50 mM NH4HCO3. 2. 6 M urea/2 M thiourea solution. 6 M urea, 2 M thiourea, 10 mM Tris–HCl, 150 mM NaCl, 1 mM PMSF, pH 8.0 with protease inhibitor cocktail (PI). 3. 80 mM DTT stock solution. 4. 300 mM iodoacetamide stock solution. 5. LysC stock solution. To make 1 mL, dilute 0.5 μg in 1 mL digestion buffer. Separate into aliquots and store at −20°C. 6. Sequencing-grade trypsin (Promega). 7. Mobile phase (e.g. 10 mM ammonium acetate–methanol– acetonitrile (30:35:35)). 8. Thermo Electron LCQ Deca XP (Thermo Scientific, San Jose, CA) coupled with an Agilent capillary HPLC 1100 series (Agilent Technologies, Palo Alto, CA, USA) and C18 column (150 mm × 4.6 mm, 5 μm, Zorbax Extend).
3. Methods The proteomic methods described can be used on microsomes isolated from different tissues. In our laboratory we typically use hepatocytes or enterocytes from the Syrian golden hamster; however, this protocol should be applicable to other rodents. The procedures for microsome (Subheading 3.2), ER, and Golgi isolations
280
Wong and Adeli
(Subheading 3.2.1) have been adapted but significantly modified from a published protocol (23). The protocols for ER (Subheading 3.2.2) and cytosol (see “PCTV Budding Assay” in Subheading 3.2.2) isolations for PCTV budding assay, as well as PCTV budding assay (Subheading 3.2.2) have been previously published (24). An overview of cell fractionation can be seen in Fig. 2. 3.1. Animal Surgery
1. Maintain male Syrian golden hamsters on a chow diet for at least 3 days prior to use and then fast overnight prior to killing. 2. Anaesthetize the animal with a continuous flow of the anaesthetic mixture. 3. Make an incision up the middle of the animal’s abdomen and prepare for liver perfusion.
Microsomal Fractionation
PCTV Budding Assay
Homogenate Incubate
8,500g, 10min
35-37°C PNS
Nuclear pellet
100,000g, 3h
cytosol
ER + cytosol + ATP
PCTVs bud from ER
microsome pellet
wash, 4,000g
0.1M sucrose
concentrated cytosol
2.6mL 0.25 M sucrose 2.6mL 0.86 M sucrose 2.6mL 1.15 M sucrose 3.0mL 1.22 M sucrose
1.15M sucrose 80 000g, 1.5h
80,000g, 3h PCTVs
\\
\
\
trans-Golgi (0.25 / 0.86M) cis-Golgi (0.86/1.15M) smooth ER (1.22M) rough ER
ER
Fig. 2. Protocols for preparation of total microsomal membranes and the PCTV budding assay. (a) Following the isolation of primary cells, centrifuge the homogenate at low speed in the ultracentrifuge to remove nuclei, mitochondria, and unbroken cells. Take the post-nuclear supernatant and spin at high speed to remove cytosol. Overlay the resulting microsomal pellet with a discontinuous sucrose gradient to isolate ER and Golgi. (b) Incubate ER with cytosol, ATP, and other factors to bud the PCTVs out of the ER. Overlay the PCTV mixture onto a continuous sucrose gradient and spin to purify the budded PCTVs.
Microsomal Proteomics 3.1.1. Primary Cell Isolation: Hepatocytes
281
1. Insert a needle into the abdominal vena cava and secure with sutures. 2. Isolate the liver from the circulatory system by blocking the thoracic aorta, caudal vena cava, abdominal aorta, and abdominal vena cava with sutures. 3. Pump 50 mL of perfusion solution at 39°C into the liver using the peristaltic pump at the highest setting. 4. Snip the portal vein to allow perfusion media to escape the liver. 5. Pump 50 mL of liver digest medium at 39°C into the liver. 6. Cut out the liver and dice with scissors in wash media. 7. Pass the perfused liver through a cell strainer. 8. Wash the hepatocytes three times with 1× PBS to remove collagenase. 9. Centrifuge the hepatocytes at 100 × g to pellet.
3.1.2. Primary Cell Isolation: Enterocytes
1. Excise approximately 10 cm of the proximal end of the small intestine and place in a 100-mm dish of ice-cold 1× PBS. 2. Rinse the intestine several times with 1× PBS using an 18-gauge needle. 3. Cut the intestine longitudinally and subsequently cut into 1-cm fragments. 4. Immerse intestinal fragments in Cell Recovery Solution for 1 h at 4°C. 5. Wash the enterocytes using 1× PBS with agitation on an orbital shaker for 5 min. 6. On the shaker, add 5 mL of 1× PBS to the fragments and use fingers to tap intestine gently. At this time, villi will start dissociating from the intestinal fragments. 7. Collect the PBS with the suspended villi and repeat 4–5 times or until villi no longer dissociate from the intestinal fragments. 8. Pellet the suspended villi with a 3-min centrifugation at 200 × g.
3.2. Microsome Isolation
1. Resuspend the cell pellet in 10–20 mL cold buffer A with PI (see Note 3). 2. Homogenize the cells using a Parr bomb cell disruption vessel at 800 psi N2 pressure for 35 min at 4°C. 3. Transfer the homogenate to a 12-mL ultracentrifuge tube and spin at 8,500 × g for 10 min at 4°C in an ultracentrifuge (Beckman Optima LE-80, SW41Ti rotor) to remove nuclei and mitochondria. 4. Take the post-nuclear supernatant (PNS) and centrifuge at 100,000 × g for 3 h at 4°C. 5. Discard the supernatant and retain microsome pellet.
282
Wong and Adeli
3.2.1. Microsome Isolation for PCTV Budding Assay
1. Resuspend 20 × 106 enterocytes in 20 mL cold buffer B supplemented with 10 mM glutamine and PI. 2. In a separate tube, add 50 μCi [3H]-oleic acid to 200 μL 10% BSA-10 mM oleate complex (see Note 4). Vortex and add to cells from step 1. 3. Incubate enterocytes at 37°C for 30 min with occasional mixing. 4. Centrifuge the cells at 1,000 × g and discard the supernatant. 5. Wash the pellet twice with 2% BSA in cold PBS. 6. Discard the supernatant and resuspend cells in 10–20 mL cold buffer A with PI. 7. Follow steps 2–5 from Subheading 3.2.
Cytosol Preparation
1. Resuspend enterocytes in 10–20 mL cold buffer C with PI. 2. Homogenize the cells using a Parr bomb cell disruption vessel at 1,000 psi N2 pressure for 40 min at 4°C. 3. Transfer the homogenate to a 12-mL ultracentrifuge tube and centrifuge at 8,500 × g for 10 min at 4°C in an ultracentrifuge (Beckman Optima LE-80, SW41Ti rotor). 4. Take the post-nuclear supernatant and centrifuge at 100,000 × g for 3 h at 4°C. 5. Take the supernatant (avoid lipid layer) and place in Amicon Ultra 15 Centrifugal tube. 6. Centrifuge at 4,000 × g until volume is less than 1 mL. 7. Add buffer C (no PI) and centrifuge until volume is less than 1 mL. 8. Repeat step 7 and concentrate until protein concentration in cytosol fraction is ∼20 mg/mL. 9. Prepare an equal volume cytosol and whole cell lysate as a positive control, for electrophoresis on 8% SDS-PAGE and transfer to PVDF membrane. 10. Immunoblot the membrane with anti-enolase antibody to determine if cytosol is present. 11. Immunoblot the membrane with anti-calnexin and/or antiGS28 antibody to verify that cytosol is not contaminated with ER and/or cis-Golgi.
3.2.2 ER and Golgi Apparatus Isolations
1. Resuspend the microsome pellet in 3 mL of 1.22 M sucrose solution. 2. In the ultracentrifuge tube, carefully overlay the microsomal suspension with 2.6 mL each of 1.15 M, 0.86 M, and 0.25 M sucrose solutions. 3. Centrifuge the sucrose step gradient at 82,000 × g for 3 h at 4°C (SW41Ti rotor).
Microsomal Proteomics
283
4. Take the pellet and 1.22 M sucrose layer as rough and smooth ER, respectively. The cis- and trans-Golgi membranes float at the 0.25/0.86 M and 0.86/1.15 M density interfaces, respectively (24). Remove by Pasteur pipette. 5. For highly purified ER, adjust the ER density to 1.22 M sucrose and repeat steps 2 and 3. 6. Fractionate by unloading 22 × 500 μL aliquots from the top of the tube. 7. Prepare equal volumes of each fraction, along with whole cell lysate as a positive control, for electrophoresis on 8% SDS-PAGE and transfer to PVDF membrane. 8. Immunoblot the membrane with anti-calnexin antibody to determine which fractions contain ER. 9. Immunoblot the membrane with anti-GS28 antibody to determine which fractions contain cis-Golgi and to verify ER is not contaminated with cis-Golgi. PCTV Budding Assay
1. Use a Bradford assay to determine the concentration of the ER and cytosol fractions (25). 2. Take 400–500 μg of ER protein (∼50 μL of a 10 mg/mL ER fraction) and combine with 40 μL of cytosol (40 μL of cytosol buffer for negative control), 100 μL of ATP mixture, 25 μL of 2.5 mM Mg2+, 25 μL of 2.5 mM Ca2+, 50 μL of 5 mM DTT, 10 μL E600 prep, and 200 μL of transport buffer for a total volume of 500 μL. 3. Agitate slightly and incubate at 35–37°C for 35 min. Remove all air bubbles. 4. During incubation, prepare a continuous gradient of 0.1–1.15 M sucrose (10 mL total). 5. Stop the reaction by placing tubes on ice. Add 700 μL cold 10 mM HEPES. 6. Overlay suspension on gradient. 7. Centrifuge at 80,000 × g (Beckman SW41Ti rotor) for 95 min. 8. Remove the top 100 μL-containing cytosolic proteins and discard. 9. Fractionate by unloading 23 × 500 μL aliquots from the top of the tube. 10. Take 100 μL of each fraction and count the DPM for 2–3 min. 11. Plot DPM vs. fraction. Pool fractions with PCTVs.
3.3. Isoelectric Focusing
1. Add 2 volumes of lysis buffer to 500 μg of microsomal protein. 2. Lyse with 25-gauge syringe ten times. 3. Centrifuge at 15,000 × g for 10 min.
284
Wong and Adeli
4. Take the supernatant. Add 1 volume of acetone. 5. Incubate at −20°C for 2 h. 6. Centrifuge at 15,000 × g for 10 min. 7. Discard the supernatant. Solubilize the pellet in 30 μL of BioRad Reagent 3, which can dissolve the most insoluble proteins. Shake 5 min. 8. Add 420 μL rehydration buffer and a pinch of bromophenol blue. 9. Place (+) acrylamide side of 24 cm Immobiline Drystrip gel face down in drystrip holder. Remove all air bubbles. 10. Cover with 5 mL Drystrip Cover Fluid. 11. Seal with Saran wrap and let sit for 12 h. 12. Place strip (+) side upwards, polyacrylamide side up onto IEF unit. 13. Attach metal brackets to contact acrylamide at each end of the strip. 14. Overlay with 4 mL of cover fluid. 15. Use the following Step and Hold pattern: (a) Step and Hold 30 V for 1 h (b) Step and Hold 150 V for 1 h (c) Step and Hold 300 V for 1 h (d) Step and Hold 1,000 V for 1 h (e) Step and Hold 2,500 V for 1 h (f) Gradient 8,000 V for 1 h (f) Step and Hold 8,000 V for 3 h (h) Step and Hold 150 V for 40 h 16. Incubate the IPG strip at room temperature in equilibration buffer with 1% DTT for 15 min. Wash 3 × 10 min with distilled water. 17. Incubate the IPG strip in equilibration buffer with 4.8% iodoacetamide for 15 min. Wash 3 × 10 min with distilled water. 3.4. 2-Dimensional Polyacrylamide Gel Electrophoresis
1. Use instructions in Ettan DALTsix electrophoresis unit to prepare the 8% SDS-PAGE gel. In short, mix distilled water, 1% APS, 1.5 M Tris–HCl, 40% acrylamide, and TEMED. Pour solution between glass plates and leave space for a stacking gel. Overlay with distilled water. 2. After 30 min, prepare the stacking gel by dissolving 0.5% agarose in 1× running buffer. Allow gel to cool to lukewarm temperature. Seal the IPG strip in agarose. Insert a single-well lane from a well comb for the prestained molecular weight marker. The stacking should polymerize within a few minutes.
Microsomal Proteomics
285
3. Insert the gel(s) into the PAGE tank. Fill the upper and lower chambers with 1× running buffer. 4. Carefully remove the well lane. Load the lane with the molecular weight marker. 5. Complete the assembly of the gel unit and connect to a power supple. Run at 60 V until dye front reaches the bottom. 6. Following electrophoresis, stain the gel with Deep Purple Total protein Stain according to the manufacturer’s protocol (Amersham Biosciences). 7. Use a Typhoon 9400 imager to view the 2DE gels using wavelengths of 457 nm for excitation and 610 nm for emission. 3.5. Mass Spectrometry Sample Preparation
1. Cut a pipette tip to size with a scalpel. Use this to excise protein spots from the 2DE gel. 2. Eject the protein plug from the pipette tip into a 2-mL tube by inserting another pipette tip to pop it out. At this point the gel plug can be stored at −20°C if necessary. 3. Crush the gel plug in the bottom of the tube with a plunger from a small syringe. 4. Destain the plug with 5 × 15 min washes of 1 mL 30 mM ammonium bicarbonate/40% acetonitrile on an orbital shaker. 5. Dry the gel in a SpeedVac centrifuge until it turns to a powder (approximately 1.5 h). 6. Dissolve 2 μL of 0.2 μg/μL modified sequencing-grade trypsin (Promega) in 50 mM acetic acid, warmed first to 30°C for 15 min, and add to the dried gel. 7. Next, add 50 μL of 50 mM ammonium bicarbonate. Put the tube on ice for 1 h. 8. After the gel has taken up all the liquid, add an additional 25 μL of 50 mM ammonium bicarbonate/1 mM CaCl2 solution. It is important to keep the gel pieces in excess liquid for the digestion process. Place the tube in a water bath set at 37°C for 18 h. 9. After the digestion, take the supernatant and set aside in a 1.5-mL tube. 10. Treat the remaining gel with 50 μL of the following solutions: 50 mM ammonium bicarbonate for one wash, 50% acetonitrile/1% trifluoroacetic acid for two washes, 20% formic acid/15% 2-propanol/25% acetonitrile for one wash, and 80% acetonitrile for one wash. After adding the solution for each wash, vortex the tube for 10 min, spin down, and then let it sit at room temperature for 15 min. 11. Extract the supernatant and pool with the first extract in the 1.5-mL tube. Dry the pooled extract in a SpeedVac for
286
Wong and Adeli
approximately 2 h. The samples can be stored at −20°C this point if necessary. 12. Add 20 μL of 0.8 M guanidine chloride/2.5% TFA to the dried sample and sit at room temperature for 15 min. 13. Sample clean-up using μZT should not be performed more than one day before MS analysis. Wet the μZT by two cycles of 50% acetonitrile solution and equilibrate with two cycles of 0.1% trifluoroacetic acid. 14. Take 5 μL of the solubilized sample and insert into the μZT for MALDI-TOF mass fingerprinting. The peptides in the sample will bind to the C18 beads by fifteen cycles of the sample solution. 15. Wash the peptides bound to the beads by five cycles of 0.1% trifluoroacetic acid solution. 16. Elute the peptides from the column by fifteen cycles of 5 μL of 0.1% trifluoroacetic acid/50% acetonitrile solution. Do not introduce any air bubbles into the column. 17. Spot 0.3 μL of the sample on the MALDI target and allow to dry. Next, spot 0.3 μL of the CHCA matrix solution on top of the sample and allow to dry. Spot each sample in triplicate. 18. Analyse the extract by applying it to a Micromass MALDIQ-TOF. The resulting peptide map can be searched against the Profound protein database. PROWL ProFound search engine available at http://prowl.rockefeller.edu/, using a focused rodent or mammalian search. 19. If proteins cannot be identified through database mining, tandem mass spectrometry (Chapter “Shotgun Protein Analysis by Liquid Chromatography-Tandem Mass Spectrometry”) can be used to sequence and identify the proteins. The resulting spectra can then be used to identify protein candidates in the National Center for Biotechnology Information (NCBI) non-redundant protein sequence database with the MASCOT search engine (Matrix Science, London). 3.6. Liquid Chromatography Tandem Mass Spectrometry Sample Preparation
1. Dissolve 50 μg protein in 6 M urea/2 M thiourea solution with protease inhibitor cocktail (PI). 2. Lyse with 25-gauge syringe ten times. 3. Add 1 volume of DTT stock solution and incubate at room temperature for 30 min. 4. Add 1 volume of iodoacetamide stock solution and incubate at room temperature for 20 min.
Microsomal Proteomics
287
5. Add 1 μg LysC and incubate for 3 h or overnight at room temperature. 6. Dilute sample with 4 volumes of digestion buffer. Add 1 μg trypsin and incubate overnight at room temperature. Digested peptides can be stored at −20°C. 7. Dry the sample in a SpeedVac. 8. Reconstitute the peptides in 150 μL mobile phase and 20 μL injected into the LC-MS/MS system. The mass spectrometer is equipped with an ion-spray source and operates in the positive ion mode. 9. The resulting spectra can be used to identify protein candidates in the National Center for Biotechnology Information (NCBI) non-redundant protein sequence database with the Mascot search engine (Matrix Science, London).
4. Notes 1. Store Cell Recovery Solution, Buffers A-C, sucrose buffers, cytosol and transport buffers, and resolving gel buffer at 4°C. 2. For PCTV Budding Assay, store E600 prep at −20°C and ATP mixture at −80°C. Also, store PCTVs at −80°C. 3. To prepare protease inhibitor cocktail (PI), dissolve one tablet in 2 mL of distilled water and use as 1:50 (PI cocktail: buffer). 4. To prepare 10% BSA-10 mM oleate complex: (a) Dissolve 1 g BSA in 9 mL of 10 mM HEPES, pH 7.2. (b) In a separate tube, dissolve sodium oleate in 1 mL of lukewarm distilled water. (c) Add oleate to BSA mixture. Vortex.
Acknowledgments The authors would like to thank Dr. Charles Mansbach II for his technical guidance and training of DMW, particularly in the isolation of intestinal PCTVs. This work was supported by operating grants from the Canadian Institutes for Health Research Heart (CIHR) and the Heart and Stroke Foundation of Ontario to KA. DMW holds a postgraduate scholarship from the Natural Sciences and Engineering Research Council of Canada (NSERC).
288
Wong and Adeli
References 1. Lescuyer, P., Strub, J. M., Luche, S., Diemer, H., Martinez, P., Van Dorsselaer, A., Lunardi, J. and Rabilloud, T. (2003) Progress in the definition of a reference human mitochondrial proteome. Proteomics. 3, 157–167. 2. Molloy, M. P., Herbert, B. R., Walsh, B. J., Tyler, M. I., Traini, M., Sanchez, J. C., Hochstrasser, D. F., Williams, K. L. and Gooley, A. A. (1998) Extraction of membrane proteins by differential solubilization for separation using two-dimensional gel electrophoresis. Electrophoresis. 19, 837–844. 3. Molloy, M. P., Herbert, B. R., Slade, M. B., Rabilloud, T., Nouwens, A. S., Williams, K. L. and Gooley, A. A. (2000) Proteomic analysis of the Escherichia coli outer membrane. Eur. J. Biochem. 267, 2871–2881. 4. Eng, J. K., McCormack, A. L. and Yates, J. R., III (1994) An approach to correlate tandem mass spectral data of peptides with amino acid sequences in a protein database. J. Am. Soc. Mass Spectrom. 5, 976–989. 5. McCormack, A. L., Schieltz, D. M., Goode, B., Yang, S., Barnes, G., Drubin, D. and Yates, J. R., III. (1997) Direct analysis and identification of proteins in mixtures by LC/MS/MS and database searching at the low-femtomole level. Anal Chem. 69, 767–776. 6. Babusiak, M., Man, P., Petrak, J. and Vyoral, D. (2007) Native proteomic analysis of protein complexes in murine intestinal brush border membranes. Proteomics. 7, 121–129. 7. Monti, M., Orru, S., Pagnozzi, D. and Pucci, P. (2005) Interaction proteomics. Bioscience Reports. 25, 45–56. 8. Jung, E., Heller, M., Sanchez, J. C. and Hochstrasser, D. F. (2000) Proteomics meets cell biology: the establishment of subcellular proteomes. Electrophoresis. 21, 3369–3377. 9. Morand, J. P., Macri, J. and Adeli, K. A. (2005) Proteomic profiling of hepatic endoplasmic reticulum-associated proteins in an animal model of insulin resistance and metabolic dyslipidemia. J. Biol. Chem. 280, 17626–17633. 10. Appenzeller-Herzog, C. and Hauri, H. P. (2006) The ER-Golgi intermediate compartment (ERGIC): in search of its identity and function. J. Cell Sci. 119, 2173–2183. 11. Hauri, H. P., Kappeler, F., Andersson, H. and Appenzeller, C. (2000) ERGIC-53 and traffic in the secretory pathway. J. Cell Sci. 113, 587–596. 12. Schweizer, A., Fransen, J. A., Bachi, T., Ginsel, L. and Hauri, H. P. (1988) Identification, by a monoclonal antibody, of a 53-kD protein associated with a tubulovesicular compart-
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
ment at the cis-side of the Golgi apparatus. J. Cell Biol. 107, 1643–1653. Klumperman, J., Schweizer, A., Clausen, H., Tang, B. L., Hong, W., Oorschot, V. and Hauri, H. P. (1998) The recycling pathway of protein ERGIC-53 and dynamics of the ERGolgi intermediate compartment. J. Cell Sci. 111, 3411–3425. Letourneur, F., Gaynor, E. C., Hennecke, S., Demolliere, C., Duden, R., Emr, S. D., Riezman, H. and Cosson, P. (1994) Coatomer is essential for retrieval of dilysinetagged proteins to the endoplasmic reticulum. Cell 79, 1199–1207. Siddiqi, S. A., Gorelick, F. S., Mahan, J. T. and Mansbach, C. M., II. (2002) COPII proteins are required for Golgi fusion but not for endoplasmic reticulum budding of the prechylomicron transport vesicle. J. Cell Sci. 116, 415–427. Siddiqi, S. A., Mahan, J. T., Siddiqi, S., Gorelick, F. S. and Mansbach, C. M., II. (2006) Vesicleassociated membrane protein 7 is expressed in intestinal ER. J. Cell Sci. 119, 943–950. Knoblach, B., Keller, B. O., Groenendyk, J., Aldred, S., Zheng, J., Lemire, B. D., Li, L. and Michalak, M. (2003) ERp19 and ERp46, new members of the thioredoxin family of endoplasmic reticulum proteins. Mol. Cell. Proteomics. 2, 1104–1119. Breuza, L., Halbeisen, R., Jeno, P., Otte, S., Barlowe, C., Hong, W. and Hauri, H. P. (2004) Proteomics of endoplasmic reticulumGolgi intermediate compartment (ERGIC) membranes from brefeldin A-treated HepG2 cells identifies ERGIC-32, a new cycling protein that interacts with human Erv46. J. Biol. Chem. 279, 47424–47253. Taylor, R. S., Wu, C. C., Hays, L. G., Eng, J. K., Yates, J. R., III and Howell, K. E. (2000) Proteomics of rat liver Golgi complex: minor proteins are identified through sequential fractionation. Electrophoresis. 21, 3441–3459. Wu, C. C., MacCoss, M. J., Mardones, G., Finnigan, C., Mogelsvang, S., Yates, J. R., III and Howell, K. E. (2004) Organellar proteomics reveals Golgi arginine dimethylation. Mol. Biol. Cell. 15, 2907–2919. Andersen, J. S. and Mann, M. (2006) Organellar proteomics: turning inventories into insights. EMBO reports. 7, 874–879. Schirmer, E. C., Florens, L., Guan, T., Yates, J. R. and Gerace, L. (2003) Nuclear membrane proteins with potential disease links found by subtractive proteomics. Science. 301, 1380–1382.
Microsomal Proteomics 23. Kumar, N. S. and Mansbach, C. M., II. (1997) Determinants of triacylglycerol transport from the endoplasmic reticulum to the Golgi in intestine. Am. J. Physiol. 273, G18–G30. 24. Kumar, N. S. and Mansbach, C. M., II. (1999) Prechylomicron transport vesicle: isolation and
289
partial characterization. Am. J. Physiol. 276, G378–G386. 25. Bradford, M. (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72, 248–254.
Chapter 18 Prefractionation Using Microscale Solution IEF Won-A Joo and David Speicher Summary Proteomics often involves systematic analyses of proteomes that are constantly changing in response to changes in the environment of the cell, tissue, or organism being analyzed. Due to limitations of all current protein profiling methods, powerful, reliable proteome prefractionation methods prior to twodimensional electrophoresis (2DE) gels or alternative non-2DE gel methods are needed for in-depth quantitative comparisons of the complex proteomes typically encountered with samples from higher eukaryotes. The microscale solution isoelectrofocusing (MicroSol IEF) fractionation method is capable of reproducibly dividing complex proteomes into as many as seven well-resolved fractions based on the proteins’ pIs on a small volume scale (∼0.65mL/fraction). When MicroSol IEF is combined with narrow pH range 2DE gels or with alternative downstream analysis methods, it can substantially increase the detection dynamic range and the total number of proteins that can be quantitatively compared. Although MicroSol IEF is reasonably reproducible, subtle variations can occur in different separations similar to the minor variations often seen in most separations of proteins. Therefore, for reliable quantitative comparisons the samples to be compared should be differentially labeled with either Cy dyes or stable isotope labels prior to mixing and separation in a single MicroSol IEF run. Larger numbers of samples can be compared across many MicroSol IEF separations by using a differentially labeled internal standard composed of equal aliquots of all samples to be compared. Key words: MicroSol IEF, Protein prefractionation, ZOOM® IEF Fractionator, Protein profiling, Narrow pH range 2DE gel, 3D protein separations.
1. Introduction Two-dimensional polyacrylamide gel electrophoresis (2DE) remains the most commonly used method for quantitatively comparing changes of proteomes, despite a number of well-known limitations. The first dimension separation utilizes isoelectrofocusing (IEF) in a polyacrylamide gel containing either soluble
David Sheehan and Raymond Tyther (eds.), Methods in Molecular Biology, Two-Dimensional Electrophoresis Protocols, vol. 519 © Humana Press, a part of Springer Science + Business Media, LLC 2009 DOI: 10.1007/978-1-59745-281-6_18
291
292
Joo and Speicher
ampholytes (1) or immobilines (2) under denaturing conditions (see also Chapter “Two-Dimensional Electrophoresis : An Overview”). Unfortunately, existing 2DE methods have inadequate resolution and insufficient dynamic range for the complex proteomes usually encountered when analyzing samples from higher eukaryotes, since even large format gels can only reproducibly resolve approximately 1,000–1,500 protein spots (3, 4). While 2DE gels are often the best method for detecting subtle alterations in specific proteins such as changes in post-translational modifications, oxidative modifications (see also Chapters “Diagonal Electrophoresis for Detection of Protein Disulphide Bridges,” “Detection of 4-Hydroxy-2-Nonenal- and 3-NitrotyrosineModified Proteins Using a Proteomics Approach,” “Proteomic Detection of Oxidized and Reduced Thiol Proteins in Cultured Cells,” “Detection of Ubiquitination in 2DE,” “Phosphoproteome Analysis by in-Gel Isoelectric Focusing and Tandem Mass Spectrometry,” and “Detection of Protein Glutathionylation”), and changes in splice forms, they are less well suited for separating and detecting some types of proteins, including hydrophobic proteins, very small or very large proteins, and very acidic or very basic proteins (see also Chapters “Solubilisation of Proteins in 2DE: An Outline,” “Difficult Proteins,” and “Protein Extraction for 2DE”). Other well-known limitations of 2DE gels include their low throughput, the high level of technical skill required to obtain quality results, and the substantial variability from gel to gel even when gels are prepared by a skilled practitioner. The most effective way to overcome the complexity and wide dynamic range of complex proteomes is to prefractionate samples into a number of less complex mixtures prior to protein profiling. Most prefractionation methods exploit the various physical and chemical properties of proteins such as size, charge, hydrophobicity, and solubility or the highly organized features of cellular components and organelles (5–7). The most important factors for optimal prefractionation are high reproducibility, good protein recovery, and high-resolution separation of complex proteomes into a relatively small number of fractions. One of the most successful prefractionation techniques is IEF, despite the fact that it is not orthogonal to subsequent 2DE gel separation. The major advantages of solution IEF are its high resolution, its compatibility with subsequent 2DE analysis, and the fact that far higher loads of prefractionated samples can produce high-quality separations on narrow pH range gels (see also Chapter “Selection of pH Ranges in 2DE”) compared with unfractionated samples. One of the oldest solution IEF methods is based on the work of Bief and colleagues, and commercialized as the Rotofor apparatus (BioRad Laboratories) (8). The Rotofor separates proteins in a rotating chamber with either 55mL or 18mL sample capacity. The lack of any separation barriers that restrict bulk liquid
Prefractionation Using Microscale Solution IEF
293
flow in the Rotofor makes it very challenging to recover fractions without mixing separated proteins, and as a result resolution is often poor to moderate with single proteins frequently observed to spread over several adjacent fractions (9, 10). Righetti and coworkers subsequently developed a complex solution IEF device, the multi-compartment electrolyzer (MCE) where chambers were separated by thin polyacrylamide membranes buffered at specific pH values and liquid was recirculated through the separation chamber using peristaltic pumps in an attempt to prevent protein precipitation in the separation chambers. This apparatus was commercialized as the IsoPrimeTM (GE Healthcare) system (11). The IsoPrime has large volumes for each fraction (30–125mL) and requires cooling in addition to sample recirculation, making it cumbersome to set up and use. More recently, a smaller MCE apparatus was developed by Herbert and Righetti (12). The MCE can accept up to seven chambers with 6mL per chamber, and the electrophoretic separation is performed on the IsoelectrIQ2 electrophoresis platform (Proteome Systems), which also doubles as an immobilized pH gradient (IPG) IEF system. Solution IEF separation devices that rely on immobiline-buffered membranes for protein separation are capable of very high-resolution fractionations because membrane partitions can be selectively made at precise pHs, and proteins with pIs differing by as little as 0.01 pH units can be separated (13). However, both the commercially available MCE instruments described have relatively large separation chambers and the instruments are relatively complex. At approximately the same time that the Proteome Systems’ MCE device was developed, our laboratory developed a smaller, simpler solution IEF device and separation method, which we called microscale solution IEF (MicroSol IEF) (14). The small volumes of these separation chambers conserved precious samples and made interfacing of high-resolution sample prefractionation with subsequent narrow pH range 2DE gels relatively seamless as samples did not need to be concentrated prior to 2DE. MicroSol IEF has been successfully used to separate Escherichia coli extracts (14), mouse and human serum proteins (15–17), and many other types of eukaryotic cell and tissue extracts. This device has been commercialized as the ZOOM® IEF Fractionator (Invitrogen Corp). The combination of MicroSol IEF prefractionation, use of high protein loads after fractionation on narrow range IPG gels, and use of multiple protein stains can dramatically increase the detection dynamic range and the total number of proteins detected in complex proteomes (see Note 1). In this chapter, we describe the MicroSol IEF method and its use for comprehensive protein profile analysis of complex proteomes. This prefractionation method is compatible with multiple alternative downstream protein profiling methods, including 1DE, narrow pH range 2DE, 2D differential in-gel electrophoresis
294
Joo and Speicher
(2-D DIGE; see also Chapters “High-Resolution 2DE” and “Two-Dimensional Difference Gel Electrophoresis”), and shotgun protein profiling methods such as multidimensional liquid chromatography (LC) interfaced with MS, i.e., LC/LC-MS/MS methods (see also Chapter “Shotgun Protein Analysis by Liquid Chromatography-Tandem Mass Spectrometry”).
2. Materials The ZOOM® IEF Fractionator is a multichamber solution IEF device that depends on isoelectric partitions to trap proteins at their pI values. The number of fractions used and the pH ranges of these fractions can be readily varied depending on the requirements of the experimental design. The commercial device is capable of separating samples into up to seven fractions, and precast partition membrane disks up to either pH 10.0 or 12.0 can be used depending upon whether very basic proteins are of interest. The materials listed are based on a five fraction separation covering the pH range 3–12. 1. MicroSol IEF apparatus (the ZOOM® IEF Fractionator, Invitrogen Corp.). 2. pH membranes (ZOOM Disks): 3.0, 4.6, 5.4, 6.2, 7.0, and 12.0 (Invitrogen Corp.). ZOOM Disks are stable for 3 months when stored at 4°C. 3. Anode buffer: Novex IEF anode buffer (50×) (Invitrogen Corp.) diluted to 1× with addition of 8M urea and 2M thiourea (final concentrations). Adjust to pH 3.0 with Novex IEF buffer (50×) if necessary. Novex IEF anode buffer (50×) is stored at room temperature and is stable for a year. 4. Cathode buffer: Novex IEF cathode buffer, pH 9–12 (10×) (Invitrogen Corp.), diluted to 1× and with addition of 8M urea and 2M thiourea (final concentrations). Adjust to pH 12.0 with NaOH if necessary. Store Novex IEF Cathode buffer (10×) at 4°C. 5. Sample buffer: 8M urea, 2M thiourea, 4% 3-[(3cholamidopropryl)dimethylammonio]-1-propanesulfonic acid (CHAPS), 1% dithiothreitol (DTT), 1% pH 3–7 and 1% pH 7–12 of ZOOM focusing buffers (Invitrogen Corp.). 6. 10% NuPAGE Bis-Tris 1DE gels (Invitrogen Corp.). 7. Centrifugal filter devices (0.5mL, 0.22μm, Millipore). 8. Ultrapure urea (GE Healthcare). 9. 3M Tris base (Bio-Rad Laboratories).
Prefractionation Using Microscale Solution IEF
295
10. 3M DTT stock solution (GE Healthcare). 11. 10M N, N-Dimethylacrylamide (DMA) stock solution (Sigma-Aldrich Inc.).
3. Methods
3.1. Sample Preparation
Proper sample preparation is a critical factor for successful MicroSol IEF fractionation. The key objectives of sample preparation are completely to solubilize and denature as many proteins as possible in the sample of interest, subsequently reduce all disulfide bonds and alkylate the cysteines to prevent reformation of disulfide bonds during fractionation or at later stages of the analysis. Furthermore, as noted earlier, one can improve quantitative comparisons by differentially labeling samples to be compared with either the Cy dyes designed for DIGE using the manufacturer’s instructions or using stable isotope labeling methods if non-gelbased downstream analysis methods are to be used. The general steps for sample preparation using plasma or serum are described as follows. For human or mouse plasma or serum an optional first step is to further reduce the complexity and dynamic range by removing the most abundant proteins using immunoaffinity columns. A range of alternative immunoaffinity resins are currently available from several commercial sources in either spin-column or HPLC-column format that can deplete 2, 6, 12, or even 20 abundant human proteins, and other columns can deplete 3 or 7 abundant mouse or rat proteins. 1. Denaturation and pH adjustment: Adjust 3mg immunodepleted serum or plasma to 8M urea (add dry high purity reagent) and 20mM Tris, pH 8.5 (use 3M Tris base stock). Ensure that the urea has completely dissolved. Using a 3M stock solution of Tris base ensures that the final volume of the protein solution remains low. The pH of the 3M Tris stock solution is not adjusted since, upon dilution in the serum, the final pH will be reduced to the desired value. 2. Reduction: Adjust the sample to 20mM DTT using a 3M stock solution, mix, blanket with argon or nitrogen gas, and place on a shaking incubator for 30min at room temperature (RT). 3. Alkylation: Add 10M dimethylacrylamide (DMA) to a final concentration of 50mM, mix, blanket with argon or nitrogen gas, and place on a shaking incubator for 30min at room temperature (RT).
296
Joo and Speicher
4. Quenching: Using the 3M stock, adjust the sample to 1% final DTT (64.8mM final), taking into account that it already contains 20mM DTT. Mix well. This not only quenches the alkylation reaction but also ensures that the sample contains a high concentration of DTT to keep any residual unmodified cysteines reduced during subsequent ZOOM-IEF fractionation. 3.2. Assembling the ZOOM® IEF Fractionator
One advantage of the ZOOM® IEF Fractionator (Fig. 1a) is the flexibility in terms of the number of fractions and pH ranges that can be utilized. Fractionation is achieved using a series of tandem sample chambers separated by thin acrylamide membranes (ZOOM Disks) containing covalently attached immobiline buffers of defined pHs (Fig. 1b). The unit is assembled prior to loading samples by following the manufacturer’s instructions. As an
Fig. 1. (a) A ZOOM® IEF Fractionator, which consists of seven Teflon sample chambers separated by up to eight acrylamide/immobiline partition membranes and/or open spacers. The insert in the lower right corner shows (from left to right) the anode side of an unplugged chamber, the cathode side of a plugged chamber showing the indentation for a membrane disk, an acrylamide/immobiline membrane disk, and a spacer that is equal in thickness to the partition membrane but with a hole equal to the bore of the chamber. (b) Schematic illustration of a ZOOM® IEF Fractionator where only five of the seven sample chambers are used for fractionation. Chambers are separated by either membrane disks or spacers (crosshatched vertical bars). The pHs of the membrane disks used to fractionate a sample into 5 pH ranges between 3.0 and 12.0 are shown above the membrane disks. In this scheme, the two blank chambers are coupled to either the anode or cathode electrode reservoir via a spacer. Membranes in the pH ranges shown can be purchased from Invitrogen Corp.
Prefractionation Using Microscale Solution IEF
297
example, the method for the configuration shown in Fig. 1b is described as follows. 1. Assemble the ZOOM® IEF Fractionator starting from the anode (+) end. Insert the anode end sealer into the chamber assembly tube followed by a spacer. 2. Place an assembled sample chamber, which contains a port plug and an O-ring, on the anode end sealer. Push down the sample chamber until the chamber is flush with the top of the tube. 3. Using forceps, place a pH 3.0 ZOOM Disk on the anode end sealer. The disk can be inserted into the chamber with either side facing down. 4. Add the next sample chamber, making sure that the port plug is in place and the O-ring is facing down. Push the chamber into the chamber assembly tube until it is flush with the top of the assembly tube. 5. Repeat steps 3 and 4 until all membrane disks and sample chambers are serially connected. Adjacent chambers should be separated by membrane disks in the following order: pH 3.0, 4.6, 5.4, 6.2, 7.0, and 12.0. As shown in Fig. 1b, the first and last chambers are blank chambers without a disk separating it from the anode and cathode chamber. 6. Insert the cathode end sealer O-ring on the groove facing down, next to the last sample chamber. 7. Attach the cathode end screw cap to the end of the chamber assembly tube. 8. Load ∼17.5mL 1× Novex IEF anode (pH 3.0) and Novex cathode buffer (pH 12.0) into each electrode reservoir of the chamber. 3.3. Separation of Alkylated Sample in ZOOM® IEF Fractionator
Depending upon the number of sample chambers to be used and the initial sample volume, the sample can be further diluted with additional sample buffer to a final volume equal to the total volumes of the chambers that will be loaded with sample (see Note 2). There is substantial flexibility in how protein samples are loaded into the device. Sample can be loaded into any single sample chamber, into several sample chambers, or into all sample chambers (see Note 3). At least 1–3mg of most types of protein extracts can be effectively separated in a single run. For example, if a total of 3mg of depleted serum or plasma is loaded into the three central sample chambers the procedure would be as follows: 1. Add 670μL of ZOOM Buffer to chambers 2 and 6. Chambers 1 and 7 will contain the anode and cathode electrode buffer, respectively.
298
Joo and Speicher
2. Dilute the sample to 670μL times 3 (the number of chambers to be loaded with sample) = 2,010μL, and remove any particulate material by either filtering the sample through a 0.22-μm Ultrafree-MC microfilter unit (Millipore Corp.) or by centrifugation at 16, 000× g for 10min at room temperature. 3. Add 670μL of sample to chambers 3, 4, and 5. If any bubbles get trapped in sample chambers, use a gel-loading tip to break the bubbles. 4. Insert sample chamber port plugs into all sample chambers. 5. Place the lid on the assembled fractionator. 6. Focus the sample by applying an electrical field. MicroSol Fractionation of a complex proteome such as serum or plasma requires a power supply with a capacity of at least 1,000V and that can operate at currents below 1mA. The separation is initially limited by current, and as the conductivity falls the voltage rises until 1,000V is achieved. The separation is then continued until a stable low current is reached, i.e., after about 1.5h at 1,000V the current should have dropped to ∼0.3mA and the rate of decrease will slow noticeably (see Note 4). 7. Collect the fractionated sample after MicroSol IEF Fractionation. To avoid crosscontamination and loss of fractionated samples, remove fractionated samples through the fill ports using a gel-loading pipette tip prior to disassembling the device and use a small volume of sample buffer to rinse each chamber. Optionally extract bound proteins from ZOOM disks (see Note 5). 3.4. Evaluation of MicroSol IEF Separation 3.4.1. Evaluation of Fractionation Efficiency on 10% 1DE Gels and Medium Range 2DE Gels
An initial rapid assessment of protein separation quality and relative recoveries in solution fractions and membrane disk extracts can be made by comparing proportional amounts of fractions and membrane disk extracts on 10% NuPAGE Bis-Tris-1DE gels (Invitrogen Corp). A good separation is characterized by having only moderate amounts of protein trapped in the membrane partitions and presence of some protein bands unique to each fraction. Representative results from separation of mouse serum into 5 fractions and 6 membrane extracts are shown in Fig. 2. Typically the majority of proteins in cell extracts and biological fluids have pHs between 4.6 and 7.0, and hence central fractions covering these pH ranges invariably contain the highest protein content and exhibit the most intense protein staining when proportional amounts of each fraction are compared (Fig. 2). The effectiveness of MicroSol IEF prefractionation can be more accurately evaluated by separating individual fractions on medium pH range 2DE gels to evaluate whether efficient separation over the expected pH range has occurred. The results of
Prefractionation Using Microscale Solution IEF
299
Fig. 2. Rapid initial evaluation of MicroSol IEF fractionation using 1DE gels. Initial analysis of equivalent percentages of each fraction and membrane disk extract from separation of major protein depleted mouse serum was performed on 10% NuPAGE Bis-Tris 1DE gels (right panel). In this experiment, all fraction volumes (including the chamber washes) were 700μL, and 10μL of each sample was loaded onto the gel. The depletion of the three most abundant proteins prior to MicroSol IEF using an Aglient MARS immunoaffinity column is shown in the left panel.
fractionating mouse serum are shown (Fig. 3). The separations of unfractionated serum on a pH 3–11 2DE gel and enlargement of the pH 4–8 regions of a pH 3–11 2DE gel are shown (Fig. 3a). Analysis of individual MicroSol IEF fractions on pH 4–7 or 5–8 2DE gels show that the vast majority of protein staining is within the expected pH boundaries defined by the membrane disks used for the separation (Fig. 3b). Of course some proteins within the expected pH range exhibit poor solubility during focusing and streak into other regions of the gel. In addition, small amounts of abundant proteins that belong in an adjacent fraction can be observed in these gels. However, typically greater than 90% of the total protein for these spots is in the correct fractions. 3.4.2. Narrow pH Range 2DE Gels
A practical strategy to achieve more comprehensive analysis of complex proteomes is to combine MicroSol IEF prefractionation with subsequent separation on narrow pH range 2DE gels (15; see also Chapter “Selection of pH Ranges in 2DE”). MicroSol IEF interfaces seamlessly with subsequent narrow pH range IPG
300
Joo and Speicher
Fig. 3. Comparison of unfractionated and MicroSol IEF fractionated mouse serum separated on 2DE gels. (a) Unfractionated mouse serum (∼40μg) was separated on an 11-cm pH 3–11 NL (left panel) IPG strip followed by SDS-PAGE using a 10% Tris-Tricine SDS gel cast in a Criterion gel cassette (Bio-Rad Laboratories). The right panel shows an enlargement of the pH 4–8 region of the pH 3–11 2DE gel. (b) MicroSol-IEF prefractionated serum fractions (2, 3, and 4) were focused on pH 4–7 or pH 5–8 IPG strips (11 cm), followed by separation on 10% Tris-Tricine SDS gels. The vertical lines show approximate locations of the pH boundaries defined by the membrane disks in the MicroSol IEF fractionation. Gels were stained using Invitrogen SilverQuest stain.
gels because similar sample buffers are used and fractions usually do not need to be concentrated prior to 2DE. To maximize spot separation and resolution of individual fractions, the IPG strips should have pH ranges as narrow as possible and be as long as possible to maximize total IEF separation distance. Ideally, the IPG gels should be slightly wider (typically ± 0.1 pH units) than the pH range of each fraction to prevent loss of proteins near the boundaries of the 2DE gels, while maximizing separation distance as much as possible. 3.4.3. 3D DIGE Analysis
As noted earlier, a particularly powerful method of analyzing MicroSol IEF fractions is to utilize DIGE technology to label samples prior to MicroSol IEF (18). This provides more reliable comparative protein profiling than quantitative comparisons between unlabeled samples separated in parallel MicroSol IEF
Prefractionation Using Microscale Solution IEF
301
fractions. It also greatly reduces the number of replicate MicroSol IEF fractionations and replicate narrow range 2DE gels needed to detect significant biological differences. Hence, this 3D DIGE analysis provides far greater depth of analysis than 2D DIGE alone, and it provides much more reliable quantitative comparisons with greater throughput than use of MicroSol IEF and subsequent narrow pH 2DE gels without differential labeling of samples to be compared.
4. Notes 1. Typically 10- to 50-fold higher protein loads can be applied to narrow pH range 2DE gels when a complex proteome such as serum or a cell lysate is initially prefractionated using MicroSol IEF as compared to optimal loads without prefractionation (15). The ability to load much larger proportional amounts of samples after this sample simplification enables detection of much lower abundance spots. In addition, duplicate gels can be stained with different sensitivity stains i.e. such as Colloidal Coomassie, Sypro Ruby, and silver stains (see also Chapter “Silver Staining of Proteins in 2DE Gels”). This multistain approach increases the dynamic range of detection by allowing quantitative comparisons of the highest abundance proteins using the lower-sensitivity stain. If MicroSol IEF pH ranges are chosen so that the complexity of each resulting narrow pH range 2DE gel is similar, it should be practical to detect 1,500–2,500 protein spots per 2DE gel. 2. The maximum amount of sample that can be effectively separated by MicroSol IEF varies somewhat depending upon sample type, loading position, and the number of sample chambers used. Typically, at least 3mg of complex samples such as serum/ plasma or cell lysates can be effectively fractionated when five or more pH range fractions are used. In some cases, separation may be improved by removing excess lipids or nucleic acids. High sample loads tend to result in high sample conductivity and protein precipitation/aggregation on the partition membrane disk surfaces. There seems to be some benefit to using very slow separations at very low initial voltages in an overnight run when protein loads substantially higher than 3mg are to be separated. In addition to high conductivity, high protein loads may overwhelm the buffering capacity of membrane disks because inevitably the pIs of some proteins will exactly match the pH of any partition membrane disk and will be trapped in the membrane. These high local protein
302
Joo and Speicher
concentrations in the pores may either block the pores causing subsequent deposition of other proteins on the membrane disk surfaces or they may simply overwhelm the buffering capacity of the immobilines in the membrane. Increasing the partition gel pore size, cross-sectional area, or buffer capacity of the membrane disks might further increase protein load capacity, but there are trade-offs with any of these strategies. For most studies a sample load capacity in the 1–3mg range is usually adequate since these amounts match well with optimal loads for subsequent narrow pH range 2DE gels. 3. In general, a sample is loaded into all separation chambers to allow it to be diluted to the largest possible volume. But for some samples this may result in poor focusing of proteins in the most acidic and most basic pH chambers, presumably because some proteins that are unstable at pH extremes may precipitate before they have sufficient time to move out of these terminal chambers. Also, proteins loaded into the most basic and acidic sample chambers that are acidic and basic, respectively, must migrate the longest distance to reach the appropriate chamber at equilibrium. Therefore, it is sometimes advantageous to avoid loading some types of samples into the most acidic and most basic sample chambers. In this case these chambers are filled with sample buffer that does not contain any protein. In addition, in some cases there may be sample-specific benefits to selected sample loading. For example, when serum or plasma samples are fractionated without depleting any major proteins, albumin comprises more than 50% of the total protein and normally severely restricts the amount of serum or plasma that can be applied to a 2DE gel (15). Therefore, a good experimental design is to isolate albumin in a single sample chamber with a final very narrow pH range to enhance detection of low abundant proteins in other fractions, i.e., mouse serum albumin can be sequestered in a single narrow range pool of pH 5.4–6.2, and in this case it may be beneficial to load the entire sample into this chamber so that the majority of the protein does not need to pass through any membrane disks. 4. If the power supply has current and power limit capacities, set maximum current to 1mA, maximum power at 1W, and maximum voltage at 1,000V. The separation should be terminated as soon as the drop in current after reaching 1,000V slows down because excessive focusing appears to be detrimental to the separation in some cases. The total focusing time for 3mg of serum or plasma should be about 3h using the conditions described earlier. Samples containing higher protein loads, high salt concentrations, or higher ampholyte concentrations will usually require much longer separation times, and in some
Prefractionation Using Microscale Solution IEF
303
cases it may not be feasible to reach 1,000V or a minimum current below 1mA. 5. Transfer contents of chambers 2–6 into individual microfuge tubes and measure the volumes. After the MicroSol IEF run, volumes in sample chambers may be reduced slightly due to electroosmosis. However, reduction of any fraction to less than 50% of its original volume indicates that chamber was poorly sealed and the quality of the separation was probably compromised. In this case, all fractions should be discarded and a new MicroSol IEF separation should be carried out. After collecting fractions from each sample chamber, a small volume of sample buffer (∼100μL) is used to rinse the inside walls of the chambers, and these rinses are combined with the corresponding fraction. All fractions are then adjusted to equal volumes (e.g., 750 or 800μL) for convenient further analyses and comparisons. Because proteins with pIs equal to the pHs of partition membrane disks will be retained in the membrane disks, it is generally recommended that the membrane disk be separately extracted with two sequential small volumes (∼100μL) of sample buffer, if sample is to be combined with adjacent solution fraction for analysis on narrow pH range 2DE gels. Alternatively, 1% SDS, 20mM Tris, 1% mercaptoethanol can be used if the membrane extract is to be analyzed only on 1DE gels. These sequential extractions should be conducted at room temperature for 30min each with agitation. References 1. O’Farrell, P.H. (1975) High resolution twodimensional electrophoresis of proteins. J. Biol. Chem. 250, 4007–4021. 2. Bjellqvist, B., Ek, K.J., Righetti, P.G., Gianazza, E., Gorga, A., Westermeier, R., and Postel, W. (1982) Isoelectric focusing in immobilized pH gradients; principle, methodology and some applications. J. Biochem. Biophys. Methods 6, 317–339. 3. Gorg, A., Weiss, W., and Dunn M.J. (2004) Current two-dimensional electrophoresis technology for proteomics. Proteomics 4, 3665–3685. 4. Williams, K.L. (1999) Genomics and proteomes: towards a multidimensional view of biology. Electrophoresis 20, 678–688. 5. Dreger, M. (2003) Subcellular proteomics. Mass Spectrom. Rev. 22, 27–56. 6. Elortza, F., Nuhse, T.S., Foster, L.J., Stensballe, A., Peck, S.C., and Jensen, O.N. (2003) Proteomic analysis of glycosylphosphatidylinositol-anchored membrane proteins. Mol. Cell. Proteomics 2, 1261–1270.
7. Zuo, X., Lee, K., and Speicher, D.W. (2004) Electrophoresis prefractionation for comprehensive analysis of proteome. In Proteome Analysis: Interpreting the Genome. Speicher D.W. (Ed.), Elsevier, New York, pp. 93–117. 8. Bier, M. (1988) Recycling isoelectric focusing and isotachophoresis, Electrophoresis 19, 1057–1063. 9. Righetti, P.G., Castagna, A., Herbert, B., Reymond, F., and Rossier, J.S. (2003) Prefractionation techniques in proteome analysis. Proteomics 3, 1397–1407. 10. Sang, T.Q., Ginter, J.M., Johnston, M.V., Larsen, B.S., and McEwen, C.N. (2003) Carrier ampholyte-free solution isoelectric focusing as a prefractionation method for the proteomic analysis of complex protein mixtures. Electrophoresis 24, 2359–2368. 11. Righetti, P.G., Wenisch, E., and Faupel, M. (1989) Preparative protein purification in a multi-compartment electrolyser with immobiline membranes. J. Chromatogr. 475, 293–309.
304
Joo and Speicher
12. Herbert, B. and Righetti, P.G. (2000) A turning point in proteome analysis: sample prefractionation via multicompartment electrolyzers with isoelectric membranes. Electrophoresis 21, 3639–3648. 13. Righetti, P.G., Wenisch, E., Jungbauer, A., Katinger, H., and Faupel, M. (1990) Preparative purification of human monoclonal antibody isoforms in a multi-compartment electrolyser with immobiline membranes . J. Chromatogr. 500, 681–696. 14. Zuo, X. and Speicher, D.W. (2000) A method for global analysis of complex proteomes using sample prefractionation by solution isoelectrofocusing prior to two-dimensional electrophoresis, Anal. Biochem. 284, 266–278. 15. Zuo, X. and Speicher, D.W. (2002) Comprehensive analysis of complex proteomes using microscale solution isoelectrofocusing prior to narrow pH range two-dimensional electrophoresis. Proteomics 2, 58–68.
16 . Echan , L.A. , Tang , H-Y. , Ali-Khan , N. , Lee, K., and Speicher, D.W. (2005) Depletion of multiple high-abundance proteins improves protein profiling capacities of human serum and plasma. Proteomics 5, 3292–3303. 17 . Tang , H-Y, Ali-Khan , N. , Echan , L.A. , Levenkova, N., Rux, J.J., and Speicher, D.W. (2005) A novel four-dimensional strategy combining protein and peptide separation methods enables detection of low-abundance proteins in human plasma and serum proteomes. Proteomics 5, 3329–3342. 18. Han, M.J. and Speicher, D.W. (2008) Microscale isoelectric focusing in solution; A method for comprehensive and quantitativet proteome analysis using 1-D and 2-D DIGE combined with MicroSol IEF prefractionation. In: Methods in Molecular Biology. Posch A. (Ed.), Humana Press, Totowa, NJ, pp. 241–256.
Chapter 19 Diagonal Electrophoresis for Detection of Protein Disulphide Bridges Brian McDonagh Summary A state of oxidative stress (OS) can occur when there is an imbalance between the rate of reactive oxygen species (ROS) production and their detoxification. Under OS conditions sulphur-containing residues are particularly susceptible to oxidation, and this can result in transient formation of intra- or inter-chain disulphide bridges. Diagonal electrophoresis is a relatively simple technique to analyse the formation of these bridges by sequential non-reducing/reducing electrophoresis. Proteins that do not form disulphides, electrophorese identically in both dimensions and form a diagonal after the second dimension, proteins that contained intra-chain disulphides lie above this diagonal, while those that formed interdisulphides fall below the diagonal. These spots can be excised, tryptic digested, and identified by mass spectrometry. Identification of those proteins, which are reversibly modified, could play an important part in coupling redox status to protein function. Key words: Cysteine, Disulphide, Diagonal electrophoresis, Oxidative stress.
1. Introduction Oxidative changes in the redox potential in the cell can lead to amino acid modifications such as carbonylation, glutathionylation, nitrosylation, and the formation of disulphide bonds. Some of these modifications such as carbonylation are irreversible and lead to inactivation of the protein, while other modifications may play a protective role in preventing the protein from further oxidative damage by the formation of disulphide bonds. As disulphide bond formation is reversible, protein function can be restored after oxidative challenge (see ref 1). This type of protection has been shown to play a role in a wide variety of protein cellular functions. Although two-dimensional electrophoresis David Sheehan and Raymond Tyther (eds.), Methods in Molecular Biology, Two-Dimensional Electrophoresis Protocols, vol. 519 © Humana Press, a part of Springer Science + Business Media, LLC 2009 DOI: 10.1007/978-1-59745-281-6_19
305
306
McDonagh
(2DE) allows the separation of up to thousands of individual proteins into discrete spots (2), the reducing conditions required for high resolution can make it difficult to visualise individual proteins involved in redox proteomics without radiolabelling. A relatively simple method developed for the analysis of both inter- and intra-chain disulphide bonds, is the use of diagonal electrophoresis (3). Although cysteines (Cys) are one of the most rarely used amino acids in proteins, they play a crucial role in their structure and function (4). The high reactivity of the thiol group is often essential in the redox centres of proteins. Under basal, nonstressed conditions, the sulphydryl (–SH) groups often remain protonated in the cytoplasm. However, under oxidative stress, sensitive redox proteins can form disulphide bonds due to a change in their redox environment. Diagonal electrophoresis involves sequential non-reducing/ reducing electrophoresis. Initially, a protein mixture is electrophoresed under non-reducing conditions, in an alkylated buffer to prevent thiol disulphide exchange. The entire lane is then excised from the one-dimensional electrophoresis gel, treated with a reducing agent, placed orthogonally on top of a second gel and electrophoresed as normal under reducing conditions (Fig. 1). After staining, proteins that do not contain disulphide bonds congregate along a diagonal since they electrophorese NON-REDUCING DIRECTION
S S
S S
-S-S-
REDUCING DIRECTION
Fig. 1. Principle of diagonal gel electrophoresis. Proteins are first electrophoresed in non-reducing conditions. Proteins with intra-chain disulphides run slightly faster due to their greater compactness. Proteins with inter-chain disulphides run slowly because of their large size due to disulphide bonds linking several polypeptide chains.
Diagonal Electrophoresis for Detection of Protein Disulphide Bridges
307
NON-REDUCING DIRECTION
Mr (x10-3)
97 66 45 29
REDUCING DIRECTION
205
20
Fig. 2. Example of diagonal electrophoresis. Silver staining reveals several proteins in a crude extract with inter-chain (boxes) and intra-chain (circles) disulphide bonds. Note that the majority of proteins electrophorese identically under both reducing and nonreducing conditions and hence form a crowded diagonal.
identically in both reducing and non-reducing conditions (Fig. 2). Proteins involved in inter-chain disulphide bonds will fall below the diagonal as they have a lower apparent molecular mass in reducing conditions. Proteins that contain intra-chain disulphide bonds will run above the diagonal as they have a larger apparent molecular mass. Protein spots can then be excised, tryptic digested, and identified by mass spectrometry. This technique has led to the identification of proteins previously not known to form disulphide bonds (5–6). It has also been used to identify proteins by immunoblotting and analysis of carbonylation (7).
2. Materials 1. Alkylating buffer solution. 50 mM Tris–HCl, pH 7.5, 2 mM EDTA, 1 mM PMSF, and 40 mM iodoacetamide (IAM). 2. Sample buffer (4×) (8). 1.25 mL 0.5 M Tris–HCl, pH 6.8, 3.0 mL glycerol, 1 mL 20% SDS and bromophenol blue. 3. 20% trichloroacetic acid (TCA). 4. Resolving acrylamide gel solution (8). 5. Stacking acrylamide gel solution (8). 6. Running buffer. 1L containing 14.4 g glycine, 3.03 g Tris base, and 1g SDS.
308
McDonagh
7. Equilibration buffer. (6 M urea, 0.375 M Tris–HCl, pH 8.8, 2% SDS, 20% glycerol) with either 2% dithiothreitiol (DTT) or 2.5% IAM. 8. Agarose solution. 0.1% agarose in running buffer plus bromophenol blue. 9. Two sets of slab gel apparatus, either ATTO and BioRad mini Protean system or alternatively slab gel apparatus with two sets of spacers of different thickness (see Note 1). 10. Power supply. 11. Rocking platform.
3. Methods 3.1. Sample Preparation
3.2. Electrophoresis
1. Protein extracts are prepared using an alkylating buffer solution. 2. Proteins are precipitated in 20% TCA and resuspended in non-reducing sample buffer (4X×) to give a final protein concentration of 50μg/15μL (see Note 2). 1. Prepare a 12% polyacrylamide resolving gel with 4% stacking gel (8) (see Note 3) 2. Add 15μL non-reducing sample buffer containing 50μg of protein to a well. Use alternate wells in the non-reducing first dimension so that it will be easier to excise the lanes. 3. One-dimensional electrophoresis (1DE) is carried out at 60 V until dye reaches the end of the gel (see Note 4). The entire lane is excised using a sharp fine scalpel (see Notes 5 and 6). 4. The entire lane is soaked for 20 min in equilibration buffer containing 2% DTT at room temperature on a rocker, followed by 20 min in equilibration buffer containing 2.5% IAM. 5. The gel lane is rinsed in running buffer and placed horizontally on a 12% resolving gel (see Note 3). Carefully slot excised lane onto slab gel using the blunt end of small spatula. It helps if the excised lane is kept moist using running buffer (see Note 7). 6. A protein standard mixture can be electrophoresed alongside diagonal gel by pipetting 5μL of mixture onto a small cube of blotting paper (see Note 8). 7. Layer warm agarose containing bromophenol blue on top of the slab gel. Ensure there are no air bubbles between gel lane and slap gel as this will interfere with protein separation. 8. Carry out electrophoresis as per normal until dye front has reached the bottom of the gel. Proteins may be detected by any of the staining techniques applicable to 2DE. Proteins
Diagonal Electrophoresis for Detection of Protein Disulphide Bridges
309
may also be electroblotted for immunoblot analysis. This technique may also be applied to radiolabelled proteins.
4. Notes 1. In the second dimension it is important to use a gel with greater thickness than first dimension. In this case the ATTO gel rig used in second dimension (1 mm thick) is comparable to the BioRad mini Protean rig in first dimension (0.75 mm thick). Alternatively in larger gels, spacers with different thicknesses can be used. Also large gels can give option of running two samples from small first-dimension rigs, alongside each other (5). 2. If using 50μg/well in the first dimension it is better to precipitate 200μg of protein and resuspend in 60μL of sample buffer overnight. This ensures complete and homogenous resuspension and also allows at least three replicates of the sample to be used. 50μg of protein is a guideline as the precise amount will depend on the abundance of proteins of interest and size of the gel rigs. Samples can be resuspended in any sample buffer so long as there are no reducing agents present. 3. Percentages of acrylamide gels can be varied to optimise resolution of proteins of interest. 4. Electrophoresis in the first dimension is carried out slowly as if voltage or current are too high; separation of proteins in the non-reducing dimension will result in streaking of the final gel. 5. The lane width should not be exceeded as it will result in poor resolution in second dimension. Also lane edges should be kept as smooth as possible by using sharp scalpel. 6. The gel lane should not be allowed to dry during excision as it will become very brittle; use running buffer to keep moist during excision. 7. Excised lanes can be stored at −70°C at this stage. 8. This allows easier detection of standards; however, it is not as accurate as for determining the molecular weight of proteins that have been electrophoresed in both dimensions.
Acknowledgment I acknowledge an Embark fellowship from the Irish Research Council for Science, Engineering and technology.
310
McDonagh
References 1. Leichert, L. and Jakob, U. (2006). Global methods to monitor the thiol-disulfide state of proteins in vivo. Antioxid. Redox Signal. 8, 763–772. 2. O’ Farrell, P. H. (1975). High resolution two-dimensional electrophoresis of proteins. J. Biol. Chem. 250, 4007–4021. 3. Sommer, A. and Traut, R. R. (1974). Diagonal polyacyrlamide-dodecyl sulphate gel electrophoresis for the identification of ribosomal proteins crosslinked with emtyl4-mercapto-butyrimidate. Proc. Natl. Acad. Sci. U.S.A. 71, 3946–3950. 4. Pe’er, I., Felder, C. E., Man, O., Silman, I., Sussman, J. L., and Beckmann, J. S. (2004). Proteomic signatures: amino acid and oligopeptide compositions differentiate among phyla. Proteins 54, 20–40.
5. Brennan, J. P., Wait, R., Begum, S., Bell, J. R., Dunn, M. J., and Eaton, P. (2004). Detection and mapping of widespread internal protein disulfide formation during oxidative stress using proteomics with diagonal electrophoresis. J. Biol. Chem. 279, 41352–42360. 6. Cumming R. C., Andon N. L., Haynes P. A., Park M., Fischer W. H., and Schubert D. (2004). Protein disulfide bond formation in the cytoplasm during oxidative stress . J. Biol. Chem. 279, 21749–21758. 7. McDonagh, B. and Sheehan, D. (2007). Effect of oxidative stress on protein thiols in the blue mussel Mytilus edulis: Proteomic identification of target proteins. Proteomics, 7, 3395–3403. 8. Laemmli, U. K. (1970). Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature, 227,680–684.
Chapter 20 High-Resolution Large-Gel 2DE Claus Zabel and Joachim Klose Summary Our two-dimensional gel electrophoresis (2DE) protocol has been continuously improved in our laboratory since its inception 30 years ago. An updated version is presented here. This protocol is a result of our experience in proteome analysis of tissue extracts, cultured cells (mammalian, yeast, and bacteria), cellular organelles, and subcellular fractions. Many modifications and suggestions emerging in our lab as well as in the literature were tested and integrated into our improved protocol if helpful. Importantly we use (a) large (46 × 30 cm) gels to achieve a high resolution and (b) ready-made gel solutions produced in large batches and stored frozen, a prerequisite, among others, for our very high reproducibility. Employing the 2DE method described here we demonstrated that protein patterns separating more than 10,000 protein spots can be obtained from mouse tissue. This is the highest resolution reported in the literature for 2DE of complex protein mixtures so far. Our 2DE patterns are of high quality with regard to spot shape and intensity as well as background. The reproducibility of the protein patterns is shown to be extremely satisfactory. New staining methods such as differential in gel electrophoresis (DIGE) and the latest 2DE gel evaluation software are compatible to our 2DE protocol. Using suitable staining protocols proteins can easily be identified by mass spectrometry. Key words: Two-dimensional gel electrophoresis, High resolution, High reproducibility, DIGE, Proteomics, Mass spectrometry.
1. Introduction Two-dimensional electrophoresis (2DE) of proteins is used for several purposes, such as resolving a distinct group of proteins (e.g., serum proteins), revealing heterogeneity of a particular protein (e.g., γ-actin, transferrin), purifying a protein or testing the purity of a protein obtained by other methods (see also Chapters “Two-Dimensional Electrophoresis (2DE): An Overview,” “Difficult Proteins,” and “Organelle Proteomics”). David Sheehan and Raymond Tyther (eds.), Methods in Molecular Biology, Two-Dimensional Electrophoresis Protocols, vol. 519 © Humana Press, a part of Springer Science + Business Media, LLC 2009 DOI: 10.1007/978-1-59745-281-6_20
311
312
Zabel and Klose
However, the most exceptional feature of this method is its potential to resolve a large number of different proteins if not all of a certain cell type or tissue. It was this particular feature of 2DE developed in 1975 (1–3) that facilitated the study of many of the biological problems of current interest from a new angle: Unanswered questions concerning gene expression, gene regulation, genetic variation, cell differentiation, embryonic development, stem cells, pathogenesis of certain diseases, and other areas could be studied based on a broad spectrum and representative number of proteins. Previously, studies like these were performed on selected proteins only, frequently chosen because of easy accessibility, and considered as model proteins (e.g., hemoglobin). Moreover, when simple procedures for extracting cell or tissue proteins are utilized, i.e., procedures which avoid protein precipitation, lyophilization, dialysis, or chromatographic fractionation, 2DE protein patterns reveal individual proteins in quantities which reflect, at least to some extent, the relative concentration of these proteins in cells or tissues ( see Chapters “Solubilization of Proteins in 2DE: An Outline,” “Difficult Proteins,” “Organelle Proteomics,” and “Preparation and Analysis of Plant and Plastid Proteomes by 2DE”). Quantitative data on gene expression are an important parameter for all studies in cell biology (4, 5). Furthermore, once proteins are separated by 2DE, specific protein analytical techniques such as blotting, partial sequencing, and peptide analysis by MS allow identification and characterization of individual proteins. After completion of sequencing the entire human genome (and the genomes of other organisms) (6–8), 2DE became an indispensable tool for the elucidation of gene function. Like in the genome era this cannot be achieved singlehandedly by one lab alone. Therefore, in order to maintain comparability of experiments highly standardized protocols for 2DE related data need to be implemented so that results can be used efficiently by the research community (9). When studying the function of all genes it is important to answer the question: Is it possible to reveal the entire protein species of a single cell type by 2DE? If 2DE is actually able to reveal each individual protein of a particular cell type, this method would reach a new level of significance in studying biological and pathological processes. Because there is probably no process in a living cell that does not alter the occurrence or concentration of at least one protein, it would be worthwhile to study 2DE protein patterns relevant to the problem under investigation in any study of basic and applied biomedical research. Unfortunately, since there is no guarantee that rare proteins such as transcription factors can be detected in these patterns, the information expected cannot be considered complete.
High-Resolution Large-Gel 2DE
313
The present estimation of the total number of genes (defined as a region of DNA sequences that is transcribed into poly-adenylated mRNA that is translated into protein) in the human genome is about 25,000 (8). Based on RNA reassociation kinetics, a typical mammalian cell may express 10,000 distinct genes (10). We estimated from 2DE protein patterns (soluble fraction of mouse brain proteins) which was generated by the large-gel technique described here that, on average, each individual protein is represented by three spots in the 2DE pattern (11). This estimation was obtained by a genetic approach: Protein spots which showed a shift in position by exactly the same distance (e.g., 2.5 mm) and direction in the 2DE pattern and, moreover, mapped to the same locus on the mouse chromosomes, when one mouse strain was genetically compared to another, were considered modifications of the same protein. Consequently, a 2DE pattern that reveals the entire set of proteins of a typical mammalian cell should consist of about 30,000 (i.e., 10,000 × 3) protein spots. This is consistent with the notion that, theoretically, a 2DE gel with dimensions of 400 × 300 mm would provide sufficient space for ∼30,000 spots if the mean spot size is 2 × 2 mm. The probability that two protein species of a cell are completely identical with regard to pI, molecular weight, and three-dimensional shape and, therefore, occupy exactly the same location on the 2DE gel is rather low (12). According to these considerations a large 2DE gel should be able to reveal the entire set of proteins of a single cell type. Still, several problems occur when considering the scenario put forward here. Cellular proteins do not distribute evenly over the entire gel area, and streaking spots occupy an area which is significantly larger than may be expected from their properties. Furthermore, a selective loss of proteins may occur in different steps after extraction from tissue to the final protein sample and subsequently during migration into the first- and second-dimension gel of 2DE (Chapters “Difficult Proteins” and “Protein Extraction for 2DE”). Most critically, however, certain protein classes may exist only in very low concentrations in a tissue, i.e., only a few protein molecules occur per cell and possibly not even in each cell of that tissue (13). Proteins involved in regulatory processes and characterized by high turnover rates may represent such classes of proteins. If rare proteins are defined as proteins present in a tissue in concentrations of ∼1 molecule per cell or per group of cells we believe, according to an experimental approach described elsewhere (14), that these proteins cannot be detected by presently available 2DE techniques. However, when defining the concentration of rare proteins as 3–30 molecules/cell, Duncan and McConkey (15) concluded from their investigations that rare proteins can indeed be detected by 2DE. Fractionation of tissue
314
Zabel and Klose
proteins followed by concentrating the extracts and running each fraction separately may be a way to overcome both the problem of restricted space on a 2DE gel and of detecting low-abundance proteins which are enriched by this procedure. Furthermore, using CyDye saturation labeling helps to detect low-abundance proteins (16, 17). Selective loss of proteins during sample preparation and during incubation of the isoelectric focusing (IEF) gel can be avoided by following our sample preparation procedure described in Chapter “Protein Extraction for 2DE.” However, hydrophobic proteins still remain problematic since they are not readily resolved in 2DE (Chapter “Difficult Proteins”). Nevertheless, in using 2DE our goal is to resolve and detect all the individual proteins of a distinct tissue. Therefore, a special procedure for extracting total tissue proteins was developed (Chapter “Protein Extraction for 2DE”), and our original 2DE technique (1, 18) was substantially altered to support the separation of such a large number of proteins (19). The main feature of the 2DE technique described here is the use of large gels with a resolving power of more than 10,000 polypeptide spots per pattern (see Note 1). A protein sample is located on the acid side of the IEF gel to avoid loss of very basic proteins. On the acidic side the complete loss of specific protein species is very unlikely since very acidic proteins (pI below 4) are very rare. In comparison, immobilized pH gradients used in long IEF gels were found to lose resolution, particularly in the basic range (19). Moreover, when comparing 2DE patterns of the same protein sample in the two different systems, more protein remains at the start position in immobilized pH gradient (IPG) gels and, after SDS-PAGE, in the IEF gel than in carrier ampholyte gels (our own observations; ref 20). Therefore, since it is our goal to obtain high resolution we use carrier ampholytes and large 2DE gels. In the following sections, we describe the large-gel 2DE technique developed in our laboratory (19) in all detail, including the preparation of 40-cm IEF gels in capillary tubes, cutting the long gels into two halves after an IEF run, transfer of the two half gels onto two 20-cm width SDS gels with a vertical separation distance of 30 cm, silver staining of the separated proteins, and drying the 2DE gels. The slightly different conditions used when IEF is performed in 20-cm gels are also described briefly. This reduction in IEF size may be employed for protein samples of lower complexity. Recipes are provided for preparing gel solutions in ready-to-use quantities to freeze and use later. Freezing large numbers of ready-made solutions from the same batch of gel solution contributes considerably to reproducibility of 2DE patterns (see Note 2)
High-Resolution Large-Gel 2DE
315
2. Materials 2.1. Equipment 2.1.1. Isoelectric Focusing
1. Glass tubes for IEF gels: The dimensions of glass tubes used for 40-cm gels and 20-cm gels are indicated in Fig. 1 (see Note 3). To fill tubes with gel solution as well as push the gels out of the tubes after IEF, nylon strings (fishing line) are used. The strings are 0.8 mm in diameter and one end is increased in diameter (0.9 mm; Fig. 1h) by melting the end of the string and fitting it quickly into the glass tube. (Glass tubes Schott, Mainz, Germany; ready-made glass tubes are available from Wita GmbH, Teltow, Germany). 2. Gel tube stand: This custom-made apparatus consisted of a tube holder and an adjustable platform (Fig. 2). A special gel solution tray with a groove wide enough to accommodate the glass tubes is placed on the adjustable platform to bring the gel solution in contact with the lower ends of the tubes (Fig. 2). 3. Apparatus for IEF: The apparatus consisted initially of an upper and a lower cylindrical buffer chamber, but we added an intermediate cylinder, available in various heights, to allow the insertion of gel tubes of various lengths (44.5 cm or 23.5 cm; available from Wita GmbH). Up to eight glass tubes may be inserted into the apparatus and fixed with screws. 4. Power supply: The power supply for IEF should be programmable and allow a user to select a voltage of at least 2,000 V (e.g., Electrophoresis Power Supply – EPS 3500 XL; GEHealthcare, Freiburg, Germany). 5. To store the gels after IEF and to facilitate the transfer of the IEF gels to the second dimension (SDS-PAGE gels), special rails with gel grooves were made from plexiglass (see Fig. 3; available from Wita GmbH) and a box with mountings to lock the grooves in an orderly way. A wire shaped as shown in Fig. 3 is used to manipulate the thin and long IEF gels.
2.1.2. SDS-PAGE
1. Polymerization stand: The polymerization stand Desaphor VA (Desaga, Heidelberg, Germany) which holds two gel cells is used. At the middle of the stand bottom a hole is drilled penetrating the plastic bottom and the silicone gasket. Afterward a cannula was inserted into this hole in such a way that the opening is located exactly between the two glass plates of the gel cell, with its other end projecting slightly from the bottom. A fine tube is attached to the free end of the cannula. In order to apply glycerol a 2-mL syringe is attached to the end of the rubber tube.
316
Zabel and Klose
Fig. 1. Schematic representation of hollow glass tubes for IEF gels and strings to draw gel solutions into tubes. Left side: Filling string is in start position. Center: The gel solution is in the tube and the filling string (a) is in polymerization position. (The lower end of the string is slightly thickened; h). The tubes were filled by dipping the end of the tubes into the separation gel solution (see Fig. 2) and lifting the string, first to marking 1 to draw in the separation gel solution (b), then to marking 2 to draw in the cap gel solution, and (c) finally to marking 3 to obtain, at the end of the tube, a small cubicle free of gel solution. At the beginning of this filling procedure wet the end of string and tube with gel solution by moving the end of the string several times out of and into the tube. After the separation gel solution is drawn in, gel solution is removed from the groove (Fig. 2) and cleaned. Subsequently cap gel solution is filled in the groove and drawn in. The residual solution at the outside of the tube ends is wiped off by filter paper after the first and second steps. The markings on the glass tubes for 40-cm IEF gels have the following distances: tube-end to marking 1:40.2 cm; marking 1 to 2: 0.8 cm; marking 2 to 3: 0.5 cm; marking 3 to tube end: 2.0 cm (total length: 43.5 cm). Right side: The upper end of a gel tube is shown indicating the various layers of solution after sample application: phosphoric acid (d); protection solution, 5 mm (e); sample solution, 5–10 μL, maximum 16 μL (f); Sephadex suspension, 2 mm (g). Sample solution in 20-cm gel tubes: 5 μL, maximum 10 μL. Phosphoric acid solution (upper running buffer) was omitted in cases of maximum sample load.
Fig. 3. Scheme of gel groove. (a) Total view. (b) Cross section of the groove holding an IEF gel. The groove is adapted to the gel cell, demonstrating how an IEF gel is brought into close proximity and can subsequently be transferred onto an SDS-PAGE gel. A wire suitably shaped (c) is used to slide the IEF gel between glass plates. (d) To store IEF gels in a freezer, gel grooves carrying the gels are attached to a metal clamp on a plate and protected by a plastic box.
High-Resolution Large-Gel 2DE
317
Fig. 2. Schematic representation of a gel tube stand. Tube holder (a) the upper buffer chamber of a Desaphor VC 260 apparatus (Desaga) was adapted to our gel tube stand. Only eight of the 24 holes of the real chamber are indicated in the scheme. Gel solution groove (b) the inner dimensions of the groove are 13 × 0.9 cm, deepest point of the curved bottom: 0.6 cm. Adjustable platform (c) used to move the gel solution containing groove up and down during gel preparation. An improved version of this apparatus is available (Wita GmbH).
318
Zabel and Klose
2. To pour the gel solution into the glass cell, use a special funnel (Fig. 4). 3. Apparatus for PAGE: A Desaphor VA 300 (DESAGA) electrophoresis chamber is used. Two gel cells can be inserted into this apparatus. The dimensions of the gel cells are indicated in Fig. 4. For analytical 2DE runs 0.75-mm gage gauge plastic are used to ensure the desired distance between the two glass plates of the cell. 4. Power supply: The power supply for PAGE should be able to sustain an electric current of 100 mA and a voltage of 1,000 V (e.g., see Subheading 2.1.1).
Fig. 4. Schematic view of Desaphor VA 300 electrophoresis apparatus gel cell (without clamps). The horizontal line below the upper edge of the glass plate indicates the surface of the separation gel. The horizontal line above the lower edge of the glass plate was scratched into the glass as a mark which indicates when to terminate an electrophoretic run. The hatched stripes on each side of the gel cell indicate the plastic spacers between the two glass plates. A special plastic funnel to pour the gel solution between the glass plates is shown in frontal view (a) and in a cross section (b). The lower part of this figure shows how an IEF gel is cut in half to transfer the 40-cm tube gel to two regular-sized SDS-PAGE gels.
High-Resolution Large-Gel 2DE 2.1.3. Gel Drying
319
Incubator: For drying up to eight 20 × 30 cm gels, the incubator type UL-60 from Memmert (alternatively, Model INP 800, Memmert, Schwabach, Germany) is used with some modifications: holes are drilled through the two side walls of the incubator on four different levels (Fig. 5), a modification carried out by Memmert on request. Drying plates serving as supports for the gels are manufactured from aluminum in a workshop (Fig. 5). All eight plates are connected by a tubing system that passes through a manometer, one on the left and one on the right side, and end in a water-saving vacuum pump, type JET-1/A4 (Genser Scientific Instruments, Rothenburg, Germany; Fig. 5). For the
Fig. 5. Scheme of the gel-drying system. The incubator (only the right half is shown) with inner dimensions of 100 × 80 (height) × 45 cm (depth) has four levels, each containing two drying plates. One plate and part of the second are shown on the top-most level. The four plates of one side are connected by a tubing system that ends in a water-saving vacuum pump. A manometer combined with a water trap is integrated in the tube system. Circuit breakers on each level of the tube system allow for generation of a vacuum for each plate individually. The vacuum pump is connected with the tube system of both the left (not shown) and the right side. A drying plate is shown in top and front view. A system of rills in the bottom of the plate drains the water away by vacuum from the wet sheets layered on the plates. The different layers are shown: (1) drying plate; (2) metal plate perforated by fine holes, 31.0 × 37.8 cm, 0.7 cm width; (3) ten layers of filter paper; (4) water-permeable clear membrane; (5) SDS-PAGE gel; (6) water-permeable clear membrane; (7) porous polyethylene sheet; (8) silicon sheet, 44 × 37 cm, 1 mm width; (9) metal plate, 45 × 36 cm, 3 mm thick. When the vacuum pump is turned on, the silicon sheet is pressed by the metal plate against the drying plate until a vacuum is generated in the area between silicon sheet and drying plate.
320
Zabel and Klose
drying procedure, various paper sheets and the gel are layered onto the drying plate (Fig. 5). The drying plate is a V4A steel plate perforated with 0.93-mm holes equally spaced at a distance of 1.9 mm, custom-made by a metal working company. Filter paper type 0859 (filtration time: 150 s, Schleicher and Schuell, Keene, USA), a water-permeable lucent membrane with porous cellophane backing (GE Healthcare), porous polyethylene sheets (GE Healthcare), a silicon sheet, and an aluminum plate with a coin-sized hole in each corner (to allow penetration of air) and equipped with a handhold are employed. 2.2. Buffers and Solutions (See Note 4) 2.2.1. Isoelectric Focusing
1. Separation gel solution: 3.5% Acrylamide, 0.3% bisacrylamide, 4% carrier ampholyte mixture, 9 M urea, 5% glycerol, 0.6% TEMED (Table 1). Prepare approximately 49 mL. The gel solution is filtered (GF/C glass microfiber filter, diameter: 7 cm; Whatman Scientific Ltd., Maidstone, Kent, UK), aliquoted to 0.975 mL, and stored at −70°C. 2. Cap gel solution: The composition is the same as the separation gel solution with exception of the acrylamide stock solution, which is replaced by the following solution: 12.0g acrylamide and 0.13 g piperazine diacrylamide are added to bidistilled water to yield a final volume of 20 mL. Only half the volume
Table 1 Gel solution for IEF gels Components
Stock solutionsa
Mixture
Final concentrations
Acrylamide
3.5 g
Piperazine diacrylamide
0.3 g in 20 mLb
10.00 mL
Carrier ampholyte mixture
8.0 mLc
5.00 mL
4.00%
Urea
–
27.00 g
54.00% (9.0 M)
Glycerol
14.3 g/50 mL
8.75 mL
5.00%
TEMED
60.0 μL/10 mL
5.00 mL
0.06%
Gel solution
48.75 mL
a
3.50% 0.30%
Chemicals were dissolved in bidistilled water; final volumes are indicated Amberlite treatment of acrylamide solution: 20 mL acrylamide solution was mixed with 1 g Amberlite MB-1A and stirred for 1 h. Immediately thereafter the solution was filtered to remove Amberlite (Amberlite may alter the pH of the solution). Amberlite was washed several times with bidistilled water on a filter before use c The carrier ampholytes mixture was a combination of 8 mL Ampholine, pH 3.5–10 (Pharmacia); 8 mL Servalyt, pH 2–11 (Serva); 24 mL Pharmalyte, pH 4–6.5 (Pharmacia); 16 mL Pharmalyte, pH 5–8; and 8 mL Servalyte, pH 6.0–9. The resulting 64 mL ampholyte mixture was aliquoted into 8-mL units, frozen in liquid nitrogen, and stored at −70°C. One 8 mL unit was used for separation and the cap gel solution. Any remaining solution was aliquoted into 60 μL portions, frozen, and used for the Sephadex suspension (see Subheading 2.3.1). The original ampholyte solutions were processed immediately after opening b
High-Resolution Large-Gel 2DE
321
of the mixture indicated in Table 1 is prepared and then filtered, aliquoted into 0.390-mL portions, and stored at −70°C. 3. Ammonium persulfate solution: 0.12g ammonium persulfate is dissolved in bidistilled water to yield a final volume of 10 mL. The solution is filtered, aliquoted into 100 μL portions, and stored at −70°C. 4. Sephadex suspension: Swell 20 g Sephadex G-200 superfine in 500 mL bidistilled water for 5 h at 90°C. Decant the supernatant water and resuspend Sephadex in 1 L 25% glycerol solution and stir for 2 h using a propeller. The glycerol solution is changed once during this time. The supernatant water is then removed from the suspension by using a filter funnel attached to a vacuum source. Subsequently, Sephadex slurry is aliquoted to 8mL units and stored at −20°C. Prior to use, each 8-mL unit is further aliquoted to 0.272 g units and stored again at − 20°C. This dual step procedure is employed to economically utilize the time consuming first step. 5. Sample protection solution: Dissolve 6 g urea and 1 g glycerol in bidistilled water to a final volume of 19 mL. From this solution 7.6 mL is mixed with 0.4 mL Servalyt, pH 2–4 (Serva). The resulting solution is filtered, aliquoted into 50 μL portions, and stored at −70°C. The original Servalyt solution is aliquoted into 0.5-mL portions and frozen at –70°C immediately after first usage. 6. Electrode solutions: The composition is indicated in Table 2. The electrode solutions are prepared fresh before each IEF run and degassed for 5 min. 7. Equilibration solution: The composition is indicated in Table 3. Two liters of the final solution is filtered, aliquoted into 20-mL portions, and stored at −70°C.
Table 2 Electrode solutions for IEF Components
Mixture
Final concentration
Phosphoric acid
25 mL
7.27%
Urea
90 g
18.00% (3 M)
Anode Solution
in 500 mLa
Ethylenediamine
20 mL
5.00%
Urea
216 g
54.00% (9 M)
Glycerol
20 g
5.00%
Cathode solution
in 400 mLa
a
Distilled water; final volume is indicated
322
Zabel and Klose
Table 3 Equilibration solution Components
Mixture
Final concentration
Tris buffer
125.00 mLa
125 mM
Glycerol
400.00 mL
40%
DTT
10.03 g
65 mM
SDS
30.00 g
3%
Equilibration solution
In 1,000 mLb
a Tris-phosphate buffer: 30.275 g Tris in bidistilled water titrated with phosphoric acid; pH 6.8 at 20°C; final volume 250 mL b Bidistilled water; final volume is indicated
2.2.2. Sodium Dodecyl Sulfate Polyacrylamide Gel Electrophoresis (SDS-PAGE)
1. Gel solution: The composition is indicated in Table 4. The acrylamide solution is mixed with buffer solution and degassed for 30 min in 1.5 L batches. SDS was added and dissolved by briefly and slowly stirring the solution, thus avoiding a new intake of air. Two 1.5 L batches are filtered and combined in this way in a large beaker (change the filter once). The gel solution is then aliquoted into 67.5 mL units, fast frozen at −70°C, and stored at −20°C. 2. Ammonium persulfate solution: 3.2 g ammonium persulfate is dissolved in bidistilled water at a final volume of 250 mL. The solution is subsequently filtered, degassed, aliquoted into 10-mL portions, and stored at −20°C. 3. Sublayering solution: 60 g glycerol is dissolved in distilled water at a final volume of 100mL and stored at 4°C. Sublayering solution: 17.251 g Tris-Base and 7.099 g Tris–HCl (Tris-buffer as indicated in Table 4 and 0.5 g SDS is dissolved in distilled water at a final volume of 500 mL). The solution is aliquoted into 20-mL portions and stored at −20°C. 4. Upper and lower electrode buffer: The composition is indicated in Table 5. The buffer is prepared fresh for each 2DE run. 5. Agarose buffer: Upper running buffer is aliquoted into 9.95mL units and stored at −20°C. 6. Bromophenol blue solution: 50mg bromophenol blue is dissolved in 150 mL distilled water, filtered, aliquoted into 2-mL portions, and stored at −20°C. 7. Protein standards: 2-DE standards (Bio-rad, Munich, Germany); Rainbow Markers (GE Healthcare); “Electran,” M.W. range:
High-Resolution Large-Gel 2DE
323
Table 4 Gel solution for SDS-PAGE Component
Stock solutiona
Acrylamide
498.000 g
Bis
6.640 g In 1,660 mLb
Mixture
15.000% 0.200% 1,600.0 mL 0.375 Mc
Tris
110.408 g
Tris–HCl
45.436 g
TEMED
0.960 g In 1,400 mLd
1,400.0 mL
–
3.2 g
SDS
Final concentration
0.030% 0.100%e
3,000.0 mLf
Gel solution a
Chemicals were dissolved in bidistilled water; final volumes are indicated The acrylamide solution is treated with Amberlite as indicated in Table 1, footnote b. To reduce processing time, the solution is exposed to Amberlite already after the 1 h treatment, the main portion of the solution is decanted from the Amberlite at the bottom and each part is filtered separately. In addition, if as much as 1,660 mL has to be treated with Amberlite we recommend dividing this volume in half c Molarity of Tris/Tris–HCl according to Laemmli (21) d pH of the buffer: 8.88–8.89 at 20°C (9.01–9.02 at 15°C); pH of the final gel solution: 8.80 at 20°C (8.96 at 15°C); pH provided by Laemmli (15) for the gel solution: 8.8. The pH of the gel solution is critical for the migration behavior of proteins: The lower the pH, the closer to the buffer front (or even within the buffer front) the proteins migrate (21) e According to Laemmli (21) f An increase in volume by adding 3.2 g SDS was ignored in this calculation b
Table 5 Electrode buffer for SDS-PAGE Component
Mixture
Final concentrationa
Tris
33.3 g
0.025 M
Glycine
158.4 g
0.192 M
SDS
11.0 g
0.100 M
Electrode buffer
In 11 Lb
a
According to Laemmli (21) Distilled water, final volume is indicated; pH of electrode buffer: 8.55 at 20°C (8.69 at 15°C); pH given by Laemmli (21): 8.3 b
324
Zabel and Klose
2,512–16,949, BDH (VWR, Darmstadt, Germany); Gibco BRL 10 kDa Protein Ladder (Life Technologies, Eggenstein, Germany). 2.2.3. Silver Staining
Six different solutions are used for the quantitative silver staining procedure with compositions listed in Table 6. The solutions are prepared fresh for each set of gels to be stained. In particular, the silver nitrate solution should only be prepared shortly before use. For protein identification by MS, a modified silver staining protocol is used (24). In this silver staining procedure proteins are not crosslinked and can be readily identified by standard mass spectrometers.
2.2.4. Difference Gel Electrophoresis (DIGE)
Apart from the classical silver staining procedure the 2DE technique described here is also compatible with difference gel electrophoresis (DIGE; see also Chapters “High-Resolution 2DE” and “Two-Dimensional Difference Gel Electrophoresis”).
Table 6 Solutions for silver staininga
Components of solutions
Amount of each component in 1 Lb of each solution
Final concentrations
Fixation solution (1)
Ethanol Acetic acid
500.000 mL 100.000 mL
50.00% 10.00%
Incubation solution (2)
Ethanol Glutardialdehyde Sodium thiosulfate Sodium acetate
300.000 mL 20.000 mL 2.000 g 41.015 g
30.00% 0.50% 0.20% 0.50 M
Silver nitrate solution (3)
Silver nitrate Formaldehyde
1.000 g 0.288 mL
0.10% 0.01%
Wash solution (4)
Sodium carbonate
25.000 g
2.50%
Developer solution (5)
Sodium carbonate Formaldehyde
25.000 g 0.288 mL
2.50% 0.01%
Stop solution (6)
Titriplex III (Merck) Thimerosal
18.500 g 0.200 g
0.05 M 0.02%
c
a
Table is based on data from literature (22, 23) Distilled water; 1 L is the final volume c pH 11.3, adjusted with sodium hydrogen carbonate b
High-Resolution Large-Gel 2DE
325
Importantly, only the total extraction protocol described in Chapter “Protein Extraction for 2DE” can be used for DIGE sample preparation. After labeling samples according to manufacturer’s instructions (CyDye DIGE Fluors (minimal dyes) for Ettan DIGE, GE Healthcare) samples are separated as described earlier. IEF and SDS-PAGE equipment are covered with metal foil to shield the fluorescent labeled proteins from light.
3. Method 3.1. Preparation of Isoelectric Focusing Gels
1. Insert two hollow glass tubes for 40-cm IEF gels with nylon strings in start position (Fig. 1) into the tube stand (Fig. 2) (see Note 5). 2. Thaw one unit of each: separation gel solution, cap gel solution, and ammonium persulfate solution. Incubate the gel solutions for 30 min at 25–30°C in a water bath. Immediately thereafter connect the vials containing the gel solution and the cap gel solution to a water-jet pump and degas both solutions for 5 min. Air bubbles usually stick to the wall and must be removed by gently knocking at the container. 3. Mix 25 μL ammonium persulfate solution (no vortexing!) into the gel solution and pour it into the gel solution groove of the tube stand (Fig. 2). Use the procedure described in the legend to Fig. 1 for filling the glass tubes with gel solution (see Note 6). 4. Mix cap gel solution with 10 μL ammonium persulfate immediately before use and proceed as described in item 3. 5. Polymerize the filled gel tubes for 30 min at room temperature (Fig. 1). 6. Remove the gel tubes from the tube stand, transfer a drop of water onto the opening of each tube end (without touching the gel surface), and tightly seal the ends with parafilm. 7. Polymerize IEF gels at 37°C for 40 min (see Note 7). 8. Cast 20-cm gels (see legend to Fig. 1) as described for 40-cm gels. 9. Use one portion of separation gel solution to produce four or five 40-cm gels.
3.2. IEF Apparatus and Sample Application
1. Insert four tubes with 40-cm gels or eight tubes with 20-cm gels into the upper chamber of the IEF apparatus with the cap gel side of the tubes directed downward.
326
Zabel and Klose
2. Fill the lower (cathode) chamber with 400 mL ethylenediamine solution. Afterward fill the lower gel-free ends of all gel tubes with ethylenediamine solution. 3. Mount the two buffer reservoirs one upon the other separated by a cylinder, its heights selected according to tube length. 4. Dry the surface of the gel-free upper end of each gel with a fine strip of filter paper. 5. Mix one portion of Sephadex suspension with 270 mg urea, 25 μL DTT, and 25 μL of ampholyte mixture. (Sephadex mixture is not reused.) 6. As described in Subheading 3.2, apply the protein sample to the anodic side of the gel. Layer Sephadex, sample, protection solution, and phosphoric acid upon the gel as shown in Fig. 1 (see Note 8). This layering is performed with fine Pasteur pipettes, except for sample application which is carried out with a microliter syringe suitable for 10 μL volumes (type 701 RN + Chaney, Hamilton, Bonadutz, Switzerland). It is important to determine protein concentrations in the samples prior to application. Sample volume applied is corrected by protein concentration to guarantee that the same amount of protein per sample is loaded. Roughly, 8 μL or 100 μg of protein is applied to one gel. 7. Fill the upper (anode) reservoir with 500 mL of phosphoric acid solution. 3.3. Isoelectric Focusing
The upper chamber of the IEF apparatus is connected to the anode, the lower chamber to the cathode of a power supply, and the electric current is set as indicated in Table 7 (see Note 9). Within a series of experiments the number of gels run in parallel is always the same. Moreover, the same electrophoresis chambers and the same power supply are used.
3.4. Equilibration and Storage of IEF gels (See Note 10)
1. Thaw one unit of equilibration solution, bring to room temperature, and distribute among all gel grooves (Fig. 3a). One gel groove per half gel is used. 2. When the IEF run is finished, the upper buffer reservoir with IEF tubes attached is removed from the apparatus and the buffer is drained. Any solutions above the IEF gels are removed from both sides. 3. Remove the first tube from the chamber. Determine the length of the IEF gel within the tube. The gel usually shrinks somewhat during the run. Half the gel length is determined. Usually the gel is extruded opposite the sample application side (here: cap gel side). At this side half the gel length is marked.
High-Resolution Large-Gel 2DE
327
Table 7 Power supply settings for 2DE SDS-PAGEa
IEF 40-cm Gels
20-cm Gels
Time schedule
Vb
Time schedule
Vb
Time schedule
mA
Vc
1h
100
1h
100
15 min
65
150–160 (start)
1h
300
1h
200
6 h ± 5 min
85
210–230 (15 min after start)
23 h
1,000
17.5 h
400
30 min
1,500
1h
650
10 min
2,000
30 min 10 min 5 min
1,000 1,500 2,000
850–900 (final)
a
Two gels in parallel Current (mA) and power (W) regulators of the power supply were adjusted to maximum c Voltage (V) regulator of the power supply was adjusted to maximum b
4. Using the nylon string (Fig. 1), extrude a 0.9-mm IEF gel directly into a gel groove (Fig. 3a). The equilibration solution helps keep the gel relaxed and movable. For 1.5-mm IEF gels, a 1-mL syringe capped with a 200-μL pipette tip filled with bidistilled water is attached opposite the sample application side. Pressing the water into the tube extrudes the IEF gel (Note 10). 5. Arrange the grooves carrying the gels of a run in a special box (Fig. 3c) and store at −70°C. The whole procedure, from finishing the IEF run to freezing the gels, must be performed as quickly as possible to restrict diffusion of proteins within the IEF gel. 3.5. Preparation of IEF Gels for SDS-PAGE (See Note 11)
1. Assemble two gel cells (Fig. 4) according to the instructions of the manufacturer using silicone grease to seal the cells instead of adhesive tape and silicone gaskets. Carefully clean the inner sides of the glass plates to remove any dust that may produce “point streaks” in the silver-stained protein patterns. Insert the two gel cells into the polymerization stand that has been placed on a horizontally adjusted plate. When employing the DIGE technique the width of the glass plates used may be of importance. Some fluorescence scanners may only be able to compensate a certain glass width before the laser beam gets out of focus. Small width differences in glass plates can
328
Zabel and Klose
be compensated for by spacers to accommodate them into the clamps holding the glass plates. 2. Thaw two units of gel solution and one pre-prepared solution of ammonium persulfate and overlay solution. Incubate the gel solutions for 30 min at 40°C in a water bath. 3. To each of the two units of gel solution, mix 4.5 mL ammonium persulfate without vortexing (to avoid strong air contact of the gel solution). Immediately thereafter pour the gel solutions into the gel cells up to the rim of the glass plates using a special funnel (Fig. 4). Remove tiny air bubbles between the glass plates using a long, thin wire. 4. Using a syringe mounted at the bottom of the polymerization stand, slowly and steadily push 1.2 mL underlay solution under the gel solution. Since in the meantime the gel solution may have dropped a few millimeters below the edge of the glass plates, gel solution should be supplied to level with the rim of the glass plates again. 5. Put a strip of parafilm onto the upper slit of each glass cell so that an air-free contact to the gel solution is warranted. Allow the gel solutions to polymerize for 30 min. 6. Immediately (not later), turn the gel cells upside-down and insert into the gel stand, but do not firmly fix them by clamps. Remove the underlay solution, rinse and cover the gel surface with overlay solution. Leave the gels for 1 h at room temperature to further polymerize. 7. Completely remove the solution on the gel surface, rinse the surface with overlay solution, and store at 4°C overnight. 3.6. SDS-PAGE Apparatus and Transfer of the IEF Gel
1. Prepare 11 l of fresh electrode buffer, fill the lower compartment of the electrophoresis apparatus, and precool to 15°C using a circulating cooling machine. Mix 1.5 L of the same buffer (use bidistilled water) with 0.2 mL bromophenol blue solution and use for the upper buffer compartment. 2. Thaw and mix one unit of pre-prepared agarose buffer with 0.1 g agarose. Dissolve the agarose at 70°C and use at 40°C to overlay IEF gels. 3. Remove overlay buffer from the two gel cells prepared previously and rinse the gel surface with the same buffer. Afterward keep the gels at an angle and remove any excess buffer. The gel cells with holding device should be placed on a level surface. 4. Take two gel grooves, e.g., one carrying the “acid” half and one the “basic” half of a 40-cm IEF gel (Note 11), from the −70°C freezer (Fig. 4). When the gels have thawed, incubation solution, if present in excess on the grooves, is removed but
High-Resolution Large-Gel 2DE
329
for a little amount to keep the gels wet and facilitate their slipping over the glass plates into the groove on top of the SDS-PAGE gels without pushing. Bring the IEF gel holding groove into contact with the edge of the glass cell (Fig. 3b) and manipulate the gel with a special wire (Fig. 3d) to transfer it onto the surface of the SDS-PAGE gel. Caution: This must be done without stretching the gel. At the end of this step, the IEF gel should be in tight contact with the SDS-PAGE gel, i.e., inclusion of air or solution between the gels must be avoided. Remove excess solution that still remains with the IEF gel. 5. Overlay the IEF gels with agarose solution up to the edges of the glass cells (see Note 12). When agarose is polymerized, insert the two cells into the electrophoresis apparatus. 6. Slowly fill the upper compartment with electrode buffer containing bromophenol blue. 3.7. SDS-PAGE
The electrophoresis apparatus is connected to a power supply using the upper buffer compartment as cathode, the lower compartment as anode. The electric current is regulated as indicated in Table 7 (see Note 13). The lower buffer temperature is kept at 15°C (see Note 14). The transfer of IEF gels to the second dimension, a process which starts after thawing the IEF gels and ends when starting electrophoresis, should be performed as quickly as possible to restrict diffusion of proteins. Furthermore, within a series of experiments, the same apparatus should always be used. SDS-PAGE is finished when the bromophenol blue line in the gels reaches a marking located two cm before the lower end of the glass plates (Fig. 4). After the run the two gels are immediately removed from the glass plates and transferred, each into a plastic trough containing 1 L fixation solution (Table 8).
3.8. Silver Staining (See Note 15)
The gels are shaken for 2 h in fixation solution and then remaining in this solution overnight. Each gel is subsequently transferred into a large, transparent plastic trough (bottom: 30 × 40 cm; Bruckle-Labo-Plast, Lörrach, Germany) which is carefully cleaned each time prior to use. The staining procedure is outlined in Table 8 (see Note 16). During the entire procedure the gels remain continuously shaking in the same trough. After each step the used solution is removed by a water-jet pump.
3.9. Gel Drying
1. Transfer a stained gel along with two sheets of Cellophane and several sheets of filter paper onto the perforated steel plate as described in Fig. 5. Prior to this step, trim the edges of the gel below and lateral to the protein pattern to the size of the steel plate. Soak cellophane and filter paper in water (cellophane for
330
Zabel and Klose
Table 8 Silver staining procedurea
Gel treatment
Volume of solution per gel (identification no. of solution as indicated in Table 6)
Length of treatment
Fixation
1 L solution 1
2h
Incubation
1 L solution 2
2h
Wash
4 L distilled water
20 min
Wash
4 L distilled water
20 min
Silver reaction
1 L solution 3
30 min
Wash
1 L distilled water
few seconds
Shake the gel just once by hand
Wash
1 L solution 4
1 min
Distribute arising gray clouds quickly
Pattern developing 1 L solution 5
3–5 min
Distribute arising gray clouds quickly, and watch the staining intensity
Stop development
1 L solution 6
20 min
Solution 6 intensifies the stain of the protein spots
Wash
1 L distilled water
10 min
Removes excess solution 6
a
Remarks Leave gel in solution 1 overnight
Table is based on data from literature (22, 23)
one h and filter paper briefly) before use. When layering the sheets on the plate, extrude excess water manually. 2. Put the steel plate, when fully loaded, onto the drying plate installed in the incubator and further proceed as described in Fig. 5. The gels dry by exposing them to vacuum and 80–90°C for 2–2.5 h. During this period the manometer drops from about 150 mbar to almost zero. 3. The drying procedure is finished when the rubber tubing outside the incubator feels cool and is dry inside the tube (use transparent polyethylene tubes). 4. Label the dried gels and store the acid and basic halves of a gel dry and in the dark. 3.10. Quality Criteria for 2DE Patterns
1. The quality of a 2-DE pattern can be judged based on three criteria: (I) Separation: In a 2DE pattern of high quality, protein spots are highly defined (sharp) and well concentrated;
High-Resolution Large-Gel 2DE
331
no streaks or background staining occurs. (II) Resolution: In a high-quality 2DE pattern of a complex protein sample (e.g., total extract of brain tissue) several thousand protein spots are visible; they are well distributed over the gel area, they do not overlap, and both very weak and tiny spots as well as very dark and large spots appear in the same pattern. (III) Reproducibility: High quality means (i) several 2DE patterns resulting from the same sample match perfectly in all areas of the gel: the absolute distance of corresponding spots from the start point of their gels is the same in both directions. Reproducibility may be defined in a less stringent way as well: (ii) The absolute position of corresponding spots of different patterns is only within a certain section of the 2DE pattern the same. (iii) The relative position of corresponding spots of different patterns is the same, i.e., the ratio of the distances of one spot to several surrounding spots is for corresponding spots of different patterns in each pattern the same. In practice, definitions (ii) and (iii) are adequate, especially for large 2DE gels. Computer programs for matching different 2DE patterns use the higher reproducibility of the relative spot position (compared to the absolute position) to shift the corresponding spots of different patterns to the same absolute position. So far we have considered the reproducibility of the position of the spots. However, patterns from different runs of the same sample should also show a high similarity with regard to the (relative) protein quantity of each spot. 2. 2DE patterns which fulfill all the high-quality criteria mentioned cannot be created. Differences in the nature of individual proteins (e.g., solubility, sensitivity to proteases) and protein samples (see later) exclude identical behavior in the same 2DE system. This becomes obvious when, within a narrow gel area, sharp and concentric protein spots are seen adjacent to streaking spots. We also observe that the quality of a 2DE pattern depends to some extent on the origin of a protein sample. For example: protein extracts from embryonic mouse tissues generally result in protein patterns of higher quality than extracts from adult tissues; brain proteins yield better results than liver proteins. Moreover, some quality criteria are not compatible with each other, e.g., revealing the weakest and finest spots of a protein pattern usually leads to streaks and overlap among dark and large spots. 3. Fig. 6 shows a protein extract from brain embryonic development day 11.5 of C57BL/6 mice. The complexity of this protein sample was relatively high and the variation in spot intensity was large. Still, protein separation and resolution
332
Zabel and Klose
Fig. 6. Silver-stained large-gel 2DE pattern of a total protein extract from brain embryonal day 11.5 of C57BL/6 mice. 2DE was performed as described in the text. The protein sample was prepared according to the total protein extraction procedure described in Chapter “Protein Extraction for 2DE.” The figure shows an example of a protein pattern generated by a very complex tissue sample still showing good resolution and low background staining.
Fig. 7. Large-gel 2DE pattern of soluble mouse testis proteins. 2DE was performed as described in the text. Sample preparation followed the fractionated extraction protocol described in Chapter “Protein Extraction for 2DE.” Sample load per gel was 155 μg protein/10 μL sample. This pattern demonstrates a maximum of resolution. The protein spots of this pattern were counted manually by dotting each individual spot with an indelible pencil on a transparent plastic foil which was put onto the dried gel (10). 10,345 spots were detected.
High-Resolution Large-Gel 2DE
333
reached a quality level that can be considered quite good. Fig. 7 shows a pattern of even higher complexity (soluble fraction of mouse testis proteins) where highly abundant protein spots occur in certain areas. Consequently, more streaks and overlapping spots appear. On the other hand, the pattern reveals a maximum number of protein spots (we counted 10,345 spots; see ref 19). Sections from this pattern in an enlarged format (Fig. 8) show that even in such a complex pattern, quality of spot separation is still quite acceptable. The fractionated extraction procedure in Figs. 7 and 8 and large-gel 2DE is used by the advanced user interested in very high resolution or certain subcellular fractions of proteins.
Fig. 8. Sections of the protein pattern shown in. (A) Section of the right (basic) side of the pattern, the most crowded area. (B) Section of a rather sparsely occupied area of the left (acid) side of the pattern shown in Fig. 7B.
334
Zabel and Klose
4. Notes 1. A large-gel format was used to increase resolution for protein mixtures. The increase of resolution is achieved by increasing the separation area and therefore the distance between protein spots. The additional space gained in this way can be exploited to reveal additional spots by increasing the sample load or staining period. Still, separation of completely superimposed spots by increasing the gel size would be improved just slightly and would not contribute to an increase in spot number. Completely superimposed spots rarely occur even in smaller-sized (20× 30 cm) SDS-PAGE gels and so an increase in area means an increase in resolution. Increasing the gel size also increases the diffusion of protein spots due to the increased running time, and this reduces the free area gained to some degree. The limited complexity of the ampholyte mixtures also limits the increase in resolution obtained by increasing gel size. Therefore, increasing size beyond a certain point does not necessarily result in a linear increase of resolution. 2. The high reproducibility that can be achieved with the 2DE technique described was demonstrated elsewhere (19). Problems in obtaining reproducible protein patterns often result from sample preparation rather than from the 2DE procedure itself. Furthermore, proteins are amphoteric in character and, therefore, may modify the pH gradient in an IEF gel, or in a distinct region of the gel in which a highly abundant protein dominates the region and therefore the pH. This problem may also occur in IPGs. Consequences of this protein behavior for reproducibility may become obvious when protein patterns of quite different sources (e.g., serum versus brain proteins) are compared. Lack in reproducibility of single spots may occur in the basic than in the acid region of the 2DE gel since using our IEF conditions. This is because under our IEF conditions the basic part of the pH gradient does not reach its equilibrium. 3. Thin IEF gels (0.9 mm) and thin SDS gels (0.75 mm) were found to yield sharper spots and therefore better resolution than thicker gels (1.5–5 mm). If, however, protein spots have to be extracted from the gels for identification by MS, 1.5mm gels in combination with higher sample load (50–60 μL (or 40 μL total extracts) containing 0.6–0.8g protein) still resulted in protein patterns comparable to thin gels. 4. To achieve high reproducibility of 2DE patterns, the gel solutions for IEF and SDS-PAGE are produced in large batches, almost complete in their composition, and stored in
High-Resolution Large-Gel 2DE
335
ready-made units at −70°C. For a set of experiments gels are prepared from the same batch of solutions. Negative effects were not observed when gel solutions were frozen for more than 2 years. Source and quality of the chemicals used should remain the same, once found to be satisfactory. This is noteworthy particularly with regard to ampholytes and SDS. 5. Cleaning the glass tubes: Used glass tubes are rinsed by pressing distilled water through the capillary canal. The tubes are then kept for 30 min at 60–80°C in a glass cylinder containing 6% Deconex 12 PA solution (Borer Chemie, Zuchwil, Switzerland). Afterward the tubes are rinsed with water as described and kept for 30 min at 95°C in 0.1 N HCl solution. The tubes are rinsed again, dried by air pressure, and stored dust free. The heating steps are performed in an incubator using preheated solutions. 6.
Layering of several solutions (separation and cap gel solution) in a 0.9-mm tube is difficult and requires some experience. When pulling the nylon strings the strain should be constant to avoid a build up of air bubbles. The solutions should be added into the tube as indicated by markings 1, 2 and 3 in Fig. 1.
7. IEF gels, when prepared and polymerized in glass tubes, were kept at 37°C for 40 min. and were then stored overnight at room temperature. This improves the reproducibility of IEF probably because polymerization of the soft gels reaches the final stage at temperatures higher than room temperature. Furthermore, we occasionally observed the appearance of diagonal streaks at some protein spots of the 2DE pattern, particularly at the prominent spots. This might be due to interactions of these proteins with still soluble components of a not fully polymerized gel. This phenomenon occurred to a lower extent, or not at all, if the gels were kept at 37°C for 40 min and were stored at room temperature overnight. 8. The function of the Sephadex layer is to protect (to some extent) the gel surface from shrinking and becoming clogged by protein precipitates which may occur when the proteins become highly concentrated above the gel shortly after starting the IEF. 9. Owing to high ionic strength of ethylenediamine in the lower electrode solution, very basic proteins (e.g., histones) accumulate at the end of the gel (in front of the cap gel) rather than moving out of the gel. If this group of proteins is of particular interest (see Chapter “Protein Extraction for 2DE,” pellet suspension), the running time for IEF should be reduced by 2 h at the 1,000 V level to obtain a better spreading of these protein spots.
336
Zabel and Klose
10. Usually, satisfactory results are obtained when IEF gels are just extruded into equilibration solution on gel grooves prior to use or freezing. Still, if solubilization problems occur the following procedure should be adhered to. IEF gels should be incubated in equilibration solution to expose the proteins to a high concentration of SDS to improve solubility before separation by SDS-PAGE. However, at this step a portion of proteins becomes lost by diffusion. The main disadvantage is that diffusion may affect proteins to a different degree: especially small proteins and proteins which do not focus perfectly (and therefore still preserve their solubility) may become preferentially lost. Therefore, we determined a maximal incubation time of 10 min. 11. A large-gel format was used to increase resolution for protein mixtures. The increase of resolution is achieved by increasing the separation area and therefore the distance between protein spots. The additional space gained in this way can be exploited to reveal additional spots by increasing the sample load or staining period. Still, separation of completely superimposed spots by increasing the gel size would be improved just slightly and would not contribute to an increase in spot number. Completely superimposed spots rarely occur even in smaller-sized (20 × 30 cm) SDS gels, and so an increase in area means an increase in resolution. Increasing the gel size also increases the diffusion of protein spots due to the increased running time, and this reduces the free area gained to some degree. The limited complexity of the ampholyte mixtures also limits the increase in resolution obtained by increasing gel size. Therefore, increasing gel beyond a certain point does not necessarily result in a linear increase of resolution. 12. Embedding the IEF gel in agarose may be omitted. This can result in sharper spots in the lower part of the SDS gel although the risk of IEF gel floating increases drastically. 13. When separating proteins by 2DE for identification, i.e., 1.5-mm IEF gels and 1.0-mm SDS-PAGE gels were used. In this case the SDS-PAGE gels were run at 87 mA and after 15 min at 116 mA. 14. The temperature during an SDS-PAGE run should be 15°C and not much lower. Otherwise a temperature gradient may arise across the gel: the temperature in the center of the gel is higher than at the two surfaces. Consequently, the protein molecules do not migrate in the same front throughout the gel width, and this affects the shape of the spots. 15. The quality of a stained protein pattern (background, staining intensity) depends on the quality of chemicals used
High-Resolution Large-Gel 2DE
337
(purity, age). This is particularly true for ethanol, formaldehyde, and glutaraldehyde. Furthermore, the quality of water is crucial for silver-stained protein pattern. Conductivity of water should be below 0.1 μS/cm. 16. For reasons unknown, thimerosal in the stop solution acts on stained spots as an intensifier and imparts a more homogeneous gray-violet color. However, the red centers in highly concentrated protein spots and some differences in spot color (problems relevant to densitometric measurements) are not avoided completely. Slowing the staining process by developing at 4°C, using less formaldehyde or increasing the bisacrylamide concentration of the separation gel did not solve this problem. In this respect, the use of Duracryl (Millipore) instead of acrylamide/bisacrylamide for the separation gel was found to be advantageous but only to a certain extent.
References 1. Klose, J. (1975) Protein mapping by combined isoelectric focusing and electrophoresis of mouse tissues. A novel approach to testing for induced point mutations in mammals, Humangenetik 26, 231–43. 2. O’Farrell, P. H. (1975) High resolution two-dimensional electrophoresis of proteins, J Biol Chem 250, 4007–21. 3. Zabel, C. (2006) Proteins are specifically altered in disease – really? Expert Rev Proteomics 3, 469–70. 4. Zabel, C., Sagi, D., Kaindl, A. M., Steireif, N., Klare, Y., Mao, L., et al. (2006) Comparative proteomics in neurodegenerative and non-neurodegenerative diseases suggest nodal point proteins in regulatory networking, J Proteome Res 5, 1948–58. 5. Zabel, C., Chamrad, D. C., Priller, J., Woodman, B., Meyer, H. E., Bates, G. P., et al. (2002) Alterations in the mouse and human proteome caused by Huntington’s disease, Mol Cell Proteomics 1, 366–75. 6. Lander, E. S., Linton, L. M., Birren, B., Nusbaum, C., Zody, M. C., Baldwin, J., et al. (2001) Initial sequencing and analysis of the human genome, Nature 409, 860–921. 7. Venter, J. C., Adams, M. D., Myers, E. W., Li, P. W., Mural, R. J., Sutton, G. G., et al. (2001) The sequence of the human genome, Science 291, 1304–51. 8. Stein, L. D. (2004) Human genome: end of the beginning, Nature 431, 915–6.
9. Gibson, F., Anderson, L., Babnigg, G., Baker, M., Berth, M., Binz, P.A., et al. (2007) MIAPE: gel electrophoresis, Nat Biotechnol 26, 863–4. 10. Anderson, W. F. (1994) Genes, genes, and more genes, Hum Gene Ther 5, 1077–8. 11. Klose, J., Nock, C., Herrmann, M., Stuhler, K., Marcus, K., Bluggel, M., et al. (2002) Genetic analysis of the mouse brain proteome, Nat Genet 30, 385–93. 12. McConkey, E. H. (1979) Double-label autoradiography for comparison of complex protein mixtures after gel electrophoresis, Anal Biochem 96, 39–44. 13. Hirsh, A. G. (1984) On the possibility of the production of many rare proteins by higher eukaryotes, J Theor Biol 110, 399–410. 14. Jungblut, P. R., Prehm, J., and Klose, J. (1987) An attempt to resolve all the various proteins of a single human cell type by twodimensional electrophoresis, J Biol Chem Hoppe-Seyler 368, 439. 15. Duncan, R., and McConkey, E. H.(1982) How many proteins are there in a typical mammalian cell?Clin Chem 28, 749–55. 16. Sitek, B., Potthoff, S., Schulenborg, T., Stegbauer, J., Vinke, T., Rump, L. C., et al. (2006) Novel approaches to analyse glomerular proteins from smallest scale murine and human samples using DIGE saturation labelling, Proteomics 6, 4337–45. 17. Sitek, B., Luttges, J., Marcus, K., Kloppel, G., Schmiegel, W., Meyer, H. E., et al.
338
Zabel and Klose
(2005) Application of fluorescence difference gel electrophoresis saturation labelling for the analysis of microdissected precursor lesions of pancreatic ductal adenocarcinoma, Proteomics 5, 2665–79. 18. Klose, J. (1983) High resolution of complex protein solutions by two-dimensional electrophoresis. In Modern Methods in Protein Chemistry - Review Articles (Tschesche, H. ed.), Walter de Gruyter Verlag, Berlin and New York, pp 49–78. 19. Klose, J., and Kobalz, U. (1995) Two-dimensional electrophoresis of proteins: an updated protocol and implications for a functional analysis of the genome, Electrophoresis 16, 1034–59. 20. Rabilloud, T., Adessi, C., Giraudel, A., and Lunardi, J. (1997) Improvement of the solubilization of proteins in two-dimensional electrophoresis with immobilized pH gradients, Electrophoresis 18, 307–16.
21. Laemmli, U. K. (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4, Nature 227, 680–5. 22. Heukeshoven, J., and Dernick, R. (1985) Simplified method for silver staining of proteins in polyacrylamide gels and the mechanism of silver staining, Electrophoresis 6, 103–12. 23. Jungblut, P. R., and Seifert, R. (1990) Analysis by high-resolution two-dimensional electrophoresis of differentiation-dependent alterations in cytosolic protein pattern of HL-60 leukemic cells, J Biochem Biophys Methods 21, 47–58. 24. Nebrich, G., Herrmann, M., Sagi, D., Klose, J., and Giavalisco, P. (2007) High MS-compatibility of silver nitrate-stained protein spots from 2-DE gels using ZipPlates and AnchorChips for successful protein identification, Electrophoresis 28, 1607–14.
Chapter 21 Silver Staining of Proteins in 2DE Gels Cécile Lelong, Mireille Chevallet, Sylvie Luche, and Thierry Rabilloud Summary Silver staining detects proteins after electrophoretic separation on polyacrylamide gels. Its main positive features are its excellent sensitivity (in the low nanogram range) and the use of very simple and cheap equipment and chemicals. The sequential phases of silver staining are protein fixation, then sensitization, then silver impregnation, and finally image development. Several variants of silver staining are described here, which can be completed in a time range from 2 h to 1 day after the end of the electrophoretic separation. Once completed, the stain is stable for several weeks. Key words: Two-dimensional electrophoresis, Staining, Silver, Fixation, Development, Detection.
1. Introduction Silver staining of polyacrylamide gels was introduced in 1979 by Switzer et al. (1) and rapidly gained popularity owing to its high sensitivity, ca. 100 times higher than staining with Coomassie blue. However, the first silver-staining protocols were not trouble-free. High backgrounds and silver mirrors were frequently experienced, with a subsequent decrease in sensitivity and reproducibility. This led many authors to suggest improved protocols, so that more than 100 different silver-staining protocols for proteins in polyacrylamide gels can be found in the literature. However, all of them are based on the same principle (2, 3) and comprise more or less four major steps.
David Sheehan and Raymond Tyther (eds.), Methods in Molecular Biology, Two-Dimensional Electrophoresis Protocols, vol. 519 © Humana Press, a part of Springer Science + Business Media, LLC 2009 DOI: 10.1007/978-1-59745-281-6_21
339
340
Lelong et al.
1. The first step is fixation, and aims at insolubilizing the proteins in the gels and removing interfering compounds present in the two-dimensional electrophoresis (2DE) gels (glycine, Tris, SDS, and carrier ampholytes present at the bottom of the gels). 2. The second step is sensitization and aims at increasing the subsequent image formation. Numerous compounds have been proposed for this purpose. All these compounds bind to the proteins, and are also able either to bind silver ion, or to reduce silver ion into metallic silver, or to produce silver sulfide (2, 3). This sensitization step is sometimes coupled with the fixation step. 3. The third step is silver impregnation. Either plain silver nitrate or ammoniacal silver can be used. 4. The fourth and final step is image development. For gels soaked with silver nitrate, the developer contains formaldehyde, carbonate, and thiosulfate. The use of the latter compound, introduced by Blum et al. (4), reduces dramatically the background and allows for thorough development of the image. For gels soaked with ammoniacal silver, the developer contains formaldehyde and citric acid. In this case, thiosulfate is better introduced at the gel polymerization step (5). Background reduction by thiosulfate can also be achieved by brief incubation in thiosulfate prior to development (6), or by inclusion in the developer. When the desired image level is obtained, development is stopped by dipping the gel in a stop solution, generally containing acetic acid and an amine to reach a pH of 7. Final stabilization of the image is achieved by thorough rinsing in water to remove all the compounds present in the gel. However, the development of downstream protein characterization methods, such as analysis by mass spectrometry (MS) starting from gel separated proteins, has brought additional constraints on silver staining. Besides the classical constraints weighing on any detection method (sensitivity, linearity, and homogeneity), the interface of the detection method with downstream methods becomes more and more important. In the case of MS, this interface comprises the compatibility with enzymatic digestion and peptide extraction, as well as the absence of staining-induced peptide modifications. While the exquisite sensitivity of silver staining is unanimously recognized, its compatibility with downstream analysis appears more problematic than staining with organic dyes (e.g., Coomassie blue). A mechanistic study (7) has shown that these problems are linked in part to the pellicle of metallic silver deposited on the proteins during staining, but are mainly due to the presence of formaldehyde during silver staining. Up to now, formaldehyde is the only chemical known able to produce a silver image of good
Silver Staining of Proteins in 2DE Gels
341
quality in protein staining, and attempts to use other chemicals have proven rather unsuccessful (7). However, besides a lowered peptide representation in silver stained gels (7), formaldehyde induces some peptide modifications, such as +12 and +30 Da adducts (8) as well as formylation (9), the latter being most likely caused by the formic acid produced upon reaction of formaldehyde with silver ions in the image development step. These problems are common to all silver-staining protocols, although their extent is variable from one protocol to another. Some guidelines for the choice of a silver-staining protocol are described in Note 1.
2. Materials 2.1. Equipment
1. Glass dishes or polyethylene food dishes. The latter are less expensive, have a cover, and can be easily piled up for multiple staining. They are however more difficult to clean, and it is quite important to avoid scratching of the surface, which will induce automatic silver deposition in subsequent staining. Traces of silver are generally easily removed by wiping the plastic box with a tissue soaked with ethanol. If this treatment is not sufficient, stains are easily removed with Farmer’s reducer (0.1% sodium carbonate, 0.3% potassium hexacyanoferrate (III), and 0.6% sodium thiosulfate). Thorough rinsing of the box with water and ethanol terminates the cleaning process. 2. Plastic sheets (e.g., the thin polycarbonate sheets sold by BioRad for multiple gel casting) Used for batch processing. 3. Reciprocal shaking platform: The use of orbital or threedimensional movement shakers is not recommended.
2.2. Reagents 2.2.1. General
Generally speaking, chemicals are of standard pro analysis grade. 1. Water: The quality of the water is of great importance. Water purified by ion exchange cartridges, with a resistivity greater than 15 MΩ/cm, is very adequate, while distilled water gives more erratic results. 2. Formaldehyde: Formaldehyde stands for commercial 37–40% formaldehyde. This is stable for months at room temperature. It should not be stored at 4°C, as this promotes polymerization and deposition of formaldehyde. The bottle should be discarded when a layer of polymer is visible at the bottom of the bottle. 3. Sodium thiosulfate solution: 10% solution of crystalline sodium thiosulfate pentahydrate in water. Small volumes of this solution (e.g., 10 mL) are prepared fresh every week and stored at room temperature.
342
Lelong et al.
4. Ethanol: Technical-grade alcohol can be used, and 95% ethanol can be used instead of absolute ethanol, without any volume correction. The use of denatured alcohol is however not recommended. 5. Citric acid solution: 2 M citric acid. Stored at room temperature for months. 6. Potassium acetate solution: 5 M potassium acetate. This is stable for months at room temperature. 7. Silver nitrate solution: 1 N silver nitrate. A 1 N silver nitrate solution (Fluka) is less expensive than solid silver nitrate and is stable for months if kept in a fridge (a dark and cold place). 8. Sodium hydroxide solution: 1 N sodium hydroxide (Fluka). 9. Ammonium hydroxide solution: 5 N ammonium hydroxide (Aldrich). Ammonium hydroxide is kept in the fridge. 2.2.2. Solutions for Silver Staining with Ammoniacal Silver and Formaldehyde Fixation
1. Fix solution I: 25% ethanol, 4% formaldehyde (see Note 2). 2. Sensitivity-enhancing solution I: 0.05% 2–7 Naphthalene disulfonate disodium salt (NDS) (Acros chemicals). 3. Ammoniacal silver solution: To prepare 500 mL of this solution (sufficient for a batch of four gels) place 480 mL of water in a flask under strong magnetic agitation. Add successively 7.5 mL of 1 N sodium hydroxide, 7.5 mL of 5 N ammonium hydroxide, and 12 mL of 1 N silver nitrate. Upon addition of silver nitrate, a transient brown precipitate forms, which should redissolve within a few seconds (see Note 3). Only clear solutions should be used. 4. Development solution I: 1 mL of 37% formaldehyde and 180 μL of 2 M citric acid per liter (see Note 4). 5. Stop solution I: 20 mL acetic acid and 5 mL ethanolamine per liter.
2.2.3. Solutions for Silver Staining with Ammoniacal Silver (10)
1. Fix solution II: 5% acetic acid/30% ethanol and 0.05% 2–7 Naphthalene disulfonate disodium salt (NDS) (Acros chemicals). 2. Ammoniacal silver solution: To prepare 500 mL of this solution (sufficient for a batch of four gels) place 480 mL of water in a flask under strong magnetic agitation. Add successively 7.5 mL of 1 N sodium hydroxide, 7.5 mL of 5 N ammonium hydroxide, and 12 mL of 1 N silver nitrate. Upon addition of silver nitrate, a transient brown precipitate forms, which should redissolve within a few seconds (see Note 3). Only clear solutions should be used. 3. Development solution I: 1 mL of 37% formaldehyde and 180 μL of 2 M citric acid per liter (see Note 4). 4. Stop solution I: 20 mL acetic acid and 5 mL ethanolamine per liter.
Silver Staining of Proteins in 2DE Gels 2.2.4. Solutions for Fast Silver Staining
343
1. Fix solution III: 5% acetic acid, 30% ethanol. 2. Sensitivity-enhancing solution II: 2 mL of 10% thiosulfate per liter. 3. Silver stain solution II: and 12.5 mL of 1 N silver nitrate per liter. 4. Development solution II: 30 g anhydrous potassium carbonate, 250 μL 37% formaldehyde and 125 μL 10% thiosulfate per liter (see Note 5). 5. Stop solution II: 40 g of Tris and 20 mL of acetic acid per liter.
2.2.5. Solutions for Long Silver Nitrate Staining for Free-Floating Gels (11)
1. Fix solution IV: 5% acetic acid/30% ethanol. 2. Sensitivity-enhancing solution III: 0.5 M potassium acetate, 25% ethanol, and 3 g potassium tetrathionate per liter (see Note 6). 3. Silver stain solution III: 12.5 mL 1 N silver nitrate per liter. 4. Development solution III: 30 g anhydrous potassium carbonate, 250 μL 37% formaldehyde, and 125 μL 10% thiosulfate per liter (see Note 5). 5. Stop solution III: 40 g of Tris and 20 mL of acetic acid per liter.
3. Methods 3.1. General Practice
Batches of gels (up to four gels per box) can be stained. For a batch of 3 or 4 medium-sized gels (e.g., 160 × 200 × 1.5 mm), 1 L of the required solution is used, which corresponds to a solution/gel volume ratio of at least 5. 500 mL of solution is used for 1 or 2 gels. Batch processing can be used for every step longer than 5 min, except for image development, where one gel per box is required. For steps shorter than 5 min, the gels should be dipped individually in the corresponding reagent(s). For changing solutions, the best way is to use a plastic sheet. This is pressed on the pile of gels with the aid of a gloved hand. Inclining the entire setup allows to empty the box while keeping the gels in it. The next solution is poured with the plastic sheet in place, which prevents the flow to fold or break the gels. The plastic sheet is removed after the solution change and kept in a separate box filled with water until the next solution change. This water is changed after each complete round of silver staining. When gels must be handled individually, they are manipulated with gloved hands. The use of powder-free, nitrile, gloves
344
Lelong et al.
is strongly recommended, as standard gloves are often the cause of pressure marks. Except for development or short steps, where occasional hand agitation of the staining vessel is convenient, constant agitation is required for all the steps. A reciprocal (“ping-pong”) shaker is used at 30–40 strokes per min. Four different silver-staining protocols are detailed. The rationale for choosing one of them according to the constraints brought by the precise 2DE protocol used and the requirements of the experimenter are described in Note 1. 3.2. Silver Staining with Ammoniacal Silver and Formaldehyde Fixation
This protocol is based on the original protocol of Eschenbruch and Bürk (12), with modifications (5, 13, 14). For optimal results, alterations must be made to gel casting. Piperazine diacrylamide is used as crosslinker in place of Bis (in a weight-to-weight substitution) (13), but this is not mandatory, and thiosulfate is added at the gel polymerization step (5). Practically, the initiating system is composed of 1 μL of TEMED, 10 μL of 10% sodium thiosulfate solution, and 10 μL of 10% ammonium persulfate solution per mL of gel mix. This ensures correct gel formation and gives minimal background upon staining. After electrophoresis, carry out silver staining as follows: 1. Rinse gels in water for 5–10 min. 2. Soak gels in fix solution I for 1 h. 3. Rinse 2 × 15 min in water. 4. Sensitize overnight in sensitivity-enhancing solution I. 5. Rinse 6 × 20 min in water. 6. Impregnate for 30–60 min in the ammoniacal silver solution. 7. Rinse 3 × 5 min in water. 8. Develop image (5–10 min) in development solution I. 9. Stop development in stop solution I. Leave in this solution for 30–60 min. 10. Rinse with water (several changes) prior to drying or densitometry.
3.3. Silver Staining with Ammoniacal Silver (10)
1. Fix the gels in fix solution II (1 h + overnight) (see Note 2). 2. Rinse in water (6 × 15 min). 3. Impregnate for 30–60 min with ammoniacal silver solution. 4. Rinse 3 × 5 min in water. 5. Develop image (5–10 min) in development solution I. 6. Stop development in stop solution I. Leave in this solution for 30–60 min. 7. Rinse with water (several changes) prior to drying or densitometry.
Silver Staining of Proteins in 2DE Gels
3.4. Fast Silver Staining
345
This protocol is based on that of Blum et al. (4), with modifications (14). 1. Soak the gels in fix solution II for at least 3 × 30 min (see Note 7). 2. Rinse in water for 3 × 10 min. 3. To sensitize, soak gels for 1 min (1 gel at a time) in sensitivityenhancing solution II. 4. Rinse 2 × 1 min in water (see Note 8). 5. Impregnate for at least 30 min in silver solution II (see Note 9). 6. Rinse in water for 5–15 s (see Note 10). 7. Develop image (10–20 min) in development solution II (see Note 5). 8. Stop development (30–60 min) in stop solution II. 9. Rinse with water (several changes) prior to drying or densitometry.
3.5. Long Silver Nitrate Staining for FreeFloating Gels (11)
1. Fix the gels in fix solution IV (3 × 30 min). 2. Sensitize overnight in sensitivity-enhancement solution IV. 3. Rinse in water (6 × 20 min). 4. Impregnate for 1–2 h with silver in a silver solution IV. 5. Rinse with water for 5–10 s (see Note 10). 6. Develop image (10–20 min) in development solution IV (see Note 5). 7. Stop development (30–60 min) in stop solution IV. 8. Rinse with water (several changes) prior to drying or densitometry.
4. Notes 1. From the rather simple theoretical basis described in the introduction, more than 100 different protocols were derived. The changes from one protocol to another are present either in the duration of the different steps or in the composition of the solutions. The main variations concern either the concentration of the silver reagent, or the nature and concentration of the sensitizers. Only a few comparisons of silver-staining protocols have been published (14, 15). From these comparisons, selected protocols have been proposed in the former sections. The choice of a protocol will depend on the constraints of the experimental setup and of the requisites of the experimentator (speed, reproducibility, etc.). The following guidelines can be suggested.
346
Lelong et al.
(a) The maximum sensitivity is not widely different from one protocol to another. The main differences are in the uniformity of the staining from one protein to another, the reproducibility of the staining, the speed of the method, and its adaptation to the various 2DE protocols. For example, the carrier ampholytes that are present in all 2DE protocols require rigorous fixation steps which are not mandatory when simple SDS gels are to be silver stained. There is a tradeoff between speed and long-term reproducibility. Fast protocols use short steps (less than 5 min) and thus a more transient chemistry, which are difficult to keep reproducible (b) Generally speaking, methods using ammoniacal silver give very uniform results, with minimal color effects and improved compatibility with MS (10). They are far more sensitive than silver nitrate-based methods for the staining of basic proteins and are therefore strongly recommended for 2DE gels with very wide pH gradients. However, these methods suffer from a number of minor drawbacks which prevent their universal use. (c) The silver reagent is very sensitive to the ammonia concentration. As ammonia is highly volatile, this introduces problems for the long term reproducibility of the method. This problem can be alleviated to a large extent by the use of commercial titrated ammonia solutions. (d) Ammoniacal silver is not compatible with all SDS gel systems. Systems using Tricine or Bicine as trailing ions are not compatible with ammoniacal silver staining. (e) Ammoniacal silver staining is not recommended for gels supported by a plastic film. Silver mirrors are frequently encountered in this case. (f) Optimal protocols for ammoniacal silver staining (e.g., Subheading 3.2) are generally time consuming. In addition, optimal results are obtained with the use of home-made gels, containing PDA as a crosslinker (13) and with thiosulfate included at the gel polymerization step (5). This prevents the use of commercial gels. Moreover, these protocols give best results when aldehydes (formaldehyde or glutaraldehyde) are used as fixers/sensitizers. This prevents any recovery of the silver-stained protein for subsequent use (e.g., MS). This drawback can be however alleviated (see Subheading 3.3) at the expense of uniformity of the staining (Figs.1, 2). (g) Silver staining being a delicate process (2, 3), the temperature control in the laboratory plays a role in the optimal silver-staining protocol. Ammoniacal silver is hampered by low temperatures (less than 18°C), while high temperatures (more than 30°C) produce a yellow background in silver nitrate staining.
Silver Staining of Proteins in 2DE Gels
347
Fig. 1. Global comparison of silver-stained gels.100 μg of proteins prepared from a complete cell extract (WEHI 274 monocytic cells) are loaded on two-dimensional gels (pH 3–10.5range, 10% acrylamide). The gels are then stained by different silver-staining methods described in the chapter. (a) ultrafast method; (b) fast silver nitrate; (c) long silver nitrate; (d) Ammoniacal silver without aldehyde fixation; (e) Ammoniacal silver with formaldehyde fixation. The Ammonical silver methods are more sensitive with basic proteins but less with acidic proteins, as shown in Fig. 2. Box: gel zone magnified in Fig. 2.
2. Other fixation processes can be used. For gels running overnight, a shorter process can be used. For ammoniacal silver staining (Subheading 3.3), fix the gel in fixing solution II for 3 × 30 min, and then rinse in water for 4 × 15 min. Return to standard protocol in step 4 (silver impregnation). (a) For silver nitrate staining, reduce fixation to a single 30-min bath (16). This will improve sequence coverage in mass spectrometry, at the expense of a strong chromatism (spots can be yellow, orange, brown, or gray), making image analysis difficult. Furthermore, ampholytes are not removed by short fixation and give a gray background at the bottom of the 2DE gels.
348
Lelong et al.
Fig. 2. Detail of silver staining in the acidic region. A small homologous gel zone is compared from the gels shown in Fig. 1. The gel panels are in the same order compared to the total gels shown in Fig. 1. Hollow spots (arrows) are obvious in gels stained with ammoniacal silver (panels D and E), but also in the gel stained with the ultrafast method (panel A).
3. The composition of the ammoniacal silver solution has an important influence on the final sensitivity. The ammonia/ silver mole ratio is in fact the key parameter (12). The solution given in this protocol has an ammonium/silver molar ratio of 3.1, which ensures maximal sensitivity, but less stability of the solution. If a brown precipitate remains in the solution, this means that the ammonia solution is no longer concentrated enough. The best remedy is to discard the ammoniacal silver solution and to prepare a new one with a new bottle of 5 N ammonium hydroxide. If this is not possible, add small aliquots of ammonia to the precipitated ammoniacal silver solution until the solution becomes clear. The sensitivity will be however lesser than usual. If reduced sensitivites are required, increase the ammonium hydroxide concentration by a factor of 1.3–2. This will progressively decrease the sensitivity. 4. In a standard analytical 2DE gel loaded with 50–100 μg of protein, the first major spots should begin to appear within 1 min. Delayed appearance indicates lower than expected
Silver Staining of Proteins in 2DE Gels
349
sensitivity, but is observed when aldehydes have not been used in the fixing process. In the latter case, sensitivity is restored by a longer development. The developer should be altered if no thiosulfate is present in the gel (e.g., use of ready-made gels). To prevent the rapid appearance of background, add 10 μL of 10% thiosulfate per liter of developer. The maximum sensitivity will not be altered by this variation. However, some spots will show a lighter color or give hollow spots. 5. When the gel is dipped in the developer, a brown microprecipitate of silver carbonate should form. This precipitate must be redissolved to prevent deposition and background formation. This is simply achieved by immediate agitation of the box. Do not expect the appearance of the major spots before 3 min of development. The spot intensity reaches a plateau after 15–20 min of development, and then background appears. Stop development at the beginning of background development. This ensures maximal and reproducible sensitivity. 6. The sensitization solution is prepared as follows. Dissolve potassium tetrathionate in water (half the desired final volume). After complete dissolution, add the required volumes of concentrated potassium acetate and ethanol. Fill up to the final volume with water. 7. The fixation process can be altered if needed. The figures indicated in the protocol are the minimum times. Gels can be fixed without any problem for longer periods. For example, gels can be fixed overnight, with only one solution change. For ultrarapid fixation, the following process can be used (16): (a) Fix in 10% acetic acid/40% ethanol for 10 min, then rinse for 10 min in water. (b) Post fix in 0.05% glutaraldehyde/40% ethanol and 100 μL/L 37% formaldehyde for 5 min. (c) Rinse in 40% ethanol for 2 × 10 min and then in water for 2 × 10 min. Proceed to step 3. 8. The optimal setup for sensitization is the following. Prepare four staining boxes containing the sensitizing thiosulfate solution, water (two boxes), and the silver nitrate solution, respectively. Put the vessel containing the rinsed gels on one side of this series of boxes. Take one gel out of the vessel and dip it in the sensitizing and rinsing solutions (1 min in each solution). Then transfer to silver nitrate. Repeat this process for all the gels of the batch. A new gel can be sensitized while the former one is in the first rinse solution, provided that the 1 min time is kept (use a bench chronometer). When several batches of gels are stained on the same day, it is necessary to prepare several batches of silver solution. However, the sensitizing and rinsing solutions can be kept for at least three batches and probably more.
350
Lelong et al.
9. Gels can be impregnated with silver for at least 30 min and at most 2 h without any change in sensitivity or background. 10. This very short step is intended to remove the liquid film of silver solution brought with the gel.
References 1. Switzer, R.C., Merril, C.R., and Shifrin, S. (1979) A highly sensitive silver stain for detecting proteins and peptides in polyacrylamide gels. Anal. Biochem. 98, 231–237 2. Rabilloud, T. (1990) Mechanisms of protein silver staining in polyacrylamide gels: a ten years synthesis. Electrophoresis 11, 785–794 3. Rabilloud, T., Vuillard, L., Gilly, C., and Lawrence, J.J. (1994) Silver staining of proteins in polyacrylamide gels: a general overview. Cell. Mol. Biol. 40, 57–75 4. Blum, H., Beier, H., and Gross, H.J. (1987) Improved silver staining of plant proteins, RNA and DNA in polyacrylamide gels. Electrophoresis 8, 93–99 5. Hochstrasser, D.F., and Merril, C.R. (1988) Catalysts for polyacrylamide gel polymerization and detection of proteins by silver staining. Appl. Theor. Electrophor. 1, 35–40 6. Wiederkehr, F., Ogilvie, A., and Vonderschmitt, D. (1985) Two-dimensional gel electrophoresis of cerebrospinal fluid proteins from patients with various neurological diseases. Clin. Chem. 31, 1537–1542 7. Richert, S., Luche, S., Chevallet, M., Van Dorsselaer, A., Leize-Wagner, E., and Rabilloud, T. (2004) About the mechanism of interference of silver staining with peptide mass spectrometry. Proteomics 4, 909–916 8 Metz, B., Kersten, G.F., Baart, G.J., de Jong, A., Meiring, H., ten Hove, J., et al. (2006) Identification of formaldehyde-induced modifications in proteins: reactions with insulin. Bioconjug. Chem. 17, 815–822 9. Oses-Prieto, J.A., Zhang, X., and Burlingame, A.L. (2007) Formation of {epsilon}-formyl-
10.
11.
12.
13.
14.
15.
16.
lysine on silver-stained proteins: Implications for assignment of isobaric dimethylation sites by tandem mass spectrometry. Mol. Cell. Proteomics 6, 181–192 Chevallet, M., Diemer, H., Luche, S., van Dorsselaer, A., Rabilloud, T., and Leize-Wagner, E. (2006) Improved mass spectrometry compatibility is afforded by ammoniacal silver staining. Proteomics 6, 2350–2354 Sinha, P., Poland, J.,Schnolzer, M., andRabilloud, T. (2001) A new silver staining apparatus for MALDI/TOF analysis of proteins after two-dimensional electrophoresis Proteomics 1, 835–840 Eschenbruch, M., and Bürk, R.R. (1982) Experimentally improved reliability of ultrasensitive silver staining of protein in polyacrylamide gels. Anal. Biochem. 125, 96–99 Hochstrasser, D.F., Patchornik, A., and Merril, C.R. (1988) Development of polyacrylamide gels that improve the separation of proteins and their detection by silver staining. Anal. Biochem. 173, 412–423 Rabilloud, T. (1992) A comparison between low background silver diammine and silver nitrate protein stains. Electrophoresis 13, 429–439 Ochs, D.C., McConkey, E.H., and Sammons, D.W. (1981) Silver staining for proteins in polyacrylamide gels: A comparison of six methods. Electrophoresis 2, 304–307 Shevchenko, A., Wilm, M., Vorm, O., and Mann, M. (1996) Mass spectrometric sequencing of proteins silver-stained polyacrylamide gels. Anal Chem. 68, 850–858
Chapter 22 Detection of 4-Hydroxy-2-Nonenal- and 3-NitrotyrosineModified Proteins Using a Proteomics Approach Rukhsana Sultana, Tanea Reed, and D. Allan Butterfield Summary Oxidative stress has been shown to be one of the mechanisms involved in a number of diseases, including neurodegenerative disorders, ischemia, cancer, etc. Oxidative stress occurs mainly due to an imbalance between oxidant and antioxidant systems. Oxidants can damage virtually all biological molecules including DNA, RNA, cholesterol, lipids, carbohydrates, proteins, and antioxidants. The oxidative modification of proteins has been shown to play an important role in a number of human diseases. And the methods to identify specific proteins that are susceptible to 4-hydroxy 2-nonenal (HNE) and 3-nitrotyrosine (NT) modifications are limited and difficult. Our laboratory uses two-dimensional polyacrylamide gel electrophoresis (2DE) in combination with western blotting to identify the specific targets of protein nitration and lipid peroxidation. This may require the analysis of thousands of individual proteins from cells and tissues, and coupling of mass spectrometry to this technique allows the identification of proteins. Since the protein levels and the protein oxidation can be obtained from 2DE and 2D blots, specific nitration or HNE modification of each protein spot can be easily calculated. Key words: Proteomics, Polyacrylamide gel electrophoresis, Western blotting, Protein nitration, 3-Nitrotyrosine, Lipid peroxidation, 4-Hydroxy-2-nonenal.
1. Introduction In recent years, oxidative stress, which is an imbalance in oxidant and antioxidant systems, has been shown to be one of the mechanisms involved in a number of diseases, including neurodegenerative disorders, ischemia, cancer, etc. Oxidants can damage virtually all biological molecules that include DNA, RNA, cholesterol, lipids, carbohydrates, proteins, and antioxidants. The indices used to study oxidative stress include: protein oxidation (protein carbonyl and 3-nitrotyrosine; 3-NT), lipid peroxidation, DNA oxidation, advanced glycation end products, and ROS formation (1–10). David Sheehan and Raymond Tyther (eds.), Methods in Molecular Biology, Two-Dimensional Electrophoresis Protocols, vol. 519 © Humana Press, a part of Springer Science + Business Media, LLC 2009 DOI: 10.1007/978-1-59745-281-6_22
351
352
Sultana, Reed, and Butterfield
One of the ways of measuring protein oxidation is quantifying the amount of 3-nitrotyrosine formation in a given sample. A number of studies support the notion that nitrosative stress also contributes to disease, for example neurodegeneration in Alzheimer’s disease (AD) (3, 7, 10–12). Reactive nitrogen species (RNS) generated by nitric oxide synthase (NOS) within a physiologically relevant concentration are not toxic and are relatively specific in their cellular targets (13). Protein nitration is a reversible and selective process and could be a cellular signaling mechanism, similar to protein phosphorylation (12, 14). RNS can also be produced in brain via overexpression of inducible and neuronal specific nitric oxide synthase (NOS: iNOS and nNOS, respectively) leading to increased levels of NO. In a neurodegenerative disease like AD, mitochondrial abnormalities occur (15), which could lead to leakage of O2−. These two species can react to produce peroxynitrite, an extremely strong oxidant that in the presence of CO2 can cause oxidative damage to proteins, lipids, and carbohydrates and might be involved in the deterioration observed in AD. The amino acids cysteine, methionine, tryptophan, phenylalanine, and tyrosine are particularly susceptible to nitration. Another marker of oxidative stress is 4-hydroxy-2-nonenal (HNE), a highly reactive lipid peroxidation product that can covalently bind to proteins by forming Michael adducts with cysteine, lysine, or histidine residues (16). HNE can cause membrane structural damage and produce diffusible secondary bioactive aldehydes and induce cell death in many cell types (17–22). In AD patients, the levels of free HNE in cerebrospinal fluid, amygdala, hippocampus, and parahippocampal gyrus were found to be significantly increased when compared with control subjects (20). Our laboratory is the first to use redox proteomics to identify oxidatively modified proteins in AD brain. We have applied our knowledge on proteomics to identify the nitrated proteins in AD IPL and hippocampus (3, 10). These identified nitrated proteins were found to be either directly or indirectly associated with AD pathology. We have also applied proteomics to identify HNEbound proteins in neurodegenerative disease (23). In this chapter, we describe the protocols we use in our laboratory to identify HNE and nitrated proteins in a given biological sample.
2. Materials 2.1. Sample Preparation for Detection of Protein Carbonyls
1. Sample homogenization buffer (pH 7.4): 10 mM HEPES, 137 mM NaCl, 4.6 mM KCl, 1.1 mM KH2PO4, 0.6 mM MgSO4, 0.5 μg/mL leupeptin (stored as an aliquot at −20°C), 0.7 μg/mL pepstatin (stored as an aliquot at −20°C), 0.5 μg/mL
Detection of 4-Hydroxy-2-Nonenal- and 3-Nitrotyrosine-Modified Proteins
353
trypsin inhibitor, 40 μg/mL PMSF dissolved in deionized water. Store at 4°C. 2. Laemmli buffer: 0.125 M Tris base pH 6.8, 4% SDS, and 20% glycerol. Store at room temperature. 3. Protein precipitation: Trichloroacetic acetic acid (100%, stored at 4°C) added at a final 7.5% of the total volume. 4. Lipid removal wash: 50:50 ethanol: ethyl acetate. Make fresh. 2.2. Isoelectric Focusing of Samples or First Dimension
1. Rehydration buffer: 8 M urea, 2 M thiourea, 2% CHAPS, 0.2% biolytes, 50 mM dithiothreitol (DTT), bromophenol blue dissolved in deionized water. Make fresh.
2.3. SecondDimension Electrophoresis
1. DTT equilibration buffer (pH 6.8): 50 mM Tris–HCl, 6 M urea, 1% SDS, 30% glycerol, 0.5% DTT dissolved in deionized water. Make fresh. 2. Iodoacetamide (IA) equilibration buffer (pH 6.8): 50 mM Tris–HCl, 6 M urea, 1% SDS, 30% glycerol, 4.5% IA dissolved in deionized water. Make fresh. 3. Tris–glycine–SDS (TGS; 10×) Running buffer is purchased from Bio-Rad (Hercules, CA). Store at room temperature. 4. Fixing solution: 10% methanol, 7% acetic acid dissolved in deionized water. Store at room temperature. 5. SYPRO ruby stain: Purchased from Bio-Rad, (Hercules, CA). Store at room temperature.
2.4. Oxyblot (Immunochemical Detection)
1. Transfer buffer: 1% running buffer (10×), 10% methanol diluted with deionized water. Store at 4°C. 2. Wash blot/phosphate-buffered saline with Tween (PBST): 0.01% sodium azide and 0.2% Tween 20 dissolved in phosphate-buffered saline (PBS). Store at room temperature. 3. Blocking buffer: 2% bovine serum albumin (BSA) in PBST. Make fresh. 4. Primary antibody solution: Anti-HNE antibody (Alpha Diagnostic, San Antonio, TX) (1:5,000) or antinitrotyrosine antibody (1:2,000) (Sigma-Aldrich, St. Louis, MO), diluted in PBST directly before use. 5. Secondary antibody solution: Antirabbit antibody conjugated to alkaline phosphatase (Sigma Aldrich, St Louis, MO) diluted in PBST (1:4,000) directly before use. 6. Developing solution: SigmaFast tablet [5-bromo-4-chloro3-indolyl phosphate/nitro blue tetrazolium (BCIP/NBT)] (Sigma Aldrich, St Louis, MO) dissolved in 10 mL deionized water. Make fresh.
354
Sultana, Reed, and Butterfield
3. Methods Two-dimensional gel electrophoresis (2DE) of biological samples involves the separation of proteins based on two physiochemical properties of proteins: isoelectric point and size (24; see also Chapters “Two-Dimensional Electrophoresis (2DE): An Overview”, “Solubilization of Proteins in 2DE: An Outline”, “Difficult Proteins”). The first step is isoelectric focusing (IEF) in which the proteins are focused according to their isoelectric point. The second step is sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS–PAGE) in which the proteins are further separated based on their migration rate and provides the two-dimensional gel map or total protein profile of a given sample. Comparisons of such profiles, or maps, help discriminate differential protein expression. This technique also allows for the screening of thousands of proteins at once, thereby providing information about post-translational modifications which may result in changes in total protein charge (i.e., possible shift in the position of the protein spot on the gel). In our laboratory, we invoke the use of a parallel analysis in which the 2DE gel map is coupled with 2D immunochemical detection on western blots of nitrated and HNE-bound proteins coupled with MS analysis to identify particular proteins of interest (3, 9, 10, 25, 26). The redox proteomics method used in our laboratory is outlined in Fig. 1. 3.1. Sample Preparation for Detection of HNEand 3-NT-Modified Proteins
1. Prepare 10% tissue or cell homogenate in homogenization buffer (10 g of tissue per 100 mL homogenization buffer). Take a small amount of homogenized tissue and sonicate for 10 s on ice. 2. Determine the protein concentrations in the sonicated samples by using BCA reagent kit from Pierce (Rockford, IL). 3. To 200 μg of protein, add 100% TCA to the samples to get a final concentration of 7.5% TCA to precipitate the protein. Incubate the samples on ice for 10 min. 4. Centrifuge the samples at 10,000 × g for 5 min at 4°C. 5. Decant the supernatant and wash the pellet four times with ice-cold ethanol: ethyl acetate (50:50) mixture (see Note 1). 6. All the washing steps should be carried out at 10,000 × g for 5 min at 4°C. 7. Dry the pellet (see Note 2). 8. After the last washing step resuspend the pellet in 200 μL rehydration buffer, and vortex at room temperature for 1 h. 9. Sonicate the samples for 10 s before loading onto IEF tray (Bio-Rad, Hercules, CA).
Detection of 4-Hydroxy-2-Nonenal- and 3-Nitrotyrosine-Modified Proteins
355
Sample (Protein mixture) X μL sample (150μg protein) + 4X μL 2N HCl Incubate 20 min at room temperature Add TCA (30% final concentration)
Incubate on ice for 10 min Centrifuge at 10,000g for 5min Wash the pellet with ethanol: ethylacetate (1:1) 3 times Add Rehydration Buffer IEF (First Dimension) 2DE
2DE Gel
2D Blot
Image Analysis (PDquest) Protein Spot Excision In gel trypsin digestion Mass Spectrometry Protein database searching for identification of HNE-bound and nitrated proteins
Fig. 1. Outline of redox proteomics workflow for detection of HNE-bound and nitrated proteins.
3.2. Isoelectrofocusing of Samples or First Dimension
1. IEF is performed with a Bio-Rad (Hercules, CA, USA) Protein IEF cell apparatus using 110-mm, pH 3–10 immobilized pH gradient (IPG) strips. 2. Transfer 180 μL of the samples into the bottom of the well in the IEF tray carefully by using a micropipette. Avoid bubbles. 3. Lift IPG strips (Bio-Rad, Hercules, CA) with forceps and remove the plastic protectant from the IPG strips. Place IPG strips on top of the sample with gel side facing down. Make sure
356
Sultana, Reed, and Butterfield
that the positive (+) end of strip is placed toward the positive (+) end of the IEF tray. This is essential for proper focusing. 4. Cover the IEF tray with plastic Bio-Rad lid (Hercules, CA), and place the IEF tray in the Protean IEF cell (Bio-Rad, Hercules, CA), and close the lid of the cell. IPG strips are actively rehydrated at 50 V (20°C) overnight. 5. After 1 h, open the lid of IEF cell (without turning off the instrument) and remove the IEF tray from the machine. Add 2 mL of mineral oil to each well to cover the IPG strips (see Note 3). 6. Place the IEF tray into the IEF apparatus. Close the lid and resume active rehydration for 16–18 h. 7. After 16–18 h of rehydration, wet paper wicks (Bio-Rad, Hercules, CA) with 8 μL of nanopure water and lift the IEF strip using forceps. Place a paper wick on both the electrodes (see Note 4). 8. Place the IEF tray into the apparatus with correct electrode connection and carry out IEF at 20°C as follows: 300 V for 2 h linear gradient; 500 V 2 h linear gradient; 1,000 V 2 h linear gradient; 8,000 V 8 h linear gradient; 8,000 V 10 h rapid gradient. 9. After completion of IEF, transfer IPG strips to an equilibration tray (Bio-Rad, Hercules, CA) and either process directly for second dimension or stored at −80°C until use (see Note 5). 3.3. SecondDimensional Electrophoresis
1. Heat overlay agarose solution (Bio-Rad, Hercules, CA, USA) (see Note 6). 2. Remove the IPG strip from the −80°C freezer and thaw at room temperature (IPG strips change color from milky white to clear upon thawing). 3. Incubate the IPG strips in 4 mL DTT equilibration buffer in a disposable equilibration tray with lid (Bio-Rad, Hercules, CA, USA) with the gel side facing up for 10 min at room temperature. Keep the equilibration tray in dark (see Note 7). 4. While waiting for equilibration, prepare 2DE gel (Bio-Rad, Hercules, CA) by rinsing the gels with DI water (inside and outside). Remove white plastic strip from the bottom of the 2DE gels. 5. Keep the gels on a stand and remove excess water using Kimwipes. 6. Prepare 1× running buffer (Bio-Rad, Hercules, CA) by diluting 100 mL of 10× TGS running buffer with 900 mL of deionized water in a graduated cylinder. 7. After 10 min transfer the IPG strips into a new well in the equilibration tray and add 4 mL of IA equilibration buffer, again with the gel side facing up for another 10 min. Carry out the equilibration in the dark (see Note 8).
Detection of 4-Hydroxy-2-Nonenal- and 3-Nitrotyrosine-Modified Proteins
357
8. After 10 min incubation in IA solution, rinse the IPG strips with 1× running buffer to remove excess equilibration buffer. 9. Place the IPG strips with gel side facing up into criterion gels (Bio-Rad, Hercules, CA, USA). The strip next to the molecular weight marker well and IPG strip should read “3–10,” when placed in the correct orientation. Do not push the IPG strip into the well. 10. Load unstained molecular weight marker (Bio-Rad, Hercules, CA) into the standard well adjacent to the IPG strip well to stain the gels and load precision stained molecular weight markers (Bio-Rad, Hercules, CA) on the gel that will be used for western blotting. 11. Add warm overlay agarose solution into the wells of Criterion gels, avoiding bubbles and then slowly push the IPG strip on either end until contact is established between the gel and IPG strip. (Strip must be in parallel contact with the gel). 12. Wait 10 min for agarose (Bio-Rad, Hercules, CA) to solidify. Meanwhile add running buffer into the tank up to the fill level marked on the gel unit. Then place the gels in the tank and fill the upper tank with running buffer. 13. Connect the power supply with the correct connections (+ve to +ve and −ve to −ve). 14. Run gels at 200 V for 65 min at room temperature, until the dye front (bromophenol blue) exits the gel into the lower tank. 15. Disconnect the power supply (Bio-Rad, Hercules, CA) and disassemble the 2DE apparatus. The gels plates are broken open to remove the gels and one corner cut from the gel to allow its orientation to be tracked. 3.4. Protein Staining
1. The gels are fixed in 50 mL fixative solution at room temperature for 60 min with gentle agitation. 2. Remove fixative solution and add 50 mL of SYPRO Ruby gel stain (Bio-Rad, Hercules, CA, USA). Incubate overnight at room temperature on a rocking platform.
3.5. Immunochemical Detection of HNE or 3-NT-Modified Proteins
1. Gels are transferred to nitrocellulose membranes (Bio-Rad, Hercules, CA) for immunochemical detection of HNE and 3-NT. 2. The samples that have been separated by SDS–PAGE are transferred to nitrocellulose membranes using a semidry transfer unit (Bio-Rad, Hercules, CA). The gels, nitrocellulose membrane, and filter papers (Bio-Rad, Hercules, CA) are soaked in cold transfer buffer for 10 min. 3. A setup of transfer is prepared in the following order: place one soaked filter paper on the transfer unit platform, followed
358
Sultana, Reed, and Butterfield
by the nitrocellulose membrane, gel, and one more filter paper. Roll a glass rod after the gel is placed onto the nitrocellulose membrane to remove bubbles that are trapped between the nitrocellulose membrane, gels, and filter paper. Once the sandwich is ready, roll the glass rod again to ensure that no bubbles are trapped in the transfer sandwich. 4. Connect power supply with correct connection (+ve to +ve and −ve to −ve). It is critically important to ensure correct orientation of the power supply or the proteins will move from the gel into the filter paper instead of the nitrocellulose membrane. 5. Close the lid of the transfer unit and activate power supply. Transfers are performed at 15 V for 2 h at room temperature. The semitransfer method of proteins not only saves time, but it requires less transfer buffer. 6. Once the transfer is complete, the transfer unit is disconnected, the unit is carefully disassembled, and the nitrocellulose membrane is taken out. Since molecular weight markers are used, we do not need to cut the ends of the nitrocellulose membrane for orientation. The gel and filter papers can then be discarded. The prestained molecular weight markers should be clearly visible on the membrane. 7. The nitrocellulose is then incubated in 50 mL of blocking buffer for 1 h at room temperature on a rocking platform (see Note 9). 8. To the blocking buffer 1:5,000 of anti-HNE antibody (Alpha Diagnostics, San Antonio, TX) or 1:2,000 anti-NT (SigmaAldrich, St. Louis, MO) is added and incubated for 1 h at room temperature on a rocking platform. 9. The primary antibody is then removed and the membrane washed three times for 5 min with 50 mL wash blot each time. 10. The secondary antibody (Anti-Rabbit alkaline phosphataseconjugated) is freshly prepared for each experiment as 1:3,000 in wash blot and the membrane is incubated on a rocking platform for 1 h. 11. The secondary antibody is discarded and the membrane washed three times for 5 min each with wash blot. 12. After final wash, the blot is developed using a Sigma Fast tablet. (Sigma, St. Louis, MO, USA) (see Note 10). 13. For optimal color intensity 30–40 min is required. After color development, the developer is drained; the membrane is washed with tap water and dried between Kim-Wipes. An example of the results produced is shown in Fig. 2.
Detection of 4-Hydroxy-2-Nonenal- and 3-Nitrotyrosine-Modified Proteins
359
Fig. 2. A representative (a). Sypro Ruby-stained 2DE gel and (b) 2D-Western blot. A box is drawn around the area that shows immunopositive spots for HNE.
3.6. Image Analysis
1. SYPRO Ruby stained gels are scanned using a UV transilluminator (λex = 470 nm, λem = 618 nm, Molecular Dynamics, Sunnyvale, CA, USA). 2. The 2D Western blots can be scanned with Adobe Photoshop on a Microtek Scanmaker 4,900. 3. Western blots and 2DE gel maps are matched using PDQuest image analysis software (Bio-Rad, Hercules, CA, USA) to determine the levels of specific HNE-bound proteins or nitrated proteins. The immunoreactivity of the western blot is normalized to the actual protein content as measured by the intensity of a protein stain such as SYPRO ruby (Bio-Rad, Hercules, CA, USA). 4. Protein spots showing a significant increase in HNE or 3-NT levels are excised from the gel and digested in-gel with trypsin for submission for mass spectrometry for mass analysis (see Chapter “Tryptic Digestion of In-Gel Proteins for Mass Spectrometry Analysis”). Finally, proteins are identified following interrogation of proteins detected databases.
360
Sultana, Reed, and Butterfield
4. Notes 1. Proper washing of the pellet is crucial to reduce the background signal on the blot. In addition, proper washing removes lipids. 2. Make sure to remove the ethanol and ethyl acetate solution completely before adding the rehydration buffer. If necessary let pellets dry 5–10 min, before adding the rehydration buffer, as excess of this organic mixture could lead to precipitation of proteins. 3. Addition of oil will prevent sample evaporation. 4. Paper wicks will prevent IPG strips from burning. 5. Cover the equilibration tray with plastic wrap, followed by wrapping of the equilibration tray with aluminum foil to prevent moisture formation. 6. Commercially available agarose solution comes with bromophenol blue dye that can be used as a tracking dye to monitor the electrophoresis. 7. DTT breaks disulfide bonds. 8. IA stabilizes the disulfide bond once it is cleaved by DTT. 9. Washing is not required between blocking and primary antibody treatment. 10. One tablet of Sigma Fast is dissolved in 10 mL of deionized water.
Acknowledgments This work was supported in part by grants from NIH [AG-10,836; AG-05119].
References 1. Butterfield, D.A. (2002) Amyloid beta-peptide (1–42)-induced oxidative stress and neurotoxicity: implications for neurodegeneration in Alzheimer’s disease brain. A review. Free Radic Res 36, 1307–1313 2. Butterfield, D.A., and Lauderback, C.M. (2002) Lipid peroxidation and protein oxidation in Alzheimer’s disease brain: potential causes and consequences involving amyloid beta-peptide-associated free radical oxidative stress. Free Radic Biol Med 32, 1050–1060
3. Castegna, A., Thongboonkerd, V., Klein, J.B., Lynn, B., Markesbery, W.R., and Butterfield, D.A. (2003) Proteomic identification of nitrated proteins in Alzheimer’s disease brain. J Neurochem 85, 1394–1401 4. Lovell, M.A., Gabbita, S.P., and Markesbery, W.R. (1999) Increased DNA oxidation and decreased levels of repair products in Alzheimer’s disease ventricular CSF. J Neurochem 72, 771–77 5. Lovell, M.A., and Markesbery, W.R. (2001) Ratio of 8-hydroxyguanine in intact DNA to
Detection of 4-Hydroxy-2-Nonenal- and 3-Nitrotyrosine-Modified Proteins free 8-hydroxyguanine is increased in Alzheimer disease ventricular cerebrospinal fluid. Arch Neurol 58, 392–396 6. Markesbery, W.R., Kryscio, R.J., Lovell, M.A., and Morrow, J.D., (2005) Lipid peroxidation is an early event in the brain in amnesic mild cognitive impairment. Ann Neurol 58,730–735 7. Smith, M.A., Harris, P.L.R., Sayre, L.M., Beckman, J.S., and Perry, G. (1997) Widespread peroxynitrite-mediated damage in Alzheimer’s disease. J Neurosci 17, 2653–2657 8. Smith, M.A., Harris, P.L.R., Taneda, S., Kutty, R.K., Sayre, L.M. , Monnier, V.M. , et al. (1994) Advanced Maillard reaction end products, free radicals, and protein oxidation in Alzheimer’s disease. Ann N Y Acad Sci 738, 447–454 9. Sultana, R., Perluigi, M., and Butterfield, D.A. (2006) Redox proteomics identification of oxidatively modified proteins in Alzheimer’s disease brain and in vivo and in vitro models of AD centered around Abeta(1–42). J Chromatogr B Analyt Technol Biomed Life Sci 833, 3–11 10. Sultana, R., Poon, H.F., Cai, J., Pierce, W.M., Merchant, M., Klein, J.B., et al. (2006) Identification of nitrated proteins in Alzheimer’s disease brain using a redox proteomics approach. Neurobiol Dis 22, 76–87 11. Dalle-Donne, I., Scaloni, A., and Butterfield, D.A. (2006) Redox Proteomics: From Protein Modifications to Cellular Dysfunction and Diseases. John Wiley and Sons, Hoboken, NJ 12. Koeck, T., Levison, B., Hazen, S.L., Crabb, J.W., Stuehr, D.J., and Aulak, K.S. (2004) Tyrosine nitration impairs mammalian aldolase A activity. Mol Cell Proteomics 3, 548–557 13. Lafon-Cazal, M., Culcasi, M., Gaven, F., Pietri, S., and Bockaert, J. (1993) Nitric oxide, superoxide and peroxynitrite: putative mediators of NMDA-induced cell death in cerebellar granule cells. Neuropharmacology 32, 1259–1266 14. Aulak, K.S., Koeck, T., Crabb, J.W., and Stuehr, D.J. (2004) Dynamics of protein nitration in cells and mitochondria. Am J Physiol 286, H30–H38 15. Beal, M.F. (1980) Mitochondrial dysfunction in neurodegenerative diseases. Biochim Biophys Acta 1366, 211–233 16. Esterbauer, H., Schaur, R.J., and Zollner, H. (1991) Chemistry and biochemistry of
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
361
4-hydroxynonenal, malonaldehyde and related aldehydes. Free Radic Biol Med 11, 81–128 Choudhary, S., Zhang, W., Zhou, F., Campbell, G.A., Chan, L.L., Thompson, E.B., et al. (2002) Cellular lipid peroxidation end-products induce apoptosis in human lens epithelial cells. Free Radic Biol Med 32, 360–369 Hashimoto, M., Sibata, T., Wasada, H., Toyokuni, S., and Uchida, K. (2003) Structural basis of protein-bound endogenous aldehydes. Chemical and immunochemical characterizations of configurational isomers of a 4-hydroxy-2-nonenal-histidine adduct. J Biol Chem 278, 5044–5051 Lovell, M.A., Xie, C., and Markesbery, W.R. (2001) Acrolein is increased in Alzheimer’s disease brain and is toxic to primary hippocampal cultures. Neurobiol Aging 22,187–194 Markesbery, W.R., and Lovell, M.A. (1998) Four-hydroxynonenal, a product of lipid peroxidation, is increased in the brain in Alzheimer’s disease. Neurobiol Aging 19, 33–36 Tamagno, E., Robino, G., Obbili, A., Bardini, P., Aragno, M., Parola, M., et al. (2003) H2O2 and 4-hydroxynonenal mediate amyloid beta-induced neuronal apoptosis by activating JNKs and p38MAPK. Exp Neurol 180,144–155 Uchida, K. (2003) 4-Hydroxy-2-nonenal: a product and mediator of oxidative stress. Prog Lipid Res 42, 318–343 Perluigi, M., Poon, H., Hensley, K., Pierce, W.M., Klein, J.B., Calabrese, V., et al. (2005) Proteomic analysis of 4-hydroxy-2nonenal-modified proteins in G93A-SOD1 transgenic mice–a model of familial amyotrophic lateral sclerosis. Free Radic Biol Med 38, 960–968 Rabilloud, T. (2002) Two-dimensional gel electrophoresis in proteomics: old, old fashioned, but it still climbs up the mountains. Proteomics 2, 3–10 Butterfield, D.A., Gnjec, A., Poon, H.F., Castegna, A., Pierce, W.M., Klein, J.B., et al. 2006. Redox proteomics identification of oxidatively modified brain proteins in inherited Alzheimer’s disease: an initial assessment. J Alzheimers Dis 10, 391–397 Butterfield, D.A., Perluigi, M., and Sultana, R. (2006) Oxidative stress in Alzheimer’s disease brain: new insights from redox proteomics. Eur J Pharmacol 545, 39–50
Chapter 23 Proteomic Detection of Oxidized and Reduced Thiol Proteins in Cultured Cells Sarah L. Cuddihy, James W. Baty, Kristin K. Brown, Christine C. Winterbourn, and Mark B. Hampton Summary The oxidation and reduction of cysteine residues is emerging as an important post-translational control of protein function. We describe a method for fluorescent labelling of either reduced or oxidized thiols in combination with two-dimensional sodium dodecyl sulphate polyacrylamide gel electrophoresis (2DE) to detect changes in the redox proteome of cultured cells. Reduced thiols are labelled with the fluorescent compound 5-iodoacetamidofluorescein. To monitor oxidized thiols, the reduced thiols are first blocked with N-ethyl-maleimide, then the oxidized thiols reduced with dithiothreitol and labelled with 5-iodoacetamidofluorescein. The method is illustrated by treating Jurkat T-lymphoma cells with hydrogen peroxide and monitoring increased labelling of oxidized thiol proteins. A decrease in labelling can also be detected, and this is attributed to the formation of higher oxidation states of cysteine that are not reduced by dithiothreitol. Key words: Thiol, Cysteine, Oxidation, Hydrogen peroxide, Jurkat T-lymphoma cells, 5-Iodoacetamidofluorescein, Maleimide, Redox proteomics.
1. Introduction Cells sense alterations in the redox status of their environment through key cysteine residues on signalling proteins. Cysteine is an ideal redox sensor. The sulphydryl group of cysteine (–SH) has a pKa of 8.3 and is predominantly protonated at physiological pH. However, the presence of positively charged residues in the local protein environment can stabilize the thiolate anion (–S−), providing a mechanism for increasing the reactivity of selected
David Sheehan and Raymond Tyther (eds.), Methods in Molecular Biology, Two-Dimensional Electrophoresis Protocols, vol. 519 © Humana Press, a part of Springer Science + Business Media, LLC 2009 DOI: 10.1007/978-1-59745-281-6_23
363
364
Cuddihy et al.
Fig. 1. Protein thiols (–SH) can be blocked with either N-ethylmaleimide (NEM), or iodoacetyl derivatives (R=H for iodoacetamide, R=fluorescein for IAF; see Note 6).
cysteine residues. The thiolate anion is a strong nucleophile that can be oxidized to sulphenic (–SOH), sulphinic (–SO2H), and sulphonic (–SO3H) acids; S-nitrosothiols (–SNO); and disulphides (R–S–S–R; see also Chapter “Diagonal Electrophoresis for Detection of Protein Disulphide Bridges”). Oxidation, either directly or via exchange reactions, has a strong influence on protein structure, metal chelation, and catalytic activity. The dynamic oxidation and reduction of cysteine residues in key cellular proteins (e.g. transcription factors, phosphatases, structural proteins, metabolic enzymes) in response to an increase in environmental or endogenous oxidants is termed redox signalling. Methods that are used to detect changes in the redox state of cellular thiol proteins make use of iodoacetyl and maleimide groups that react with thiols (Fig. 1). Derivatives of these compounds with additional charge (1), mass (2), fluorescence (3), radioactivity (4), or biotin derivatives (5) enable labelling and monitoring of changes in redox state. The majority of cellular thiols are present in the reduced form, so to optimize the detection of oxidized thiol proteins Gitler and colleagues (4) developed a technique in which reduced thiols are irreversibly blocked with an alkylating agent, before the oxidized thiols are reduced and labelled. In this chapter we describe a technique that utilizes 5-iodoacetamidofluorescein (IAF) to label thiol proteins (Fig. 2). The labelling of reduced thiols is described, but the major focus is detection of oxidized thiol proteins. Cells are lysed in the presence of N-ethylmaleimide (NEM) to rapidly and irreversibly block all reduced thiols, then oxidized thiols are reduced with dithiothreitol (DTT) and labelled with IAF (6, 7). The labelled proteins are focused on an isoelectric
Proteomic Detection of Oxidized and Reduced Thiol Proteins in Cultured Cells
365
Fig. 2. Reduced or oxidized thiol proteins are labelled with a fluorescent iodoacetamide derivative 5-iodoacetamidofluorescein (IAF) by distinct protocols. To detect oxidized thiols the reduced thiols are blocked with NEM, the sample is treated with DTT and then the newly reduced thiols labelled with IAF. In comparison, for reduced thiols all sites are immediately labelled with IAF (6, 7).
focusing (IEF) strip and separated on a sodium dodecyl sulphate polyacrylamide gel electrophoresis (SDS–PAGE) gel. We have used this two-dimensional electrophoresis (2DE) method to identify proteins that are sensitive to exogenous hydrogen peroxide and to disruption of the cellular thioredoxin system (7). The treatment of Jurkat T-lymphoma cells with a single bolus of hydrogen peroxide is used to illustrate the method in the current chapter. The technique is very sensitive, and many of the proteins susceptible to oxidation are not sufficiently abundant to enable identification. However, monitoring the patterns of oxidation can provide useful information. One example is the comparison of different exogenous oxidants, and the method will also detect a disruption of redox homeostasis as a result of endogenous oxidative stress or during activation of signal transduction pathways.
366
Cuddihy et al.
2. Materials 2.1. Cell Culture and Blocking
1. RPMI 1640 media (Invitrogen, Carlsbad, CA) supplemented with 10% foetal bovine serum (FBS) (Invitrogen). 2. Jurkat T-lymphoma cells (obtained from American Type Culture Collection, Rockville, MD, USA) maintained in RPMI1640 supplemented with FBS, 2 mM glutamine, 100 U/mL penicillin, and 100 μg/mL streptomycin at 37°C in a humidified atmosphere with 5% CO2. Passage cells every 1–3 days when they reach a density of approximately 1 × 106 cells/ mL. 3. Hydrogen peroxide (H2O2): 100 mM stock in MilliQ-water. The concentration is determined using the extinction coefficient of H2O2 (ε240 = 43.6 M−1/cm). 4. Phosphate-buffered saline (PBS): 140 mM NaCl, 13 mM KCl in 10 mM sodium phosphate buffer, pH 7.4. 5. Blocking buffer: 100 mM N-ethylmaleimide (NEM) (Sigma, St Louis, MO) (12.5 mg/mL) in extract buffer containing 50 mM NaCl, 1 mM EDTA, 1 mM EGTA, and protease inhibitors (Complete™, Roche Diagnostics, Mannheim Germany) in 40 mM HEPES buffer, pH 7.4. 6. Lysis reagent: 25% CHAPS stock in MilliQ-water.
2.2. IAF Labelling and Sample Preparation
1. Reducing agent: 75 mM dithiothreitol (DTT, Sigma) in MilliQ-water and store at −20°C. 2. Prepare 10 mM IAF (5 mg/mL) (Invitrogen) in DMSO and store at −20°C (see Note 1). 3. Protein is assayed using the Dc Protein Assay Kit (cat# 500– 0112, Bio-Rad, Hercules, CA) (see Note 2). 4. Desalting columns (cat# 732–6221; Bio-Rad). 5. Sample/rehydration buffer: 7 M urea (16.8 g), 2 M thiourea (6.1 g), 10 mM DTT (0.061 g, see Note 3), 4% CHAPS (1.6 g), and 0.2% Bio-Lyte 3–10 ampholytes (Bio-Rad) (200 μL of a 40% stock) in 40 mL. Add a few crystals of bromophenol blue. Aliquot and store at −20°C.
2.3. Isoelectric Focusing and SDS–Polyacrylamide Gel Electrophoresis
1. IEF strip (17cm pH 3–10 Readystrip™), electrode wicks, focusing trays, and the Protean IEF Cell (Bio-Rad). 2. Equilibration solution: 6 M Urea (180 g), 20% Glycerol (100 mL), 2% SDS (10 g), and a few crystals of bromophenol blue in 375 mM Tris–HCl, pH 8.8 (100 mL of a 1.875 M stock, which is 22.7 g Tris-base in 100 mL MilliQ-water, pH to 8.8 with HCl) in 500 mL total volume.
Proteomic Detection of Oxidized and Reduced Thiol Proteins in Cultured Cells
367
3. SDS–PAGE running buffer (10×): 0.248 M Tris base (30 g), 1.92 M glycine (144 g), and 1% SDS (10 g), made to 1 L with MilliQ-water (see Note 4). Make to 1× with MilliQ-water prior to use. 4. 0.5% Agarose solution: 2.5 g agarose in 500 mL of 1× SDS– PAGE running buffer with a few crystals of bromophenol blue. 5. 2nd-dimension SDS–PAGE can be any large-format system – we use the Protean II xi System (Bio-Rad) cast with a 12% gel composed of (per gel): 15 mL 40% acrylamide (37.5:1, BioRad, this is a neurotoxin when unpolymerized, extreme care should be taken), 12.5 mL 2 M Tris–HCl, pH 8.8, 21.75 mL MilliQ-water, 250 μL 20% SDS, 250 μL 10% ammonium persulphate, 25 μL tetramethylethylenediamine (TEMED). 2.4. Detection and Quantification
1. Silver stain solutions: Solution 1: 0.2 g of anhydrous sodium thiosulphate in 1 L of MilliQ-water. Solution 2: 2 g silver nitrate and 750 μL of formaldehyde to 1 L of MilliQ-water, freshly prepared. Solution 3: 60 g sodium carbonate, 20 mL silver stain solution 1, 500 μL formaldehyde, and 980 mL of MilliQ-water. Solution 4: 20 g of sodium EDTA in 1 L of MilliQ-water.
3. Methods 3.1. Cell Culture and Harvesting
1. Cell culture methods will be specific to the cell lines being used (we used Jurkat T-lymphoma cells). 2. Prior to exposure to a stimulus, count and resuspend the cells in fresh medium at an appropriate density, then equilibrate them in the incubator for 1 h (see Note 5). For Jurkat cells this is 1 × 106 cells/mL media. Ensure an untreated control is included. 3. Treat the cells with the stimulus. We used 200 μM hydrogen peroxide for 10 min. 4. After treatment, pellet cells at 1,000 × g for 4 min, resuspend in 1 mL cold PBS and re-pellet at 10,000 × g for 1 min.
3.2. IAF Labelling and Sample Preparation of Reduced Thiols
1. After pelleting, resuspend cells in extract buffer (pH 7.4) containing 1% CHAPS and 200 μM IAF to a density of 5 × 107 cells/mL and incubate in the dark at room temperature for 30 min (see Notes 1 and 6). 2. Separate insoluble material from the extract by centrifugation at 16,000 × g for 5 min.
368
Cuddihy et al.
3. Remove excess IAF by passing 75 μL of labelled protein extract through two sequential spin columns that have each been equilibrated with 3 × 400 μL sample/rehydration solution (see Note 7). 3.3. IAF Labelling and Sample Preparation of Oxidized Thiols
1. As described in Subheading 3.1, cells are pelleted, washed with PBS, and re-pelleted. Resuspend the pellet in NEM blocking buffer to a density of 5 × 107 cells/mL and incubate at room temperature for 15 min. (Lyse cells in 1% CHAPS for a further 15 min and pellet insoluble material at 16,000 × g for 5 min (see Note 9). 2. After lysis, remove excess NEM by passing 75 μL of extract through a spin column equilibrated with 3 × 400 μL of extract buffer (see Notes 7 and 10). 3. Reduce oxidized thiols in the filtrate by incubating with 1 mM DTT for 10 min at room temperature. Protein concentration should be assayed at this point (we use the Dc Protein Assay Kit; Bio-Rad). 4. Add IAF to a final concentration of 200 μM and incubate the extract in the dark at room temperature for 10 min. Subsequent steps should be carried out with minimal exposure to light (see Notes 1 and 6). 5. Remove excess IAF is by passing 75 μL of labelled protein through two sequential spin columns that have each been previously equilibrated three times with 400 μL sample/rehydration solution (see Note 7).
3.4. Isoelectric Focusing and 2D SDS– Polyacrylamide Gel Electrophoresis
1. IEF strips are rehydrated with the protein sample in a focusing tray by pipetting the labelled sample in 300 μL along one edge of a channel (maximum of 300 μg of protein for a 17 - cm strip). Remove the plastic backing from the strip using forceps and lay the strip gel-side down on the sample, so that it is evenly distributed along the strip (see Note 11). Cover the strip with 2–3 mL of mineral oil (Bio-Rad) to prevent dehydration by evaporation. Wrap the tray with aluminium foil and place on a level surface overnight. 2. After rehydration, use forceps to remove the loaded strip and drain the oil by letting it run off the strip and gently blotting the plastic backing and ends of the strip on a paper towel. Use forceps to wet four electrode wicks per strip and place them directly on top of the electrode wires in the focusing tray (two at each end). Lay the strip gel side down in the focusing channel and ensure good contact between the gel and the electrode by depressing the strip lightly. Overlay the strip with 2–3 mL mineral oil. Place lid on tray and insert tray into the peltier
Proteomic Detection of Oxidized and Reduced Thiol Proteins in Cultured Cells
369
platform, ensuring that the electrodes of the focusing tray contact the colour-coded electrodes of the Protean IEF cell. 3. Run the IEF program (in dark) as per the manufacturers’ instructions. For 17 cm strips (pH 3–10) we used 300 V for 1 h; 1,000 V for 1 h; 3,000 V for 1 h; 6,000 V for 75,000 Vh (~12.5 h); 500 V hold (see Note 12). 4. Prepare large-format SDS–PAGE gels as described in Subheading 2.3. Pour the gel to within 1 cm of the top. Prepare running buffer (1×) (2 L for Protean xi). 5. After focusing, equilibrate strips for SDS–PAGE by placing them gel side up in a focusing tray and covering them with 2–3 mL of equilibration buffer. Keep the tray wrapped in aluminium foil and incubate on a shaker with fast rotation for 10 min. Transfer strip to a new channel and repeat. 6. Use forceps to remove the strip from the equilibration tray. Use clean scissors to remove the ends of the strip so that it fits neatly on top of the gel (see Note 13). 7. Using forceps gently slide the strip into position on top of the gel, avoiding formation of bubbles between the strip and the gel (see Note 14). Once the strip is in position, pipette in 2–3 mL of melted agarose to hold it in place. 8. Allow the agarose to set and then assemble the running rig (as per manufacturers’ instructions). For the Protean xi, connect the cooling core to a cold water tap, and ensure there is no leakage of buffer from the upper tank. Cover running rig with a cardboard box to limit quenching of the fluorescent tags and then run slowly overnight (10–15 mA/gel) until the dye front has run off the bottom of the gel. 3.5. Detection and Quantitation
1. At the completion of electrophoresis, scan both control and treated gels for IAF fluorescence using a fluorescent imager with an excitation wavelength of 488nm and an emission/ detection wavelength of 530 nm. We use a Bio-Rad Molecular Imager® FX (Bio-Rad Laboratories). Figure 3 illustrates IAF-labelled oxidized thiol patterns comparing control Jurkat T lymphocytes with cells exposed to 200 μM hydrogen peroxide for 10 min. Enlarged areas of increased (Fig. 4a) and decreased (Fig. 4b) spot intensities are shown. 2. Ensure there are no saturated pixels in the scan as this will prevent accurate quantification of spot intensity (see also Chapter “Troubleshooting Image Analysis in 2DE”). 3. Gel images of resolved IAF-labelled thiol proteins can be analyzed with 2DE gel analysis software (we use PDQuest™ from Bio-Rad). The analysis software is used to generate a list of spots with at least a twofold increase or decrease in fluorescence in response to the stimulus. Each spot nominated by
370
Cuddihy et al.
Fig. 3. Jurkat T-lymphocytes were treated for 10 min with 200 μM hydrogen peroxide, then prepared for labelling of oxidized thiols as described, and separated by 2DE. A total of 52 spots consistently changed between untreated cells (left) and treated cells (right), with 28 spots increasing in intensity (box, enlarged in Fig.4a) and 24 spots decreasing (oval, enlarged in Fig. 4b) (7).
Fig. 4. (a) IAF-labelled protein spots can increase in intensity indicating increased oxidation. MALDI-TOF MS identified this series of spots as GAPDH, a highly sensitive protein linking redox balance with energy metabolism. (b) Protein spots can also decrease in intensity indicating either over-oxidation of the protein, or modifications that change either pI or molecular weight. MALDI-TOF MS identified these spots as members of the peroxiredoxin family. The peroxidative cysteine of the peroxiredoxins is oxidized to a sulphinic acid by hydrogen peroxide.
the software should be visually inspected to determine if the change was authentic or an artefact of spot detection. An estimation of spot molecular weight (MW) and isoelectric point (pI) can be made by comparison with the positions of proteins with known MW and pI (see Note 15).
Proteomic Detection of Oxidized and Reduced Thiol Proteins in Cultured Cells
371
4. Following visualization of IAF fluorescence, visualize total protein by silver staining (see also Chapter “Silver Staining of Proteins in 2DE Gels”). Fix gels with a 50% ethanol, 10% acetic acid solution for 30 min and then wash with a 5% ethanol, 1% acetic acid solution for 15 min. Wash gels three times with MilliQ-water for 5 min each before addition of solution 1 for 2 min. Thoroughly wash gels three times with MilliQ-water for 30 s and then incubate with solution 2 for 30 min. Wash gels twice with MilliQ-water for 20 s each before addition of solution 3. Quickly wash off Solution 3 with MilliQ-water as soon as colour development reaches the desired intensity. Stop colour development by addition of solution 4. Acquire gel images using an appropriate imager (we use the Bio-Rad Fluor-S™ MultiImager). Figures 5 and 6 compare IAF and silver stain protein patterns. 5. Protein spots of interest are cut out of the gel for identification by mass spectrometry (see Notes 16 and 17; Chapter “Tryptic Digestion of In-Gel Proteins for Mass Spectrometry Analysis”). This can be done with a fluorescent spot cutter or manual excision by printing out an actual size image of the gel, placing a
Fig. 5. Specific labelling of oxidized thiol proteins is highly selective and sensitive. In untreated Jurkat T lymphocytes there is a 40,000-fold dynamic range in fluorescence intensity between the most intense and least intense protein spots. While the IAF and silver stain patterns share many similarities there are also regions where IAF labelling is more sensitive than silver stain (box; see Fig.6 for enlargement).
372
Cuddihy et al.
Fig. 6. IAF fluorescence and silver staining show many matching spots (small arrow); however; there are also some spots that are visible by IAF but are at such low abundance that they are not detectable by silver stain (large arrow). Silver staining of 2DE gels post-IAF scanning also ensures that observed spot intensity changes are not due to changes in protein expression.
glass plate between the image and the gel, which will not have any visible spots, and using the image as a “map” to excise the spots of interest. Rescan the gel to check spot excision.
4. Notes 1. IAF is extremely light sensitive as the iodoacetamide group is highly susceptible to iodine loss, leading to lower than expected protein labelling and higher non-specific staining. The 10 mM stock must be kept in the dark at −20°C and is stable for approximately two weeks. When de-halogenated the IAF forms a channel in 2DE gels, causing distortion and streaking of the spots and appearing as a large “blotch” at approximately 10 kDa. This can be prevented by using two desalting spin columns to remove excess IAF after protein labelling. Note that this is not typically necessary for onedimensional SDS–PAGE (1DE) where excess IAF runs off with the dye front. 2. Protein concentration should be assayed after NEM blocking and DTT reduction, but prior to addition of IAF. The Dc Protein Assay kit is used as DTT does not interfere with the colour development in this particular assay. 3. For oxidized thiols it is not necessary to include DTT in the sample/rehydration buffer, as all thiols have been blocked with either NEM or IAF. Do not heat the sample/rehydration buffer past 37°C as this increases the likelihood of protein carbamylation by urea.
Proteomic Detection of Oxidized and Reduced Thiol Proteins in Cultured Cells
373
4. Never adjust the pH of the running buffer (10X) as this affects ion migration which affects how the protein runs in the gel. Buffer must be made to 1X prior to use. 5. It is important to let the cells re-equilibrate in the new medium. For Jurkat cells the stress of pelleting and resuspending in fresh media can induce some temporary redox changes which will dissipate after 1 h. 6. Reduced thiols may be labelled with maleimide or iodoacetyl derivatives. Iodoacetyl derivatives only react with the thiolate anion, and therefore the selectivity of labelling will be dependent on the pKa of individual cysteine residues and the pH of the extract buffer. The efficiency of maleimide labelling is not pH dependent. 7. Micro-spin columns can be purchased pre-packed containing 10 mM Tris–HCl buffer, pH 7.4. This must be exchanged with an appropriate buffer prior to use. It is preferable to use a swing bucket rotor in order to ensure even packing of the bedding. Fixed-angle rotors cause the matrix to pack on an angle and cause the sample to pass unevenly through the column, leading to inefficient removal of the excess IAF. Equilibration involves passing 400 μL of an appropriate buffer through the spin column at 1,000 × g for 1 min, removing the filtrate, and repeating a total of three times. 8. NEM is cell permeable and at high concentrations will block most intracellular thiols in intact cells. However, intracellular thiols involved in multiprotein aggregates or in the vicinity of lipids will not be efficiently blocked. In these cases prolonged NEM blocking prior to lysis can lead to oxidation of a subset of proteins, thereby restricting the blocking step to 15 min. As soon as the cell is lysed with detergent the NEM is able to access all thiols and the remaining cysteines will be blocked. 9. Different cell lines may require higher concentrations of CHAPS or alternate detergents for efficient lysis. 10. NEM must be removed by spin column in order to prevent reaction with DTT. The concentrations of DTT and IAF were empirically determined so that there is sufficient DTT to reduce oxidized protein thiols, but not so much that it will interfere in IAF labelling. Increasing DTT concentrations will result in lower IAF labelling efficiency. 11. Dilute protein sample with hydration buffer to the appropriate volume if necessary. Make sure there are no air bubbles between the sample and the gel of the strip, as this will prevent even loading which can lead to streaking.
374
Cuddihy et al.
12. At the end of focusing the strips can be prepared for immediate SDS–PAGE, or they may be stored at −80°C, drained of oil, gel side up in a tray wrapped in aluminium foil. 13. Approximately 0.5 cm must be cut from each end of the strip to fit the Protean xi gel. This may not be necessary with other systems. Typically the proteins at the very edges of the strips are not focused and run as streaks. 14. If MW markers are to be included use a small rectangle of filter paper and drop the marker solution onto it. This marker can be placed next to the strip on top of the SDS–PAGE gel and fixed with agarose, and will run normally into the gel under current. 15. Separate 2DE gels can be run using labelled protein standards (available commercially from Bio-Rad) for comparative purposes. Standards are hen egg white conalbumin (MW: 76 kDa; pI: 6.0, 6.3, 6.6), bovine serum albumin (MW: 66.2 kDa; pI: 5.4, 5.6), bovine muscle actin (MW: 43 kDa; pI: 5.0, 5.1), rabbit muscle GAPDH (MW: 36 kDa; pI: 8.3, 8.5), bovine carbonic anhydrase (MW: 31 kDa; pI: 5.9, 6.0), soybean trypsin inhibitor (MW: 21.5 kDa; pI: 4.5), equine myoglobin (MW: 17.5 kDa; pI: 7.0). 16. The likelihood of obtaining clean mass spectrometry data from the gels is significantly improved if the gels are allowed to set overnight prior to running the second dimension as this reduces the amount of unpolymerized acrylamide contaminant in the sample. Keratin contamination can be reduced by the use of forceps and gloves for all handling of protein samples, IEF strips, and SDS–PAGE gels. 17. The purity of the excised spot can be improved by labelling the oxidized protein samples with a biotinylated NEM or iodoacetamide, followed by streptavidin purification of labelled proteins. In this way highly abundant unlabelled proteins that may run at the same position as the protein spot of interest will be removed.
Acknowledgments The authors would like to thank Juliet Pullar, Andrew Cox, and Rachel Wilkie for valuable discussions during method development. Financial support has come from the Royal Society Marsden Fund and the Health Research Council of New Zealand. Sarah Cuddihy is supported by an FRST Post-Doctoral Fellowship and Kristin Brown has a TEC Top Achievers Doctoral Scholarship.
Proteomic Detection of Oxidized and Reduced Thiol Proteins in Cultured Cells
375
References 1. Thomas, J. A., Zhao, W., Hendrich, S., and Haddock, P. (1995) Analysis of cells and tissues for S-thiolation of specific proteins. Methods Enzymol. 251, 423–429 2. Guo, Z. Y., Chang, C. C., Lu, X., Chen, J., Li, B. L., and Chang, T. Y. (2005) The disulfide linkage and the free sulfhydryl accessibility of acyl-coenzyme A:cholesterol acyltransferase 1 as studied by using mPEG5000-maleimide. Biochemistry 44, 6537–6546 3. Wu, Y., Kwon, K. S., and Rhee, S. G. (1998) Probing cellular protein targets of H2O2 with fluorescein-conjugated iodoacetamide and antibodies to fluorescein. FEBS Lett. 440, 111–115 4. Gitler, C., Zarmi, B., and Kalef, E. (1997) General method to identify and enrich vicinal thiol
proteins present in intact cells in the oxidized, disulfide state. Anal. Biochem. 252, 48–55 5. Bayer, E.A., Safars, M., and Wilchek, M. (1987) Selective labeling of sulfhydryls and disulfides on blot transfers using avidin-biotin technology: studies on purified proteins and erythrocyte membranes. Anal. Biochem 161, 262–271 6. Baty, J. W., Hampton, M. B., and Winterbourn, C. C. (2002) Detection of oxidant sensitive thiol proteins by fluorescence labeling and two-dimensional electrophoresis. Proteomics 2, 1261–1266 7. Baty, J. W., Hampton, M. B., and Winterbourn, C. C. (2005) Proteomic detection of hydrogen peroxide-sensitive thiol proteins in Jurkat cells. Biochem. J. 389, 785–795
Chapter 24 Detection of Ubiquitination in 2DE Brian McDonagh Summary Ubiquitination involves the tagging of proteins with one (mono-) or more (poly-) ubiquitin molecules. Primarily the role of ubiquitination involves mainly short-lived and regulatory proteins being tagged with a poly-ubiquitin tail, thus introducing a hydrophobic patch that allows the protein to be identified and degraded by the 26S proteasome. Transfer of ubiquitin to the lysine residue of a target protein is a multi-step ATP-dependent process. The functions of ubiquitination have been extended in recent years to all areas of biology, many of them proteasome independent. As a small fraction of any protein may potentially be ubiquitinated, this may explain the wide range and large number of proteins that have been identified as being tagged with ubiquitin in the literature. This chapter outlines a general method for an indication of ubiquitination levels and identification of ubiquitinated proteins by two-dimensional electrophoresis in combination with immunoblotting. Key words: Ubiquitin, Ubiquitination, 26S proteasome, Protein turnover, Anti-ubiquitin.
1. Introduction It is almost 30 years since the discovery of the ubiquitin proteasome system (UPS). Its significance in biology was acknowledged when the 2004 Chemistry Nobel Prize was awarded to Aaron Ciechanover, Avram Hershko, and Irwin Rose, who first established its main features. Ubiquitin is a 76-residue protein with a molecular mass of about 8 kDa that is highly conserved in all eukaryotes. Classically, ubiquitination is a multi-step process in which an ubiquitin molecule is covalently linked to a lysine residue, which can then act as a substrate for the formation of a poly-ubiquitin tail. The poly-ubiquitin tail can then serve as a recognition signal for the 26S proteasome, where the ubiquitin is recycled and the protein is degraded (1). David Sheehan and Raymond Tyther (eds.), Methods in Molecular Biology, Two-Dimensional Electrophoresis Protocols, vol. 519 © Humana Press, a part of Springer Science + Business Media, LLC 2009 DOI: 10.1007/978-1-59745-281-6_24
377
378
McDonagh
In recent years, the functions of ubiquitin have been extended to a regulatory role that affects all areas of biology. Indeed it is now associated with a number of proteasome-independent functions including endocytosis, signalling, replication, and transcription (for review see (2)). The transfer of ubiquitin to a target protein is a multi-step ATP-dependent process that involves ubiquitin activating enzymes, ubiquitin-conjugating enzymes, and ubiquitin ligase. Ubiquitination is also balanced by the presence of deubiquitinating enzymes. Together with the 26S proteasome these make up the UPS. The 26S proteasome consists of a 20S catalytic core capped on either end by the 19S regulator. This regulator is required for the recognition and binding of ubiquitin-conjugated proteins, allowing deubiquitination and entry into the catalytic core (for review see (3)). Many of the substrates for the 26S proteasome are short-lived regulatory proteins, not necessarily damaged or denatured, but which require tagging with a poly-ubiquitin tail as a hydrophobic patch to be recognised by the 26S proteasome (4–5). Nevertheless ubiquitin is also involved in turnover of misfolded proteins, such as when endoplasmic reticulum (ER)-associated proteins involved in protein folding are damaged or altered during disease progression. Generally, this could lead to the possibility that at least a small fraction of any protein may be ubiquitinated and explain why such a large number of proteins have been identified as having been tagged with ubiquitin. Ubiquitination therefore necessitates sensitive methods of detection. Previously used methods include incubation with his-tagged ubiquitin or ubiquitin tagged with FLAG epitope and subsequent purification or detection (6–7). As the number of ubiquitinated proteins can be very large, a more general method for an indication of ubiquitin levels and identification of ubiquitin proteins is to use two-dimensional electrophoresis (2DE) in combination with immunoblotting with an anti-ubiquitin antibody. Proteins can subsequently be identified using mass spectrometry (MS). Proteolysis of ubiquitin proteins produces a unique peptide ubiquitination site that contains a lysine with an iso-peptidelinked glycine–glycine sequence and can be identified by MS analysis due to its mass shift of 114.1 Da and the lack of proteolytic cleavage of modified lysine residues (6). Interestingly it has been shown from 2DE immunoblotting that the ubiquitination profile can differ largely from the carbonylation profile of oxidised proteins (8). It has been suggested that carbonylated proteins may be degraded without ubiquitin conjugation, possibly by the 20S proteasome. Oxidised proteins can have partial unfolding with the exposure of hydrophobic regions and do not require the hydrophobic patch that a poly-ubiquitin tail would confer (4). Nevertheless exposure to oxidative stress has increased ubiquitination levels in proteins (Fig. 1). This
Detection of Ubiquitination in 2DE 4
pI
7
4
CONTROL
pI
379 7
CdCl2
Fig. 1. Protein ubiquitination on exposure of mussels to cadmium. Mussels were exposed to CdCl2 and separated in 2DE, followed by blotting with anti-ubiquitin (right-hand panel). For comparison, an anti-ubiquitin blot of control is also shown (left-hand panel).
increase may be due to damage to the UPS itself, as both the 26S proteasome and ubiquitin-conjugating system are inhibited and inactivated by oxidative stress, which would lead to an increase in the number of ubiquitinated proteins (9). Alternatively oxidative damage to proteins involved in the folding of newly formed proteins in the ER such as protein disulphide isomerase or chaperone-like proteins could lead to an increase in misfolded and hence ubiquitinated proteins.
2. Materials 1. Total protein stain for membranes BLOT FastStain™ (GenoTechnology) is reversible and sensitive to at least 2 ng. 2. Plastic containers for membrane incubations. 3. Rocking platform. 4. PBST: Phosphate-buffered saline (PBS) containing 0.05% Tween 20. 5. Blocking solution: 1% BSA in PBST. 6. Primary antibody against ubiquitin, (Rabbit anti-ubiquitin from DakoCytomation was used in Fig. 1). 7. Secondary antibody (Goat anti-rabbit horseradish peroxidase from Sigma used in Fig. 1). 8. Chemiluminescent substrates. 9. X-ray film or chemiluminescent detector.
380
McDonagh
3. Methods 1. After electroblotting on to nitrocellulose or PVDF, reversibly stain membranes (see Chapter “Immunoblotting 2DE Membranes”). This is to ensure that there was a reproducible, high resolution and if quantifiably stained, comparable levels of total protein (see Note 1). 2. After membrane stain has been removed, membranes should be blocked using blocking solution for one hr at room temperature on a rocker. 3. Wash membrane 3 × 10 min with PBST on rocker. 4. Dilute primary antibody, anti-ubiquitin in blocking buffer (1:2,000) and incubate with membrane overnight on a rocker at 4ºC (see Note 2). 5. Wash membrane 3 × 10 min with PBST at room temperature. 6. Dilute secondary antibody in PBST (1:1,000) and incubate with membrane for 1 h at room temperature on rocker (see Note 2). 7. Ubiquitinated proteins are detected using chemiluminesence. In Fig. 1, Pierce Super Signal West Pico Chemiluminescent substrates were used according to manufacturer’s instructions (see Note 3). 8. Ubiquitinated proteins can be visualised either in dark room using X-ray film or by a chemiluminescent detector.
4. Notes 1. Ensure that reversible protein stain of membranes is compatible with immunodetection and is sensitive for detection of proteins in 2DE. 2. Antibody dilutions should be optimised before use with 1D SDS–PAGE. There may be differences between manufacturers and even stocks of antibodies. 3. Ensure that the entire membrane is covered in chemiluminesent substrates and for an equal period of time.
Acknowledgments I acknowledge an Embark fellowship from the Irish Research Council for Science, Engineering and Technology.
Detection of Ubiquitination in 2DE
381
References 1. Hershko, A., and Ciechanover, A. (1998). The ubiquitin system. Annu. Rev. Biochem. 67, 427–479 2. Mukhopadhyay, D., and Riezman, H. (2007). Proteasome-independent functions of ubiquitin in endocytosis and signalling. Science 315, 201–205 3. Kaiser, P., and Huang, L. (2005). Global approaches to understanding ubiquitination. Genome Biol. 6, Article 233 4. Shringarpure, R., Grune, T., Mehlase, J., and Davies, K. J. A. (2003). Ubiquitin conjugation is not required for the degradation of oxidised proteins by proteasome. J. Biol. Chem. 278, 311–318 5. Shang, F., Gong, X., and Taylor, A. (1997). Activity of ubiquitin-dependent pathway in response to oxidative stress. Ubiquitin activating enzyme is transiently up-regulated. J. Biol. Chem. 268, 15405–15411
6. Peng, J., Schwartz, D., Elias, J.E., Thoreen, C.C., Cheng, D., Marsischky, G., et al. (2003). A proteomics approach to understanding protein ubiquitination. Nat. Biotechnol. 21, 921–926 7. Sato, K., Hayami, R., Wu, W., Nishikawa, T., Nishikawa, H., Okuda, Y., et al. Nuclophosmin/ B23 is a Candidate substrate for the BRCA1BARD1 ligase. J. Biol. Chem. 279, 30919–30922 8. McDonagh, B., and Sheehan, D. (2006). Redox proteomics in the blue mussel Mytilus edulis: carbonylation is not a prerequisite for ubiquitination in acute free radical mediated oxidative stress. Aquat. Toxicol. 79, 325–333 9. Powell, S.R., Wang, P., Divald, A., Teichberg, S., Haridas, V., McCloskey, T. W., et al. (2005). Aggregates of oxidised proteins (lipofuscin) induce apoptosis through proteasome inhibition and dysregulation of proapoptotic proteins. Free Radic. Biol. Med. 38, 1093–1101
Chapter 25 Phosphoproteome Analysis by In-Gel Isoelectric Focusing and Tandem Mass Spectrometry Sarka Beranova-Giorgianni, Dominic M. Desiderio, and Francesco Giorgianni Summary Protein phosphorylation is central to most signaling events in eukaryotic cells. Large-scale analysis of protein phosphorylation in vivo is a highly challenging undertaking that requires powerful analytical and bioinformatics tools; numerous phosphoproteomic methodologies that use various combinations of these tools have been developed recently. This chapter describes an in-gel isoelectric focusing–liquid chromatography–tandem mass spectrometry (IEF–LC–MS/MS) analytical strategy for phosphoproteome mapping. The strategy encompasses seven steps: (1) extraction of proteins from the biological system under study (e.g., a tissue); (2) separation of the protein mixture by isoelectric focusing in an immobilized pH gradient (IPG) strip; (3) protein fixation followed by sectioning of the IPG strip; (4) digestion of the proteins in each gel section; (5) enrichment of phosphopeptides in the digests by immobilized metal ion affinity chromatography; (6) analysis of the enriched digests by LC–MS/MS; and (7) identification of the phosphopeptides/proteins through database searches, and assignment of the sites of phosphorylation in these proteins. Key words: Phosphoproteome, Isoelectric focusing, Immobilized pH gradient strip, Immobilized metal ion affinity chromatography, IMAC, Mass spectrometry.
1. Introduction Protein phosphorylation is a key post-translational modification in eukaryotic cells. Many cellular processes are directly controlled by phosphorylation, and disruptions in phosphorylation-mediated cell signaling events are associated with various diseases. An estimated one-third of all proteins are phosphorylated, comprising tens of thousands of distinct phosphorylation sites (1). David Sheehan and Raymond Tyther (eds.), Methods in Molecular Biology, Two-Dimensional Electrophoresis Protocols, vol. 519 © Humana Press, a part of Springer Science + Business Media, LLC 2009 DOI: 10.1007/978-1-59745-281-6_25
383
384
Beranova-Giorgianni, Desiderio, and Giorgianni
System-wide characterization of protein phosphorylation will improve the understanding of disease pathologies and of cellular behavior in general. Large-scale identification and characterization of phosphoproteins in complex biological systems is an extremely challenging task because of the large number of phosphorylation sites present in a phosphoproteome; low stoichiometry of phosphorylation; wide abundance range of phosphorylated proteins; and the dynamic nature of protein phosphorylation (2). In recent years, tandem mass spectrometry (MS/MS; see also Chapter “Shotgun Protein Analysis by Liquid Chromatography-Tandem Mass Spectrometry”) has emerged as an effective tool for analysis of in vivo phosphorylation. MS/MS provides data that are diagnostic of the amino acid sequence of a phosphorylated peptide and of the exact localization of the phosphorylation site(s). To probe effectively complex phosphoproteomes, MS/MS-based analytical strategies typically include steps for selective enrichment of phosphorylated peptides, in combination with multidimensional separations at the protein and/or peptide level. Phosphopeptide enrichment is commonly achieved via immobilized metal ion affinity chromatography (IMAC); for multidimensional separations, various methods have been employed, including cationexchange chromatography, reversed-phase chromatography, and sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS– PAGE) (3–6). In this chapter, an in-gel isoelectric focusing–liquid chromatography–tandem mass spectrometry (IEF–LC–MS/MS) strategy suitable for the analysis of mammalian phosphoproteomes is described. The general principle of this strategy is shown in Fig. 1. Proteins are extracted from the sample under study, and the protein mixture is fractionated by in-gel IEF in a conventional IPG strip. Following IEF, the proteins are fixed and the strip is divided into a set of gel sections. The proteins in each section are digested with trypsin, and the digests are subjected to IMAC to enrich for phosphorylated peptides. Enriched digests are analyzed by LC–MS/MS, and the LC–MS/MS datasets are used for searches of a protein sequence database to identify the phosphorylated peptides/proteins and to assign the sites of phosphorylation. The in-gel IEF–LC–MS/MS methodology incorporates in-gel IEF for an initial separation of the proteins based on their pI characteristics. This simplification of the protein mixture, combined with selective isolation of phosphorylated peptides, provides an increased coverage of the phosphoproteome. The methodology relies on a proven IEF technology with commercially available IPG strips that offer a high degree of reproducibility, high loading capacity, and flexibility in experimental design (IPG strips of various lengths and pH gradient ranges are available). Standard 2DE equipment is used, and multiple separations can be performed simultaneously. Because
Phosphoproteome Analysis by In-Gel Isoelectric Focusing
385
Tissue/Cells Protein Extraction Protein Separation by in-gel IEF Processing of IPG Strips Protein Digestion Phosphopeptide Enrichment by IMAC LC-MS/MS Database Searches (Phosphopeptide/Protein Identification; Assignment of Phosphorylation Sites ) Fig. 1. A flowchart depicting the in-gel IEF–LC–MS/MS strategy for phosphoproteomics. In this approach, proteins extracted from a biological sample are first fractionated by in-gel IEF using a conventional IPG strip. After IEF, the strip is divided into sections and the proteins in each section are digested with trypsin. Selective enrichment for phosphopeptides is achieved by IMAC, and the enriched digest is analyzed by LC–MS/MS. The LC–MS/MS data are used in database searches to identify the phosphopeptides/proteins and to characterize the location of their phosphorylation sites.
the information about the apparent pI of the identified phosphoprotein is preserved, data from in-gel IEF–LC–MS/MS can be linked to 2DE data. To date, the in-gel IEF–LC–MS/MS methodology has been applied to phosphoproteome analysis of human pituitary tissue (7) and of a human prostate cancer cell line.
2. Materials 2.1. Protein Extraction
1. Homogenizer (PowerGen 125; Fisher Scientific, Pittsburgh, PA). 2. Probe sonicator (Fisher Scientific, Pittsburgh, PA). 3. TRIZOL reagent kit (Invitrogen, Carlsbad, CA); phosphatase inhibitor cocktail (Sigma, St. Louis, MO). To prepare the TRIZOL solution, add 5 μL of phosphatase inhibitor cocktail per 1 mL of the TRIZOL reagent. 4. Phosphate-buffered saline (PBS; Sigma St. Louis, MO)
386
Beranova-Giorgianni, Desiderio, and Giorgianni
5. Ettan 2-D Quant protein concentration assay kit compatible with the IEF rehydration buffer (Amersham Biosciences, Piscataway, NJ, USA). 6. IEF rehydration buffer: urea (7 M), thiourea (2 M), CHAPS detergent (2%), IPG buffer (carrier ampholyte mixture; 2%), dithiothreitol (0.3%), and a trace of bromophenol blue dye (see Note 1). All reagents from Amersham Biosciences except bromophenol blue (Sigma, St. Louis, MO). 7. Guanidine–HCl wash solution: 0.3 M guanidine–HCl in 95% ethanol. 8. Siliconized 1.5-mL microcentrifuge tubes (Fisher Scientific, Pittsburgh, PA). 9. Other reagents: chloroform; ethanol; isopropanol; trifluoroacetic acid (TCA). All reagents should be of high purity. 2.2. In-Gel IEF
1. Multiphor II 2-DE system (Amersham Biosciences, Piscataway, NJ), which consists of a Multiphor II flatbed electrophoresis unit, a MultiTemp II thermostatic circulator, and an EPS 3501 XL power supply. 2. Accessories for Multiphor II: Immobiline DryStrip reswelling tray for the rehydration of IPG strips and an Immobiline DryStrip kit (containing a running tray, IPG strip aligner, electrodes, and electrode paper strips). 3. Immobilized pH gradient gels: pH 3–10, 11-cm IPG strips (Amersham Biosciences, Piscataway, NJ, USA). 4. DryStrip cover fluid (oil) from Amersham Biosciences. 5. IEF strip wash solution 1: 20% aqueous trichloroacetic acid (TCA). 6. IEF strip wash solution 2: 10% acetic acid/50% methanol/40% water. 7. IEF strip wash solution 3: 200 mM ammonium bicarbonate.
2.3. Processing of the IPG Strip; Protein Digestion; IMAC Enrichment
1. Trypsin solution: Dissolve 10 μg sequencing grade trypsin (Promega, Madison WI) per mL of 50 mM ammonium bicarbonate. 2. Peptide extraction solution: 60% acetonitrile/35% water/5% TFA. 3. Peptide reconstitution solution: 10% aqueous acetic acid. 4. IMAC columns (Phosphopeptide Isolation Kit; Pierce, Rockford, IL). 5. IMAC wash solution 1: 0.1% aqueous acetic acid. 6. IMAC wash solution 2: 90% water/10% acetonitrile/0.1% acetic acid. 7. IMAC elution solution 1: 200 mM sodium phosphate buffer, pH 8.4.
Phosphoproteome Analysis by In-Gel Isoelectric Focusing
387
8. IMAC elution solution 2: 20% acetonitrile/80% 200 mM sodium phosphate buffer, pH 8.4. 9. ZipTip C18 pipette tips (Millipore, Billerica, MA). 10. ZipTip elution solution: 50% acetonitrile/50% water/0.1% TFA. 11. Gel-loading pipette tips, 100 μL (Fisher Scientific, Pittsburgh, PA). 12. Siliconized microcentrifuge tubes, 1.5 mL and 0.6 mL (Fisher Scientific, Pittsburgh, PA). 13. Other reagents: trichloroacetic acid; ammonium bicarbonate; sodium phosphate monobasic; sodium hydroxide; methanol; acetic acid; acetonitrile; trifluoroacetic acid. All reagents should be of high purity. 14. Vacuum centrifuge (Eppendorf Vacufuge; Fisher Scientific, Pittsburgh, PA). 2.4. LC–MS/MS
1. Nanoflow chromatograph (Dionex; Sunnyvale, CA) coupled to a nanoelectrospray-linear ion trap mass spectrometer (ThermoElectron, San Jose, CA). 2. LC column: Picofrit™ column (360 μm OD, 75 μm ID, 15 μm tip ID) from New Objective (Woburn, MA) packed with 9–10 cm of C18 silica-based reversed-phase packing material (5 μm, 200Å MAGIC C18) from Michrom Bioresources (Auburn, CA) (see Note 2). 3. LC solvents: water (J.T. Baker, Phillipsburg, NJ); acetonitrile (J.T. Baker, Phillipsburg, NJ); formic acid (EM ScienceEMD Chemicals, Gibbstown, NJ) (see Note 3).
2.5. Bioinformatics
1. Bioworks 3.2. with TurboSEQUEST for database searches. 2. Additional bioinformatics tools for phosphoproteomics: SCANSITE (scansite.mit.edu; a tool for predictions of kinase-specific motifs or phosphorylation-specific proteinprotein interactions); Phosphosite (www.phosphosite.org; database of known phosphorylation sites); Phosida (www. phosida.org; database of known phosphorylation sites). These resources can be accessed free of charge.
3. Methods 3.1. Protein Extraction with TRIZOL Reagent
The TRIZOL-based method is well suited for small and large tissue specimens and for cultured cells. With this method, cellular proteins are separated from RNA and DNA (8). Prolonged storage of the protein extract may result in problems with subsequent solubilization.
388
Beranova-Giorgianni, Desiderio, and Giorgianni
3.1.1. Protein Extraction from Tissue (see Note 4)
1. Quickly weigh the frozen tissue specimen and mince it in small pieces; place the pieces in a 14-mL plastic tube and rinse them three times with 5 mL of ice-cold phosphatebuffered saline. 2. Add the TRIZOL solution (i.e., TRIZOL reagent with phosphatase inhibitors) at a volume of 4 mL of TRIZOL per gram of tissue. 3. Homogenize the mixture until full disruption of the tissue is achieved (during homogenization, keep the mixture on ice; alternate 10-s homogenization period with 20-s rest period). 4. After homogenization, sonicate the sample with the probe sonicator (alternate 10 s sonication with 20 s of rest; repeat this step three times). 5. Vortex the sample for 4 h at 4 °C. 6. Add 0.2 mL of chloroform per 1 mL of TRIZOL reagent used. Centrifuge the sample at 12,000 × g for 10 min. The mixture will separate into three phases: upper (aqueous) phase, interphase, and lower (organic) phase (see Note 5). Remove and discard the upper (aqueous) phase. 7. To the lower phases (interphase and organic), add 0.3 mL of ethanol per 1 mL of TRIZOL reagent used, and mix the sample by inversion (see Note 6). Centrifuge the mixture at 2,500 × g for 5 min. Draw off the supernatant and collect it in a clean tube. 8. To the supernatant collected in Step 7, add 1.5 mL of isopropanol per 1 mL of TRIZOL reagent used, and mix the sample by inversion (see Notes 7 and 8). Centrifuge the mixture at 12,000 × g for 10 min. Remove and discard the supernatant. 9. Wash the protein with guanidine–HCl wash solution (2 mL of the wash solution per 1 mL of TRIZOL reagent used); repeat the washing procedure three times (see Note 9). 10. Briefly vortex the protein pellet in 2 mL of ethanol. Then store the pellet in ethanol for 20 min, and centrifuge at 7,500 × g for 5 min. Remove and discard the ethanol solution. 11. Dissolve the protein pellet in an appropriate volume of IEF rehydration buffer containing no dye (see Note 10). 12. Determine protein concentration in the rehydration buffer (see Note 11). 13. Perform IEF of the sample; or aliquot the sample and store the aliquots at −80°C until further use (see Note 12).
3.1.2. Protein Extraction from Cultured Cells
1. Pellet cells by centrifugation and wash them with ice-cold phosphate-buffered saline.
Phosphoproteome Analysis by In-Gel Isoelectric Focusing
389
2. Add TRIZOL solution to the cell pellet at 1 mL of TRIZOL per 5 × 106 cells. 3. Vortex the mixture intermittently until full lysis of the cell pellet is achieved (see Note 13). 4. After cell lysis, proceed with protein extraction as described in Subheading 3.1.1, steps 6–13. 3.2. In-Gel IEF
These instructions assume the use of a Multiphor II system. For other IEF systems, instructions provided by the manufacturer should be followed.
3.2.1. Sample Application
1. Centrifuge the TRIZOL-extracted sample (i.e., protein dissolved in the rehydration buffer; Subheading 3.1) at 21,000 × g for 30 min (see Notes 14 and 15). 2. Mark the plastic backing of IPG strip(s) with a permanent marker to outline the sections to be cut later (see Notes 16–18). 3. Pipette an exact volume of the sample solution into a slot in the DryStrip reswelling tray (see Notes 14 and 19). 4. Peel off the protective cover from the IPG strip, and place the strip gel-side down on top of the sample solution (see Note 20). To prevent evaporation, cover the strip in the slot with 3 mL of oil. Allow the covered IPG strip to rehydrate overnight.
3.2.2. Isoelectric Focusing (see Note 21)
1. Set the MultiTemp circulator to 20°C. 2. Pour 4 mL of oil (cover fluid) onto the flatbed cooling plate of the Multiphor II unit; position the DryStrip running tray onto the cooling plate, and connect the electrode wires to appropriate positions. 3. Pour 10 mL of oil into the DryStrip running tray, and place the plastic IPG strip aligner into the tray. 4. Prepare two 100-mm-long electrode paper strips, and moisten them with 0.5 mL of water; blot off excess water. 5. Remove the rehydrated IPG strip carrying the protein sample from the reswelling tray; rinse the surface of the gel with water; wipe off oil by sweeping the plastic backing of the strip over a sheet of filter paper. 6. Place the IPG strip into a groove in the strip aligner, gel-side up, and in the correct orientation (see Notes 22 and 23). 7. Place the moist electrode paper strips (step 4) across the anodic and cathodic ends of the IPG strip; the electrode paper strips must be in contact with the gel ends of the IPG strip. Position the electrodes over the electrode paper strips. To cover the IPG strip, pour 70–80 mL of oil into the tray.
390
Beranova-Giorgianni, Desiderio, and Giorgianni
8. Place the lid onto the electrofocusing chamber, connect the cables to the power supply, and perform IEF according to the following protocol: 0–100 V (gradient over 1 min); 100 V (fixed for 120 min); 100–500 V (gradient over 1 min); 500– 3,500 V (gradient over 90 min); 3,500 V (fixed for 6 h) (see Note 24). 9. After completion of IEF, remove the IPG strip from the tray; wipe off excess oil; loosely wrap the IPG strip in plastic wrap; store the strip at −20°C until further processing. 3.3. Processing of the IPG Strip; Protein Digestion; and IMAC Enrichment
Steps 1–5 are performed with the whole IPG strip (see Note 25). 1. Rinse with 15 mL of water for 30 s.
3.3.1. Processing of the IPG Strip
3. Incubate with 15 mL of IEF strip wash solution 2 (see Subheading 2.2) for 15 min, with the solution changed every 5 min.
2. Incubate in 15 mL of IEF strip wash solution 1 (see Subheading 2.2) for 30 min.
4. Wash with 15 mL of water for 5 min. 5. Incubate in IEF strip wash solution 3 (see Subheading 2.2) for 5 min. 6. Using a clean scalpel, separate the strip into gel sections; place each gel section into a siliconized 0.6-mL microcentrifuge tube (see Note 26). Steps 7 and 8 are performed after sectioning of the IPG strip: 7. To each gel section, add 100 μL of acetonitrile; repeat twice (see Note 27). 8. Dry the gel sections for 30 min in a vacuum centrifuge. 3.3.2. Protein Digestion
1. Rehydrate each gel section with 100 μL of trypsin solution (see Subheading 2.3). 2. Incubate the samples in a water bath at 37°C overnight. 3. After digestion, acidify each sample with 0.5 μL of TFA, and add 100 μL of acetonitrile. 4. Centrifuge the samples at 10,000 × g for 1 min; transfer each supernatant into a clean 0.6-mL microcentrifuge tube. 5. To extract the remaining peptides from each gel section, add 50 μL peptide extraction solution (see Subheading 2.3); vortex for 10 min; centrifuge at 10,000 × g for 1 min; draw off the supernatant and combine with digest solution obtained in step 4. 6. Dry the samples in a vacuum centrifuge.
3.3.3. Phosphopeptide Enrichment with IMAC
The IMAC procedure involves selective binding of phosphorylated peptides and removal of nonphosphorylated peptides; the phosphopeptides are then eluted with a suitable buffer (see Note 28).
Phosphoproteome Analysis by In-Gel Isoelectric Focusing
391
1. Reconstitute each digest (obtained in step 6; Subheading 3.3.2) in 50 μL of peptide reconstitution solution (see Subheading 2.3). 2. Apply each sample to the IMAC column; incubate for 1 h with occasional swirling of the contents. Remove the bottom seal of the column; place the column in a microcentrifuge tube; centrifuge the tube at 1,000 × g for 1 min. Discard the eluate from the column. 3. Wash the IMAC column with 40 μL of IMAC wash solution 1 (swirl the contents); centrifuge the column at 1,000 × g for 1 min; discard eluate. Repeat this washing step one more time. 4. Wash the IMAC column with 40 μL of IMAC wash solution 2 (swirl the contents); centrifuge the column at 1,000 × g for 1 min; discard the eluate. Repeat this washing step one more time. 5. Wash the IMAC with 40 μL of water (swirl the contents); centrifuge the column at 1,000 × g for 1 min; discard eluate. Repeat this washing step one more time. 6. To elute the phosphopeptides, add 30 μL of IMAC elution solution 1 to the column; incubate for 5 min; centrifuge the column at 1,000 × g for 1 min; collect the eluate (see Note 29). Repeat the elution one more time, and combine the eluates. 7. For additional phosphopeptide elution, add 30 μL of IMAC elution solution 2; incubate for 5 min; centrifuge the column at 1,000 × g for 1 min; collect the eluate and combine with eluates from step 6. 8. To the combined eluates from steps 6 to 7, add 0.9 μL of TFA, and reduce the volume of the sample to 2–5 μL in a vacuum centrifuge. 9. To the enriched digest, add 15 μL of 0.1% aqueous TFA. Desalt the enriched digest with a ZipTip column, following instructions provided by the manufacturer. Elute the peptides from the ZipTip with 3 μL of ZipTip elution solution. 10. Add to each eluate 3 μL of 0.5% aqueous acetic acid (see Note 30). 11. Store peptide samples at 4°C until LC–MS/MS analysis (see Note 31). 3.4. LC–MS/MS
To obtain data diagnostic of the peptide sequence and of the location of the phosphorylation site(s), the IMAC-enriched digest from each section of the IPG strip is analyzed by LC–MS/MS. The peptide mixture is separated by nanoflow reversed-phase HPLC, using a gradient of water/methanol (see Note 32). The peptides separated by nanoLC are introduced online into the mass spectrometer and analyzed. The MS measurements are performed
392
Beranova-Giorgianni, Desiderio, and Giorgianni
in data-dependent mode, in which the instrument executes one MS scan to determine the peptide molecular masses, followed by acquisition of peptide sequence-diagnostic MS/MS (product-ion) spectra for several of the most abundant ions from the MS scan. This cycle is repeated throughout the entire run, thus producing extensive MS/MS data for peptides in a given digest. In our experiments, the mass spectrometer cycles between acquisitions of a full-scan MS spectrum, followed by seven MS/MS scans. 3.5. Database Searches
The MS/MS data are used to search a protein sequence database (SWISSPROT or NCBInr) to identify the phosphopeptides/proteins in each section of the IPG strip (see Note 33). Following interrogation of the proteins sequence database, the phosphopeptide hits are evaluated by manual inspection of the database search outputs and of the corresponding MS/MS spectra. This evaluation has two objectives: confirmation of the correct amino acid sequence of the phosphopeptide which establishes the presence of the phosphorylated form of the corresponding protein in the biological system under study; and assignment of the exact phosphorylation site(s) in the phosphopeptide (see Note 34). Further information on the characterized phosphoproteins/ sites can be obtained using various bioinformatics resources available over the Internet (see Subheading 2.5 and Chapter “Database Interrogation Algorithms for Identification of Proteins in Proteomic Separations”).
4. Notes 1. The rehydration buffer (without bromophenol blue dye and DTT) can be prepared in advance and frozen in aliquots. The dye (which would interfere with protein assay) and DTT should be added just before use. 2. Columns may be packed in house, e.g., by using a pressure vessel. Alternatively, prepacked columns for nanoflow LC are commercially available from New Objective (Woburn, MA). 3. High-purity solvents must be used (see recommendations in the main text) and extreme care must be taken to avoid contamination. Dedicated glassware should be used for mobilephase preparation. 4. A face shield, gloves, and protective clothing should be worn during tissue processing. Rules for handling of biohazard materials and for disposal of biohazard waste must be followed.
Phosphoproteome Analysis by In-Gel Isoelectric Focusing
393
5. The aqueous phase contains RNA; the interphase and the organic phase contain DNA and proteins. 6. This will precipitate the DNA. 7. This will precipitate the proteins. 8. Alternatively, the protein precipitation can be performed with acetone instead of isopropanol. For acetone precipitation, a volume of acetone five times the volume of the supernatant solution is added, the sample is mixed by inversion, and centrifuged at 12,000 × g for 10 min. (If 1.5-mL tubes are used, the sample will have to be split into aliquots prior to addition of acetone to accommodate the acetone volume). The supernatant is discarded, and the protein pellet is processed further as described in the main text. 9. During each wash cycle, the protein pellet should be stored in the wash solution for 20 min and then centrifuged at 7,500 × g for 5 min. 10. The appropriate volume of rehydration solution will depend on the amount of protein that was extracted; on the desired protein load for IEF; and on the type of IPG strip to be used (see Notes 14 and 15). A rough estimate of the protein content of a particular sample (tissue, cells) is typically known to the investigator, or it should be obtained in a separate experiment, if possible. 11. The protein assay used must be compatible with the components of the rehydration buffer. 12. Prolonged storage of the protein pellet can make subsequent solubilization difficult. 13. For LNCaP cells, solubilization of the cell pellet takes up to 60 min. 14. The suitable protein load will depend on the type of IPG strip. For our experiments to date, we use 250–300 μg of protein per 11-cm, pH-3–10 strip. 15. The volume of the sample may need to be adjusted with additional rehydration buffer in order to obtain the required protein concentration. Note that IPG strips require a specific sample volume for rehydration. 16. Marking the IPG strip prior to IEF facilitates subsequent sectioning of the gel (Subheading 3.3). 17. In our experiments to date, ten gel sections were used (each section was 1.1 cm). 18. Several IPG strips may be run in parallel and the corresponding sections from each strip may be combined for further processing.
394
Beranova-Giorgianni, Desiderio, and Giorgianni
19. The final sample volume applied to the IPG strip is 200 μL. This volume is appropriate for 11-cm IPG strips from Amersham Biosciences. Strips of different length and/or from a different manufacturer may require a different volume for proper strip rehydration. 20. There should be no bubbles in the sample solution that is in contact with the gel; pipette the solution slowly. 21. Steps 1–4 should be completed before removal of the IPG strip from the reswelling tray. 22. The IPG strips have markings indicating the anodic end. 23. The IPG strip should not be left exposed for more than 5 min. 24. This IEF protocol is used in our laboratory for separations of proteins from human pituitary or human prostate. Note that optimum IEF conditions are sample dependent and that the protocol may need to be adjusted to achieve optimum focusing. 25. The steps involving incubations of the whole strip can be performed in a glass tube or in a suitable tray. 26. The dull edge of the scalpel should be placed almost flat against the strip at the appropriate mark (see Note 16), and the gel piece should be scraped off the plastic backing and transferred into the microcentrifuge tube. 27. Gel-loading pipette tips must be used; regular tips would clog. 28. A number of different IMAC columns are commercially available. Furthermore, other phosphopeptide enrichment strategies such as the use of titanium dioxide columns have been recently developed (9). The procedure described in the main text is for Ga(III)-based minispin columns from Pierce. 29. Before the phosphopeptide elution step, make sure that the seal is placed on the bottom of the IMAC column. After incubation of the column with the elution buffer, carefully remove the seal; place the column in a clean siliconized microcentrifuge tube and centrifuge. 30. The sample volume may need to be adjusted depending on the particular injection setup for LC–MS/MS. In our work, we use manual injection through a six-port valve. 31. We normally store the dried digests (produced in step 6; Subheading 3.3.2) at −20°C; we then process the digests via IMAC and ZipTip in smaller batches, so that LC–MS/ MS of each batch can be completed in 1–2 days. 32. The LC conditions used in our laboratory are: (a) sample injection volume: 2 μL; (b) gradient of mobile phases A and B (A = 98% water/2% methanol/0.05% formic acid; B = 90% methanol/10% water/0.05% formic acid): 5 min
Phosphoproteome Analysis by In-Gel Isoelectric Focusing
395
initial isocratic elution with 0% B, followed by a linear gradient 0–90% B in 90 min, 15 min isocratic elution with 90% B, and a linear gradient 90–0% B in 5 min; (c) flow rate: 200 nL/min; (d) to minimize carryover, the injector port is washed between injections with 200 μL of a solution of 50% methanol/50% water, followed by 200 μL of 98% water/2% methanol/0.05% formic acid. 33. Typically, the search parameters include: fully tryptic peptides are considered; maximum of two missed cleavages; differential modification of +80 Da on S, T, and Y; and differential modification of +16 Da on M. 34. Validation of the phosphopeptides retrieved in database searches is a time-consuming but essential step in phosphoproteome analysis. For validation of the correctness of the phosphopeptide sequence, we consider peptides retrieved by TurboSEQUEST as top matches with thresholds for Xcorr values of ≥ 2.5 for doubly charged precursor ions and ≥ 3.5 for triply charged precursor ions. For these peptides, we then evaluate the quality of the corresponding MS/MS spectra. The criteria that we consider for MS/MS spectra are: a good signal-to-noise ratio with most of the abundant product ions assigned (additional product ions can be assigned to dissociation processes not considered by SEQUEST); the presence of continuous stretches of y- and/or b-ion series; and the presence of a phosphate-diagnostic product ion (10). Next, we apply the same procedure to examine the phosphopeptides whose Xcorr values are between 2.0 and 2.5 (doubly charged precursor ions). All phosphopeptides that pass the sequence validation are compiled in a summary table. Next, we examine in detail the product-ion patterns in the MS/MS spectra for the phosphopeptides from this summary table to confirm the assignment of the site(s) of phosphorylation. If the exact location of the site cannot be verified, we note this fact in the summary table.
References 1. Cohen, P. (2001) The role of protein phosphorylation in human health and disease. The Sir Hans Krebs Medal Lecture. Eur. J. Biochem. 268, 5001–5010 2. Reinders, J., and Sickmann, A. (2005) Stateof-the-art in phosphoproteomics. Proteomics 5, 4052–4061 3. Molina, H., Horn, D. M., Tang, N., Mathivanan, S., and Pandey, A. (2007) Global proteomic profiling of phosphopeptides using electron transfer dissociation tandem mass spectrometry. Proc. Natl. Acad. Sci. U. S. A. 104, 2199–2204
4. Nousiainen, M., Sillje, H. H., Sauer, G., Nigg, E. A., and Korner, R. (2006) Phosphoproteome analysis of the human mitotic spindle. Proc. Natl. Acad. Sci. U. S. A. 103, 5391– 5396 5. Olsen, J. V., Blagoev, B., Gnad, F., Macek, B., Kumar, C., Mortensen, P., et al. (2006) Global, in vivo, and site-specific phosphorylation dynamics in signaling networks. Cell 127, 635–648 6. Villen, J., Beausoleil, S. A., Gerber, S. A., and Gygi, S. P. (2007) Large-scale phosphorylation
396
Beranova-Giorgianni, Desiderio, and Giorgianni
analysis of mouse liver. Proc. Natl. Acad. Sci. U. S. A. 104, 1488–1493 7. Beranova-Giorgianni, S., Zhao, Y., Desiderio, D. M., and Giorgianni, F. (2006) Phosphoproteomic analysis of the human pituitary. Pituitary 9, 109–120 8. http://tools.invitrogen.com/content/sfs/ manuals/15596018%20pps%20Trizol%20 Reagent%20061207.pdf
9. Larsen, M. R., Thingholm, T. E., Jensen, O. N., Roepstorff, P., and Jorgensen, T. J. (2005) Highly selective enrichment of phosphorylated peptides from peptide mixtures using titanium dioxide microcolumns. Mol. Cell Proteomics. 4, 873–886 10. DeGnore, J. P., and Qin, J. (1998) Fragmentation of phosphopeptides in an ion trap mass spectrometer. J. Am. Soc. Mass Spectrom. 9, 1175–1188
Chapter 26 Detection of Protein Glutathionylation Elisabetta Gianazza, Ivano Eberini, and Pietro Ghezzi Summary Recent studies indicate that protein glutathionylation is an important regulatory mechanism. The development of redox proteomics techniques to identify proteins undergoing glutathionylation has a key role in defining the importance of this post-translational modification, although the available methods are not yet comparable to those for the study of other modifications like phosphorylation. We describe here methods that have been successfully employed to identify in vitro glutathionylated proteins. Key words: Glutathione, Glutathionylation, Redox regulation, Thiols, Disulfides, Cysteine.
1. Introduction Proteins can undergo different forms of oxidative post-translational modifications (PTM). Some are irreversible, such as nitration of tyrosine or carbonylation of various amino acid residue side chains, and are normally considered only in the context of cellular toxicity. The thiol group of cysteine can exist both in its reduced (sulfhydryl) and in its oxidized (disulfide) form. It has long been thought that, in a given protein, each cysteine could be assigned a specific condition, i.e., as a free sulfhydryl or as a structural disulfide. It has also been known that, while all cysteines in extracellular secreted proteins are oxidized to disulfides, in the cytoplasm virtually all cysteines are present as free sulfhydryls, due to the strongly reducing environment set by the high concentration of reduced glutathione (γ-glutamylcysteinylglycine; GSH) as well as by the high value of the ratio: reduced GSH/oxidized GSH (GSSG) (1–3). However, the identification of oxidized thiols through proteomic procedures has demonstrated that many cytoplasmic David Sheehan and Raymond Tyther (eds.), Methods in Molecular Biology, Two-Dimensional Electrophoresis Protocols, vol. 519 © Humana Press, a part of Springer Science + Business Media, LLC 2009 DOI: 10.1007/978-1-59745-281-6_26
397
398
Gianazza, Eberini, and Ghezzi
proteins can undergo different reversible forms of cysteine oxidation. These modifications include S-nitrosylation, formation of sulfinic acids (4), of protein disulfides (5), and of mixed disulfides with GSH to form glutathionylated proteins (6, 7). Glutathionylation (sometimes also termed glutathiolation or glutathionation), which can be catalyzed by glutaredoxin, represents the reaction between protein cysteines (PSH) and GSH to yield PSSG. The process may result from direct oxidation (PSH + GSH + oxidant), from thiol-disulfide exchange (PSH + GSSG) or from nitrosylation (PSH + GSNO, also yielding PSSG). These findings hint at the existence of “redox-sensitive” proteins whose functions can be regulated by the redox cycling of specific cysteines through various types of reversible oxidation. These proteins could possibly act as sensors of the redox state of the cell, which would explain how the redox state can control signal transduction and metabolic pathways (“redox regulation” reviewed in ref 8).
2. Materials 1. (Protocol A Subheading 3.2.1) Unconjugated anti-GSH antibodies are commercially available both in polyclonal (e.g., from Advanced Targeting Systems, Calbiochem, Cell Sciences, Chemicon, GeneTex, Novus) and in monoclonal form (e.g., from Chemicon, GeneTex, QED Bioscience, ViroGen). 2. (Protocol B Subheading 3.2.2) GSH– and GSSG–Sepharose (Sigma, St.Louis, CO, USA). 3. Protocol B binding buffer: 50 mM H3PO4/Na, pH 7.5, containing protease inhibitors. 4. (Protocol C ***Subheading “Enzymatic Reaction with N-Ethylmaleimide-Biotin After Dethiolation”) N-ethylmaleimide and N-ethylmaleimide-biotin are available from Sigma; glutaredoxin is available from IMCO Corporation. 5. (Protocol D ***Subheading 3.2.3.2) Sulfosuccinimidyl-6(biotinamido)hexanoate and sulfo-N-hydroxysuccinimidebiotin (Pierce); an octadecylsilane column (e.g., Hypersil® from Thermo Scientific); biotinamidocaproic acid 3-sulfoN-hydroxy-succinimide ester, avidin-peroxidase conjugate, and streptavidin-agarose (all from Sigma). 6. (Protocol E Subheading 3.2.4) The following chemicals and ready-made solutions are required: Hanks’ Balanced Salt Solution (HBSS) culture medium (Invitrogen, Carlsbad, CA, USA); acetic acid, low electroendosmosis agarose, bromophenol blue, chloroform, Coomassie Brilliant Blue R250, cycloheximide, diamide, N-ethylmaleimide, GelBond®,
Detection of Protein Glutathionylation
399
glycerol, glycine, HEPES, hydrochloric acid, methanol, trichloroacetic acid, tris-hydroxymethylaminomethane, urea (all from Sigma); Amplify®, Immobiline II® monomers (pK 4.6, 6.2, 7.0, 8.5, 9.3), Ampholine® and Pharmalyt® carrier ampholytes (pH range: 3–10, 4–6, 5–7), l-[35S] cysteine (specific activity, 1,000 Ci/mmol) (all from GEAmersham, Amherst, MA, USA); acrylamido buffer of pK 0.8 (Fluka, Buchs SG, CH); acrylamide, N,N -bis-acrylamide, ammonium persulfate, sodium dodecylsulfate, TEMED (all from BioRad, Hercules, CA, USA).
3. Methods 3.1. Quantification on a Single Protein
Hemoglobin (Hb) is the single, high-abundance protein whose glutathionylation is currently investigated as a clinical marker of oxidative stress (9). Various types of glutathionylation of Hb (HbSSG) have been characterized in erythrocytes (RBC): (1) a mixed disulfide bond formed between GSH and normal Hb; (2) a disulfide bond between C93 of metHb β-chain and GSSG; and (3) a disulfide bond between the other cysteine residues of metHb α-chain and/or metHb β-chain and GSSG (10, 11). Upon oxidative stress, globin α-chains preferentially associate with RBC membrane (12). Most procedures for the quantification of HbSSG involve RBC lysate fractionation by liquid chromatography (LC), either as reverse-phase (13–18) or as cation exchange chromatography (19, 20), with on-line or off-line detection/identification by electrospray ionization mass spectrometry (ESI-MS). In one report matrix-assisted laser desorption time of flight (MALDI-TOF) was applied to the whole lysate (21).
3.2. Analysis in a Complex Sample
A number of procedures for studying glutathionylation in a proteomic setup are listed in Table 1. Protocols A–C do not involve metabolic labeling and may therefore be applied to all types of samples. Protocols A and B rely on the natural PTM moiety, whereas C makes use of the enzyme glutaredoxin to exchange a tagged chemical for covalently bound GSH. Conversely, protocols D and E entail metabolic labeling and may only be applied to a cell culture setup; the tag is either biotin, in D, or radioactivity (35S), in E. Tagged antibodies are used for detection on blots of the natural PTM, tagged avidin for the detection of the biotinylated derivatives. Covalent redox chromatography on the natural PTM or avidin affinity chromatography on the biotinylated derivatives may be used as preparative procedures eventually leading to the identification of the modified proteins. After incorporation
Protocol A
Protocol C
Purification of biotinylated derivatives by avidin affinity
Metabolic incorporation of biotin-cysteine in the presence of cycloheximide or of biotinglutathione (ethyl ester) in the absence of cycloheximide
Protocol D
All steps performed under nonreducing conditions
Reaction with glutaredoxin + Purification of N-ethylmaleimide-biotin glutathionylated proteins by chromatography Purification of biotinylated derivatives by avidin affinity on glutathioneagarose
Protocol B
Protocol E Metabolic incorporation of 35 S-cysteine in the presence of cycloheximide
For cell culture samples
PostelectroImmunodetection with MS (identification) Avidin-peroxidase MS (identification Avidin-peroxidase MS (identification 1DE or 2DE folon blots on blots lowed by autophoresis anti-glutathione (detection) (detection) radiography/ procedures Ab or blot overlay MS (detection/ with biotinylated identification) GST (detection)
2DE
Sample preparation
Cell culture
Protocol step
For in vivo, ex vivo, in vitro samples
Protocol suitability
Table 1 Procedures for the analysis of glutathionylated proteins in complex samples
400 Gianazza, Eberini, and Ghezzi
Detection of Protein Glutathionylation
401
of 35S-cysteine, two-dimensional electrophoresis (2DE) may be used allowing at once detection and identification. 3.2.1. Protocol A: Immunological Detection
Literature reports (a list is provided in Table 2) most often deal with in vitro treatment of purified/recombinant proteins (22–24) or to the PTM characterization of immunoprecipitated components (25). The use of anti-GSH antibodies to reveal PSSG in immunoblots has allowed identification of abundant proteins, such as actin, as a glutathionylation target in human fibroblasts (26). Recent studies have applied immunological techniques to standard proteomic procedures leading, for instance, to the identification of over 20 proteins whose glutathionylation was induced by anticancer agents (27). Another potentially useful technique, whose principle is very similar to that of using anti-GSH antibodies, is based on the high affinity of GSH transferase (GST) for GSH. GST overlay can be used to visualize glutathionylated proteins in one-dimensional electrophoresis (1DE) separations (28).
3.2.2. Protocol B: Redox Chromatography on Glutathione-Agarose
GSH-agarose is effective for affinity purification of proteins containing GSH-binding sequences. On the other hand, two reports – (29) and in more detail (30) – describe its application to purification of PSSG: molecules with exposed thiolate anions readily forming disulfide linkages with immobilized GSH, in a process mimicking the natural glutathionylation reaction in cells. As a preparative procedure this protocol allows quantitative recovery of bound proteins under mild, nondenaturing conditions. 1. Equilibrate GSH– and GSSG–Sepharose matrices with binding buffer. 2. Incubate 1 mg aliquots of protein (from cells sonicated in binding buffer) with 100 μL resin, in rotating tubes, for 2 h at 4°C. 3. Wash the beads with binding buffer fortified with 0.25 M NaCl. 4. Elute bound proteins with 10 mM dithiothreitol (DTT) in the same buffer. 5. Analyze total lysate, bound and unbound fractions by SDSPAGE.
3.2.3. Biotin-Avidin Protocols Protocol C: Enzymatic Reaction with N-Ethylmaleimide-Biotin After Dethiolation
This multistep procedure (31) is most suitable to monitor baseline PTM. 1. Block reactive –SH with excess N-ethylmaleimide (40 mM for 5 min in culture medium). 2. Lyse cells by sonication and add glutaredoxin at a ratio 1:8–1:15 vs. total protein specifically to remove bound GSH.
402
Gianazza, Eberini, and Ghezzi
3. Incubate for 2 min at 37°C in the presence of 0.5 mM GSH. 4. Tag the freed –SH with 5 mM N-ethylmaleimide-biotin, for 5 min at room temperature (Caution: glutaredoxin has variable affinity for various PSSG substrates (32)). Protocol D
1. Metabolic Labeling with Biotin-Cysteine or Biotin-Glutathione (Ethyl Ester) Biotinylation of Cysteine 1. To 29 mg of cysteine in 2 mL of 10× PBS, add 120 mg of the water-soluble biotinylation reagent sulfosuccinimidyl-6(biotinamido)hexanoate. 2. Leave to derivatize for 1 h at room temperature. The selected pH ensures specificity of reaction at the amino vs. the sulfydryl group. 3. Purify biotin-cysteine by HPLC on an octadecylsilane column (33). Biotinylation of GSH 1. To 25 mg of GSH in PBS add 50 mg of the water-soluble amino reactive biotinylation reagent biotinamidocaproic acid 3-sulfo-N-hydroxy-succinimide ester. 2. Leave to derivatize for 1 h before the addition of 250 mg BSA in 50 mL buffer to quench the excess reagent. 3. After 20 min, spin the solution through a 10-kDa cut-off centrifugal filter unit to remove unreacted and biotinylated BSA (29). Biotinylation of GSH ethyl ester 1. React sulfo-N-hydroxysuccinimide-biotin with GSH ethyl ester at a 1:1 molar ratio (4:3 by weight) in 50 mM NaHCO3, pH 8.5. 2. After 1 h at room temperature, terminate the reaction by the addition of a fivefold molar excess of NH4HCO3 (34). Labeling with biotin-cysteine requires inhibition of protein synthesis, biotin-GSH (29), or of its membrane-permeant analog, biotin-GSH ethyl ester (34), do not. Biotin-cysteine (33) and biotin-GSH (29) have been used in ex vivo experiments, added at a concentration of 0.5 mM to the bicarbonate perfusion buffers of isolated heart preparations. Their concentration was confirmed spectrophotometrically using Ellman’s reagent with cysteine as a standard. Biotin-GSH ethyl ester has been used in cell culture assays, added to the medium as a 1:100 dilution of the reaction mixture (approximately 250 μM free –SH) (34). 2. Detection of PSSG on blots with Streptavidin-HRP Avidin-peroxidase conjugate is commercially available from more than 20 companies in various forms. For instance, the aque-
Detection of Protein Glutathionylation
403
ous solution marketed by Sigma is to be used at 1:400,000 dilution, in combination with chemiluminescent (ECL) detection. 3. Purification of PSSG This involves the following steps: 1. Clarification: Incubate 50 μL/mg of soluble protein for 30 min at 4°C with streptavidin-agarose blocked with 100-fold excess biotin. 2. Binding: Incubate at 4°C for 30 min to overnight, using 100 μL streptavidin-agarose for every 100 mg tissue or 1 mg protein; buffers may contain up to 0.01% SDS. 3. Washing: With buffers containing 0.1% Triton X100 or with RIPA buffer (1% NP-40, 0.1% SDS, 0.5 mg/mL sodium deoxycholate, 150 mM NaCl, and 50 mM Tris–HCl, pH 7.5) followed by PBS containing 2 mM EDTA and 0.1% SDS. 4. Elution: With 2% SDS, or with 20 mM DTT, or with 0.1% SDS-10 mM DTT. 3.2.4. Protocol E: Metabolic Labeling with 35 S-Cysteine
Cell Culture Procedures
The most general procedure for assessing protein glutathionylation involves metabolic labeling of the intracellular GSH pool with 35S-cysteine while inhibiting protein synthesis (35, 36). The main steps in the protocol involve: extraction, 1DE or better 2DE, staining, fluorography, and spot identification by mass spectrometry (MS) (6, 7). Control experiments include: evaluation of constitutive glutathionylation, i.e., in the absence of oxidative stress; reversibility of labeling upon reduction with DTT; discrimination from cysteinylation by showing that labeling is inhibited by blocking GSH synthesis with buthionine sulfoximine. While stable in acidic buffers, PSSG lose their tags at neutral to alkaline pH. Their analysis, with special reference to the IEF step, has then to be as quick as possible. To comply with this requirement we have devised and tested the protocol outlined in the following flow-sheet. The main recipes and experimental conditions are detailed in Table 2. MS analysis of several spots from 2DE maps obtained in this way confirmed that identified proteins had migrated to their expected pI (6, 7). The protocol thus appears adequate to give equilibrium focusing on immobilized pH gradients (IPG) over a timescale closer to typical carrier ampholyte (CA)-IEF protocols, without resorting to voltage extremes (37). Figure 1 shows a typical result obtained using this methodology. Radiolabeling the intracellular GSH pool is carried out according to the protocol originally described by Rokutan et al. (35) and later optimized by us for studying T lymphocytes (38). Briefly, cells are cultured in HBSS (which provides glucose to allow some degree of metabolism, but does not contain any thiol) in the presence of
404
Gianazza, Eberini, and Ghezzi
Table 2 Procedures for 2DE of glutathionylated proteins: solutions for casting an 8-cm-long 4–10 NL IPG slab pH 4 solutiona
pH 10 solutionb
146 μL
pK 0.8c
17 μL
138 μL
pK 4.6d
–
162 μL
pK 6.2d
49 μL
–
pK 7.0d
65 μL
–
pK 8.5d
24 μL
–
pK 9.3d
46 μL
–
1 M AcOHe
15 μL
720 μL
T30 C4f
720 μL
1,060 μL
Glycerol
–
5.4 mL
Water to
5.4 mL
3.3 μL
TEMED
2.86 μL
40% APSg
5 μL
Transfer to a gradient mixer 5 μL a
To the mixing chamber of the gradient mixer To the reservoir of the gradient mixer c From Sigma-Aldrich d From GE-Amersham or Sigma-Aldrich e Prepared with 60 g, or 57.2 mL, glacial acetic acid/L f Prepared with 28.8 g acrylamide and 1.2 g N,N′-bis-acrylamide (BioRad) per 100 mL final volume; to be filtered before use; storage at 4°C; shelf life: few months g Prepared with 200 mg ammonium persulfate and 440 μL ddwater; storage at 4°C; shelf life: 1 week b
a: Coomassie stain
b: fluorography
protein Mr , kDa
1 2
30 3 20
4 5
10
size of 2 nd dimension, cm
70 50 40
6 4
5 6 protein pI, pH units
7
8 9 10
1
2 3 4 5 6 size of 1st dimension, cm
7
8
Fig. 1. 2DE of the extract from T lymphocytes oxidatively stressed with 1 mM diamide. Proteins were separated on the basis of pl (4–10 nonlinear gradient, left to right ) and molecular weight (12% SDS-PAGE, top to bottom). (a) Coomassie blue stain. (b) Fluorography displaying the glutathionylation pattern.
Detection of Protein Glutathionylation
405
high concentrations of cycloheximide to prevent protein synthesis and incorporation of radiolabeled cysteine. Nonincorporated radioactivity is then washed away and glutathionylation stimulated by the addition of diamide, a thiol-specific oxidant (39). 1. Cell number and incubation medium. These experimental conditions have to be tailored to the relevant cell type. In our studies we used 2 × 107 T lymphocytes (7) or 1 × 107 hepatoma cells (6), in HBSS supplemented with 20 mM HEPES, pH 7.4, to strengthen pH buffering and containing 50 μg/mL cycloheximide (“HBSS/cycloheximide”). 2. Radiolabeling. After 20 min in a CO2 incubator, add 8 μCi/ mL of l-[35S] cysteine (specific activity, 1,000 Ci/mmol) to the culture and incubate for 30 min (see Note 1). Then wash the cells with HBSS/cycloheximide (by centrifugation if nonadherent, such as T lymphocytes) and resuspend the pellet in the same solution at the original cell density. 3. Treatment with oxidants. Add diamide (0.2–1.0 mM in the case of T lymphocytes) to the cultures and incubate for 5–10 min (see Note 2). 4. Measurement of incorporated radioactivity. Pellet the cells by centrifugation and resuspend in medium containing 50 mM N-ethylmaleimide to minimize thiol-disulfide exchange (see Note 3). 5. Precipitate the proteins with 5% TCA, wash extensively until no radioactivity is present in the supernatants, and count in a liquid scintillation counter to assess the extent of thiolation. This test can be conveniently performed on small aliquots (1 × 106 cells) while setting up the optimal experimental conditions. Protein Extraction Procedures
The purpose of this treatment is: (1) to acidify the sample to further stabilize disulfide bonds; and (2) to remove nonproteinbound radioactivity for a safe sample handling. For T lymphocytes we successfully applied the following procedure: 1. Resuspend the pellet from 1–2 × 107 cells in 0.2 mL of HBSS. 2. Acidify the suspension with 20 μL of 0.1N HCl; then extract with 1 volume of chloroform, 4 volumes of methanol, and 3 volumes of water. This step will also remove lipids.
Analytical Procedures
3. Wash the proteins at the phase interface with 3 volumes of methanol, air dry, and process for electrophoresis (see Note 4). Day 1: Preparation of the IPG (see Note 5) and of the SDSPAGE slabs. 1. Assemble a 125 × 260 × 0.5 mm3 polymerization cassette using a U-framed silanized glass plate (80-1,106-89 from GEAmersham), a foil of GelBond film, a plain glass plate, and a few clamps.
406
Gianazza, Eberini, and Ghezzi
2. Cast an 8-cm-high IPG 4–10 nonlinear (NL) from the limiting solutions in Table 2. 3. Allow the IPG to polymerize for 1 h at 50°C. 4. Disassemble the cassette and wash the slab for 1 h in 1 L of 1% glycerol, with shaking on an orbital platform. 5. Lay the slab (upside up) on a plain glass plate, put in front of a cool fan (see Note 6), and let dry; this takes 60–90 min, depending on temperature and relative humidity. 6. Reassemble the cassette, fill with reswelling solution, prepared according to Table 3, and degass. 7. Allow the gel to rehydrate overnight at room temperature in a moist box. 8. Polymerize the required number of SDS-PAGE slabs, with size 160 × 140 × 1.5 mm3. Optimal polyacrylamide concentration in the running gel may vary depending on the sample (see Note 7). Day 2: Running the first and second dimensions (see Note 8) A. IEF 1. Turn on the thermostatic unit and set it at 15°C. 2. Cover the cooling plate of the electrophoretic chamber (e.g., Multiphor II, Amersham/GE Healthcare) with Parafilm (see Note 9). 3. Drain excess urea solution from the reswelling cassette. Open the cassette, and – if required – blot dry the surface of the IPG and/or clear from any dust (see Note 10). 4. Cut twelve-millimeter-wide strips in the IPG matrix while leaving a continuous plastic backing (see Note 11).
Table 3 Procedures for 2DE of glutathionylated proteins: reswelling solution for the 4–10 NL IPG slaba Chemicals
Amount
Urea
5.76 g
4–6.5 Pharmalyte
1.2 mL
CA mix
150 μL
4–6 Ampholine 5–7 Ampholine 3–10 Pharmalyte
1.2 mL 1.6 mL 2.0 mL
Water to
12 mL
3.5–10 Ampholine
2.0 mL
a
CA mix composition
To be degassed for overnight reswelling
Detection of Protein Glutathionylation
407
5. Mark the individual lanes on the back with a felt tip for sample identification. 6. Align the slab on the cooling plate; a couple of milliliters of distilled water allow a thorough contact between the surfaces (see Note 12). 7. Cut electrode strips (e.g., 18-1,004-40 from Amersham/ GE Healthcare) to size, moisten in distilled water, and overlay to the gel at its anodic and cathodic edges. 8. Apply the samples NEAR THE ANODE on stacks of Paratex pads (e.g., 80-1,129-46 from Amersham/GE Healthcare) (see Note 13). 9. Align the electrodes to the electrode strips, secure the lid in place, and connect the cables to the power supply. 10. Carry out the focusing according to the scheme (time and voltage drop) in Table 4. B. Interfacing IPG strip to SDS-PAGE gel 1. Take the gel out of the electrophoresis chamber and remove Paratex pads as well as electrode strips after moistening with a stream of distilled water. 2. Equilibrate the samples with 3% SDS in electrode buffer (40) for 15 min, on a shaking platform at room temperature (see Note 14).
Table 4 Procedures for 2DE of glutathionylated proteins: running conditions for the first- and second-d steps of 2DE Protocol
Voltagea
Duration
1d: nonlinear 4–10 IPG
150 V
1h
300 V
1h
550 V
1h
850 V
1 h 30 min
1,500 V
2h
2,000 V
45 min
Protocol
Currentb
Duration
2d: SDS-PAGE on 12%T PAA
50 mA
ca. 2 h
a
With electrode distance of 8 cm Per slab (140 × 160 × 1.5 mm3)
b
408
Gianazza, Eberini, and Ghezzi
3. Trim the lanes corresponding to individual samples by cutting the plastic support flush to one side and approximately 2 mm away from the other. 4. Mount pairs of them head to tail on SDS gels in a layer of agarose (see Note 15). 5. Once the agarose is set, move the SDS slabs with their overlays to the vertical electrophoretic chamber for the SDS-PAGE separation. C. SDS-PAGE The conditions for SDS-PAGE are detailed in Table 4. The run is terminated when the tracking dye has moved 8 cm down the running gel, which takes approximately 2 h (see Note 16). D. Staining Stain the gels overnight with 0.3% Coomassie Brilliant Blue R250 in 30% ethanol-10% acetic acid on an orbital shaker. Day 3 on: Detection procedures E. Destaining Destain the gels in 30% ethanol-10% acetic acid on an orbital shaker. F. Fluorography (see Note 17) 1. Following the manufacturer’s instructions, impregnate the gels with one of the commercial scintillation reagent solutions (Amplify, GE-Amersham; En3Hance, or Enlightening, NEN) (see Note 18) by incubation for 15–60 min on an orbital shaker at room temperature. 2. Soak the gel in 10% glycerol for 1 h on an orbital shaker at 4°C (see Note 19). 3. Lay the gels over a square of Whatman 3-MM paper, cover with Saran wrap, and vacuum dry at 60–80°C. 4. In a dark room, expose the dried gels to appropriate X-ray film (see Note 20). 5. Store the autoradiography cassette at –70°C for several days (see Note 21). 6. Develop the film. Scan film and dried gel for image analysis. G. Phosphorimaging (see Note 16) 1. Blank the phosphor screen on a light source for about 20 min. 2. Carefully seal the wet gels (see Note 21) in Saran wrap and place in the exposure cassette, recording the coordinates of the exposed area. 3. Expose the screen, at room temperature, for some hours.
Detection of Protein Glutathionylation
409
4. Remove the gels from the cassette and scan the phosphor screen with a phosphorimager. 5. Capture the Coomassie pattern with a scanner and analyze the two images. Protein Identification
Protein identification may be carried out along standard procedures based on tryptic digestion and peptide mass fingerprinting, as in our investigations using radiolabeled GSH (7) (see also Chapters “Shotgun Protein Analysis by Liquid Chromatography-Tandem Mass Spectrometry,” and “Tryptic Digestion of in-Gel Proteins for Mass Spectrometry Analysis”). Obviously, as these standard procedures require protein reduction (usually with dithiothreitol (DTT) or tris-(2-carboxyethyl)phosphine (TCEP)) and alkylation (usually with iodoacetamide), it is impossible in this way to detect by MS any glutathionylated peptide since the GSH moiety is removed in the reduction step. To be able to detect glutathionylated peptides nonreducing conditions are needed (41). We successfully carried out tryptic digestion in a solution of 50 mM ammonium bicarbonate in 80% acetonitrile (42). Using nonreducing conditions, it is likely that some targets may not be available for tryptic cleavage, thus resulting in lower peptide coverage. Of note, the alkaline pH required for trypsin digestion will not result in loss of disulfidebound GSH if the sample was previously alkylated. In fact, instability of disulfides in alkaline pH is due to thiol-disulfide exchange with thiols in the sample and may be prevented by alkylating agents (43).
3.3. Pro and Cons of the Different Strategies for Detecting PSSG
The techniques available for the detection of glutathionylated proteins are far behind those for the identification of phosphoproteins, which rely mostly on immunological techniques taking advantage of antibodies that recognize specific phosphopeptides or phosphoaminoacids. This bias is mainly due to the fact that anti-GSH antibodies were generated using an antigen consisting of GSH bound to different amino acids to protein haptens instead of bona fide glutathionylated proteins. On the other hand, the main drawback of techniques based on the incorporation of labeled GSH (such as the one described here in Subheading 3.2.4) is that with this approach we are unable to study or identify basally glutathionylated proteins (i.e., PTM without exposure to oxidants). In fact, proteins highly susceptible to glutathionylation may already be fully glutathionylated under normal conditions and may not incorporate any further GSH when exposed to oxidative stress. Although the reported techniques have greatly improved our knowledge in the field, we believe that current limitations should never be overlooked.
410
Gianazza, Eberini, and Ghezzi
4. Notes 1. The radioactive compound suitable for labeling the intracellular GSH pool varies according to the cell population studied. Lymphocytes have been reported to take up cysteine but not cystine, while macrophages would need cystine instead (44). Radioactive GSH is membrane impermeable and cannot substitute for radioactive cysteine. 2. Depending on the biological problem and cell type, other stimuli may be applied. For instance, we used diamide for T lymphocytes or hepatoma cells (6, 7) but also 0.1–1.0 mM H2O2 for 5 min or, in the case of primary rat hepatocytes, 0.1 mM menadione for 2 h. Concentrations should be adjusted to prevent toxic effects. Short exposure times (5–10 min) are preferred for this reason and also to minimize removal of the label by dethiolation (38). 3. Protein-bound GSH is rapidly lost by dethiolation (38), which is largely dependent on thiol-disulfide exchange with residual protein thiols and thus can be effectively prevented by alkylation (43). 4. This protein extraction procedure works well with lymphocytes but if a different protein preparation turned insoluble, then alternative procedures should be considered (e.g., in ref (6)). 5. The IPG slabs are home-made according to the recipe in (45) and processed essentially as in (46). Using home-made slabs (47) makes it possible to cut IPG strips of any width and to load a large sample volume and/or a high protein amount per strip. It also facilitates control of the pH of the sample loading area, avoiding or minimizing exposure of redox PTM proteins to alkaline buffers. 6. The area where this procedure is carried out must be CLEAN. 7. Refer to e.g., (48) for selection of the experimental conditions for second-dimension gels and details on SDS-PAGE slab polymerization. The requirement for gel drying before phosphorimaging prevents the use of gradient gels for the second dimension. A 2-cm-high stacking gel helps narrowing the protein spots migrating from a wide first-dimension strip. 8. Wear disposable powder-free GLOVES throughout the following procedures. All disposable plasticware and all liquids must be collected for a check of their radioactive contamination before disposing of them appropriately: in our experience, however, the radioactive count in waste material was NEVER above background.
Detection of Protein Glutathionylation
411
9. To avoid possible contamination by radioactive material, two stretches of Parafilm, 25-cm long, are aligned side by side at the center without overlap, and made lay flat with distilled water. Excess liquid is blotted from the edges. 10. WET, low-linting paper wipes should be used. 11. The gel slab is aligned on millimeter paper xeroxed on a transparency, stuck by the edges with double-side tape (49). Gel is removed, anode to cathode, by cutting through with a long, straight knife, then sliding it against the plastic support, to leave 12-mm-wide strips for sample application. At least eight such strips, 8 mm apart, may be obtained from a standard slab. It is advised to leave wide gel strips along the edges, in order to prevent drying of the sample lanes during the run. Shreds of gel not removed by the knife are scraped with a straight scalpel blade. Cutting the strips should be carried out, and the slab moved to the electrophoresis chamber, as quickly as possible to avoid drying out and urea crystallization. 12. In this step, the slab is held by the edges (top and bottom), bent backward, and carefully lowered in order for the water to spread without getting above the plastic support. Paper tissues are lined up to the anodic and cathodic edges to collect excess liquid. 13. One such pad holds up to 25 μL; up to 4–5 layers may be overlaid. To prevent sample losses to the anodic strip, the stack of dry pads is aligned 3–4 mm from the strip, the sample solution is pipetted slowly onto it (from the cathodic rim), and the wet stack is finally pushed (with the same pipette tip) 1–2 mm from the strip. 14. Depending on the size of the gel and on the size of the container, 50–100 mL is required. A saturated solution of bromophenol blue is added, at a 1:100 ratio, to provide for a tracking dye. 15. A solution of 0.7% low electroendoosmosis agarose in electrode buffer is prepared about 30 min in advance by boiling the suspension in a water bath. The mouth of the flask is loosely covered with a smaller, inverted flask or with a piece of aluminum foil; the suspension is mixed thoroughly on a stirring, heating plate. The solution is then maintained at 60–65°C either by transferring the flask to a thermostatted bath or by lowering the setting of the heating plate. Any liquid above the SDS-PAGE slabs is removed by inverting them on a towel. Agarose sol is poured in the rim above the gel; the individual sample strips are inserted promptly in the warm solution, flush side down, and aligned with a flat
412
Gianazza, Eberini, and Ghezzi
spatula. Agarose setting takes some time, so that it is easy to reorient the strips, in case a mistake had occurred. 16. This selection allows maintaining an approximately square area for spot resolution, avoiding dispersion of the radioactive signal over a wide region, preventing electrophoresis of the anionic front into the lower buffer compartment, and cuts running time to a minimum. With the chosen setup the area available for protein resolution in the running gel is higher than with a Midget apparatus, letting also room for a thick stacking gel layer, see earlier. 17. Fluorography and phosphorimaging are alternative options for detecting 35S-labeled macromolecules. Phosphorimaging is between 2 and 10 times more sensitive than autoradiography, its output being linear within four logs. Phosphorimaging does not require chemicals or consumables, but the price of equipment and screens is extremely high. 18. As an alternative to commercial products, 2,5-diphenyloxazole (PPO) may be used as a 20% solution in acetic acid according to Skinner and Griswold (50). 19. The fluor precipitates and the gels turn opaque white. 20. Intensifying screens are marginally effective with low-energy isotopes such as 35S. 21. The minimal activity detected after 24 h exposure in a 0.1-cm2-band is 400 Beq (51). 22. For longer exposure times, it is advised to dry the gels.
Acknowledgments Work supported in part by AFM and by MIUR (FIRB 2003), to EG, and Fondazione CARIPLO, Milano (to PG).
References 1. 2.
3.
Thornton, J.M. (1981) Disulphide bridges in globular proteins. J Mol Biol 151, 261–287. Rietsch, A., and Beckwith, J. (1998) The genetics of disulfide bond metabolism. Annu Rev Genet 32, 163–184. Fahey, R.C., Hunt, J.S., and Windham, G.C. (1977) On the cysteine and cystine content of proteins: differences between intracellular and extracellular proteins. J Mol Evol 10, 155–160.
4.
5.
Saurin, A.T., Neubert, H., Brennan, J.P., and Eaton, P. (2004) Widespread sulfenic acid formation in tissues in response to hydrogen peroxide. Proc Natl Acad Sci U S A 101, 17982–17987. Brennan, J.P., Wait, R., Begum, S., Bell, J.R., Dunn, M.J., and Eaton, P. (2004) Detection and mapping of widespread intermolecular protein disulfide formation during cardiac oxidative
Detection of Protein Glutathionylation
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
stress using proteomics with diagonal electrophoresis. J Biol Chem 279, 41352–41360. Fratelli, M., Demol, H., Puype, M., Casagrande, S., Villa, P., Eberini, I., et al. (2003) Identification of proteins undergoing glutathionylation in oxidatively stressed hepatocytes and hepatoma cells. Proteomics 3, 1154–1161. Fratelli, M., Demol, H., Puype, M., Casagrande, S., Eberini, I., Salmona, M., et al. (2002) Identification by redox proteomics of glutathionylated proteins in oxidatively stressed human T lymphocytes. Proc Natl Acad Sci U. S. A. 99, 3505–3510. Ghezzi, P., Bonetto, V., and Fratelli, M. (2005) Thiol-disulfide balance: From the concept of oxidative stress to that of redox regulation. Antioxid Redox Signal 7, 964–972. Bursell, S.E., and King, G.L. (2000) The potential use of glutathionyl hemoglobin as a clinical marker of oxidative stress. Clin Chem 46, 145–146. Johnson, R.M., Buck, S., Schneider, H., Sampaio, I., Gage, D.A., Shen, T.L., et al. (1997) Identification of pre-γ-globin. Hemoglobin 21, 143–153. Mawatari, S., and Murakami, K. (2004) Different types of glutathionylation of hemoglobin can exist in intact erythrocytes. Arch Biochem Biophys 421, 108–114. Murakami, K., and Mawatari, S. (2003) Oxidation of hemoglobin to methemoglobin in intact erythrocyte by a hydroperoxide induces formation of glutathionyl hemoglobin and binding of α-hemoglobin to membrane. Arch Biochem Biophys 417, 244–250. Naito, C., Kajita, M., and Niwa, T. (1999) Determination of glutathionyl hemoglobin in hemodialysis patients using electrospray ionization liquid chromatography-mass spectrometry. J Chromatogr B Biomed Sci Appl 731, 121–124. Niwa, T., Naito, C., Mawjood, A.H., and Imai, K. (2000) Increased glutathionyl hemoglobin in diabetes mellitus and hyperlipidemia demonstrated by liquid chromatography/electrospray ionization-mass spectrometry. Clin Chem 46, 82–88. Naito, C., and Niwa, T. (2000) Analysis of glutathionyl hemoglobin levels in diabetic patients by electrospray ionization liquid chromatography-mass spectrometry: Effect of vitamin E administration. J Chromatogr B Biomed Sci Appl 746, 91–94. Takayama,F.,Tsutsui,S.,Horie,M.,Shimokata,K., and Niwa, T. (2001) Glutathionyl hemoglobin in uremic patients undergoing hemodialysis and continuous ambulatory peritoneal dialysis. Kidney Int Suppl 78, S155–S158.
413
17. Piemonte,F.,Pastore,A.,Tozzi,G.,Tagliacozzi,D., Santorelli, F.M., Carrozzo, R., et al. (2001) Glutathione in blood of patients with Friedreich’s ataxia. Eur J Clin Invest 31, 1007–1011. 18. Sampathkumar, R., Balasubramanyam, M., Sudarslal, S., Rema, M., Mohan, V., and Balaram, P. (2005) Increased glutathionylated hemoglobin (HbSSG) in type 2 diabetes subjects with microangiopathy. Clin Biochem 38, 892–899. 19. Pastore, A., Mozzi, A.F., Tozzi, G., Gaeta, L.M., Federici, G., Bertini, E., et al. (2003) Determination of glutathionyl-hemoglobin in human erythrocytes by cation-exchange highperformance liquid chromatography. Anal Biochem 312, 85–90. 20. Al-Abed, Y., VanPatten, S., Li, H., Lawson, J.A., FitzGerald, G.A., Manogue, K.R., and Bucala, R. (2001) Characterization of a novel hemoglobin-glutathione adduct that is elevated in diabetic patients. Mol Med 7, 619–623. 21. Biroccio, A., Urbani, A., Massoud, R., di Ilio, C., Sacchetta, P., Bernardini, S., et al. (2005) A quantitative method for the analysis of glycated and glutathionylated hemoglobin by matrix-assisted laser desorption ionizationtime of flight mass spectrometry. Anal Biochem 336, 279–288. 22. Caplan, J.F., Filipenko, N.R., Fitzpatrick, S.L., and Waisman, D.M. (2004) Regulation of annexin A2 by reversible glutathionylation. J Biol Chem 279, 7740–7750. 23. Sadidi, M., Geddes, T.J., and Kuhn, D.M. (2005) S-thiolation of tyrosine hydroxylase by reactive nitrogen species in the presence of cysteine or glutathione. Antioxid Redox Signal 7, 863–869. 24. Chen, F.C., and Ogut, O. (2006) Decline of contractility during ischemia-reperfusion injury: Actin glutathionylation and its effect on allosteric interaction with tropomyosin. Am J Physiol Cell Physiol 290, C719–C727. 25. Clavreul, N., Bachschmid, M.M., Hou, X., Shi, C., Idrizovic, A., Ido, Y., et al. (2006) S-glutathiolation of p21ras by peroxynitrite mediates endothelial insulin resistance caused by oxidized low-density lipoprotein. Arterioscler Thromb Vasc Biol 26, 2454–2461. 26. Pastore, A., Tozzi, G., Gaeta, L.M., Bertini, E., Serafini, V., Di Cesare, S., et al. (2003) Actin glutathionylation increases in fibroblasts of patients with Friedreich’s ataxia: a potential role in the pathogenesis of the disease. J Biol Chem 278, 42588–42595. 27. Townsend, D.M., Findlay, V.J., Fazilev, F., Ogle, M., Fraser, J., Saavedra, J.E., et al. (2006) A glutathione S-transferase pi-activated
414
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
Gianazza, Eberini, and Ghezzi prodrug causes kinase activation concurrent with S-glutathionylation of proteins. Mol Pharmacol 69, 501–508. Cheng, G., Ikeda, Y., Iuchi, Y., and Fujii, J. (2005) Detection of S-glutathionylated proteins by glutathione S-transferase overlay. Arch Biochem Biophys 435, 42–49. Eaton, P., Wright , N., Hearse, D.J., and Shattock, M.J. (2002) Glyceraldehyde phosphate dehydrogenase oxidation during cardiac ischemia and reperfusion. J Mol Cell Cardiol 34, 1549–1560. Niture, S.K., Velu, C.S., Bailey, N.I., and Srivenugopal, K.S. (2005) S-thiolation mimicry: quantitative and kinetic analysis of redox status of protein cysteines by glutathioneaffinity chromatography. Arch Biochem Biophys 444, 174–184. Lind, C., Gerdes, R., Hamnell, Y., SchuppeKoistinen, I., von Löwenhielm, H.B., Holmgren, A., and Cotgreave, I.A. (2002) Identification of S-glutathionylated cellular proteins during oxidative stress and constitutive metabolism by affinity purification and proteomic analysis. Arch Biochem Biophys 406, 229–240. Chai, Y.C., Hoppe, G., and Sears, J. (2003) Reversal of protein S-glutathiolation by glutaredoxin in the retinal pigment epithelium. Exp Eye Res 76, 155–159. Eaton, P., Byers, H.L., Leeds, N., Ward, M.A., and Shattock, M.J. (2002) Detection, quantitation, purification, and identification of cardiac proteins S-thiolated during ischemia and reperfusion. J Biol Chem 277, 9806–9811. Sullivan, D.M., Wehr, N.B., Fergusson, M.M., Levine, R.L., and Finkel, T. (2000) Identification of oxidant-sensitive proteins: TNF-α induces protein glutathiolation. Biochemistry 39, 11121–11128. Rokutan, K., Thomas, J.A., and Johnston, R.B. Jr. (1991) Phagocytosis and stimulation of the respiratory burst by phorbol diester initiate S-thiolation of specific proteins in macrophages. J Immunol 147, 260–264. Thomas, J.A., Park, E.M., Chai, Y.C., Brooks, R., Rokutan, K., and Johnston, R.B. Jr. (1991) S-thiolation of protein sulfhydryls. Adv Exp Med Biol 283, 95–103. Sanchez, J.-C., and Hochstrasser, D.F. 1998. High-resolution,IPGbased,minitwo-dimensional gel electrophoresis. In 2-D Proteome Analysis Protocols. (ed. A.J. Link). Humana Press, Totowa, NJ, pp. 227–233. Ghezzi, P., Romines, B., Fratelli, M., Eberini, I., Gianazza, E., Casagrande, S., et al. (2002) Protein glutathionylation: coupling and uncoupling of glutathione to protein thiol groups in
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
lymphocytes under oxidative stress and HIV infection. Mol Immunol 38, 773–780. Kosower, N.S., and Kosower, E.M. (1987) Formation of disulfides with diamide. Methods Enzymol 143, 264–270. Laemmli, U.K. (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680–685. Russell, W.K., Park, Z.Y., and Russell, D.H. (2001) Proteolysis in mixed organic-aqueous solvent systems: applications for peptide mass mapping using mass spectrometry. Anal Chem 73, 2682–2685. Casagrande, S., Bonetto, V., Fratelli, M., Gianazza, E., Eberini, I., Massignan, T., et al. (2002) Glutathionylation of human thioredoxin: a possible crosstalk between the glutathione and thioredoxin systems. Proc Natl Acad Sci U. S. A. 99, 9745–9749. Rossi, R., Cardaioli, E., Scaloni, A., Amiconi, G., and Di Simplicio, P. (1995) Thiol groups in proteins as endogenous reductants to determine glutathione-protein mixed disulphides in biological systems. Biochim Biophys Acta 1243, 230–238. Gmunder, H., Eck, H.P., Benninghoff, B., Roth, S., and Droge, W. (1990) Macrophages regulate intracellular glutathione levels of lymphocytes: evidence for an immunoregulatory role of cysteine. Cell Immunol 129, 32–46. Gianazza, E., Giacon, P., Sahlin, B., and Righetti, P.G. (1985) Non-linear pH courses with immobilized pH gradients. Electrophoresis 6, 53–56. Gianazza, E. 2002. Casting immobilized pH gradients. In The Protein Protocol Handbook, 2nd ed. (ed. J.M. Walker). Humana Press, Totowa, NJ, pp. 169–180. Gianazza, E., Rocco, A.G., Marchetto, A., and Vergani, L. (2007) Immobilized pH gradients with electrodic plateaus (and other unusual procedures for 2DE). Electrophoresis 28, 2953–2956. Hames, B.D. (1990) One-dimensional polyacrylamide gel electrophoresis. In Gel Electrophoresis of Proteins: A Practical Approach. (eds. B.D. Hames, and D. Rickwood). IRL, Oxford, pp. 1–147. Gianazza, E., Astrua-Testori, S., Giacon, P., and Righetti, P.G. (1985) An improved protocol for 2D maps of serum proteins with immobilized pH gradients in the first dimension. Electrophoresis 6, 332–339. Skinner, M.K., and Griswold, M.D. (1983) Fluorographic detection of radioactivity in polyacrylamide gels with 2,5-diphenyloxazole in acetic acid and its comparison with existing procedures. Biochem J 209, 281–284.
Detection of Protein Glutathionylation 51. Laskey, R.A. (1980) The use of intensifying screens or organic scintillators for visualizing radioactive molecules resolved by gel electrophoresis. Methods Enzymol 65, 363–371. 52. de Souza, G.A., Godoy, L.M., Teixeira, V.R., Otake, A.H., Sabino, A., Rosa, J.C., Dinarte,
415
A.R., Pinheiro, D.G., Silva, W.A., Jr., Eberlin, M.N., et al. (2006) Proteomic and SAGE profiling of murine melanoma progression indicates the reduction of proteins responsible for ROS degradation. Proteomics 6, 1460–1470.
Chapter 27 Activity-Based Protein Profiling of Protein Tyrosine Phosphatases Chad Walls, Bo Zhou, and Zhong-Yin Zhang Summary The ability to accurately monitor the dynamics involved with the activity and state of a specific protein population in a complex biological system represents one of the major technological challenges in studying systems biology. Over the past several years a number of groups have attempted to spearhead this new frontier of systems biology by developing enzyme family-specific activity-based chemical probes linked to appropriate reporter groups that by nature only target and subsequently tag the active form of these enzymes. In this work, we will highlight the methods used to characterize activity-based probes as to their utility in biological contexts. Specifically, we will address activity-based protein profiling of the protein tyrosine phosphatases, a highly conserved enzyme family responsible for the phospho-tyrosine hydrolysis reaction, a ubiquitous reaction that is absolutely essential to the regulation of a myriad of cellular processes. Key words: Activity-based protein profiling (ABPP), Activity-based probes (ABPs), Protein tyrosine phosphatases (PTPs), Biotin probe, Fluorescent probe, Systems biology.
1. Introduction A more detailed look into cellular processes as it pertains to global alterations to protein function in a complex web of interactions is the paradigm of “systems biology.” To date, being able to understand a specific protein population with regard to its distinguishable traits in a dynamic setting such as subgroups that are being positively or negatively regulated (e.g., inactive vs. active states) rather than just the mere observation of natural
David Sheehan and Raymond Tyther (eds.), Methods in Molecular Biology, Two-Dimensional Electrophoresis Protocols, vol. 519 © Humana Press, a part of Springer Science + Business Media, LLC 2009 DOI: 10.1007/978-1-59745-281-6_27
417
418
Walls, Zhou, and Zhang
abundance remains a formidable task both technologically and methodologically. Activity-based protein profiling (ABPP) represents a relatively new concept in the era of further defining the proteome on a “systems-wide” scale as more than just quantifiable differences in protein expression levels, a methodology currently employed solely by mass spectrometry (MS). ABPP attempts to address, for the first time, protein function in a dynamic setting, thus allowing for a specific protein population to be characterized as to both its natural abundance and its activity; two dimensions that do not necessarily correlate with each other biologically. Over the past several years a number of groups have attempted to spearhead this new frontier of “system’s biology” by developing enzyme family-specific activity-based chemical probes (ABPs) linked to specific reporter groups that by nature only target and subsequently tag the active form of these enzymes both in vitro and in vivo. The enzyme families that have been studied using this ABPP approach are the serine hydrolases (1–3), metalloproteinases (4), protein tyrosine phosphatases (5, 6), 26S proteasome (7), and protein and lipid kinases (8). Protein tyrosine phosphatases (PTPs) represent one of two extremely critical and evolutionarily conserved enzyme families responsible for the reversible tyrosine phospho-transfer reactions that have defined the life processes of multicellular organisms for over 600 million years. PTPs catalyze the dephosphorylation reaction of phospho-tyrosine residues on substrate proteins and thus represent important signaling enzymes that serve as key regulatory components in various signal transduction pathways (9, 10). Defective or inappropriate regulation of PTP activity leads to aberrant tyrosine phosphorylation, which contributes to the development of many human diseases (11). The sequencing of the human genome has revealed more than 100 PTPs (12) of which little is known about their physiological role and dynamic regulation despite having a very detailed understanding of the mechanism by which they carry out the phospho-tyrosine hydrolysis reaction (13). Knowing the details of this catalytic mechanism provides a major victory toward the development of chemical probes that specifically target and irreversibly modify the nucleophilic cysteine residue in the PTP signature motif ((H/V)C(X)5R(S/T). All ABPs must undergo rigorous testing as to their utility in covalently labeling the active forms of their respective targets both in a simplistic environment (e.g., probe with purified target protein) and a more complex one (e.g., a cell lysate or in vivo). In this work, the general methodology that remains consistent to all these endeavors will be highlighted as it pertains to the
Activity-Based Protein Profiling of Protein Tyrosine Phosphatases
419
O P OH OH
Br O OH OH
P Br
NH O
O (CH2)4
NH
H
O
H N
S
SO3H O
H
I
S
N H
N
O
N
II
Fig. 1. Structures of the biotinylated PTP α-Bromobenzylphosphonate probe I and the rhodamine-conjugated PTP α-Bromobenzylphosphonate probe II.
characterization and experimental utility of both the α-bromobenzylphosphonate (BBP)-based biotin- and fluorophore-conjugated PTP probes (I and II respectively; Fig.1) (5, 6).
2. Materials 2.1. Kinetic Characterization of the a-Bromobenzylphosphonate PTP Probes I and II
1. PTP Probes I and II. Prepare as 100 mM stocks in 100% DMSO, store at −20°C. 2. Yersinia PTP (YopH). Expressed and purified from Escherichia coli (14). 3. Labeling buffer. 50 mM sodium succinate, 1 mM EDTA, 1 mM dithiothreitol (DTT), 150 mM NaCl, pH 6.0 (see Note 1). 4. PTP assay buffer. 50 mM sodium succinate, 1 mM EDTA, 1 mM DTT, 150 mM NaCl, pH 6.0, supplemented with 20 mM pNPP (200 μL total volume).
2.2. Specificity of the a-Bromobenzylphosphonate PTP Probes I and II
1. Labeling reaction buffer for the PTP enzymes. 50 mM sodium succinate buffer (pH 6.0) containing 1 mM EDTA and 1 mM dithiothreitol (DTT). Adjust the ionic strength to 150 mM with NaCl (see Note 1). 2. PTP enzymes. PTP1B, SHP2, HePTP, YopH, and FAP-1 (classical cystolic PTPs), PTPα and DEP-1 (receptor-like PTP),
420
Walls, Zhou, and Zhang
VHR, PRL-3, and Cdc14 (dual-specificity phosphatases), and LMW-PTP. 3. Labeling reaction buffers for all non-PTP enzymes: (a) Potato and prostatic acid phosphatases, 100 mM sodium acetate, pH 5.0, 1 mM EDTA. (b) Protein phosphatase 1 (PP1) and λ-phosphatase, 50 mM 3, 3-dimethylglutarate, pH 7.0, 2 mM MnCl2. (c) Alkaline phosphatase, 50 mM Tris–HCl, pH 9.0, 1 mM MgCl2. (d) Papain, lysozyme, Grb2-Src homology 2 (SH2), Shc-phosphotyrosine-binding (PTB), and SNT1-PTB, 50 mM 3,3-dimethylglutarate, pH 7.0, 1 mM EDTA. (e) Glyceraldehyde-3-phosphate dehydrogenase, 15 mM sodium pyrophosphate, pH 8.5, 7.5 mM NAD, 1 mM DTT. (f) Calpain, 100 mM imidazole, pH 7.3, 10 mM CaCl2, 1 mM DTT. (g) Trypsin, 100 mM Tris–HCl, pH 8.5, 1 mM EDTA. (h) Chymotrypsin, 50 mM Tris–HCl, pH 7.8, 50 mM CaCl2. (i) Protein phosphatase 2B, 50 mM 3,3-dimethylglutarate, pH 7.0, 0.3 μM calmodulin, 2 mM MnCl2. (j) Src kinase, 100 mM Tris–HCl, pH 7.0, 1 mM EDTA. (k) Thermolysin, 50 mM HEPES, pH 7.0, 5 mM CaCl2. 4. Quenching solution. SDS-loading buffer (sample buffer) (62.5 mM Tris–HCl, pH 6.8, 2.5% SDS, 10% glycerol, 10 mM DTT, 0.2% bromophenol blue. 5. SDS-PAGE buffers and reagents as highlighted in the SDSPAGE materials section. 6. (For use if using the biotin-conjugated PTP probe I ) – Western blotting reagents as highlighted in Subheading 2.8. 7. (For use if using the fluorophore-conjugated PTP probe II ) – Typhoon 9400 scanner (Amersham Biosciences, Inc.) in fluorescence mode (green: 532 nm; red: 633 nm) for detection of the fluorescent probes. ImageQuant 3.3 (Molecular Dynamics) for fluorescence quantification. 8. (For use if using the fluorophore-conjugated PTP probe II ) – Silver staining reagents as described in Subheading 2.7. 2.3. Sensitivity of the a-Bromobenzylphosphonate PTP Probes I and II
1. PTP Probes I and II. Prepare as 100 mM stocks in 100% DMSO and store at −20°C. 2. Yersinia PTP (YopH). Expressed and purified from E. coli (14) (see Note 2). 3. Labeling buffer. 50 mM sodium succinate, 1 mM EDTA, 1 mM DTT, 150 mM NaCl, pH 6.0 (see Note 1).
Activity-Based Protein Profiling of Protein Tyrosine Phosphatases
421
4. Quenching solution. SDS-loading buffer (sample buffer); 62.5 mM Tris–HCl, pH 6.8, 2.5% SDS, 10% Glycerol, 10 mM DTT, 0.2% bromophenol blue. 2.4. Effect of H2O2 on PTP Activity Monitored by the a-Bromobenzylphosphonate PTP Probes I and II both in vitro and in vivo
1. Labeling buffer. 50 mM sodium succinate, 1 mM EDTA, 1 mM DTT, and 150 mM NaCl, pH 6.0 (see Note 1). 2. Hydrogen peroxide (H2O2) (Sigma). 3. PTP1B. Expressed and purified from E. coli (15). 4. Human-colon-carcinoma-derived HCT-116 cell culture. 5. SDS-PAGE, gel staining, cell culture, lysis, and protein assay reagents and buffers as described in their respective materials sections. Fluorescence scanning as described previously. Note: supplement lysis buffer with 2 mM PTP probe.
2.5. PTP Profiling in Cancer Cells Monitored the a-Bromobenzylphosphonate PTP Probes I and II
2.6. Cell Culture, Lysis, and Protein Assay
1. MOA231 (lung), MDA-MB-435S (breast), MCF-7 (breast), HepG2 (liver), SKOV3 (ovary), HeLa (cervices), and HCT-116 (colon) cell cultures. 2. Labeling buffer. 50 mM sodium succinate, 1 mM EDTA, 1 mM DTT, and 150 mM NaCl, pH 6.0 (see Note 1). 3. Cell culture, lysis, SDS-PAGE, and staining reagents and buffers as described in their respective materials sections. Fluorescence scanning as described previously. 1. Dulbecco’s Modified Eagle’s minimum essential Medium (DMEM) (Gibco/BRL, Bethesda, MD) supplemented with 10% fetal bovine serum (Life Technologies, Inc.), penicillin (50 units/mL), streptomycin (50 μg/mL), and L-glutamine (2 mM). 2. Dulbecco’s Phosphate-Buffered Saline (DPBS; 10×) (Gibco/ BRL, Bethesda, MD) made 1× before using as cell washsolution. 3. Cell lysis buffer. 50 mM MES, pH 6.0, 150 mM NaCl, 1.0% Triton X-100, 0.1% SDS, 10% Glycerol, 5 mM EDTA, supplemented with fresh 10 μg/mL leupeptin, 5 μg/mL aprotinin, 1 mM PMSF, and 1 mM DTT. 4. Teflon cell scrapers (Fisher). 5. Bradford Protein Assay (Amersham Biosciences).
2.7. SDS-Polyacrylamide Gel Electrophoresis (SDS-PAGE)
1. Running Buffer (5×). 125 mM Tris-HCl, 960 mM glycine, 0.5% SDS. Store at room temperature. 2. Thirty percent (30%) acrylamide/bis solution. 37.5:1 with 2.6% C (Fisher). 3. N,N,N,N ¢-Tetramethyl-ethylenediamine (TEMED, Fisher).
422
Walls, Zhou, and Zhang
4. Ammonium persulfate (APS). Prepare a fresh 10% solution in water. Immediately freeze down small single-use aliquots at −20°C (do not store these aliquots for long). 5. Resolving gels. 10 and 12.5% SDS-PAGE gels. 6. SDS-loading buffer (sample buffer). 62.5 mM Tris–HCl, pH 6.8, 2.5%SDS, 10% glycerol, 10 mM DTT, 0.2% bromophenol blue. 7. EZ-Run Protein Standard (Prestained molecular weight marker) (Fisher). 8. GelCode Blue Stain Reagent (Pierce). 9. SilverQuest (Silver Staining Kit) (Invitrogen). 2.8. Western Blotting
1. Transfer buffer (10×). 30 g Tris-HCl, 144 g Glycine, 10 g SDS. Dilute to 1× before use. 2. NitroBind, Cast, Pure Nitrocellulose 0.22 μm (GE Water and Process Technologies). 3. Whatman Chromatography Paper (Schleicher & Schuell). 4. TBS-T. Tris-buffered saline with 1% Tween 20 (TBS-T). 5. Blocking buffer. 5% nonfat dry milk in TBST. 6. Primary antibody dilution buffer (for antibiotin western blot). TBST-containing nonfat dry milk (5%). Primary antibody (antibiotin-horseradish peroxidase (HRP) conjugate (Cell Signaling Technology, Beverly, MA)). 7. HRP substrate (ECL Western Blotting Detection Reagents) (Amersham Biosciences).
3. Method In this work, we will highlight the characterization and utility of both the α-Bromobenzylphosphonate (BBP)-based biotin- and fluorophore-conjugated protein tyrosine phosphatase (PTP) probes (5, 6). (BBP) has been shown to be a quiescent affinity inactivator of the Yersinia PTP YopH (16). Figure 1 illustrates both the biotin- (I) and fluorophore- (II) conjugated PTP probes. The biotin conjugate allows for target protein enrichment from a much more complex matrix via affinity-based purification methodology (an approach that, theoretically, is much better suited for later protein identification methodologies like MS). Contrary to this asset of target protein enrichment via affinity-based purification is that detection and identification can only be done after affinity purification, gel electrophoresis, and western blot (all approaches that act to decrease protein yield in preparation for
Activity-Based Protein Profiling of Protein Tyrosine Phosphatases
423
direct visualization or MS). These inherent liabilities make this approach far less sensitive toward visualization when compared to its fluorophore-based counterpart. The fluorescent probe allows for “direct” in-gel visualization of target protein conjugate and, through a number of studies, has proven to be far more sensitive than the biotin-based approach. However, the fluorophore-based approach toward protein isolation for subsequent MS analysis, mechanistically, is more difficult than the biotin-based approach (from a complex matrix-type setting) because target protein must be cut from a gel among a myriad of total cellular protein run at the same time. Both probes are ideal when used to attain goals that base their achievement off of the assets inherent to each probe’s specific chemistry. The methodology used to characterize these probes in a biological setting will be described here. 3.1. Kinetic Characterization of the a-Bromobenzylphosphonate PTP Probes I and II
The Yersinia PTP YopH provides an excellent model to explore the in vitro kinetic mechanism of probe-induced enzyme inactivation as well as relevant kinetic constants associated with the probe’s concentration-dependent inhibitory nature against PTP activity toward “synthetic” or “biological” substrates. Previous work in ABPP has shown that ABPs display differential probe reactivity toward members of the same enzyme family. It is crucial that the probe display inactivating properties that are “general” to the entire family and not “specific” toward any particular member of the group. Para-nitrophenyl phosphate (pNPP) is a widely used molecule in the assessment of PTPase activity. The rate of hydrolysis of this molecule directly correlates with the competence of the enzyme for catalytic activity. We have previously shown that both probes I and II display very similar inactivating kinetics toward the standard PTP YopH (a necessary observation showing that the probe conjugates are relatively inert toward reaction progress) (5, 6). 1. Initiate the labeling/inactivation reaction by adding a 5-μL aliquot of 1 μM YopH stock (kept on ice) to a 45-μL solution containing appropriately diluted probe (0.2, 0.5, 1.0, 1.5, 2.0, and 3.0 mM) (make sure to keep the final DMSO concentration at 5% for all concentrations to minimize experimental error) at 25°C. 2. At appropriate time intervals, remove aliquots of 2 μL from the reaction and subsequently add them to 200 μL of the PTP assay buffer highlighted in the Subheading 2 at 30°C. 3. After 3 min, quench each pNPP reaction by adding 50 μL of 5N NaOH. Transfer 200 μL of the quenched reaction mixture to a 96-well plate and measure the absorbance at 405 nm (At), which reflects the residual activity of YopH in the labeling reaction at time point taken. 4. Obtain the pseudofirst-order inactivation constants (kobs) at various concentrations of the probe by fitting the data to
424
Walls, Zhou, and Zhang
Eq. 1: where A0, At, and A∞ are activities of the enzyme at incubation time zero, t, and infinity, respectively. At A∞ ⎛ A0 − A∞ ⎞ − kobs ·t = − e . A0 A0 ⎜⎝ A0 ⎟⎠
(1)
5. Perform a nonlinear regression fit of kobs vs. probe concentration to Eq. 2. This will yield the maximal inactivation constant ki and dissociation constant Ki. kobs = 3.2. Specificity of the a-Bromobenzylphosphonate PTP Probes I and II
ki × [I ]
. K I + [I ]
(2)
A measure of the utility of an ABP is its ability to discriminate successfully not only between PTP and non-PTP enzymes, but also between a PTP enzyme that is in an active conformation (one exposing the active site for catalysis) and a PTP enzyme that is, by any means, in an inactive state. Furthermore, the probe must engage the active site chemistry in such a way that the product of the reaction is a fully inactive and covalently bound enzyme. The PTP family of enzymes share a highly conserved signature motif ((H/V)C(X)5R(S/T), with the cysteine residue being the active site nucleophile responsible for the initiation of the phospho-tyrosine hydrolysis reaction. The specificity of the probe relies on its recognition of this particular active site architecture. A very important test is that it distinguishes not only between PTP and non-PTP phosphatases, but also between the PTP signature motif cysteine residue and other non-PTP enzymes that also initiate catalysis via a nucleophilic reaction involving a cysteine residue. 1. Incubate the PTP probes (1 mM for the biotin-based probe I) (0.2 mM for the fluorophore-based probe II) with the pre-equilibrated PTPs (or non-PTP enzymes) (25 μM for the biotin-based probe I) (1 μM for the fluorophore-based probe II) at 25°C for 15 min. All enzymes and appropriate buffers are described in Subheading 2. 2. After the incubation time period, prepare the enzyme samples to be resolved in a gel matrix using SDS-PAGE (appropriate buffers are highlighted in Subheading 2). First, quench the labeling reaction by adding one volume of the quenching solution (see Subheading 2.2.). 3. For use if using the biotin-conjugated PTP probe I ) – Divide each sample into halves (one-half goes to one 12.5% SDS-PAGE gel for later staining and the other half to another 12.5% SDSPAGE gel for later Western blot). Load 3 μg protein per lane. 4. After electrophoresis perform a western blot by transferring the proteins from one gel to a nitrocellulose membrane overnight at 4°C (see Chapter “Immunoblotting 2DE Membranes”). Use the appropriate transfer, blocking and washing buffers as detailed in Subheading 2.8. Treat the membrane for 2 h
Activity-Based Protein Profiling of Protein Tyrosine Phosphatases
425
with antibiotin-horseradish peroxidase (HRP) conjugate at a dilution of (1:1,000) in TBS-T-containing 5% nonfat dry milk at 25°C. Treat the antibiotin-HRP-treated blot with HRP substrate for 1 min. Expose the chemiluminescent membrane blot to X-ray film in a dark room. 5. Stain the other gel with Coomassie blue for protein visualization following the appropriate manufacturer’s recommendations. 6. (For use if using the fluorophore-conjugated PTP probe II ) – Immediately scan the gel (after the gel run is completed) containing the fluorescently labeled enzymes using a Typhoon 9400 scanner (Amersham Biosciences) in fluorescent mode, with green (532 nm) and red (633 nm) lasers with preset optimum emission filters. Quantify the scanned images with ImageQuant 3.3 (Molecular Dynamics) (see also Chapter “Troubleshooting Image Analysis in 2DE”). 7. After fluorescence scanning, silver stain the gel (according to the manufacturer’s instructions) to directly visualize equal protein loading between the lanes. 3.3. Sensitivity of the a-Bromobenzylphosphonate PTP Probes I and II
It has been shown in a number of studies, including our own (6), that the fluorophore-based approach is far more sensitive (in the order of 1,000-fold more sensitive) than the affinity-based biotin approach. This phenomenon is due to a number of reasons including the physical and chemical nature of fluorescence as well as the biotin method inherently being deleterious to the preservation of target protein (i.e., many “sample-losing” steps prior to analysis). Though target protein visualization is naturally a positive aspect to ABPP, it is the relative quantitative information gained from both approaches that is most critical to subsequent protein identification methods such as MS (which requires a certain protein amount to be present, relatively speaking, to obtain sufficient fragmentation (structural) information on sample peptides for proper identification via database similarity searches to take place). Setting up a comparison of probe sensitivities toward a standard PTP enzyme will be described here. 1. Using the Yersinia PTP (YopH) as a standard protein (see Subheading 2.1), incubate the enzyme (25 μM) with both the biotin- and fluorophore-conjugated probes (2 mM) for 60 min in the labeling buffer (see Subheading 2.1) at 25°C. Set up a serial dilution of the YopH enzyme concentration so that the following “gram” amounts can be loaded into each successive lane: 200, 100, 50, 20, 10, 5, 2, 1, 0.5, 0.2, 0.1, 0.05, 0.02, and 0.01 ng. 2. Quench the labeling reactions by adding an equal volume of the quenching solution highlighted in Subheading 2. Heat the resulting mixture at 75°C for 5 min. 3. Load the entire sample for all experimental groups. Resolve the proteins by using 12.5% SDS-PAGE gels.
426
Walls, Zhou, and Zhang
4. Immediately scan the gel containing the fluorescently labeled YopH enzymes using a Typhoon 9400 scanner (Amersham Biosciences) in fluorescent mode, with green (532 nm) and red (633 nm) lasers with preset optimum emission filters. Quantify the scanned images with ImageQuant 3.3 (Molecular Dynamics). 5. The enzymes tagged with the biotin-conjugated probe will have to be visualized by Western blotting using antibiotin antibody. Set up a western blot of this gel by following the western blotting procedure described in Subheading 3.2. 3.4. Effect of H2O2 on PTP Activity Monitored by the a-Bromobenzylphosphonate PTP Probes I and II Both in vitro and in vivo
It has been experimentally shown that physiologically, hydrogen peroxide (H2O2) acts as a second messenger. One of its physiological targets is the PTP family of enzymes, where it acts as a “transient” negative regulatory modification to PTP activity through direct engagement of the nucleophilic cysteine residue of the PTP signature motif ((H/V)C(X)5R(S/T). Hydrogen peroxide treatment, in vitro, of a standard PTP provides a unique opportunity to observe the probes selectivity toward only the active form of the PTP. In vivo, H2O2 treatment of cultured cells can provide insights into the inhibition properties of this molecule to specific PTP enzymes of interest as H2O2 and the PTP probe act as competitive inhibitors. 1. To carry out the in vitro labeling experiment, incubate (1 μM) PTP1B and H2O2 (1 mM) in a 20-μL reaction at 25°C in the labeling buffer highlighted in Subheading 2.4. Follow the protocol in Subheading 3.1 to kinetically characterize this reaction in the presence and absence of H2O2. 2. To establish an understanding of the probe’s selectivity for active PTP in a complex protein mixture, set the probe against a whole cell lysate generated from a Human-colon-carcinomaderived HCT-116 cell line treated with and without H2O2. 3. First, establish a cell culture of Human-colon-carcinoma-derived HCT-116 cells (reagents as described in the cell culture section of the materials). Upon reaching a suitable confluency, treat the HCT-116 cells with or without H2O2 (1 mM) for 5 min. 4. After the 5-min incubation period, immediately pour off the cell medium and wash the cell bed with ice-cold 1×PBS (twice). Collect the cells in a sufficient volume (1 mL) (for the size of the plates the cells are grown on) of ice-cold PBS (1×) by scraping them from the dish with a sterilized Teflon cell scraper. Pipette the volume of PBS (1×) and cell suspension in an appropriate collection vessel (suitable for high-speed centrifugation). Centrifuge this cell suspension at 1,000 × g at 4°C for 10 min. After this procedure the cell pellet will be clearly visible. 5. Aspirate the PBS (1×) from the cell pellet.
Activity-Based Protein Profiling of Protein Tyrosine Phosphatases
427
6. Resuspend the cell pellet in the following lysis buffer. 50 mM MES (pH 6.0), 150 mM NaCl, 1.0% Triton X-100, 0.1% SDS, 10% glycerol, 5 mM EDTA, 10 μg/mL leupeptin, and 5 μg/mL aprotinin, 1 mM PMSF containing 2 mM fluorescent probes (predetermine the volume of lysis buffer to be used). 7. Let the probes incubate with the cell lysate proteins for 30 min on ice. Perform a control experiment at the same time with the cells not treated with H2O2. 8. After completion of the labeling reaction, clear the two cell lysates (control and experimental) of a nonprotein cellular debris by centrifugation at 13,000 × g for 15 min in a prechilled 4°C centrifuge. 9. Pipette the supernatants into new collection tubes and discard the pellets. Determine the relative protein concentrations by Bradford assay. 10. After establishing a relative protein concentration for each of the experimental groups, use these concentrations to load a comparative amount of each of the experimental group’s lysate proteins to a 10% SDS-PAGE gel (load 50 μg total protein per lane). Also, load a prestained molecular weight marker (8 μL) for sample molecular weight determinations. 11. Visualize the fluorescent probes using the method described earlier. After fluorescence scanning, silver stain the gel (according to the manufacturer’s instructions) to directly visualize equal protein loading between the two lanes. 12. The enzymes tagged with the biotin-conjugated probe will have to be visualized by Western blotting using antibiotin antibody. Set up a western blot of this gel by following the western blotting procedure (see Subheading 3.2). 3.5. PTP Profiling in Cancer Cells Monitored with the a-Bromobenzylphosphonate PTP Probes I and II
Protein tyrosine phosphatases play extremely critical roles in the regulation of a myriad of cellular process. Multicellular organisms use reversible tyrosine phosphorylation to alter protein structure and function during intracellular signal transduction. Cancer has been described as a disease of malfunctioning cells, and in many cases it is the “deregulation” of the reversible phospho-tyrosine protein states that underlies the transformation process. The PTP ABPs offer a unique approach to gaining possible experimental insight into the underlying molecular mechanisms that are contributing to the “transformed” state of various authentic cancer cell lines if PTPs are involved. 1. Establish MOA231 (lung), MDA-MB-435S (breast), MCF-7 (breast), HepG2 (liver), SKOV3 (ovary), HeLa (cervices), and HCT-116 (colon) cell cultures in the growth medium described in (Subheading 2.5) under a humified atmosphere of 5% CO2. 2. When the cells reach a suitable confluency perform the collection and lysate protocol as described in Subheading 3.4.
428
Walls, Zhou, and Zhang
A difference from this protocol is that the lysis buffer does not contain ABP and the labeling reaction is to be done after cell debris removal as previously described. Incubate the cells on ice in lysis buffer for 30 min. 3. After a relative protein concentration has been determined for each of the cell lysates, incubate 50 μg of protein from each lysate with 1 mM of the fluorescent probe (1 mM biotin-based probe can be used) for 1 h at 25°C in the labeling buffer described in (Subheading 2.1). 4. Perform steps 10 and 11 from the previous section (Subheading 3.4). 5. The enzymes tagged with the biotin-conjugated probe will have to be visualized by Western blotting using antibiotin antibody. Set up a western blot of this gel by following the western blotting procedure described in Subheading 3.2.).
4. Notes 1. The α-bromobenzylphosphonate-based probes undergo significant solvolysis in solution with a pH value >7 (5). In pH 6.0 buffer and at room temperature, the probes are stable for at least 1 h (unpublished data). Store the probes at −20°C in DMSO. Dilute to the desired concentration in the pH 6.0 buffer before experiments. 2. For detection limit of the probe labeling, a serial dilution of the probe-labeled YopH was made in 2×SDS-loading buffer containing 0.01 mg/mL BSA as a carrier protein.
Acknowledgments This work was supported in part by grants from the National Institutes of Health DK68447 and CA69202.
References 1. Liu, Y., Patricelli, M.P., and Cravatt, B.F. (1999) Activity-based protein profiling: The serine hydrolases. Proc. Natl. Acad. Sci. USA 96, 14694–14699. 2. Kidd, D., Liu, Y., and Cravatt, B.F. (2001) Profiling serine hydrolase activities in complex proteomes. Biochemistry 40, 4005–4015.
3. Patricelli, M.P., Giang, D.K., Stamp, L.M., and Burbaum, J.J. (2001) Direct visualization of serine hydrolase activities in complex proteomes using fluorescent active site-directed probes. Proteomics 1, 1067–1071. 4. Saghatelian, A., Jessani, N., Joseph, A., Humphrey, M., and Cravatt, B.F. (2004) Activity-based
Activity-Based Protein Profiling of Protein Tyrosine Phosphatases
5.
6.
7.
8.
9.
10.
probes for the proteomic profiling of metalloproteases. Proc Natl. Acad. Sci. USA 101, 10000–10005. Kumar, S., Zhou, B., Liang, F., Wang, W.-Q., Huang, Z., and Zhang, Z.-Y. (2004) Activitybased probes for protein tyrosine phosphatases. Proc. Natl. Acad. Sci. USA 101, 7943–7948. Kumar, S., Zhou, B. Lian, F., Yang, H., Wang W-Q, and Zhang, Z.-Y. (2006) Global Analysis of protein tyrosine phosphatase activity with ultra-sensitive fluorescent probes. J. Proteome Res. 5, 1898–1905. Verdoes, M., Florea, B., Menendez-Benito, V., Maynar, C.J., Witte, M.D., Van der Linden, W.A., et al. (2006) A fluorescent broad-spectrum proteasome inhibitor for labeling proteasomes in vitro and in vivo. Chem. Biol. 13, 1217–1226. Yee , M. , Fas, S.C., Stohlmeyer, M.M. , Wandless, T.J., and Cimprich, K.A., (2005) A Cell-permeable, activity-based probe for protein and lipid kinases. J. Biol. Chem. 280, 29053–29059. Tonks, N.K., and Neel, B.G. (2001) Combinatorial control of the specificity of protein tyrosine phosphatases. Curr. Opin. Cell Biol. 13, 182–195. Li, L., and Dixon, J.E. (2000) Form, function, and regulation of protein tyrosine phosphatases
11.
12.
13.
14.
15.
16.
429
and their involvement in human diseases. Semin. Immunol. 12, 75–84. Zhang, Z.-Y. (2001) Protein tyrosine phosphatases: Prospects for therapeutics. Curr. Opin. Chem. Biol. 5, 416–423. Alonso, A., Sasin, J., Bottini, N., Friedberg, I., Osterman, A., Godzik, A., et al. (2004) Protein tyrosine phosphatases in the human genome. Cell 117, 699–711. Zhang, Z.-Y. (2003) Mechanistic studies on protein tyrosine phosphatases. Prog. Acid Res. Mol. Biol. 73, 171–220. Zhang, Z.-Y., Clemens, J.C., Schubert, H.L., Stuckey, J.A., Fischer, M, W.F., Hume, D.M., et-al. (1992) Expression, purification, and physicochemical characterization of a recombinant Yersinia protein tyrosine phosphatase J. Biol. Chem. 267, 23759–23766. Puius, Y.A., Zhao, Y., Sullivan, M., Lawrence, D.S., Almo, S.C., and Zhang, Z.-Y. (1997) Identification of a second aryl phosphatebinding site in protein-tyrosine phosphatase 1B: A paradigm for inhibitor design. Proc. Natl. Acad. Sci. USA 94, 13420–13425. Taylor, W.P., Zhang, Z.-Y., and Widlansky, T.S. (1996) Quiescent affinity inactivators of protein tyrosine phosphatases. Bioorg. Med. Chem. 9, 1515–1520.
Chapter 28 Active Protease Mapping in 2DE Gels Zhenjun Zhao and Pamela J. Russell Summary Proteases act as the molecular mediators of many vital biological processes. To understand the function of each protease, it needs to be separated from other proteins and characterized in its natural, biologically active form. In the method described in this chapter, proteases in a biological sample are separated under nonreducing conditions in 2DE gels. A specific small protease substrate, tagged with a fluorescent dye, is copolymerized into the SDS gel in the second dimension. After electrophoresis, the proteins are renatured by washing the gel with Triton X-100 solution or Milli Q water to remove SDS. The gel is then incubated in a protease assay buffer. The hydrolysis of the tagged specific substrate by the renatured protease releases the free fluorescent dye, which fluoresces in situ. The fluorescent spots indicate the location of the specific proteases in the gel and the specificity of the proteases. Key words: Proteases, Substrate-specific in-gel protease assay, Protein renature.
1. Introduction Proteases, representing approximately 2% of the total number of proteins present in all types of organisms (1, 2), act as the molecular mediators of many vital biological processes, such as embryonic development, immune responses, wound healing, and cancer metastasis, and assist in the processing of cellular information (3). To understand the function of each protease, it needs to be separated from other proteins and characterized in its natural biologically active form. Due to the use of large amounts of chaotropes, detergents, and reducing agents, the two-dimensional electrophoresis (2DE) technique has limitations for investigating biologically
David Sheehan and Raymond Tyther (eds.), Methods in Molecular Biology, Two-Dimensional Electrophoresis Protocols, vol. 519 © Humana Press, a part of Springer Science + Business Media, LLC 2009 DOI: 10.1007/978-1-59745-281-6_28
431
432
Zhao and Russell
active proteases. Traditionally, gelatin or casein zymography has been the standard method for separating and characterizing active proteases. It is suitable for nonspecific proteases only; moreover, the presence of macromolecules in sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDSPAGE) gels reduces gel resolution and makes it difficult to cut the spots (4). In theory, another problem of 2DE gelatin or casein zymography is the presence of gelatin or casein and their fragments hydrolyzed by the proteases, often with unknown specificity, in gel spots making the identification of the proteases by mass spectrometry (MS) difficult. Recently, a new technique, activity-based protein profiling, has been developed. This technique utilizes active site-directed probes to label active proteins before separation and then profile the functional status of enzymes in gels (5, 6). Covalent modification of the enzymes by the probes may change the isoelectric points (pIs) of the target proteases and may interfere with trypsin digestion. This technique has been limited to onedimensional electrophoresis (1DE) gels so far. The method presented in this chapter, substrate-specific in-gel protease assay (7), has been developed and applied successfully in identifying a protease in 2DE gels from a cell culture medium (8). Instead of macromolecules such as gelatin or casein, a small protease substrate tagged with a fluorescent dye (a nonfluorescent molecule conjugate) is copolymerized into separation gels in the second dimension (9). After separation by 2DE, proteins are renatured by removing SDS from the gels. The gels are then incubated in a protease assay buffer. Protease hydrolysis of the tagged specific substrate releases the free fluorescent dye, which fluoresces in situ. The fluorescent spot, which contains the specific protease, can then be visualized, excised, and subjected to trypsin digestion and MS analysis to identify the protease. The position of the spot also gives information about the protease’s pI and molecular weight. The procedure presented here is for the mapping of active proteases with specificity for lysine or arginine, such as trypsin and urokinase-type plasminogen activator, from a cell culture medium, separated in a large SDS gel (18 cm × 20 cm × 1 mm). A 4-methyl-coumaryl-7-amide (MCA) tagged small peptide, Boc-Gln-Ala-Arg-MCA, is used as the specific substrate. After hydrolysis, free 7-amino-4-methyl-coumarin is released and can be detected with an excitation wavelength of 346 nm and an emission wavelength of 442 nm. By using appropriate substrates, other proteases with different specificities could be detected (see Note 1).
Active Protease Mapping in 2DE Gels
433
2. Materials 2.1. Equipment
1. Protein desalting and concentrating products, such as Centricon™ Plus-80 (Millipore, Billerica, MA). 2. Dry strip reswelling tray. 3. Gel-casting apparatus. 4. Isoelectric focusing (IEF) electrophoresis apparatus, such as Bio-Rad Protean IEF cell (Bio-Rad, Hercules, CA). 5. SDS gel electrophoresis apparatus, such as Protean II xi/XL system (Bio-Rad, Hercules, CA). 6. Fluorescence imaging system, such as Bio-Rad Fluor-S imaging system (Bio-Rad, Hercules, CA).
2.2. Reagents
In order to retain the bioactivities of the proteases, no reducing agents are used in any of the solutions. 1. Immobilized pH gradient (IPG) strip rehydration solution: 4 M urea, 2% CHAPS, 2% IPG buffer pH 6–11, 0.002% bromophenol blue. Store in aliquots (300 μL) at −20°C. 2. IPG strips 18-cm (GE Healthcare, Piscataway, NJ). 3. Mineral oil. 4. Electrode wicks. 5. 10% Ammonium persulfate (APS). Freeze in single use aliquots at −20°C. 6. TEMED. 7. Boc-Gln-Ala-Arg-MCA (Molecular Probes, Eugene, OR) stock solution (20 mM in Milli Q water). Freeze in aliquots at −20°C and protect from light. The solution is stable for a few months. 8. Standard gel preparation solutions: 40% acrylamide/Bis solution (37.5:1), 10% SDS (Bio-Rad, Hercules, CA). 9. Equilibration solution: 2 M urea, 2% SDS, 20% glycerol, 0.375 M Tris–HCl, pH 8.8. Freeze in aliquots at −20°C. 10. Triton X-100 2.5%. Store at 4°C. 11. Protease assay buffer: Glycine buffer 0.2 M. Store at 4°C.
434
Zhao and Russell
3. Methods 3.1. In-Gel Sample Rehydration of IPG Strips
1. Collect serum-free cell culture medium (RPMI 1640) from cell culture flasks. Centrifuge at low speed (200 × g for 5 min) to remove cellular debris and then concentrate the sample to around 4 mg/mL protein using a Centricon™ Plus-80 with a cut-off value of 30 kDa (see Note 1). Sonicate the sample to reduce viscosity. 2. Mix the samples with the rehydration solution in a ratio of 2:5. Leave the mixture at room temperature for 10 min, and then centrifuge at 10,000 × g for 3 min. 3. Level the dry strip reswelling tray. 4. Pipette an appropriate amount (350 μL) of the supernatant into the grooves of the dry strip reswelling tray. 5. Peel off the protective cover sheet from an 18-cm IPG strip, and position it such that the gel of the strip is in contact with the sample (gel side down). 6. Cover the IPG strip and the sample with mineral oil and let the strip rehydrate overnight at room temperature.
3.2. IEF Separation (The First Dimension)
1. Remove the rehydrated IPG strip carrying the protein sample from the groove with tweezers, and then blot it gently with a sheet of filter paper to remove excess sample. 2. Position the IPG strip in a groove of a Bio-Rad Protean IEF cell (gel side up, cut short to fit in if necessary). Leave a 0.5–0.7 cm gap between the end of the strip and the electrode in each side. The acidic end of the IPG strip must face toward the anode. 3. Cut two electrode wicks of 1.2–1.5 cm length. Soak the wicks in Milli Q water. Blot them with filter paper and place the wicks longitudinally on top of the aligned IPG strip at the cathodic and anodic gel ends with 3–5 mm overlap with the strip. 4. Cover the strip with mineral oil. 5. Assemble electrodes. 6. Connect electric power. Increase voltage from 50 to 100; 200; 500; 1,000; and 2,500 V during the first 10 h, followed by 5,000 V for 60,000 Vh. The conditions may need to be changed for different samples. 7. Disconnect power. Remove the IPG strip with tweezers and put it in a groove of a dry strip reswelling tray. Add the equilibration
Active Protease Mapping in 2DE Gels
435
solution to soak the strip for 45 min at room temperature with gentle shaking. 3.3. SDS-PAGE (The Second Dimension)
1. Substrate gel casting: Mix 15 mL of 40% acrylamide/Bis, 12.5 mL of 1.5 M Tris–HCl, pH 8.8, 21.25 mL of Milli Q water, 0.5 mL of 10% SDS, 0.25 mL of 10% APS. Mix well. Then add 0.5 mL of Boc-Gln-Ala-Arg-MCA stock solution (20 mM in Milli Q water) into the mixture (1.0% of the total final volume, final concentration 200 μM) (see Notes 2–4). Mix well. Add 25 μL of TEMED. Mix well and cast the gel (12%) overnight. 2. Blot the equilibrated IPG strip with filter paper to remove excess solution. 3. Fill the gap at the top of the cast gel with molten embedding solution. 4. Immediately load the IPG strip through the solution, onto the top of the SDS-PAGE gel. Let the solution cool down to room temperature (about 15–20 min). 5. Start electrophoresis. Set the current at 5 mA/gel for 1 h and then 10 mA/gel to completion (overnight).
3.4. Renaturation of Proteases and Fluorescence Development
1. Take out the gel and wash it in cold 2.5% Triton X-100 five times, soaking the gel for 5 min in each wash with gentle shaking. If the gel is to be subjected to MS identification afterward, wash the gel in cold Milli Q water the same way seven times (see Note 5). 2. Soak the gel in 0.2 M glycine buffer at 37°C overnight (see Note 6).
3.5. Fluorescence Imaging
1. Carefully place the gel on the sample plate of a Bio-Rad Fluor-S imaging system. Add a small amount of the glycine buffer on the topside of the gel to prevent gel drying. 2. Select scanning transillumination model and UV excitation wavelength 290–365 nm. Use a coumarin band pass filter (425–480 nm) for the emission and a cooled (10 ± 0.1°C) CCD camera with an imaging array of 1,340 × 1,040, pixel size of 6.8 × 6.8 μm and pixel depth of 12-bit. As little as 1 ng of trypsin can be visualized (see Note 7). 3. Capture and analyze the image using Quantity One software (Bio-Rad, Hercules, CA) (see Note 8). The fluorescent spots on the gel are the proteases with the appropriate substrate specificity. The spots can be excised and subjected to trypsin digestion and MS identification as described in Chapter “Tryptic Digestion of In-Gel Proteins for Mass Spectrometry Analysis.”
436
Zhao and Russell
4. Notes 1. Aqueous samples should be particle-free and appropriately concentrated or diluted. The molecular weight cut-off value of the Centricon™ Plus-80 should be lower than the molecular weights of the targeted proteins. The concentration of salts and/or detergents in the samples should be low enough to prevent interference with electrophoresis. Dry powder samples should be solubilized in Milli Q water if possible or directly in rehydration solution, at an appropriate concentration. The protein concentration of the samples should be high enough to allow visualization of the protease spots, which depends on protease activity and the substrate used. 2. To prepare the substrate gels, the substrate conjugates should be soluble in the gel solution and not migrate during electrophoresis. Choice of the right substrate conjugate is important for the success of the detection. 3. After addition of the Boc-Gln-Ala-Arg-MCA stock solution to gel solution, a white cloud appears; it disappears after mixing. 4. The detection of other proteases with different specificities depends on the availability of their labeled specific substrates. As far as we are aware, many specific protease substrates labeled with a fluorescent dye for different proteases are available from Invitrogen (Molecular Probes) (http://probes. invitrogen.com). 5. Triton X-100 is not compatible with MS identification. If the gel spots are to be subjected to MS identification, do not use Triton X-100 to renature the proteins. Water can be used as an alternative in this case, but the efficiency is not as good as when using Triton X-100 solution. Washing too many times will reduce the detection sensitivity. 6. The conditions, such as buffer, temperature, time, etc., for development of the fluorescence (hydrolysis of the substrate) are different for different proteases and substrate conjugates. They should be optimized for each detection. 7. The sensitivity of the detection depends on the fluorescent intensity of the fluorescent tag, the activity of the given proteases against the substrate and the sensitivity of detection of the imaging system. It may be improved if a better substrate conjugate or a better camera is developed. 8. The fluorescent band intensity correlates linearly with the amount of trypsin in the range of our test (2.5–80 ng; Fig. 1). It is possible to calculate the amount of the proteases if their standards are available.
Active Protease Mapping in 2DE Gels
a
b
pH 8.5
pH 9.3
pH 11
pH 8.5
pH 9.3
437
pH 10.5
40.0 ng
80.0 ng
10.0 ng
20.0 ng
2.5 ng
5.0 ng
25kDa
90 y = 0.0009x + 7.0332 R2 = 0.9741
80 70 60 50 40 30 20 10 0 0
20000
40000
60000
80000
100000
Adjusted volume (total counts) Fig. 1. Assay of trypsin on MCA 2DE gel. Different amounts of trypsin were separated by IEF and then electrophoresed in an MCA SDS-PAGE gel. After renaturating and development of fluorescence, the trypsin spots were visualized. (a) Image of fluorescent spots. (b) Linear regression of trypsin amount and spot counts. (This figure is reprinted from Zhao and Russell (7) with permission of the authors and Electrophoresis.
References 1. Barrett, A.J. (2004) Bioinformatics of proteases in the MEROPS database. Curr Opin Drug Discov Devel 7, 334–41. 2. Rawlings, N.D., Morton, F.R., and Barrett, A.J. (2006) MEROPS: the peptidase database. Nucleic Acids Res 34, D270–2. 3. Saklatvala, J., Nagase, H., and Salvesen, G. (Eds) (2003) Proteases and the Regulation of Biologival Processes, Portland Press, London. 4. Yatsuda, A.P., Bakker, N., Kritjgsveld, J., Knox, D.P., Heck, A.J., and de Vries, E. (2006) Identifica-
tion of secreted cysteine proteases from the parasitic nematode Haemonchus contortus detected by biotinylated inhibitors. Infect Immun 74, 1989–93. 5. Sieber, S.A., Niessen, S., Hoover, H.S., and Cravatt, B.F. (2006) Proteomic profiling of metalloprotease activities with cocktails of active-site probes. Nat Chem Biol 2, 274–81. 6. Jessani, N., and Cravatt, B.F. (2004) The development and application of methods for activity-based protein profiling. Curr Opin Chem Biol 8, 54–9.
438
Zhao and Russell
7. Zhao, Z., and Russell, P.J. (2003) Trypsin activity assay in substrate-specific one- and twodimensional gels: a powerful method to separate and characterize novel proteases in active form in biological samples. Electrophoresis 24, 3284–8. 8. Zhao, Z., Raftery, M.J., Niu, X.M., Daja, M.M., and Russell, P.J. (2004) Application of in-gel protease assay in a biological sample:
characterization and identification of urokinasetype plasminogen activator (uPA) in secreted proteins from a prostate cancer cell line PC-3. Electrophoresis 25, 1142–8. 9. Yasothornsrikul, S., and Hook, V.Y. (2000) Detection of proteolytic activity by fluorescent zymogram in-gel assays. Biotechniques 28, 1166–8, 70, 72–3.
Chapter 29 Two-Dimensional Difference Gel Electrophoresis Gert Van den Bergh Summary The introduction of two-dimensional fluorescent difference gel electrophoresis has enabled the extensive screening of differential protein expression levels with higher confidence and greater sensitivity than using the classical two-dimensional electrophoresis (2DE) approach. Using this technology, multiple protein samples can be labeled with up to three different fluorescent dyes. These labeled protein samples are mixed and applied on the same 2DE gel, subsequently scanned and analyzed by specialized software tools. The possibility to run two or more protein samples on a single gel, as well as the introduction of an internal standard on each gel drastically reduces the gel-to-gel variability and thus results in higher levels of certainty with regard to the differential character of the expressed proteins. Key words: Two-dimensional electrophoresis, Two-dimensional difference gel electrophoresis, DIGE, CyDyes, Proteomics, Fluorescence.
1. Introduction Numerous technological developments, such as differential display PCR (1), suppressive subtractive hybridization (2) or cDNA microarrays (3), have enabled the rapid and in depth analysis of differential mRNA expression levels of large sets of gene products in parallel. Notwithstanding the wealth of information these approaches have provided regarding the molecular mechanisms underlying physiological processes or diseases, proteins are the ultimate effector molecules of the cell. The quantitative study of the proteome should therefore be given high priority, which is true for two more reasons. Firstly, quantitative variations in
David Sheehan and Raymond Tyther (eds.), Methods in Molecular Biology, Two-Dimensional Electrophoresis Protocols, vol. 519 © Humana Press, a part of Springer Science + Business Media, LLC 2009 DOI: 10.1007/978-1-59745-281-6_29
439
440
Van den Bergh
mRNA levels are not always correlated with corresponding changes in the expression of proteins (4), and secondly, many proteins contain post-translational modifications that are not detectable at the transcriptional level at all but are essential for their activation and functional regulation. Large-scale, quantitative analysis and comparison of protein expression levels, also called functional proteomics, is generally performed by two different approaches: gel-free or gel-based. In gel-free proteomics, proteins or enzymatically cleaved peptides derived from proteins are separated, quantified, and identified by liquid chromatography (LC) and mass spectrometry (MS), whereas in gel-based proteomics, proteins are first separated and quantified by two-dimensional electrophoresis (2DE), followed by MS identification of the differentially expressed proteins. While gel-free proteomics is more capable of detecting hydrophobic proteins such as membrane proteins, gel-based proteomics methods can easily detect post-translationally modified proteins. Moreover, it is easier to compare a large number of experimental conditions by 2DE than by MS. Since comparative gel-based proteomics depends on the quantitative separation and visualization of a large fraction of the expressed proteins, traditional 2DE is currently not the optimal choice, since it suffers from a rather high level of variability between different gels. Moreover, traditional staining methods such as coomassie brilliant blue or silver staining (see Chapter “Silver Staining of Proteins in 2DE Gels”) provide either a low sensitivity or a limited linear range, impairing the quantitative aspect of protein visualization. The introduction of two-dimensional difference gel electrophoresis (DIGE), a multiplexing approach for 2DE based on fluorescent protein visualization, proved to be of enormous importance for the renaissance of gel-based proteomics (5–8) (see also Chapter “High-Resolution 2DE”). With this technique, proteins are labeled with up to three spectrally distinct fluorescent dyes that have no discernable influence on the electrophoretic mobility of the proteins. Two to three labeled samples can therefore be run on a single 2DE gel. The protein spot patterns are visualized by sequentially scanning the gel at the excitation and emission wavelengths of each of the dyes used. The quantitative analysis of differential protein expression levels is then performed by using appropriate software (see Chapter “Troubleshooting Image Analysis in 2DE”), a process that is facilitated by the relative comparison of protein expression levels obtained from the same gel, reducing the gel-to-gel variability inherent to 2DE. This variability can be further reduced by using an internal standard, a mixture of protein samples from every experimental condition, which is applied on every gel (Fig. 1) (7). As a result, the matching of spots and their quantification across different gels is greatly facilitated and enables DIGE to be
Two-Dimensional Difference Gel Electrophoresis
441
Fig. 1. Schematic overview of the DIGE technique, incorporating three fluorescent dyes and an internal standard.
employed as a high-throughput screening method for the detection of differential protein expression. Although recently, new dyes have been developed that enable saturation labeling of the proteins and therefore provide a much higher sensitivity than the minimal labeling dyes (9), in this chapter we will only describe the standard DIGE approach using an internal standard and minimal labeling dyes.
442
Van den Bergh
2. Materials 2.1. Sample Preparation
1. 2DE lysis buffer. 7M urea, 2M thiourea, 4% CHAPS, 40mM Tris-base. Store in 1-mL aliquots at −20°C or use fresh. Do not allow warming above room temperature, since this can result in the degradation of urea to isocyanate, leading to the carbamylation of proteins. Add 40mL complete protease inhibitor (Roche, Basel, Switzerland) to 1 mL aliquot prior to use (stock solution: 1 tablet in 2mL HPLC-grade water. Store for a maximum of 1month at −20°C) (see Note 1). 2. Bath sonicator (Model 5,510, Branson Ultrasonics, Danbury, USA). 3. PBS buffer. 0.1M NaH2PO4, pH 7.4. 4. Protein Assay kit (Bio-Rad, Hercules, CA, USA). 5. ELISA plate reader (Labsystems Multiskan RC, Thermo Electron Corporation, Brussels, Belgium).
2.2. Experimental Setup and Protein Labeling with Fluorescent Dyes
1. pH indicator strips. 2. Cy2, Cy3 and Cy5 minimal labeling dyes (GE Healthcare Life Sciences, Uppsala, Sweden). 3. Dimethylformamide (DMF), waterfree. 4. 10mM Lysine solution.
2.3. Isoelectric Focusing (IEF)
1. Immobilized pH gradient (IPG) strips, such as the Immobiline DryStrips (GE Healthcare Life Sciences). We use 24cm strips with pH gradients from pH 3–11 nonlinear for starting experiments. In later experiments, we sometimes use strips with narrower pH gradients (pH 4–7 or 6–9) to focus on a specific subset of proteins. 2. Destreak rehydration solution with 0.5% pharmalytes (GE Healthcare Life Sciences). 3. IPG strip reswelling tray (GE Healthcare Life Sciences). 4. Paraffin oil as cover fluid. 5. IPGphor cup loading manifold (GE Healthcare Life Sciences). 6. IPGphor IEF unit (GE Healthcare Life Sciences). 7. 2× lysis buffer. 7M urea, 2M thiourea, 4% CHAPS, 2% Dithiothreitol (DTT), 40mM Tris-base. Store in 1 mL aliquots at −20°C or use fresh.
2.4. SDS-PAGE
1. Gel casting equipment. 2. Low fluorescence glass plates (GE Healthcare Life Sciences). 3. Vacuum grease. 4. Acrylamide gel solution. For a 12.5% T gel, we prepare 900mL of the following solution, which is enough for casting 14 gels: 375mL acrylamide stock solution (30% T, 2.6%
Two-Dimensional Difference Gel Electrophoresis
443
C, BioRad Hercules CA, USA), 225mL Tris buffer (1.5M Tris–HCl, pH 8.8), 300mL water. 5. TEMED and ammonium persulfate (APS). 6. 100mL Displacing solution. 0.375M Tris–HCl, pH 8.8, 50% glycerol, bromophenol blue. 7. Water saturated butanol. 8. Two liter of 10× SDS running buffer. 250mM Tris, 1.92M glycine, 1% SDS, approximately pH 8.3. The pH of this solution should not be adjusted. Dilute to obtain 1× and 2× SDS running buffer. 9. 100mL agarose sealing solution, containing 0.5% agarose in 2× SDS running buffer and a few grains of bromophenol blue. 10. Equilibration solution A. 50mM Tris–HCl, pH 6.8, 6M urea, 30% glycerol, 2% SDS and 1% DTT. 11. Equilibration solution B. 50mM Tris–HCl, pH 6.8, 6M urea, 30% glycerol, 2% SDS and 2.5% iodoacetamide. 2.5. Gel Imaging and Data Analysis (See Also Chapter “Troubleshooting Image Analysis in 2DE”)
1. Fluorescent gel imager capable of the detection of Cy2, Cy3, and Cy5 with high resolution and high sensitivity (e.g., Ettan DIGE Imager or Typhoon 9,400 Variable Mode Imager, GEHealthcare Life Sciences, or equivalent). 2. Image analysis software (DeCyder Differential Analysis Software, GEHealthcare Life Sciences, or equivalent). 3. Fixation solution. 50% methanol, 5% acetic acid in water.
2.6. Protein Identification
1. Coomassie blue G-250 protein staining solution (Bio-Rad) or other protein staining solutions. 2. Sterile razor blades or a spot-picking robot. 3. Laminar flow hood. 4. Acetonitrile (ACN), trifluoroacetic acid (TFA), formic acid (FA) (Merck Eurolabo, Leuven, Belgium). 5. Digestion solution: 100ng modified porcine trypsin (Promega, Leiden, The Netherlands, sequencing grade), 12.5mM NH4HCO3 and 5% ACN in water. 6. ZipTip C18 reversed-phase chromatography pipette tips (Millipore, Bedford, MA, USA). 7. Mass spectrometer.
3. Methods 3.1. Sample Preparation
1. Collect fresh or frozen cells or tissue. In our experiments, we usually collect protein samples from a small area of mammalian brain, namely the neocortex. To this end, 200mm thick cryostat
444
Van den Bergh
sections are cut and an area of gray matter of approximately 160mm2 (+/−32mm3) is cut from these sections on dry ice to prevent protein degradation. Collected tissue is rapidly transferred to a volume of ice-cold lysis solution. For the above-described amount of tissue, we use a standard volume of 100mL lysis solution. This volume should be changed according to the volume of the cells or tissue to be lysed (see Note 2). 2. Homogenize brain tissue on ice. Avoid excessive foaming of the lysis buffer to prevent protein loss. 3. Sonicate in a bath sonicator for approximately 1min at room temperature. 4. Put protein sample at room temperature for 1h, to allow for complete solubilization of the proteins in the lysis buffer. 5. Sonicate again for 1min at room temperature. 6. Clear the protein lysate by centrifugation for 20min at 13,000×g at 4°C. 7. Dialyze the supernatant against water for 2–3h, to decrease the salt concentration of the protein sample. Place the supernatant in dialysis tubing with a cut-off of 500Da. 8. Determine the protein concentration of the dialysate. We utilized a modified Bradford assay, as previously described (10), but any protein quantification method that is compatible with the contents of the lysis solution (high concentrations of urea, thiourea, CHAPS…) can be used. Make a 2mg/mL stock solution of ovalbumin and prepare a standard dilution series, adding 400mL of a solution containing 500, 300, 100, 50, 25, and 12.5mg ovalbumin in PBS and 300mL 0.1M NaOH. Prepare the samples to be measured by mixing 1 mL of protein sample, 30mL 0.1M NaOH and 69mL PBS buffer. Place 20mL of the samples and ovalbumin standard in triplicate on a flat-bottomed 96-well plate. Add 200mL of Bio-Rad protein assay solution (diluted 1/10 in HPLC-grade water), read the absorbance at 595nm with an ELISA plate reader and determine protein concentrations. 9. Store the protein samples at −70°C in aliquots to reduce freeze-thaw cycles. 3.2. Experimental Setup and Protein Labeling with Fluorescent Dyes
Because of its relatively high cost and complexity, the design of a DIGE experiment is of utmost importance. When comparing two experimental conditions, it is possible to directly analyze both samples on a single gel to determine relative protein expression levels. However, in most cases where multiple samples have to be compared, it is advantageous to include an identical internal standard on every single gel alongside the experimental samples, since this standard can be used to normalize the experimental samples across different gels (by dividing the volume of every
Two-Dimensional Difference Gel Electrophoresis
445
spot on each gel by the volume of the same spot from the internal standard), thereby eliminating gel-to-gel variability. But even in a comparison of two samples, it is useful to include an internal standard, because of the requirement to run several biological replicates on different gels to obtain valid statistical information. In the experiments performed in our lab, where we routinely use Cy2 for the labeling of the internal standard, we always ensure that one half of the biological replicates of an experimental condition is labeled with Cy3, the other half with Cy5, in order to prevent dye-specifically labeled proteins to falsely show up as differentially expressed. We also always try to pair the samples under investigation ad random, to prevent any experimental bias due to technical reasons. An example of an experimental setup with three conditions is given in Table 1. 1. Prepare the internal standard. This standard is made of equal aliquots from every protein sample that will be investigated in the experiment. Make a larger amount of the internal standard than strictly necessary (50mg of protein for each gel that will be run) to compensate for gels that have to be repeated due to technical reasons or additional experimental conditions that are added late in the experimental design. 2. Check the pH of the protein samples and internal standard. The optimal pH for the labeling reaction is between pH 8.0 and 9.0. Adjust the pH only if it is higher than pH 9.0 or lower than pH 6.0. 3. Make a working solution of the CyDyes at a concentration of 400pmol/mL, by diluting them in waterfree DMF (see Note 3). 4. Mix 50mg of protein sample with 400pmol fluorescent dye (see Note 4).
Table 1 Example of an experimental setup for an experiment with three biological conditions (A–C) and four replicates (1–4). 50mg of each sample is loaded onto each gel. The internal standard consists of equal amounts of protein from samples A 1 to C 4 Gel Number
Cy3
Cy5
Cy2
1
A1
B1
Internal standard
2
C1
A2
Internal standard
3
B2
C2
Internal standard
4
A3
C3
Internal standard
5
B3
A4
Internal standard
6
C4
B4
Internal standard
446
Van den Bergh
5. Vortex sample and incubate for 30min at 4°C in the dark. 6. Add 1mL of 10mM lysine and incubate 15min at 4°C in the dark. This will block the labeling reaction of the remaining free dyes and will prevent cross-labeling after mixing the samples. 7. Mix the two samples, labeled with Cy3, Cy5, and the internal standard, labeled with Cy2 that have to be run on the same gel and store on ice. 3.3. Isoelectric Focusing (IEF)
The IEF method described here is based on the IPGphor IEF system, using broad pH gradient IEF strips to obtain a general overview of the proteins present in the samples. 1. To reconstitute the dried and frozen IPG strips to their original volume, they need to be rehydrated with destreak rehydration solution that at the same time provides the appropriate conditions to perform IEF (see Note 5). Remove the plastic cover sheets of the IPG strips and place them overnight (at least 10h) with their gel surface down in the correct volume of rehydration buffer containing 0.5% IPG buffer (with the same pH gradient as the gels) in a IPG drystrip reswelling tray. For a 24cm gel, the required volume is 450mL. Cover the strips with 2mL paraffin oil to prevent dehydration and carbon dioxide absorption. 2. Prepare the cup loading manifold for IEF. Place the ceramic manifold tray on the surface of the IEF unit. Place the rehydrated IPG strips with their gel side up in the grooves of the manifold tray, with their anodic end pointing towards the marked direction. Place the sample cups on the strips and pipette about 20mL of paraffin oil in the cups to check for leaks. If no leaks are found, the manifold tray and IPG strips should be covered as soon as possible with 100mL paraffin oil to prevent drying of the strips (see Note 6). 3. Wet paper wicks evenly with 150mL of distilled water and place them at the end of the IPG strips. One third of the paper wick should cover the end of the strip. Place the electrodes on the paper wicks, where the wicks and strips overlap. 4. Add an equal volume of 2× lysis buffer to the sample mixture. Pipette this mixture in the sample cups and close the lid. Make sure the sample is completely covered with paraffin oil. 5. Start the IEF run. For a broad pH gradient on a 24cm strip and 50mg protein per sample we use the following protocol: 3h at a constant voltage of 300V, 3h at a constant voltage of 600V, a gradual increase in voltage over 6h to 1,000V, a 3h gradual increase towards 8,000V and a final 4h focusing at a constant 8,000V. The temperature should be 20°C and the current 50mA/gel (see Note 7). 6. After the run, remove the IPG strips and proceed to the second-dimension or store them at −80°C between two plastic sheets for up to 3 months.
Two-Dimensional Difference Gel Electrophoresis
3.4. SDS-PAGE 3.4.1. Casting of Gels
447
1. Prepare the gel caster. Wash each item (gel caster, glass plates, separator sheets, blank cassette inserts) with water and detergent and thoroughly rinse with tap water and HPLC-grade water. Glass plates can be further cleaned by washing them with ethanol. Dry glass plates with lint-free cloth and keep away from dust. Any residual dust could result in artefactual spots on the gel images. Tilt the gel caster backwards and fill it, starting with a separator sheet. Alternate between a gel cassette (large and small glass plate) and separator sheet. If the caster is not filled by the number of gel cassettes required, add additional blank cassette inserts and separator sheets until the stack in the caster is even with the edge of the caster. 2. Lubricate the sealing gasket of the gel caster lid with vacuum grease, to ensure a liquid-tight seal and close the lid. 3. Prepare an appropriate volume of acrylamide gel solution. For a full 14-gel set, this amounts to 0.9L. Immediately prior to gel casting, add 250mL TEMED and 250mg APS in 1mL water to start the polymerization reaction. 4. Pour the gel solution in the gel caster through a funnel, taking care not to introduce any air bubbles. Fill to about 4cm of the top of the small glass plates. 5. Remove the funnel and add displacing solution to the balance chamber in small volumes, until the bottom V-well of the gel caster is filled with this solution. The remaining acrylamide solution is forced into the gel cassettes to the final gel height. The required volume of displacing solution will be between 50 and 100mL. 6. Immediately pipette 2mL of butanol onto each gel to ensure completely level gel surfaces. Allow for overnight polymerization of the gels before running the second-dimension electrophoresis to complete the polymerization reaction. After that, the gels can be stored in sealed plastic bags at 4°C for up to 1week.
3.4.2. Equilibration of IPG Strips
1. Prepare 100mL of a 0.5% agarose sealing solution in 2× SDS running buffer, including a few grains of bromophenol blue. Dissolve the agarose by boiling. 2. To saturate the separated proteins with SDS, the IPG strips need to be equilibrated with a solution containing SDS. Place the IPG strips in equilibration solution A for 15min at room temperature. We generally use 5mL/strip, although higher volumes could result in better equilibration and therefore less vertical streaking and better transfer of proteins from first to second-dimension. The DTT in this solution reduces the disulfide bonds. 3. Incubate the strips for another 15min in equilibration solution B (5–mL/ strip). The iodoacetamide in this solution will alkylate the free thiol groups of the reduced –SH groups, but more importantly will help to remove any remaining DTT that could result in vertical streaking (see Note 8).
448
Van den Bergh
3.4.3. SDS-PAGE
1. Prepare the electrophoresis unit following the instructions of the manufacturer. The protocol here is for the Ettan Dalt 12 system. Turn the pump valve to circulate and fill the separation unit with 7.5L 1× SDS running buffer. Turn the pump on and set the temperature at 12°C. 2. Prepare the gel cassettes for loading of the IPG strips by rinsing the top of the gel with water and drain. Remove any residual polyacrylamide from the surface of the glass plates. Store the gel cassettes upside down until loading of the IPG strip to prevent drying of the gel surface. 3. Remove one IPG strip from the equilibration solution, using forceps, and rinse it in fresh 2× SDS running buffer. Dry the strip by briefly placing it on its side on absorbent paper. Put the SDS-PAGE gel in an upright position in the cassette rack, with the small glass plate in front. Using two forceps, place the IPG strip with its plastic backing against the large glass plate just above the small glass plate and with the acidic side of the strip pointing to the left. Gently slide the strip between the two glass plates using a small spatula or plastic spacer, only pushing against the plastic background and not making contact with the surface of the IPG strip or the SDS-PAGE gel, until it firmly rests against the SDS-PAGE gel. Make sure the gel surface of the IPG strip does not touch the small glass plate. 4. Pipette 2mL of agarose solution on top of the IPG strip to seal it in place and allow it to solidify. Repeat steps 1–4 for every IPG strip. 5. Slide the gel cassettes in the separation unit, and fill unused slots with blank cassette inserts. 6. Fill the upper buffer chamber with 2.5L of 2× SDS running buffer. 7. Start the run at 40mA for 30min and continue with an overnight run at about 15mA/gel at 20°C. Running the second-dimension overnight gives the best resolution and facilitates the scanning of the gels on the following day. Stop the electrophoresis run when the bromophenol blue front reaches the lower edge of the gel. Leave the gels in the cooled running buffer in the apparatus until they are scanned.
3.5. Gel Imaging and Data Analysis
1. Remove the gel from the electrophoresis apparatus, clean with water and lint free cloth to remove dust particles from the glass plates. Transfer the gel to the gel scanner. All information that will follow is based on our experience with the Ettan DIGE imager, but is easily adaptable to other gel imagers (see Chapter “Troubleshooting Image Analysis in 2DE”). 2. To determine the optimal exposure times for each fluorescent dye present in the gel, perform a test scan for a small portion of the first gel of the batch, scanning it at a 100 mm resolution.
Two-Dimensional Difference Gel Electrophoresis
449
Table 2 Scan settings for the three CyDyes using an Ettan DIGE Imager CyDye Excitation filter (nm)
Emission filter (nm) Exposure time (s)
Cy2
480+/−30
530+/−40
1.0
Cy3
540+/−25
595+/−25
0.5
Cy5
635+/−30
680+/−30
2.0
The maximal pixel value should be close to 65,000 counts for maximal sensitivity, but should not exceed this value to avoid saturation. For laser based scanners, the voltage of the photomultiplier tubes has to be optimized for maximal sensitivity. 3. Scan each gel for each fluorescent dye with the exposure times as determined in the previous step. Resolution is at 100mm and the correct dye chemistry setting should be checked (minimal labeling). We usually use the exposure settings as given in Table 2. After scanning, gels can be stored in fixation solution, if required. See Fig. 2 for a representative example of DIGE gel images obtained by scanning with the Ettan DIGE imager. 4. Perform image analysis using the desired software. A very capable software package for the analysis of DIGE gel images is the DeCyder 2D software (GE HealthCare Life Sciences), which is specifically designed with the use of an internal standard in mind. A typical analysis consists of the following two modules: 5. In the differential in-gel analysis (DIA) module, perform intra-gel analysis. On each gel, protein spots are co-detected on the Cy2, Cy3, and Cy5 images and the Cy3 and Cy5 gel images are normalized with respect to the Cy2 image of the internal standard. Absolute protein abundances are therefore replaced by the normalized volume ratios of Cy3/Cy2 and Cy5/Cy2. 6. Perform inter-gel analysis using the biological variation analysis (BVA) module (Fig. 3. After automatic matching of all spots on each gel with the spots on a master gel, manually confirm or correct the matching for a large number of spots. Although taking some time, this step is required to achieve a correct statistical analysis. The relative abundances of all spots are then plotted against the normalized internal standard and a student’s t-test or one-way ANOVA test is applied to generate a list of protein spots that are differentially regulated between the different experimental conditions.) 3.6. Protein Identification
The identification of the proteins present in the differentially regulated spots is a procedure that consists of several steps. These steps are ultimately dependent on the type and brand of MS that one has access to. Therefore, the description that will follow below
450
Van den Bergh
Fig. 2. Representative Cy2, Cy3, and Cy5 images obtained from a DIGE gel, scanned with the Ettan DIGE imager. (a) An internal standard was labeled with Cy2. Two different protein samples from mammalian brain were stained with (b) Cy3 and C. Cy5.
should be adapted for each specific MS instrument. In general, a preparative 2DE gel is run that is matched to the protein spots visualized by DIGE. The spots of interest are excised from the gel, either manually or with a spot-picking robot. The proteins in the excised spots are digested with trypsin and the resulting
Two-Dimensional Difference Gel Electrophoresis
451
Fig. 3. Screen shot from the DeCyder biological variation analysis (BVA) workspace, showing the protein table view. This view is divided in four panes, namely (a) The image view, showing Cy3, Cy5, or Cy2 images of the gels in the experiment. (b) A 3-D view showing a three-dimensional representation of the gel region around the selected spot from the gel images in (a). (c) A graphical view shows a graphical quantitative representation of the selected spot from all gels in the experiment. In this example, spot 1,426 was selected in an experiment comparing protein expression levels of mammalian brain at different postnatal ages. Six experimental conditions were compared containing three data points each. The thick black line connects the means of each experimental condition, while the dashed lines each connect two points that were run on a single DIGE gel, labeled with either Cy3 or Cy5. D. Table view shows the statistical data from the experiment, with the selected spot indicated by the gray bar.
peptides are concentrated and purified. This is followed by analysis of these peptides by MS and identification of the protein by protein identification search engines (see Chapter “Database Interrogation Algorithms for Identification of Proteins in Proteomic Separations”). 1. Run one or more preparative 2DE gels as described above (Subheadings 3.4 and 3.5), except that 500mg or more of the sample of interest should be loaded onto each gel (see Note 9). Optionally, label 50mg of this sample with one of the CyDyes to facilitate the matching of the preparative gels with the DIGE images. In this case, the gel has to be imaged prior to detection of the proteins by coomassie blue staining, as described in Subheading 3.6. 2. Rinse the gels twice (5min) with 200mL of water. Stain the gels with 200mL coomassie blue staining solution overnight
452
Van den Bergh
and destain the gel by rinsing several times with water, until the blue spots are visible and the background is translucent. The gels can be stored in water for several days at 4°C prior to spot excision (see Note 10). 3. Excise the gel spots of interest using a sterile razor blade or a spot-picking robot using a picking list generated in DeCyder (see Note 11). 4. Perform an enzymatic digestion of the proteins in the gel spots using trypsin. Pool identical spots from different gels. Wash the gel cubes twice for 15min with milliQ water, followed by two 15min rinses in 50% ACN and two 5min washes with milliQ water. Dry the gel pieces in a SpeedVac vacuum centrif u g e and incubate overnight at 37°C in 40 m L digestion solution. Extract the tryptic peptides from the gel pieces by two washes with 100mL and 50mL of a 5:95 ACN:water solution and sonication in a bath sonicator for 30min. Concentrate the supernatant using SpeedVac vacuum centrifugation. 5. Desalt and concentrate the tryptic peptides using ZipTip C18 reversed-phase chromatography. Rinse the column by pipetting five times with 10mL 100% ACN, followed by five times 50% ACN and five times with 1% TFA in water. Dilute the peptide mixture in 1% TFA in water and apply to the column by pipetting up and down. Wash the column five times with 10mL of 1% TFA followed by five times 10mL 1% FA. Elute the peptides from the column with 2mL of 50% ACN and 1% FA. 6. Analyze the peptide mixture by MS, either by peptide mass fingerprinting on MALDI-TOF, or by generating peptide sequence tags on a tandem mass spectrometer, possibly in combination with LC. Submit the ion mass lists to protein identification search engines such as MASCOT to search the relevant protein databases to obtain unambiguous protein identification.
4. Notes 1. All water used during sample preparation and two-dimensional electrophoresis should be HPLC-grade or similar quality, having a resistance of 18.2MW. 2. A literature search should be performed in order to use the most efficient protein extraction protocol for each particular cell or tissue type. For many tissues or cell types, reproducible procedures have been described. Further optimization might be necessary to obtain optimal results. Try to keep the extraction protocol as simple as possible to prevent protein loss or the introduction of non-biological sources of variation.
Two-Dimensional Difference Gel Electrophoresis
453
3. To prevent degradation of the reactive N-hydroxy-succinimidyl group of the dyes, storage of the dyes should always be done in waterfree conditions. Therefore, always use fresh waterfree DMF for diluting the CyDyes and store the diluted working solution maximally up to 1week at −20°C. For prolonged storage of the dye, evaporate all solvents using a SpeedVac evaporator and store at 4°C or −20°C on silica gel. 4. 400pmol CyDye per sample is the amount of dye recommended by GE Healthcare Life Sciences. However, depending on the protein samples or to decrease the costs of the experiment, the amount of dye can be lowered to 200pmol, without any appreciable effect on the sensitivity of the protein detection. 5. Instead of the destreak solution, proteins can also be reduced during IEF by the addition of DTT to the rehydration solution. However, using destreak solution results in less horizontal streaking of the more basic proteins in the gel. 6. IEF is also possible in individual focusing trays. This enables the addition of the sample during IPG strip rehydration, by mixing the sample with the rehydration solution. This makes it also possible to place a low voltage across the rehydrating strip, that ensures a better uptake of high molecular weight proteins (11). 7. The total volthours should be in the range of 40–50kVh. The number of volthours should be increased when narrower pH gradients are used or higher amounts of protein are applied. However, to prevent over-focusing and associated horizontal streaking artifacts, the total number of volthours should never surpass 100kVh. 8. Do not exceed the given times of reduction and alkylation as this could result in the loss of protein. Perform this step as close to running the second-dimension as possible. 9. If the differences in expression levels between the spots are not large, it is possible to use the internal standard as the sample on the preparative gel(s), since only one sample then has to be run to excise all the differential spots. If differential spots are absent in several of the gels, it is usually better to separate the samples where these spots show the highest expression levels, to ease the identification of these spots. 10. Alternative protein stains are the fluorescent dyes SyproRuby or DeepPurple. The use of silver staining (see Chapter “Silver Staining of Proteins in 2DE Gels”) is discouraged, since this stain generally interferes with MS analysis, resulting in a lower chance of protein identification. 11. If cutting manually, do not cut the gel plugs too large, to avoid diluting the protein with non-protein contaminants. Perform all steps in a dust-free environment (laminar flow hood, gloves…) to prevent keratin contamination.
454
Van den Bergh
References 1. Liang, P., and Pardee, A. B. (1992) Differential display of eukaryotic messenger RNA by means of the polymerase chain reaction. Science 257, 967–971. 2. Diatchenko, L., Lau, Y. F., Campbell, A. P., Chenchik, A., Moqadam, F., Huang, B., et-al. (1996) Suppression subtractive hybridization: a method for generating differentially regulated or tissue-specific cDNA probes and libraries. Proc. Natl. Acad. Sci. U S A 93, 6025–6030. 3. Schena, M., Shalon, D., Davis, R. W., and Brown, P. O. (1995) Quantitative monitoring of gene expression patterns with a complementary DNA microarray. Science 270, 467–470. 4. Anderson, L., and Seilhamer, J. (1997) A comparison of selected mRNA and protein abundances in human liver. Electrophoresis 18, 533–537. 5. Ünlü, M., Morgan, M. E., and Minden, J. S. (1997) Difference gel electrophoresis: a single gel method for detecting changes in protein extracts. Electrophoresis 18, 2071–2077. 6. Tonge , R. , Shaw, J. , Middleton , B. , Rowlinson, R., Rayner, S., Young, J., et al. (2001) Validation and development of fluorescence two-dimensional differential gel electrophoresis proteomics technology. Proteomics 1, 377–396.
7. Alban, A., David, S. O., Bjorkesten, L., Andersson, C., Sloge, E., Lewis, S., and Currie, I.(2003) A novel experimental design for comparative two-dimensional gel analysis: Two-dimensional difference gel electrophoresis incorporating a pooled internal standard. Proteomics 3, 36–44. 8. Van den Bergh, G., and Arckens, L. (2004) Fluorescent two-dimensional difference gel electrophoresis unveils the potential of gel-based proteomics. Curr. Opin. Biotechnol. 15, 38–43. 9. Shaw, J., Rowlinson, R., Nickson, J., Stone, T., Sweet, A., Williams, K., and Tonge, R. (2003) Evaluation of saturation labelling two-dimensional difference gel electrophoresis fluorescent dyes. Proteomics 3, 1181–1195. 10. Qu, Y., Moons, L., and Vandesande, F. (1997) Determination of serotonin, catecholamines and their metabolites by direct injection of supernatants from chicken brain tissue homogenate using liquid chromatography with electrochemical detection. J. Chromatogr. B Biomed. Sci. Appl. 704, 351–358. 11. Görg, A., Obermaier, C., Boguth, G., and Weiss, W. (1999) Recent developments in twodimensional gel electrophoresis with immobilized pH gradients: wide pH gradients up to pH 12, longer separation distances and simplified procedures. Electrophoresis 20, 712–717.
Chapter 30 Protein Expression Profiling Brian P. Bradley, Bose Kalampanayil, and Michael C. O’Neill Summary Protein expression profiling is defined in general as identifying the proteins expressed in a particular tissue, under a specified set of conditions and at a particular time, usually compared to expression in reference samples. This information is useful in drug discovery and diagnosis as well as in understanding response mechanisms at the protein level. We may identify all the proteins responding to a particular stimulus and select those whose expression changes most. Or we may isolate significant protein variables and then identify them. These definitive sets of proteins (protein expression signatures; PES) are specific to diseases, toxicants, physical stresses, and to degrees of stress severity. Here we describe a method, based on machine learning, for isolating the sets of proteins, before identifying them by name, which classify accurately the treatment classes in a study. The principle in this chapter is that if proteins associated with known classes of interest can be used to identify unknown classes then the proteins are definitive for diagnosis. The proteins in each class, including controls, are converted to digital data and serve as input to artificial neural network (ANN) models. Multiple two-dimensional electrophoresis (2DE) gel patterns are included in each treatment class. A training subset of digitized individual, not composite, gel images is used to construct an ANN model which is then applied to a test set of images. Successful classification of the unknown (test) data confirms that the variables included in the model are indeed significant in discrimination among the classes. In the study described here the misclassifications were 5% or less using the ANN models. The ANN method seems to be a useful complement to image analysis, described in Chapter “Troubleshooting Image Analysis in 2DE”. The reduction in protein variables permits multivariable statistics such as cluster and discriminant analyses. Key words: Protein variables, Neural networks, Classification, Diagnosis, Machine learning, Protein expression signatures.
David Sheehan and Raymond Tyther (eds.), Methods in Molecular Biology, Two-Dimensional Electrophoresis Protocols, vol. 519 © Humana Press, a part of Springer Science + Business Media, LLC 2009 DOI: 10.1007/978-1-59745-281-6_30
455
456
Bradley, Kalampanayil, and O’Neill
1. Introduction In protein expression profiling we begin with the simple fact that proteins contain specific information about stressful conditions internal or external to an organism. To some this means constructing protein microarrays analogous to DNA microarrays (1). Alternatively, on a more practical level, it is now possible to identify all the expressed proteins under particular conditions or in a particular tissue (2; see also Chapters “Two-Dimensional Electrophoresis (2DE): An Overview,” “Solubilization of Proteins in 2DE: An Outline,” and “Difficult Proteins”). In the approach discussed in this chapter, we describe how artificial neural networks (ANN) can reduce a set of expressed proteins to those needed to classify the conditions of interest. The method has special relevance to diagnosis (or prognosis) of a particular stress. There is increasing interest in statistical analysis to simplify protein responses (3–8), and iterative procedures such as ANN have been used with considerable success by one of us in microarray analyses (9). The basic idea in using protein patterns in diagnosis is to separate irrelevant variability (among samples, tissues or organisms) from that among treatments of interest. Composite 2D or 3D images from control samples are subtracted from composites for each treatment, thus producing what we call protein expression signatures (PES). These consist to of proteins or subunits differentially up- or down- regulated which are then compared among chemicals, species, concentrations, or other treatments to isolate the components which make the PES specific. The methods are illustrated in Bradley et al. (10). The imaging software, guided by a patient operator, can compare simultaneously more treatments than should ever be included in one experiment. PES may be obtained for any stressor, level of stressor, class of stressor, indeed for any classification desired. A PES for generic stress might include heat shock proteins and possibly acute phase proteins, non-specific proteins responding to stress. The methods used to derive PES in this laboratory have been based on image analysis with versions of MELANIE and PDQuest (BioRad) software. The principles are described elsewhere in this volume (e.g. see Chapter “Troubleshooting Image Analysis in 2DE”). The main finding of the PES work so far is the high level of specificity, now well established (11–15), both to specific toxicants, diseases, and even physical stresses as well as to levels of exposure. The thousands of variables available, even without the important post-translational modifications, make this specificity highly likely a priori.
Protein Expression Profiling
457
The PES method reduces the number of protein variables by eliminating those not consistently present or consistently absent from treatment classes. However it seems that intra-class variability can be used in an analysis if only to create a measure of error variance. Thus proteins can be treated as a multivariable system and analysed by conventional methods, statistical, and otherwise. One such method is an iterative procedure based on machine learning. The ANN is a parallel computational model inspired by the structure of brains and nerve systems. A typical neural network consists of a group of inter-connected processing units, which are also called neurons. Each neuron makes its independent computation based upon a weighted sum of its inputs and passes the results to other neurons in an organized fashion. Neurons receive input data from the input layer, in the present case the proteins, while output is generated to users from the output layer, in this case the treatment groups being profiled. Neural networks are trained to solve problems or make predictions by analysing data with both input and output information that is under supervision. ANN performance is measured in two ways, training and testing. Data move from the input layer to a hidden layer, then to the output layer during training. A commonly used version of ANN is back propagation (16), a form of supervised learning where the error rate is sent back through the network to alter the weights to improve prediction and decrease error. A hidden layer of neurons determines the relationship between the input and output layers through the assignment of weights. Examples of untrained and trained networks are shown in Fig. 1. A more detailed diagram of a trained network is shown in Fig. 2. The ANN method complements the imaging mentioned earlier in this volume (see Chapter “Troubleshooting Image Analysis in 2DE”) (see Note 1). The method depends on gels being manipulated for consistency, also necessary for the imaging analysis. Assays of digitized proteins are then used to teach the computer the systematic differences among treatments. The model constructed (i.e. what the computer has learned) is then tested on a different set of treatments and control proteins. If the classification is perfect or close to it, this confirms that what the computer learned was in fact the set of proteins needed to accurately identify each treatment. These proteins can be the basis of a diagnostic assay. Questions of statistical confidence in and generalization of the proteins isolated follow the initial models, but are not discussed here (Note 2). The general approach to protein profiling, of which ANN is one example, is to treat the thousands of proteins as a multivariable
458
Bradley, Kalampanayil, and O’Neill
Fig. 1. Trained and untrained networks. In the top diagram the weights are arbitrary, prior to minimizing errors by the back propagation method. In the lower diagram the network is fully trained to identify the one output class from the six input neurons. Neurons 1, 3, 4, and to a lesser extent 5 appear to be most predictive.
system. The important step, in our view, is not identifying all the proteins expressed under a particular condition but rather first to isolate the minimum number of protein variables needed to classify (detect) the condition with near perfect accuracy. So, if statistical analysis has greater appeal than ANN or if you wish to do both, the method of data preparation, if not the analysis, is relevant to you.
Protein Expression Profiling
459
Fig. 2. A network with six input neurons, two middle layer neurons, and three output neurons. The specific input datum is 001010. The weights at this time are indicated by the arrows to each of the inter-layer lines. The values shown inside the middle layer neurons are obtained after summing the weights on the incoming lines and applying the sum to the transform function. The source process is respected at the output neurons.
2. Materials 2.1. The Data
The original data were derived from a PhD thesis on UVb damage and repair in strains of Japanese medaka (Oryzias latipes).The gels were quite consistent, run at that time without the benefit of immobilized pH gradient (IPG) strips (see Chapter “Two-Dimensional Electrophoresis (2DE): An Overview”). More important, there were 22 treatment classes, providing a stern test for the ANN models. Medaka larvae were exposed to various combinations of UV light treatments by Tina Armstrong (17). There were four pigment strains (leucophoreless, orange, white, and wild type) and 6 relative levels of UVb (0, 2, 5, 10, 20, and 30% fluence rates). Gels were produced using methods described elsewhere in this volume (e.g. see Chapters “Two-Dimensional Electrophoresis (2DE): An Overview” and “Organelle Proteomics”). The gels we used for analysis were run on the Mini-Protean II system by BioRad. Scanning was done on a cheap scanner, the 636 by Epson.
2.2. Gel Preparation
Gels used were selected from high-quality images without any processing prior to being transformed to digits. We recommend the method described later using imaging software to remove features which are not protein spots and to align the gels. The software with which we are most familiar is PDQuest (Biorad), but we have used MELANIE versions with equal success.
460
Bradley, Kalampanayil, and O’Neill
2.3. Co-Ordinates for the Protein Spots
Image Quant (Molecular Dynamics, USA) version 5.2 was used to create co-ordinates for the protein spots on each image. The software is available on-line with simple directions for adjusting the mesh size of the grid, as described later.
2.4. Neural Network Analysis
The software used for the neural network analysis was from NeuralWorks, based on a supervised learning method known as back propagation. This method has yielded excellent classification results (16) (see Notes 5, 6).
3. Methods 3.1. Gel Scanning
We begin with the 2DE gel images, obtained by methods discussed extensively in this volume. Each 2DE gel is scanned as a TIFF image on a flatbed scanner with a transparency unit attached to a computer. The neural net analysis requires gel-togel (or image to image) consistency. This means that protein spots should appear in close to the same position in every image. Also, of course, a spot should represent a protein. There are steps that are important in preparing the 2DE images for analysis. Some of these are included in the proprietary software such as Melanie.
3.2. Image Adjustment
Gel images can be adjusted by commercially available software such as the Melanie series or PDQuest. The following are the relevant steps, embedded in the software and based on proprietary algorithms: (a) Gel cleanup – removal of markings on the image such as streaks (b) Detection of the protein spots and removal of all others (c) Warping of the images for proper alignment When the gel images are adjusted in this way, for making composites for each treatment, they are ready for the next step in preparing the protein data for ANN analysis. Composite or matched images are not produced for ANN analysis or for standard statistical analysis. Variation within treatment groups is part of the analysis.
3.3. Grid Co-Ordinates
The protein data are next reduced to numbers (binary in this case but quantitative if necessary for diagnosis) by first imposing a grid (ImageQuant, NIH) on each gel. The mesh or number of rows and columns depends on the (population) density of the spots and on the uniformity of the gels. The grid size is the smallest possible with particular spots being within the same grid coordinates or square on each gel image (Note 3). The grid mesh must be appropriate for all gels in the experiment (Fig. 3a).
Protein Expression Profiling
3.4. Dots to Digits
461
Figures 3b, c illustrate the final step in data preparation, the conversion of proteins (dots) to digits. In the present case, described later (Notes 4, 5), a binary scale (0 or 1) was sufficient for complete success in digital pattern recognition by the ANN models. The output of the gel reader (Image Quant) is converted to a listing of row by column (900 for example) real numbers, one per square and stored in ‘list txt’, an Excel file in our case. The logical steps required (pseudo code), from Answer Code Translation Language, are as follows: For 1 to N (gel images, 66 for example) ‘What is answer code for this?’ Read answer code For 1 to 900 do Read number from example) = x
list
txt
(from
an
Excel
file
for
If x > background, write output file = 1 Otherwise write output file ‘0’ End New line in output file, write Add the output file answer code, e.g. ‘100’, to the end of the input example New line in output file End Output file example:#1 0010001…..100 (903 digits, including the three digits to indicate the class) #2 1001001….011 (903 digits) The output file including the code for the treatment becomes the input file for the ANN.
Fig. 3. Stages in digitizing proteins, from dots to numbers. The numbers are binary here but could be semi-quantitative or continuous based on spot volume.
462
Bradley, Kalampanayil, and O’Neill
3.5. Running the Neural Network
The network is first run in training mode (Note 5). In this phase, the network is not only given the protein pattern of the sample, it is also provided with the correct classification for that sample. Initially the network has all random weights of both signs; it knows nothing; it outputs a value around 0.5 for each output neuron. But it knows what it was supposed to produce. Suppose the output codes are 0 0 1 for control, 0 1 0 for treatment A, and 1 0 0 for treatment B. If the example was a control, it should have produced 0 0 1. The program looks at the error it just made at each output neuron and, using a gradient descent algorithm, it changes the weights on the lines coming to that output neuron in a direction to reduce the error just made. It continues this process while it is presented examples of all three types until the error is satisfactorily small (less than 5% perhaps) or until it has correctly classified all training examples. At this point, all the weights in the network are permanently fixed (see the legend in Fig. 1 for untrained and trained networks). Actual weights of a trained network are shown in Fig. 2, cited earlier. If the training sample has been representative of the greater population, the network should now be ready to classify new
Table 1 Sample output (one Medaka strain) from a trained neural network 0.000000
0.000000
0.000000
0.000000
0.000000
1.000000
0.025053
0.027320
0.022863
0.022137
0.034030
0.941891
0.000000
0.000000
0.000000
0.000000
1.000000
0.000000
0.012317
0.024978
0.016354
0.019436
0.928853
0.029998
0.000000
0.000000
0.000000
1.000000
0.000000
0.000000
0.010167
0.014812
0.025595
0.933654
0.016040
0.024093
0.000000
0.000000
1.000000
0.000000
0.000000
0.000000
0.027377
0.031309
0.920119
0.016226
0.016754
0.023446
0.000000
1.000000
0.000000
0.000000
0.000000
0.000000
0.028056
0.918995
0.032352
0.022820
0.020346
0.025785
1.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.933288
0.022873
0.016772
0.009487
0.014149
0.013151
First row gives expected output of treatment 1 (the lowest of six uvb levels in the test case, code 000001). The second row gives actual output. The actual value is 0.94 which is close to 1. Similarly for the other five treatments. Results of all analyses are given later (Notes 5–7)
Protein Expression Profiling
463
samples without benefit of the answers. If the training set is not broad, there is the possibility that the network will memorize each training example by some exclusive ‘tell’ and thereby fail to generalize on the problem. This sort of ‘overtraining’ can generally be recognized easily because it is characterized as producing excellent training results and poor test results. An example of the output from the ANN analysis is shown in Table 1.
4. Notes 1. In this chapter we have described a method not intrinsic to 2DE gel electrophoresis. While the ANN method is intended to reduce the number of protein variables in a sampled proteome, the concept of variables rather than proteins with names and functions will not appeal to all readers of this volume. Our intent is not to replace some existing methods but rather to add formal data analysis, even an iterative method, to the profound advances in genome sequencing (ergo, protein databases), in protein separation, in pattern imaging and most of all in mass spectrometry (MS). Subtraction assays followed by high-throughput LC tandem mass spectrometry (sequencing) have great appeal. Depending on the database, you can end up with a definitive list of proteins involved in the response of interest. Then you decide which may be important. Our modest proposal is to start the identification with a subset of number of candidate protein variables (‘spots’) which diagnose or predict a condition of interest, rather than ‘draining the pond’ and identifying everything in it. 2. Obviously the analysis is only as good as the samples and to produce protein indicators for general use the analysis will need to be repeated over a wide variety of populations. This is assuming the goal is a diagnostic tool which can be applied to all ages and populations. We do not know at present if it is realistic to expect robust sets of indicators applicable to all populations, even subsets of one species. 3. The question of grid mesh is important. We do not know how forgiving the analysis is when many spots are not clearly in one location or square, from one image to another. We seek to include the largest number of spots possible while maintaining consistency of location of corresponding spots. We have not done the direct test of how grid mesh relates to accuracy of ANN classification. 4. The ANN method has been demonstrated on Medaka. Scanned 2DE gels from an experiment on Medaka fish
464
Bradley, Kalampanayil, and O’Neill
exposed to five levels of UVb radiation and a control were arranged in a grid using Image Quant software (Molecular Dynamics, USA). A grid (30 × 30), found to be optimal, was superimposed on each gel, and a table of channel pixel volume-percents converted to binary scale was produced as input to the neural net model. The data were divided into training and test sets; training and testing were performed as described previously. A total of 66 gels from four strains of medaka fish was tested using the neural net model. The 900grid squares were graded as containing a protein spot or not (intensity data could also have been used). In this case the input to our network was a set of 900 0’s or 1’s representing unoccupied and occupied squares on the grid, one for each individual fish in the sample. 5. During training, the networks were trained to an error level below 0.05, and then tested with previously unseen data. The networks consisted of 1,800 inputs (coarse coded two digits per data point, 01 or 10 for spots absent or present); 25 hidden neurons; and 4, 6, or 22 output neurons depending on the classification tested. Four strains were tested independently for various treatments. Of three gel images from each level of UVb, two were used to train and one to test. Training and testing were repeated in round-robin fashion for a total of 18 test patterns from the six treatments. As mentioned earlier, mortality of the wild-type strain reduced the levels and thus the number of tests run. As seen in Table 2 ANN models correctly classified 89–100% of the times during the independent testing of each strain. Testing was done by strain ignoring treatment (four outputs with 4–6 tests). In a total of 66 tests, with the round-robin design, ANN models correctly identified the treatments 62 times (Table 2). Similarly when levels of uvb were tested ignoring or pooling strains the result was also 62 of 66 correct. The probability of any of these results by chance a priori is essentially zero. An important inference from the latter results is that classification can be at different levels. For example proteins defining exposure to a toxicant at any level or defining the type of toxicant or diagnosing a disease regardless of severity could be isolated by ANN models. 6. The results with the Medaka data were confirmed in tomato data collected as part of a PhD thesis by the middle author. Proteins from root tissues of tomato plants exposed to four levels of salt over three time periods were analysed as described for Medaka. There were four tomato plants per treatment class, a total of 48. The results of the ANN analysis in Table 3 show a low error rate, around 6%. What the models were detecting were the time and level of exposure of individual plants based on significant bias in proteins expressed and
Protein Expression Profiling
465
Table 2 Summary of results (inferred from output as shown in Table 1) from Medaka fish exposed to six levels of uvb Strains/Groups
Treatments (UV-B Fluence rates)
Correct recognition of treatment
Leucophoreless
0, 2%, 5%, 10%, 20%, 30%
18/18
Orange-red
0, 2%, 5%, 10%, 20%, 30%
16/18
White
0, 2%, 5%, 10%, 20%, 30%
18/18
Wild type
0, 5%, 20%, 30%
11/12
Separate strains
All treatments
62/66
All strains
Separate treatments
62/66
Wild-type fish did not survive the highest UVb fluence rates; thus, only four levels are included in the wild-type data. Eighteen images were tested in three sets (two images for training and one for testing, six treatments per strain). Results are shown also for strain classes with treatment ignored and for treatments with strain ignored
Table 3 Results for tomato classifications by protein profiles Time
Treatments (Nacl-mM)
Correct recognition of treatments
Day 1
0
0 error/16 tests
100 200 300 Day 3
0
2 errors/16 tests
100 200 300 Day 7
0
1 error/16 tests
100 200 300 Total
3 errors/48 tests
Both time and concentration are discernible in the protein patterns
466
Bradley, Kalampanayil, and O’Neill
disregarding protein variation not related to the treatment classes. Such pattern recognition in a biological system seems no less remarkable in tomato than in Medaka. 7. Results of the ANN analyses reported here are corroborated by the results of DNA microarray analyses done by the third author of this chapter (9). He has shown in this article and in other unpublished work a remarkably high level of accuracy of prognosis and of diagnosis of forms of cancer in particular. In one study he showed with lymphoma microarray data that a sample of around 60 genes out of a total of 4,026 genes was almost 100% accurate for prognosis and diagnosis of lymphoma. The prognostic and diagnostic sets were actually completely distinct. 8. Measures of statistical significance or level of confidence are not routinely parts of ANN models. However multiple trials may provide a measure of error (or confidence) not possible with one ANN. It is feasible to train a series of ANNs using, say, 90% of the examples for training and holding back 10% for testing. A different 10% can be tested in a second ANN and so on. In this way, with the training of ten ANNs, each input can be found in a test set one time and can, therefore, be independently evaluated. The fact that one ends up with ten ANNs is not an impediment to analysis since any future examples could be submitted to all ten ANNs for evaluation, with a majority poll deciding the classification. These ANNs are, of
Table 4 Numbers of proteins determined to be distinct to class and thus useful for separating treatment classes Strains Treatments
Wild type
White
Orange-red
Leucophoreless
0
12
6
13
3
2%
NA
8
9
5
5%
11
12
6
5
10%
NA
0
12
6
20%
10
7
5
7
30%
10
10
7
8
Total spots
43
43
52
34
Protein variables identified by differentiation of the original output, a proprietary method
Protein Expression Profiling
467
course, likely to be very similar if their training sets differ only slightly. A second major advantage of ANNs follows from this. Not only are ANNs trained to the specific question and tested for generality, but they can be asked for a quantitative assessment of how they got the correct answer. Numerical partial differentiation of the ANN with respect to a given test input example allows one to see the ANN evaluation of the relative impact of each protein in arriving at the correct answer for this particular input. This is not a statistical add-on but an inherent property of the trained network. Given this ability, ANNs should be able to identify relatively small protein subsets that will significantly outperform the initial total protein sets in classification. The results of such an analysis of the trained network from Medaka are shown in Table 4. We realize that numbers are not names and the proteins in the subsets will need to be identified to satisfy most people. 9. The desired product of the ANN or any other profiling method is not the statistical or quantitative result in itself. In the end we need proteins, with names and functions, not just variables. And these proteins collectively will predict outcomes and diagnose conditions with close to 100% accuracy. The sheer number of protein variables available for our models makes this goal feasible for any type of condition needing detection.
Acknowledgments The authors are grateful to Dr Tina Armstrong for her excellent 2DE gels produced in the course of her PhD research with Dr Bradley. We acknowledge the organizers of two conferences at which this work was presented. One was a conference arranged by the Systems Biology Initiative at West Virginia University in Morgantown in summer 2005. The second was arranged by the SWIRE Institute of Hong Kong University in January 2007. As helpful as these conferences were all the remaining ambiguities are the responsibility of the authors.
References 1. Tomlinson, I.M., and Holt, L.J. (2001). Protein profiling comes of age. . Genome Biology 2, 1004.1––1004.3. 2. Ferguson, P.L., and Smith, R.D. (2003). Proteome analysis by mass spectrometry. . Annu. Rev. Biophys. Biomol. Struct. 32, 399–424.
3. Han, M-K., Hong, M-Y., Lee, D., Lee, D-E., Noh, G-Y., Lee, J-H.et al. (2006). Expression profiling of proteins in L-threonine biosynthetic pathway of Escherichia coli by using antibody microarray. Proteomics 6, 5929–5940.
468
Bradley, Kalampanayil, and O’Neill
4. Yamanaka, H., Yakabe, Y. Saito, K., Sekisima M., and Shirai T. (2007). Quantitative proteomic analysis of rat liver for carcinogenicity prediction in a 28-day repeated dose study. Proteomics 7, 781––795. 5. Biron, D.G., Brun, C., Lefeure, T., Lebarhenchon, C., Loxdale, H.D., Chevenet, F., et al. (2006). The pitfalls of proteomics experiments without the correct use of bioinformatics tools. Proteomics 6, 5577–5596. 6. Urfer, W., Grzegorc, M., and Jung, K. (2006) Statistics for proteomics: A review of tools for analyzing experimental data. Pract. Proteomics 1–2, 48–55. 7. Anderson, T.J., Tchernyshyov, I., Diez, R., Cole, R.N., Geman, D., Dang, C.V.et al. (2007). Discovering robust protein biomarkers for disease from relative expression reversals in 2-D DIGE data. Proteomics 7, 1197–1207. 8. Lehesranta, S.J., Koistinen, K.M., Masset, N., Davies, H.V., Shepherd, L.V.T., McNicol, J.W.et al. (2007) Effects of agricultural production systems and their components on protein profiles of potato tubers. Proteomics 7, 597–604. 9. O’Neill, M., and Song, L. (2003) Neural network analysis of lymphoma microarray data: prognosis and diagnosis near-perfect. BMC Bioinformatics 4, 13–25. 10. Bradley, B.P., Shrader, E.A., Kimmel, D.G., and Meiller, J.C. (2002) Protein expression
11.
12.
13.
14.
15.
16. 17.
signatures: an application of proteomics. Marine Environ. Res. 54, 373–377. Blom, A., Harder, W., and Matin, A. (1992) Unique and overlapping pollutant stress proteins of Escherichia coli. Appl. Environ. Microbiol. 58, 331–334. Kultz, D., and Somero, G.N. (1996) Differences in protein patterns of gill epithelial cells of the fish Gillichthys mirabilis after osmotic and thermal acclimation. Comp. Physiol. 166B, 88–100. Ostergaard, M., Rassmussen, H.H., Neilsen, H.V., Vorum, H., Orntoft, T.F., Wolf, H.et al. (1997) Proteome profiling of bladder squamous cell carcinomas: Identification of markers that define their degree of differentiation. Cancer Res. 57, 4111–4117. Shrader, E.A., Henry, T.R., Greeley, M.S., and Bradley, B.P. (2003) Proteomics in zebrafish exposed to endocrine disrupting chemicals. Ecotoxicology 12, 485–488. Shepard, J. L., Olsson, B., Tedengren, M., and Bradley, B.P. (2000). Protein expression signatures identified in Mytilus edulis exposed to PCBs, copper, and salinity stress. Marine Environ. Res. 50, 337–340. Werbos, P.J. (1994) The Roots of Back Propagation. New York: Wiley. Armstrong, T.N., Reimschussel, R., and Bradley, B.P. (2002) DNA damage, histological changes and DNA repair in larval Japanese medaka (Oryzias latipes) exposed to ultraviolet-B radiation. Aquat.Toxicol. 58, 1–14.
Chapter 31 C-Terminal Sequence Analysis of 2DE-Separated Proteins Bart Samyn, Kjell Sergeant, and Jozef Van Beeumen Summary The overall study of post-translational modifications (PTMs) of proteins is gaining strong interest. Beside phosphorylation and glycosylation, truncations of the nascent polypeptide chain at the N- or C-terminus are by far the most common types of PTMs. Nevertheless, little attention has been paid to the development of approaches that allow a systematic analysis of these proteolytic processing events. Here we present a protocol that allows the identification of the C-terminal sequence of proteins separated by twodimensional polyacrylamide gel electrophoresis (2DE). For each purified protein, a peptide mixture is generated by cleavage of the protein with cyanogen bromide. During incubation with carboxypeptidases only the original C-terminal fragment forms a ladder. Ladder readout is performed using MALDI mass spectrometry. 2DE-separated proteins from Shewanella oneidensis were chosen as a model system to investigate the effectiveness of the approach. Key words: MALDI-MS.
Post-translational
modification,
C-terminus,
Proteolytic
processing,
2DE,
1. Introduction Until recently, the only methods that allow direct confirmation of the termini of proteins are chemical degradation techniques. N-terminal protein sequencing by Edman degradation is still the method of choice to determine N-terminal proteolytic processing. A chemical method for C-terminal sequence analysis should provide a complementary approach. Despite the fact that such methods have been developed, there remain a number of unsolved problems, such as their limited sensitivity (20–100 pmol) and their low repetitive yields (1). Furthermore, several
David Sheehan and Raymond Tyther (eds.), Methods in Molecular Biology, Two-Dimensional Electrophoresis Protocols, vol. 519 © Humana Press, a part of Springer Science + Business Media, LLC 2009 DOI: 10.1007/978-1-59745-281-6_31
469
470
Samyn, Sergeant, and Beeumen
amino acids require chemical modification, and it remains difficult to sequence through proline (2). Alternatively, methods have been described that allow the isolation of the C-terminal peptide (3–4). However, they have rarely been applied, most likely due to problems associated with the recovery of the peptide and the need for larger sample amounts (4–5). Recently, we described a novel ladder approach that allows the systematic identification of the C-terminus of proteins (6–7). The mass spectrometry (MS)-based, enzymatic, ladder-sequencing approach can be applied to proteins separated by two-dimensional polyacrylamide gel electrophoresis (2DE; Fig. 1a). The purified protein in each gel spot is chemically cleaved with cyanogen bromide (CNBr). This reagent cleaves at the C-terminal side of methionine, converting it into a homoserine lactone (hsl), which can undergo hydrolysis to form homoserine. Acidic conditions favor the formation of hsl, whereas under basic conditions the free acid is formed. The unseparated peptide mixture is subsequently incubated with a carboxypeptidase (CP). Internal peptides ending in an hsl are resistant to CP action. Therefore, only the original C-terminal fragment, having a free carboxyl group, is accessible to enzymatic degradation and forms a ladder (Fig. 1b). Ladder readout is performed by matrix-assisted laser desorption ionization (MALDI) mass spectrometry (MALDI–MS). To illustrate the applicability of this approach at a proteomic scale, we characterized a number of C-terminal sequences from 2DE-separated proteins of the bacterium Shewanella oneidensis MR-1 (6). Since this bacterium is able to reduce a variety of metal substrates, it is of considerable interest to researchers involved in bioremediation (8). Therefore, a total protein extract from aerobically grown Shewanella MR-1 was separated by 2DE. After Coomassie staining, 25 spots were randomly selected and subjected to the new protocol (Fig. 2). In general, a positive identification of the C-terminus will depend on the length and ionization capacity of the generated CNBr fragments. In this approach, MS analysis was performed using MALDI ionization which generates predominantly singly charged ions, allowing a simple and straightforward interpretation of the ladders. Experiments were performed on a MALDI TOF/TOF instrument with MS/MS capabilities. The use of a α-cyano-4-hydroxycinnamic acid (α-CHCA) matrix produces interference (matrix clusters) in the low MW mass range and, therefore, presents a challenge for the analysis of peptides with a MW below 0.7–1.3 kDa. However, the strength of the ladder sequence approach appears to be more pronounced with larger peptides. In our experience, the upper mass limit for the analysis of ladder sequences in reflectronMALDI–MS, providing enough resolution and accuracy to identify amino acids by 0.1 Da difference, is restricted to C-terminal fragments with a MW of ±5 kDa.
C-Terminal Sequence Analysis of 2DE-Separated Proteins
471
Fig. 1. Schematic representation of the different steps in the C-terminal sequencing method. (a) Proteins separated by 2DE are cleaved in gel with CNBr after destaining. Removal of excess salts is accomplished by washing the gel pieces prior to chemical cleavage. The resulting CNBr fragments are extracted from the gel or the PVDF, pooled and dried, and finally redissolved in buffer for incubation with CPY. (b) After reduction, iodoacetamide (IAA) reacts with the ionized, free –SH groups of Cys at alkaline pH values, resulting in the formation of a carboxyamidomethylcysteine (camC) derivative (Δm = + 57.02 Da). CNBr cleavage results in the formation of internal fragments ending at a homoserine lactone (hsl) derivative. Only the peptide containing the original C-terminal sequence (Xxx-Yyy-Zzz) is accessible for enzymatic degradation (CPY) and forms a ladder. MALDI analysis of the ladders is performed on a 4700 TOF/TOF instrument (6).
In this study, 21 proteins (with MW ranging from 10 to 57 kDa) out of 25 spots could be identified by combining the peptide mass fingerprint analysis of the CNBr fragments and Cterminal sequence information to search against a database on a local MASCOT server (9). Upon incubation with CPY, we observed ladder formation for 11 proteins, and one to ten amino
472
Samyn, Sergeant, and Beeumen
Fig. 2. 2DE analysis of Shewanella oneidensis proteins. 2DE-separated proteins from aerobically grown Shewanella MR-1. ±100 μg of total cell extract is loaded on an IPG (4–7) strip and analyzed as described (see Chapter “De Novo Sequence Analysis of N-Terminal Sulfonated Peptides After in-Gel Guanidination”). The numbered spots are subjected to C-terminal sequence analysis (6).
acids were cleaved from the C-terminal peptide (Table 1). In some spots, we observed four CNBr fragments, e.g., in spot seven, resulting from the cleavage of the translation elongation factor Ts. However, the C-terminal fragment had a MW of 5,430.9 Da; too large to be measured in the reflectron mode. Database analysis of the other identified proteins of which the C terminus was not characterized (ten spots) indicated that all of them, except two, had a C-terminal peptide with a MW below 1,000 Da or higher than 5,000 Da (Table 2). Peptide fragments with such MW cannot be analyzed at this sensitivity with sufficient resolution in the reflectron mode. The fact that we did not observe ladder formation for spots 15 and 24, both having a C-terminal peptide with a MW in the correct range, indicated that these fragments were not efficiently extracted or did not ionize well.
1,254.68
30.5
[GI_24373198][NP_717241]
7
2,339.25
Conserved hypothetical protein
29.3
2,090.04
6
1,254.68
[GI_24376191][NP_720235]
27.1 2,026.28 2,416.36
5
Uridine phosphorylase
[GI_24375619][NP_719662]
Conserved hypothetical protein
2,951.66
20.4
[GI_24374881][NP_718924]
4
1,514.93 2,470.37 2,537.33
Hypothetical protein S00554
1,288.68
19.7
[GI_24372148][NP_716190]
3
2,673.63 3,667.29
Universal stress protein family
15.6
1,367.86
2
1,489.81
[GI_24375179][NP_719222]
12.5
Asn201-Met212
Thr85-Met107
Ile257-Glu274
Lys233-Ala252 Leu14-Met36
Ala2-Met13
Lys165-Lys191
Leu162-Leu174 Arg97-Met118 Ser119-Met141
Ser145-Met156
Ala27-Met50 Val109-Lys143
Ile51-Met63
Ser89-Lys122
Ser2-Met15
MW (kDa) MW Obs (Da) CNBr fragment
c
3,495.93
1
Spot
b
Ribosomal protein L7/L12
[GI_24371821][NP_715863]
Protein
a
1,253.60
2,338.14
2,089.04
2,025.26 2,415.32
1,253.65
x
x
(continued)
+(FKATYSE)
+(LA) x
x
+(RK)
+(KL) x x
1,497.89f 2,469.29 2,536.24 2,950.61
x
x +(AVCPVLVVK)
x
+(EIK)
x
CPX-sequencee
1,287.62
2,672.38 3,665.95
1,366.74
3,494.87
1,488.76
MW Calc d (Da)
Table 1 C-terminal sequence analysis of 2DE-separated proteins of Shewanella oneidensis (first published in reference 6)
C-Terminal Sequence Analysis of 2DE-Separated Proteins 473
Fructose-bisphosphate aldolase II
38.8
3,037.39 3,272.52 3,730.61
1,680.95
11
1,041.58
[GI_24346525][Aan54007]
37.3
1,062.70 2,337.26
10
Leucine dehydrogenase
[GI_24374179][NP_718222]
Malate dehydrogenase
1,761.12
32.3
[GI_24372359][NP_716401]
9
2,083.48 2,280.46 2,637.73 2,771.91 3,070.18 3,293.19 3,451.42
Methylisocitrate lyase
31.9
2,015.48
8
[GI_24371943][NP_715985]
MW Obs c MW (kDa) (Da) 1,738.95 1,840.02 3,457.69
Spot
b
Translation elongation factor Ts
Protein
a
Table 1 (continued)
Glu313-Met340 Ala2-Met31 Thr67-Met99
Tyr341-Leu355
Ala335-Ala344 Trp45-Met65
Ile125-Met133
3,036.49 3,271.65 3,729.73
1,679.89
1,061.60 2,320.08f
1,040.47
1,760.01
x x x
+(L)
+(AKA) x
x
+(FVK)
x x x x x +(DK) x
2,082.12 2,279.07 2,636.27 2,770.42 3,068.64 3,291.62 3,449.82g Val136-Met154 Thr210-Met230 Glu113-Met135 Ala155-Met182 Ile183-Met209 Gln266-Lys292 Met1-Met32 Leu296-Lys311
x
x x x
2,014.14
1,737.90 1,838.97 3,456.73
CPX-sequencee
Val231-Met249
Lys218-Met233 Ala2-Met19 His169-Met200
CNBr fragment
MW Calc d (Da)
474 Samyn, Sergeant, and Beeumen
12
43.5 2,555.45
2,532.63 Asp370-Ala394
Pro114-Met135 2,554.43
2,531.41 +(GAGVVAKIIA)
x
b
NCBI Entrez entries (http://www.ncbi.nih.gov/Entrez/) Spot number according to the position on the 2DE (Fig. 2). The position of the CNBr fragments in the protein sequence is indicated in Arabic numbers c MW observed in positive mode reflectron analysis (singly protonated) d MW calculated by using the residual monoisotopic values with Met → hsl, and Cys → >camC e x indicates that no ladder formation was observed, whereas + indicates ladder formation (the observed amino acid sequence is indicated from N – >C) f CNBr fragment containing an oxidized Trp (Δm = 16 Da) g Indicates a CNBr fragment with a missed Met-Xxx cleavage
a
Elongation factor Tu
[GI_416942][P33169]
C-Terminal Sequence Analysis of 2DE-Separated Proteins 475
476
Samyn, Sergeant, and Beeumen
Table 2 Proteins of Shewanella oneidensis identified by PMF-analysis of CNBr fragments without characterization of the C termini (first published in ref.6) Proteina
Spotb
MW (kDa)
Calc MW of C-terminal peptide (Da)c
Not identified
13
–
–
Not identified
14
–
–
[GI_32171551] [Q8EAH2] 30S ribosomal protein S6 15
15.0
3,206.5
[GI_24372802] [NP_716844.1] Purine nucleoside phosphorylase
16
25.6
817.47
[GI_24373389] [NP_717432.1] Hypothetical protein 17 SO1825
25.2
5,185.78
[GI_24375475] [NP_719518.1] Aerobic respiration control ArcA
18
27.2
5,687.93
Not identified
19
–
–
[GI_24347242] [AAN54551.1] Alcohol dehydrogenase II
20
40.0
7,147.7
Not identified
21
–
–
[GI_30315823] [Q8EBR0] 2-Phosphoglycerate dehydratase
22
45.6
7,843.13
[GI_24373359] [NP_717402.1] Trigger factor
23
47.6
717.4
[GI_24376221] [NP_720265.1] ATP synthase F1, alpha subunit
24
55.1
4,189.11
[GI_24372295] [NP_716337.1] Chaperonin GroEL
25
57.0
150.06
a
NCBI Entrez entries (http://www.ncbi.nih.gov/Entrez/) Spot number according to the position on the 2DE (Fig. 2) c Calculated MW based on residual monoisotopic mass values b
It should be noted that longer incubation times with CP may result in a decrease in ion intensity of some of the fragments. Therefore, in order to obtain a complete sequence, without gaps, it is necessary to analyze the ladder fragments at different time points (10). Furthermore, the rate at which amino acids are cleaved by CP depends to a great extent on the peptide sequence. In experiments where only a few C-terminal residues are removed, indicating that this peptide represents the C-terminal fragment, the peptide can be subjected to MALDI MS/MS fragmentation to obtain more sequence information (Table 1). In those cases the C-terminal CNBr fragment was selected for MS/MS analysis. This approach is demonstrated using the CNBr peptide map
C-Terminal Sequence Analysis of 2DE-Separated Proteins
477
Fig. 3. C-Terminal sequence analysis of 2DE-separated proteins from Shewanella MR-1. (a) Mass spectrum in the positive reflectron mode of CNBr fragments from the gel spot containing uridine phosphorylase. (b) Upon incubation with CPX only two C-terminal amino acids are removed from the C-terminal fragment at m/z 2,026.28 (Lys233Ala252, NCBI Entrez [GI_24375619]). (c) MALDI MS/MS fragmentation of the C-terminal fragment (precursor labeled with an asterisk) results in a continuous stretch of b-ions from which 13 C-terminal amino acids are identified (indicated in one-letter code) (6).
478
Samyn, Sergeant, and Beeumen
of uridine phosphorylase from which, upon incubation with CP, only two C-terminal amino acids were removed (Fig. 3). The peptide at m/z 2,026.28 was selected as precursor and yielded, upon fragmentation, a continuous stretch of b-ions from which 13 C-terminal amino acids were identified.
2. Materials 1. Acetonitrile (ACN), assay (GC) min. 99.9% (Biosolve). 2. Ammonium acetate, GR for analysis ACS reagent (Merck). 3. Ammonium hydrogen carbonate, Biochemica ultra >99.5% (T) (Fluka). 4. Carboxypeptidase Y (CPY), sequencing grade from yeast, Roche Applied Science. 5. Cyanogen bromide (CNBr), 5 M in ACN, Sigma. 6. Trifluoroacetic acid (TFA), R3 40 mL, Applied Biosystems. 7. α-CHCA; >98%(TLC) powder, Sigma.
3. Methods 3.1. 2de
2DE was performed essentially as described in chapter “De Novo Sequence Analysis of N-Terminal Sulfonated Peptides After inGel Guanidination”.
3.2. Wash Procedure for Gel Spots (see Note 1)
1. After staining and destaining the 2DE gel, excise the gel spot of interest and transfer to a clean Eppendorf. 2. Wash the gel spot twice with 200 mM NH4HCO3/50% ACN/ MQ, prepared by dissolving 3.16 g ammonium bicarbonate in 100 mL ACN and 100 mL Milli-Q water (MQ). Add 150 μL to the vial and keep at 37°C for 30 min. Discard the wash solvent and repeat the wash step. 3. Discard the wash solvent again, and dehydrate the gel spot with 40 μL ACN (100%). 4. Dry the gel spot briefly in a Speedvac.
3.3. CNBr Cleavage (see Note 2)
1. Reswell the dried gel spot in 5 μL MQ water/15 μL TFA, and add 5 μL CNBr (5 M in ACN) to start the CNBr-cleavage. 2. Wrap the Eppendorf vial in aluminum foil and incubate at 4°C for 16 h (CNBr is light sensitive). Caution: TFA is corrosive
C-Terminal Sequence Analysis of 2DE-Separated Proteins
479
and CNBr is highly toxic and corrosive; these reagents should only be used under a fume hood by experienced personnel wearing protective clothing and safety glasses. 3. After incubation, pipette the supernatant in a new Eppendorf. Extract the CNBr peptides from the gel spot by adding 50 μL 70% ACN containing 0.1% TFA. Keep the vial at 37°C for 30 min and transfer the supernatant to the same Eppendorf tube. Repeat the extraction step and dry the pooled fractions in a Speedvac. 4. The dried peptides can be frozen at −20°C for several months. 3.4. Incubation with Carboxypeptidase Y (see Notes 3 and 4)
1. Redissolve the dried peptide sample in 10 μL 10 mM ammonium acetate (pH 5.4) and briefly vortex to improve solubilization. Note that the cleavage rate of carboxypeptidases depends on reaction conditions such as pH, ion strength, and substrate concentration. 2. Redissolve 20 μg sequencing-grade CPY in 70 μL MQ water. According to the manufacturers’ instructions, this results in a stock solution containing 5 pmol/μL of enzyme and 40 mM sodium citrate buffer (pH 6.0). Divide the stock solution into several smaller aliquots (±3 μL) and store at −20°C. 3. Defrost one carboxypeptidase stock aliquot and add to the redissolved peptide mixture in an enzyme to sample ratio of 1/50. Take 1 μL sample aliquots of this reaction mixture at 0, 1, 3, 10, 20, and 30 min and transfer to individual vials. Immediately, mix each sample aliquot with 1 μL matrix solution (5 mg α-CHCA in 700 μL 50% ACN containing 0.1% TFA).
3.5. Mass Analysis (see Note 5)
1. In this study, we used an Applied Biosystems 4700 Proteomics Analyzer with TOF/TOF optics (Applied Biosystems). The instrument uses a 200-Hz frequency-tripled Nd:YAG laser operating at a wavelength of 355 nm. For MS/MS, ions generated by the MALDI process were accelerated at 8 kV through a grid at 6.7 kV into a short, linear, field-free drift region. In this region, the ions passed through a timedion-selector device that is able to select one peptide, from a mixture of peptides, for subsequent fragmentation in the collision cell. After a peptide at a given m/z was selected by the timed-ion-selector it passed through a retarding lens where the ions were decelerated and then passed into the collision cell, which was operated at 7 kV. The collision energy is defined by the potential difference between the source and the collision cell (1 kV). After passing through the cell, the ions (both intact peptide ion and fragments) were accelerated in the second source region at 15 kV, passed through a second, field-free, linear drift region into the reflectron and
480
Samyn, Sergeant, and Beeumen
finally, to the detector. The detector amplifies and converts the signal to electric current, which is observed and manipulated on a PC-based operating system. For high-resolution analysis, the instrument is operated in the reflectron mode. After the MALDI process generates the peptide ions, they are accelerated at 20 kV through a grid at 14 kV into the first, short, linear, field-free drift region. After this point, the rest of the instrument can be treated as a continuation of this region until the ions enter the reflectron and then reach the detector where, as before, the signal at the detector is amplified and converted to an electrical current. 2. One microliter of the sample/matrix mixture is spotted on a stainless steel (192-well) MALDI target plate and allowed to air dry at room temperature. Insert into the mass spectrometer and subject to MALDI–MS analysis. The samples are allowed to air dry at room temperature and are then inserted into the mass spectrometer. Prior to MALDI–MS analysis, the instrument is externally calibrated with a mixture of Angiotensin I, Glu-fibrino-peptide B, ACTH (1-17), and ACTH (18-39). For MS/MS experiments, the instrument is externally calibrated with fragments of Glu-fibrino-peptide.
4. Notes 1. Care must be taken during sample manipulation to avoid oxidation. Artifactual oxidation of methionine to sulfoxide means that CNBr is inhibited from reacting with the sulfur atom of the methionine, which may explain the absence of certain CNBr-fragments. For a short period (days), store gels in MQ water containing 5 mM DTT. For longer storage (weeks–months), keep excised gel spots at −80°C in a closed recipient. 2. CNBr cleaves at the C-terminal side of methionine, converting it into a hsl (Δm = −48 Da), which can undergo hydrolysis to form homoserine (Δm = −29.99 Da). CNBr is considered to be one of the most useful reagents for chemical cleavage of polypeptides because of its high specificity and near-quantitative cleavage yield. Due to the low content of methionines usually found in proteins (1.5%) CNBr cleavage generally produces only a few peptide fragments. In acidic conditions the formation of hsl is favored over that of homoserine, whereas under basic conditions the free acid is formed. Traditionally the cleavage is performed in 70% formic acid, but this often
C-Terminal Sequence Analysis of 2DE-Separated Proteins
481
results in formylation of the amino acids. In our study, the best results were obtained using CNBr in 70% aqueous TFA. 3. The CPY activity will drop significantly after repetitive cycles of freezing and thawing. Control the CPY activity by analyzing a standard peptide with known sequence prior to each experiment. We evaluated the specificity of carboxypeptidase Y (CPY) using several test peptides. According to the known specificity, we observed a slow cleavage of C-terminal Gly. Unexpectedly, the presence of Phe or Thr also slows down or even inhibits ladder generation, although this strongly seems to depend on the peptide sequence. 4. Compared to in-gel trypsin digestion, the CNBr-approach is less sensitive. Nevertheless, sub-microgram detection is adequate for most proteomics applications and is comparable to the detection limit of Coomassie-stained gel spots. Furthermore, it should be noted that most exopeptidases have a Km-value in the range of 5–50 mM, which means that they are operating at 50% maximum velocity when a protein concentration of 5 pmol/μL is used. 5. A positive identification of the C terminus will depend on the length and ionization capacity of the generated CNBr fragments. In our approach, MS analysis was performed using MALDI ionization, which predominantly generates singly charged ions. The use of the α-CHCA matrix produces interference in the low MW mass range and therefore presents a challenge for the analysis of peptides with a MW below 0.7– 1.3 kDa. In our experience, the upper mass limit for the analysis of ladder sequences in RE-MALDI–MS, providing enough resolution and accuracy to identify the amino acid sequence, is restricted to C-terminal fragments of ±5 kDa. However, the use of other MS techniques might allow the identification of larger C-terminal fragments.
Acknowledgments B.S. is a Postdoctoral fellow of the Fund for Scientific ResearchFlanders (F.W.O.-Vlaanderen, Belgium). References 1. Samyn, B., Hardeman, K., Van der Eycken, J., and Van Beeumen, J. Applicability of the alkylation chemistry for chemical C-terminal protein sequence analysis. Anal. Chem. 72, 1389–1399 (2000).
2. Hardeman, K., Samyn, B., Van der Eycken, J., and Van Beeumen, J. An improved chemical approach toward the C-terminal sequence analysis of proteins containing all natural amino acids. Protein Sci. 7, 1593–1602 (1998).
482
Samyn, Sergeant, and Beeumen
3. Sechi, S., and Chait, B.T. A method to define the carboxyl terminal of proteins. Anal. Chem. 72, 3374–3378 (2000). 4. Kosaka, T., Takazawa, T., and Nakamura, T. Identification and C-terminal characterization of proteins from two-dimensional polyacrylamide gels by a combination of isotopic labeling and nano-electrospray Fourier transform ion cyclotron resonance mass spectrometry. Anal. Chem. 72, 1179–1185 (2000). 5. Zhou, X.W., Blackman, M.J., Howell, S.A., and Carruthers, V.B. Proteomic analysis of cleavage events reveals a dynamic two-step mechanism for proteolysis of a key parasite adhesive complex. Mol. Cell. Proteomics 3, 565–576 (2004). 6. Samyn, B., Sergeant, K., Castanheira, P., Faro, C., and Van Beeumen, J. A novel method for C-terminal sequence analysis in the proteomic era. Nat. Methods 2, 193–200 (2005).
7. Samyn, B., Sergeant, K., and Van Beeumen, J. A method for C-terminal sequence analysis in the proteomic era (proteins cleaved with cyanogen bromide). Nat. Protocols 1, 317–322 (2006). 8. Glasauer, S., Langley, S., and Beveridge, T.J. Intracellular iron minerals in a dissimilatory iron-reducing bacterium. Science 265, 117– 119 (2002). 9. Perkins, D.N., Pappin, D.J., Creasy, D.M., and Cottrell, J.S. Probability-based identification by searching sequence databases using mass spectrometry data. Electrophoresis 20, 3551–3567 (1999). 10. Patterson, D.H., Tarr, G.E., Regnier, F.E., and Martin, S.A. C-terminal ladder sequencing via matrix-assisted laser desorption mass spectrometry coupled with carboxypeptidase Y time dependent and concentration-dependent digestions. Anal. Chem. 67, 3971–3978 (1995).
Chapter 32 Shotgun Protein Analysis by Liquid Chromatography-Tandem Mass Spectrometry Kazuishi Kubota, Toshiyuki Kosaka, and Kimihisa Ichikawa Summary Two-dimensional electrophoresis (2DE) is an excellent technology for the analysis of complex protein mixtures, but it has drawbacks, such as hardly detecting very hydrophobic proteins. Shotgun protein analysis is one of the major technologies used to compensate for the weaknesses of 2DE. In this approach, total proteins are digested as a mixture and the digested peptides are separated by one-dimensional or multidimensional chromatography and introduced into a tandem mass spectrometer. Since the shotgun approach is the primary strategy in proteomics besides 2DE, a great number of related methodologies have been developed. In this chapter, we would like to introduce the simplest protocol, in which proteins are digested in solution and the digested peptides are analyzed by one-dimensional reversed-phase liquid chromatography coupled with tandem mass spectrometry (LC–MS/MS), as a starting point for shotgun protein analysis. Key words: Shotgun proteomics, In-solution digest, Tandem mass spectrometry, LC–MS/MS, Database search.
1. Introduction Two-dimensional electrophoresis (2DE) is an excellent technology for the analysis of complex protein mixtures due to its extremely high resolution, which can separate proteins with different posttranslational modifications (PTMs; see Chapters “Detection of 4-Hydroxy-2-Nonenal- and 3-Nitrotyrosine-Modified Proteins Using a Proteomics Approach,” “Proteomic Detection of Oxidized and Reduced Thiol Proteins in Cultured Cells,” “Detection
David Sheehan and Raymond Tyther (eds.), Methods in Molecular Biology, Two-Dimensional Electrophoresis Protocols, vol. 519 © Humana Press, a part of Springer Science + Business Media, LLC 2009 DOI: 10.1007/978-1-59745-281-6_32
483
484
Kubota, Kosaka, and Ichikawa
of Ubiquitination in 2DE, Phosphoproteome Analysis by InGel Isoelectric Focusing and Tandem Mass Spectrometry,” and “Detection of Protein Glutathionylation”). The combination of 2DE and protein identification techniques by mass spectrometry (MS) has been an important strategy for the proteomics of biological materials research. However, 2DE-based protein analysis has several drawbacks. For example, the loading capacity of 2DE gels limits the dynamic range of the detected proteins. Also, it is notoriously difficult to observe hydrophobic proteins by a general protocol (see Chapter “Difficult Proteins”). Finally, small proteins are not shown on a standard 2DE gel (1, 2). One of the most popular alternative methodologies for the analysis of protein mixtures is shotgun protein analysis. In this approach, proteins are not separated by gel electrophoresis, but are digested enzymatically as a mixture in solution, and the digested peptides are introduced to tandem mass spectrometry (MS/MS). This gel electrophoresis separation-free approach is referred to as “shotgun” by analogy to shotgun genome sequencing. The shotgun approach has been made possible owing to the recent development of three key components: sophisticated tandem mass spectrometers, protein identification algorithms from MS/MS data, and the development of protein and DNA sequence databases (3–5). Compared with 2DE-based protein analysis, shotgun protein analysis offers advantages in throughput, sensitivity, dynamic range, unbiased detection of proteins with extremes in pI, molecular weight, and hydrophobicity. However, this methodology is also far from perfect. Because protein digestion takes place first, theoretically, the identified proteins in shotgun analysis lose protein information such as molecular weight or pI, and proteins with different PTMs are treated as a whole. Obviously, 2DE and shotgun analysis have their own strengths and weaknesses, but they are complementary to each other (6). The simplest version of shotgun protein analysis uses onedimensional liquid chromatography (LC) coupled with MS/MS (LC–MS/MS). In most cases, digested peptides are separated by nanoflow reversed-phase LC and introduced to MS/MS online. To analyze simple protein mixtures such as immunoprecipitated proteins, this one-dimensional separation is sufficient. However, when analyzing more complex protein mixtures such as mammalian whole cell lysate, multidimensional separation of the digested peptides is necessary. The representative combination is an online two-dimensional chromatography using strong cation exchange and reversed-phase separations, and is known as multidimensional protein identification technology (MudPIT; 7, 8). Furthermore, to obtain quantitative information in shotgun protein analysis, several strategies have been developed. Most of the strategies may be divided into two categories: label-free methods (9, 10) and stable isotope labeling methods. The labeling methods are
Shotgun Protein Analysis by Liquid Chromatography-Tandem
485
further divided into two groups: chemical labeling methods such as ICAT and iTRAQ (11, 12), and metabolic labeling methods such as SILAC (13, 14). Since shotgun proteomics is the primary approach in proteomics other than 2DE, a great number of methods have been developed and published. In this chapter, we would like to introduce one-dimensional LC–MS/MS analysis of in-solution protein digests as a starting point for shotgun protein analysis.
2. Materials 2.1. In-Solution Digestion
1. Urea solution: Weigh 0.48 g urea in a 1.5-mL Eppendorf tube. Add 620 μL water (see Note 1), 50 μL of 1 M Tris– HCl, pH 8.0, 20 μL of 0.5 M ethylenediamine tetraacetic acid (EDTA), and 0.5 μL of 10% dodecyl-β-maltoside (DM) (see Note 2). This urea solution must be prepared just before use (see Note 3), and the final formulation is 8 M urea, 50 mM Tris–HCl, pH 8.0, 10 mM EDTA, 0.005% DM in water. 2. Dithiothreitol (DTT) solution: 100 mM (15.4 mg/mL) in water. Prepare just before use. 3. Iodoacetamide solution: 200 mM (40.4 mg/mL) in water. Prepare just before use. 4. Urea dilution buffer: 50 mM Tris–HCl, pH 8.0, 0.005% DM in water. Store at 4°C. 5. Trypsin stock solution: Dissolve 20 μg of lyophilized trypsin (Sequencing Grade Modified Trypsin, Promega) in 200 μL of 5 mM acetic acid (final 100 ng/μL trypsin). Store this stock trypsin solution in single use (20 μL) at −20°C and avoid multiple freeze–thaw cycles. Dilute 20 μL stock trypsin solution with 20 μL urea dilution buffer (final 50 ng/μL trypsin) just before use (see Note 4). 6. 10% formic acid: Neat formic acid is diluted with nine volumes of water in a draft chamber. Store at 4°C.
2.2. LC–MS/MS
1. Electrospray ionization (ESI) tip column (see Note 5): A fused silica capillary (150 μm I.D.) is pulled to make an electrospray needle with a P-2000 Laser-based micropipette puller (Sutter Instrument). The processed electrospray needle is packed with a reversed-phase resin (BEH C18, 1.7 μm, Waters) with a Helium pressure cell (Mass Evolution) at 1,000 psi until it is approximately 5 cm. The packed ESI tip column is washed well with 0.05% formic acid/water.
486
Kubota, Kosaka, and Ichikawa
2. Trap column (see Note 6): A fused silica capillary (150 μm I.D.) is pulled and packed with a reversed-phase resin (Inertsil ODS-3, 3 μm, GL Science) as described earlier. Cut a needle and the other end from the packed capillary to become 2.5 cm length, and set it up in a capillary sample trap column assembly (Upchurch Scientific). 3. Nanoflow liquid chromatography system: DiNa (KYA Tech). 4. Solvent A (0.05% formic acid/water): Add 500 μL of highgrade formic acid (i.e., for LC–MS or for amino acid sequence analysis) to 1 L of water (see Note 7). Store at room temperature, and degas it well by sonication in a vacuum just prior to use. 5. Solvent B (0.05% formic acid/acetonitrile): Add 500 μL of high-grade formic acid to 1 L of HPLC-grade acetonitrile (see Note 7). Store at room temperature, and degas it well by sonication in a vacuum just prior to use. 6. Tandem mass spectrometer: a hybrid quadrupole time-offlight tandem mass spectrometer (Q-tof2, Waters). MS/MS spectra are obtained in data-dependent mode. 2.3. Data Analysis
1. Peak list generation software: MassLynx (Waters). 2. Database search software: Mascot (Matrix Science).
3. Methods 3.1. In-Solution Digestion
Sample preparation must be done in the cleanest possible place; otherwise, the majority of proteins identified will be keratin and related proteins. In many laboratories, sample preparation is performed on a clean bench in an isolated room used only for MS sample preparation, where a person must change shoes and wear new gloves and clothes. It is essential to remember that human beings are the main source of contaminants. In the following protocol, the methanol/chloroform precipitation is used to clean up the sample. Although this method was reported about 20 years ago (15) and appears old fashioned, it is very effective in removing many additives, including detergents, with good recovery for small amounts of proteins. If the sample does not have any additives disturbing the LC–MS/MS analysis, this part of the protocol (1–8) can be omitted. In such a case, the sample could be dried using a centrifuge evaporator and be dissolved with the urea solution.
Shotgun Protein Analysis by Liquid Chromatography-Tandem
487
1. Prepare 50 μL protein solution in a 500-μL protein lowbinding tube (Protein LoBind, Eppendorf) (see Note 8). 2. Add 200 μL methanol to the solution (total 250 μL), and vortex well for 5 s. 3. Add 50 μL chloroform to the solution (total 300 μL), and vortex well for 5 s. 4. Add 150 μL water to the solution (total 450 μL), and vortex well for 20 s. 5. Centrifuge the tube at more than 10,000 × g for 2 min at room temperature. A phase separation will be achieved. The proteins are precipitated at the chloroform (lower phase)methanol/water (upper phase) interphase. Carefully remove the upper phase (around 400 μL), so as not to touch the interphase, and discard it. 6. Add 200 μL methanol to the lower chloroform phase (total around 300 μL), and vortex well for 10 s. The phase separation will disappear. 7. Centrifuge the tube at more than 10,000 × g for 2 min at room temperature. The proteins are precipitated at the outer bottom of the tube. In the case of a small amount of proteins, you may not see a pellet, but there will be one. Carefully remove almost all the liquid from the tube, so as not to touch the protein pellet. 8. The protein pellet is dried completely in a centrifuge evaporator. Five to ten min should be sufficient. The dried pellet can be stored at −20°C until use. Prepare urea solution during the drying if you plan to digest the protein continuously. 9. Add 6.25 μL urea solution to the protein pellet immediately, and vortex the tube for more than 10 s to dissolve the pellet. Centrifuge the tube briefly at more than 2,000 × g. 10. Prepare DTT solution, and add 0.625 μL DTT solution to the protein solution (final 10 mM DTT). Mix well, and centrifuge briefly. The proteins are reduced at 37°C for 20 min. 11. Prepare the iodoacetamide solution while cooling the protein solution, and add 0.625 μL iodoacetamide solution to the protein solution (final 20 mM iodoacetamide). Mix well, and centrifuge briefly. The proteins are alkylated at room temperature for 20 min in the dark. 12. Add 42.5 μL urea dilution buffer to the protein solution (total 50 μL) to reduce urea concentration (see Note 4). Vortex well for 5 s and centrifuge briefly at more than 2,000 × g.
488
Kubota, Kosaka, and Ichikawa
13. Prepare trypsin solution, and add 5 μL trypsin solution to the protein solution (see Note 9). Mix well, and centrifuge briefly at more than 2,000 × g. Confirm whether the pH is around 8 using pH indication paper by sampling 0.5 μL. The proteins are digested at 37°C for at least 7 h to overnight. 14. Add 5 μL of 10% formic acid to the protein digest to stop the enzymatic reaction. Mix well, and centrifuge briefly. Take out 0.5 μL of the sample and check that the pH is less than five using pH indication paper. The protein digest can be stored at 4°C for a short term (within 1 week) and at −20°C for a long term until use. 3.2. LC–MS/MS
The LC–MS/MS conditions for shotgun protein analysis depend on the LC–MS/MS system. As an example, in this chapter we describe the LC–MS/MS conditions we use in our laboratory. 1. Set up the liquid chromatography system as shown in Fig. 1. Prepare fresh solvents, and flush the pumps well with them.
Fig. 1. The nanoliquid chromatography system used in our laboratory.
Shotgun Protein Analysis by Liquid Chromatography-Tandem
489
Connect the trap column and the ESI tip column, and flush these columns with 50% solvent B for 5 column volumes. 2. Flow 1% solvent B at 150 nL/min, and adjust the needle position of the ESI tip column and electrospray voltage for a proper spray. 3. Check the LC–MS/MS system sensitivity by a running standard peptide mixture, for example 100 fmol of apomyoglobin tryptic digest. A typical gradient program is shown in Fig. 2. 4. Run the sample digested as described in Subheading 3.1 with a proper LC gradient program (see Note 10) and appropriate MS setup (see Note 11). To improve the reliability of the results, more than one experiment is desirable for protein identification by shotgun protein analysis. 3.3. Data Analysis
1. Make a peak list of tandem mass spectra by MassLynx (see Note 12). A typical condition is shown in Fig. 3. 2. Search the peak data against an appropriate sequence database (see Note 13) using Mascot (see Note 14; see also Chapter “Database Interrogation Algorithms for Identification of Proteins in Proteomic Separations”). Typical parameters for the identification of human proteins are shown in Fig. 4 (see Note 15). 3. Interpret the Mascot results and identify the proteins in the sample. Typical criteria for protein identification in shotgun analysis are described as follows (see Note 16): Step I: We call the protein “observed” when all three prerequisites below are fulfilled. (a) There are at least two top-hit peptides (shown in bold red). (b) The sum of the Mascot score of these top-hit peptides exceeds the protein significant score.
Fig. 2. Example of a gradient program for nano-LC–MS/MS analysis.
490
Kubota, Kosaka, and Ichikawa
Fig. 3. Example of parameters for generating peak list from MS/MS spectra obtained with Qtof2.
Fig. 4. Example of search parameters for Mascot MS/MS ion search.
Shotgun Protein Analysis by Liquid Chromatography-Tandem
491
(c) At least one peptide has a top-hit score which is more than twofold of the second-hit score. Step II: We regard the observed protein as “identified” after observing it at least twice in independent LC–MS/MS runs.
4. Notes 1. The purity of the water is a very important issue. We use water that has more than 18 MΩ cm resistivity and less than 5 ppb total organic content, which is produced by a MilliQ system (Millipore). Some researchers believe that commercial bottled water should be used for MS analysis instead of water processed by a purification system. However, we use so-called Milli-Q water because we haven’t experienced any severe problems. However, this situation may depend on the quality of the source water, instrument maintenance, or other reasons. 2. A very small amount of DM (a kind of alkyl glycoside detergent) is added to the sample to prevent protein/peptide adsorption to the sample tube. This detergent is eluted after most of the peptides in our LC condition, and thus it doesn’t affect the LC–MS/MS analysis. 3. Urea in the solution accumulates isocyanic acid which leads to the carbamylation of proteins. Thus, the urea solution cannot be stored and must be prepared just before use. 4. Modified trypsin is a trypsin modified by reductive alkylation to suppress trypsin autolysis. It is maximally active in the range of pH 7–9 and is reversibly inactivated under pH 4–5. Thus, it should be stored in an acidic solution and used in the active pH range. Modified trypsin is active at less than 1 M urea; therefore, the urea concentration in the protein solution must be diluted before adding trypsin solution. The trypsin concentration in this protocol seems to be high because trypsin autolysis peptides hardly interfere in our LC–MS/MS system. The trypsin concentration should be optimized for each LC–MS/MS system. 5. Some companies, e.g., New Objective, sell equivalent electrospray needles and prepacked ESI tip columns. 6. Several companies, e.g., Upchurch Scientific, LC Packings, sell equivalent trap columns. 7. Detergents and polymers are typical and bothersome contaminants in LC–MS. To prevent such contamination, you
492
Kubota, Kosaka, and Ichikawa
should always use the same glass bottle for each solvent without washing these bottles with detergents. When you first use a bottle, you should wash it with several cycles of solvent A and solvent B, followed by rinsing well with water. 8. Several manufacturers make special sample tubes for protein low binding including a siliconized tube. Regrettably, some brands are not compatible with MS analysis and show many contaminant peaks. Thus, when you first use a new brand of tube, rinse it with a small amount of LC solvent and check it by analyzing the solvent by LC–MS/MS. 9. You can change to proteases other than trypsin, such as lysyl endopeptidase (Lys-C), chymotrypsin, and protease V8 (Glu-C). Each enzyme has a different cleavage specificity, optimum pH range, favorable reaction temperature, and different resistance against urea concentration. For example, Achromobacter Lys-C (Wako) cleaves at the carboxylic side of lysine. It shows little autolysis and is active in 4 M urea. Thus, this enzyme can be stored in water and can be used at a higher urea concentration for digestion than trypsin. 10. Depending on the sample complexity, you should choose a suitable LC gradient program. We often use a shallower gradient program (0.5% per min or 0.25% per min) when analyzing a complex protein mixture such as cell lysate or immunoprecipitated proteins. Please be careful that the excess shallow gradient will broaden the peptide elution peak, which will deteriorate the sensitivity. 11. You should optimize many parameters to obtain the best results, i.e., MS scan time, number of acquired MS/MS spectra per MS spectra, MS/MS scan time, collision energy, exclusion mass lists, data-dependent acquisition condition, calibration, etc. Since LC performance influences these parameters, you should optimize your parameters by analyzing a standard peptide mixture similar to the objective sample under the same LC conditions. 12. The software for generating a peak list from tandem mass spectra is usually attached to a mass spectrometer. For example, MassLynx is supplied by Waters, and Bioworks is by Thermo Scientific. In addition, software such as Mascot distiller can generate a peak list by an original algorithm. 13. You should choose a proper database against which you search because each database has its own strengths and weakness (16). For example, the National Center for Biotechnology Information (NCBI) nonredundant (nr) database has almost all public amino acid sequences from unlimited species (see Chapter “Database Interrogation Algorithms for Identification of Proteins in Proteomic Separations”). Thus,
Shotgun Protein Analysis by Liquid Chromatography-Tandem
493
this database gives one of the highest possibilities that you can identify the protein. Furthermore, if the protein you seek has not been submitted to the database, you may identify a homolog of the protein from a related species (17). However, it will take a long time to perform the search, the identified protein list will be full of redundant proteins, and the sensitivity for protein identification will drop. On the other hand, the IPI database is a nonredundant database from model species (18). Thus, if the IPI database includes the species of protein that you are seeking, you will get the right answer in a simple protein list with a much shorter search time and higher sensitivity. 14. Although we mainly use Mascot search engine for protein identification, there are many similar search engines, such as SEQUEST (Thermo Scientific), X! Tandem (The Global Proteome Machine Organization), and OMMSA (NCBI). 15. You should optimize many parameters depending on the mass spectrometer and experimental conditions (16), i.e, fixed modifications, variable modifications, enzyme specificity, max number of missed cleavages, peptide tolerance in MS spectra, peptide tolerance in MS/MS spectra, etc. 16. You should define your own criteria for protein identification using your own data set. In general, the criteria for shotgun analysis are stricter than those for gel-based analysis. To minimize false-positive identifications, various approaches have been used (16, 19, 20). Our approach, multiple observation and multiple peptides, appears very simple, but it works well in our laboratory. Whatever approach you take, you should validate it by your own data set. For example, if you digest a known protein mixture such as Universal Proteomics Standard Set (Sigma-Aldrich), you would analyze its digestion by LC–MS/MS, and then search for the MS/MS data against a database. Since you know the right answer, you can easily determine both the false-positive and false-negative rate and compare using a variety of criteria.
Acknowledgments We would especially like to thank T. Yoneyama-Takazawa, R. Nakano, and Y. Fukushima for helping with the method optimization. We also acknowledge M. Kubota for her encouragement and helpful discussions.
494
Kubota, Kosaka, and Ichikawa
References 1. Rabilloud, T. (2002) Two-dimensional gel electrophoresis in proteomics: old, old fashioned, but it still climbs up the mountains. Proteomics. 2, 3–10 2. Ong, S. E., and Pandey, A. (2001) An evaluation of the use of two-dimensional gel electrophoresis in proteomics. Biomol. Eng. 18, 195–205 3. Hunt, D. F., Henderson, R. A., Shabanowitz, J., Sakaguchi, K., Michel, H., Sevilir, N., et al (1992) Characterization of peptides bound to the class I MHC molecule HLA-A2.1 by mass spectrometry. Science. 255, 1261–1263 4. McCormack, A. L., Schieltz, D. M., Goode, B., Yang, S., Barnes, G., Drubin, D., and Yates, J.R. 3rd (1997) Direct analysis and identification of proteins in mixtures by LC/ MS/MS and database searching at the lowfemtomole level. Anal. Chem. 69, 767–776 5. Dongre, A. R., Eng, J. K., and Yates, J.R. 3rd (1997) Emerging tandem-mass-spectrometry techniques for the rapid identification of proteins. Trends. Biotechnol. 15, 418–425 6. Kubota, K., Kosaka, T., and Ichikawa, K. (2005) Combination of two-dimensional electrophoresis and shotgun peptide sequencing in comparative proteomics. J. Chromatogr. B. 815, 3–9 7. Link, A. J., Eng, J., Schieltz, D. M., Carmack, E., Mize, G. J., Morris, D. R., et al. (1999) Direct analysis of protein complexes using mass spectrometry. Nat. Biotechnol. 17, 676–682 8. Washburn, M. P., Wolters, D., and Yates, J.R. 3rd (2001) Large-scale analysis of the yeast proteome by multidimensional protein identification technology. Nat. Biotechnol. 19, 242–247 9. Liu, H., Sadygov, R. G., and Yates, J.R. 3rd (2004) A model for random sampling and estimation of relative protein abundance in shotgun proteomics. Anal. Chem. 76, 4193–4201 10. Ishihama, Y., Oda, Y., Tabata, T., Sato, T., Nagasu, T., Rappsilber, J., and Mann, M. (2005) Exponentially modified protein abundance index (emPAI) for estimation of absolute protein amount in proteomics by the
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
number of sequenced peptides per protein. Mol. Cell. Proteomics. 4, 1265–1272 Gygi, S. P., Rist, B., Gerber, S. A., Turecek, F., Gelb, M. H., and Aebersold, R. (1999) Quantitative analysis of complex protein mixtures using isotope-coded affinity tags. Nat. Biotechnol. 17, 994–999 Aggarwal, K., Choe, L. H., and Lee, K. H. (2006) Shotgun proteomics using the iTRAQ isobaric tags. Brief. Funct. Genomic. Proteomic. 5, 112–120 Oda, Y., Huang, K., Cross, F. R., Cowburn, D., and Chait, B. T. (1999) Accurate quantitation of protein expression and site-specific phosphorylation. Proc. Natl. Acad. Sci. U.S.A. 96, 6591–6596 Ong, S. E., Blagoev, B., Kratchmarova, I., Kristensen, D. B., Steen, H., Pandey, A., and Mann, M. (2002) Stable isotope labeling by amino acids in cell culture, SILAC, as a simple and accurate approach to expression proteomics. Mol. Cell. Proteomics. 1, 376–386 Wessel, D., and Flugge, U. I. (1984) A method for the quantitative recovery of protein in dilute solution in the presence of detergents and lipids. Anal. Biochem. 138, 141–143 Nesvizhskii, A. I. (2006) Protein identification by tandem mass spectrometry and sequence database searching. Methods Mol. Biol. 367, 87–120 Liska, A. J., and Shevchenko, A. (2003) Expanding the organismal scope of proteomics: cross-species protein identification by mass spectrometry and its implications. Proteomics. 3, 19–28 Kersey, P. J., Duarte, J., Williams, A., Karavidopoulou, Y., Birney, E., and Apweiler, R. (2004) The international protein index: an integrated database for proteomics experiments. Proteomics. 4, 1985–1988 Nesvizhskii, A. I., and Aebersold, R. (2005) Interpretation of shotgun proteomic data: the protein inference problem. Mol. Cell. Proteomics. 4, 1419–1440 MacCoss, M. J. (2005) Computational analysis of shotgun proteomics data. Curr. Opin. Chem. Biol. 9, 88–94
Chapter 33 De Novo Sequence Analysis of N-Terminal Sulfonated Peptides After in-Gel Guanidination Kjell Sergeant, Jozef Van Beeumen, and Bart Samyn Summary In this protocol, we describe an approach in which two-dimensional electrophoresis (2DE)-separated proteins are guanidinated in-gel prior to enzymatic cleavage. In contrast to previously described techniques, this procedure allows the extracted tryptic peptides to be N-terminally sulfonated without any further sample purification. The protocol was applied on a proteomic study of 2DE-separated proteins from Halorhodospira halophila, an extremophilic eubacterium with an unsequenced genome at the moment of analysis. Key words: Chemically assisted fragmentation, Guanidination, Sulfonation, De Novo sequence analysis.
1. Introduction The rapid and accurate identification of proteins is the primary goal of modern proteomics (see Chapters “Two-Dimensional Electrophoresis: An Overview,” “Solubilization of Proteins in 2DE: An Outline,” “Difficult Proteins,” and “Organelle Proteomics”). The speed of identification we know today has only been attained after the development of sophisticated algorithms (e.g., SEQUEST, Mascot) for the identification of proteins based on MS and MS/MS spectra (see Chapter “Database Interrogation Algorithms for Identification of Proteins in Proteomic Separations”). Despite their success, their utility depends largely on the quality of the generated data and, more importantly, on the availability of database information of the protein under investigation. When the sequence of a protein is not known,
David Sheehan and Raymond Tyther (eds.), Methods in Molecular Biology, Two-Dimensional Electrophoresis Protocols, vol. 519 © Humana Press, a part of Springer Science + Business Media, LLC 2009 DOI: 10.1007/978-1-59745-281-6_33
495
146080_Sheehan_33.indd 495
3/12/2009 4:18:58 PM
496
Sergeant, Beeumen, and Samyn
it is necessary to determine its partial or complete amino acid sequence using either manual or automated methods for de novo sequence analysis. Due to the difficulties in de novo sequence interpretation of mass spectra, either manually or by using algorithms, high-throughput proteomic approaches are limited to those species for which a high amount of sequence information is available (1). Although de novo sequencing of underivatized peptides using matrix-assisted laser desorption ionization –time of flight-time of flight (MALDI–TOF/TOF) has recently been demonstrated, the interpretation of fragment spectra from peptides originating from unknown proteins strongly depends on the use of automated search routines or on manual interpretation. TOF/TOF fragmentation analysis of underivatized peptides yields multiple, incomplete fragment ion series, which are often difficult to interpret. Attempts to improve de novo interpretation have focused on the simplification of the fragmentation pattern by derivatizing the N- or C-terminus of peptides with charged groups (2–4).The introduction of a sulfo group, as developed by Keough et al., facilitates MS/MS fragmentation of singly charged peptide ions by providing a second, “mobile” proton, which lowers amide bond strength and allows more facile unimolecular decay (4). This approach, reviewed in 2003, was optimized for use in combination with both matrix-assisted laser desorption ionization mass spectrometry (MALDI–MS) and electrospray ionization– mass spectrometry (ESI–MS) analysis and, more recently, the use of other reagents has been reported (5). The sulfonation reaction however results in modification of both the N-termini and the ε-amino groups of lysine-containing peptides. It has been demonstrated that introduction of sulfonic acid groups to tryptic peptides is possible only at the N-terminus following guanidination of lysine ε-amines (6). Guanidination, a derivatization wherein lysines are converted into homoarginines (+42 Da), can selectively and quantitatively be performed with O-methylisourea at high pH and does not affect the peptide amino terminus or other side chains (6, 7). In most laboratories, the required desalting step is commonly performed by using reversed-phase microextraction columns (8, 9). However, this step is difficult to automate and loss of analytes can therefore not be avoided (10). We compared the tandem mass spectrometric behavior of guanidinated and sulfonated peptides with that of their native analogues using a MALDI–TOF/TOF instrument (11). More recently, we reported a novel approach in which gel-separated proteins are guanidinated in-gel prior to enzymatic cleavage. In contrast to previously described techniques, this procedure allows extracted tryptic peptides to be N-terminal sulfonated without any further sample purification (see Fig. 1) (12). The derivatized peptides were subsequently fragmented using a MALDI–TOF/
146080_Sheehan_33.indd 496
3/12/2009 4:18:58 PM
De Novo Sequence Analysis of N-Terminal Sulfonated Peptides
497
Fig. 1. Schematic representation of the workflow for N-terminal sulfonation with (right) and without (left) in-gel guanidination. The selected spots are excised from the stained gel. After in-gel guanidination the excess of reagents is removed during destaining of the gel plugs. In the in-gel approach samples are now further processed as in standard trypsin digestion protocols and can be analyzed without further cleanup. When guanidination is performed on peptides after tryptic digestion of the proteins, one more cleanup step is required prior to sulfonation and mass spectrometric analysis. hR homoarginine; SO3 sulfo-acid containing derivatization.
TOF instrument. The approach facilitates the de novo sequence analysis and allows us to obtain longer stretches of amino acid sequence information. Finally, we demonstrated that the information obtained can be used to identify proteins using sequence similarity search algorithms (13, 14). In one study, we applied our improved MS identification approach to identify a number of two-dimensional electrophoresis (2DE)-separated proteins from Halorhodospira halophila (14). (Partial) sequences of tryptic peptides were submitted to homology search for identification of the corresponding protein (see Fig. 2). The total protein extract (250 μg) from Halorhodospira halophila grown anaerobically under green/blue light was separated by 2DE. The proteins were guanidinated in-gel and desalted/destained in one single step as described herein. Subsequently, guanidinated proteins were enzymatically cleaved with trypsin and, after extraction, the peptides were sulfonated. The peptide mass fingerprint (PMF) was analyzed both in the positive and the negative ion mode (see Fig. 2a, b) as previous
146080_Sheehan_33.indd 497
3/12/2009 4:18:58 PM
498
Sergeant, Beeumen, and Samyn
Fig. 2. (a) MALDI MS reflectron spectrum (positive ion mode) of the translation elongation factor Ts (spot 3) tryptic peptides after performing the in-gel procedure; (b) MALDI MS reflectron spectrum (negative ion mode) of the same sample; (c) MS/MS spectrum (positive ion mode) of the precursor at m/z 1869.8, the loss of the sulfonation label is indicated (−184 Da). All labeled ions are y-ions. de novo-derived sequence information is indicated in the one-letter code.
146080_Sheehan_33.indd 498
3/12/2009 4:18:59 PM
De Novo Sequence Analysis of N-Terminal Sulfonated Peptides
499
experiments indicated that sulfonic acid-derivatized peptides have poorer positive-ion sensitivity than the corresponding native peptides (4). The introduction of a negative charge usually leads to a decrease or even loss in signal intensity in the positive mode (11). By way of example, the protein in spot three was identified as translation elongation factor Ts (gi:121998247). Upon sulfonation, five peptides were subjected to MALDI MS/MS analysis. In all fragmentation spectra we observed a complete y-ion series that could easily be interpreted, facilitating de novo sequencing. Figure 2c shows the fragmentation spectrum of the precursor selected at m/z 1869.8. The de novo-derived sequence information was combined in one search query and analyzed using three homology search algorithms (14).
2. Materials 2.1. 2DE
1. Reswelling buffer: 6 M urea, 4% CHAPS, trace amount of bromophenol blue; this can be prepared and stored in single use aliquots at −20°C. 2. DTT (Fluka): Add DTT just before use. 3. Acrylamide/bisacrylamide solution (National diagnostics): store in a dark place. 4. Separation buffer: 1.5 M Tris–HCl, pH 8.8, 4% SDS: store at room temperature. 5. Ammonium persulfate (APS) solution (Amersham): Prepare 10% APS solution in water: prepare just prior to use. 6. N,N,N,N ′-tetramethyl-ethylenediamine (TEMED): (Fluka): Because the quality of TEMED decreases, it should only be used for a limited period and the stock should be renewed regularly: store at room temperature. 7. Water-saturated butanol: Mix equal volumes of water (MQ) and isobutanol; shake vigorously and allow the separation in two phases; use the top phase: store at room temperature. 8. Equilibration solution: 6 M urea/30% glycerol/2% SDS/50 mM Tris–HCl; Weigh the compounds just prior to use and dissolve in a prepared solution of 50 mM Tris–HCl/30% glycerol that can be stored at room temperature. 9. Agarose (Amersham): Dissolve 400 mg agarose in 100 mL MQ and a trace amount of bromophenol blue: Store this solution after preparation at room temperature and melt prior to use.
146080_Sheehan_33.indd 499
3/12/2009 4:19:01 PM
500
Sergeant, Beeumen, and Samyn
2.2. In-Gel Guanidination and Destaining
1. O-methylisourea hemisulfate (Across): Store at 4°C. Prepare solutions fresh prior to use. 2. 7N Ammonium hydroxide: Dilute a 30% NH4OH solution (Merck) in an equal volume of MQ: Store and manipulate concentrated NH4OH under a fume hood and avoid direct inhalation of irritating and toxic fumes.
2.3. Tryptic Digestion
1. Trypsin stock solution (Promega): the content of 1 vial (20 μg) is dissolved in 200-μL reconstitution buffer as indicated by the manufacturer. Always store trypsin in solution at −20°C and thaw on ice. 2. N,N -diisopropylethylamine (DIEA)(Applied Biosystems): is a volatile, highly corrosive base; manipulate under a fume hood, store at 4°C.
2.4. Sulfonation and Spotting of Samples on MALDI Plates
1. 2-Sulfobenzoic acid cyclic anhydride (Fluka): Store at 4°C and make sure that the vial is properly sealed (it will react when moist). 2. Tetrahydrofuran (THF): Toxic and irritating solvent; store at 4°C and manipulate under a fume hood.
3. Methods The method, as described herein, begins with a protein extract that is separated using 2DE. It can also be applied to identify proteins separated by SDS–PAGE (see Note 1). We describe the use of 2-sulfobenzoic acid cyclic anhydride as reagent for sulfonation. However, the protocol can also be performed using other reagents for N-terminal sulfonation, including the water-compatible 4-sulfophenyl isothiocyanate (SPITC) (15) and CAF reagent (16). 3.1. 2DE
1. These instructions assume performance of passive sample loading during the reswelling of 18-cm-long IEF-gel strips. They can be easily adapted to cup loading of samples using the different formats that have been developed for single-dimensional electrophoresis and 2DE (see Note 1). 2. Dissolve 3 mg DTT in 1 mL reswelling buffer. 3. Add the reswelling buffer, which contains DTT, to 300 μg of protein extract to obtain a volume of 337.5 μL and add 12.5 μL 3–10 ampholyte stock solution (Bio-Rad). 4. Apply the whole mixture to 18-cm-long immobilized pH gradient (IPG) strips and perform a rehydration procedure for 6–8 h at room temperature. Perform isoelectrofocusing (IEF)
146080_Sheehan_33.indd 500
3/12/2009 4:19:01 PM
De Novo Sequence Analysis of N-Terminal Sulfonated Peptides
501
using the Multiphor II (Amersham Biosciences) at 18°C by using a standard program provided by the manufacturer (see Note 2). Ensure that the strips are completely covered with mineral oil during IEF to avoid drying. 5. Rinse the glass plates and assemble the gels in a gel caster (Bio-Rad) using 1-mm thick spacers. Check the assembly by filling it with MQ water (must be carefully removed afterward). Pour an acrylamide solution (for two 18 × 18 cm2 gels: 3.3 mL 30% acrylamide/Bis Solution/4.1 mL MQ/2.5 mL separation buffer/50 μL APS solution/10 μL TEMED) between the glass plates leaving 0.5 cm at the top. Overlay the gel solution with water-saturated isobutanol and allow the gel to polymerize for several hours at room temperature (see Note 3). 6. Take the strips out of the IEF apparatus and gently rinse them with MQ to remove residual mineral oil. Equilibrate the strips by gently shaking them for 10 min at room temperature while they are completely covered with 5 mL of the equilibration solution containing 1% DTT. Remove the solution containing DTT, replace it with equilibration solution containing 2.5% iodoacetamide, and leave at room temperature for another 10 min. 7. Gently rinse the top of the gel with water (MQ) to remove residual isobutanol and apply the equilibrated IEF gel strip on top of the gel. Fill the remaining space between the plates with the 0.4% agarose solution ensuring that air bubbles between the gel strip and the SDS–PAGE gel are removed. Mount the gels in a gel electrophoresis apparatus and fill the upper and lower buffer chambers with electrophoresis buffer. 8. Connect the gel unit (Protean II Bio-Rad) to a power supply and apply current. Run the gels at 8°C and approximately 100 mAh/gel until the bromophenol blue front reaches the bottom of the gel. 9. Prepare staining solution by dissolving 2 g colloidal Coomassie brilliant blue G-250 in 1 L 34% methanol/17% ammonium sulfate/3% phosphoric acid, and mix vigorously. 10. Remove the gel from the cassette and fix the proteins by incubating the gel for 30 min in 50% ethanol/MQ/2% phosphoric acid. Discard the fixation solution and pour the solution of Coomassie blue over the gel, completely covering it. The gel is stained after 4 h incubation in Coomassie blue solution. 11. Destain the background with 30% methanol overnight (see Note 4).
146080_Sheehan_33.indd 501
3/12/2009 4:19:01 PM
502
Sergeant, Beeumen, and Samyn
3.2. In-Gel Guanidination and Destaining
1. Excise the protein spots of interest and transfer them to a clean Eppendorf vial (see Note 5). 2. Prepare a 7.5 M O-methylisourea hemisulfate solution by dissolving 50 mg of the compound in 51 μL water (MQ). Vortex the solution vigorously. 3. Add 5 μL MQ, 11 μL 7N ammonium hydroxide, and 3 μL of the O-methylisourea solution to the gel spots. Incubate this mixture for 2 h at 65°C. 4. Discard the remainder of the O-methylisourea solution and add 150 μL 200 mM ammonium bicarbonate in 50% ACN/ MQ. Wash the gel plugs for 30 min at 30°C, discard the solution, and repeat the wash step (see Notes 6 and 7). 5. Discard the solvents and completely dry the gel plug.
3.3. Tryptic Digestion
1. Prepare a trypsin solution by diluting 1 μL of a 0.1 μg/μL stock solution with 49 μL 50 mM ammonium bicarbonate (pH 7.8). Keep the solution on ice. 2. Add 8 μL of this trypsin solution to the dried gel spots and keep the tubes on ice for 45 min. 3. Add 10 μL 50 mM ammonium bicarbonate (pH 7.8) to cover the reswollen gel plugs with fluid and incubate in a warm water bath at 37°C for 16 h. 4. Spin the tubes down for a brief moment to recover droplets on the tube walls and transfer the liquid to a clean tube. 5. Add 35 μL 60% ACN/MQ/0.1% DIEA and keep it in a warm water bath at 30°C for 30 min. 6. Spin down briefly and pool the supernatant with the liquid from step 4. 7. Repeat steps 5 and 6 and dry the pooled extraction solvents (see Note 8).
3.4. Sulfonation and Spotting of Samples on MALDI Plates
1. Add 4 μL 12.5 mM ammonium bicarbonate in 50% ACN/ MQ to the dried extracts and vortex briefly to improve solubilization of the peptides. 2. Transfer 0.7 μL to a clean Eppendorf tube and mix with 0.7 μL of the matrix solution that has been prepared by dissolving 7 mg α-cyano-4-hydroxycinnamic acid (Sigma-Aldrich) in 1 mL 50% ACN/MQ containing 0.1% TFA. Spot 0.7 μL of the sample-matrix mixture on a stainless steel target plate and allow to air dry at room temperature. 3. Prepare the sulfonation reagent by dissolving 2 mg of 2-sulfobenzoic acid cyclic anhydride in 1 mL of dry THF. Vortex until the reagent is completely dissolved.
146080_Sheehan_33.indd 502
3/12/2009 4:19:01 PM
De Novo Sequence Analysis of N-Terminal Sulfonated Peptides
503
4. Add 2 μL of the prepared sulfonation reagent to 2 μL of the resolubilized peptides, vortex briefly, and allow to react at room temperature for 15 min. 5. Transfer 0.7 μL to a clean Eppendorf tube and mix with 0.7 μL of the matrix solution, spot 0.7 μL of the sample-matrix mixture on a stainless steel target plate, and allow to air dry at room temperature. 3.5. Mass Spectrometry
We used a 4700 MALDI–TOF/TOF mass analyzer (Applied Biosystems), but mass analysis can be performed by using other mass spectrometers as well documented (6, 16, 17). Although the use of electrospray ionization (ESI) MS for the analysis of sulfonated peptides has been described before (18), it is less commonly used as this generates multiple charged ions. 1. Insert the stainless steel target plate in the mass spectrometer and check the calibration in both the MS and MS/ MS modes thoroughly; adjust the calibration if needed (see Note 9). 2. Analyze the spots of nonsulfonated samples in the positive MS mode. 3. Analyze the spots of sulfonated samples in the positive MS mode. Identify those peptides that are sulfonated by the characteristic mass difference of 184 Da (for peptides sulfonated with 2-sulfobenzoic acid cyclic anhydride) between the analyses in steps 2 and 3. 4. Select the sulfonated peptides and use them as precursor in positive mode MS/MS experiments (see Note 10). 5. Analyze the spots of sulfonated samples in the negative MS mode. Identify those peptides that are sulfonated by the characteristic mass difference of 182 Da (for peptides sulfonated with 2-sulfobenzoic acid cyclic anhydride) between the analyses in steps 2 and 5. 6. Select the sulfonated peptides, add 2 Da, and use this mass as precursor in positive mode MS/MS experiments (see Note 10). 7. Determine the sequence of the fragmented peptides by calculating the mass differences between consecutive peaks in the y-ion series.
3.6. Database Searching/Protein Identification
146080_Sheehan_33.indd 503
Several software tools have been developed that allow the use of peptide sequences for homology searching. In previous studies, we used the following three algorithms: MS Blast (http://dove.embl-heidelberg.de/Blast2/msblast. html),
3/12/2009 4:19:01 PM
504
Sergeant, Beeumen, and Samyn
FASTS
(http://fasta.bioch.virginia.edu/fasta_www/cgi/)
and MS Homology (http://prospector.ucsf.edu/ucsfhtml4.0/ mshomology.htm) The three algorithms ask for the use of a different query format. Therefore, we refer to the respective websites for precise information on the format of the submitted data (see Note 11). 1. Determine the sequences of all fragmentation spectra from a single spot. Mass differences of 113 Da between consecutive y-ions are designated as “I,” differences of 128 Da always as “Q.” Assemble all sequences in a single search string. 2. Submit the search using the three algorithms. 3. Consider a protein identified only when each of the three algorithms results in a significant score, giving the highest probability to proteins with similar functions. 4. It has been demonstrated that indirect evidence can add to the significance of an identification. Therefore, the identifications were further validated by using information such as the cleavage specificity of trypsin, and sequence information resulting from known preferential fragmentation patterns of sulfonated peptides (11), etc. (see Note 12).
4. Notes 1. The protocol described here can also be used if SDS–PAGE is used for sample preparation. However, the possibility to obtain de novo sequence information can be hampered if too many peptides are present in one gel band. One MS/MS spectrum may contain fragments from different peptides which makes interpretation less straightforward. However, SDS–PAGE can be used for sample preparation after prefractionation or for the analysis of smaller subproteomes. Finally, the size of the SDS– PAGE gel plugs can be larger than those excised from 2DE gels. Therefore, the volumes must be adjusted with respect to the protocol described. In our experience, this has no impact on the final result. 2. The program we use is: 150 V (30 ), 150 V (120 ), 300 V (30 ), 300 V (45 ), 3,500 V (90 ), 3,500 V (540 ), 500 V (10 ), and hold at 500 V. For samples containing high concentrations of salts or other contaminating substances it is often required to prolong the first steps in this program. Such prolongation, at low voltages, allows the removal of salts and has no detrimental effects on the gel separation.
146080_Sheehan_33.indd 504
3/12/2009 4:19:01 PM
De Novo Sequence Analysis of N-Terminal Sulfonated Peptides
505
3. Once the gels are poured and polymerized, they can be stored at 4°C for several days. 4. Because the Coomassie stain is not-saturating, the time lapses for staining and destaining of the background can be made longer to the convenience of the experimenter. 5. If the protocol is performed on an SDS–PAGE gel, an additional reduction and alkylation step must be incorporated. Reduction is done by incubating the excised gel spots with 15 μL 10 mM DTT in 7 M GuHCl/0.3 M Tris-HCl (pH 9) at 55°C for 45 min. Subsequent alkylation is performed be adding 5 μL 55 mM iodoacetamide, followed by an additional incubation in the dark (45 min). 6. Destaining of gel plugs can also be performed prior to the guanidination reaction. Therefore, two extra wash steps are inserted after excision of the selected gel spots. 7. More intense colored spots might require the use of additional destaining steps to avoid interference of residual dye during MS analysis. 8. After drying, the pooled extractions can be stored at −20°C for several months. 9. For calibration, we use a mixture of four peptides: Angiotensin I, Glu-fibrino-peptide B, ACTH (1–1), and ACTH (18–39). The same set of peptides can be used to calibrate both in negative- and positive-mode analyses. During calibration, the obtained mass accuracy must be within 25 ppm. For the calibration in the MS/MS mode, the most abundant fragments from Glu-fibrino-peptide B are used. Alternatively, other calibration mixtures can be applied. 10. On the MALDI–TOF/TOF instrument, fragmentations are performed in “CID-off” mode. There is no gas present in the fragmentation cell in this mode and fragmentation is only initiated by energy differences between the source and the CID cell. Because of the lower fragmentation energy in this mode, fragmentation of the peptide bond will be dominant and internal fragment ions are generally not observed (19). 11. We noticed previously that the score obtained can vary using the FASTS algorithm and, even more pronounced, using the MSBlast algorithm. This is according to the order of the peptide sequences in the search query (13). 12. Different proteins comigrating in the same spot can be identified by resubmitting those peptide sequences that were not matched in a first round of database searching. In this way, we were able to identify seven proteins in a single SDS– PAGE band (preliminary results).
146080_Sheehan_33.indd 505
3/12/2009 4:19:01 PM
506
Sergeant, Beeumen, and Samyn
References 1. Liska, A. J., and Shevchenko, A. (2003). Expanding the organismal scope of proteomics: cross-species protein identification by mass spectrometry and its implications. Proteomics 3, 19–28. 2. Burlet, O., Yang, C., and Gaskell, S. J. (1992). Influence of cysteine to cysteic acid oxidation on the collision-activated decomposition of protonated peptides: evidence for intraionic interactions. J Am Soc Mass Spectrom 3, 337–44. 3. Roth, K. D., Huang, Z. H., Sadagopan, N., and Watson, J. T. (1998). Charge derivatization of peptides for analysis by mass spectrometry. Mass Spectrom Rev 17, 255–74. 4. Keough, T., Youngquist, R. S., and Lacey, M. P. (1999). A method for high-sensitivity peptide sequencing using post-source decay matrix-assisted laser desorption ionization mass spectrometry. Proc Natl Acad Sci U S A 96, 7131–6. 5. Keough, T., Youngquist, R. S., and Lacey, M. P. (2003). Sulfonic acid derivatives for peptide sequencing by MALDI MS. Anal Chem 75, 156A–65A. 6. Keough, T., Lacey, M. P., and Youngquist, R. S. (2000). Derivatization procedures to facilitate de novo sequencing of lysine-terminated tryptic peptides using postsource decay matrix-assisted laser desorption/ionization mass spectrometry. Rapid Commun Mass Spectrom 14, 2348–56. 7. Beardsley, R. L., and Reilly, J. P. (2002). Optimization of guanidination procedures for MALDI mass mapping. Anal Chem 74, 1884–90. 8. Cagney, G., and Emili, A. (2002). De novo peptide sequencing and quantitative profiling of complex protein mixtures using masscoded abundance tagging. Nat Biotechnol 20, 163–70. 9. Hellman, U., and Bhikhabhai, R. (2002). Easy amino acid sequencing of sulfonated peptides using post-source decay on a matrix-assisted laser desorption/ionization time-of-flight mass spectrometer equipped with a variable voltage reflector. Rapid Commun Mass Spectrom 16, 1851–9. 10. Beardsley, R. L., Karty, J. A., and Reilly, J. P. (2000). Enhancing the intensities of lysineterminated tryptic peptide ions in matrixassisted laser desorption/ionization mass spectrometry. Rapid Commun Mass Spectrom 14, 2147–53.
146080_Sheehan_33.indd 506
11. Samyn, B., Debyser, G., Sergeant, K., Devreese, B., and Van Beeumen, J. (2004). A case study of de novo sequence analysis of N-sulfonated peptides by MALDI TOF/TOF mass spectrometry. J Am Soc Mass Spectrom 15, 1838–52. 12. Sergeant, K., Samyn, B., Debyser, G., and Van Beeumen, J. (2005). De novo sequence analysis of N-terminal sulfonated peptides after ingel guanidination. Proteomics 5, 2369–80. 13. Samyn, B., Sergeant, K., Carpentier, S., Debyser, G., Panis, B., Swennen, R., et al. (2007). Functional proteome analysis of the banana plant (Musa spp.) using de novo sequence analysis of derivatized peptides. J Proteome Res 6, 70–80. 14. Samyn, B., Sergeant, K., Memmi, S., Debyser, G., Devreese, B., and Van Beeumen, J. (2006). MALDI TOF/TOF de novo sequence analysis of 2D-PAGE separated proteins from Halorhodospira halophila, a bacterium with unsequenced genome. Electrophoresis 27, 2702–11. 15. Gevaert, K., Demol, H., Martens, L., Hoorelbeke, B., Puype, M., Goethals, M., et al. (2001). Protein identification based on matrix assisted laser desorption/ionization-post source decaymass spectrometry. Electrophoresis 22, 1645–51. 16. Flensburg, J., Tangen, A., Prieto, M., Hellman, U., and Wadensten, H. (2005). Chemically-assisted fragmentation combined with multi-dimensional liquid chromatography and matrix-assisted laser desorption/ionization post source decay, matrix-assisted laser desorption/ionization tandem time-of flight or matrix-assisted laser desorption/ionization tandem mass spectrometry for improved sequencing of tryptic peptides. Eur J Mass Spectrom 11, 169–79. 17. Keough, T., Lacey, M. P., and Strife, R. J. (2001). Atmospheric pressure matrix-assisted laser desorption/ionization ion trap mass spectrometry of sulfonic acid derivatized tryptic peptides. Rapid Commun Mass Spectrom 15, 2227–39. 18. Bauer, M. D., Sun, Y., Keough, T., and Lacey, M. P. (2000). Sequencing of sulfonic acid derivatized peptides by electrospray mass spectrometry. Rapid Commun Mass Spectrom 14, 924–9. 19. Pashkova, A., Moskovets, E., and Karger, B. L. (2004). Coumarin tags for improved analysis of peptides by MALDI-TOF MS and MS/ MS. 1. Enhancement in MALDI MS signal intensities. Anal Chem 76, 4550–7.
3/12/2009 4:19:01 PM
Chapter 34 Tryptic Digestion of In-Gel Proteins for Mass Spectrometry Analysis Mai-Loan Huynh, Pamela Russell, and Bradley Walsh Summary Identification and characterization of proteins are ultimately the goal in proteomic analysis. In order to identify a protein trypsin is commonly used to digest protein into peptides which can be analyzed by matrix-assisted laser desorption ionization mass spectrometry (MALDI-MS) or liquid chromatographytandem mass spectrometry (LC-MS/MS). This chapter describes a tryptic digestion method for digestion of proteins in one-dimensional (1DE) or two-dimensional (2DE) polyacrylamide gels. The method involves cutting target protein bands or spots, removal of protein stain, reduction and alkylation of native protein, digestion and finally extraction of peptides for mass spectrometry analysis. The method is simple and reasonably sensitive that many in-gel proteins that are barely visible with Coomassie blue stain have been successfully identified. Key words: In-gel tryptic digestion, Peptide mass fingerprinting, Protein identification.
1. Introduction An appropriate digestion method is vital for successful identification and characterization of in-gel proteins. Generally to identify a protein an enzyme or a chemical is used to specifically cleave the amino acid sequence into a mixture of various length peptides (1). The exact mass of the peptides can be measured by matrix-assisted laser desorption ionization mass spectrometry (MALDI-MS) or electrospray mass spectrometry (ESI-MS) technique to generate a peptide map which can be matched against the theoretical digest of known protein databases to identify the
David Sheehan and Raymond Tyther (eds.), Methods in Molecular Biology, Two-Dimensional Electrophoresis Protocols, vol. 519 © Humana Press, a part of Springer Science + Business Media, LLC 2009 DOI: 10.1007/978-1-59745-281-6_34
507
508
Huynh, Russell, and Walsh
protein (2,3). Additionally the tryptic peptides can be subjected to tandem mass spectrometry (MS/MS) where each peptide can be further fragmented to produce a ladder of peptides for amino acid sequencing to allow highly confident identification or confirmation of the protein (4). Although there is a range of chemicals and enzymes commercially available trypsin is the most commonly used enzyme for digestion of proteins. Trypsin is a serine protease that specifically cleaves protein at the C-terminal side of arginine or lysine producing tryptic peptides of appropriate lengths that can be analyzed by current mass spectrometry instruments. The location of the basic residue (R or K) at the terminus of the tryptic peptides increases ionization and hence improves signal intensity of the mass spectral peaks (5). The tryptic digestion method described in this chapter was derived and simplified from a number of tryptic digestion methods that have been described in the past twenty years (6,7). Basically the tryptic digestion procedure involves cutting target protein bands or spots removal of protein stainreduction and alkylation of native protein digestion and finally extraction of peptides for MS or MS/MS analysis. All these steps are important and required appropriate handling for maximum peptide recovery.
2. Materials 2.1. Equipment and Consumables
1. A light box for cutting gel spots or bands.
2.2. Reagents
All solutions including stock solution of 100 mM ammonium bicarbonate should be freshly prepared within the day of conducting the experiment to avoid or minimize contaminants.
2.2.1. Tryptic Digestion of In-Gel Proteins
1. Cleaning solution 1. 70% ethanol in ultrapure H2O.
2. C18 ziptip® (Millipore Corporation).
2. Cleaning solution 2. 1% acetic acid in ultrapure H2O. 3. Destaining solution. 50% acetonitrile/50 mM ammonium bicarbonate. Mixing 100% acetonitrile in 100 mM ammonium bicarbonate1:1. 4. Dehydration solution. 100% acetonitrile. 5. Reducing/alkylating solution. 5 mM tributyl phosphine (TBP)10 mM acrylamide in 50 mM ammonium bicarbonate solution. 6. Trypsin buffer. 50 mM ammonium bicarbonate (dilute the 100 mM ammonium bicarbonate in ultrapure H2O). 7. Promega Trypsin. Sequencing-grade modified trypsin (Cat# V5111).
Tryptic Digestion of In-Gel Proteins for Mass Spectrometry Analysis 2.2.2. C18. Ziptip Treatment
509
1. Activation solution. 70% acetonitrile/0.1% trifluoroacetic acid. 2. Equilibration solution. 0.1% trifluoroacetic acid. 3. Washing solution. 0.1% trifluoroacetic acid. 4. Matrix elution solution for MALDI-MS. 8 mg/mL recrystallized α-Cyano-4-hydroxycinnamic acid in 70% acetonitrile/0.1% trifluoroacetic acid. 5. Elution solution for LC-MS/MS. 70% acetonitrile/0.1% formic acid (or appropriate elution solution for the LC-MS/MS method).
3. Method It is very important to avoid all sources of keratin, saliva and polymer contaminants. The presence of the contaminant peaks whether at low or high concentration can cause suppression of the target peptides resulting in complication in identification of the target protein. Always conduct the experiment in a clean area. Ensure samples solutions and the pipette tips are covered when not in use. At present proteins that are invisible with Coomasie Blue stain may not be identified. To increase the probability of identifying weak spots two or three spots of the same protein can be pooled to improve the MS data. In a 1DE gel a dark protein band can be identified with a small piece (1.5-mm cube) but for a weak protein band the whole band is required for identifying the protein. 3.1. Tryptic Digestion for In-Gel Proteins
1. Print out the gel image and number the desired spots or bands for cutting. 2. Decontaminate the light box surface with cleaning solution 1 and wipe off with Kimwipe paper; apply the cleaning solution 2 and wipe off with Kimwipe paper. Repeat the cleaning if necessary. 3. Before cutting add 200 μL of ultrapure H2O into a new 1.5mL Eppendorf tube for each sample. A 96-well multi-titer plate can be used to process a large number of samples. 4. Use a modified 1-mL tip to excise the protein spot or band and transfer the gel piece into the sample container (see Note 1). Pipette up and down to release the spot and also to wash the tip. Use the same tip to excise the rest of the protein spots or bands. 5. Commercial silver stain kits differ slightly. Hence follow the manufacturer’s instruction for destaining procedure (see Note
510
Huynh, Russell, and Walsh
2). For Coomasie blue stains discard the solution in the sample and add 100 μL destaining solution per gel piece. Cover and leave the sample at room temperature for 30 min or at 37°C for 15 min. Discard the washed solution. 6. For heavily stained spots repeat the destaining step once or twice (see Note 3). 7. Add 50 μL of dehydration solution per gel piece. After 5 min the gel should turn opaque; otherwise discard the solutionadd another 50 μL of dehydration solution per gel piece. When the gel piece turns opaque place the uncovered sample in 37°C oven for 15 min to dry completely. For 2DE gels proteins are usually reduced and alkyated prior to electrophoresis; hence ignore steps 8–10 and continue from step 11 (see Note 4). 8. For 1DE gels or native gels add 50 μL reduction-alkylation solution per gel piece. Allow to incubate at room temperature for 60 min (see Note 5). 9. Discard the reduction-alkylation solution (see Note 6) add 100 μL of ultrapure H2O to the gel piece mix briefly then discard the solution. Repeat the washing once. 10. Add 50 μL of dehydration solution per gel piece. After 5 min the gel should turn opaque; otherwise discard the solution and add another 50 μL of dehydration solution per gel piece. When the gel turns opaque discard the solution. Place the uncovered sample at 37°C oven for 15 min to dry completely. 11. Immediately before digestion resuspend the lyophilized Promega trypsin in the trypsin buffer to the final concentration of 5 μg/mL (see Note 7). 12. Add 30 μL of trypsin solution per gel piece. Ensure the gel piece is totally submerged with the trypsin solution; completely cover the sample to prevent evaporation and incubate in an oven at 37°C for 4 h or at 30°C overnight. Discard the unused diluted trypsin. 13. Cool the sample in a freezer for 5 min and then sonicate it in an ultrasonic water bath for 10 min to allow peptides to diffuse out of the gel. For MALDI-MS analysis the peptide solution requires further treatment with C18 ziptip. For LC-MS/MS analysis the peptide solution can be analyzed with or without C18 ziptip treatment (see Note 8). 3.2. C18 Ziptip Treatment
1. Set a micropipette to 10 μL insert a C18 ziptip in the activation solution and pipette up and down three cycles (see Note 9).
Tryptic Digestion of In-Gel Proteins for Mass Spectrometry Analysis
511
2. Pipette 10 μL of equilibration solution and then dispense on a Kimwipe tissue. Repeat the equilibration two times. 3. In the peptide sample pipette up and down ten times to bind peptides. For larger sample volumes ensure the entire sample passes through the ziptip at least once to bind all peptides. 4. Pipette 10 μL of washing solution and then dispense on a Kimwipe tissue. Repeat the washing twice. 5. For MALDI-MS analysis pipette approximately 1.5 μL of the matrix elution solution; on a cleaned MALDI target plate rest the ziptip tip on a well and then pipette up and down ten times to elute peptides. Avoid bubbles. 6. Leave the sample to air dry before inserting into MALDI instrument (Fig. 1). 7. For LC-MS/MS analysis pipette 5–10 μL of the elution solution; in a new vial pipette up and down ten times to elute peptides. The peptide sample can be placed at 37°C for a few min to evaporate acetonitrile before the LC-MS/MS analysis.
Fig. 1. A typical MALDI-MS profile for tryptic digestion of a medium Coomassie brilliant blue stained spot from 2DE.
512
Huynh, Russell, and Walsh
4. Notes 1. Cut a 1-mL tip so that the tip is of about 1.8–2 mm internal diameter. Attach the tip to a pipette press the whole pipette down to excise spot/band and then use the plunger to pick up or release the spot into the container. A scalpel is useful for 1DE bands but a new scalpel can become rusty within minutes and the rust can affect the MS data. A blank gel piece should always be cut and processed in parallel in order to identify trypsin autoproteolytic peaks and the matrix background peaks. 2. Silver is a sensitive stain for visualizing protein but this stain is not recommended as it is incompatible with mass spectrometry analysis resulting in poor data and difficulty in identifying protein. Alternatively Sypro Ruby stain can be used for sensitive spot detection with UV light and then Coomassie brilliant blue stain can be used to restain protein for cutting at normal light. 3. Excessive destaining of the protein gel can cause loss of protein. For weakly stained gel only wash 1–2 times; for heavily stained gel three washes are enough as the stain will not significantly affect the protein identification. 4. In general proteins should be denatured, reduced and alkylated for enzymatic digestion to reach completion (8). Proteins in 2DE gels are usually reduced and alkylated before the seconddimension electrophoresis; hence reduction and alkylation are not necessary. However proteins in 1DE gels are usually reduced but not alkylated; therefore reduction and alkylation are required for successful identification. 5. The combination of reduction and alkylation in single step (9,10) was quite simple and effective. Alternatively 50 μL of 10 mM DDT in 50 mM ammonium bicarbonate solution can be used to reduce protein. However the sample must be heated at 95°C for 20 min or at 60°C for 50 min. After cooling down to room temperature 50 μL of 50 mM acrylamide can be used to alkylate the protein. Although iodoacetamide can be used instead of acrylamide the use of iodoacetamide may produce some protein alkylated with iodoacetamide and some protein alkylated with free acrylamide monomer in the gel. 6. Reducing and alkylating solutions are generally toxic or carcinogenic; therefore they should be discarded in a biohazard waste container. 7. Sequencing-grade trypsin is stable at pH <4; to save trypsin for future uses first resuspend a 20-μg vial with 100 μL of
Tryptic Digestion of In-Gel Proteins for Mass Spectrometry Analysis
513
100 mM acetic acid (resuspension buffer can be stored at −20°C) make aliquots and store at −20°C to avoid multiple freeze thaw cycles. Just before digestion dilute the aliquot to 5 μg/mL with the trypsin buffer. 8. For MALDI-MS analysis concentration and desalting sample with C18 ziptip significantly improve the MS data. For LC-MS/MS analysis depending on the loading volume it may be necessary to concentrate the peptide sample with a C18 ziptip before loading. 9. Avoid introducing air into the ziptip by keeping pressure on the plunger all the way down after dispensing solution only releasing the plunger when aspirating solution again. Air bubbles in the ziptip can cause ineffective treatment and poor peptide recovery.
References 1. Kellner, R. (1994) Chemical and Enzymatic Fragmentation of Proteins. In Microcharacterization of Proteins (Kellner, R. ed.), pp. 11–27. VCH, New York 2. Henzel, W. J., and Watanabe, C. (2003) Protein identification: the origins of peptide mass fingerprinting. J. Am. Soc. Mass Spectrom. 14, 931–942 3. Yates, J.R. III, Speiche, S. Griffin, P.R., and Hunkapiller, T. (1993) Peptide mass maps: a highly informatic approach to protein identification. Anal. Biochem. 214, 397–408 4. Yates, J.R. III, Eng, J.K., McCormack, A.L., and Schieltz, D. (1995) Method to correlate tandem mass spectra of modified peptides to amino acid sequences in the protein database. Anal. Chem. 67, 1426–1436 5. Westermeier, R., and Naven, T. (2002) Proteomics in Practice. Wiley-VCH Weinheim Germany 6. Wilm, M., Shevchenko A., Houthaeve T., Breit, S., Scheweigerer, L., Fotsis, T., and Mann, M. (1996) Femtomole sequencing of
7.
8.
9.
10.
proteins from polyacrylamide gels by nanoelectrospray mass spectrometry. Nature 379, 466–469 Shevchenko, A., Wilm, M., Vorm, O., and Mann, M. (1996) Mass spectrometric sequencing of proteins from silver-stained polyacrylamide gels. Anal. Chem. 68, 850–858 Kellner, R., Lottspeigh, F., and Meyer, H.E. (ed.) (1994) Microcharacterizaion of Proteins. VCH, NY Herbert, B., Galvani, M., Hamdan, M., Olivieri, E., MacCarthy, J., Pedersen, S., and Righetti, P. G. (2001) Reduction and alkylation of proteins in preparation of twodimensional map analysis: whywhenand how? Electrophoresis 22, 2046–2205 Herbert, B., Molloy, M.P., Gooley, A.A., Walsh, B.J., Bryson, W.G., and Williams, K.L., (1998) Improved protein solubility in twodimensional-gel electrophoresis using tributyl phosphine as a reducing agent. Electrophoresis 19, 845–851
Chapter 35 Database Interrogation Algorithms for Identification of Proteins in Proteomic Separations Patricia M. Palagi, Frédérique Lisacek, and Ron D. Appel Summary Protein identification is a key aspect in the investigation of proteomes. Typically, in a 2-DE gel-based proteomics analysis, the spots are enzymatically digested and the resulting peptide masses are measured, producing mass spectra. Peptides can also be isolated and fragmented within the mass spectrometer, leading to tandem mass spectra. For protein and peptide identification, an algorithm matches the mass spectra and other empirical information against a protein database to define if a protein is already known or novel. A variety of different programs for protein identification with database interrogation has been developed. This chapter focuses on the use of the software Aldente and Phenyx, for MS and tandem MS identification, respectively. Key words: Peptide-mass fingerprinting, Tandem mass spectrometry, Protein identification, Database search, Bioinformatics.
1. Introduction Protein identification plays a central role in the investigation of proteomes, especially when making use of 2-DE gels for sample separation followed then by mass spectrometry (MS) analysis. Typically, the spots of a 2-DE gel are enzymatically digested and the resulting peptide masses are measured, producing an MS spectrum, also called a “peptide-mass fingerprint” (PMF). Peptides can also be isolated and fragmented within the mass spectrometer, leading to MS/MS spectra, also called “Peptide Fragmentation Fingerprints” (PFF). For protein and peptide identification, the user matches the mass spectra and other empirically acquired
David Sheehan and Raymond Tyther (eds.), Methods in Molecular Biology, Two-Dimensional Electrophoresis Protocols, vol. 519 © Humana Press, a part of Springer Science + Business Media, LLC 2009 DOI: 10.1007/978-1-59745-281-6_35
515
516
Palagi, Lisacek, and Appel
information against a protein database to define if a protein is already known or novel. The method used to match the experimentally produced PMF spectra against theoretical spectra calculated from protein database sequences is called “peptide-mass fingerprinting.” The method used to match PFF spectra is called “peptide fragmentation fingerprinting.” In both methods the matches between experimental and theoretical spectra are scored to provide a measure of similarity between the two spectra, a higher score indicating a higher likelihood that the corresponding protein or peptide is the target one. Although PMF can produce excellent results when searching databases for species of small and fully sequenced genomes, PFF is a more specific and sensitive identification method and is better adapted when searching in larger databases or when working with complex mixtures of peptides. The sequence information given by the amino acid fragmentation in MS/MS spectra increases the chances of finding true positive hits in database searches. It is possible that even a single peptide (a single MS/MS spectrum) will correctly identify a protein, but it depends on the number of amino acids in its sequence (which means the peptide coverage on the identified protein). However, PFF identification is not a perfect method. Many MS/MS spectra collected during an experiment may not be assigned to any peptide. Possible reasons for these nonmatches can be the presence of experimental contaminants appearing before or during MS acquisition, the badquality spectra with noise and unusual fragmentation for certain mass spectrometers, spectra derived from proteins not present in the database, or products of alternatively spliced genes not annotated in the database, etc. In general, current experimental workflows increasingly use both PMF and PFF methods in a complementary approach to increase the number of confidently identified proteins. A variety of different programs for protein identification with database interrogation has been developed, and they have been already extensively listed in the literature (1–3). In this chapter, we will focus on the use of the software Aldente for PMF identification and Phenyx for PFF identification. We consider that the reader is familiar with mass spectrometry techniques as well as with the kind of data generated by MS equipment.
2. Materials 2.1. Software and Hardware
Aldente, a tool that allows to identify proteins using PMF data (4), is available through any web browser such as Firefox or Internet
Database Interrogation Algorithms for Identification of Proteins
517
Explorer, and does not require particular hardware. Aldente is accessible from the ExPASy server (5). Its main features are: • The use of a robust method (the Hough transform) to find the deviation function of the mass spectrometer and resolve peptide match ambiguities. In particular, the method is relatively insensitive to noise (see Note 1). • A tuneable score parametrization where the user can choose the parameters that will be considered in the score and in which proportion. • An extensive use of the UniProtKB/Swiss-Prot (6, 7) annotations such as protein mature form, post-translational modifications (PTM) and alternative splicing, offering a degree of protein characterization as part of the identification procedure (see Note 2). • The possibility for the user to define one or more chemical amino acid modifications (besides the usual oxidation of methionine, acrylamide adducts on cysteine residues, alkylation products on cysteine residues, etc.) and to define the contributions of these modifications to the score. Phenyx, a software platform for the identification and characterization of proteins and peptides from MS/MS data, has a public version available through a web interface and a commercial version to be locally installed on the user’s computer. The public version is available at http://phenyx.vital-it.ch/pwi/. In both versions, the main features are: • A robust and flexible scoring algorithm based on the OLAV algorithm (8). It ensures a high accuracy of results while dealing with a wide variety of instruments data and scoring models (see Note 3). • A two-round analysis strategy that increases the sequence coverage by searching for combinatorial modifications or other special features (see Note 4). • An extensive use of UniProtB (Swiss-Prot/TrEMBL) annotations, as for Aldente. • A conflict resolution algorithm that distinguishes alternative matches and decides which is the most probable (see Note 5). • A crossvalidation structure which allows the comparison of the results of different runs and the comparison with other MS search engine results (for example, Mascot and Sequest results). 2.2. Data Format
Currently, a wide variety of vendor-specific mass spectrometry formats exists despite the fact that they contain almost the same information, e.g., the mass/charge values, less frequently the intensities of the peaks, and for the MS/MS spectra the parent
518
Palagi, Lisacek, and Appel
Table 1 Experimental data formats Acronym Aldente Phenyx Comments BTDX
x
File format defined by Bruker Daltonics
x
Format defined by SEQUEST/Thermo Electro and originally designed to list single spectra only; multiple spectra can be stored in a single file by concatenating single spectrum DTA files together
PKL
x
Format defined by Waters Corp; multiple spectra can be stored in a single file by concatenating single spectrum files together, leaving a blank line between spectra
MGF
x
Format defined by Matrix Science; multiple spectrum ASCII representation for platform-specific ASCII formats
x
Evolving XML standard (set by HUPO Proteomics Standard Initiative, PSI)
x
Evolving XML standard (set by the Institute for Systems Biology, Seattle)
DTA
mzData
x
x
mzXML PKT
x
Produced by the Data Explorer software of Applied Biosystems;
PKM
x
Voyager software of Perseptive Biosystems or the GRAMS software
mass value. A subset of these formats is given in Table 1. The programs described here were designed to cope with most of these formats.
3. Methods 3.1. PMF Identification: Submitting Data
Aldente, or any other PMF tool, compares experimentally measured, user-specified peptide masses with theoretical peptides calculated for all proteins in a database. Isoelectric point, molecular weight, the species (or group of species), and other information can be specified in order to restrict the number of candidate proteins, to reduce false-positive matches, and to increase the confidence on top-list protein(s).
Database Interrogation Algorithms for Identification of Proteins
519
1. Open the Aldente tool from the ExPASy Tools page (http:// www.expasy.org/tools/aldente/). 2. In the “Sample” section, enter a sample query name, and the pI and Mw estimations, if known (see Notes 6 and 7). 3. Enter the list of experimental peaks (see Note 8) directly into the text area (peptide masses with or without peak intensities), or upload a file in one of the accepted formats detailed in Table 1. 4. In the “Protein” section, select the database to search among the UniProtKB/Swiss-Prot database or the UniProtKB/ TrEMBL database or a database of known contaminants (see Note 9). 5. Select the species (or group of species) if you want to restrict your query (see Note 10). 6. Determine the accepted mass tolerance at the protein level by setting the upper and lower mass limits. This information is used to filter the results but is not taken into account when calculating the score (in contrast to the Mw estimation mentioned earlier). 7. Define the upper and lower pI limits accepted at the protein level. 8. In the “Peptide” section, specify the enzyme that was used to generate the peptides. In order to take into account partial cleavages, you can specify 0 or 1 missed cleavage sites to be allowed. 9. Assign the isotopic resolution of the experimental masses and the charge state of the peptides (see Note 11). 10. Specify the chemical modifications occurring on the unknown protein before cleavage, and the way to take them into account in the score (see Note 12). 11. You may also choose whether the PTMs annotated in SwissProt are taken into account or not (e.g., only experimentally proven or also computationally predicted ones) and the way to consider them in the score (see Notes 13 and 14). 12. In the “Thresholds” section, determine the peptide-mass tolerance, which corresponds to the estimated internal precision of the mass spectrometer: the instrument’s accuracy can be specified, either with an absolute value in Daltons (Da, field Spectrometer shift max) or with a relative value in parts per million (ppm, field Spectrometer slope max), or with both (see Note 15). Less accurate peptide-mass data will require a larger mass tolerance and will result in a lower accuracy of your search (see Note 16). 13. Define a minimum number of peptide-mass hits required for a matching protein to be included in the result list. The default value is 4.
520
Palagi, Lisacek, and Appel
14. Finally, in the “Output” section, specify the maximum number of proteins to be displayed in the Aldente output, and select the “Submit” button to send your query to ExPASy. Depending on whether you provide your e-mail address or not, the results will be sent to you by e-mail or displayed directly in your browser. If you are following this chapter with the Aldente search query open, you must have noted that we did not mention how to deal with the “Scoring” section or some other parameters. For a start, accept the default parameters as they are. Otherwise, you may change these values to have your own aperçu. 3.2. PMF Identification: Interpreting the Output
The output result window of Aldente is divided into three parts. The top part provides the date of the query, the database release number, and current number of entries, and some statistics about the search. The protein statistics give the total number of proteins in the selected database and taxa, the number of proteins in the protein mass range, the number of proteins with enough peptides in the peptide-mass tolerance, the number of proteins with the minimum number of hits after alignment, and the number of displayed proteins (see Note 17). Then the peptide statistics give the number of peptides generated in the mass range of your sample, the number of peptides matching a peak of the sample, and the average number of theoretical peptides per protein in the mass range. The second part is a summary of the n best-matching proteins from the database (where n is the parameter that you have chosen in the “Output” section) with a “quick jump” link to detailed peptide information provided further down in the same page. In the output example given in Fig. 1, there are 15 best-matching proteins in the list. The best match (the top protein) is highlighted in green by Aldente. Its score is 8.48 and its coverage is 24%.. The third part shows, for each matching protein, the detailed information about the matching peptides, with individual scores, differences between the experimental and calculated masses, information regarding PTM or chemical modification if any, peptide positions, and sequences. Finally the protein sequence is visualized with identified peptides in blue and uppercase, and the digestion enzyme loci (K, R for trypsin) in red. Aldente results are displayed online or sent by e-mail in html, XML, or text formats. In the html format, you will find direct links to FindMod, GlycoMod, and FindPept. These tools help to further characterize matching proteins by predicting potential PTMs, finding potential single-amino acid substitutions and potential unspecific cleavages. The html output result also contains direct links to PeptideMass, for helping the interpretation of the PMF results, and to BioGraph, for the graphical representation of the theoretical spectrum (9). Relevant input data and/or information about the matching database entry are automatically transferred to those programs.
Database Interrogation Algorithms for Identification of Proteins
521
Fig. 1. First part of the Aldente output, showing the summary of the best-matching proteins from the database. The top of the page summarizes some statistics about processed proteins and peptides, followed by the list of best-matching proteins, with related information.
You can launch a new Aldente search directly from the result output. It allows you to submit a second search with slightly modified parameters, i.e., with modified molecular weight or pI ranges, number of missed cleavages, taxonomic range; or to resubmit a stored query at a later stage. This is particularly useful if the initial identification was unsuccessful or ambiguous, or to query a new release of the database. 3.3. MS/MS Identification
Phenyx identifies and characterizes proteins and peptides by comparing experimentally measured MS/MS fragmentation masses with theoretical ion fragmentation values from protein databases. As in Subheading 3.1, the species (or group of species), the instrument type, and other information can be specified in order to restrict the number of candidate proteins and reduce false-positive matches.
522
Palagi, Lisacek, and Appel
1. Connect to Phenyx at http://phenyx.vital-it.ch/pwi. Create a new account to login or use the “guest” account (see Note 18). 2. Click on the Submission tab to start the query and enter a title for the submission. 3. Select the database to search among UniProt/Swiss-Prot, UniProt/TrEMBL, the International Protein Indexes (IPI) of human, mouse and rat proteomes, and NCBInr. 4. Select the species (or group of species) if you want to restrict your query (see Note 10). 5. Specify the type of instrument used to generate the MS/MS data and the scoring model (see Note 19). 6. Select the amino acid modifications among known chemical modifications, PTMs, crosslinking, etc. (see Note 9). Once selected, they appear in the AA Modif. Details table (the table at the right). 7. Specify the enzyme that was used to generate the peptides. In order to take into account partial cleavages of the proteins, you can allow up to four missed cleavage sites (see Note 20). 8. Determine the Parent error tolerance, which corresponds to the estimated deviation between experimental parent ion masses and the theoretical masses. These values can be given in Daltons or in parts per million. 9. In the “Acceptance Parameters” window, define the minimum accepted number of amino acids in the peptide lengths, the minimum accepted z-score values and the maximum accepted p-Values for the peptides, and the minimum AC value (see Note 21). 10. Upload the list of experimental peaks in one of the accepted formats detailed in Table 1. 11. Click on the “Submit” button to send your query to Phenyx Calculation Unit. A progress bar appears showing the process evolution (see Note 22). With these parameters, Phenyx will calculate the first round of queries. A second round can be calculated to narrow the results obtained from the first round of scoring. Only the proteins that fulfilled the first-round criteria are processed during the second round. Choose to do two rounds of calculations by checking the box corresponding to round 2 and adapt the given parameters (see Note 4). 3.4. MS/MS Identification: Interpreting the Output
Phenyx shows the output results of identified proteins and peptides in various different views. When the job is completed, the Proteins Overview window pops up. In this view, you can parse the identified proteins and see the identified peptides for each of them.
Database Interrogation Algorithms for Identification of Proteins
523
If you click on one of the spectrum titles in the compound field of the result tab, the Compounds Overview pops up. Here, you can browse the results spectrum by spectrum. In this view, you may click on the peptide sequences and reach the Peptide Match Details view, or on the Protein Details link and reach the Protein Match Details view. You can then compare different job results (for example sets of MS/MS data) in the Result Comparison window, which is available from the user console. The top part of the Protein Overview shows a link to the chosen parameters, a link to the Compounds overview, and a table summarizing the best-scoring proteins (see Fig. 2 for the example). In this table, the proteins are sorted in descending order of their score values. For each protein, these information are detailed: the accession number (AC), the identification code (ID), the effective score (i.e., the sum of the best score per valid peptide sequence), the number of valid peptide matches (see Note 23) and the total number of peptide matches found for the given protein, the percentage coverage of all amino acids of the valid peptides, the official protein name and common synonyms. By clicking in a row of this table, the row is highlighted and more information about this protein is presented at the right and at the bottom of the window. At the right, the information such as the ID, the AC, and the database name are repeated, together with a set of proteins that have matched peptides in common with the highlighted protein. At the bottom,
Fig. 2. Protein overview window from Phenyx output result.
524
Palagi, Lisacek, and Appel
the list of matched peptides for the highlighted protein details the following data (some of them are also found in other views): 1. The Auto column gives the automatic validation proposed by Phenyx. A plus sign (+) means that the peptide match is considered valid by Phenyx and a minus sign (−) means that the peptide match did not fulfill at least one of the specified submission criteria. 2. The User column gives the same kind of validation but now done by the user. As long as the user does not validate the peptides, the validity assigned by Phenyx appears by default in this space. 3. The amino acid sequence of the peptide. 4. A link to the BLAST search tool on the ExPASy server. A new window pops up and the amino acid sequence is automatically copied to the tool. You will then be able to easily search for similar proteins. 5. The charge of the theoretical peptide (z), the experimental mass-to-charge ratio (m/z), the difference between the observed m/z of the parent ion and theoretical mass of the peptide (d m/z), the z-Score, and p-Value. 6. The values of the start and end positions of the peptide in the protein sequence. 7. The number of missed cleavages found and the chemical modifications undergone on the amino acids. This information is also given in the Protein Match Details that pops up when you click on any of the Protein Details links. In this window (see Fig. 3 for the example), there are also two different graphical representations of the protein sequence and its coverage by identified peptide matches. One of the graphics gives a large overview with all amino acids represented by a one-letter code. The other gives a rough bar code representation of the protein. Clicking in one of the peptide sequences will open in a new window the Peptide Match Details. In the top of this window (see Fig. 4 for the example), there are the theoretical calculated mass of peptide, the m/z and charge of the observed parent ion, the number of peaks in the mass spectrum, the difference between the theoretical m/z of the peptide and the observed m/z of the parent ion and the peptide z-Score. Following this information, there is an interactive graphical representation of the mass spectrum (ion types and intensity values can be changed and the graphic is adapted accordingly). An extra table is given in this window showing the summary of all ions that were used to explain the peptide match. There is one column per peptide amino acid and one row per fragment series. The cells in this table are color coded giving an idea of the most intense peaks of the spectra.
Fig. 3. Protein Match Details window from Phenyx output result.
Fig. 4. Peptide Match Details window from Phenyx output result.
526
Palagi, Lisacek, and Appel
In the Compound Overview window (see Fig. 5 for the example), all identified peptides of a job are listed according to the order of appearance of their corresponding spectra (i.e., compounds) in the loaded file(s). Each peptide is associated with its sequence, a z-Score and the difference between observed and theoretical ion masses, as well as a list of proteins that have actually matched this specific peptide (i.e., proteins that have been considered to have this peptide in common). Multiple matches for a given spectrum appear with the letter M in the “Multi” column. The peptide matches are color coded in the “Sequence” column according to their definition: automatically considered as valid (white), automatically rejected (gray), winner of the conflict resolution according to the Phenyx algorithm (green), and the loser of the conflict resolution (red). A winner is a valid sequence that respects z-Score and p-Value thresholds specified by Phenyx and successfully satisfies the conflict resolution algorithm. On the other hand, a loser is a valid sequence that also respects z-Score and p-Value thresholds specified by Phenyx but does not successfully satisfy the conflict resolution algorithm. For a spectrum, there can be multiple winners and/or multiple losers. Finally, in this window, you will then validate the peptides by clicking in the “User” column.
Fig. 5. Compounds overview window from phenyx output result.
Database Interrogation Algorithms for Identification of Proteins
527
4. Notes 1. The algorithm has no major difficulties when working with very crowded spectra (i.e., with a high number of input masses, e.g., >100) and with a large number of theoretical masses. Consequently, increasing the number of possible peptides by taking into account combinations of missed cleavages, post-translational modifications, alternative splicing, and chemical modifications, is conceivable. However in addition to increasing the number of true-positive peptide matches, there is also a risk of increasing the number of false-positive hits. 2. Aldente does not do any de novo prediction of post-translational modifications on proteins. All modified peptides shown in the results will be the verification of an event documented in UniProtKB/Swiss-Prot. However, Aldente can match peptides whose modifications are documented in UniProtKB/Swiss-Prot as “potential” or “by similarity”, and thus allows predicted post-translational modifications to be validated. 3. Olav’s scoring function takes into account many physicochemical parameters of the fragment ions such as fragmentation pattern probabilities, the presence of different ion series (a, b, y etc), peak intensities, and residue modifications (as annotated in the UniProtKB/Swiss-Prot database). This gives Phenyx the ability to efficiently discriminate between true and false positive matches. In addition, the score models were optimized for specific mass spectrometers by using a training set of validated identified proteins. 4. The first round is typically run with stringent parameters on an extended database to achieve a rough isolation of proteins of the sample while keeping combinatorial explosion under control. The second round is performed with looser parameters on a limited protein database built from the potentially identified proteins. Unspecific cleavages, mutations, or a wider range of post-translational modifications can be explored during the second round. 5. A conflict can occur during scoring when a mass spectrum matches more than one peptide sequence in the selected protein databases. A given spectrum should ideally correlate to a unique molecular structure, except if a spectrum represents a mixture of peptides. Whenever possible, Phenyx’s scoring algorithm distinguishes all possible matches and decides which are the most probable. Otherwise, it reports all sequences matching with an acceptable score and the user makes the decision.
528
Palagi, Lisacek, and Appel
6. The Aldente search form can be seen in a global view with all the parameters displayed in one page (tab “All) or in a section view with the parameters grouped in a certain logical way. To jump to one of these sections, click on the respective tabs “Sample,” “Protein,” “Peptide,” “Thresholds,” “Scoring,” and “Output.” 7. If pI or Mw fields are left blank, these parameters will not be used in the Aldente score. This is the case for most optional parameters. 8. You should avoid using peptide masses known to be from autodigestion of an enzyme (e.g., trypsin), or other artifactual peaks (e.g., matrix peaks). Press the button “Remove known masses.” Aldente will mark the known masses with a double slash and hence will not use them in the query. 9. Multiple selections are possible by holding down the “Ctrl” key. This feature is available in most of Aldente’s drop-down menus. 10. A “single species matching” is used for molecularly well-defined species or ideally species that have been the subject of a genome project. This is the case, for example, when you are analyzing an E. coli sample that you then match against only the E. coli proteins in the database. Otherwise, you would better do a “cross-species matching.” For example, if working with proteins from Drosophila alpina, you may wish to either match your proteins against all proteins from insects or against the fully sequenced Drosophila melanogaster. Important things to know when doing cross-species matching: protein pI is frequently poorly conserved, but protein mass is generally very well conserved (10). You should take this into consideration when setting your pI and Mw values. On the other hand, peptide masses are not well conserved across species boundaries. The poor conservation of peptidemass data is expected, as a single amino acid substitution in any peptide can drastically change its mass. 11. The “Monoisotopic Mass” option is useful in the mass prediction of small peptides (<3,000 Da) that can often be isotopically resolved on mass spectrometers. The (M + H) + option will calculate all peptide masses with an extra hydrogen atom, to give values for singly charged peptides as found in the MALDI analysis. 12. Aldente supports any user-defined chemical modifications to amino acids. You may specify the amino acid where the modification should appear and the chemical formula of the product to add/remove on this locus. Two types of modifications can be applied. Use fixed modification whenever amino acids should be modified, and specify in the tolerance
Database Interrogation Algorithms for Identification of Proteins
529
field the number of exceptions allowed. For example, for carboxymethylation on cysteine (CAM) with fixed option and a tolerance of 1, the program will generate the theoretical peptide with all cysteines modified, and peptides with all but one cysteine modified. On other hand, use variable modification when residues may or may not be modified, and specify in the tolerance field the maximum number of modified residues expected. For example, for methionine oxidation (MSO) with variable option and a tolerance of 2, the program will generate the theoretical peptides with 0, 1, or 2 methionines modified. The program supports every combination of possible modifications (except N- or O-linked glycosylation or other complex modifications like glycan phosphatidyl-inositol anchors) occurring on the same locus. In an advanced mode, you can specify, for each modification, a factor to be applied on the score to penalize peptides with (un)modified loci. 13. For all types of PTMs annotated in UniProtKB/Swiss-Prot, two modes of peptide modification are available: fixed or variable. In fixed mode, the program will generate the theoretical peptide with all modified loci. In variable mode, the program will generate theoretical peptides with all combinations of modified or unmodified loci. If several PTMs (or chemical modifications; see Note 12) are possible at the same sequence position, the program will generate the theoretical peptides corresponding to every possible combination. 14. UniProtKB/Swiss-Prot annotation distinguishes between experimentally proven and computationally predicted posttranslational modifications, as well as those inferred by similarity (7). The UniProtKB/Swiss-Prot document “A primer on UniProtKB/Swiss-Prot annotation” (http://www.expasy.org/ txt/annbioch.txt) describes how these qualifiers are used. 15. Mass spectrometers usually have a mass-dependent error associated with mass measurements, which cannot be uniformly expressed in Daltons. The use of ppm can therefore be more accurate. If both Spectrometer shift max and Spectrometer slope max have been specified, the program will combine them with a logical OR. 16. Current available mass spectrometers are capable of acquiring masses with single decimal point resolution; however, it may depend on the machine calibration and the use of internal standards. Values of 0.2 Da or 200 ppm tolerance would be fair, whenever possible. 17. Aldente performs two runs. In the first one, it keeps the n best proteins within the protein and peptide-mass tolerance ranges set by the user. In the second pass, it finds the best line fitting the maximum of hits (matching peptide masses) only
530
Palagi, Lisacek, and Appel
for those n best proteins. Because of a higher precision in the second step, some hits from the first step may be removed, which means that the number of hits in the output result can be smaller than the number of hits you have requested. 18. A personal account allows saving the personal settings and to retrieve old jobs (old queries). The guest login, although fully functional, does not give you access to personal settings and jobs report, but is a good way of starting anyway. 19. The algorithm that calculates the score between matches of experimental and theoretical masses is based on several instrumental properties such as the type and charge state of the fragment ions produced. If the exact instrument that was used to generate the MS/MS data is not listed then chose a similar method. 20. The cleavage mode allows normal cleavages, on both termini of the protein as defined by the cleavage rules, or half-cleavages, at just one end specifically and the other end nonspecifically (either the C- or N-terminus). 21. Peptides shorter than the specified number of amino acids (Minimum Peptide Length field) are reported in the peptide match results, but they do not contribute to the protein score. In fact, they are labeled as invalid in the Phenyx results. The Minimum Peptide z-Score is also known as the peptide or match score. The distribution of calculated scores is compared to that of random peptide sequences. The mean and the variance are calculated. The z-Score measures the deviation of the peptide score from the distribution’s mean. The z-Score is a normalized score and thus independent of the size of the search space. The Maximum Peptide p-Value is the probability that a peptide matches by chance with this score or better. A p-Value has a maximum of 1.0. The larger the search space, the higher the p-value since the chance of a peptide being a random match increases. The lower the p-Value, the more significant is the match. For a single search (or set of sampled peptides), you can compare z-Scores. However, when two or more searches are performed on different size spaces, you first need to look at the p-Values before comparing z-Scores. The AC Score is the minimum significant value for a protein’s accession number score. It is the sum of the best scores per valid peptide sequences. Protein matches scoring lower than this value are rejected from the identified list of proteins. 22. You will need to allow cookies handling in your browser; otherwise, the queries will not be sent by Phenyx. Accepting pop-up windows is also highly recommended. 23. A valid peptide has the Acceptance Parameters values higher than the thresholds set by the user (see Subheading 3.3, step 9).
Database Interrogation Algorithms for Identification of Proteins
531
References 1. Palagi, P. M., Hernandez, P., Walther, D., and Appel, R. D. (2006) Proteome informatics I: Bioinformatics tools for processing experimental data. Proteomics. 6, 5435–5444. 2. Hernandez, P., Müller, M., and Appel, R. D. (2006) Automated protein identification by tandem mass spectrometry: Issues and strategies. Mass Spectrom Rev. 25, 235–254. 3. Johnson, R. S., Davis, M. T., Taylor, J. A., and Patterson, S. D. (2005) Informatics for protein identification by mass spectrometry. Methods. 35, 223–236. 4. Tuloup, M., Hernandez, C., Coro, I., Hoogland, C., Binz, P. A., and Appel, R. D. (2003) Aldente and BioGraph: An Improved Peptide Mass Fingerprinting Protein Identification Environment (Ducret, A., Eberle, A.N., Faupel, M.-D. et al., eds), FontisMedia, Lausanne, Switzerland. 5. Gasteiger, E., Gattiker, A., Hoogland, C., Ivanyi, I., Appel, R. D., and Bairoch, A. (2003) ExPASy: The proteomics server for indepth protein knowledge and analysis. Nucleic Acids Res. 31, 3784–3788. 6. Bairoch,A.,Apweiler,R.,Wu,C.H.,Barker,W.C., Boeckmann, B., Ferro, S., Gasteiger, E.,
7.
8.
9.
10.
Huang, H., Lopez, R., Magrane, M., Martin, M. J., Natale, D. A., O’Donovan, C., Redaschi, N., and Yeh, L. S. (2005) The Universal Protein Resource (UniProt). Nucleic Acids Res. 33 Database Issue, D154–D159. Farriol-Mathis, N., Garavelli, J. S., Boeckmann, B., Duvaud, S., Gasteiger, E., Gateau, A., Veuthey, A. L., and Bairoch, A. (2004) Annotation of post-translational modifications in the Swiss-Prot knowledge base. Proteomics. 4, 1537–1550. Colinge, J., Masselot, A., Giron, M., Dessingy, T., and Magnin, J. (2003) OLAV: towards high-throughput tandem mass spectrometry data identification. Proteomics. 3, 1454–1463. Gasteiger, E., Hoogland, C., Gattiker, A., Duvaud, S., Wilkins, M. R., Appel, R. D., and Bairoch, A. (2005) Protein Identification and Analysis Tools on the ExPASy Server. In The Proteomics Protocols Handbook (Walker, J.M., ed), Humana Press, NJ, pp. 571–607. Wilkins, M. R., Williams, K. L. (1997) Crossspecies protein identification using amino acid composition, peptide mass fingerprinting, isoelectric point and molecular mass: a theoretical evaluation. J Theor Biol. 186, 7–15.
Chapter 36 Creating 2DE Databases for the World Wide Web Christine Hoogland, Khaled Mostaguir, and Ron D. Appel Summary With the development of the Internet, a growing number of two-dimensional electrophoresis (2DE) databases have become available (60 in 2009, for a total of 425 image maps). By linking the two components constituting 2DE databases, gel images and protein information, the active hypertext links provide a powerful tool for data integration, in addition to navigation from one database to another. This chapter shows how to prepare the necessary files to build a federated 2DE database in order to make it available over the Internet and how to further update it. Key words: Federated database, Internet, Data integration.
1. Introduction A federated two-dimensional electrophoresis (2DE) database is a database containing 2DE data (gel images, protein description, experimental information at the separation level and at the postseparation analysis level) that are accessible on the World Wide Web (WWW). These data are dynamically linked to other similar databases through active hypertext links over the Internet. Some rules have been proposed to define a federated 2DE database (1) that can be summarized as follows: the database must be remotely accessible on the WWW through keywords search and graphical query (basically by clicking on a spot in a 2DE gel image), and it must have bidirectional cross-references (i.e., must be linked) to at least a reference protein database called the main index. Existing fully or partially federated databases (60 in 2009, for a total of 425 image maps) are listed on the ExPASy proteomics website (2) at: http://www.expasy.org/ch2d/2d-index.html. David Sheehan and Raymond Tyther (eds.), Methods in Molecular Biology, Two-Dimensional Electrophoresis Protocols, vol. 519 © Humana Press, a part of Springer Science + Business Media, LLC 2009 DOI: 10.1007/978-1-59745-281-6_36
533
534
Hoogland, Mostaguir, and Appel
More recently, new tools have been developed to enhance proteomics data integration. The Make2D-DB package (3) is such a tool, specifically designed automatically to interconnect federated 2DE databases, which become simultaneously accessible from a single web portal (see an example at http://www. expasy.org/world-2dpage/). In this chapter, we will focus on how to prepare the necessary files to build a federated 2DE database in order to make it available over the Internet, and how to update it further.
2. Materials 1. Computer connected to the Internet, such as an Apple Macintosh®, a Windows-based PC, or a Unix X-Windows system. 2. WWW browser such as Mozilla family, Netscape® Navigator, or Microsoft® Internet Explorer. Free copies of WWW browser can generally be downloaded from Internet. See for example http://www.mozilla.org/. 3. HTTPD Web server daemon such as Apache freely available at http://www.apache.org 4. Perl interpreter (version 5.6 or higher, retrievable from http://www.perl.org/), including DBI module and PostgreSQL driver DBD::Pg. 5. PostgreSQL server (version 7.4 or higher, freely available at: http://www.postgresql.org) 6. Make2D-DB, freely available at: http://www.expasy.org/ ch2d/make2ddb/. Make2D-DB is software designed to create, convert, publish, interconnect, and keep up-to-date federated 2DE databases. 7. Optionally, Melanie 2D gel analysis software (GeneBio S.A.). A free viewer can be downloaded from Internet (http://www. expasy.org/melanie/).
3. Methods To create a 2DE database, one needs at least one gel image and a list of protein identifications with their X and Y coordinates on that gel. The method described here will show one way among others to create and install such a 2DE database, using
Ctreating 2DE Databases for the World Wide Web
535
the Make2D-DB package (3) as it greatly simplifies the task. It is recommended to refer to the Make2D-DB user manual for more extensive documentation on how to get maximum benefit from the tool (see Note 1). Assuming that Perl, a HTTP server, and PostgreSQL are installed and running, the following operations must be carried out: 3.1. Data Preparation
The first step is to prepare the data itself (gel image and protein list) (see Note 2). These data files have to be placed in the same directory; let us call this directory “Data”. This directory should be on the same computer as the web server.
3.1.1. Gel Image
The gel image(s) may be stored in image format such as gif, png, jpg, etc. For each gel of the database, one should create 2 corresponding images in such format. One is the image in full size (either original or with annotation such as axis legends), the other is a small one (with approximately a size of 100 pixels × 100 pixels), to be used as an icon on the web server. Both the image files should have exactly the same name, although the small image name should be preceded by the prefix “small_”.
3.1.2. Protein Identification
The list of protein identification may be stored in tab-delimited text files (see Note 3), such as export from spreadsheet software (i.e., Excel). The mandatory columns are the following: • SPOT: the spot identifier (a number) • X: x-coordinate of the spot on the gel image (the width value) in pixel • Y: y-coordinate of the spot on the gel image (the height value) in pixel • MW: experimental molecular weight (from the gel), in Dalton • PI: experimental isoelectric point (from the gel), only for 2DE gel • AC: the protein identifier (typically from the main index) The following columns are optional but highly recommended: Mapping Methods: the identification method(s) used (such as PMF, MS/MS, etc.) Then experimental data corresponding to this identification should be detailed. These data can be given in different ways, either as literal text information (see Note 4), or as text file (for supported formats, see Note 5) located in the same computer or as link to another resource, such as MS repositories (for example: PRIDE, http://www.ebi.ac.uk/pride/). Create a separate protein list file for each gel image. This file should be named exactly as the gel image, although the extension should instead be “.txt”.
536
Hoogland, Mostaguir, and Appel
3.1.3. Private Data
Private data are data hidden from the public view, i.e., data only visible to authorized users whenever connected. Hidden data can be defined at different levels (gel, protein, spot, and identification method) and should be listed in separate files: • hiddenGels.txt, in which are listed the gels to be hidden as a whole. • hiddenEntries.txt, in which are listed the protein accession numbers to be hidden completely. • hiddenSpots.txt, in which are listed the spot identifiers to be hidden (see Note 6).
3.2. Server Configuration
Run the Make2D-DB package with option “config” to set up the configuration files for both the database installation and the Web server pages (see Note 7). This step will create three files: • include.cfg, the configuration file for the database conversion/creation process • 2d_include.pl, the configuration file for the Web server interface • existing.maps, the configuration file for the gel images list and related data During this step, the Make2D-DB script will ask for information mainly based on specific user definitions (database administrator and query user usernames and passwords), paths and locations of necessary files, default values in case of missing information, display and specific settings (database name, interface colours, copyright, etc.), detailed information on gel preparation (protocols according to current standards(4) as plain text or link to remote repository), and gel image (width, height, pI/Mw ranges, sample species and/or tissue, etc.).
3.3. Database Creation and Installation
Run the Make2D-DB package with option “transform” to proceed to database installation. This step consists in verifying data consistency, creating the database relational structure, loading the data into the database, and installing the associated web server. The tool also connects to the Internet to gather some additional external data (such as UniprotKB/Swiss-Prot descriptions, gene names, keywords, etc., or NCBI taxonomy, etc.) related to the proteins (see Note 8).
3.4. Database Navigation
Once created, you have access to your 2DE database through a WWW browser. Using the Make2D-DB package provides many interesting built-in features (Fig. 1). As an example, the home page offers detailed statistics of the database content (computed automatically), various search engines (to search for data by protein name or description, gene name, authors, spot number, identification methods, full text, pI/Mw range), as well as graphical queries (by selecting a spot on a chosen reference map). It is
Ctreating 2DE Databases for the World Wide Web
537
Fig. 1. Web interface of a 2DE database created with the Make2D-DB package. (1) Home page of the database with detailed description of the current content. (2) Quick links to search engines (either textual or graphical). (3) Queries may be performed either on only one chosen database, or on a selection of project(s) (if any), and even on any remote database built with the same package. (4) Example of a graphical search on one chosen reference map. Crosses show identified proteins. (5) The NiceView of a database entry shows its content in a user-friendly view. (6) General information describing the protein entry (name, origin, references), provided with links to specific online databases such as PubMed or NCBI’s taxonomic classification. (7) Icons of reference maps available for the shown protein. Annotations including here experimental pI/MW, peptide masses or tandem mass spectrometry data, mapping methods, etc. (8) Spectra viewer for one specific MS data. (9) Additional data automatically integrated from the UniProtKB/Swiss-Prot database.
also possible to retrieve in a table all the protein entries identified on a given reference map, with all 2DE information (spot number, experimental pI/Mw, mapping procedure, references). Moreover, the built-in interface allows querying from the same place of other 2DE databases set-up with the same tool package. More importantly it allows the database being referenced through the World-2DPAGE portal (http://www.expasy.org/ world-2dpage/) thus leading to a large 2DE databases network. 3.5. Database Update
To add or modify 2DE data such as adding a new gel image, adding new protein identification to existing gel image(s), or modifying existing data, follow the instructions to prepare the new
538
Hoogland, Mostaguir, and Appel
files (Subheading 3.2; see above), to be put in the “Data” directory. Then run the Make2D-DB package with option “update” to proceed to database update. External data gathered during the installation process may also be updated through the administrator interface of the database. It is recommended to execute such updates regularly, to make sure the database is always synchronized with those external resources (UniProt definitions, taxonomy, cross-references to remote 2DE databases, etc.).
4. Notes 1. The Make2D-DB user manual is accessible at: http://www. expasy.org/ch2d/make2ddb/. A complete description on how to prepare the data can be found at: http://www.expasy. org/ch2d/make2ddb/2.Readme_preparation.html 2. For testing purpose, a test database can be built with a set of sample data, included in the distribution of the Make2D-DB package. In that case, Subheading 3.1 can be omitted. 3. Alternatively, one could prepare a plain text file in SWISS2DPAGE-like format (a full description of this format can be found at: http://www.expasy.org/ch2d/manch2d.html) or a Melanie spot report (exported as text or XML). 4. The experimental information should be given in SWISS2DPAGE-like format; you may refer to the Make2D-DB user manual for a full description: http://www.expasy.org/ch2d/ make2ddb/2.Readme_preparation.html#annotations 5. Mass Spectrometry (either PMF or MS/MS) data files may be mzML (previously known as mzData), mzXML, dta, mgf, pkl, etc. 6. It is possible to hide one, several, or all spots for a specific protein entry, or spots corresponding to a specific identification method (such as PMF, MS/MS, etc.) for a specific gel/ protein or not. All combinations of spot, protein, identification method and/or gel can be specifically hidden. 7. For a complete description on configuration details, refer to corresponding README file at: http://www.expasy.org/ ch2d/make2ddb/3.Readme_configuration.html 8. It is highly recommended that you provide protein accession numbers similar to those of Swiss-Prot/TrEMBL (cf. UniProtKB user manual at: http://www.expasy.org/ sprot/userman.html) or at least provide cross-references to
Ctreating 2DE Databases for the World Wide Web
539
corresponding Swiss-Prot/TrEMBL entries to get the maximum use of this external data-gathering feature.
Acknowledgments Our thanks go to Frédérique Lisacek who read the manuscript and provided valuable comments.
References 1. Appel, R.D., Bairoch, A., Sanchez, J.C., Vargas, J.R., Golaz, O., Pasquali, C., and Hochstrasser, D.F. (1996) Federated 2-DE database: a simple means of publishing 2-DE data. Electrophoresis 17, 540–546. 2. Hoogland, C., Sanchez, J.C., Walther, D., Baujard, V., Baujard, O., Tonella, L., et al. (1999) Two-dimensional electrophoresis resources available from ExPASy. Electrophoresis 20, 3568–3571.
3. Mostaguir, K., Hoogland, C., Binz, P.A., and Appel, R.D. (2003) The Make 2D-DB II package: conversion of federated two-dimensional gel electrophoresis databases into a relational format and interconnection of distributed databases. Proteomics 3, 1441–1444. 4. Taylor, C.F. (2006) Minimum reporting requirements for proteomics: A MIAPE Primer. Proteomics 6, 39–44.
INDEX A Active protease mapping ....................................... 431–437 Activity-based protein profiling (ABPP)............... 417–428 Affinity chromatography ........................................... 55–56, 399 depletion .................................................................... 56 tags............................................................................. 88 Albertsson, P.-A. ............................................................. 67 Alkylation .......................................................25–26, 35, 60 Alzheimer’s Disease ....................................................... 352 Antibody glutathione............................................................... 398 selection ........................................................... 105–106 Arabidopsis ......................................................... 76–77, 206 growth conditions .................................................... 210 Artificial neural networks (ANN) ..................455, 457–459 Autoradiography................................................................ 8
B Bacillus subtilis.............................................36, 59, 132, 141 Bacterial proteins ................................................... 131–144 preparation....................................................... 134–135 Bathymodiolus azoricus............................................ 197–198 Biotin avidin... ............................................................ 401–403 switch assay................................................................ 95 Biotinylation ...............................................95–96, 402, 419 Blue native polyacrylamide gel electrophoresis (BN-PAGE) ........................................ 241–257 first dimension ......................................................... 244 MS analysis in ................................................. 250–257 one dimesional (1D) ........................................ 246–247 second dimension ............................................ 244–245 two dimensional (2DE) ................................... 247–249 Brain proteins .................................................181–185, 189
C C-18 reversed phase ........................................................ 72 Carbamylation ........................................................... 24–25 Carrier ampholyte (CA) .............. 25, 27, 32, 227–228–229, 232–233, 320–322 Casein ............................................................................ 432 Caenorhabditis elegans............................................. 145–169 culture.. ............................................................ 161–162 dauer larvae ...............................................147–153, 161
differential protein expression.......................... 149–152 egg proteome ............................................153–157, 162 Carboxypeptidase Y ....................................................... 479 Cell culture.. ............................. 367, 388–389, 403, 405, 421 suspension................................................................ 210 Centrifugation gradient...................................................................... 67 Chaotropes .....................................................22–23, 26–27 CHAPS ........................................................... 24, 188–189 Charge-coupled device (CCD)...................................... 115 Chemiluminescence................................106–107, 379–380 Chloroplast isolation ................................................................... 207 proteome.............................................................. 74–76 Chromatographic two-phase partition ...................... 67–68 Chromosomes human metaphase ............................................ 259–270 preparation of .................................................. 260–261 Chymotrypsin................................................................ 492 CNBr cleavage................................................471, 478–479 Co-depletion ................................................................... 56 Codon adaptation index (CAI)........................................ 49, 58 bias index (CBI) .................................................. 49, 58 Collision-induced dissociation (CID) ....................... 72, 84 Colloidal Coomassie staining ..............................8, 88, 134, 138–139, 147–148 Combinatorial library ................................................ 56–57 Coumarin tag ............................................................ 90–91 Critical micellar concentration (CMC) ........................... 27 Critical micellar temperature (CMT) .............................. 27 C-terminal sequencing .......................................... 469–481 Cup loading ......................................................7, 35, 39, 60 CyDye™ fluors .........107, 167, 224–227, 441–443, 450–451 Cysteine oxidation .......................................................... 363–374 pKa.............. ............................................................ 363 S-nitrosylation ......................................................... 398 tagging. ................................................................ 73, 74
D Dansyl tag.................................................................. 90–91 Databases......................................................... 33, 533–539 searching ................................... 392, 395, 503, 515–531 Dauer larvae, see Caenorhabditis elegans
541
TWO-DIMENSIONAL ELECTROPHORESIS PROTOCOLS 542 Index DeCyder™ software...........................................73, 120, 451 Delta2D™ software ........................................................ 120 De novo sequencing .................. 72, 86–87, 91–92, 495–499 Dephosphorylation ........................................................ 418 Derivatization ........................................................ 496–499 Destreak reagent .............................................................. 35 Detergents ................................................23–25, 27–28, 71 Diagonal gel electrophoresis .................................. 305–309 Diamide......................................................................... 404 Difference gel electrophoresis (DIGE)..........9–10, 88, 107, 114–115, 118, 139–140, 153–157, 159–160, 162–164, 224–229, 233–234, 300–301, 324–325, 439–453 Diffusion coefficient (D) ................................................. 32 Disulphide bridge..............................................22, 25–26, 305–306 exchange .................................................................. 398 Dithioerythritol (DTE) ................................................... 35 Dithiothreitol (DTT) ...................................................... 35 Dynamic range ................... 47–48, 51, 56–57, 71, 122, 292
E Edman degradation ......................................................... 78 Electroendoosmosis ................................................7, 32, 36 Electrophoresis free-flow ........................................................ 52–53, 67 preparative ........................................................... 52–55 vertical 201 Electrospray ionization (ESI), see mass spectrometry (MS) Endoplasmic reticulum (ER) ..........................277, 282–283 Enterocyte preparation .................................................. 281 Equalizer technology ....................................................... 95 Escherichia coli .......................................................... 58, 132 N-Ethylmaleimide..........................................364, 372–373 biotin... ............................................................ 401–402 Etioplasts ............................................................... 206–207 Expression proteomics ................................................... 274
F FASTS .......................................................................... 504 Ficoll ............................................................................... 67 Fluorescein .................................................................... 364 Fluorography ................................................................... 10 Free response operator curve ((FROC) ...................................................... 116 Functional proteomics ........................................... 274, 440
G Gel drying.. .............................................319–320, 329–330 imaging ........................................................................ 9 Gelatine ......................................................................... 432 Gelfox™ software ........................................................... 120 Genome ........................................................................... 76
Glutathione ................................................................... 397 Glutathionylation .................................................. 397–412 Guanidination ........................................................... 89–90 Golgi apparatus preparation ...................277–278, 282–283 GRAVY ........................................................................... 58 Grid coordinates .................................................... 460–461 Guanidination ....................................89–91, 472, 495–505
H Halorhodospira halophila ....................................92, 495, 497 Hamster, Syrian golden ................................................. 276 Heart proteins ................................................181–185, 189 Helicobacter pylori ............................132–133, 136, 140–143 Hemoglobin .................................................................. 399 Hepatocyte preparation ................................................. 281 High-performance liquid chromatography (HPLC) ......................................................... 72 High-resolution electrophoresis ............................ 311–337 Human genome ...................................................................... 61 brain proteome project (HUPO) ............................. 234 4-Hydroxy-2-nonenal ........................................... 351–360
I ICAT ....................................................................... 73, 485 Image analysis.............................. 113–128, 160, 165, 250, 449 acquisition..........120–121, 164, 168, 435, 443, 448–449 adjustment ............................................................... 460 Quant™ software .......................................122, 164, 460 software ............................................119–120, 121–123 warping .....................................................140–141, 460 ImageMaster™ software ......................................... 120, 165 Imidazole tag ............................................................. 90–91 Immobilines............................................................... 32–33 Immobilized metal affinity chromatography (IMAC) ............................................... 390–391 Immobilized pH gradient (IPG) .....................7, 26, 32–33, 49–50, 159, 227–230, 386–387, 390, 404–407, 434, 447 Immunoaffinity capture ................................................... 68 Immunoblotting ............... 78, 103–109, 353, 379–380, 401 optimization ............................................................ 108 Inclusion compounds....................................................... 27 In-gel digestion ......................................164–165, 509–510 Iodoacetamide ............................................................... 471 5-Iodoacetamidofluorescein (IAF) ........................ 363–364 detection .......................................................... 369–372 labelling ........................................................... 366–368 IPGMAKER program .................................................... 36 Isoelectric beads..... ..................................................................... 54 focusing (IEF) ..................... 6–7, 19, 25, 137, 200–201, 229–230, 283–284, 315, 320–322, 325–328, 353, 368–369, 386, 389–390, 434–435
TWO-DIMENSIONAL ELECTROPHORESIS PROTOCOLS 543 Index microscale solution .................................... 291–303 “off-gel”.......................................................... 54–55 point, see pI iTRAQ ...................................................59, 73, 93–94, 485 Isotope-coded affinity label (ICPL) ................................ 94 Isotope-coded affinity tag (ICAT) ...................... 73, 93–95
J Jurkat T-lymphoma cells........................................ 365–366
L Large format gels...................... 34, 171, 332–334, 485–486 Liquid chromatography-tandem MS (LC-tandem MS)........................... 78, 85, 485–486, 488–489 Liquid-gel transition ....................................................... 25 Liver proteins ........................................................ 181–188 Lys-C (lysyl endopeptidase ........................................... 492 Lysyl endopeptidase (Lys-C)......................................... 492
M Make2D-DB ..................................................534, 536, 538 Marine molluscs .................................................... 197–203 Mascot ............................................................486, 489–491 MasLynx................................................................ 486, 489 Mass spectrometry (MS) ............................71–74, 250–257 electrospray ionization (ESI) ..........71, 83–84, 485, 507 FT-ICR ..................................................................... 57 MALDI...........................................71–72, 83–84, 160, 164–165, 285–286, 471, 477, 507, 511 tandem (MS/MS) ............................................... 72–73, 84–85, 286–287, 384–385, 385–387, 391–392, 477, 484, 496–499, 503, 505, 508, 511, 521–527 Medaka...........................................................459, 462, 465 Medium format gels ...................................................... 143 Melanie™ software ..........................................120, 456, 534 MICADO ....................................................................... 36 Microsomes preparation .........................277–278, 281–282 Micromal proteomics ............................................ 273–287 MIPS ............................................................................... 36 Mitochondrial proteome ............................76–78, 242–243 Mitochondrion .............................................................. 241 preparation............................................................... 246 Mouse tissues ......................... 176–181, 225, 252–257, 300 MS Blast........................................................................ 503 MS Homology .............................................................. 504 Mussel ........................................................................... 379 Multi-compartment electrolyser (MCE) ....................................50–51, 53–54, 61 Multidimensional liquid chromatography (MDLC) ........................................................ 77
N 3-Nitrotyrosine ...................................................... 351–360 Non-covalent interactions ......................................... 22–25
Nonlinear gradient .......................................................... 34 Nycodenz ......................................................................... 67
O o-Methyl isourea .............................................................. 89 One-dimensional electrophoresis (1DE) .......................................59, 70, 133–134 Organelle 65–69, 172, 259, 273–276 lysis....... ................................................................... 214 markers ...................................................................... 68 Oxidation, of cysteine ............................................ 364, 368 Oxidative phosphorylation ............................................... 241–242 stress.... .....................................305–306, 351–352, 399 Oxyblot.......................................................................... 353
P PDQuest™ software ................ 116, 120, 125, 369–370, 456 Peptide fragmentation ........................................ 91–93, 515 mass fingerprinting (PMF)............37, 84, 160–161, 515–516, 518–521 Percoll ...............................................67, 211–212, 264–265 pH altering....................................................................... 23 gradient................................................................ 31–37 narrowing of ...................................37–41, 299–300 Phenyx ........................................................................... 517 pI ....................................................................... 6–7, 26–27 Phosphopeptide ..............................................390–391, 395 Phosphorimaging .................................................... 11, 412 Phosphorylation .................................................... 383–395 Plasmodium falciparum ..................................................... 76 Plastid proteome.................................................... 205–219 Post-source decay ............................................................ 84 Post-translational modification (PTM)...............73, 95–96, 383–395, 397–398, 469 Prechylomicron transport vesicle (PCTV) budding assay ............................................... 283 Prefractionation ..................................... 41, 41–42, 58, 172, 181–188, 193–194, 206, 291–303 Progenesis™ software ............................................. 120, 139 Prostate specific antigen (PSA) ....................................... 39 Protease ................................................................. 431, 437 in-gel digestion .........................................485, 486–488 V8 (Glu-C).............................................................. 492 Protein aggregation ................................................................ 26 alkaline .......................................................... 48, 58–61 carbonyls .......................................................... 351–353 complexes ............................................................ 70–71 dephosphorylation ................................................... 418 depletion ........................................................ 55–56, 95 difficult ................................................................ 47–48 enrichment ................................................................ 95
TWO-DIMENSIONAL ELECTROPHORESIS PROTOCOLS 544 Index Rotofor ............................................................ 53, 292–293 Ruditapes decussatus ................................................ 197–198
Protein (Continued) expression profiling .......................................... 455–467 extraction ..................................171–196, 199, 214–216 fractionated ......................... 50, 172, 181–188, 193–194 fragmentation .............................87, 496–499, 502–503 glutathionylation.............................................. 397–412 high-abundance ................................................... 55–56 high mass ................................................................... 53 hydrophobic ......................................................... 11, 70 interactions .......................................................... 21 import to mitochondria ....................................... 76–77 large..... ...................................................................... 11 loading. .............................................136–137, 192–193 low abundance ..................................................... 48–52 membrane-bound .............................. 11, 27, 48, 54, 58, 70, 77, 215–216 neutral........................................................................ 58 nuclear.................................................................. 27, 36 plasma.. ................................................................ 56–57 phosphorylation ............................................... 383–395 precipitation............................................... 26, 210–211 prefractionation .....................................41–42, 58, 172, 181–188, 193–194, 206, 291–303 protein interactions .............................................. 21–25 ribosomal ................................................................... 36 sequencing ................................................84–85, 89–90 solubility .............................................................. 25–27 solubilization ................................19–25, 188–189, 208 total...........................................................172, 176–181 unfolding ............................................................. 22–23 water-soluble ........................................................... 214 Proteome definition ................................................................. 221 prediction............................................................. 74–77 Proteomics ..................................................................... 3–4 organelle .............................................................. 65–69 Proteominer™............................................................. 56–57 Proteomweaver™ software .............................................. 120
S-cysteine .................................................................... 403 Saccharomyces cerevisiae ............................................... 39, 76 SameSpots™ software..................................................... 120 SDS ................................................................................. 22 PAGE.. ............................................230–231, 315–316, 322–324, 328–329, 447–448 SEQUEST .................................................................... 395 Shewanella oneidensis .......................................470, 473–477 Shotgun proteomics........................ 52, 78, 84, 93, 483–493 SILAC........................................................................... 485 Silver staining ....................................................8, 219, 324, 329–330, 339–350 equipment ................................................................ 341 procedures........................................................ 343–350 solutions........................................................... 342–343 Small format gels ........................................................... 143 Sonication.............................................................. 189–190 Spot detection ...................................8–9, 115–117, 126–127 digitizing.......................................................... 461–462 editing.............................................................. 123–125 matching ...................................................116, 126–127 normalization....................................117–118, 124–125 Staining..... ............................................8–9, 105, 114–115, 222–223, 245–246, 440 Statistics in image analysis .......................................118, 125–126 Streptavidin-horseradish peroxidise ............................................. 402–403 Stripping immunoblots.......................................... 107–108 Subcellular fractionation ............... 32, 66–69, 211–213, 274 Sucrose gradient ...................................................... 67, 280 Sulphonation ..............................................91–93, 495–505 Sypro stains ........................................ 8, 114–115, 219, 357
Q
T
Quantification by tagging ............................................................ 93–95 Quantity One software .................................................. 435
Tandem MS, see Mass spectrometry Tagging applications .......................................................... 86–87 chemical......................................................... 83, 86–96 Thiolate anion ............................................................... 363 Thiol detection ...................................................... 363–374 Thioredoxin ................................................................... 365 Thiourea .......................................................................... 26 Time of flight (TOF) ...................................................... 84 T-lymphocytes ............................................................... 405 Tomato .......................................................................... 465 Tryptic digestion............................................160, 164–165, 502, 507–513
R Radiolabelling...................................................... 10–11, 34 Reflectron MS spectrum ............................................... 498 Resolution.............................................................. 115, 331 high.................................................................. 221–238 Resolving power .............................................................. 32 Rhodamine .................................................................... 419 Rice proteins .......................................................... 209, 212
S 35
TWO-DIMENSIONAL ELECTROPHORESIS PROTOCOLS 545 Index Two-dimensional electrophoresis (2DE) ...................................135–138, 291–294 databases .......................................................... 533–539 development of .................................3–4, 47–48, 83–84 first dimension .................................................6–7, 137, 200–201, 208, 216–217, 266–267, 283–284, 315, 320–322, 325–328, 353, 355–356, 368–369, 404–407, 434–435, 442, 446, 500–501 highly-reduced ..................................262–263, 266–268 interfering substances ........................................ 5–6, 20 large gel format ................................................ 311–337 limitations of ............................... 11–12, 28, 41, 69–70, 173–174, 291–292, 311–314, 431–432, 484 principle of....................................................... 4–5, 103 radical-free ........................................262–263, 266–268 second dimension ....................................7–8, 137–138, 159, 201, 208, 217–219, 267–268, 284–285, 315–316, 322–324, 328–329, 353, 356–357, 369, 407–408, 435, 442–443, 447, 501 Two-phase partitioning ................................................... 68 Tyrosine phosphatases ........................................... 417–428
U Ubiquitin proteasome system (UPS) ............................. 377 Ubiquitination ....................................................... 377–380 Urea ............................................................23–27, 188–189
V V8 protease(Glu-C) ...................................................... 492
W Western blotting ............................ 103–109, 245, 249–250, 353, 359, 379–380, 422 Worldwide web (www) .......................................... 533–539
X X-ray film ...................................................................... 380
Z Zoom gels .........................................................12, 31–2, 39 Zoom® IEF fractionator .................................294, 296–298