Proteomic Profiling - Methods and Protocols

Methods in
Molecular Biology 1295
Anton Posch Editor
Proteomic
Profiling
Methods and Protocols
METHODS
IN
MOLECULAR BIOLOGY
Series Editor
John M. Walker
School of Life and Medical Sciences
University of Hertfordshire
Hatfield, Hertfordshire, AL10 9AB, UK
For further volumes:

http://www.springer.com/series/7651
Proteomic Profiling
Methods and Protocols
Edited by
Anton Posch
Bio-Rad Laboratories GmbH, Munich, Germany
Editor
Anton Posch
Bio-Rad Laboratories GmbH
Munich, Germany
ISSN 1064-3745
ISSN 1940-6029 (electronic)
Methods in Molecular Biology
ISBN 978-1-4939-2549-0
ISBN 978-1-4939-2550-6 (eBook)
DOI 10.1007/978-1-4939-2550-6
Library of Congress Control Number: 2015933382
Springer New York Heidelberg Dordrecht London
Springer Science+Business Media New York 2015
This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the material is
concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction
on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation,
computer software, or by similar or dissimilar methodology now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not
imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and
regulations and therefore free for general use.
The publisher, the authors and the editors are safe to assume that the advice and information in this book are believed to
be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty,
express or implied, with respect to the material contained herein or for any errors or omissions that may have been made.
Printed on acid-free paper
Humana Press is a brand of Springer
Springer Science+Business Media LLC New York is part of Springer Science+Business Media (www.springer.com)
Preface
Sample Preparation Methods and Protein Techniques for Proteomic Profiling
This volume is a comprehensive continuation and extension of a book called 2D PAGE:
Sample Preparation and Fractionation which was published in 2008.
This book presents the latest developments of the main pillars of protein analysis,
namely sample preparation, separation, and characterization. Individual technologies of
each pillar combined into complementary and robust workflows render proteomic analysis
of complex biological samples even more powerful and are the prerequisite to gain maximum
value from biological samples in a single experiment.
In this volume, basic but important sample preparation protocols are described again,
followed by sophisticated procedures to enrich for specific protein classes and completed by
the detailed description of integrated workflows for comprehensive protein analysis and
characterization. The authors of the individual chapters are well-known protein biochemists, and all of them have set value to provide a detailed representation of their lab work and
to share important tips and tricks for a successful and reproducible employment of their
precious protocols in other laboratories.
This book is for students of Biochemistry, Biomedicine, Biology, and Genomics and
will be an invaluable source for the experienced, practicing scientist, too.
Munich, Germany
Anton Posch
Contents
Preface. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1 Mechanical/Physical Methods of Cell Distribution
and Tissue Homogenization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Stanley Goldberg
2 Sample Preservation Through Heat Stabilization of Proteins:
Principles and Examples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Mats Born
3 Isolating Peripheral Lymphocytes by Density Gradient Centrifugation
and Magnetic Cell Sorting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Frederic Brosseron, Katrin Marcus, and Caroline May
4 Investigating the Adipose Tissue Secretome: A Protocol
to Generate High-Quality Samples Appropriate for Comprehensive
Proteomic Profiling. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Simon Gddeke, Jorg Kotzka, and Stefan Lehr
5 Methods for Proteomics-Based Analysis of the Human Muscle Secretome
Using an In Vitro Exercise Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Mika Scheler, Martin Hrab de Angelis, Hadi Al-Hasani,
Hans-Ulrich Hring, Cora Weigert, and Stefan Lehr
6 Urinary Pellet Sample Preparation for Shotgun Proteomic Analysis
of Microbial Infection and HostPathogen Interactions . . . . . . . . . . . . . . . . .
Yanbao Yu and Rembert Pieper
7 A Protocol for the Parallel Isolation of Intact Mitochondria
from Rat Liver, Kidney, Heart, and Brain . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Sabine Schulz, Josef Lichtmannegger, Sabine Schmitt, Christin Leitzinger,
Carola Eberhagen, Claudia Einer, Julian Kerth, Michaela Aichler,
and Hans Zischka
8 Isolation of Mitochondria from Cultured Cells and Liver Tissue
Biopsies for Molecular and Biochemical Analyses. . . . . . . . . . . . . . . . . . . . . . .
Sabine Schmitt, Carola Eberhagen, Susanne Weber, Michaela Aichler,
and Hans Zischka
9 Dynamic Range Compression with ProteoMiner:
Principles and Examples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Lei Li
10 Qualitative and Quantitative Proteomic Analysis of Formalin-Fixed
Paraffin-Embedded (FFPE) Tissue . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Omid Azimzadeh, Michael J. Atkinson, and Soile Tapio
vii
v
xi
1
21
33
43
55
65
75
87
99
109
viii
Contents
11 Full-Length Protein Extraction Protocols and Gel-Based Downstream

Applications in Formalin-Fixed Tissue Proteomics . . . . . . . . . . . . . . . . . . . . . .
Alessandro Tanca, Sergio Uzzau, and Maria Filippa Addis
12 Enrichment of Low-Abundant Protein Targets by Immunoprecipitation
Upstream of Mass Spectrometry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Barbara Kaboord, Suzanne Smith, Bhavin Patel, and Scott Meier
13 Principles of Protein Labeling Techniques . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Christian Obermaier, Anja Griebel, and Reiner Westermeier
14 Isolation of Extracellular Vesicles for Proteomic Profiling . . . . . . . . . . . . . . . .
Dong-Sic Choi and Yong Song Gho
15 A Protocol for Exosome Isolation and Characterization: Evaluation
of Ultracentrifugation, Density-Gradient Separation, and Immunoaffinity
Capture Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
David W. Greening, Rong Xu, Hong Ji, Bow J. Tauro,
and Richard J. Simpson
16 Chloroplast Isolation and Affinity Chromatography for Enrichment
of Low-Abundant Proteins in Complex Proteomes . . . . . . . . . . . . . . . . . . . . .
Roman G. Bayer, Simon Stael, and Markus Teige
17 Depletion of RuBisCO Protein Using the Protamine Sulfate
Precipitation Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Ravi Gupta and Sun Tae Kim
18 Step-by-Step Preparation of Proteins for Mass Spectrometric Analysis . . . . . . .
Thomas Franz and Xinping Li
19 Identification of Protein N-Termini Using TMPP or Dimethyl Labeling
and Mass Spectrometry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Jingjing Deng, Guoan Zhang, Fang-Ke Huang, and Thomas A. Neubert
20 Optimization of Cell Lysis and Protein Digestion Protocols
for Protein Analysis by LC-MS/MS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Dominic Winter, Alireza Dehghani, and Hanno Steen
21 Comprehensive Protocol to Simultaneously Study Protein
Phosphorylation, Acetylation, and N-Linked Sialylated Glycosylation . . . . . . .
Marcella Nunes Melo-Braga, Mara Ibez-Vea, Martin Rssel Larsen,
and Katarzyna Kulej
22 Protein Profiling and Phosphoprotein Analysis by Isoelectric Focusing . . . . . .
Giuseppina Maccarrone and Michaela D. Filiou
23 Principles and Examples of Gel-Based Approaches
for Phosphoprotein Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Birgit Steinberger and Corina Mayrhofer
24 Neutral Phosphate-Affinity SDS-PAGE System for Profiling
of Protein Phosphorylation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Emiko Kinoshita-Kikuta, Eiji Kinoshita, and Tohru Koike
25 Enrichment and Identification of Bacterial Glycopeptides
by Mass Spectrometry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Nichollas E. Scott and Stuart J. Cordwell
117
135
153
167
179
211
225
235
249
259
275
293
305
323
355
Contents
26 In-Gel Peptide IEF Sample Preparation for LC/MS Analysis. . . . . . . . . . . . . .

Tom Berkelman, Sricharan Bandhakavi, and Aran Paulus
27 Western Blotting Using In-Gel Protein Labeling as a Normalization
Control: Stain-Free Technology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Jennifer E. Gilda and Aldrin V. Gomes
28 2-D Western Blotting for Evaluation of Antibodies Developed
for Detection of Host Cell Protein . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Tom Berkelman, Adriana Harbers, and Sricharan Bandhakavi
29 Free Flow Electrophoresis for Separation of Native Membrane
Protein Complexes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Lutz Andreas Eichacker, Gerhard Weber, Ute Sukop-Kppel,
and Robert Wildgruber
30 Three-Dimensional Electrophoresis for Quantitative Profiling
of Complex Proteomes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Sergio Mauro, Bertrand Colignon, Marc Dieu, Edouard Delaive,
and Martine Raes
31 A Bead-Based Multiplex Sandwich Immunoassay to Assess
the Abundance and Posttranslational Modification State of -Catenin . . . . . . .
Nicola Groll, Cornelia Sommersdorf, Thomas O. Joos, and Oliver Poetz
32 Identification of SUMO E3 Ligase-Specific Substrates Using the HuProt
Human Proteome Microarray . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Eric Cox, Ijeoma Uzoma, Catherine Guzzo, Jun Seop Jeong,
Michael Matunis, Seth Blackshaw, and Heng Zhu
33 Amyloid-Binding Proteins: Affinity-Based Separation,
Proteomic Identification, and Optical Biosensor Validation . . . . . . . . . . . . . . .
Alexei Medvedev, Olga Buneeva, Arthur Kopylov, Oksana Gnedenko,
Alexis Ivanov, Victor Zgoda, and Alexander A. Makarov
34 Proteomic Profiling by Nanomaterials-Based Matrix-Assisted
Laser Desorption/Ionization Mass Spectrometry for High-Resolution
Data and Novel Protein Information Directly from Biological Samples . . . . . .
Suresh Kumar Kailasa and Hui-Fen Wu
Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
ix
369
381
393
415
427
441
455
465
479
497
Contributors
MARIA FILIPPA ADDIS Porto Conte Ricerche, Tramariglio, Alghero(SS), Italy
MICHAELA AICHLER Research Unit Analytical PathologyInstitute of Pathology,
Helmholtz Center Munich, German Research Center for Environmental Health,
Neuherberg, Germany
HADI AL-HASANI Institute of Clinical Biochemistry and Pathobiochemistry,
German Diabetes Center, Duesseldorf, Germany; German Center for Diabetes
Research (DZD), Duesseldorf, Germany
MARTIN HRAB DE ANGELIS Institute of Experimental Genetics, Helmholtz Zentrum
Mnchen, German Research Center for Environmental Health, Neuherberg, Germany;
Center of Life and Food Sciences Weihenstephan, Technische Universitt Mnchen,
Freising-Weihenstephan, Germany; German Center for Diabetes
Research (DZD), Duesseldorf, Germany
MICHAEL J. ATKINSON Institute of Radiation Biology, Helmholtz Zentrum Mnchen,
German Research Center for Environmental Health, Neuherberg, Germany; Technical
University of Munich, Munich, Germany
OMID AZIMZADEH Institute of Radiation Biology, Helmholtz Zentrum Mnchen,
German Research Center for Environmental Health, Neuherberg, Germany
SRICHARAN BANDHAKAVI diaDexus, South San Francisco, CA, USA
ROMAN G. BAYER Department of Ecogenomics and Systems Biology, University of Vienna,
Vienna, Austria
TOM BERKELMAN Bio-Rad Laboratories, Hercules, CA, USA
SETH BLACKSHAW Solomon H. Snyder Department of Neuroscience, Johns Hopkins
University School of Medicine, Baltimore, MD, USA; Institute for Cell Engineering,
Johns Hopkins University School of Medicine, Baltimore, MD, USA; Department
of Ophthalmology, Johns Hopkins University School of Medicine, Baltimore, MD, USA;
Center for High-Throughput Biology, Johns Hopkins University School of Medicine,
Baltimore, MD, USA
MATS BORN Denator AB, Gothenburg, Sweden
FREDERIC BROSSERON Deutsches Zentrum fr Neurodegenerative Erkrankungen (DZNE) e.V.,
Bonn, Germany
OLGA BUNEEVA Institute of Biomedical Chemistry, Moscow, Russia
DONG-SIC CHOI Department of Life Sciences, Pohang University of Science and Technology,
Pohang, Republic of Korea
BERTRAND COLIGNON Dpartement Sciences du Vivant, Centre wallon de Recherches
agronomiques, Gembloux, Belgium; URBC-NARILIS, Universit de Namur,
Namur, Belgium
STUART J. CORDWELL School of Molecular Bioscience, The University of Sidney, Sidney,
Australia; Discipline of Pathology, School of Medical Sciences, The University of Sidney,
Sidney, Australia; Charles Perkins Centre, The University of Sidney, Sidney, Australia
xi
xii
Contributors
ERIC COX Biochemistry, Cellular and Molecular Biology Graduate Program, Johns
Hopkins University School of Medicine, Baltimore, MD, USA; Solomon H. Snyder
Department of Neuroscience, Johns Hopkins University School of Medicine, Baltimore,
MD, USA; Department of Pharmacology and Molecular Sciences, Johns Hopkins
University School of Medicine, Baltimore, MD, USA
ALIREZA DEHGHANI Institute for Biochemistry and Molecular Biology, University of Bonn,
Bonn, Germany
EDOUARD DELAIVE URBC-NARILIS, Universit de Namur, Namur, Belgium
JINGJING DENG Department of Biochemistry and Molecular Pharmacology, Kimmel
Center for Biology and Medicine at the Skirball Institute, New York University School
of Medicine, New York, NY, USA
MARC DIEU URBC-NARILIS, Universit de Namur, Namur, Belgium
CAROLA EBERHAGEN Institute of Molecular Toxicology and Pharmacology, Helmholtz Center
Munich, German Research Center for Environmental Health, Neuherberg, Germany
LUTZ ANDREAS EICHACKER Center of Organelle Research, University of Stavanger,
Stavanger, Norway
CLAUDIA EINER Institute of Molecular Toxicology and Pharmacology, Helmholtz Center
MICHAELA D. FILIOU Max Planck Institute of Psychiatry, Munich, Germany
THOMAS FRANZ Max Planck Institute for Biology of Ageing, Cologne, Germany
YONG SONG GHO Department of Life Sciences, Pohang University of Science
and Technology, Pohang, Republic of Korea
JENNIFER E. GILDA Department of Neurobiology, Physiology, and Behavior, University
of California, Davis, CA, USA
OKSANA GNEDENKO Institute of Biomedical Chemistry, Moscow, Russia
SIMON GDDEKE Institute of Clinical Biochemistry and Pathobiochemistry, German Diabetes
Center, Leibniz Center for Diabetes Research at Heinrich Heine University, Duesseldorf,
Germany; German Center for Diabetes Research (DZD), Duesseldorf, Germany
STANLEY GOLDBERG Glen Mills Inc., Clifton, NJ, USA
ALDRIN V. GOMES Department of Neurobiology, Physiology, and Behavior, University
of California, Davis, CA, USA; Department of Physiology and Membrane Biology,
University of California, Davis, CA, USA
DAVID W. GREENING Department of Biochemistry, La Trobe Institute for Molecular
Science, La Trobe University, Melbourne, Australia
ANJA GRIEBEL SERVA Electrophoresis GmbH, Heidelberg, Germany
NICOLA GROLL Department of Protein Analytics, NMI Natural and Medical Sciences
Institute at the University of Tuebingen, Reutlingen, Germany
RAVI GUPTA Department of Plant Bioscience, Pusan National University, Miryang,
Republic of Korea
CATHERINE GUZZO Department of Biochemistry and Molecular Biology, Bloomberg School
of Public Health, Johns Hopkins University, Baltimore, MD, USA
ADRIANA HARBERS Bio-Rad Laboratories, Hercules, CA, USA
HANS-ULRICH HRING Division of Endocrinology, Diabetology, Angiology, Nephrology,
Pathobiochemistry and Clinical Chemistry, Department of Internal Medicine, University
of Tbingen, Tbingen, Germany; Institute for Diabetes Research and Metabolic Diseases
of the Helmholtz Zentrum Mnchen at the University of Tbingen, Tbingen, Germany;
German Center for Diabetes Research (DZD), Duesseldorf, Germany
Contributors
xiii
FANG-KE HUANG Department of Biochemistry and Molecular Pharmacology, Kimmel

MARA IBEZ-VEA Department of Biochemistry and Molecular Biology, University
of Southern Denmark, Odense, Denmark
ALEXIS IVANOV Institute of Biomedical Chemistry, Moscow, Russia
JUN SEOP JEONG Department of Pharmacology and Molecular Sciences, Johns Hopkins
HONG JI Department of Biochemistry, La Trobe Institute for Molecular Science, La Trobe
University, Melbourne, Australia
THOMAS O. JOOS Department of Protein Analytics, NMI Natural and Medical Sciences
BARBARA KABOORD Thermo Fisher Scientific, Rockford, IL, USA
SURESH K. KAILASA Department of Applied Chemistry, S. V. National Institute
of Technology, Surat, India
JULIAN KERTH Institute of Molecular Toxicology and Pharmacology, Helmholtz Center
SUN TAE KIM Department of Plant Bioscience, Pusan National University, Miryang,
Republic of Korea
EIJI KINOSHITA Department of Functional Molecular Science, Graduate School
of Biomedical Sciences, Hiroshima University, Hiroshima, Japan
EMIKO KINOSHITA-KIKUTA Department of Functional Molecular Science, Institute
of Biomedical and Health Sciences, Hiroshima University, Hiroshima, Japan
TOHRU KOIKE Department of Functional Molecular Science, Institute of Biomedical
and Health Sciences, Hiroshima University, Hiroshima, Japan
ARTHUR KOPYLOV Institute of Biomedical Chemistry, Moscow, Russia
JOERG KOTZKA Institute of Clinical Biochemistry and Pathobiochemistry, German
Diabetes Center, Leibniz Center for Diabetes Research at Heinrich Heine University,
Duesseldorf, Germany; German Center for Diabetes Research (DZD), Duesseldorf,
Germany
KATARZYNA KULEJ Department of Biochemistry and Molecular Biology, University
MARTIN RSSEL LARSEN Department of Biochemistry and Molecular Biology, University
STEFAN LEHR German Center for Diabetes Research (DZD), Neuherberg, Germany;
Institute of Clinical Biochemistry and Pathobiochemistry, German Diabetes Center,
Duesseldorf, Germany
CHRISTIN LEITZINGER Institute of Molecular Toxicology and Pharmacology, Helmholtz
Center Munich, German Research Center for Environmental Health, Neuherberg,
Germany
XINPING LI Max Planck Institute for Biology of Ageing, Cologne, Germany
LEI LI Bio-Rad Laboratories, Hercules, CA, USA
JOSEF LICHTMANNEGGER Institute of Molecular Toxicology and Pharmacology, Helmholtz
Center Munich, German Research Center for Environmental Health, Neuherberg,
Germany
GIUSEPPINA MACCARRONE Max Planck Institute of Psychiatry, Munich, Germany
ALEXANDER A. MAKAROV Engelhardt Institute of Molecular Biology of Russian Academy of
Sciences, Moscow, Russia
xiv
Contributors
KATRIN MARCUS Medizinisches Proteom-Center, Ruhr-Universitt Bochum, Bochum, Germany

MICHAEL MATUNIS Department of Biochemistry and Molecular Biology, Bloomberg School
of Public Health, Johns Hopkins University, Baltimore, MD, USA
SERGIO MAURO Dpartement Sciences du Vivant, Centre wallon de Recherches
agronomiques, Gembloux, Belgium
CAROLINE MAY Medizinisches Proteom-Center, Ruhr-Universitt Bochum, Bochum, Germany
CORINA MAYRHOFER Institute of Animal Breeding and Genetics, University of Veterinary
Medicine, Vienna, Austria; Institute of Biotechnology in Animal Production,
Department for Agrobiotechnology, University of Natural Resources and Applied Life
Sciences Vienna, Tulln, Vienna, Austria
ALEXEI MEDVEDEV Department of Proteomic Research and Mass Spectrometry,
Institute of Biomedical Chemistry, Moscow, Russia
SCOTT MEIER Thermo Fisher Scientific, Rockford, IL, USA
MARCELLA N. MELO-BRAGA Department of Biochemistry and Molecular Biology,
University of Southern Denmark, Odense, Denmark
THOMAS A. NEUBERT Department of Biochemistry and Molecular Pharmacology, Kimmel
CHRISTIAN OBERMAIER SERVA Electrophoresis GmbH, Heidelberg, Germany
BHAVIN PATEL Thermo Fisher Scientific, Rockford, IL, USA
ARAN PAULUS Thermo Fisher Scientific, San Jose, CA, USA
REMBERT PIEPER The J Craig Venter Institute, Rockville, MD, USA
OLIVER POETZ Department of Protein Analytics, NMI Natural and Medical Sciences
MARTINE RAES URBC-NARILIS, Universit de Namur, Namur, Belgium
MIKA SCHELER Institute of Experimental Genetics, Helmholtz Zentrum Mnchen,
German Research Center for Environmental Health, Neuherberg, Germany;
German Center for Diabetes Research (DZD), Duesseldorf, Germany
SABINE SCHMITT Institute of Molecular Toxicology and Pharmacology, Helmholtz Center
SABINE SCHULZ Institute of Molecular Toxicology and Pharmacology, Helmholtz Center
NICHOLLAS E. SCOTT School of Molecular Bioscience, The University of Sidney, Sidney,
Australia; Centre for High-Throughput Biology, Department of Biochemistry
and Molecular Biology, The University of British Columbia, Vancouver, Canada
RICHARD J. SIMPSON Department of Biochemistry, La Trobe Institute for Molecular
Science, La Trobe University, Melbourne, Australia
SUZANNE SMITH Thermo Fisher Scientific, Rockford, IL, USA
CORNELIA SOMMERSDORF Department of Protein Analytics, NMI Natural and Medical
Sciences Institute at the University of Tuebingen, Reutlingen, Germany
SIMON STAEL Department of Ecogenomics and Systems Biology, University of Vienna, Vienna,
Austria; VIB Department of Plant Systems Biology, Gent University, Gent, Belgium
HANNO STEEN Department of Pathology, Boston Childrens Hospital and Harvard
Medical School, Boston, MA, USA
BIRGIT STEINBERGER Institute of Animal Breeding and Genetics, University of Veterinary
Medicine, Vienna, Austria; Department for Agrobiotechnology, Institute of Biotechnology
Contributors
in Animal Production, University of Natural Resources and Applied Life Sciences

Vienna, Tulln, Vienna, Austria
UTE SUKOP-KPPEL FFE Service GmbH, Feldkirchen, Germany
ALESSANDRO TANCA Porto Conte Ricerche, Tramariglio, Alghero (SS), Italy
SOILE TAPIO Institute of Radiation Biology, Helmholtz Zentrum Mnchen,
BOW J. TAURO Department of Biochemistry, La Trobe Institute for Molecular Science,
La Trobe University, Melbourne, Australia
MARKUS TEIGE Department of Ecogenomics and Systems Biology, University of Vienna,
Vienna, Austria; Department of Applied Genetics and Cell Biology, University
of Natural Resources and Life Sciences, Vienna, Austria
IJEOMA UZOMA Department of Pharmacology and Molecular Sciences, Johns Hopkins
SERGIO UZZAU Porto Conte Ricerche, Tramariglio, Alghero (SS), Italy
GERHARD WEBER FFE Service GmbH, Feldkirchen, Germany
SUSANNE WEBER Institute of Experimental Genetics, Helmholtz Center Munich,
CORA WEIGERT Division of Endocrinology, Diabetology, Angiology, Nephrology,
Pathobiochemistry and Clinical Chemistry, Department of Internal Medicine,
University of Tbingen, Tbingen, Germany; Institute for Diabetes Research and
Metabolic Diseases of the Helmholtz Zentrum Mnchen at the University of Tbingen,
Tbingen, Germany; German Center for Diabetes Research (DZD), Duesseldorf,
Germany
REINER WESTERMEIER SERVA Electrophoresis GmbH, Heidelberg, Germany
ROBERT WILDGRUBER FFE Service GmbH, Feldkirchen, Germany
DOMINIC WINTER Institute for Biochemistry and Molecular Biology, University of Bonn,
Bonn, Germany; Department of Pathology, Boston Childrens Hospital and Harvard
Medical School, Boston, MA, USA
HUI-FEN WU Department of Chemistry, Medical Sciences and Nanotechnology,
National Sun Yat-Sen University, Kaohsiung, Taiwan
RONG XU Department of Biochemistry, La Trobe Institute for Molecular Science,
La Trobe University, Melbourne, Australia
YANBAO YU The Craig Venter Institute, Rockville, MD, USA
VICTOR ZGODA Institute of Biomedical Chemistry, Moscow, Russia
GUOAN ZHANG Department of Biochemistry and Molecular Pharmacology, Kimmel
HENG ZHU Department of Pharmacology and Molecular Sciences, Johns Hopkins
University School of Medicine, Baltimore, MD, USA; Center for High-Throughput
Biology, Johns Hopkins University School of Medicine, Baltimore, MD, USA
HANS ZISCHKA Institute of Molecular Toxicology and Pharmacology, Helmholtz Center
xv
Chapter 1
Mechanical/Physical Methods of Cell Distribution
and Tissue Homogenization
Stanley Goldberg
Abstract
This chapter covers the various methods of Mechanical Cell Disruption and Tissue Homogenization that
are currently commercially available for processing minute samples (<1 mL) to larger production quantities.
These mechanical methods of lysing do not introduce chemicals or enzymes to the system. However,
the energies needed when using these harsh methods can be high and destroy the very proteins
being sought.
The destruction of cell membranes and walls by these harsh methods is effected by subjecting the
cells (1) to shearing by liquid flow, (2) to exploding by pressure differences between inside and outside of
cell, (3) to collision forces by impact of beads or paddles, or (4) a combination of these forces. Practical
suggestions to optimize each method, where to acquire such equipment, and links to reference sources
are included.
Key words Cell disruption, Bead mills, BioNeb cell disruption, Cell disruption vessel, Douce tissue
grinder, Dyno-Mill, French Press G-M, Gaulin high-pressure homogenizer, High-pressure homogenizers, Megatron, Microfluidics, Mixer-Mill, Mortar, Pestle, Nitrogen Parr vessel, Opposed jet
homogenization, Parr nitrogen vessel, Polytron, Potter-Elvehjem tissue grinders, Pressure vessel,
Sonicator, Sonitube, Tissue grinders, Tissue homogenization, Ultrasonic processor, Electro Water
Separation
Introduction
The need to release cell components without introducing encumbering
chemicals or enzymes suggests the use of mechanical methods of
lysing. The destruction of cell membranes and walls by these
harsh methods is effected by subjecting the cells (1) to shearing
by liquid flow, (2) to exploding by pressure differences between
inside and outside of cell, (3) to collision forces by impact of beads
or paddles, or (4) a combination of these forces.
Generally speaking, any of the techniques described here can,
to some degree, disrupt any cells or tissues. For more difficult
materials, just the increase of motivation force or time of exposure
will improve breakage. However, the use of excessive force is
Anton Posch (ed.), Proteomic Profiling: Methods and Protocols, Methods in Molecular Biology, vol. 1295,
DOI 10.1007/978-1-4939-2550-6_1, Springer Science+Business Media New York 2015
Stanley Goldberg
limited due to the generation of detrimental heat and/or shear

that can ruin the desired proteins. In addition, excess force will
accelerate wear and ultimately damage the equipment.
By judicious use of the equipment one can select from a gentle
nicking of the cell to release intact organelle up to a vigorous action
to release membrane bound proteins. Some methods are suitable
to handle tissues only, others for free cells only, and some are suitable
for both. Some techniques are capable of processing only small
quantities of material while others are limited to handling larger
amounts.
Tissues that are difficult to break down include heart muscle,
lung, intestine, and skin. On the other hand, some fragile mammalian
cells can be broken by just a moderate shaking of the suspended
cells. Free cells that are difficult to process include those that are
extremely small size (below 0.25 m) bacteria, and the tough
yeasts and spores. Plant materials and seeds will need higher energy
inputs for proper maceration. Table 1 provides an overview of the
Table 1
Methods overview, trade names, websites
Technique
Trade name(s)
Websites
Bead Impactshaking vessel
Mixer Mill [1]

Mini Bead Beater [2]
www.RETSCH.de
www.BIOSPEC.com
Bead Impactagitator shaft
DYNO-MILL [3]
www.WAB.ch
www.GLENMILLS.com
Rotor/statorshear
by spinning shaft
Polytron [4]
www.KINEMATICA.ch
Mortar/pestleshear by
mechanical pressure
Potter-Elvehjem tissue
grinders [5]
www.WHEATON.com
High-pressure batchliquid
expansion
French Press G-M [6]
www.GLENMILLS.com
High-pressure batchgas
expansion
Parr vessel [7]
www.PARRINST.com
High-pressure flowhigh
velocity liquid shear
APV Gaulin [8]
www.SPX.com
High-pressureopposed
liquid streams
Microfluidizer [9]
www.MICROFLUIDICSCORP.com
Dropletlow pressure
flow droplet nebulizing
BioNeb [10]
www.GLASCOL.com
Ultrasonicshear by
collapsing bubbles
Sonicator [11]
Sonitube
www.SONICATOR.com
www.GLENMILLS.com
Electromotive force
Electro Water
Separation [12]
www.ORIGINOIL.com
Mechanical/Physical Cell Disruption Methods
Table 2
Suitable subjects and capacity for each method
Bacteria
Yeast,
Algae,
Fungus,
Spores
Seeds
Plants
Tissues
Capacity
Bead impact shaking vessel
S/M
Bead impact agitator shaft
M/L
Rotor/stator shear by spinning shaft
S/M/L
Mortar/pestle shear
by mechanical pressure
S/M
High-pressure batch liquid expansion
S/M
High-pressure batch gas expansion
S/M
High-pressure flow high velocity shear
S/M/L
High-pressure flow opposed

liquid streams
S/M/L
Droplet low pressure droplet

nebulizing
S/M
Ultrasonic shear collapsing bubbles
S/M/L
Electro water separation
M/L
Type of biomaterial to be lysed

Technique
Suitability: Ygeneral good practice; Nnot recommended; ?not known or marginal success
Quantities: S = small 0.125 mL; M = medium 10500 mL; L = 250 mL to many liters
mechanical methods with at least one example of commercial

equipment and related websites while Table 2 describes suitable
subjects and capacity for each method.
2
2.1
Bead Impact Methods: Shaking Vessel

Theory
All bead devices open the cells or homogenize tissues by throwing

the beads (also called grinding media) against the cells/tissue.
Also the accelerated beads generate strong shear in the liquid buffer surrounding the cells/tissues, which also pulls then apart. Two
methods to accelerate the grinding media (beads) are (1) by shaking the entire container or (2) by a spinning agitator within a
container (see next section). The shaking container method is
usable for tissues as well as free cells. For extremely small samples
of 0.2 mL to somewhat larger quantities of 50 mL, the shaking of
the vessel is the method of choice. The motion can be of differing
geometries depending upon what equipment is selected. Shaking
can only be done in batch operation thus limiting the amount of
Stanley Goldberg
Fig. 1 Bead impactshaken vesselMini Bead Beater-1 equipment
Fig. 2 Bead impactshaken vesselinside viewbeads moving in jar
materials than can be processed. The equipment is very low cost,

durable, and simple to operate requiring minimal training.
Materials needed to operate any of these shaking include the
cells/tissue, grinding media (beads), liquid phase such as buffer,
the container, and the equipment. Variables include the bead
selection (density, diameter, and quantity), speed of agitation, cell
concentration, and duration of run (Figs. 1 and 2).
2.2 Practical
Aspects
Denaturing of proteins due to high temperature or excessive

shear is to be considered in all mechanical disruption/homogenization equipments. Also, the wear of the grinding media
and container into the samples needs to be evaluated.
Hints for successful temperature control include: (1) Prechilling of samples and containers; (2) Runs of short duration
with rest time to allow for re-chilling samples on ice; (3) Use
of fewer beads and/or extra buffer to act a heat sink; (4)
Reduced degree of shaking vigor.
3
3.1
Detrimental contamination of the batch due to the wear of

either the grinding media or container walls is usually rare due
to the insignificant amounts involved. A hint to mitigate any
contamination problem is to use inert materials of construction such as using beads and containers of zirconium oxide
stabilized with yttria (95 %/5 %; specific gravity 6.0).
Beads size: For small diameter cells (e.g. bacteria) use beads of
0.100.5 mm diameter. For larger cells (e.g. yeast, algae,
hyphae) use beads of 0.51.25 mm. Glass (sp. gr. 2.5) is a
good starting material due to low cost. For homogenizing
plant or animal tissues that have been previously chopped with
a razor, beads of 1.05.0 mm diameter are used.
Bead density: If additional energy is needed to improve breakage/homogenization of tough cells, then use higher density
materials. These material types include ceramic (zirconium
oxide family) with specific gravities from 3.8 to 6.0, stainless
steel of sp. gr. 7.0+, and tungsten carbide of sp. gr. 14.2+.
Also, the use of larger diameter beads from 2 to 20 mm can
improve breakage. Recently, success has been reported with
SiC grit with its sharp edges, and with stainless steel ballcones
that have a wedged edge at the equator.
Time savings can be achieved by ganging several samples into

96-well titer plates rather then running one sample at a time.
The larger models can accommodate these plates.
A simple way to evaluate the suitability of bead shaking can be

done as follows. In a test tube place some glass beads and buffered cells/tissues. Hold the tube against a vortex shaker for
13 min. If breakage/homogenization is realized, then bead
shaking has promise. Switching to suitable equipment as
described above will reduce repetitive strain to the technician
holding the test tubes.
Bead Impact Methods: Stirred Agitated Beads

Theory
For modest sample quantities of 50 mL and scaling up to industrial amounts of several thousand liters, the agitation of the beads
by a turning agitator within the vessel is the method of choice. In
this class of agitated bead mills, the beads and the cell suspension
are loaded into a chamber. Into the mix is placed one or more
spinning discs that accelerate the beads. The beads striking the
cells combined with the shearing by the moving liquid phase will
disrupt the cells.
Stanley Goldberg
Fig. 3 Bead impactagitated beadsDyno-Mill with 600 mL chamber

equipment
These units are normally used for disruption of free microorganism cells (bacteria, yeasts, hyphae, mycelia) and not for tissue
samples. For lesser quantities, see previous section on shaking container method. Materials needed to operate the agitated bead mills
include the cells/tissue, grinding media (beads), liquid phase such
as buffer, cooling ice or jacket fluids, and the mill equipment.
Variables include the bead selection (density, diameter, and quantity), speed of agitation, cell concentration, and duration of run
(Figs. 3 and 4).
3.2 Practical
Aspects

shear is to be considered in all mechanical disruption/homogenization equipments. Also, the wear of the grinding media
and container into the samples needs to be evaluated.
Hints for successful temperature control include: (1) Pre-chilling

of samples to below 5 C. (2) Reduce residence time by
increased feed rates and then do a second pass with inter-stage
cooling. (3) Slower tip speed of Agitator discs to 6 m/s. (4)
Lower the concentration of cells if viscosity increases contribute
to excessive heating.
Detrimental contamination of the batch due to the wear of

either the grinding media or container walls is usually rare due
to the insignificant amounts involved. A hint to mitigate any
contamination problem is to use inert materials of construction. For example, when iron would be harmful, do not use
steel beads, switch to glass or ceramics. The least wear is reportedly
seen when using ceramic beads and containers that are fabricated from zirconium oxide stabilized with yttrium (95 %/5 %;
specific gravity 6.0).
Fig. 4 Bead impactagitated stirred beadsinside viewspinning agitators

moving beads against suspended cells. (1) Input, (2) Output, (3) Agitator discs, (4)
Beads, (5) Cooling jacket, (6) Bead screen
4
4.1
For small diameter cells (e.g. E. coli of 0.251.0 m) use beads

of 0.100.5 mm diameter. For larger cells (e.g. yeast, algae,
hyphae) use beads of 0.51.25 mm. Glass (sp. gr. 2.5) is a
good starting material due to low cost. If addition energy is
needed to improve breakage/homogenization, then switch to
ceramic (zirconium oxide family) with specific gravities from
3.8 to 6.0.
A quick preliminary test can be run with standard lab equipment. Load a beaker with the buffered cells/tissue along with
some beads. Place on a magnetic stirrer (or use an overhead
stirrer) and spin the beads. If some breakage is seen then the
bead mill is a good candidate for processing the cells in
question.
RotorStator Homogenizer
Theory
Rotor/Stator homogenizers consist of a rapidly spinning paddle

contained within an open-ended tube with slots near the working
end. The turning paddle pushes liquid out the slots, creating a lowpressure region that draws fresh suspension up from the open end.
As the tissue is pulled up, there is a stretching action. Then, as the
Stanley Goldberg
Fig. 5 RotorstatorShear by spinning shaftPolytron equipment
material is forced between the narrow gaps, there is a cutting

action. The working part of the equipment that contacts the
samples is called the generator.
The Rotor/Stator design consists of two or more coaxial interlocking rows of teeth. The internal rotor(s) is driven with a motor
running at speeds between 3,000 rpm for the larger industrial units
to 27,000 rpm for the smaller lab units. The size of the motor is
chosen based on the size of the equipment and application. Rotor
tip speeds of up to 50 m/s can be achieved. Usually the gap
between rotor and stator is around 300500 m (Figs. 5 and 6).
4.2 Practical
Aspects
Choice of the rotor/stator geometry depending upon the

application. Some examples of specialized generators include
extended knives to cut into a larger tissue sample, low-foaming
generators, and easy-cleaning units that may be autoclaved.
Next choose the correct size generator for the sample volume
to be homogenized. Factory tables provide the volume ranges
suitable for each generator thus selecting the correct parts is
quite easy. For example, a sample of size from 0.5 mL to 5 mL
Fig. 6 Rotorstatorshear by spinning shaftinside view
would be best processed with a generator of 5 mm diameter.

As the quantity of material to be processed at one time
increases, larger sizes of generators and motors are needed.
Interchangeable generators are designed to fit onto only certain motors. Therefore, one must decide which motor is the
most suitable for the particular range of applications and then
be sure to select only those generators designed to fit onto
that motor.
The homogenizers currently available can process sample volumes anywhere from less than 0.5 mL up to 150,000 L. Batch
equipment (e.g. Polytron) is used for small sample quantities
from less than 0.5 mL though larger units can handle several
hundred liters. For larger industrial in-line system for either
continuous or re-circulation flow, there are single-, double-, or
even triple-stage rotor/stator configurations (Megatron).
Mortar and Pestle Tissue Grinders: Shear by Mechanical Pressure

This class of simple devices disrupts cells and homogenizes tissues
by the pressure and friction generated when a moving pestle
pinches the samples against the wall of the mortar. Though usually
manual, there are electrically driven units available. The previous
equipment class of Rotor/Stator equipment is an alternative design
of this technique. Selection of proper mortar depends upon samples to be processed. Materials of construction are of glass, stainless
steel, Teflon, and plastics. Some units available include: Wheaton
Potter-Elvehjem Tissue Grinders 2 mL samples and larger.
One practical aspect to use these devices for bacteria is to freeze
the sample with liquid nitrogen.
10
6
6.1
Stanley Goldberg
High-Pressure Batch: Expanding Fluids

Theory
There are two widely used cell disruption methods that employ
rapidly expanding fluids from within the cell to explode the cell
membranes. The French Press G-M uses liquid under pressure,
and the Parr cell disruption vessel uses compressed gases. Since
these are bath operations, they are only suitable for small quantities
of less than about a liter.
6.1.1 The French

Press G-M
The French Press G-M design consists of a stainless steel cylinder

(called Pressure Cell) fitted with an exit valve at one end and a
piston/plunger at the other end. Up to 35 mL of suspend-free
cells in buffer are loaded into the pressure cell. With the exit valve
closed the piston is pressed against the liquid by a hydraulic press,
called the French Press G-M. Once a suitable pressure (up to
40 kpsi) is achieved throughout the liquid and within the cell body,
then the outlet valve is opened to allow the cell suspension to drip
out at a slow rate of about 1 mL/min (920 drops/min). This
exposure to atmospheric pressure, being much lower than the
pressure that was forced within the microorganisms body causes
the liquid to rush out, and thereby rupturing the cell membrane.
Normally, bacteria (at 40 kpsi) and yeast (at 20 kpsi) are handled in
the French Press G-M, not tissues, plants, nor seeds (Fig. 7).
6.1.2 The Parr Cell

Disruption Vessel
The Parr cell disruption vessel is a pressure vessel into which the
sample to be disrupted is placed along with a dip tube fitted with
an exit valve. Suitable gas such as nitrogen at 2 kpsi is forced into
the vessel and dissolved into the cells. When the exit valve is opened
the gas pressure is suddenly released causing the nitrogen to come
out of the solution within the cells as expanding bubbles. This
action stretches the membranes of each cell until they rupture and
releases the contents of the cell. Although sometimes referred to as
explosive decompression, nitrogen decompression is actually a
gentle method.
This method is suited for treating mammalian and other
membrane-bound cells, for treating plant cells, for releasing virus
from fertilized eggs, and for treating fragile bacteria. It is not recommended for untreated bacterial cells, unless using various
pretreatment procedures to weaken the cell wall. Yeast, fungus,
spores, and other materials with tough walls do not respond well
to this method (Figs. 8 and 9).
6.2 Practical
Aspects
Keeping the samples cold is achieved by pre-chilling of the

pressure cell. After the cells exit the valve, immediately cool by
keeping the collection beaker on ice. There are no provisions
to chill during the disruption process.
Removal of air prior to closing the exit valve will minimize the
amount of oxygen degradation of the released proteins.
6.2.1 French Press G-M
11
Fig. 7 High-pressure batchLiquid expansion French Press G-M and pressure

cells (35 mL and 3.7 mL) equipment
Fig. 8 High-pressure batchgas expansionParr vessel equipment
12
Stanley Goldberg
Fig. 9 High-pressure batchgas expansioninside view
6.2.2 Parr Cell

Disruption Vessel
During the paced release of materials, the pressure in the

Pressure Cell will drop. This needs to be balanced by periodically re-pressurizing the system by running the hydraulic press.
Individual cells such as lymphocytes, leukocytes, tissue culture

cells, or very fragile bacterial cells will not require pretreatment. Tissues must usually be pre-minced to ensure that they
not plug the exit dip tube and discharge valve.
The intended use of a homogenate generally determines the

composition of the suspending medium. Isotonic solutions are
commonly used. Solutions with higher concentrations will
tend to stabilize the nucleus and organelles. Conversely, very
dilute solutions will pre-stretch the cells by osmotic pressure
and will render them more susceptible to disruption by the
Vessel method.
Very small quantities of calcium chloride, magnesium acetate,

or magnesium chloride added to the suspending medium will
stabilize the nuclei when differential rupture is desired. Ratios
of approximately 10 mL of suspending medium to 1 g of wet
cells are commonly used to prepare the cell suspension.
Small sample quantities can be held inside of a smaller test tube or

beaker placed in the vessel. The inner container should be
approximately twice the volume of the suspension to be treated,
and the dip tube adjusted to reach the bottom of the container.
7
7.1
13
To cool a small inner vessel, it can be floated on ice water

within the vessel. The dip tube will hold it in place. Alternatively,
the entire external unit may be chilled.
Degree of disruption can be controlled by the amount of gas

pressure introduced. The greater the pressure, the more
homogenization. The use of moderate pressures will reduce
the disruptive forces and thus leave nuclei, active mitochondria, and other organelles intact.
High-Pressure Flow: Shear Through a Valve or Tube

Theory
In high-pressure homogenizers, the liquid stream of suspended

cells is forced at high pressure down a narrow channel or across the
small gap of a valve. This accelerates its speed, thereby stretching
and shearing cells. In some designs, the moving stream is subsequently and abruptly impacted against an obstacle to further damage the cells membrane. Two versions of impact are (1) where the
stream is directed to slam against an impingement wall (trade name
Gaulin), and (2) where the stream is split into two legs of a Y,
and these lines are then directed at one another in an interaction
chamber where they collide, further disrupting the cells (trade
name Microfluidics). These devices are used for suspended-free
cells, not for tissue samples.
7.1.1 High-Pressure
Valve with Impingement
Wall: Gaulin
The construction of the high-pressure homogenizer consists of a

positive displacement pump, a homogenizing valve, and sometimes
an impingement wall or ring. Though pumps may consist of one,
two, three, or five plungers, most smaller laboratory homogenizers
and those for cell breakage have only one plunger.
Attached to the pump is a homogenizing valve assembly that
may consist of one or two stages. For cell disruption a single-stage
valve is needed, typically consisting of three parts: a seat (bottom
part), a valve (top part), and an impact (wear) wall or ring.
By adjusting the gap or clearance between the valve and seat,
the flow area in the homogenizing valve is controlled. When the
flow area is reduced, pressure within the pump discharge manifold
increases. When the flow area is increased, the pressure is reduced.
The high pressure generated by the pump is converted to fluid
velocity and heat as the fluid is discharged from the restricted area
in the homogenizing valve. For cell disruption, the microorganisms are disrupted due to various mechanisms associated with the
fluid velocity. At 100 MPa, the fluid velocity can be as high as
450 m/s (Figs. 10 and 11).
7.1.2 High-Pressure
Flow Narrow Tubes
or Opposed Jets:
Microfluidics
These processors disrupt cells by applying a combination of shear

and impact forces onto the cells. The media that contains the cells
is forced through the narrow channels of the proprietary interaction chamber of the processor. Inside these channels, with typical
14
Stanley Goldberg
Fig. 10 High-pressure flowliquid shearSPX equipment
Fig. 11 High-pressure flowliquid shearinside view
dimensions of 75300 m, the fluid achieves velocities up to

400 m/s. High pressures (up to 275 MPa) are required to generate the high velocities inside the interaction chamber. The resulting shear rates can be up to 10,000,000/s and are the highest
commercially available. An alternative configuration splits the
15
Fig. 12 High-pressure flow-opposed liquid streamsMicrofluidizer inside view
stream of pressurized cell suspension fluid into two legs. These are
then directed at one another in an interaction chamber where they
collide, disrupting the cells.
This equipment is suitable for a variety of free cells (bacteria,
yeasts, mammalian cells), but not seeds, tissue samples, nor plant
materials. The design of identical fluid channels in both laboratoryscale and large-scale units allows for direct scale-up from the smallest laboratory unit (14 mL batch) directly to large production
units (tens of L/min) (Fig. 12).
7.1.3 Practical Aspects
Using High-Pressure Valve
with Impingement Wall:
Gaulin
To process a small sample size, first a liquid compatible with the

continuous phase of the slurry is added to the feed hopper of
the homogenizer. The machine is started and the pressure is set.
When the liquid level reaches the very bottom of the feed hopper, the cell slurry can be added quickly. The cell slurry will
push the liquid ahead of it. When the slurry is observed in the
discharge, a sample can be taken. In this way, limited sample is
not lost while waiting for pressure to rise to operational levels.
If the product is very sensitive to heat, then it may be necessary

to cool the equipment prior to introducing cell suspension and
to cool the suspension immediately after it discharges from the
homogenizer. A cooling coil is connected to the discharge
tube of the homogenizer and is immersed in an ice water bath.
Since the homogenizer is a positive displacement pump, there
must be no valves or restrictions in this discharge line that
could potentially shut off flow. Rapid chilling will minimize
16
Stanley Goldberg
losses due to the temperature rise of 2.5 C per 10 MPa of

pressure generated during the nearly adiabatic heating during
pressurization.
7.1.4 Practical Aspects:

High-Pressure Narrow
Tubes/Opposed Jets
To improve breakage, higher pressures are used up to the

pumps limits. High pressure can shorten valve life.
The material should be fully thawed before processing; ice may

plug the chambers.
Multiple passes decrease the particle artifacts size. Purification

methods to be used after cell disruption may determine the
desired particle size.
Optimization with respect to process pressure and number of

passes may be needed to increase the yield and facilitate subsequent purification steps.
Most models are autoclavable and air driven, though some are
electric-hydraulic systems. In many circumstances, especially when
samples are processed multiple times, the Microfluidizer processors require sample cooling before and/or after processing.
Low Pressure: Shear by Droplets Impingement

Invented at Indiana University, the BioNeb disruption system disrupts cells by a low-pressure droplet method. In the process of
droplet formation, large molecules or cells suspended in the liquid
being nebulized are forcefully distributed from the liquid into the
forming secondary droplet. This creates a transient laminar flow in
the microcapillary nebulization channel formed between the
surface of the liquid and the forming secondary droplet. The laminar flow in the capillary channel exerts sufficient shearing forces to
break cells. The shearing force created depends on the gas pressure
applied (10250 psi), the type of gas (nitrogen, argon, etc.) and
the viscosity of the liquid. By varying these parameters, it is possible to precisely regulate the magnitude of the force applied during
nebulization (Fig. 13).
9
9.1
Ultrasonic Processors: Shear by Collapsing Bubbles

Theory
The use of sound waves in fluids can disrupt cells. The operation
starts with normal electrical current (50 Hz or 60 Hz) being transformed to 20,000 Hz. This electrical signal is fed to a piezo-electric
crystal causing it to oscillation at this high frequency. The vibrations move a titanium metal HORN about 515 m. The shape of
the horn amplifies this motion to 100150 m/cycle. By placing
the horns endthe tipinto fluid, the tip moves the liquid
17
Fig. 13 Droplet low pressure nebulizerBioNeb equipment
forward (away) and then retracts (back) quicker than the liquid can
return. During the return stroke, the pressure in the system drops
below the vapor pressure of the liquid so boiling occurs (cavitation). As the liquid flows back, the bubbles collapse. This bubble
collapsing imparts the energy needed to disrupt the cells (Fig. 14).
9.2 Practical
Aspects

shear is often noted. The equipment has ONOFFON cycle
adjustments available. During the OFF periods, the samples can
cool. Lowered amplitude settings may reduce protein damage.
Hint to improve bubble formation is to keep the system as cold

as possible.
Time savings can be achieved by ganging several samples into

96-well titer plates rather then running one sample at a time.
Multi models can accommodate these plates. Also, some models have several tips fitted on a single horn that can handle
several samples simultaneously.
Larger quantities up to 100 L/h flow rates can be processed in

the Sonitube. Here the entire pipe for 15 in. (35 cm) oscillates
and ultrasonically processes all fluids pumped through.
18
Stanley Goldberg
Fig. 14 Ultrasonicshear by collapsing bubblesSonicator equipment
10
10.1
Extraction Across an Electromotive Field

Theory
10.2 Practical
Applications
Extracting non-polar lipids from microalgae are achieved using a

lipid extraction device having an anode and a cathode that forms a
channel and defines a fluid flow path through which an aqueous
slurry is passed. An electromotive force is applied across the channel at a gap distance in a range from 0.5 mm to 200 mm to cause
the non-polar lipids to be released from the algae cells (Fig. 15).
The non-polar lipids can be extracted at a high-throughput

rate and with low concentrations of polar lipids such as
phospholipids and chlorophyll.
Used to concentrate algae and other microorganisms from

aqueous systems for harvesting and concentrating cell masses.
Still experimental for cell lysing or product extracting.
Clearing water streams.
Acknowledgements
Superb assistance was rendered by the following people who are
most conversant in their noted equipment areas. Bead Milling
by Tim Hopkins (Biospec Products) and Harald Frommherz
19
Electro Water SeparationTM

The OriginOil Two-Stage Process
Electro-Flotation
Concentrate
Electrical Pulses
Contaminated
Water
Clear
Water
Electro-Coagulation
Electrical Pulses
Fig. 15 Electro Water Separation
(W. A. Bachofen AG); Rotor-Stator by Roger Munsinger

(Kinematica USA/AG); High pressure flowing liquid by Toni
Parker (SPX) and Steven Mesite (Microfluidics/IDEX); High
pressure batch liquid by Robert Borgon (University of Central
Florida); High pressure batch gas by Kay Frere and Lisa Randolph
(Parr Instruments); Low pressure flowing gas by James Jacso
(Glas-Col); and Ultrasonic processors by Andrea Coppola and
Marc Lusting (QSonica); Electro Water Separation by Devin
Angus and Jean-Louis Kindle (Origin Oil).
My book editor Dr. Anthon Posch has been exceptionally supportive in both technical as well as motivational matters. My wife
Robin and sons Evan and Adam are the sparkle of my life.
References
1. Reinkemeier M, Rocken W, Leitzmann C
(1996) A rapid mechanical lysing procedure
for routine analysis of plasmids from lactobacilli, isolated from sourdoughs. Int J Food
Microbiol 29:93104
2. Rezwan M, Laneelle MA, Sander P, Daffe M
(2007) Breaking down the wall: fractionation
of mycobacteria. J Microbiol Methods 68(1):
3239
3. Jung J, Xing X, Matsumoto K (2001) Kinetic
analysis of disruption of excess activated sludge
by Dyno Mill and characteristics of protein
release for recovery of useful materials.
Biochem Eng J 8:17
4. Lizotte E, Tremblay A, Allen BG, Fiset C

(2005) Isolation and characterization of subcellular protein fractions from mouse heart.
Anal Biochem 345:4754
5. Lindeskog P, Haaparanta T, Norgard M,
Glaumann H, Hansson T, Gustafsson JA
(1986) Isolation of rat intestinal microsomes:
partial characterization of mucosal cytochrome
P-450. Arch Biochem Biophys 244:492501
6. Borgon RA, Verity N (2012) Quantitative biological methods. Pearson Learning Solutions,
Boston, MA
7. Autuori F, Brunk U, Peterson E, Dallner G
(1982) Fractionation of isolated liver cells after
20
Stanley Goldberg
disruption with a nitrogen vessel and sonication. J Cell Sci 57:113

8. Kelly WJ, Muske KR (2004) Optimal operation of high-pressure homogenization for
intracellular product recovery. Bioprocess
Biosyst Eng 27:2537
9. Mehlhorn I, Groth D, Stockel J, Moffat B,
Reilly D, Yansura D, Willett WS, Baldwin M,
Fletterick R, Cohen FE, Vandlen R, Henner
D, Prusiner SB (1996) High-level expression
and characterization of a purified 142-residue
polypeptide of the prion protein. Biochemistry
35:55285537
10. Vinatier J, Herzog E, Plamont MA, Wojcik

SM, Schmidt A, Brose N, Daviet L, El
Mestikawy S, Giros B (2006) Interaction
between the vesicular glutamate transporter
type 1 and endophilin A1, a protein essential
for endocytosis. J Neurochem 97:11111125
11. Jaki BU, Franzblau SG, Cho SH, Pauli GF
(2006) Development of an extraction method
for mycobacterial metabolome analysis. J Pharm
Biomed Anal 41:196200
12. Eckelberry N (2013) US patent application
publication, US 2013/0211113 A1, 15 Aug
2013
Chapter 2
Sample Preservation Through Heat Stabilization
of Proteins: Principles and Examples
Mats Born
Abstract
Due to post-sampling changes, caused by residual enzyme activity in the sample, levels of analytes can
change from their in vivo levels so that they no longer accurately reflect conditions in the living system.
The Stabilizor system accomplishes elimination of enzyme activity through heat-induced denaturation of
enzymes by permanently altering the 3D protein structure of the enzymes. Heat stabilization can be introduced in the workflow either directly after sampling, with the instrument just next to where the sample is
taken, or prior to sample homogenization and extraction, when samples are heat denatured directly from
a frozen state. Initially, heat stabilization was developed to enable mass spectrometric analysis of neuropeptides. Heat stabilization has since been further developed and applied to a range of samples and downstream protein analysis techniques such as western blot, 2D gels and phosphorylation analysis with LC-MS.
Key words Heat inactivation, Post-sampling change, Phosphorylation, Post-translational modification,
Brain, Neuropeptides
Introduction
Biosampling for subsequent analysis of molecular biomarkers is a
corner stone in modern medicine and biological research.
Measured levels of various analytes are the basis for diagnosis and
the overall understanding of biology. It is thus of uttermost
importance that measured levels accurately reflect actual in vivo
levels as close as possible. Due to post-sampling changes, caused
by residual enzyme activity in the sample, levels of analytes can
change from their in vivo levels so that they no longer accurately
reflect conditions in the living system. In the living organism, levels of analytes are tightly controlled by cellular signaling and
homeostasis is maintained. When a sample is removed from its
natural environment, a dramatic signaling cascade is initiated. In the
case of a tissue biopsy the removal results in loss of blood flow, which
subsequently leads to low oxygen levels and a switch to anaerobic
metabolism. This results in low energy levels and a drop in sample pH.
21
22
Mats Born
All these changes, low oxygen, low ATP and low energy are signals
which are sensed on a cellular and protein level and results in an
active response from the still living cells in the sample. This results
in changes in protein levels as well as post-translational modifications (PTMs). This is a very fast process which initiates within
seconds of sampling and leads to measurable changes [1]. In addition
to the initial response, primarily an active regulated response,
which takes place from the moment of sampling until stabilization,
by freezing or heat inactivation, there is a second response, mainly
reactive in nature, which occurs during sample preparation due
to residual enzyme activity in the homogenate [2]. During this
phase, enzymes released from their cellular confinement and controls are free to interact with whatever substrates they encounter.
Both phases of enzyme driven change post-sampling must be
adequately addressed and minimized in order to successfully
measure in-vivo relevant levels of analytes.
The standard way of addressing the problem of enzymatically
driven post-sampling changes have been to add chemical inhibitors
and work fast while keep the samples cool during homogenization and
extraction. This is only partly successful as enzyme inhibitors are
selective and reduce enzyme activity rather than eliminate it [3].
A few years ago, an alternative approach, to preserving sample
quality in protein studies, was introduced, the Stabilizor system [3].
The Stabilizor system accomplishes elimination of enzyme activity
through heat-induced denaturation of enzymes by permanently
altering the 3D protein structure of the enzymes. In contrast to
chemical inhibitors, heat-induced enzyme denaturation is a general
principle affecting all protein-modifying enzymes, virtually reducing their activity to zero in the stabilized sample. Heat stabilization
can be introduced in the workflow either directly after sampling
with the instrument present just next to where the sample is taken
or just prior to sample homogenization and extraction when samples are heat denatured directly from a frozen state.
Heat stabilization has been applied to a range of samples and
downstream analysis techniques. It has proven itself as a valuable
tool in sample preparation for protein analysis. Initially, heat stabilization was developed to enable mass spectrometric analysis of
neuropeptides. The applicability has since been expanded and it
has been advantageously applied to a diverse array of protein analysis
techniques including western blot with phospho-specific antibodies,
top-down 2D PAGE, bottom-up phospho-shot gun, and MALDI
imaging [4]. As the Stabilizor system induces protein denaturation
and loss of protein 3D structure it is not suitable for sample preparation prior to analysis of enzyme activity, multi-protein complexes
or formalin-fixed epitopes, e.g. immuno-histochemistry. Using
western blots, stabilization of protein phosphorylation levels has
been shown. Flash-frozen and heat-stabilized samples either frozen
directly or left at room temperature for 30 min were compared.
Heat Stabilization of Proteins
23
In heat-stabilized samples detected levels of phospho-proteins were

more or less unaffected by 30 min at room temperature, while the
non-treated samples left at room temperature showed markedly
lower phosphorylation levels compared to directly flash-frozen or
heat-stabilized, see Fig. 1 [5]. In a different study, a direct comparative analysis of heat-stabilized versus flash-frozen samples was
done using top-down 2D PAGE analysis of proteins. In that study,
a majority of the low molecular spots differentially detected had
a higher intensity in the flash-frozen samples, whereas for the differentially detected high molecular weight proteins a majority was
detected with higher levels in the stabilized samples. In many cases,
the low molecular weight proteins detected with higher intensity
in flash-frozen samples was identified as having a higher theoretical mass as compared to their gel position indicating protein
degradation by protein cleavage and degradation, see Fig. 2 [6].
In another application of the technology using Pro-Q Diamond
staining of phosphorylated proteins separated using 2D PAGE, clear
Fig. 1 Quantitation of phospho-proteins in hippocampal lysates shows that heat-stabilized levels remain
unchanged while frozen tissue levels are decreased with 30 min room temperature incubation. Signals from
replicate Western blots were normalized to actin. Histograms represent average values for each group of
samples relative to 100 % for frozen tissues at 0 min RT (white bars); black bars, stabilized tissues 0 min at
RT; grey bars, frozen tissues 30 min at RT; striped bars, stabilized tissues 30 min at RT. Errors bars are the
standard error of the mean. Statistical significance was determined by the unpaired Students t-test. *p < 0.05;
**p < 0.001; ***p < 0.0001. Reprinted from J. Neurosci. Methods, 196(1), 99106. Ahmed, MM., and Gardiner,
KJ. Preserving protein profiles in tissue samples: differing outcomes with and without heat stabilization (2011),
with permission from Elsevier
24
Mats Born
Fig. 2 2-DE analysis of differentially expressed proteins from stabilized or snap-frozen mouse cortex. Spots
with differential intensity between the two samples are highlighted in color: higher intensity spots in heatstabilized samples (green), higher intensity spots in snap-frozen samples (red and blue). Numbers refer to
protein spots that were subjected to MS analysis. Reprinted from Proteomics, 9(19), 443344. Robinson, AA.,
Westbrook, JA., English, JA., et al., Assessing the use of thermal treatment to preserve the intact proteomes of
post-mortem heart and brain tissue. (2009), with permission from Wiley
differences can be seen between the flash-frozen and the stabilized

sample indicating shifts between phosphorylation species of the
same proteins, see Fig. 3 [7]. Large-scale analysis of tyrosine phosphorylations using antibody enrichment and LC-MS/MS analysis
have been done to compare phosphorylation levels in heatstabilized and flash-frozen samples. A clear trend toward higher
phosphorylations levels in heat-stabilized samples and consequently lower levels in flash-frozen samples was detected. This is
consistent with previous results showing a general decrease in most
phosphorylations post-sampling due to residual phosphatase activity both prior to freezing and during sample preparation, see Fig. 4.
Materials
2.1 The
Stabilizor System
The Stabilizor system consists of the Stabilizor T1 instrument and

Maintainor sample cards used to hold the sample during treatment and subsequent storage.
1. The Stabilizor T1 instrument is a bench-top instrument which
at its core holds a conductive heating unit. Besides ordinary
house hold electricity the instrument does not require other
inputs or special considerations.
25
Fig. 3 Phosphorylation states of murine PPIA isoforms in (a) stabilized and (b) unstabilized homogenates and
ATP synthase subunit a in stabilized (c) and unstabilized (d) homogenates, indicate a shift in phosphorylation
levels between isoforms between stabilized (a & c) and unstabilized (b & d) samples. In the unstabilized
sample (d), progressive phosphorylation of ATP synthase subunit a (21) resulted in a 550 % increase in the
higher phosphorylation state (40). The sum of mean spot volumes (62, 66, 79) were nearly identical in stabilized and unstabilized samples, respectively, as well as the sum of the mean volumes for spots 40, 53, and 21
were nearly identical in stabilized and unstabilized samples indicating that there were no preferential loss of
protein from either sample. Experimental pI values are provided on the abscissa. Reprinted from Electrophoresis,
32(16), 22062215, Smejkal, GB., Rivas-Morello, C., Chang, JH., et al., Thermal stabilization of tissues and the
preservation of protein phosphorylation states for two-dimensional gel electrophoresis (2011) with permission
from Wiley
2. The Maintainor Tissue sample card has a clam-shell design with

dual rigid plastic frames with thin Teflon-based plastic films.
The sample is enclosed between the plastic films and the card is
closed airtight prior to use in the Stabilizor instrument.
2.2 Solutes
for Resolubilization
of Heat-Denatured
Samples
Heat denaturation of proteins cause permanent alteration of the

3D fold of proteins, which generally results in low solubility in
water-based buffers. Efficient resolubilization of denatured proteins requires denaturing buffers. For proteomic analysis, the preferred buffers are either SDS- or urea-based (see Note 1).
1. SDS buffer for solubilization of denatured proteins: 1 % SDS.
Add 50 mL ddH2O to a 200 mL beaker. Weigh 1 g of SDS and
transfer to the beaker (see Note 2). Use a bit of water to rinse
weighing boat as SDS tends to stick. Mix until SDS dissolves
and transfer content to a 100 mL graduated cylinder. Fill cylinder with more water to make 100 mL 1 % SDS solution.
Store at +4 C.
2. Urea buffer for solubilization of denatured proteins: 8 M urea,
50 mM TrisHCl, pH 7.5. Add 25 mL of 50 mM TrisHCl,
26
Mats Born
Fig. 4 Percent distribution of ratios of median intensity of 1,107 peptides with tyrosine phosphorylation
between snap-frozen (SF) and heat-stabilized (Stabilized) samples isolated from mice brain tissue. 34 % of
peptides are detected at intensities at least 50 % higher or only in stabilized samples whereas only 4 % are
detected at intensities at least 50 % higher or only in snap-frozen samples, indicating a preservation of phosphorylation levels in stabilized samples. Mice brains were collected in triplicates and either just snap-frozen
or heat-stabilized using the Stabilizor system prior to freezing. Proteins were extracted using a urea-based
extraction buffer with protease and phosphatase inhibitors and turned into peptides using trypsin. Peptides
with tyrosine phosphorylations were affinity-enriched using a tyrosine-specific antibody (Cell Signaling,
#9411). Peptide intensities were analyzed using LC-MS/MS
pH 7.5 to a 100 mL beaker. Weigh 24.1 g of urea and transfer

to the beaker. Mix until dissolved and transfer content to a
100 mL graduated cylinder (see Note 3). Fill cylinder with
more 50 mM TrisHCl, pH 7.5 buffer to make 50 mL of 8 M
urea, 50 mM TrisHCl, pH 7.5 solution. Prepare the urea
extraction buffer fresh.
Methods
Enzyme inactivation using heat-induced enzyme denaturation is a
preparative step prior to a large range of protein as well as small
molecule analysis techniques. Heat inactivation in the Stabilizor
T1 is performed using a number of predefined programs depending on the state of the sample, fresh or frozen, as well as the need
to preserve sample morphology. Once the sample has been heatstabilized, the use of correct extraction buffers and homogenization
is vital in order to resolubilize proteins and enable analysis. In this
section, the standard ways for using the Stabilizor T1 instrument
will be covered. Special considerations for homogenization and
resolubilization of heat-inactivated samples will be covered in
Subheading 3.3.
3.1 Heat
Stabilization of Fresh
Samples Directly
After Removal
from the Source
27
In order to minimize enzymatic driven change it is vital that the time

between sampling and stabilization is kept as short as possible.
When using the Stabilizor T1 instrument with fresh samples the
instrument should be next to the sample source and focus should
be on getting the samples heat-stabilized as fast and reproducible
as possible.
1. Prepare all material required during sample extraction, stabilization, and subsequent storage.
2. Start the Stabilizor T1 instrument and select the Auto Fresh
method, if preservation of structural morphology is not needed,
or Fresh (Structural preserve) preservation of structural morphology is needed (see Note 4).
3. Collect samples one at a time.
4. Start the post-mortem clock at the moment of death; if sampling starts with sacrifice (see Note 5).
5. Preferentially sacrifice mice without the use of anesthesia,
which have been shown to affect phosphorylation levels [8]
(see Note 6).
6. Extract sample from source and place in on open Maintainor
Tissue card (see Note 7).
7. Position the Maintainor Tissue card in the sample sledge of the
Stabilizor T1 instrument and press the START button on the
touch screen.
8. When stabilization is complete, the Maintainor Tissue card
with sample is ejected from the instrument.
9. After stabilization, store samples frozen at 80 C, either in the
card or transfer to storage device of your choice, or proceed
with homogenization and extraction (see Note 8).
10. After heat inactivation proceed with homogenization and
extraction as outlined in Subheading 3.3.
3.2 Heat
Stabilization
of Frozen Samples
Already frozen samples can also benefit from heat inactivation as

some enzymes reactive during thawing and can induce change during
homogenization and after extraction. This can be useful in situations
where it is impossible or unpractical to have the instrument close to
the site of sampling and for already biobanked samples. When using
the Stabilizor T1 instrument with frozen samples it is important that
the samples are kept frozen until they enter the instruments and
brought directly from a frozen state to a denatured state.
1. Start the Stabilizor T1 instrument and select the Auto Frozen
method, if preservation of structural morphology is not
needed, or Frozen (Structural preserve) if preservation of structural morphology is needed (see Note 9).
2. Keep samples on dry ice or in 20 C freezer making sure the
samples remain frozen (see Note 10).
28
Mats Born
3. Place an open Maintainor Tissue card on dry ice or in 20 C

freezer and let it equilibrate for a few minutes (see Note 7).
4. Using a chilled forceps transfer the frozen sample to the cold
card and close it.
5. Quickly position the Maintainor Tissue card in the sample sledge
of the Stabilizor T1 instrument (see Note 11). Make sure it is
closed and press the START button on the touch screen.
6. When stabilization is complete the Maintainor Tissue card
with sample is ejected from the instrument. After stabilization,
store samples frozen at 80 C, either in the card or transfer to
storage device of your choice (see Note 8).
7. After heat inactivation proceed with homogenization and
extraction as outlined in Subheading 3.3.
3.3 Homogenization
and Resolubilization
of Heat-Stabilized
Samples
Proteins in heat-stabilized samples will be denatured and this

requires some special considerations concerning homogenization
and resolubilization buffers. Heat-treated tissue is generally more
brittle then non-heated tissue so homogenization is generally not
a problem but mechanical homogenization is needed for all tissues
except brain. The choice of extraction buffer is on the other hand
crucial and a strongly denaturing buffer is needed to resolubilize
the denatured proteins in the sample. Amount of buffer to sample
is also vital for good resolubilization.
3.3.1 Denaturing
Extraction Buffer
Due to the denatured state of proteins in heat-inactivated samples,

a denaturing buffer must be used to efficiently resolubilize proteins
from heat-stabilized samples. Depending on downstream analytical technique, buffers based on urea and/or SDS have been found
to work well. Buffers based on high, >8 M urea, are ideal for
extracting samples for 2D PAGE and chromatographic workflows
such as LC-MS. On the other hand, SDS, >1 % SDS, based buffers
are suitable for SDS-PAGE and western blot workflows. When
using SDS buffers, it is recommended to maximize their effectiveness by heating the buffer (>90 C) before adding it to the sample
and after homogenization as instructed below. Other components
such as detergents and buffering agents can be added as long as the
concentrations of the denaturing agents are not affected.
3.3.2 Buffer
to Sample Ratio
To ensure full dispersion of proteinprotein complexes formed

during cooling after heat stabilization, it is important that sufficient buffer is added to the sample.
1. Weigh samples to be extracted (see Note 12).
2. Calculate the amount of buffer needed to reach a buffer-tosample ratio greater than 10 (>10 l buffer/mg sample).
3. Add buffer to sample and proceed directly with homogenization.

3.3.3 Homogenization
of Tissue in 8 M Urea
Buffer Using Micro pestle
Grinding Followed by
Microtip Rod Sonication
29
Although heat-stabilized samples will usually homogenize easily, it

is very important to ensure that the initial homogenization step is
thorough to facilitate the resolubilization of proteins. Soft tissue
such as brain can be homogenized just using a microtip sonication
rod, firmer tissues, e.g. liver or heart, require physical homogenization, e.g. micro pestle grinding. A combination of physical homogenization followed by microtip sonication rod has been found to
be efficient.
1. Place the sample in a Low Bind Eppendorf tube.
2. Crush the sample using a micro pestle until no structure
remains (see Note 13).
3. Add sufficient buffer to the sample so the buffer-to-sample
ratio is above 10, e.g. >10 l buffer/mg sample.
4. Continue homogenizing the samplebuffer mixture for at least
2 min until no pieces remain and the mix appears homogeneous.
5. Use a microtip sonication rod (Vibra-Cell, Sonics and Materials)
to give the mixture ten bursts of 2 s duration at 40 W setting
(see Note 14).
6. Centrifuge homogenate for 10 min at 13,000 g to pellet
insoluble material.
7. Collect supernatant and continue with analysis or store homogenate frozen for later analysis.
3.3.4 Homogenization
of Tissue in Hot 1 % SDS
Buffer Using Micro pestle
Grinding Followed
by Heating and Microtip
Rod Sonication
1. Heat buffer solution to near boiling, >90 C, in a water bath.

2. Place the sample in a Low Bind Eppendorf tube.
3. Crush the sample using a micro pestle until no structure
remains (see Note 13).
4. Add a sufficient volume of the hot buffer to the sample so the
buffer-to-sample ratio is above 10, e.g. >10 l buffer/mg
sample.
5. Continue homogenizing the samplebuffer mixture for at least
2 min until no pieces remain and the mix appears homogeneous.
6. Heat tube containing samplebuffer homogenate in a heating
block set to 95 C for 5 min.
7. Use a microtip sonication rod (Vibra-Cell, Sonics and
Materials) give the mixture ten (10) bursts of 2 s duration at
40 W setting (see Note 14).
8. Once repeat steps 6 and 7.
9. Centrifuge homogenate for 10 min at 13,000 g to pellet
insoluble material.
10. Collect supernatant and continue with analysis or store homogenate frozen for later analysis.
30
Mats Born
3.4 Analysis
of Extracted HeatStabilized Samples
Heat-stabilized samples can be analyzed just as other samples once

homogenized and extracted as described above. As long as the buffer is compatible with the intended downstream technique no extra
steps are needed. In cases where levels of denaturant interfere with
analysis, e.g. in solution antibody-based techniques like ELISA,
dilution can be useful. Removal of denaturants by buffer exchange,
e.g. size exclusion chromatography, can be problematic as denatured proteins have lower solubility in non-denaturing buffers.
This can cause precipitation of proteins as the denaturant is
removed and should be done with caution.
Notes
1. Denaturing agents of buffers should be >1 % SDS or >8 M
urea. In addition to denaturing agents other components can
be added, e.g. detergents and salts, depending on the needs of
the specific work flow, the SDS and/or urea contents may
however not be lowered.
2. Use dust mask to avoid getting powdered SDS into the lungs
during weighing. It is also a good idea to make a 10 % SDS
stock solution to use for further dilutions to minimize the
number of times powdered SDS have to be handled.
3. As urea dissolves the solution becomes cold. Do not heat to
speed up solubilization as this may create isocyanic acid which
will cause protein carbamylation in later steps. Isocyanic acid
can also form naturally as the urea buffer ages. To avoid this,
always prepare urea containing extraction buffers fresh prior to
use.
4. Both Fresh methods, Quick and Structural preserve, will inactivate enzymes in the sample. The Quick method will compress
the sample to make it thinner, giving a faster heating throughout the sample. The Structure preserve on the other hand use
minimal compression to preserve sample structure. The sample
is thus thicker and stabilization takes longer time.
5. The Post mortem clock is basically a timer keeping track of the
time between sacrifice and stabilization. It is important to try
to keep this time as short and comparative between samples as
possible. The use of the timer helps to focus attention on the
dissection and avoids unnecessary delays. As dissection is an art
which needs practice it is quite common for the dissection time
to decrease by ~50 % between the first and the fifth animal
before leveling out. It is thus good practice to either start with
control animals not part of the study or a group where time is
not as important.
31
6. Cervical dislocation or a scientific Guillotine are examples of

preferred sacrificial techniques.
7. Card can remain open without support if first folded 180. To
minimize air in cavity after closing it, push the upper foil
inwards toward the sample prior to placing the sample in the
cavity.
8. Although the heat stabilization will inactivate enzymes, other
protein-modifying processes, e.g. oxidation, are still active and
could induce protein changes in the sample. Stable levels of
phosphorylations have been shown for up to 24 h in room
temperature after heat stabilization but it is not recommended
to keep stabilized samples at ambient temperatures for more
than a few hours at most [9].
9. Both Frozen methods, Quick and Structural preserve, will
inactivate enzymes in the sample. The Quick method will
compress the sample to make it thinner, giving a faster
heating throughout the sample. The Structure preserve
on the other hand use minimal compression to preserve
sample structure. The sample is thus thicker and stabilization takes longer time.
10. The algorithms controlling treatment time for the Frozen
methods are for treating samples between 78 C and
+20 C. If samples are transported in liquid nitrogen, 196 C,
they need to be equilibrated either at 20 C or on dry ice so
they are not too cold when treated as this will result in under
treatment and residual enzymatic activity.
11. If the Maintainor Tissue cards have been cooled on dry ice
longer than 1 min, hold the card between thumb and index
finger just over the orange seal during transfer to thaw it
slightly. If the orange seal has frozen solid the needle can have
difficulty penetrating the septum.
12. This is greatly facilitated if samples are collected in pre-weighted
containers. Weigh samples one by one transferring them
quickly from dry ice to the scale and read the weight as soon as
reasonable stable. Moisture in the air will start to deposit on
the sample container during weighing so the scale will not fix
on a value.
13. Pre-crushing the sample prior to adding buffer makes micropestle homogenization much easier. If the buffer is added first
it can be quite difficult to hunt the small slippery pieces swirling around in the buffer.
14. Have sonicator tip in buffer while active and do not dip it up
and down to avoid foaming. This is especially relevant with
SDS-based buffers.
32
Mats Born
References
1. Skld K, Alm H, Scholz B (2013) The impact
of biosampling procedures on molecular data
interpretation. Mol Cell Proteomics 12(6):
14891501
2. Stingl C, Sderquist M, Karlsson O et al (2014)
Uncovering effects of ex vivo protease activity
during proteomics and peptidomics sample
extraction in rat brain tissue by oxygen-18
labeling. J Proteome Res 13(6):28072817
3. Svensson M, Born M, Skld K et al (2009)
Heat stabilization of the tissue proteome: a new
technology for improved proteomics. J Proteome
Res 8(2):974981
4. Kultima K, Skld K, Born M (2011) Biomarkers
of disease and post-mortem changesheat stabilization, a necessary tool for measurement of
protein regulation. J Proteomics 75(1):
145159
5. Ahmed MM, Gardiner KJ (2011) Preserving
protein profiles in tissue samples: differing out-
6.
7.
8.
9.
comes with and without heat stabilization.

J Neurosci Methods 196(1):99106
Robinson AA, Westbrook JA, English JA et al
(2009) Assessing the use of thermal treatment
to preserve the intact proteomes of postmortem heart and brain tissue. Proteomics
9(19):44334444
Smejkal GB, Rivas-Morello C, Chang JH et al
(2011) Thermal stabilization of tissues and the
preservation of protein phosphorylation states
for two-dimensional gel electrophoresis.
Electrophoresis 32(16):22062215
Li X, Friedman BA, Roh MS, Jope RS (2005)
Anesthesia and post-mortem interval profoundly influence the regulatory serine phosphorylation of glycogen synthase kinase-3 in
mouse brain. J Neurochem 92:701704
Born M (2011) Methodology and technology
for stabilization of specific states of signal transduction proteins. Methods Mol Biol 717:91100
Chapter 3
Isolating Peripheral Lymphocytes by Density
Gradient Centrifugation and Magnetic Cell Sorting
Frederic Brosseron, Katrin Marcus, and Caroline May
Abstract
Combining density gradient centrifugation with magnetic cell sorting provides a powerful tool to isolate
blood cells with high reproducibility, yield, and purity. It also allows for subsequent separation of multiple
cell types, resulting in the possibility to analyze different purified fractions from one donors sample.
The centrifugation step divides whole blood into peripheral blood mononuclear cells (PBMC), erythrocytes, and platelet-rich plasma. In the following, lymphocyte subtypes can be consecutively isolated from
the PBMC fraction. This chapter describes enrichment of erythrocytes, CD14-positive monocytes and
CD3-positive T lymphocytes. Alternatively, other cell types can be targeted by using magnetic beads
specific for the desired subpopulation.
Key words Peripheral blood mononuclear cells (PBMC), Magnetic-activated cell sorting (MACS),
Fluorescence-activated cell sorting (FACS)
Introduction
Isolation of blood cells for research purposes should ensure optimal
experimental conditions by providing maximum specificity, purity,
yield, speed, and reproducibility [1, 2]. Today, numerous methods
for enrichment of different cell types exist, the most common
being density gradient centrifugation and several variants of antibody-based immunoaffinity methods [3, 4]. Density gradients are
typically used to fractionate full blood into platelet-rich plasma,
erythrocytes, and peripheral blood mononuclear cells (PBMCs)
composed of lymphocytes, monocytes, and macrophages [5, 6].
Cell (sub)types can be efficiently obtained by use of immunoaffinity purifications [4]: These methods make use of antibodies
targeting characteristic proteins like cell surface receptors, for
example to attach labels for instrumental supported isolations like
fluorescence-activated cell sorting (FACS) [7]. Likewise, antibodies are immobilized on supporting material with properties
suited for enrichment methods (e.g., magnetic cell sorting, MACS).
33
34
Frederic Brosseron et al.
The MACS technology has been frequently used and described

by us and others in terms of efficiency and yield [8, 9].
In this chapter, we describe the use of density gradient centrifugation combined with MACS sorting for subsequent isolation of erythrocytes, monocytes, and T lymphocytes from human
full blood (Fig. 1). This strategy offers multiple advantages: It is
possible to obtain multiple cell types from one blood donation,
enabling researchers to investigate multiple targets and make optimal use of the donor sample. The procedure can be easily adapted
to other cell types by choice of the respective MACS beads,
which are commercially available. Furthermore, magnetic cell sorting is of ease of use and therefore also suited for users inexperienced
with blood cell isolation.
Diluted whole blood

Pancoll
Density gradient
centrifugation
Platelet-rich plasma
PBMCs
Pancoll
Erythrocytes
PBMCs
MACS
CD14+ MACS
CD14+ Monocytes
Erythrocytes
CD3+ MACS
CD3+ T-Lymphocytes
FACS
Fig. 1 Overview of workflow. He isolation procedure begins with a density gradient centrifugation which fractionates whole blood into erythrocytes, PBMCs, and platelet-rich plasma. Fractions of intereste.g., erythrocytescan be preserved for analysis directly after the gradient. The PBMC fraction can be further subdivided
by subsequent cell type-specific magnetic cell sorting, as shown in this protocol for CD14-positive monocytes
and CD3-positive T lymphocytes
Lymphocyte Isolation by Magnetic Cell Sorting
35
Materials
2.1 Technical
Equipment
1. Laboratory area for handling of human biomaterial according

to local regulations.
2.1.1 General
2. Laminar flow hood, HEPA-filtered and UV-sterilized.

3. Containment for biofluid waste.
4. Water squirt bottle.
5. Disinfectant.
6. Ice box.
7. 4 C Freezer.
8. 80 C Freezer.
2.1.2 Blood Sampling
1. 20 mL syringes.
2. Butterfly or similar needle.
3. Adapters to connect needle to syringes.
2.1.3 Density Gradient
1. Centrifuge for 50 mL tubes, swing bucket rotor, adjustable

acceleration and breaking, cooling option.
2. Sterile serological pipettes.
3. Pipetting device or Peleus ball for serological pipettes.
4. Sterile 15 and 50 mL conical centrifuge tubes.
5. Stand for centrifuge tubes.
6. 5 mL Plastic Pasteur pipette.
7. Automatic cell counter or microscope with Neumann chamber.
2.1.4 MACS Isolation
1. Magnetic stirring unit.

2. 500 mL cell culture bottles.
3. Sterile filtering device.
4. Precision scales.
5. pH-meter.
6. MACS magnet (Mini MACS and Midi MACS or Vario MACS).
7. MACS MS-columns.
8. MACS LS-Columns.
9. Cell filter fitting to MACS columns.
12. Centrifuge for 15 and 50 mL tubes, swing bucket rotor, adjustable acceleration and breaking, cooling option.
36
15. Micropipettes.
16. Sterile micropipette tips.
2.1.5 Washing of Purified
Monocytes and
T Lymphocytes
1. Centrifuge for 15 and 50 mL tubes, swing bucket rotor, adjustable acceleration and breaking, cooling option.
6. Table centrifuge for 1.5 mL reaction tubes, cooling option.
7. 1.5 mL sterile polypropylene reaction tubes.
8. Stand for reaction tubes.
9. Micropipettes.
2.1.6 Quality Control

Using FACS

2. Micropipettes.
4. Vortex.
5. Flow cytometer.
6. FACS-vials fitting to flow cytometer.
2.2 Buffers
and Solutions
1. Phosphate buffered solution (PBS), sterile, calcium-free,

magnesium-free (PBS/).
2.2.1 General
2.2.2 Blood Sampling
1. Heparin 25,000 solution.
2.2.3 Density Gradient
1. Hanks buffered salt solution (HBSS).

2. Pancoll human, mentioned in the following as Pancoll.
2.2.4 MACS Isolation
1. MACS buffer: 0.5 % BSA, 0.06 % EDTA in PBS/. Adjust to

pH 7.2 using NaOH and HCl, sterile filter and store at 4 C
for a maximum of 1 week (see Note 1).
2. MACS CD14 microbeads.
3. MACS CD3 microbeads.
2.2.5 Quality Control

Using FACS
1. Anti-CD14-FITC.
2. Anti-CD3-PE.
3. Anti-CD19 FITC.
4. Anti-CD56 PE.
3
3.1
37
Methods
Blood Sampling
1. Blood withdrawal is to be performed by authorized medical

personal only according to local legislative regulations.
2. Prepare 4 20 mL syringes by placing 50 l Heparin 25,000 as
anticoagulant into the inlet of the syringe before blood sampling (see Note 2).
3. Fill each syringe by venipuncture of cubital vein (see Note 3).
4. Slowly invert syringes three times to ensure homogenous mixing of blood and anticoagulant.
5. Proceed to density gradient immediately.
3.2
Density Gradient
1. Let Pancoll human adjust to room temperature for 1 h

(see Note 4).
2. Dilute 20 mL venous blood 1:1 with HBSS in a 50 mL tube.
Invert three times to ensure homogenous mixing (see Note 5).
3. Place 10 mL Pancoll in a 50 mL centrifuge tube.
4. Layer diluted venous blood slowly and carefully on top of the
Pancoll, using an automatic pipetting device on the lowest
setting or a Peleus ball (see Fig. 2).
5. Centrifuge for 20 min, RT, 1,250 g using a swing bucket
rotor, with settings for acceleration and deceleration as low as
possible (see Note 6).
Diluted
blood
Pancoll
Fig. 2 Preparation of density gradient. To prepare the density gradient, begin by placing a small volume of
diluted blood as slow as possible on top of the Pancoll at the edge of the centrifuge tube. Stabilize the pipette
by pressing it at the edges of the centrifuge tube. Wait until the blood has distributed over the Pancoll, then
slowly move the pipette up while continuously letting the diluted blood flow out. Avoid too fast outlet of the
blood as this will cause disturbances in the gradient, resulting in poor performance
38
6. The gradient should result in four clearly separated phases,

listed in order of increasing density: 1st supernatant (yellowish,
remaining serum constituents and platelets), 2nd PBMCs (grey
phase), 3rd Pancoll (clear), and 4th erythrocytes (dark red) at
the tube bottom (see Fig. 1).
7. Perform all following steps consequently on ice or at 4 C with
precooled buffers and reagents (4 C).
8. Prepare 2 50 mL centrifuge tubes with 10 mL cold
PBS/ each.
9. Remove supernatant of gradient using a serological pipette to
1 cm above the PBMC phase.
10. Collect the grayish PBMCs by using a plastic Pasteur pipette,
starting at the edge of the tube and later moving across the
plane of the PBMC layer. Place PBMCs in the prepared 50 mL
tubes containing the PBS/ (see Subheading 3.4 and Note 7).
11. Remove remaining Pancoll using a serologic pipette.
12. Use remaining erythrocyte suspension for washing (see
Subheading 3.3).
3.3 Preparation
of Erythrocytes
1. Fill up tube from step 12 in Subheading 3.3 to 50 mL using

PBS/.
2. Centrifuge the erythrocytes for 8 min, 4 C, 470 g with
medium acceleration and deceleration to remove remaining
contaminants from the erythrocytes (see Note 8).
3. Discard the supernatant and the top 12 mL of the erythrocyte
fraction to ensure complete removal of contaminants.
4. Repeat steps 2 and 3 with the erythrocyte fraction.
5. Collect purified erythrocytes in one centrifuge tube and store
at 80 C until usage.
3.4 Washing
of PBMCs
1. Fill centrifuge tube from step 3.3. 10 to 50 mL with PBS/

and invert 3 times to wash the PBMC fraction.
2. Centrifuge the PMBCs for 8 min, 4 C, 470 g to remove
remaining contaminants (see Note 9).
3. Discard supernatant and resuspend the PBMCs in 1 mL PBS/.
4. Unite PBMCs from two tubes into one fresh 50 mL centrifugation tube, fill up to 50 mL with PBS/ and repeat step 2.
5. Discard supernatant and resuspend PBMCs from all remaining
tubes in 1 mL PBS.
6. Take up an aliquot for cell counting. Use a light microscope
combined with a Neubauer counting chamber or an automated
cell counter according to the manufacturers instructions.
7. Preserve 3 200,000 cells for FACS analysis of the PBMC
fraction (see Note 10).
39
8. After cell counting, wash PBMCs a third time in PBS/ and

remove supernatant.
9. Use PBMC pellet for MACS cell isolation (see Subheading 3.6).
3.5 Isolation
of CD14+ Monocytes
and CD3+ T
Lymphocytes
1. Resuspend the PBMC pellet from step 9 in Subheading 3.5 in

MACS buffer in a 15 mL centrifugation tube according to
the manufacturers instructions.
2. Add CD14+ MACS beads according to the manufacturers
instructions and vortex.
3. Incubate 4 C for 15 min.
4. Fill tube to 15 mL with MACS buffer, invert 3 times and
centrifuge for 10 min, 4 C, 300 g.
5. Discard supernatant and resuspend the cell pellet in MACS
buffer according to the manufacturers instructions.
6. Purify CD14+ monocytes using a MACS MS column in a
Mini MACS magnet, cell filter attached, according to the
manufacturers instructions.
7. Collect both flow-through (contains CD14 depleted PBMCs)
and elution fraction (contains CD14+ monocytes) in a separate
15 mL centrifugation tube.
8. Wash elution fraction (see Subheading 3.7).
9. Fill flow-through to 15 mL with MACS buffer and wash as in
step 13.
10. Resuspend PBMCs from flow-through in MACS buffer for
CD3+ sorting according to the manufacturers instructions.
11. Add CD3+ MACS beads according to the manufacturers
instructions and vortex.
12. Incubate 4 C for 15 min.
13. Fill tube to 15 mL with MACS buffer, invert 3 times and centrifuge for 10 min, 4 C, 300 g.
14. Discard supernatant and resuspend the cell pellet in MACS
buffer according to the manufacturers instructions.
15. Purify CD3+ T lymphocytes using a MACS LS column in a
Midi MACS magnet, cell filter attached, according to the manufacturers instructions.
16. Collect elution fraction (contains CD3+ T lymphocytes) in a
15 mL centrifugation tube.
17. Wash elution fraction (see Subheading 3.7).
3.6 Washing
of Purified Monocytes
and T Lymphocytes
1. Resuspend cell pellets from step 7 in Subheading 3.6 or step 16

in Subheading 3.6 in 1 mL PBS/.
2. Fill to 15 mL with PBS/, invert 3 times, and centrifuge for
10 min, 4 C, 300 g.
40
3. Discard supernatant, resuspend in 1 mL PBS/ and count cells.

4. Preserve 3 200,000 cells for FACS analysis of the purified cells.
5. Transfer the resuspended cells into a 1.5 mL reaction tube and
fill with PBS/.
6. Centrifuge the CD14+ cells for 5 min, at 4 C and 1,000 g.
7. Remove supernatant completely and store the dry cell pellet
at 80 C until usage.
3.7 Quality Control
Using FACS
1. For each isolated fraction (PBMCs, Monocytes, T lymphocytes), label 3 FACS vials as follows:
(a) Non-labeled control
(b) CD14 FITC + CD3 PE
(c) CD19 FITC + CD56 PE
2. Pipette 200,000 cells of the respective isolated fraction in
each vial.
3. Incubate cells with FACS antibodies according to the vial
labels and manufacturers instructions.
4. Measure on a flow cytometer according to the manufacturers
instructions (see Note 11).
5. Determine composition of fractions by comparing ratios of
CD14+ (monocytes), CD3+ (T lymphocytes), CD19+ (B lymphocytes), and CD56+ (NK cells). The isolation of monocytes
and T lymphocytes by MACS should provide purities > 85 %
(Fig. 3).
Notes
1. The MACS buffer is best prepared on the day before the cell
isolation.
2. The described protocol was originally designed for large yields
for proteomics experiments without further culture of the cells
and is therefore based on a large blood donation of 80 mL [9].
The procedure is adaptable to any other volume by proportionally adjusting volumes of Pancoll and MACS beads.
Magnetic cell sorting can also be performed directly from full
blood, but the density gradient will help to pre-purify and concentrate the PBMCs in a small volume for MACS isolation.
3. The appropriate blood sampling system/adapters should be
chosen by authorized and experienced medical personal.
4. Cold Pancoll has a different density than warm Pancoll
which can cause inefficient separation in density gradient.
Remove Pancoll from freezer before blood sampling is performed to ensure that there is no time delay between sampling
and gradient.
41

MACS
CD14+ Fraction
PBMC-Fraction
CD3+
FL2-Height
FL2-Height
CD3+
102
101
unlabeled
cells
0.2%
59.1%
103
28.3%
12.4%
CD14+
100
100
101
102
Unlabeled
cells
4.0%
0.4%
3.1%
92.5%
103
102
101
CD14+
100
100
103 104
101
102
103
104
FL1-Height
FL1-Height
MACS
CD3+ Fraction
FL2-Height
CD3+
94.9%
0.4%
2.4%
2.3%
103
102
101
CD14+
Unlabeled
cells
100
100
101
102
103 104
FL1-Height
Fig. 3 Purity control by FACS. Lymphocyte fractions analyzed by FACS. In fractions isolated from PBMCs by
MACS, proportions of the selected cells are drastically increased: In this example, the PBMC fractions consist
of 46 % of CD3+ cells (T lymphocytes) and 3 % of CD14+ cells (monocytes). By MACS sorting, these proportions are increased to 88 % (CD14+ fraction) and 95 % (CD3+ fraction)
5. This step is necessary to adjust the density of the blood for the
gradient.
6. Fixed angle rotors and strong acceleration or deceleration will
cause inefficient separation in gradient.
7. When collecting the PBMCs, include parts of the supernatant
and Pancoll to ensure complete harvest of the PBMCs. The
following washing and magnetic cell isolation will remove the
contaminants.
8. Erythrocytes will not form a pellet, but a concentrated
suspension.
9. This step can be performed together with step 2 in
Subheading 3.4.
10. The necessary amount of cells for FACS analysis depends on
the used instrument and measurement protocols.
11. FACS analysis should be performed in a lab with high experience
in analysis of mononuclear cells.
42
Acknowledgement
This work was supported by P.U.R.E. (Protein Unit for Research
in Europe), a project of Nordrhein-Westfalen, a federal state of
Germany, the Bundesministerium fr Bildung und Forschung
(NGFNplus, FZ 01GS08143), Germany, and the German Center
for Neurodegenerative Diseases (DZNE e.V.) within the Helmholtz
Association.
References
1. Blonder J, Issaq HJ, Veenstra TD (2011)
Proteomic biomarker discovery: its more than
just mass spectrometry. Electrophoresis 32:
15411548
2. Mallick P, Kuster B (2010) Proteomics: a pragmatic perspective. Nat Biotechnol 28:695709
3. Bennett S, Breit SN (1994) Variables in the
isolation and culture of human monocytes that
are of particular relevance to studies of HIV.
J Leukoc Biol 56:236240
4. Dainiak MB, Kumar A, Galaev IY et al (2007)
Methods in cell separations. Adv Biochem Eng
Biotechnol 106:118
5. Syrovets T, Tippler B, Rieks M et al (1997)
Plasmin is a potent and specific chemoattractant
for human peripheral monocytes acting via a
cyclic guanosine monophosphate-dependent
pathway. Blood 89:45744583
6. de Almeida MC, Silva AC, Barral A et al (2000)

A simple method for human peripheral blood
monocyte isolation. Mem Inst Oswaldo Cruz
95:221223
7. Gorelik L, Kauth M, Gehlhar K et al (2008)
Modulation of dendritic cell function by
cowshed dust extract. Innate Immun 14:
345355
8. Mayer A, Lee S, Lendlein A et al (2011)
Efficacy of CD14(+) blood monocytes/macrophages isolation: positive versus negative
MACS protocol. Clin Hemorheol Microcirc
48:5763
9. Brosseron F, May C, Schoenebeck B et al
(2012) Stepwise isolation of human peripheral
erythrocytes, T lymphocytes, and monocytes
for blood cell proteomics. Proteomics Clin
Appl 6:497501
Chapter 4
Investigating the Adipose Tissue Secretome: A Protocol
to Generate High-Quality Samples Appropriate
for Comprehensive Proteomic Profiling
Simon Gddeke, Jorg Kotzka, and Stefan Lehr
Abstract
In this chapter, we describe in detail how to prepare a sample containing the complete entity of secretion
products from murine primary adipocytes, which are suitable for comprehensive and sensitive secretome
analysis. The underlying protocol should be seen as a starting point guiding through critical steps of the
complex workflow in order to approximate to the real secretome in the context of different sample types
used for the diverse research questions the protocol has to be carefully adjusted.
Key words Secretome analysis, Adipokines, Murine adipose tissue, Pre-adipocytes
Introduction
Obesity, which is favored by imbalanced energy supply and
energy consumption coupled with a sedentary lifestyle is considered as an epidemic disease and represents a burden for almost
all societies [1]. Therefore, overweight today is the major risk
factor for developing various metabolic complications such as
insulin resistance, type 2 diabetes, non-alcoholic liver disease,
and cardiovascular diseases [26].
Over the last decade we have learned that adipose tissue, besides
its predominant role within energy homeostasis also represents a
major endocrine organ, releasing a wide variety of signaling and
mediator proteins. These so-called adipokines seem to be causally
involved in the development of a wide variety of metabolic diseases.
Over the last years, several attempts have been made to characterize
the adipokines by utilizing diverse proteomic profiling approaches.
This has led to a catalog of adipokines comprising several hundred
potentially secreted peptides and proteins [7]. Nevertheless, the limited overlap of identified adipokines in published studies impressively
43
44
Simon Gddeke et al.
illustrates that the analysis of tissue-specific secretomes is an extreme

challenging business.
Within the complex workflow including sample preparation,
concentration, proteomic profiling and extensive bioinformatics,
the most striking point is to discriminate the real secretome from
contaminating proteins. Therefore, a meaningful characterization
inalienable starts with high-quality samples and requires a reproducible processing as crucial points for successful identification of
secreted proteins. Particularly during sample collection and culturing processes considerable contaminations could occur from different sources, e.g. release of cellular proteins due to cell damage
or contamination of high-abundant proteins derived from culture,
i.e. FCS. Due to the fact that the expected concentrations of adipokines are low (pg to ng/ml), artificial proteins have the potential
to shift the dynamic range of the secretome sample dramatically.
In the light of these possible pitfalls, a careful evaluation of sample
preparation and culture setup to minimize contamination are the
most crucial steps in order to obtain relevant secretome data [7, 8].
Materials
2.1 General
Hardware
and Consumables
1. Incubator (37 C; 5 % CO2; 95 % humidity).

2. Sterile cell culture bench.
3. Centrifuge.
4. Ultracentrifuge.
5. Laboratory scissors (double sharped); 145 mm.
6. Scalpel grip Nr. 4 L.
7. Scalpel blades Nr. 22.
8. Forceps.
9. 6-Well-culture plates.
10. Polypropylene (PP) centrifugation tubes, 15 ml.
11. Polypropylene (PP) centrifugation tubes, 50 ml.
12. Sterile serological pipettes, 10 ml.
13. Glass funnel, pasteur pipettes.
14. Incubation rotator.
15. Steam-sanitizer.
16. Special accuracy scale.
2.2
Tissue Biopsies
2.3 Isolation
of Pre-adipocytes
1. Hanks Balanced Salt Solution (HBSS) w/o Calcium and

Magnesium (pH 7.4).
1. Cell Strainer, 70 m, Nylon (BD Falcon).
2. Mesh, 150 m.
Secretome Analysis of Adipose Tissue
45
3. Mesh, 75 m.
4. Syringe filters, 25 m.
5. Special accuracy scale.
6. Collagenase (NB8, Clostridium Histolyticum).
7. Phosphate buffered saline (PBS) w/o Calcium and Magnesium,
sterile.
8. Collagenase solution (750 U collagenase/g adipose tissue):
250 U/ml in HBSS (pH 7.4) (see Note 1) supplemented with
3 mM CaCl2 (see Note 2) and 4 mM glucose. Sterile filtration
(10 ml syringe), G26 cannula (1), sterile syringe filters
(0.2 m, 33 mm).
9. HBSS to dilute the collagenase: add CaCl2 (3 mM).
10. Erythrocyte lysis buffer: 8.29 g/l NH4Cl, 0.99 g/l K2HPO4,
0.04 g/l EDTA, pH 7.3 sterile filtration.
11. Basal medium: DMEM/F-12 (17.5 mM glucose), 1.25 g/l
NaHCO3, 16 mg/l Biotin, 8 mg/l Calcium-D-pantothenate,
pH 7.3, sterile filtration. Before use add 5 ml/l Gentamycin.
12. Basal medium w/o phenol-red: DMEM, w/o phenol red,
w/o FCS, 10 pmol/ml insulin. Before use add 5 ml/l
Gentamycin.
2.4 Differentiation
of Pre-adipocytes
1. Vacuum filtration unit, 500 ml (0.22 m).

2. Ethylendiaminetetraacetic acid (EDTA) solution pH 8.0
(1 mg/ml).
3. Dulbeccos Modified Eagles Medium/F-12 (DMEM/F-12,
17.5 mM glucose).
sterile.
5. Cultivation medium: Basalmedium (Subheading 2.3, item 12),
10 % FCS, 5 ml/l Gentamycin.
6. Troglitazon (TZO) stock solution: Solve 5 mg in 1 ml
Dimethylsulfoxide (DMSO).
7. Differentiation medium #1 (for 50 ml) (see Note 3): 50 ml
basal medium (Subheading 2.3, item 12), 5 l TZO stock
solution (see Note 4), 25 l Apo-Transferrin stock solution,
100 l Hydrocortisol stock solution, 5 l T3 stock solution,
150 l insulin stock solution.
8. Differentiation medium #2: Differentiation medium #1 w/o
TZO, supplement with 10 % FCS.
9. Apo-Transferrin stock solution: Solve 50 mg in ml ddH2O.
10. Hydrocortisol stock solution (50 g/ml): Solve 1 mg in 1 ml
EtOHabs and dilute in basal medium w/o phenol-red.
46
Simon Gddeke et al.
11. Trijodthyronin (T3) stock solution (20 g/ml): Solve 5 mg in

1 ml 1 N NaOH and dilute in ddH2O.
12. Insulin stock solution (=104 M): Solve 5 mg in 860 l 0.01 N
HCl and dilute in 10 ml ddH2O.
2.5 Quality Control
of Differentiation by
Oil Red O Protocol
1. Oil Red O stock solution: Mix stock solution in 15 ml PP centrifugation tubes with ddH2O (6:4, v/v) and incubate for at
least 2 h at room temperature in the dark. Filtrate the solution
before use and centrifuge for 10 min at 3,200 g.
2. Hemalaun solution: Dissolve Hematoxylin in ddH2O (1 mg/ml)
and add sodium iodate (0.2 mg/ml) and potassium sulfate
(50 mg/ml) while agitating (see Note 5). After mixing add
chloride-hydrate (50 mg/ml) and crystalline citric acid (1 mg/ml)
(see Note 6).
3. Fixation solution: Mix 15 ml Picric acid with 5 ml Formol
(37 %) and 1 ml acetic acid (100 %).
4. Oil Red O stock solution: 0.3 g Oil Red O solve in 100 ml
isoproanol (99 %) (see Note 7).
2.6 Collecting
Secreted Peptides/
Proteins
1. Mesh, 50 m.
2. Ethylendiaminetetraacetic acid (EDTA) solution pH 8.0
(1 mg/ml).
sterile.
2.7 Concentration
via Centrifugal Filter
Concentrator
1. Centrifugal filter concentrator (Amicon Ultra 15, 3 kDa).
Methods
In this chapter we describe in detail how to prepare a specimen,
containing the complete entity of secretion products from murine
primary adipocytes, which are suitable for comprehensive and sensitive secretome analysis. The underlying protocol should be seen
as a starting point guiding through critical steps of the complex
workflow in order to approximate to the real secretome in the context of different sample types used for the diverse research questions the protocol has to be carefully adjusted (Fig. 1).
3.1
Tissue Biopsies
1. On the day of surgery, after a 6 h fast, mice are sacrificed and

adipose tissue specimens are obtained from the subcutaneous,
visceral or brown depots.
2. Harvest the tissue as a whole part, not in small pieces.
3. Tissue specimens were immediately transferred in HBSS and
transported on ice to the laboratory (see Note 8).
47
Fig. 1 Workflow scheme to generate secretome sample from differentiated mouse adipocytes
3.2 Isolation
of Pre-adipocytes
(Figs. 2, 3, and 4)
1. Dissect tissue from fibrous material and visible blood vessels

and cut into fragments of 510 mg (see Note 9).
2. Transfer the mashy tissue in a 50 ml tube and add collagenase
mixture (see Note 10).
3. Incubate fat tissue fragments with vigorous
(250 cycles/min) for at least 30 min at 37 C.
shaking
4. Centrifugation: 5 min, 240 g, RT (see Note 11).

5. Re-suspend the pellet in HBSS (1:10, v/v).
6. The supernatant, more precisely the floating mature adipocytes, is harvested and transferred to another 50 ml tube with
a Pasteur pipette (see Note 12).
7. Stop collagenase reaction by adding of 0.01 M EDTA solution
(1:10, v/v) (see Note 13).
8. Mix the solution thoroughly by carefully shaking the tube
(at least 5 times).
9. Centrifugation: 5 min, 240 g, 4 C.
10. Discard the supernatant by aspiration and re-suspend the pellet
in HBSS (1:10, v/v).
11. Pool both pellets (steps 5 and 10), i.e. the stromal-vascular
cell fraction, and transfer the cell suspension in a 50 ml tube.
12. Add erythrocyte-lysis buffer (1:10, v/v) and incubate after
mixing the suspension on ice 10 min at longest (see Note 14).
13. Thereafter the tube is filled up to 50 ml with HBSS.
14. Centrifugation: 5 min, 240 g, 4 C.
in basal medium (1:10, v/v).
16. This suspension is filtered without pressure through a 150 m
filter (Image 3).
17. Thereafter flow-through is filtered without pressure through
another filter (75 m) (see Note 15).
18. Centrifuge the flow-through: 5 min, 240 g, 4 C (see Note 16).
48
Simon Gddeke et al.
Fig. 2 Digested and centrifuged adipose tissue resulting in floating mature

adipocytes
Fig. 3 Adipocyte-free pellet before (left) and after erythrocyte lysis (right; the
pellet should be white and free of erythrocytes)

in DMEM/F-12 medium supplemented with 20 % FCS (1:1,
v/v) (see Note 17).
20. Inoculate the cells into a 35 mm dish at a density of approx.
50,000 cells/cm2.
3.3 Differentiation
of Pre-adipocytes
1. Incubate the cells in a humidified 5 % CO2 atmosphere at 37 C.

2. Incubate the cells in cultivation medium until reaching
confluence.
49
Fig. 4 150 m filter (left) and 75 m mesh (right)
3. After that the cells are repeatedly washed with 1 PBS to

remove non-adhering material and incubated in differentiation
medium #1 for another 3 days.
4. Thereafter the cells are washed with 1 PBS to remove nonadhering material and incubated in differentiation medium #2
(see Note 18).
5. The cells are differentiated after at least 7 days.
6. The mature adipocytes can now be cultivated for up to 14 days
before becoming apoptotic.
7. In the phase of establishment a lipid staining with oil red is
recommended.
3.4 Quality Control
of Differentiation by
Oil Red O Protocol
(Fig. 5)
Oil Red O protocol is an assay performed to detect mature adipocytes in histological visualization of fat cells by staining neutral fat
1. Soak off the culture medium and wash the cells very gently
with 1 PBS two times.
2. Discard the supernatant by aspiration and add the fixationsolution.
3. Incubate at RT for 2 h (see Note 19).
4. Discard the fixation solution and wash with 1 PBS two times
(see Note 20).
5. Accordingly incubate the cells in 40 % isopropyl alcohol for
5 min.
6. Discard the isopropyl-solution and add the Oil Red O solution.
50
Simon Gddeke et al.
Fig. 5 Oil Red O staining of differentiated pre-adipocytes (400)
7. Incubate at RT for 15 min (see Note 21).

8. Discard the Oil Red O solution and perform a short incubation
(~5 s) in isopropyl alcohol.
9. Discard the isopropyl solution and subsequently wash briefly
with ddH2O.
10. The wet cells are incubated with the hemalaun solution for
2 min (see Note 22).
11. Afterwards the staining solution is discarded and tap water is
added (see Note 23).
3.5 Collecting
Secreted Peptides/
Proteins
1. Wash the differentiated pre-adipocytes (mature adipocytes)

carefully two times with 1 PBS supplemented with 3 mM
CaCl2 (see Note 24).
2. Thereafter add basal medium w/o phenol-red and incubate
for 4 h.
3. Wash the cells carefully two times with 1 PBS supplemented
with 3 mM CaCl2.
4. Add basal medium w/o phenol-red and incubate the cells
for 26 h.
5. Collect the supernatant in 15 ml polypropylene tube and centrifuge for 30 min at 4,900 g, 4 C.
6. Transfer the supernatant to another tube, determine the protein concentration and add EDTA solution and protease inhibitors (1).
7. Concentrate the supernatants with filter concentrator to a
protein concentration of 2 g/l and store it at 80 C.
51
Fig. 6 Visualization and quality control of secretome samples using 2-dimensional gel electrophoresis. Aliquots
(40 g protein each) of (a) secretome of differentiated, primary adipocytes, (b) total lysate of differentiated
primary adipocytes and (c) Fetal Calf Serum (FCS). Samples were separated by 2-dimensional gel electrophoresis according to the 2D-ToGo workflow [9] standard operation procedure established in our group. Accordingly,
in the first dimension samples in ReadyPrep Sample Buffer (8 M urea, 3 % CHAPS, 50 mM DTT, 0.2 % (w/v)
Bio-Lyte 3/10 ampholytes) were separated by isoelectric focusing (IEF) using pH 310 linear ReadyStrip
IPGstrips (pH 310, 11 cm) performed on a Protean i12 electrophoresis unit (Bio-Rad) and in the second
dimension by SDS-PAGE using Criterion TGX Stainfree Any KD precast gels (Bio-Rad). After electrophoretic separation spot pattern were acquired and documented without further staining, using a ChemiDoc MP
(Bio-Rad) equipped with Image Lab Software
3.6 Concentration
via Centrifugal Filter
Concentrator
Secretome samples frequently are in the range of only a few g/ml

and need a further concentration step to enable comprehensive
analysis of the complex secretome.
1. The samples are centrifuged for 45 min at 85,000 g and 4 C
to get rid of remaining cell debris.
2. The supernatants were transferred to a filter concentrator with
small pore size (3 kDa) (see Note 25).
3. Centrifuge at 4,000 g and 4 C to a final volume of 100 l
(see Note 26).
The secretome sample now is appropriate for comprehensive
targeted and non-targeted proteomic profiling. Complexity and
sample quality could be evaluated, for example, by 2-DE (Fig. 6),
which provides a powerful and reproducible profiling tool.
Notes
1. Consider the unique activity of each collagenase-charge: e.g.
PZU activity of 1.0 = 1,000 U/mg. When weighing the collagenase wearing a surgical mask is recommended.
2. The CaCl2 is crucial for activation of the collagenase.
3. Do not use this medium for more than a week.
4. TZO must be diluted in DMSO, not in aqueous solvent.
5. The solution should get a violet-blue staining.
52
Simon Gddeke et al.
6. The mixture should get a red-violet staining, filtrate the solution with a 150 m paper filter before use.
7. Only use the solution for 12 weeks until precipitation.
8. The buffer has to be freshly prepared, sterile filtered and stored
on ice.
9. For rough dissection the use of a forceps and medical scissors
is recommended. Therefore, you mince the adipose tissue, e.g.
in a weighing pan.
10. Collagenase mixture = tissue/collagenase solution (1:1, v/v).
11. In all of the following centrifugation steps, the use of a swingout rotor is necessary to ensure adequate separation of the
fractions.
12. At this point harvested adipocytes can be used for further analysis. On the one hand, it is possible to generate secretome from
the floating mature adipocytes and furthermore they can be
used for Western blot and qPCR analyses.
13. Do not use a solution of more than 0.5 M of EDTA, because
otherwise cells tend to clot. Using serum/FCS is not efficient
enough to stop the collagenase reaction.
14. This step is crucial to remove contaminating erythrocytes. It is
recommended to monitor the reaction, because ammonium
chloride might damage the cells when incubating too long.
15. Both filtering steps are needed to isolate the pre-adipocytes,
because using only a single step/mesh might end up with
plugging of the mesh and less yield of pre-adipocytes.
16. Never centrifuge for more than 10 min, because the cells are
destroyed otherwise.
17. If not planned to work on parallel with the mature adipocytes.
18. This medium was changed every 23 days, until full
differentiation.
19. This incubation step can be elongated up to 24 h if needed.
20. Wash until the yellowish dye of the washing buffer is extincted.
21. The incubation time depends upon the staining. Better monitor with microscope.
22. For visualizing of the nucleus cells can stained with hemalaun.
23. Use enough water to cover all the cells. Use tap water to ensure
presence of HO-ions which activate the metalhematoxylin
complex. The incubation will probably take longer than 10 min
until blue staining.
24. The addition of CaCl2 is crucial to wash away the FCS totally.
This enables a subsequent 2D gel analysis. Wash very carefully,
because attachment of cells is not very strong.
53
25. The use of a 3 kDa filter is of importance to ensure retaining

all kinds of proteins in the solution.
26. The retention recovery performance of the filters is ~90 % so
that an additional concentration step of the flow-through
might be useful.
References
1. Misra A, Khurana L (2008) Obesity and the
metabolic syndrome in developing countries.
J Clin Endocrinol Metab 93:930
2. Alberti KG, Eckel RH, Grundy SM et al (2009)
Harmonizing the metabolic syndrome: a joint
interim statement of the International Diabetes
Federation Task Force on Epidemiology and
Prevention; National Heart, Lung, and Blood
Institute; American Heart Association; World
Heart Federation; International Atherosclerosis
Society; and International Association for the
Study of Obesity. Circulation 120:16401645
3. Golay A, Ybarra J (2005) Link between obesity
and type 2 diabetes. Best Pract Res Clin
Endocrinol Metab 19:649663
4. Moore JB (2010) Non-alcoholic fatty liver disease: the hepatic consequence of obesity and the
metabolic syndrome. Proc Nutr Soc 69:211220
5. Despres JP, Lemieux I, Bergeron J et al
(2008) Abdominal obesity and the metabolic
6.
7.
8.
9.
syndrome: contribution to global cardiometabolic risk. Arterioscler Thromb Vasc Biol 28:
10391049
Hauner H, Petruschke T, Russ M et al (1995)
Effects of tumour necrosis factor alpha (TNF
alpha) on glucose transport and lipid metabolism of newly-differentiated human fat cells in
cell culture. Diabetologia 38:764771
Lehr S, Hartwig S, Sell H (2012) Adipokines: a
treasure trove for the discovery of biomarkers
for metabolic disorders. Proteomics Clin Appl
6:111
Brown KJ, Formolo CA, Seol H et al (2012)
Advances in the proteomic investigation of the
cell secretome. Expert Rev Proteomics 9:
337345
Posch A, Franz T, Hartwig S et al (2013)
2D-ToGo workflow: increasing feasibility and
reproducibility of 2-dimensional gel electrophoresis. Arch Physiol Biochem 119:108113
Chapter 5
Methods for Proteomics-Based Analysis of the Human
Muscle Secretome Using an In Vitro Exercise Model
Mika Scheler, Martin Hrabe de Angelis, Hadi Al-Hasani,
Hans-Ulrich Hring, Cora Weigert, and Stefan Lehr
Abstract
Over the last decade, the skeletal muscle as a secretory organ gained in importance. A growing number of
peptides are described which are produced and released by the muscle fibers and work in an autocrine,
paracrine, and endocrine fashion. The contraction-induced secretion of these myokines is considered to
contribute to the health-promoting effects of exercise. To gain further insights into the molecular processes that occur during contraction an in vitro exercise model, electric pulse stimulation (EPS), was established. Recent publications show that this model is suitable to electro-stimulate human skeletal muscle cells
and thus mimic muscle contraction in vitro. Here, we provide a detailed protocol for the proteomics-based
analysis of the human muscle secretome, starting with the cultivation of human myotubes and their electric
pulse stimulation, ending with sample preparation for targeted and untargeted proteome analysis of the
cell culture supernatant. This whole workflow should allow deeper insights into the complex nature of the
muscle secretome and the identification of new myokines which might help to understand the crosstalk of
the working muscle with different organs and the beneficial effects of exercise.
Key words Secretome, Electric pulse stimulation, Myokine, Human myotubes, 2D-DIGE, LC-MS
Introduction
It is commonly accepted that regular physical activity entails
multiple health-promoting effects, plays a key role in the prevention and treatment of metabolic and cardiovascular diseases [1]
and reduces systemic inflammation [2]. For decades, it was hypothesized that circulating factors are produced and released from the
contracting muscle and enable the communication between the
energy-demanding muscle and the energy-supplying organs like
adipose tissue and liver during exercise [3]. Interleukin 6 (IL-6)
was first discovered as such an exercise factor when a higher
expression and release of this cytokine was detected in human
skeletal muscle after contraction [4] and the putative function of
IL-6 as enhancer of endogenous glucose production in exercising
55
56
Mika Scheler et al.
humans was described [5]. With this cytokine the concept of

myokines, peptides produced and released by the muscle fibers
which work in an autocrine, paracrine and endocrine fashion arose
[6]. Exercise seems to be an important stimulus for the expression
and secretion of several myokines including amongst others IL-7,
IL-8, Leukemia inhibitory factor (LIF), Myostatin, and Irisin [6].
Although over the last decade many attempts have been made
to identify the complex human muscle secretome, the knowledge
on myokines and their expression and release especially during
contraction is still very limited and incomplete. The global
proteomics-based search for exercise factors in human plasma
samples is still very challenging due to a concentration range of
several orders of magnitude of the different secreted protein species
as recently discussed [7]. Proteins and peptides released from muscle fibers are supposed to reach only very low molecular concentrations in the circulation. Certainly, one additional problem is the
identification of the muscle cell per se as secretory unit of these
circulating exercise factors. Several myokines are not only secreted
by the muscle but also by other tissues, e.g. adipose tissue and liver.
To overcome these limitations, electric pulse stimulation (EPS)
has been applied to human myotubes and the suitability of this
model for the identification of myokines was tested. During the
last years EPS has been established as in vitro exercise model, which
enables the analysis of molecular processes that are occurring during exercise in differentiated skeletal muscle cells. In this model,
the in vivo motor nerve activation of the muscle is replaced by
electric pulses leading to Ca2+ transients that induce contraction
and de novo sarcomere formation in murine C2C12 myotubes and
human primary myotubes [811]. In human muscle cell culture,
the EPS-induced contraction leads to an increased glucose uptake
and expression of several proteins involved in oxidative phosphorylation [9, 12]. In a recent paper, we were able to show that EPS of
human myotubes leads to an activation of the MAP kinase signaling cascade which results amongst others in an altered myokine
expression and that this in vitro exercise model is indeed suitable to
analyze the contraction-induced secretome profile of muscle cells
[11]. In this study, we compared the transcriptional profile of control cells vs. EPS cells and found 153 significantly upregulated
transcripts with fold changes > 1.3 times including 35 genes encoding for secreted proteins. Amongst the top 10 regulated genes with
highest fold increases, 7 were supposed to be secreted as proteins.
In an immunobead-based assay 12 of 23 analytes were significantly
enriched in the EPS supernatant. Using a similar approach, Raschke
et al. analyzed conditioned medium of EPS and control cells and
identified 48 contraction-induced myokines [13]. These results
indicate that the EPS model can be used to study the human
muscle secretome by a global proteomics-based approach.
Analysis of the Human Muscle Secretome
57
Here, we provide a detailed protocol for the cultivation of

human myoblasts, differentiation to myotubes with subsequent
electric pulse stimulation and sample preparation of the supernatant for targeted and untargeted proteomics-based analysis.
Materials
2.1 General
Consumables
1. Pipetboy, pipet and suitable tips.

2. 1.5 mL microcentrifuge tubes.
3. 150 cm2 cell culture dish.
4. 6-well cell culture dish (see Note 1).
2.2
Cell Culture
1. Primary skeletal muscle cells are obtained from percutaneous

needle biopsies performed on the lateral portion of quadriceps
femoris (vastus lateralis) of human subjects (see Note 2).
2. Cloning medium: 1:1 mixture of -MEM and Hams F-12,
20 % FBS, 2 mM glutamine, 1 % chicken embryo extract,
100 U/mL penicillin, 100 g/mL streptomycin, 0.5 g/mL
amphotericin B.
3. Fusion medium: -MEM, 2 % FBS, 2 mM glutamine, 100 U/L
penicillin, 100 g/mL streptomycin, 0.5 g/mL
amphotericin B.
4. Fusion medium without FBS and phenol red: -MEM without
phenol red, 2 mM glutamine, 100 U/L penicillin, 100 g/mL
streptomycin, 0.5 g/mL amphotericin B.
5. Trypsin solution suitable for cell culture.
6. CO2 incubator (5 % CO2), 37 C.
7. Phosphate buffered saline (PBS) with and without Mg 2+
and Ca2+.
2.3 Electric Pulse

Stimulation
1. Power supply: C-Pace EP Culture Pacer (IonOptix, Dublin,

Ireland).
2. Electrodes: C-Dish (IonOptix, Dublin, Ireland).
2.4 Targeted
Proteomics
1. Multiplex bead-based immunoassay, e.g. Bio-Plex Pro Assay

(Bio-Rad, Hercules, CA), Milliplex Multiplex Assay (Merck
Millipore, Darmstadt, Germany).
2. Suitable suspension array system and software for the measurement and data analysis.
2.5 Untargeted
Proteomics
1. Falcon tubes (50 mL).

2. Protein concentrators (3 kDa cutoff).
3. Equipment for gel-based or gel-free protein separation technique and MS-based protein analysis [14].
58
3
3.1
Mika Scheler et al.
Methods
Cell Culture
1. Thaw, seed, and proliferate human skeletal muscle cells

( see Note 2) in Cloning medium on 150 cm2 plates. Change
medium every 23 days until cells are 8090 % confluent.
2. Trypsinate cells and seed at least 20,000 myoblasts per well on
6-well dishes.
3. When cells are 80 % confluent change medium and differentiate cells to multinucleated myotubes (Fig. 1) with Fusion
medium for 67 days. Change medium every 23 days.
3.2 Electric Pulse

Stimulation
1. Change medium directly before start of stimulation (see Note 3).

2. Insert electrodes, set culture dish including electrodes into
incubator and wire with power supply (Fig. 1; see Note 4).
3. Turn on power supply, adjust the proper program and start
stimulation. The intensity and duration of stimulation has to
be established for each cell type. We use 14 V, 5 Hz and 2 ms
in the basic mode for up to 24 h. The RNA expression of several secreted proteins is increased after less prolonged stimulation, but 24 h are needed to significantly increase glucose
consumption and lactate production [11]. Application of
higher frequencies (30 Hz) for 4 h leads to an increased phosphorylation of p70S6K1 and expression of MYOD, both
depicting high-intensity resistance training leading to increased
muscle protein synthesis and muscle hypertrophy [11]. For the
proteomics-based investigation of secreted proteins, it is necessary to minimize the contamination with proteins released
from damaged cells. The absence of EPS-induced cell damage
should be verified by the quantification of cytosolic or mitochondrial localized proteins in the supernatant, e.g. by creatine
kinase (CK) activity, also used as plasma marker for muscle
fiber damage [15] (see Note 5). The influence of the applied
EPS protocol on the cell viability might also be tested using,
e.g. the XTT assay. The EPS protocol of 14 V, 5 Hz, 2 ms for
24 h does not decrease the viability of human myotubes or
induce the release of CK (Fig. 2a, b; adapted from [11]).
Fig. 1 (continued) myotubes (immunostaining shows MyHC II, a fast-type skeletal muscle myosin in green;
nuclei are shown in blue (DAPI staining)). 2. After 67 days of differentiation myotubes are electro-stimulated. Therefore, the electrodes dip into the media in the cell culture dish and the whole sandwich is put into
the incubator. A cable connects the electrode (C-dish) with the power supply (C-Pace EP system) outside (after
IonOptix, Dublin, Ireland; detailed brochure: Culture Pacing System). 3. For targeted analysis of the muscle
secretome small amounts of media are sufficient whereas untargeted analysis needs larger amounts of
media which should be free of FBS and phenol red
59
1. Cell culture
Satellite cell
isolation
Myoblasts
Myotubes
Power supply
2. EPS
Incubator
3. Proteomics
Normal
Fusion media
Targeted immunoassay
Fusion media
phenolred
FCS
Untargeted approach
Separation
(gel-based vs. gel-free)
MS-based analysis
for protein identification
Fig. 1 Workflow showing secretome analysis of human myotubes. 1. Skeletal muscle cells are obtained from
percutaneous needle biopsies performed on the lateral portion of quadriceps femoris (vastus lateralis) of
human subjects. Satellite cells are prepared, myoblasts are cultivated and differentiated to multinucleated
Mika Scheler et al.
Absorbance 450nm
(fold change)
b 250
1.2
CK activity (U/L)
60
1
0.8
0.6
0.4
0.2
200
15
10
5
0
0
con
LDH activity (U/L)
con
EPS
EPS
SN
250
con
EPS
lysate
200
15
10
5
0
con
EPS
SN
con
EPS
lysate
Fig. 2 EPS does not induce cytotoxicity. Cell viability was measured by XTT cell viability assay (a). Measured
absorbance at 450 nm of supernatant is shown as fold change of EPS vs. control supernatant (means SEM;
n = 5). Creatine kinase (CK; b) and Lactate dehydrogenase (LDH; c) activities were measured in the supernatant (SN, filled bars) and in Triton X-100 lysates of the cells (striped bars). Values are shown in U/L (means SEM;
n = 58; p < 0.05) adapted from [11]
3.3 Collection
of Medium
for Targeted
Secretome Studies
Usually cell culture medium contains FBS to stimulate cell proliferation or in case of serum-reduced media like Fusion media to
maintain the viability of the cell. The high abundance of serum
proteins in the FBS requires additional enrichment of low-abundant
proteins, e.g. using Proteominer technology or metabolic labeling
with deuterated amino acids [16, 17]. An efficient and sensitive
way to analyze protein signatures in FBS containing supernatant is
via commercially available multiplex bead-based immunoassays
(see Notes 6 and 7). The following steps are necessary to collect
conditioned medium and prepare it for measuring with multiplex
bead-based immunoassays.
1. Directly after end of stimulation, take off medium and put on
ice until proceeding with the next step (see Note 8).
2. To remove detached cells, spin supernatant with 2,700 g for
4 min at 4 C.
3. Transfer supernatant in a new tube and store at 80 C until
used for further analysis.
61
4. When using a validated multiplex kit, proceed according to the

manufacturers instructions (see Notes 810).
5. Protein concentrations of the analytes can be calculated with
the appropriate optimized standard curves using the appropriate
software.
3.4 Collection
of Medium
for Untargeted
Proteomics
An untargeted proteomic profiling approach enables the discovery

of novel contraction-induced myokines and helps to identify specific
and sensitive biomarkers secreted from the muscle. Untargeted
approaches can either be performed by gel-based separation techniques, e.g. 1- and 2D electrophoresis and subsequent staining of
the gels or by gel-free separation via HPLC or CE. Proteins can be
visualized on gels by fluorescence labeling before electrophoresis
(e.g. DIGE) or by staining of the gels after electrophoresis (e.g.
silver staining). In all cases, subsequent protein analysis is usually
MS-based [14]. In the following part the sample preparation for
both separation techniques is described. For LC-MS/MS only
small samples sizes are needed whereas for 1- or 2D electrophoresis huge amounts of protein are necessary (see Note 11).
Nevertheless, latter application is the method of choice for visualization of the sample composition, its complexity and quality.
1. For untargeted profiling Fusion medium without FBS has to
be used during stimulation to prevent the contamination of
proteins included in the serum. To decrease any serum carryovers
wash cells 2 with PBS containing Ca2+/Mg2+ and 1 with
standard PBS (see Note 12).
2. Use 2 mL/well Fusion medium without FBS and phenol red
(see Note 13) and start stimulation according to Subheading 3.2.
3. Directly after stimulation end, take off supernatant and spin
down detached cells with 800 g for 5 min at 4 C to remove
any cell debris.
4. Remaining insoluble material is separated by a further centrifugation step (45,000 g, 5 min, 4 C).
5. Supernatant can be frozen at 80 C for further proceeding at
this point.
6. Either continue directly with the unfrozen supernatant or thaw
the supernatant on ice. Since the native protein concentration is
in the lower g/mL range, the supernatant has to be concentrated by spin concentration devices (cutoff 3 kDa) until a
protein concentration of 13 mg/mL is reached (see Note 14).
7. Concentrated supernatant is stored at 80 C until further
analysis via gel-based or gel-free protein separation technique
and subsequent MS-based protein analysis [14].
62
Mika Scheler et al.
Notes
1. Be sure to use 6-well dishes that fit with the C-dishes (electrodes).
The IonOptix handbook proposes 6-well dishes of several
companies. To avoid contaminations, heat the electrodes at
100 C for 3 h. After experiment soak electrodes in demineralized water for several days to neutralize from salts that might
have formed on the carbon electrodes and sterilize again.
2. Experiments are performed on the first and second passage of
subcultured cells.
3. We use 2 mL medium per well (6-well dish). Be sure that the
electrodes dip into the medium and do not contact the bottom
of the dish containing the cells.
4. Be careful that the cable is not kinked by the incubator door
since it can be damaged then.
5. Lactate dehydrogenase (LDH) is another marker for muscle
fiber damage [15]. After 24 h of EP-stimulation, LDH is significantly increased in the total cell lysates of EPS cells and also
in the supernatant. These data suggest that elevated levels of
cytosolic proteins in the secretome needs to be considered
carefully and do not necessarily indicate higher percentage of
damaged cells (Fig. 2c; adapted from [11]).
6. For the selection of candidate myokines it might be helpful to
do a whole genome transcriptome analysis or quantitative
Real-time PCR prior to secretome study to find EPS-regulated
proteins.
7. The multiplex assays have several advantages. Only a small
sample volume is necessary due to its high sensitivity.
Additionally, it is easy to handle, enables an exact quantification in a concerted system. The biggest advantage compared
to commercial available enzyme-linked immunosorbent assay
(ELISA) is the measurement of several proteins simultaneously
in one assay.
8. During immunoassay antibodies might cross-react with proteins
of FBS, hence a sample of unconditioned medium (reference
medium) is urgently needed for multiplex measurement. Spin
down the reference medium according to methods in step 2 in
Subheading 3.3 and store supernatant at 80 C until used for
further analysis.
9. When mixing individual singleplexes check x-Plex Bead
Regions and avoid using 2 singleplexes with similar x-Plex
Bead Regions in one assay.
10. For the conditioned medium of human myotubes, neat samples
were used (50 L sample/well). In our case, all measured ana-
63
lytes were in the assay working range. If you are above the
upper limit of quantification make sure to dilute the samples
properly.
11. For comprehensive analysis of the secretome via 2DE, a protein concentration up to mg/mL range is required. In the special case of DIGE technology at least 150 g protein is essential.
Thus, huge amounts of medium are necessary.
12. To get rid of the complex protein mixture included in FBS,
cells are washed 2 times with PBS containing Ca2+/Mg2+. This
is used to wash adherent cell cultures when one merely wishes
to wash and have the cells remain adherent because adhesion
proteins require divalent cations to function. The third washing step is done with PBS without Ca2+/Mg2+ to decrease salt
concentration in the medium.
13. For untargeted proteomics the medium should be free of phenol red because it might interfere with protein assays performed
to determine protein concentration for subsequent gels.
14. Since the native protein concentration is very low, it is essential
to concentrate the supernatant to achieve a working concentration for subsequent gels. It might be necessary to transfer
the supernatant during concentration process in an even
smaller protein concentrator tube to gain smaller volumes and
the required protein concentration. Be aware that with every
change of the device material gets lost. To minimize the loss,
rinse the membranes of the larger device with flow through
and add it to the eluent of the smaller device. Additionally, take
care of protein precipitations that might form.
Acknowledgements
This study was supported in part by a grant from the Leibniz
Gemeinschaft (SAW-FBN-2013-3) to C. Weigert and by a grant
from the German Federal Ministry of Education and Research
(BMBF) to the German Center for Diabetes research (DZD e.V.;
No. 01GI0925).
References
1. Pedersen BK (2009) The diseasome of physical
inactivityand the role of myokines in muscle
fat cross talk. J Physiol 587:55595568
2. Gleeson M (2007) Immune function in sport and
exercise. J Appl Physiol (1985) 103:693699
3. Goldstein MS (1961) Humoral nature of the
hypoglycemic factor of muscular work.
Diabetes 10:232234
4. Ostrowski K, Rohde T, Zacho M et al (1998)

Evidence that interleukin-6 is produced in
human skeletal muscle during prolonged running. J Physiol 508(Pt 3):949953
5. Febbraio MA, Hiscock N, Sacchetti M et al
(2004) Interleukin-6 is a novel factor mediating glucose homeostasis during skeletal muscle
contraction. Diabetes 53:16431648
64
Mika Scheler et al.
6. Pedersen BK, Febbraio MA (2012) Muscles,

exercise and obesity: skeletal muscle as a secretory organ. Nat Rev Endocrinol 8:457465
7. Weigert C, Lehmann R, Hartwig S et al (2014)
The secretome of the working human skeletal
musclea promising opportunity to combat
the metabolic disaster? Proteomics Clin Appl
8:518
8. Fujita H, Nedachi T, Kanzaki M (2007)
Accelerated de novo sarcomere assembly by
electric pulse stimulation in C2C12 myotubes.
Exp Cell Res 313:18531865
9. Lambernd S, Taube A, Schober A et al (2012)
Contractile activity of human skeletal muscle
cells prevents insulin resistance by inhibiting
pro-inflammatory
signalling
pathways.
Diabetologia 55:11281139
10. Manabe Y, Miyatake S, Takagi M et al (2012)
Characterization of an acute muscle contraction model using cultured C2C12 myotubes.
PLoS One 7:e52592
11. Scheler M, Irmler M, Lehr S et al (2013)
Cytokine response of primary human myotubes in an in vitro exercise model. Am J
Physiol Cell Physiol 305:C877C886
12. Nikolic N, Bakke SS, Kase ET et al (2012)
Electrical pulse stimulation of cultured
13.
14.
15.
16.
17.
human skeletal muscle cells as an in vitro

model of exercise. PLoS One 7:e33203
Raschke S, Eckardt K, Bjorklund Holven K
et al (2013) Identification and validation of
novel contraction-regulated myokines released
from primary human skeletal muscle cells.
PLoS One 8:e62008
Wohlbrand L, Trautwein K, Rabus R (2013)
Proteomic tools for environmental microbiologya roadmap from sample preparation to
protein identification and quantification.
Proteomics 13:27002730
Fernandez-Gonzalo R, Lundberg TR, AlvarezAlvarez L et al (2014) Muscle damage
responses and adaptations to eccentricoverload resistance exercise in men and
women. Eur J Appl Physiol 114(5):
10751084
Frobel J, Hartwig S, Passlack W et al (2010)
ProteoMiner and SELDI-TOF-MS: a robust
and highly reproducible combination for biomarker discovery from whole blood serum.
Arch Physiol Biochem 116:174180
Eichelbaum K, Winter M, Berriel Diaz M et al
(2012) Selective enrichment of newly synthesized proteins for quantitative secretome analysis. Nat Biotechnol 30:984990
Chapter 6
Urinary Pellet Sample Preparation for Shotgun
Proteomic Analysis of Microbial Infection
and HostPathogen Interactions
Abstract
Urine is one of the most important biofluids in clinical proteomics, and in the past decades many potential
disease biomarkers have been identified using mass spectrometry-based proteomics. Current studies mainly
perform analyses of the urine supernatant devoid of cells and cell debris, and the pellet (or sediment) fraction is discarded. However, the pellet fraction is biologically of interest. It may contain whole human cells
shed into the urine from anatomically proximal tissues and organs (e.g., kidney, prostate, bladder, urothelium, and genitals), disintegrated cells and cell aggregates derived from such tissues, viruses and microbial
organisms which colonize or infect the urogenital tract. Knowledge of the function, abundance, and tissue
of origin of such proteins can explain a pathological process, identify a microbe as the cause of urinary tract
infection, and measure the human immune response to the infection-associated pathogen(s). Successful
detection of microbial species in the urinary pellet via proteomics can serve as a clinical diagnostic alternative to traditional cell culture-based laboratory tests. Filter-aided sample preparation (FASP) has been
widely used in shotgun proteomics. The methodology presented here implements an effective lysis of cells
present in urinary pellets, solubilizes the majority of the proteins derived from microbial and human cells,
and generates enzymatic digestion-compatible protein mixtures using FASP followed by optimized desalting procedures to provide a peptide fraction for sensitive and comprehensive LC-MS/MS analysis. A highly
parallel sample preparation method in 96-well plates to allow scaling up such experiments is discussed as
well. Separating peptides by nano-LC in one dimension followed by online MS/MS analysis on a
Q-Exactive mass spectrometer, we have shown that more than 1,000 distinct microbial proteins and 1,000
distinct human proteins can be identified from a single experiment.
Key words Urine proteomics, Urinary pellet, Microbial infection, Hostpathogen interaction,
Filter-aided sample preparation (FASP), 96FASP
Abbreviations
FASP
HCD
LC
Filter-aided sample preparation

Higher energy collision disassociation
Liquid chromatography
65
66
MS
MWCO
TIC
UTI
Mass spectrometry
Molecular weight cut off
Total ion chromatogram
Urinary tract infection
Introduction
Urine is a sample source of high importance for biomarker discovery because it is easily available and collected non-invasively in
large quantities [1, 2]. The identity and quantity of proteins
excreted into urine may reflect pathological conditions that can be
traced to different organs in the body, particularly the kidneys,
prostate, and urogenital tract [3]. Currently, most urine proteomic
studies focus on the analysis of the urinary supernatant, that is the
soluble fraction of the collected urine sample following centrifugation at 1,5005,000 g for 515 min [4, 5]. The resulting urinary
pellet is frequently discarded. However, the urinary pellets, especially those from patients with urinary tract infection (UTI), which
is one of the most common conditions that lead to hospital visits
[6], contain not only pathogenic microbes, in most cases bacteria
that colonize the urinary tract of the patient, but also host proteins
associated with the inflammatory process following colonization
with the microbial pathogen. Inflammation may include activities
such as recognition of pathogen molecular patterns, cytokine
release, leukocyte recruitment, lymphocyte recruitment, complement activation, immunoglobulin secretion, fibrin deposition,
release of iron-sequestering proteins and direct microbial killing
via enzymatic activities and permeabilization or disintegration of
bacterial membranes [68]. The presence and relative quantity of
such proteins serve as a diagnostic indicator of infection and inflammation [7]. Non-pathogenic bacteria not inducing inflammatory
responses may also be identified from urinary pellets [9].
Our laboratory reported the first metaproteomic analysis of
urinary pellets derived from patients diagnosed with either asymptomatic bacteriuria or UTI, and identified the microbial causes of
bacteriuria [9]. Most recently, our laboratory developed a highthroughput urinary sample preparation approach, 96FASP (96well filter-aided sample preparation), for quantitative shotgun
proteomic analysis [10]. This method promises to be the prototype of an economical method for the diagnosis of urinary tract
infections and inflammation in the future. This article describes an
extensively used, robust step-by-step procedure pertaining to the
preparation of urinary pellet samples using the FASP approach.
The optional 96FASP method that is adapted to process multiple
samples simultaneously is described as well.
Urinary Pellet Proteomics
67
Materials
2.1 Cell Lysis

and FASP
1. Sartorius Vivacon 500 filter device (30,000 MWCO).

2. MultiScreen Ultracel-10 Filter Plate 10 kDa (Merck-Millipore,
USA).
3. Bench-top centrifuge (for example, Eppendorf 5415R or
equivalent).
4. (Optional) Plate centrifuge (for example, Eppendorf 5810R
or equivalent).
5. Misonix Sonicator 3000 Ultrasonic Cell Disruptor.
6. Pre-casted SDS PAGE gel (for example, NUPAGE 412 %).
7. SpeedVac (for example, Labconco Refrigated CentriVap
Concentrator, or equivalent).
8. TMN buffer: 40 mM TrisHCl, pH 8.1, 5 mM MgCl2 and
100 mM NaCl.
9. Lysostaphin solution: 10 mg/ml in water (AMBI Products;
from Staphylococcus simulans).
10. Mutanolysin solution: 2 mg/ml in water (Sigma-Aldrich;
from Streptococcus globisporus).
11. NaOH solution: 100 mM in water.
12. UA buffer: 8 M urea in 50 mM TrisHCl, pH 8.1. UA buffer
should be prepared freshly each day.
13. USED lysis buffer: 8 M Urea, 1 % SDS, 5 mM Na-EDTA,
50 mM DTT. USED buffer should be prepared freshly each day.
14. IAA solution: 0.05 M iodoacetamide in 50 mM TrisHCl,
pH 8.1. IAA solution should be prepared freshly each day.
15. ABC buffer: 50 mM ammonium bicarbonate in water.
16. Trypsin solution: trypsin (Promega sequencing grade); stock
concentration 0.1 g/l and stored at 80 C.
2.2 StageTip
Desalting
1. Empore C18 Extraction disks (3 M, catalog number: 2215).

2. Activation buffer: 100 % methanol.
3. Wash and equilibration buffer: 0.5 % acetic acid in H2O.
4. Elution solution-I: 0.5 % acetic acid, 60 % acetonitrile, and
40 % H2O.
5. Elution solution-II: 0.5 % acetic acid, 80 % acetonitrile, and
20 % H2O.
2.3
LC-MS/MS
1. XCalibur software (version 2.2, Thermo Scientific).

2. Proteome Discoverer (version 1.4, Thermo Scientific).
3. HPLC solvent A: 0.1 % formic acid in water.
4. HPLC solvent B: 0.1 % formic acid in acetonitrile.
68
5. Trap column: C18 PepMap100, 300 m 5 mm, 5 m, 100

(Thermo Scientific, USA).
6. Analytical column: PicoFrit, 75 m 10 cm, 5 m BetaBasic
C18, 150 (New Objective, USA).
7. Ultimate 3000 nano-LC system coupled to Q-Exactive mass
spectrometer (Thermo Scientific, USA).
Methods
An overview of the protein sample preparation for urinary pellets is
provided in Fig. 1, as explained in detail in the following procedures. The schematic also shows the downstream applications (e.g.,
LC-MS/MS and database search) for urinary proteome analysis.
3.1 Lysis of Urinary

Pellets and Collection
of Lysates
1. Take urine samples (see Note 1), centrifuge at 3,000 g for

15 min at 4 C, discard the urine supernatant and recover the
pellet, retaining ~1.0 ml of residual urine supernatant to avoid
disturbing the pellet.
2. Add ~10 ml ice-cold phosphate-buffered saline PBS, gently
shake the tube and centrifuge for another 15 min at 3,000 g.
Discard the supernatant and store the wet pellet in 80 C or
process immediately.
3. Add USED buffer (see Note 2) consistent with an approximate volume ratio of 4:1 (buffer volume/urinary pellet volume) to lysis cells and solubilize the contents in the urinary
pellet. If the urinary pellet volume is very small, a minimum of
100 l USED buffer volume can be added.
4. Vortex vigorously for 10 s a few times to resuspend the pellet;
pipette up and down to resuspend the pellet if the vortexing step
was not sufficient to homogenize the pellet. Incubate for 30 min
at room temperature and vortex or flick tube occasionally.
5. Using Misonex Sonicator (with the water bath attached, not
the probe), put the urinary lysate into an ice/water-filled water
bath. Set the program on amplitude 9 for nine 45-s cycles with
45-s intermittent cooling of the urinary lysate sample(s).
6. Let the urinary lysate sample(s) sit for approximately five more
minutes. Spin urinary lysate sample(s) in a bench-top centrifuge at maximum speed (14,000 g) for 10 min, and transfer
lysate supernatant to a new 1.7 ml microtube (see Note 3).
Freeze the urinary lysate at 80 C if necessary.
3.2 FASP Processing

Urinary Pellet Samples
for Shotgun
Proteomics
1. Before processing pellet lysate using FASP, flush the filters

with 200 l NaOH (100 mM) and 200 l UA buffer once
separately.
2. Aliquot a volume equivalent to 10100 g total protein quantity
into Vivacon 500 filter device, and mix with 200 l UA buffer.
69
Fig. 1 An overview of the urinary pellet sample preparation for shotgun proteomics. The procedures are explained
in detail in this chapter. Briefly, the urine samples are first spin down to collect pellet fraction. The pellets are
then lysed and digested following FASP protocol. A protocol of 96-well format FASP is also illustrated. Afterwards,
the protein digests are cleaned using StageTip and analyzed by nanoLC-MS/MS. The host and pathogen protein
identifications can be obtained by searching a metaproteome database which includes human proteins as well
as the proteins of common pathogens related to your study (in this case, the urinary tract infection)
3. Spin at 14,000 g for 1030 min.

4. Add 200 l UA buffer and repeat the spin at 14,000 g to
concentrate until the volume in the filter unit is reduced to
~10 l.
5. Add 100 l of the IAA solution (final concentration 50 mM)
and mix by vortexing filter unit for 1 min. Incubate without
mixing for 20 min in the dark at ambient temperature.
6. Spin at 14,000 g for 1030 min. Discard flow-through from
collection tube.
70
7. Add 200 l of UA buffer and spin at 14,000 g for 1030 min.

The final sample volume in the filter unit should be 20 l or
less.
8. Add 200 l of ABC buffer and spin at 14,000 g for 1015 min.
Add 200 l of ABC buffer and repeat spin step once more.
9. Add 100 l of ABC buffer to filter unit and trypsin so that the
final protein:trypsin ratio is 100:1. Mix by vortexing for 1 min.
Transfer the filter to a new filter collection tube.
10. Incubate at 37 C overnight in a temperature-controlled incubator (1216 h).
11. On the next day, add 200 l of ABC buffer, spin the filter unit
at 14,000 g for 1015 min and collect the filtrate (with the
peptide mixture) into a maximum recovery tube.
12. Add 100 l of ABC buffer to filter unit and trypsin so that the
final protein:trypsin ratio is 100:1, and re-incubate the tube at
37 C for another 46 h.
13. Add 200 l of ABC buffer, spin the filter unit after this second
digestion step at 14,000 g for 1015 min and collect the second filtrate (with the peptide mixture) into the same maximum recovery tube.
14. Add another 200 l of ABC buffer, spin the filter unit after the
wash step at 14,000 g for 1015 min and collect the wash
(with residual peptides) into the same maximum recovery
tube. The final volume in the collection tube is ~600 l.
15. Dry the peptide solution using a Speed-Vac (this may take
23 h). The tryptic peptide mixture is ready now for desalting
with the StageTip method.
3.3 (Optional)
96-Well Filter Plate
Processing Urinary
Pellet Samples
for Shotgun
Proteomics
1. Instead of using individual Vivacon filters to process urinary

pellet samples separately, one can use 96-well filter plate
to process multiple samples simultaneously. This could be a
good option when tens or hundreds of samples have to be
analyzed [10].
2. All the following procedures are the same as described above
for single filter processing; there are a couple of changes:
(a) Each spin takes place on plate centrifuged at 2,600 g for
4590 min.
(b) The collection plate should be polypropylene-based (e.g.,
to reduce possible non-specific bindings).
(c) The lid of the 96-well filter plate cannot seal the plate very
well; to avoid sample drying out after overnight incubation, parafilm could be used to warp and tightly seal the
lid. In the meantime, at least 100 l ABC buffer should be
added into each well for overnight digestion.
3.4 Peptide
Desalting Using
StageTip Protocol
71
1. This method is adapted from a published protocol [11]; several changes have been made to optimally fit the preparation
of urinary pellet samples (see Note 4).
2. Prepare StageTip by punching out small discs (13 layers
depending on the sample amount) of C18 Empore filter using
a 22 G flat-tipped syringe and ejecting the discs into P200
pipette tips. Ensure that the disc is securely wedged in the bottom of the tip.
3. Activate a tip by forcing 200 l methanol through the tip. Use
this step to check if the StageTip is leaky or overtight.
4. Force 200 l Elution Solution-II through the tip.
5. Equilibrate tip by forcing 200 l Wash and equilibration buffer through the tip. The tip is now ready for sample loading.
6. Resuspend the dried peptides into 100 l of Wash and equilibration buffer, and vortex for 10 min.
7. Load the 100 l peptide solutionmade in step [5] above
by forcing them through the C18-StageTip. Do not discard
the flow-through, but collect it into the original tube.
8. Reload the peptide solution onto the tip. This step may be
repeated two or three times.
9. Wash the tip with 200 l Wash buffer, and repeat this step one
or two times. The flow-through during this step can be
discarded.
10. Elute the peptides with 200 l Elution Solution-I (once) and
200 l Elution Solution-II (twice). Collect all of the eluates
(~600 l) into one maximum recovery microtube.
11. Dry the peptide eluates in the Speed-Vac (this may take 12 h).
12. Store then at 80 C, or resuspend it with LC solvent A for
immediate LC-MS/MS analysis.
3.5 Nano-LC-MS/MS
and Computational
Analysis
1. Resuspend the dried peptide samples into 2050 l LC solvent

A, centrifuge at maximal speed for 35 min, transfer to sample
vials and load to HPLC autosampler.
2. The LC-MS/MS analysis is operated by an Ultimate 3000
nano LC system and Q Exactive mass spectrometer. Around
25 l of the samples were first loaded onto a trap column at
high flow rate 5 l/min, and then separated on a PicoFrit analytical column at a nano flow rate of 300 nl/min. For a 2-h
LC-MS run, a linear gradient was applied from 100 % solvent
A to 35 % solvent B over 100 min, followed by a steeper gradient to 80 % solvent B over 15 min. The column was re-equilibrated with solvent A for 5 min. Eluting peptides were acquired
in data-depended mode using XCalibur software and top10
method as described before [10].
72
3. The acquired raw files are then processed using the Proteome
Discoverer software. The protein database involved in this
study contains UniProt human protein sequences and common urinary tract pathogens [12]. They can be downloaded
from UniProt website (http://www.uniprot.org/). MS search
parameters are similar to published previously [10].
4. Figure 2 shows a typical nanoLC-MS/MS analysis of the urinary pellet samples. The base peak of the TIC shows the eluted
peptides from HPLC column are acquired by mass spectrometer in 120 min. Shown in the upper panel is a representative
full MS scan collected at retention time 75.28 min. The mass
of each peak (peptide) are accurately measured by the Orbitrapbased mass analyzer at high resolution (e.g., 70,000), and
then sequenced in the HCD-based collision cell. The resulting
Fig. 2 A typical LC-MS analysis of the urinary pellet sample. The full mass (MS) and fragmentation (MS/MS)
are recorded by high resolution and high accuracy mass spectrometer. A representative MS scan at a given
time (for example, t = 75.3 min, as indicated by the red bar and arrows in the lower panel) with mass and
charge state information (for example, m/z = 668.3510, z = 2) is shown in the middle panel. Most of the ions in
this spectrum will be isolated and fragmentized afterwards. Then, the bioinformatics tool (for example, Sequest
algorithm) will assign the most confident amino acid sequence to each peak based on the sophisticated scoring system. Detailed explanations of the figure are provided in this chapter
73
fragments are measured again by the Orbitrap analyzer with

high resolution (e.g., 17,500). With high quality MS and MS/
MS data (mass errors are shown for each peak in the figure),
database search engine (e.g., Sequest) will assign each peak a
unique amino acid sequence and its corresponding protein, as
indicated by triangles (red, E.coli proteins; black, human
proteins) in the MS scan. The peak m/z = 846.0721 is magnified in the right window. The cysteine in protein TRFL has a
carbamidomethyl modification. The gray diamond ( ) shows
a keratin peptide (VDALMDEINFMK, = 2.74 ppm;
[K2C5_HUMAN]). In this single LCMS analysis, 2,198 proteins, including human (1,139) and bacterial proteins (1,059),
were successfully identified.
Notes
1. Usually 550 ml of urine is collected from human subjects. The
preferred approach is to process the urine sample immediately
after collection by centrifugation (see step 2). It is possible to
store the urine at 4 C for up to 6 h before centrifugation.
Finally, the entire urine sample may be stored at 80 C at the
clinical site, shipped to the site of proteomic analysis and thawed
prior to centrifugal separation of urinary supernatant and urinary pellet. The freezethaw step may alter the composition of
the urinary pellet. For a given project with multiple samples, a
distinct urine storage and processing method should be selected.
2. In the case that gram-positive bacteria with thick cell walls,
such as Streptococcus pneumoniae and Staphylococcus aureus,
are suspected to be present in urine samples, resuspend the
urinary pellet samples in a volume of TMN buffer to have an
approximate volume ratio of 1:10, usually less than 200 l.
Pipette the suspension up and down a few times in a 1.5 ml
tube. Add lysostaphin and mutanolysin to a final concentration of 20 g/ml. Mix gently to homogenize the enzymes and
suspended cells. Incubate the sample in the 37 C shaker-incubator. Take out the tube every 30 min and check if a pellet
collects at the bottom. If so, briefly vortex every 30 min.
Complete digestion after a 3 h incubation.
After pre-treatment with lysostaphin and mutanolysin,
add up to 600 l USED buffer into the lysate.
3. To estimate the protein concentration, take 10 l aliquot of
the supernatant to another new microtube, mix with SDS
loading buffer and run it in a SDS-PAGE gel. Load 2 and 5 g
BSA standards in the same gel. Coomassie Blue (CB)-G250
stain the gel followed by its destaining with standard procedures [13]. From the overall CB-G250 staining intensity of
urinary pellet lysate bands, estimate the total protein amount
74
in the lane based on BSA staining intensities. The protein concentration could also be measured by tryptophan fluorescence
as reported before [14].
4. Instead of using syringe to manually push solvents through
the StageTips, one can use pipette tip adaptors (commercially
available from The Nest Group, MA) which fit the 1.5- or 2.0mL microtubes well. This way, all the processing steps with
syringe can now be done using a bench-top centrifuge [15].
Acknowledgments
This work was supported in part by Grant NIH-1R01GM103598
(National Institute of General Medical Sciences).
References
1. Decramer S, de Peredo AG, Breuil B, Mischak
H, Monsarrat B, Bascands J-L, Schanstra JP
(2008) Urine in clinical proteomics. Mol Cell
2. Wood SL, Knowles MA, Thompson D, Selby
PJ, Banks RE (2013) Proteomic studies of urinary biomarkers for prostate, bladder and kidney cancers. Nat Rev Urol 10:206218
3. Barratt J, Topham P (2007) Urine proteomics:
the present and future of measuring urinary
protein components in disease. Can Med
Assoc J 177:361368
4. Adachi J, Kumar C, Zhang Y, Olsen J, Mann
M (2006) The human urinary proteome contains more than 1500 proteins, including a
large proportion of membrane proteins.
Genome Biol 7:R80
5. Rodrguez-Surez E, Siwy J, Zrbig P, Mischak
H (2014) Urine as a source for clinical proteome analysis: from discovery to clinical application. Biochim Biophys Acta 1844:884898
6. Nielubowicz GR, Mobley HLT (2010) Hostpathogen interactions in urinary tract infection. Nat Rev Urol 7:430441
7. Pieper R, Suh M, Fouts D, Nelson K (2012)
Metaproteomic method for diagnosis of bacteriuria, urogenital tract and kidney infections
from urinary pellet samples. US Patent App.
13/728,106
8. Weichhart T, Haidinger M, Hrl WH, Semann
MD (2008) Current concepts of molecular
defence mechanisms operative during urinary
tract infection. Eur J Clin Invest 38:2938
9. Fouts D, Pieper R, Szpakowski S, Pohl H,

Knoblach S, Suh M-J, Huang S-T, Ljungberg
I, Sprague B, Lucas S, Torralba M, Nelson K,
Groah S (2012) Integrated next-generation
sequencing of 16S rDNA and metaproteomics
differentiate the healthy urine microbiome
from asymptomatic bacteriuria in neuropathic
bladder associated with spinal cord injury. J
Transl Med 10:174
10. Yu Y, Suh M-J, Sikorski P, Kwon K, Nelson
KE, Pieper R (2014) Urine sample preparation
in 96-well filter plates for quantitative clinical
proteomics. Anal Chem 86:54705477
11. Rappsilber J, Mann M, Ishihama Y (2007)
Protocol for micro-purification, enrichment,
pre-fractionation and storage of peptides for
proteomics using StageTips. Nat Protoc
2:18961906
12. Foxman B (2010) The epidemiology of urinary tract infection. Nat Rev Urol 7:653660
13. Shevchenko A, Tomas H, Havlis J, Olsen JV,
Mann M (2007) In-gel digestion for mass
spectrometric characterization of proteins and
proteomes. Nat Protoc 1:28562860
14. Winiewski JR, Du K, Mann M (2013)
Proteomic workflow for analysis of archival
formalin-fixed and paraffin-embedded clinical
samples to a depth of 10 000 proteins.
Proteomics Clin Appl 7:225233
15. Yu Y, Smith M, Pieper R (2014) A spinnable
and automatable StageTip for high throughput peptide desalting and proteomics. Protoc
Exchange. doi:10.1038/protex.2014.033
Chapter 7
A Protocol for the Parallel Isolation of Intact Mitochondria
from Rat Liver, Kidney, Heart, and Brain
Sabine Schulz, Josef Lichtmannegger, Sabine Schmitt,
Christin Leitzinger, Carola Eberhagen, Claudia Einer,
Julian Kerth, Michaela Aichler, and Hans Zischka
Abstract
Mitochondria are key organelles for cellular energy production and cell death decisions. Consequently, a
plethora of conditions which are toxic to cells are known to directly attack these organelles. However,
mitochondria originating from different tissues differ in their sensitivity to toxic insults. Thus, in order to
predict the potential organ-specific toxicity of a given drug or pathological condition at the mitochondrial
level, test settings are needed that directly compare the responses and vulnerabilities of mitochondria from
different organs. As a prerequisite for such test strategies, we provide here a robust, prompt, and easy-tofollow step-by-step protocol to simultaneously isolate functional and intact mitochondria from rat liver,
kidney, heart, and brain. This isolation procedure ensures mitochondrial preparations of comparable purity
and reproducible quantities which can be subsequently analyzed for organ-specific mitochondrial toxicity.
Key words Mitochondria, Liver, Heart, Brain, Kidney
Introduction
The prime role of mitochondria in cell metabolism reflects the fact
that these organelles are crucial in controlling the cells fate, i.e.
survival or death, most importantly by balancing the cellular energy
demands. However, metabolic differences and metabolite preferences exist in the different healthy tissues of our body [1]. Whereas
brain tissue relies on glucose as the major metabolite, liver, especially in the postresorption phase, relies on fatty acids [1].
Consistent with these metabolic preferences, marked differences in
the molecular composition of the respective mitochondrial populations are known [24]. Thus, mitochondria delicately adapt at the
molecular level to the metabolic state of their tissue origin.
75
76
Sabine Schulz et al.
Consequently, upon exposure to toxic insults that target

mitochondria, these organelles may respond differently and in an
organ-specific fashion. For example, succinate is an important substrate for mitochondrial ROS production in brain, heart, kidney,
and skeletal muscle mitochondria whereas fatty acids generate significant quantities of oxidants in kidney and liver mitochondria [5].
Moreover, liver and brain mitochondria dramatically differ in their
response toward calcium challenges [6] and, importantly, were
shown to significantly differ in their sensitivity toward elicit of the
mitochondrial permeability transition, a cell death causing event
[7]. Thus, severe tissue-dependent differences in the mitochondrial sensitivities toward detrimental impacts exist. In order to
investigate such organ-specific mitochondrial differences, test
strategies are needed that minimize inter-experimental variability.
Preferably, mitochondria to be compared should be isolated in parallel on the same day, under appropriate conditions each, and from
the same animal in order to subsequently and simultaneously analyze them for their individual susceptibilities.
Here we provide a protocol to isolate mitochondria retaining
their metabolic functionality, from the same animal simultaneously
and from at least four different rat tissues, namely liver, kidney,
heart, and brain. This protocol is a combination of classical isolation procedures for mitochondria from individual sources based on
differential centrifugation, introduced by Palades group [8] but
containing several adaptations according to published methods
[9]. As these crude mitochondrial-enriched fractions still contain other relevant cellular compartments, particularly peroxisomes
and lysosomes, they are subjected to further Percoll gradient
centrifugation resulting in significant purification [10]. Such purified mitochondria are comparably homogenous but retain their
inherent structural peculiarities depending on their tissue origin
(Figs. 1 and 2). Importantly, these isolated mitochondria present
with intact membrane structures as they do not only efficiently
build up an inner transmembrane potential upon substrate addition (e.g. succinate), but also retain this membrane potential for at
least 60 min, highly indicative of preserved functional integrity
(Fig. 3). Thus, the mitochondrial populations isolated from the
same animal appear similar in sample purity and functional integrity, being therefore highly suitable for subsequent differential sensitivity testing. In total, the isolation procedure takes about 2.5 h
(see Note 1) and the tissue-dependent mitochondrial average yield
is given in Table 1. This protocol may be adapted to isolations of
mitochondria from other tissues, e.g. pancreas or spleen for comparative purposes (see Note 2).
Isolation of Intact Mitochondria from Rat Liver, Kidney, Heart, and Brain
77
Fig. 1 Electron micrographs of liver, kidney, heart, and brain mitochondria in situ (Scale bars equal 2 m for 1,600
magnification, and 500 nm for 10,000 and 20,000 magnifications, respectively). Hepatocyte mitochondria
appear round or elongated in shape, with an electron-dense matrix, mostly tubular cristae and dense mitochondrial granules. Kidney tubule cell mitochondria contain densely packed elongated mitochondria oriented in right
angles to the base of the cell. Myocardial muscle cell mitochondria are round or elongated in shape comprising
numerous, closely arranged cristae. Myelinated axon mitochondria appear with lamellar and tubular cristae
78
Fig. 2 Electron micrographs of liver, kidney, heart, and brain mitochondria isolated in parallel by the herein
presented protocol from the same animal (scale bars equal 5 m for 5,000, 1 m for 20,000, and 500 nm
for 50,000 magnifications). Homogeneous mitochondrial populations are obtained. However, as observed in
situ, depending on their tissue origin isolated mitochondria appear with significant structural differences
Materials
2.1 Isolation
Procedure
Components
1. Isofluran anesthesia facility (Sigma Delta Vaporizer, Penlon,

Abingdon, UK) and Isofluran (IsoFlo, Ecuphar, Greifswald,
Germany).
2. Teflon-glass homogenizer (B. Braun Biotech Homogenisator
Potter S, Melsungen, Germany) with manual hub and compatible pestles for 30 and 15 mL volumes.
Fig. 3 A time stable mitochondrial membrane potential in freshly isolated mitochondria confirms the functional integrity of their inner membranes. Assessment of the
mitochondrial membrane potential was done using the fluorescent dye Rhodamine
123 according to published protocols [13]. (a) Due to accumulation within mitochondria and fluorescence quenching of the positively charged dye upon an intact,
negative inside membrane potential, a low fluorescence is indicative for an existing
membrane potential. Upon addition of the protonophor FCCP as internal control,
dissipation of the mitochondrial membrane potential occurs, which can be followed
by a steep increase in fluorescence. (b) As a semi-quantitative measure for the time
stable mitochondrial integrity the fluorescence ratio before and immediately after
FCCP addition is calculated using the time points t = 0 (first enlarged light grey
diamond) and t = 60 (second enlarged grey diamond) minutes
Table 1
Average tissue-dependent yield of mitochondria given
as mg mitochondrial protein per g tissue wet weight [14]
Tissue
Mitochondria [mg/g w.w.]
Animals
Liver
17.7 5.1
11
Kidney
3.9 1.5
11
Heart
1.2 0.6
11
Brain
0.4 0.3
11
80
3. Rough glass-in-glass homogenizer 5 mL volume

(Homogenisator LAT, Reiss Laborbedarf e.K., MainzMombach, Germany).
4. Petri dishes, 60 mm diameter (Nunclon, Sigma-Aldrich,
Taufkirchen, Germany).
5. Blade, 39 mm cutting width (NeoLab, Heidelberg, Germany).
6. Centrifugation tubes: 50 mL (Polypropylene, Beckman
Coulter, Krefeld, Germany) and 15 mL (transparent, preferably
glass, Corex, DuPont Instruments, Neu-Isenburg, Germany).
7. Percoll solution density 1.130 mg/mL (GE Healthcare,
Munich, Germany).
8. Prepare all solutions using ultrapure water (Prepared by purifying deionized water to attain a resistivity of 18.2 M-cm
at 25 C).
2.2 Isolation
Procedure: Buffers
and Solutions
1. 0.9 % NaCl: Weigh 0.9 g NaCl into a 100 mL graduated flask,

add water up to 100 mL and let stir at room temperature until
NaCl is solved.
2. Isolation buffer with BSA: 0.3 M sucrose, 5 mM TES, 0.2 mM
EGTA, 0.1 % BSA (w/v), pH 7.2. Weigh 102.8 g sucrose,
1.146 g TES, 76 mg EGTA and 1 g BSA (fatty acid free) and
transfer to a 1 L glass beaker. Add ultrapure water to a volume
of ~800 mL and let stir for 30 min at RT to solve all chemicals
(see Note 3). Adjust the pH diligently to 7.2 with 5 M KOH
at RT (see Note 4). Transfer to a 1 L graduated flask and add
water up to 1 L and mix thoroughly. Store isolation buffer
with BSA at 4 C and use within 1 week.
3. Isolation buffer without BSA: 0.3 M sucrose, 5 mM TES,
0.2 mM EGTA, pH 7.2. Weigh 102.8 g sucrose, 1.146 g TES,
and 76 mg EGTA and transfer to a 1 L glass beaker. Add ultrapure water to a volume of ~800 mL and let stir for 30 min at
RT to solve all chemicals (see Note 3). Adjust the pH diligently to 7.2 with 5 M KOH at RT (see Note 4). Transfer to
a 1 L graduated flask and add water up to 1 L and mix thoroughly. Store isolation buffer without BSA at 4 C and use
within 1 week.
4. Isolation buffer for Percoll gradients (IPP buffer): 0.3 M
sucrose, 10 mM TES, 0.2 mM EGTA, 0.1 % BSA (w/v),
pH 6.9. Weigh 51.4 g sucrose, 1.146 g TES, 38 mg EGTA
and 0.5 g BSA (fatty acid free) and transfer to a 1 L glass beaker. Add ultrapure water to a volume of ~400 mL and let stir
for 30 min at RT to solve all chemicals (see Note 3). Adjust the
pH diligently to 6.9 with 5 M KOH at RT (see Note 4).
Transfer to a 500 mL graduated flask and add water up to
500 mL and mix thoroughly. Store IPP buffer at 4 C and use
within 1 week.
81
5. Prepare gradient solutions AC as follows:

Gradient solution A: supplement 60 mL of IPP buffer with
1.43 g sucrose.
Gradient solution B: supplement 60 mL of IPP buffer with
2.65 g sucrose.
Gradient solution C: supplement 60 mL of IPP buffer with
9.24 g sucrose.
Gradient solutions AC can be stored at 4 C for 1 week.
Methods
3.1 Preparation
of Percoll Gradients
1. Prepare gradient layer solutions with 18, 30, and 60 % Percoll

as follows (see Note 5):
18 % gradient layer solution: supplement 24.6 mL of gradient
solution A with 5.4 mL Percoll.
30 % gradient layer solution: supplement 21 mL of gradient
solution B with 9 mL Percoll.
60 % gradient layer solution: supplement 12 mL of gradient
solution C with 18 mL Percoll.
2. Using a Peleus ball (see Note 6), pour 4 mL of the 60 % gradient layer solution into a centrifuge tube. Carefully overlay
4 mL of the 30 % gradient layer solution. Rest two 30/60 %
gradients on ice, avoiding jolts.
3. Overlay the remaining 30/60 % gradients with 4 mL of the
18 % gradient layer solution, using a Peleus ball (see Note 6).
Rest the 18/30/60 % gradients on ice, avoiding jolts.
3.2 Tissue
Harvesting
1. Rats are sacrificed according to ethical guidelines and national

legislation [11].
2. Immediately rinse the organs with 10 mL ice cold 0.9 % NaCl,
injecting the solution with a syringe into the portal vein (vena
portae).
3. Remove liver, kidneys, heart, and brain placing them into ice
cold isolation buffer with 0.1 % BSA. The heart should be
placed into a small Petri dish filled with ice cold isolation buffer with 0.1 % BSA.
4. Liberate each organ from remaining fat or surrounding tissue,
leaving the organs intact and rinse free from blood with ice
cold isolation buffer with 0.1 % BSA.
5. Incise the heart to open the cardiac chambers and remove
clotted blood if necessary.
6. Weigh liver, kidneys, heart, and brain separately (see Note 7).
82
3.3 Homogenization
Procedure (See Note 1)
3.3.1 Liver, Kidney,
and Brain
1. Mince liver, kidneys, and brain with scissors.

2. Transfer minced liver into a 30 mL Teflon-glass homogenizer
and minced kidney and brain separately into a 15 mL Teflonglass homogenizer and fill up the homogenizer with ice cold
isolation buffer with 0.1 % BSA to the maximum.
3. Homogenize the tissue with 510 strokes at 800 rpm using
the Homogenisator Potter S (see Note 8).
4. Transfer homogenized tissue to centrifuge tubes and dilute
with ice cold isolation buffer with 0.1 % BSA using a total volume of ~160 mL for liver (4 50 mL centrifugation tubes),
~40 mL for kidney (1 50 mL centrifugation tube), and
~28 mL for brain (2 15 mL centrifugation tubes).
5. Seal the centrifugation tubes with Parafilm and invert.
3.3.2 Heart (See Note 9)
1. Cut the heart in ~500 mg pieces (23 pieces) and wash out
residual blood if necessary.
2. Remove the cardiac fibrous skeleton carefully.
3. Transfer 1 500 mg piece into a 60 mm Petri dish with 2 mL
of ice cold isolation buffer with 0.1 % BSA.
4. Mince 500 mg heart tissue with scissors to obtain ~1 mm tissue pieces.
5. Mince further using a blade to obtain 0.30.5 mm pieces.
6. Transfer 0.30.5 mm heart tissue pieces into a 5 mL rough
glass-in-glass homogenizer and fill up with ice cold isolation
buffer with 0.1 % BSA to 5 mL.
7. Homogenize heart tissue with five strokes by hand, carefully
rotating the pestle during each stroke. Transfer homogenized
heart tissue to a centrifuge tube.
8. Repeat steps 37 until heart tissue is homogenized.
9. Dilute with ice cold isolation buffer with 0.1 % BSA using a
total volume of ~30 mL (2 15 mL centrifugation tubes).
10. Seal the centrifugation tubes with Parafilm and invert.
3.4 Centrifugation
Steps (See Note 10)
1. Centrifuge liver, kidney, brain, and heart at 800 g for 10 min

at 4 C to free homogenates from cell debris and nuclei.
2. Transfer the supernatants to fresh centrifugation tubes.
3. Repeat steps 1 and 2 once.
4. Centrifuge liver, kidney, and heart homogenates at 9,000 g
for 10 min at 4 C and discard the supernatants.
5. Centrifuge brain homogenate at 20,000 g for 10 min at 4 C
and discard the supernatant (see Note 11).
6. The pellets obtained by steps 4 and 5 comprise a crude mitochondrial fraction. Resuspend this crude liver mitochondria
pellet in 46 mL and the crude kidney mitochondria pellet in
83
1 mL. Overlay 1 mL of each of the resuspended pellets on the

18 % phase of a 18/30/60 % Percoll gradient, using 46 gradients for the liver (see Note 12), and 1 gradient for the kidney.
To apply Percoll gradient centrifugation to crude heart and
brain mitochondria, resuspend each of the crude mitochondria
pellets in 4 mL of 18 % gradient layer solution and overlay on
the 30 % phase of a 30/60 % Percoll gradient.
7. Centrifuge the liver and kidney Percoll gradients at 9,000 g
for 10 min at 4 C and the brain and heart Percoll gradients
at 29,000 g for 10 min at 4 C.
8. Harvest the mitochondrial fraction at the boundary of the
30/60 % Percoll phases using a Pasteur pipette (see Note 13
and cf. Fig. 4). Transfer the mitochondrial fractions to a fresh
centrifugation tube, using two tubes with a maximum capacity
of ~50 mL for liver and 1 tube each for kidney, heart, and
brain with a maximum capacity of 15 mL.
9. Fill all centrifugation tubes with ice cold isolation buffer without BSA to maximum capacity and centrifuge liver, kidney,
heart, and brain Percoll gradient derived fractions at
9,000 g for 10 min at 4 C.
10. Discard the supernatants, resuspend the pellets and refill centrifugation tubes to maximum capacity with ice cold isolation
buffer without BSA. Seal the centrifugation tubes with
Parafilm and centrifuge liver, kidney, heart, and brain
Percoll gradient derived fractions at 9,000 g for 10 min at
4 C (see Note 14).
Fig. 4 Density gradient purification of mitochondrial populations using 18/30/60 % Percoll steps according to
Subheading 3.4. The arrow indicates mitochondria gathering at the 30/60 % Percoll interphase. Liver and
kidney mitochondria appear as prominent bands despite turbid 18 and 30 % Percoll phases. Heart and brain
mitochondria gather in a faint but visible band. For the latter Percoll gradients appear with whitely clouded more
intense 18 and 30 % phases than the actual mitochondria band at the 30/60 % Percoll interphase
84
11. Discard the supernatant and resuspend the mitochondrial

pellet in ice cold isolation buffer without BSA using ~3 mL for
liver, ~500 L for kidney, and 50300 L for heart and brain
(see Note 15). Keep the isolated mitochondria on ice for further investigation (see Note 16) or freeze at 80 to 178 C
(see Note 17).
Notes
1. The isolation procedure is ideally performed by two persons,
particularly at the homogenization steps. Recommended
apportionment: person A homogenizes liver, kidney, and
brain, person B takes responsibility for the homogenization of
the heart. This is because heart muscle-tissue is homogenized
with one additional mincing step and thereafter manually
homogenized. The heart homogenization procedure takes
approximately the same amount of time as the homogenization of liver, kidney, and brain together.
2. Be aware that mitochondria isolation from pancreas or spleen
may require supplementation with protease inhibitors during
tissue homogenization in order to retain mitochondrial
functionality.
3. EGTA dissolves only slowly, causing the pH to fall to ~3. Since
the supplied amount of EGTA is relatively small, care must be
taken to ensure complete solubilization by visual inspection.
4. Do not re-adjust the pH once the intended pH is exceeded.
5. The indicated volumes are sufficient for up to seven Percoll
gradients. For the liver 46 Percoll gradients are needed (see
Note 12), and 1 each for kidney, heart, and brain. Prepare
additional gradients if necessary. For heart and brain Percoll
gradients prepare only a two-phase 30/60 % Percoll
gradient.
6. The use of a Peleus ball is critical as the flow rate of a commonuse electric pipettor causes the Percoll-layers to mix.
Alternatively, a syringe can be used. Do not aim to empty the
pipette (or syringe) when releasing 4 mL, as final emptying
may also destroy the Percoll layers.
7. Weighing of organs is recommended to estimate the required
number of Percoll gradients for the liver (see Note 12).
Further, the organ weight is needed to calculate the obtained
mitochondrial protein per mg organ wet weight (cf. Table 1).
8. Tissue homogenization is critical, in the sense of liberating a
maximal amount of mitochondria from the cells without
destroying them immediately thereafter due to excessive
homogenization. Therefore, perform a minimum of strokes,
85
avoiding strong pressure from the pestle to occur onto the

homogenizator bottom. Liver, kidney, and brain tissues are
smooth and in our experience, 510 strokes lead to a homogenous result in all occasions. Liver homogenate is more brownish and dense than kidney homogenate, which is a little more
ginger in color. Brain homogenate is whitely beige. Observation
of more reddish homogenates point to remaining blood contamination, bearing the risk of oxidative stress and is likely to
worsen the functionality of isolated mitochondria.
9. Tissue homogenization is critical (see Note 8). To sufficiently
homogenize the tough heart tissue, ensure the removal of the
complete cardiac skeleton, as this fibrous tissue impedes mincing and fails homogenization. The mincing steps must be carried out thoroughly as heart tissue pieces greater than 0.5 mm
in size also fail homogenization. During homogenization twist
the pestle back and forth, avoiding strong pressure from the
pestle to occur onto the homogenizator bottom. Perform a
maximum of five strokes, even if some tissue pieces remain. The
homogenization works best if 500 mg tissue is not exceeded.
10. For an optimal centrifugation procedure, two centrifuges are
required.
11. Be careful when removing the supernatant, as the brain pellet
is labile.
12. Use one Percoll gradient for ~2.5 g of liver wet weight.
Typically 46 Percoll gradients are used for the liver.
13. A mitochondrial band is formed at the 30/60 % Percoll
interphase. For liver and kidney, these bands are prominent,
yet heart and brain mitochondrial bands appear only faint. To
harvest the mitochondrial band, dip a Pasteur pipette (3 mL
volume) just above the phase boundary, carefully aspirating
the interphase. Immediately stop aspiration the moment the
first band vanishes. Do not continue aspiration for additional
overlying bands, even if they may appear denser.
14. Be careful when removing the supernatant, as the pellet may
be labile. In this case Percoll still resides in the mitochondria
pellet, making an additional washing step necessary. Decant as
much supernatant as possible without losing the pellet and
refill the centrifugation tube with ice cold isolation buffer
without BSA to maximum capacity and repeat the centrifugation at 9,000 g for 10 min at 4 C. Now the brain mitochondria pellet should be sufficiently stabile to remove the
supernatant completely, without pellet detachment.
15. Resuspension with the specified volumes will lead to a concentration of ~50 g/L liver mitochondria, 1020 g/L kidney mitochondria, 38 g/L heart mitochondria, and
12 g/L brain mitochondria. Do not vortex mitochondria.
86
16. Isolated mitochondria remain functional intact for at least

60 min (cf. Fig. 3).
17. Depending on the subsequent experiments, isolated mitochondria may either be frozen at 80 C or in liquid nitrogen
according to Fleischer [12], who has reported that the latter
condition preserves enzymatic activity.
Acknowledgements
We would like to acknowledge E.E. Rojo for critical reading of the
manuscript. This study was supported in parts by the Deutsche
Forschungsgemeinschaft (DFG) grant RU742/6-1 to H.Z.
References
1. Lffler G, Petrides PE (1990) Biochemie und
pathobiochemie, 5th edn. Springer, Berlin
2. Mootha VK et al (2003) Integrated analysis of
protein composition, tissue diversity, and gene
regulation in mouse mitochondria. Cell
115(5):629640
3. Veltri KL, Espiritu M, Singh G (1990) Distinct
genomic copy number in mitochondria of different mammalian organs. J Cell Physiol
143(1):160164
4. Vijayasarathy C et al (1998) Variations in the
subunit content and catalytic activity of the
cytochrome c oxidase complex from different
tissues and different cardiac compartments.
Biochim Biophys Acta 1371(1):7182
5. Tahara EB, Navarete FD, Kowaltowski AJ
(2009) Tissue-, substrate-, and site-specific
characteristics of mitochondrial reactive oxygen species generation. Free Radic Biol Med
46(9):12831297
6. Andreyev A, Fiskum G (1999) Calcium
induced release of mitochondrial cytochrome c by different mechanisms selective
for brain versus liver. Cell Death Differ 6(9):
825832
7. Berman SB, Watkins SC, Hastings TG (2000)
Quantitative biochemical and ultrastructural
comparison of mitochondrial permeability
transition in isolated brain and liver mitochondria: evidence for reduced sensitivity
8.
9.
10.
11.
12.
13.
of brain mitochondria. Exp Neurol 164(2):

415425
Hogeboom GH, Schneider WC, Pallade GE
(1948) Cytochemical studies of mammalian
tissues; isolation of intact mitochondria from
rat liver; some biochemical properties of mitochondria and submicroscopic particulate material. J Biol Chem 172(2):619635
Pallotti F, Lenaz G (2007) Isolation and subfractionation of mitochondria from animal
cells and tissue culture lines. Methods Cell Biol
80:344
Petit PX et al (1998) Disruption of the outer
mitochondrial membrane as a result of large
amplitude swelling: the impact of irreversible
permeability transition. FEBS Lett 426(1):
111116
Close B et al (1997) Recommendations for
euthanasia of experimental animals: part 2.
DGXT of the European Commission. Lab
Anim 31(1):132
Fleischer S (1979) Long-term storage of mitochondria to preserve energy-linked functions.
Methods Enzymol 55:2832
Zamzami N, Metivier D, Kroemer G (2000)
Quantitation of mitochondrial transmembrane
potential in cells and in isolated mitochondria.
14. Schmitt S et al (2014) Mitochondrion 19(Pt A):

113-123. doi:10.1016/j.mito.2014.06.005.
Chapter 8
Isolation of Mitochondria from Cultured Cells and Liver
Tissue Biopsies for Molecular and Biochemical Analyses
Sabine Schmitt, Carola Eberhagen, Susanne Weber,
Michaela Aichler, and Hans Zischka
Abstract
We recently reported a new method to isolate functionally intact mitochondria from cell culture and small
tissue samples (Schmitt et al., Anal Biochem 443(1):6674, 2013). This method comprises a semiautomated cell rupture, termed pump controlled cell rupture system (PCC), which can be precisely adjusted
to the specific cellular source of isolation and which can be tightly controlled (Schmitt et al., Anal Biochem
443(1):6674, 2013). Here we provide a detailed hands-on protocol of this PCC method which results in
an efficient cell breakage but preserving the mitochondrial integrity. Upon subsequent purification steps,
the obtained mitochondrial fraction meets the quality and purity required for molecular analyses, e.g. proteomic comparisons, as well as for biochemical analyses, e.g. determination of diverse enzymatic activities.
Key words Mitochondria, Cell culture, Biopsies, Balch homogenizer
Introduction
Mitochondria are the cellular powerhouses and key integrators of
cell death decisions [1]. Consequently, conditions that impair mitochondria can lead to severe human diseases. A prime example may
be the intoxication of hepatocyte mitochondria by genetically caused
excessive copper burdens in Wilson disease [2, 3]. While augmented
mitochondria-dependent cell death is a major obstacle in neurodegenerative disorders [4, 5], avoidance of cell death is a hallmark of
cancer [6]. Consequently, the identification of specific mitochondrial targets to either stabilize or impair their structure or function is
a central aim in biomedical research [7]. Typically, the identification
of such targets is achieved by comparing mitochondria isolated from
healthy controls to mitochondria from diseased tissues, by proteomics, immuno-blotting or enzymatic measurements, etc. Such
analyses may identify molecular and functional mitochondrial alterations that can assist in understanding disease progression [8], or in
determining new targets for specific therapeutic intervention [9], or
87
88
Sabine Schmitt et al.
even provide important insights why conventional therapeutic interventions may help in some patients but fail in others [10].
Such comparative mitochondrial investigations of normal versus pathological situations require suspensions of isolated mitochondria with high and comparable purity. A most critical step in
the procedure to isolate mitochondria is the initial step of cell rupture. Especially if the starting material is limited, e.g. in cases of
cultured cells or tissue biopsies, an efficient cell membrane breakage is needed to obtain enough mitochondria for subsequent purification steps. However, if this is forcibly approached, a damage of
the mitochondrial membranes could occur. Thus, cell rupture or
tissue homogenization conditions have to be balanced between
efficient plasma membrane rupture and consequent mitochondrial
yield and mitochondrial integrity. To this end, means are needed
that can be precisely adjusted to the respective source of isolation.
We recently reported [11] that this can be accomplished using the
pump controlled cell rupture system (PCC). PCC consists of a
current version of the Balch-homogenizer (Isobiotec, Germany)
[12] coupled to a high precision pump. Within the Balchhomogenizer, the cells or the tissue have to pass an exactly adjustable clearance (Fig. 1). This allows for defining and controlling the
shear forces, which affect the cells and the mitochondria in the
course of the homogenization process. The high precision pump
further increases the reproducibility and controllability of the overall homogenization process.
In order to further extend the practical value of this method,
with a special focus on potential experimental pitfalls, we provide
here an easy-to-follow protocol for the use of PCC and downstream centrifugation steps to isolate mitochondria from cultured
cells and minute tissue samples (Fig. 2). Taken together, the herein
described protocol reproducibly yields isolated mitochondria from
cultured cells or minute tissue samples that meet the quality and
purity required for molecular as well as for functional analyses.
Materials
Prepare all solutions using ultrapure water (Prepared by purifying
deionized water to attain a resistivity of 18.2 M-cm at 25 C).
2.1 Isolation
of Mitochondria
from Cell Culture
or Tissue Biopsies
1. Pump controlled cell rupture system (PCC): cell homogenizer

(Isobiotec, Germany), tungsten carbide balls (Isobiotec,
Germany), 1 ml gastight glass syringes (internal diameter
4.608 mm, SGE Supelco, USA) (see Note 1), pump elite
(Harvard apparatus, USA) (see Note 2), manual lift table.
Store the manual metal-lift table at 4 C (see Note 3).
2. Isolation buffer: 300 mM sucrose, 5 mM TES, 200 M EGTA,
pH 7.2. Transfer 102.8 g sucrose, 1.146 g TES and 76 mg
Semi-Automated Method to Isolate Mitochondria
89
Fig. 1 Cartoon of the pump controlled cell rupture system (PCC). A high precision
pump (1) ensures, via gastight syringes (2), the continuous sample delivery (3) in a
constant rate to the Balch-homogenizer (4). Cell breakage occurs upon passage
through a defined clearance (square), which is adjusted by selecting tungsten carbide balls of different diameters (5). The cell homogenate is collected in a second
syringe (6) and can be re-subjected to the homogenizer, which is thermally equilibrated by a cooling plate (7). In this protocol, the syringe that initially contains the
sample (cell suspension or tissue pieces) is referred to as syringe B and the second
syringe is referred to as syringe A. Reproduced with permission from: Schmitt S,
Saathoff F, Meissner L, Schropp EM, Lichtmannegger J, Schulz S, Eberhagen C,
Borchard S, Aichler M, Adamski J, Plesnila N, Rothenfusser S, Kroemer G, Zischka
H (2013) A semi-automated method for isolating functionally intact mitochondria
from cultured cells and tissue biopsies. Anal Biochem 443:6674
EGTA to a 1,000 ml glass beaker. Add water to a volume of

800 ml. Stir until all chemicals are dissolved and adjust pH
with 5 M KOH. Transfer the buffer to a 1 L measuring cylinder, fill it up to 1,000 ml with water and store at 4 C. The
buffer can be used for 1 week.
90
Fig. 2 Flow chart for the isolation of mitochondria from cultured cells and liver tissue biopsies for molecular
and biochemical analyses
2.2 Purification
of the Isolated
Mitochondria
91
1. 40 % w/v Nycodenz: transfer 40 g Nycodenz to a volumetric

flask. Add 10 mM TrisHCl (pH 7.4) to 100 ml. Store at 4 C
for up to 3 months.
2. 33 % w/v Nycodenz: Add 3.3 ml of the 40 % Nycodenzsolution
and 0.7 ml isolation buffer to a 15 ml conical tube and vortex
thoroughly. Store at 4 C for no longer than 1 week.
5. Ultracentrifuge (L70, Beckman, USA), swing out rotor
(SW55Ti, Beckman, USA) centrifugation tubes (Beckman
Coulter, Thinwall, Ultra-Clear, USA), 1 ml single use syringes
(Omnifix-F solo, Braun Melsungen, Germany), hypodermic
needle (Sterican 26Gx1, Braun Melsungen, Germany).
Methods
3.1 Isolation
of a Crude
Mitochondrial Fraction
from Cultured Cells
and Liver Tissue
1. Pre-cool the cell homogenizer on ice for at least 10 min.

Adjust the syringe pump settings concerning the syringe diameter (4.608 mm for 1 ml SGE syringe) and flow rate (700 l/
min). Equilibrate the syringes with isolation buffer. Insert the
tungsten carbide ball (see Note 4 and Table 1) into the cell
homogenizer and screw it down. Flush the cell homogenizer
three times with isolation buffer (see Note 5). Keep on ice
until starting the homogenization.
2. For mitochondria isolation from cell culture, harvest the cells
and determine the cell concentration and viability (see Note 6).
Pellet the cells, discard the supernatant and resuspend the cells
in isolation buffer to reach a concentration of 5 106 cells per
ml (see Note 7). Store on ice.
3. For mitochondria isolation from liver tissue biopsy, weigh the
tissue (see Note 8) and transfer it to a 10 ml glass beaker. Add
1 ml isolation buffer and scissor the liver tissue into pieces of
about 1 mm3 in size. Add isolation buffer to reach a
concentration of 3040 mg of tissue per ml isolation buffer
(see Note 9). Store on ice.
4. Place the cell homogenizer on the lift table. Fasten syringe A
on the cell homogenizer. Fill syringe B with 1 ml of the cell
suspension or of liver tissue and fasten it on the cell
homogenizer.
92
Table 1
Experimental settings and mitochondrial yield (9,000 g pellet) for the isolation of mitochondria from
diverse cell types (in parts reproduced with permission from Schmitt S, Saathoff F, Meissner L,
Schropp EM, Lichtmannegger J, Schulz S, Eberhagen C, Borchard S, Aichler M, Adamski J, Plesnila N,
Rothenfusser S, Kroemer G, Zischka H (2013) A semi-automated method for isolating functionally
intact mitochondria from cultured cells and tissue biopsies. Anal Biochem 443:6674)
Concentration
[cells/ml]
Clearance
[m]
Number
of strokes
Yield of crude
mitochondria
[g/5 106 cells]
[ratio
RFU FCCP]
Primary rat hepatocytes
3 106
10
1,230
1.8
McA 7777
5 106
10
146
1.6
5 10
82
1.6
H4IIE
7 10
70
1.9
HepT1
5 106
114
1.5
Huh6
5 106
10
60
1.6
HepG2
5 10
128
1.6
HEK 293
5 106
70
1.7
HeLa
5 106
107
1.7
BeWo
5 10
10
90
1.7
1205Lu
5 10
10
110
1.6
Panc02
5 106
162
1.5
MEF
5 106
90
1.6
Fao
All isolated mitochondrial suspensions efficiently built up an inner transmembrane potential (m) which remained
stable for at least 1 h
5. With a constant flow rate (700 l/min) the cell suspension is

pressed three or six times (see Note 10 and Table 1) through
the cell homogenizer. Transfer the homogenate to a 2.0 ml
tube on ice. In order to recover the whole homogenate, fill
one syringe with 1 ml isolation buffer, process it once through
the cell homogenizer (flow rate 700 l/min) and transfer it
also to the 2.0 ml tube (see Note 11).
6. Pass the liver tissue once (see Note 10) through the cell
homogenizer (flow rate 700 l/min) and transfer the
homogenate to a 2.0 ml tube on ice. Readjust the empty
syringe A on the cell homogenizer. Fill syringe B (i.e. the
syringe that initially contained the tissue sample) with 1 ml
isolation buffer, pump it once through the cell homogenizer
(flow rate 700 l/min) and transfer it also to the 2.0 ml tube
(see Note 12). Fill syringe A with 1 ml isolation buffer and
rinse the system manually (see Note 13).
93
7. Repeat steps 4 and 5 or 4 and 6 until all of the cells or tissue

are/is homogenized (see Note 14).
8. Centrifuge the homogenate for 5 or 10 min (cells or tissue,
respectively) at 800 g at 4 C.
9. Transfer the supernatant to a fresh 2.0 ml tube and centrifuge
it for 10 min at 9,000 g at 4 C.
10. Carefully discard the supernatant, slowly resuspend the pellets
(crude mitochondrial fraction) in 30 l isolation buffer. If
working with multiples of 5 106 cells, you may pool the
obtained crude mitochondrial fractions now.
11. Determine the protein concentration of the crude mitochondrial fraction. This is necessary to determine the number of
Nycodenz density gradients needed for further purification.
12. After the isolation, rinse the syringes 34 times with
ddH2O. Clean the used tungsten carbide ball and the cell
homogenizer with isopropanol and rinse them subsequently
with ddH2O. Dry the tungsten carbide ball thoroughly.
3.2 Purification
of the Crude
Mitochondrial
Fraction by Nycodenz
Density Gradient
Centrifugation
1. Prepare the Nycodenz density gradient. Add 500 l 18 %

Nycodenz to a 5 ml centrifuge tube (Thinwall, Ultra-Clear,
Beckman Coulter, USA). Slowly syringe 300 l of 24 or 33 %
Nycodenz (for cultured cells or liver tissue, respectively)
below the 18 % layer (see Note 15).
2. Load 650950 g of the crude mitochondrial fraction on the
discontinuous Nycodenz density gradient (see Note 16) and
centrifuge it for 15 min in an ultracentrifuge at ca. 110,000 g
(swing out rotor SW55Ti, Beckman Coulter, USA) at 4 C.
3. Collect the mitochondria from the phase boundary (24 %/18 %
or 33 %/24 % for cell culture mitochondria or rat liver tissue
mitochondria, respectively) (see Note 17). Transfer the purified mitochondria into a fresh 2.0 ml tube and adjust to 2 ml
with isolation buffer. Centrifuge for 10 min at 9,000 g (4 C)
to remove Nycodenz remnants.
4. Discard the supernatant and carefully resuspend the pellet in
30 l isolation buffer. Purified mitochondria typically yield
20 % in mg protein of the starting amount of the crude mitochondrial isolates subjected to the Nycodenz density
gradient.
5. The purified mitochondrial fraction may now be subjected to
molecular or biochemical analyses (Fig. 3 and table 1). For
example upon addition of respiratory substrate (e.g. succinate)
intact mitochondrial preparations should build up a membrane potential that remains stable for at least 1 h [11, 13].
94
Fig. 3 Monitoring of the isolation and purification of mitochondria from H4IIE cells via a Nycodenz density
gradient (a) by immunoblot analysis (b) and electron microscopy (c)
Notes
1. We do not recommend using plastic syringes due to their lack
of stability. In our hands, best results were obtained with
syringes from the given supplier.
2. This syringe pump model can be precisely adjusted and withstands the back-pressure that emerges during the homogenization process.
3. The manual metal-lift table can thereby assist the cooling of
the cell homogenizer during the homogenization process.
Two manual lift tables should be at hand if more than 25 106
cells will be homogenized. Exchanging against a pre-cooled
lift table every 30 min assists in effective cooling throughout
the homogenization process.
95
4. We determined a clearance of 6 or 10 m to be successful for

mitochondrial isolations from the tested cell lines given in
Table 1 in terms of yield and mitochondrial integrity. We
therefore recommend such initial settings which may, however, be further optimized. A clearance of 4 m is not recommended as it impaired mitochondrial functionality. For liver
tissue, a clearance smaller than 18 m resulted in a congestion
of the cell homogenizer.
5. Ensure that the system is air bubble free, as they alter the shear
forces during the homogenization. Even though clearances of
6 or 8 m already cause a noticeable back-pressure, one should
be able to manually pump buffer through the cell homogenizer. If a strong resistance of the buffer flow via the PCC
chamber can be encountered in such pre-isolation tests,
remove the tungsten carbide ball, rinse the cell homogenizer
thoroughly with isopropanol and ultrapure water and assemble the system again as described in step 1 of Subheading 3.1.
6. We typically use cell suspensions with more than 80 % viability.
A lower viability might be accompanied by an impairment of
mitochondrial structure or function, complicating subsequent
isolation.
7. 5 106 cells yield crude mitochondria of at least 60 g in the
cell lines tested so far (Table 1). Less than 3 106 or more than
10 1010 cells per ml decreased either the absolute or the relative yield of isolated mitochondria, respectively.
8. The weight of tissue is required to calculate how much isolation buffer has to be added to reach a concentration of
3040 mg tissue per ml.
9. A higher concentration overloads the system. Lower concentrations decrease the relative yield and increase the time expenditure of the homogenization. As the functional and structural
stability of isolated mitochondria is time-dependent, it is advisable to minimize the endurance of the whole isolation process
as much as possible.
10. The number of strokes needed to homogenize the cells
depends on the respective cell line (Table 1). One stroke
represents a passage from syringe B (i.e. the syringe that
initially contains the sample) to syringe A, and vice versa. For
most cell lines, less than three strokes were insufficient to
break the cell membrane, whereas more than six strokes
impaired mitochondrial functionality or structural integrity.
For rat liver tissue, we found that more than one stroke did
not markedly increase the mitochondrial yield.
11. The volume within the cell homogenizer approximately accounts
for 500 l. Therefore, it is important to rinse the system once
with isolation buffer to recover the whole homogenate and to
96
avoid residual contaminants in the cell homogenizer. If subsequently, mitochondria from other sources shall be isolated, we
recommend at least 34 cleaning passages of fresh isolation
buffer each. Alternatively, the whole homogenization unit
may be disassembled, cleaned, and reassembled again before
continuing.
12. Some small pieces of the liver tissue may enter the cell homogenizer, but not pass the clearance. We therefore recommend
that the second stroke starts with the syringe used for the first
stroke (syringe B). Tissue pieces that remained within the cell
homogenizer can now pass the clearance with this second
stroke, thereby increasing the yield without re-stressing already
isolated mitochondria.
13. In order to control the number of passages for the single
pieces of tissue, the system unit needs to be cleared from
remaining tissue pieces before starting with the next 3040 mg
of tissue to be processed.
14. Subsequent separation of mitochondria from unbroken cells/
nuclei in a first centrifugation step at 800 g is less efficient if
the homogenate is kept on ice for more than 45 min. A plausible explanation may be a progressive clotting of mitochondria, nuclei, and cell debris over time resulting in marked
mitochondrial loss in the 800 g pellet. Thus, in the case of
multiple isolations subsequent centrifugation steps should not
start later than 3040 min after the initial homogenization.
15. In order to avoid phase intermixing of the Nycodenz density
step gradient we recommend slowly layering the 300 l of 24
or 33 % Nycodenz (for cultured cells or liver tissue, respectively) with a syringe below the 18 % Nycodenz. The phase
boundary is slightly visible for the 18 %/24 % and clearly visible for the 18 %/33 % gradient.
16. If less than 650 g of the crude mitochondrial fraction are
loaded on the Nycodenz gradient, the mitochondrial interphase might be barely visible, overloading (more than 950 g
of the crude mitochondrial fraction) may impair the purification efficiency.
17. To collect the mitochondria, slowly lower a pipette mounted
with a 200 l tip just above the phase boundary, carefully aspirating the interphase. Immediately stop aspiration, the moment
as the mitochondrial band vanishes.
Acknowledgement
We would like to acknowledge E.E. Rojo and A. Simmons for critical
reading of the manuscript. This study was supported in parts by the
Deutsche Forschungsgemeinschaft (DFG) grant RU742/6-1 to H.Z.
97
References
1. Green DR, Kroemer G (2004) The pathophysiology of mitochondrial cell death. Science
305(5684):626629
2. Zischka H, Lichtmannegger J (2014)
Pathological mitochondrial copper overload in
livers of Wilsons disease patients and related
animal models. Ann N Y Acad Sci 1315:615
3. Zischka H et al (2011) Liver mitochondrial
membrane crosslinking and destruction in a rat
model of Wilson disease. J Clin Invest
121(4):15081518
4. Lin MT, Beal MF (2006) Mitochondrial dysfunction and oxidative stress in neurodegenerative diseases. Nature 443(7113):787795
5. Winklhofer KF, Haass C (2010) Mitochondrial
dysfunction in Parkinsons disease. Biochim
Biophys Acta 1802(1):2944
6. Hanahan D, Weinberg RA (2000) The hallmarks of cancer. Cell 100(1):5770
7. Schmitt S et al (2014) Why to compare absolute numbers of mitochondria. Mitochondrion
19(A):113123
8. Schulz S et al (2013) Progressive stages of
mitochondrial destruction caused by cell toxic
9.
10.
11.
12.
13.
bile salts. Biochim Biophys Acta 1828(9):

21212133
Galluzzi L et al (2010) Mitochondrial gateways to cancer. Mol Aspects Med 31(1):
120
Aichler M et al (2013) Clinical response to
chemotherapy in oesophageal adenocarcinoma
patients is linked to defects in mitochondria.
J Pathol 230(4):410419
Schmitt S et al (2013) A semi-automated
method for isolating functionally intact mitochondria from cultured cells and tissue biopsies. Anal Biochem 443(1):6674
Balch WE, Rothman JE (1985) Characterization of protein transport between successive
compartments of the Golgi apparatus: asymmetric properties of donor and acceptor activities in a cell-free system. Arch Biochem
Biophys 240(1):413425
Zamzami N, Metivier D, Kroemer G (2000)
Quantitation of mitochondrial transmembrane
potential in cells and in isolated mitochondria.
Chapter 9
Dynamic Range Compression with ProteoMiner:
Principles and Examples
Lei Li
Abstract
One of the main challenges in proteomics investigation, protein biomarker research, and protein purity
and contamination analysis is how to efficiently enrich and detect low-abundance proteins in biological
samples. One approach that makes the detection of rare species possible is the treatment of biological
samples with solid-phase combinatorial peptide ligand libraries, ProteoMiner. This method utilizes hexapeptide bead library with huge diversity to bind and enrich low-abundance proteins but remove most of
the high-abundance proteins, therefore compresses the protein abundance range in the samples. This work
describes optimized protocols and highlights on the successful application of ProteoMiner to protein
identification and analysis.
Key words Low-abundance proteins, Proteomics, ProteoMiner, Dynamic range compression,
Plasma, Serum, Saliva, Host cell proteins
Introduction
Concentrations of different proteins in biological samples can span
as much as 12 orders of magnitudes. For example, 99 % of the protein content in plasma and serum samples is only about 20 highabundance proteins, whereas the other thousands of proteins are
less than 1 % of the plasma or serum proteome [1, 2]. After decades
of research and accumulated knowledge on high-abundance proteins due to their relatively easy access, the discovery and analysis of
the rare species in proteomes attracts more and more attentions.
ProteoMiner technology is based on a combinatorial hexapeptide
library bound to chromatographic beads [36]. It can be used to
decrease levels of high-abundance proteins for researchers to enrich and
detect more mid- to low-abundance proteins in complex samples and
unveil the proteomes deeply and comprehensively. When a complex
biological sample is added to the beads, proteins bind to their specific
ligands through combination of various types of interactions including
ionic interaction, hydrophobic interaction, hydrogen bonding and van
99
100
Lei Li
Fig. 1 (a) Scheme of the ProteoMiner technology to remove most of high-abundant proteins and enrich lowabundant proteins, thus compressing the protein concentration range in biological samples. (b) Twodimensional gel electrophoresis of human serum samples before and after treatment by ProteoMiner beads,
adapted from [31]. The samples were first separated on a pH 58 IPG strip, then an 816 % Criterion TrisHCl
gel before visualization by the Flamingo Fluorescent Gel Stain
der Waals force, so high-abundance proteins saturate their specific

ligands and the excess proteins are washed away. In contrast, mediumand low-abundance proteins not saturating their specific ligands will
mostly stay bound to their peptide ligands and are therefore enriched
relatively to the high-abundance proteins during the process (Fig. 1).
The result is a reduction in the dynamic range which allows for the
detection of medium- and low-abundance proteins.
The ProteoMiner technology has been extensively applied to
variety of sample types [6] including plasma and serum [3], urine
[7], bile [8], platelets [9], red blood cell extract [10, 11], egg
white extract [12], egg yolk extract [13], CSF [14], saliva [15,
16], venom [17], milk whey fraction [18, 19], sea urchin coelomic
fluid [20], plant leaf [21], seeds [22], peel and pulp [23], phloem
exudates [24, 25], plant latex [26], bacteria [3], and even beverage
and wine [27, 28]. The ProteoMiner beads have also been successfully used on detection, identification, adsorption, and removal of
impurities or host cell proteins from purified proteins [29, 30].
Even though ProteoMiner beads have a large diversity of
hexapeptides, the analysis in different groups [31, 32] using twodimensional gel electrophoresis, immunoassay and mass spectrometry, indicated a high level of consistency from sample to sample
when processed with similar and variable bead volumes.
Principles and Examples of ProteoMiner
200 l
500 l
1 ml
101
5 ml
250 kD
150 kD
100 kD
75 kD
50 kD
37 kD
25 kD
20 kD
15 kD
Fig. 2 Effect of sample to beads ratio on protein enrichment profile. Different amount of human serum samples
were treated with 20 l ProteoMiner beads in triplicate. The eluates were separated on a 420 % criterion
TrisHCl gel and visualized by Bio-Safe Coomassie Stain. The red arrows indicate increasing amount of proteins with higher sample to beads ratio
This chapter describes the standard protocol optimized for

plasma and serum samples to compress proteomic dynamic range.
Results with other sample types will vary depending on the amount
of protein in the samples as well as the dynamic range that each
sample type contains. The ratio of protein to beads is crucial for
optimal performance of ProteoMiner kits, so best results will be
obtained by optimization of ratio of proteins to ProteoMiner beads
for each individual sample type. With increasing amount of serum
sample loaded onto same amount of ProteoMiner beads, some
proteins can be enriched more than the others (Fig. 2).
Materials
1. Microcentrifuge or vacuum manifold.
2. ProteoMiner Protein Enrichment Large Capacity Kit (BioRad, #163-3007 or #163-3009) and Small Capacity Kit (BioRad, #163-3006 or #163-3008). These kits contain 100 l
(Large Capacity) or 20 l (Small Capacity) of ProteoMiner
beads, PBS wash buffer (150 mM NaCl, 10 mM NaH2PO4,
pH 7.4), lyophilized elution reagent (8 M urea, 2 % CHAPS)
and rehydration reagent (5 % acetic acid), and are intended for
simple one-step elution of the bound proteins.
3. ProteoMiner Sequential Elution Large Capacity Kit (Bio-Rad,
#163-3011) and Small Capacity Kit (Bio-Rad, #163-3010).
These kits contain 100 l (Large Capacity) or 20 l (Small
Capacity) of ProteoMiner beads, PBS wash buffer (150 mM
NaCl, 10 mM NaH2PO4, pH 7.4) and 4 elution reagents
102
Lei Li
(Elution Reagent 1: 1 M NaCl, 20 mM HEPES, pH 7.4;

Elution Reagent 2: 200 mM glycine, pH 2.4; Elution Reagent
3: 60 % ethylene glycol in water; Elution Reagent 4: 33.3 %
2-propanol, 16.7 % acetonitrile, 0.1 % trifluoroacetic acid),
and are intended for sequential elution of the bound
proteins.
4. Micro Bio-Spin P-6 Gel Columns in 10 mM TrisHCl buffer,
pH 7.4 (Bio-Rad, #732-6221 or #732-6222).
5. Exchange buffer: 7 M Urea, 2 M thiourea, 2 % CHAPS with
or without 30 mM TrisHCl, pH 8.5.
6. 2 ml microcentrifuge tubes with and without caps.
Methods
For Subheadings 3.13.5, the ProteoMiner Large Capacity Kits
are used as the example. The bold and italic volumes in parenthesis
are for the ProteoMiner Small Capacity Kits.
3.1 Column
Preparation
and Equilibration
Vacuum at about 16 mmHg can replace centrifugation for column

preparation, sample binding, and sample wash steps if desired.
1. First remove the top cap and then snap off the bottom cap from
each of the ProteoMiner spin columns you will be using. Keep
both top and bottom caps for usage throughout the protocol.
If beads settle in top cap, replace after removing bottom plug
and centrifuge with top cap on column. To use bottom cap as
a plug, invert and firmly place to bottom of spin column.
2. Place the column in a collection tube and centrifuge at
1,000 g for 3060 s to remove the storage solution.
3. Replace the bottom cap and add 600 l (200 l) of wash buffer, then replace top cap. Invert column end-to-end several
times over a 5 min period.
4. Remove bottom cap, place the column in a collection tube
and centrifuge at 1,000 g for 3060 s to remove the wash
buffer.
5. Repeat steps 3 and 4 one more time.
3.2 Sample Binding

(See Note 1)
Samples should be free of precipitate. If needed, centrifuge samples

at 10,000 g for 10 min to clarify or pretreat the samples to remove
incompatible or interfering materials (see Note 2).
Replace bottom cap and add sample to column, for example
1 ml (200 l) of serum or plasma. Replace top cap and invert column end-to-end on a platform or rotational shaker for 2 h at room
temperature (see Note 3).
3.3
Sample Wash
103
1. Remove bottom cap, place column in a collection tube and

centrifuge at 1,000 g for 3060 s.
2. Replace bottom cap and add 600 l (200 l) of wash buffer to
column. Replace top cap and invert from end-to-end over a
5 min period.
centrifuge at 1,000 g for 3060 s.
4. Repeat steps 2 and 3 three more times.
3.4
Elution
1. After all wash buffers have been removed, replace the bottom
and bottom caps and add 600 l (200 l) deionized water.
Attach top cap and invert end-to-end for 1 min.
2. Remove caps, place column in a collection tube and centrifuge
at 1,000 g for 3060 s to remove water.
If using vacuum up to this point, you will now need to switch
to centrifugation.
3. Attach bottom cap to the column (take caution to ensure the
bottom cap is tightly attached). Add 100 l (20 l) of rehydrated elution reagent to the column and replace top cap.
Lightly vortex for 5 s (see Note 4).
4. Incubate column at room temperature, lightly vortex several
times over a period of 15 min.
5. Remove caps, place in a clean collection tube and centrifuge at
1,000 g for 3060 s. This elution contains your eluted proteins.
6. Repeat steps 57 two more times.
7. Store elution at 20 C or proceed with downstream analysis.
For some downstream applications, we recommend a clean-up
of your sample prior to analysis. See Subheading 3.6 below for
the protocol in details.
3.5 Sequential
Elution (Alternative
to Subheading 3.4
After Subheading 3.3)
1. Carefully add 200 l wash buffer on top cap and all sides of the
column to ensure none of the beads are stuck to the cap and
sides of the column.
2. Centrifuge at 1,000 g for 3060 s. Discard collected material.
If using vacuum up to this point, you will now need to switch
to centrifugation.
3. Attach bottom cap to the column (take caution to ensure the
bottom cap is tightly attached). Add 100 l (20 l) of elution
reagent 1 to spin column. Incubate at room temperature and
lightly vortex several times over a period of 10 min.
centrifuge at 1,000 g for 3060 s to collect the elution. This
elution contains your eluted proteins.
5. Repeat steps 3 and 4 two more times and collect both elutions
in the same tube from the above step.
104
Lei Li
6. Repeat steps 35 for the other three elution reagents.

7. Store elutions at 20 C or proceed with downstream analysis.
3.6 Elution Clean-Up
with Micro Bio-Spin
columns
Some downstream applications of the ProteoMiner-treated elutions may require to remove incompatible reagents. Below is an
easy and convenient method using the Micro Bio-Spin P-6 gel column for buffer exchanging the ProteoMiner eluates with compatible reagents.
1. Invert the column several times to resuspend the settled gel
and remove any bubbles.
2. Snap off the tip and place column in a 2 ml microcentrifuge
tube. Remove cap. Allow the excess packing buffer to drain by
gravity to top of gel bed. (If column does not begin to flow,
push cap back into column and remove.) Discard the buffer.
3. Place the column back into the 2 ml tube. Centrifuge for
1 min at 1,000 g to remove the packing buffer.
4. Block the bottom of the Bio-Spin 6 column with the tip and
add 500 l of the downstream application compatible buffer
(for example, the Exchange buffer in Subheading 2). Invert
the column several times to resuspend the settled gel.
5. Remove the bottom tip and place the column in a clean 2 ml
microcentrifuge tube. Remove cap. Allow the excess packing
buffer to drain by gravity to top of gel bed. Discard the buffer.
6. Place the column back into the 2 ml tube. Centrifuge for
1 min at 1,000 g to remove the buffer.
7. Repeat steps 46 two more times.
8. Place the column in a clean 2 ml microcentrifuge tube.
Carefully apply up to 50 l of ProteoMiner eluates to the column and centrifuge for 2 min at 1,000 g. The collected solution is buffer-exchanged to the compatible reagents.
Notes
1. As listed examples in the introduction, variety of sample types
from different species and sources have been used successfully
with the ProteoMiner beads. In principle, the more proteins are
incubated with the ProteoMiner beads, the more degree of
dynamic range compression can be reached, whereas extremely
low abundant proteins can be much enriched compared to the
high abundant proteins. As a guideline, 50 mg of serum or
plasma proteins are recommended for 100 l ProteoMiner
beads. In return, a bit more than 1 mg of proteins can be bound
to and eluted from the beads, so that up to 50-fold dynamic
range compression can be obtained. Depending on the needs
and goals of the specific research, it would be wise to keep in
105
mind of this while designing the experiment and deciding the ratio
between sample and ProteoMiner beads. Much higher sample to
beads ratio has been tested (Fig. 2) with still increasing enrichment
of a few proteins. Regular buffers based on phosphate buffered
saline (PBS) or Tris buffered saline (TBS) with or without mild
detergents in samples are recommended for the ProteoMiner
beads because of their less interference with binding of proteins to
ProteoMiner peptides (see Notes 24 below in more details).
2. Biological samples often contain interfering or incompatible
materials other than proteins which researchers are interested
in. It is crucial to remove them before applying samples onto
ProteoMiner beads. Dialysis, gel filtration, centrifugation and
protein precipitation are a few examples of pretreatment methods, and can be selected according to the sample types and
sizes, protein concentrations, and downstream applications.
The quality of the samples before ProteoMiner treatment can
be simply analyzed by gel electrophoresis or a UV/Vis spectrophotometer. The ProteoMiner process can be viewed as
affinity purification with complex proteinpeptide interactions. Interfering materials and incompatible solutions for
regular protein affinity purification will have some effects on
binding of certain proteins to their specific peptide ligands.
Buffer compatibility for wash and elution steps can often refer
affinity purification guidance and principles.
3. It is recommended to optimize the sample to beads ratio in order
to achieve the desired performance. Higher sample to beads ratio
tends to enrich more mid- to low-abundant proteins (Fig. 2). If
the protein concentration is low in the samples, concentrating
may be needed. Otherwise, the equilibrated ProteoMiner beads
can be taken out of the column and incubated with samples in
another tube. After sample binding is done, the beads can be
spun down and transferred back into the column for wash and
elution. The sample binding incubation time and temperature in
this step are the recommendations for plasma and serum samples.
Incubation at 4 C overnight would be another way researchers
can often use for their specific samples. If using plasma, clumping may occur after 1 h of binding; this is expected and will not
negatively impact the sample preparation. Heparinized plasma is
not compatible with this kit.
4. For 2-D users who plan to use the ProteoMiner treated samples on DIGE, this elution reagent will require clean up and
pH adjustment. As an alternative you may elute with DIGE
buffer. However, this may result in a decreased yield and number of protein spots. For a more complete elution if it would
not interfere with downstream applications, the Laemmli
sample buffer with 1 % SDS and reducing agents like
-Mercaptoethanol or DTT can be added and incubated with
ProteoMiner beads at 95 C for 5 min before elution.
106
Lei Li
References
1. Anderson N, Anderson N (2002) The human
plasma proteome: history, character, and diagnostic prospects. Mol Cell Proteomics 11:
845867
2. Anderson N, Polanski M, Pieper R et al (2004)
The human plasma proteome: a nonredundant
list developed by combination of four separate
sources. Mol Cell Proteomics 3:311326
3. Thulasiraman V, Lin S, Gheorghiu L et al
(2005) Reduction of the concentration difference of proteins in biological liquids using a
library of combinatorial ligands. Electrophoresis
26:35613571
4. Guerrier L, Thulasiraman V, Castagna A et al
(2006) Reducing protein concentration range
of biological samples using solid-phase ligand
libraries. J Chromatogr B 833:3340
5. Boschetti E, Righetti P (2008) Hexapeptide
combinatorial ligand libraries: the march for
the detection of the low-abundance proteome
continues. Biotechniques 44:663665
6. Boschetti E, Righetti P (2013) Low abundance protein discovery: state of the art and
protocols. Elsevier, Amsterdam
7. Castagna A, Cecconi D, Sennels L et al (2005)
Exploring the hidden human urinary proteome via ligand library beads. J Proteome Res
4:19171930
8. Guerrier L, Claverol S, Finzi L et al (2007)
Contribution of solid-phase hexapeptide
ligand libraries to the repertoire of human bile
proteins. J Chromatogr A 1176:192205
9. Guerrier L, Claverol S, Fortis F et al (2007)
Exploring the platelet proteome via combinatorial, hexapeptide ligand libraries. J Proteome
Res 6:42904303
10. Sim C, Bachi A, Cattaneo A et al (2008)
Performance of combinatorial peptide libraries
in capturing the low-abundance proteome of
red blood cells. 1. Behavior of mono- to hexapeptides. Anal Chem 80:35473556
11. Bachi A, Sim C, Restuccia U et al (2008)
Performance of combinatorial peptide libraries
in capturing the low-abundance proteome of
red blood cells. 2. Behavior of resins containing individual amino acids. Anal Chem 80:
35573565
12. D'Ambrosio C, Arena S, Scaloni A et al (2008)
Exploring the chicken egg white proteome
with combinatorial peptide ligand libraries.
J Proteome Res 7:34613474
13. Farinazzo A, Restuccia U, Bachi A et al (2009)
Chicken egg yolk cytoplasmic proteome,
mined via combinatorial peptide ligand libraries. J Chromatogr A 1216:12411252
14. Shores K, Udugamasooriva D, Kodadek T

et al (2008) Use of peptide analogue diversity
library beads for increased depth of proteomic
analysis: application to cerebrospinal fluid.
15. Bandhakavi S, Stone M, Onsongo G et al
(2009) A dynamic range compression and
three-dimensional peptide fractionation analysis platform expands proteome coverage and
the diagnostic potential of whole saliva.
16. Bandhakavi S, Van Riper S, Tawfik P et al
(2010) Hexapeptide libraries for enhanced
protein PTM identification and relative abundance profiling in whole human saliva.
17. Calvete J, Fasoli E, Sanz L et al (2009)
Exploring the venom proteome of the western
diamondback rattlesnake, Crotalus atrox, via
snake venomics and combinatorial peptide
ligand library approaches. J Proteome Res
8:30553067
18. D'Amato A, Bachi A, Fasoli E et al (2009)
In-depth exploration of cows whey proteome
via combinatorial peptide ligand libraries.
19. Liao Y, Alvarado R, Phinney B et al (2011)
Proteomic characterization of human milk
whey proteins during a twelve-month lactation
period. J Proteome Res 10:17461754
20. Fasoli E, D'Amato A, Righetti P et al (2012)
Exploration of the sea urchin coelomic fluid
via combinatorial peptide ligand libraries. Biol
Bull 222:93104
21. Fasoli E, DAmato A, Kravchuk A et al (2011)
Popeye strikes again: the deep proteome of
spinach leaves. J Proteomics 74:127136
22. Esteve C, D'Amato A, Marina M et al (2012)
Identification of olive (Olea europaea) seed
and pulp proteins by nLC-MS/MS via combinatorial peptide ligand libraries. J Proteomics
75:23962403
23. Fasoli E, Righetti P (2013) The peel and pulp
of mango fruit: a proteomic samba. Biochim
Biophys Acta 1834:25392545
24. Frhlich A, Gaupels F, Sarioglu H et al (2012)
Looking deep inside: detection of lowabundance proteins in leaf extracts of
Arabidopsis and phloem exudates of pumpkin.
Plant Physiol 159:902914
25. Gaupels F, Sarioglu H, Beckmann M et al
(2012) Deciphering systemic wound responses
of the pumpkin extrafascicular phloem by
metabolomics and stable isotope-coded protein labeling. Plant Physiol 160:22852299

26. DAmato A, Bachi A, Fasoli E et al (2010)
In-depth exploration of Hevea brasiliensis
latex proteome and hidden allergens via
combinatorial peptide ligand libraries.
J Proteomics 73:13681380
27. Cereda A, Kravchuk A, DAmato A et al
(2010) Proteomics of wine additives: mining
for the invisible via combinatorial peptide
ligand libraries. J Proteomics 73:17321739
28. Cereda A, Kravchuk A, DAmato A et al (2010)
Noahs nectar: the proteome content of a glass
of red wine. J Proteomics 73:23702377
29. Fortis F, Guerrier L, Righetti P et al (2006)
A new approach for the removal of protein
impurities from purified biologicals using
107
combinatorial solid-phase ligand libraries.

Electrophoresis 27:30183027
30. Fortis F, Guerrier L, Areces L et al (2006) A
new approach for the detection and identification of protein impurities using combinatorial
solid phase ligand libraries. J Proteome Res
5:25772585
31. Li L, Sun C, Freeby S et al (2009) Protein
sample treatment with peptide ligand library:
coverage and consistency. J Proteomics
Bioinform 2:485494
32. Dwivedi R, Krokhin O, Cortens J et al (2009)
Assessment of the reproducibility of random
hexapeptide peptide library-based protein normalization. J Proteome Res 9:11441149
Chapter 10
Qualitative and Quantitative Proteomic Analysis
of Formalin-Fixed Paraffin-Embedded (FFPE) Tissue
Omid Azimzadeh, Michael J. Atkinson, and Soile Tapio
Abstract
Formalin-fixed, paraffin-embedded (FFPE) tissue has recently gained interest as an alternative to fresh/
frozen tissue for retrospective protein biomarker discovery. However, during the formalin fixation proteins
undergo degradation and cross-linking, making conventional protein analysis technologies challenging.
Cross-linking is even more challenging when quantitative proteome analysis of FFPE tissue is planned. The
use of conventional protein labeling technologies on FFPE tissue has turned out to be problematic as the
lysine residue labeling targets are frequently blocked by the formalin treatment. We have established a
qualitative and quantitative proteomics analysis technique for FFPE tissues that combines label-free proteomic analysis with optimized protein extraction and separation conditions.
Key words Label-free proteomics, Biomarker discovery, Mass spectrometry, Formalin-fixed paraffinembedded (FFPE), Peptide modification
Introduction
For decades, formalin-fixed and paraffin-embedded (FFPE) tissue has
been the standard for histopathological analysis due to the ease of
sample preparation and stability in long-term storage. Clinical archives
frequently combine stored FFPE samples with data on diagnosis and
outcome. Such samples hold unique information on biological pathways and cellular processes leading to disease, but suitable molecular
technologies are required to access the information [13].
For a long time, quantitative proteomic studies on archival material have been considered to be an almost impossible task, primarily
due to the harsh and irreversible procedures needed for fixation and
the loss of macromolecular integrity during prolonged storage. The
changes occurring during cross-linking of proteins during the fixation
of FFPE tissue have been investigated in different studies [46]. The
consensus is that the major consequence of formaldehyde fixation is
the generation of a methylol modification at free lysine residues [4, 7].
Several studies have been conducted to evaluate different components
109
110
Omid Azimzadeh et al.
used for protein extraction, including buffer components, detergents,

pH, temperature, and pressure [4, 810].
Most chemical labels used in quantitative proteomics target
lysine residues, leading to inefficient labeling of FFPE material
where lysine is no longer available [1113]. Recently, label-free
approaches have been proposed as an alternative for quantification
of proteome profiles from FFPE samples [14, 15].
Here, we provide a tested methodological protocol suitable for
the quantitative and qualitative proteomic analysis of archival tissue
combining optimal protein extraction and separation with gelbased proteomics [4]. We successfully used this method to study
the FFPE archived cardiac tissue from sham- and total bodyirradiated C57BL/6 mice [4, 16]. For quantitative proteome analysis, we used label-free approach [17] to detect putative molecular
biomarkers of ionizing radiation.
This methodology can be transferred to other tissues and treatments. The protocol will facilitate the development of future proteomic analysis techniques for FFPE tissue and provide a tool for
the validation in archival samples of biomarkers of exposure, prognosis, and disease.
Materials
All solutions are prepared using HPLC grade water. All buffers
contain protease inhibitor cocktail following the manufacturers
instructions (Complete, Roche Diagnostics).
2.1 Buffers
and Solutions
for Sample
Preparation
1. Rehydration buffer: Graded series of ethanol (100, 95, and

70 %) (v/v) in water.
2. Wash buffer: 20 mM TrisHCl, pH 7.5, 0.5 % (w/v)
beta-octylglucoside.
3. Lysis buffer: 20 mM TrisHCl, pH 8.8, 2 % (w/v) SDS, 1 %
(w/v) beta-octylglucoside, 200 mM DTT, 200 mM glycine.
2.2 SDSPolyacrylamide Gel

Electrophoresis
(SDS-PAGE)
1. 2 Laemmli sample buffer: 62.5 mM TrisHCl, pH 6.8, 2 %

(w/v) SDS, 25 % (w/v) glycerol, 0.01 % (w/v) bromophenol
blue, 5 % -mercaptoethanol.
2. Homogeneous pre-cast SDS-PAGE gels (12 %).
3. SDS-PAGE running buffer: 192 mM glycine, 0.1 % SDS
(w/v), 25 mM Tris base.
4. Electrophoresis equipment: running chamber and power supply.
5. Colloidal Coomassie Blue staining buffer: 0.08 % (w/v)
Coomassie Blue G250 (CBB G250), 1.6 % (v/v) orthophosphoric acid, 8 % (w/v) ammonium sulfate, and 20 %
(v/v) methanol.
FFPE Qualitative and Quantitative Proteomics
2.3 Processing
of Gel-Resolved
Proteins and Tryptic
Digestion
111
1. Trypsin (Sigma-Aldrich).
2. 100 % acetonitrile (ACN).
3. 50 mM NH4HCO3 in 30 % ACN (w/v), pH 8.0.
4. 10 mM NH4HCO3, pH 8.0.
5. 80 % (v/v) ACN, 1 % (v/v) trifluoroacetic acid (TFA).
6. 5 % (v/v) ACN, 0.5 % (v/v) TFA.
7. Trypsin digestion equipment: shaker, incubator, and SpeedVac
centrifuge.
8. Peptide separation and identification equipment: Highperformance liquid chromatography (HPLC) and electrospray
ionization tandem mass spectrometry (ESI-MS/MS).
2.4 Experimental
Kits
1. 2D-Clean-Up-Kit (Bio-Rad).
2.5
1. Progenesis software (Nonlinear).
Software
2. Bradford reagent.
2. Mascot (Matrix Science).

3. PANTHER bioinformatics tool (Protein ANalysis THrough
Evolutionary Relationships).
4. Database for Annotation, Visualization, and Integrated
Discovery (DAVID).
Methods
3.1 Protein
Extraction from FFPE
Tissue
1. Cut 20 m sections from the FFPE tissue blocks after initial

trimming to remove air exposed surfaces (see Note 1).
2. Place FFPE tissue sections (20 m thick, 80 mm2 wide) on
microscope slides and deparaffinize by incubating twice with
xylene for 10 min at room temperature before rehydration in
a graded series of ethanol (100, 95, and 70 %) for 10 min each.
3. Scrap the tissue sections from the slides and transfer to a reaction cup (see Note 1).
4. Wash the tissue sections with wash buffer described in
Subheading 2.1 for 15 min at room temperature with slight
shaking.
5. Incubate all samples in lysis buffer at 100 C for 20 min, and
then at 80 C for 2 h with shaking (see Note 2).
6. Centrifuge the extracts for 30 min at 14,000 g at 4 C.
7. Precipitate the protein extract with the 2D clean-up kit following the manufacturers instructions (see Note 3).
8. Estimate the protein concentration in the lysate by the
Bradford assay (see Notes 4 and 5).
112
9. Dissolve protein pellet (see step 7) with 1 Laemmli buffer at

a concentration of 15 g/L.
3.2 SDS-PAGE
and Protein Staining
1. Load sample wells of the 412 % SDS-PAGE gel with protein

solution and separate equal amounts (50100 g protein per
lane) [18] (see Note 6).
2. Fill electrophoresis chamber with SDS-PAGE running buffer.
Run electrophoresis gels at 90 V for the first ~15 min at room
temperature (until the blue dye moved out of the stacking
gel). Thereafter, increase voltage to 120 V.
3. After electrophoresis incubate 1D PAGE gels with Colloidal
Coomassie solution overnight on a shaker at 4 C.
4. Destain the gels with water and transfer to a clean container
and store in water at 4 C until further analysis.
3.3 Processing
of Gel-Resolved
Proteins and Tryptic
Digestion
For the identification and quantification of proteins, slice each

SDS-PAGE lane horizontally into five pieces. Digest proteins into
peptides prior to mass spectrometry analysis as follows:
1. Destain the gel pieces further and rinse with buffer containing
50 mM NH4HCO3 in 30 % acetonitrile (ACN) (v/w), pH 8.0.
2. Equilibrate the gel pieces in 10 mM NH4HCO3, pH 8.0, prior
to proteolytic digestion by shrinking them in 100 % ACN and
rehydrating in 10 mM NH4HCO3.
3. Add 0.10.2 g of trypsin per gel piece and incubate overnight at 37 C (see Note 7).
4. Elute the digested proteins using 80 % (v/v) ACN, 1 % (v/v)
trifluoroacetic acid (TFA).
5. Dry the eluates in a SpeedVac centrifuge.
6. Resuspend the dried samples in 20 l 5 % (v/v) ACN, 0.5 %
(v/v) TFA for subsequent high-performance liquid chromatography (HPLC) separation and ESI-MS/MS analysis (see
Notes 810).
3.4 An Example
of Label-Free
Proteomics on FFPE
Sample
Due to the protein cross-linking during FFPE preparation, conventional labeling approach cannot be used for quantitative proteomic analysis on FFPE tissue. Formaldehyde, used in tissue
fixation, leads to a 30 Da (methylol) modification mainly on lysine
residues and inhibits classical labeling methodology that is targeted
to lysine residues. Alternatively, FFPE proteome profile has been
recently quantified using non-labeling approaches. For label-free
proteomics, the acquired spectra should be identified and quantified using the Progenesis software [16, 17] (see Note 11). To
interpret the observed alterations in proteome, the significantly
differentially expressed proteins in all samples should be analyzed
using bioinformatics tools such as PANTHER or DAVID [19, 20].
Classify the proteins using Gene ontology (GO) categories [21].
113
The described methodology has been used to study the FFPE

archived cardiac tissue from sham- and total body-irradiated
C57BL/6 mice [4, 16]. The samples were analyzed using labelfree approach to detect putative molecular biomarkers of ionizing
radiation. The differentially expressed proteins were classified using
GO categories (cellular components, molecular function, and biological processes). Differentially expressed proteins grouped
according to molecular functions are shown in Fig. 1. In our case,
proteins with oxidoreductase activity represented the largest group.
The GO analysis of biological processes indicated that the deregulated proteins were mainly involved in lipid, carbohydrate, phosphate and nucleic acid metabolism as well as molecular transport
(Fig. 1b). The GO cellular compartment analysis showed that most
deregulated proteins belonged to the mitochondrial proteome.
Functional annotation clustering analysis using the DAVID software showed that the proteins belonged to different mitochondrial
compartments such as membranes, electron transport machinery,
matrix, and channels (Fig. 1c). Furthermore, radiation-induced
Fig. 1 Radiation-induced alteration of the cardiac mitochondrial proteome. Significantly differentially regulated
proteins were classified for GO categories Molecular function (a), Biological process (b), and Cellular
compartment (c) using the PANTHER and DAVID bioinformatics tools. The amount of deregulated proteins is
indicated as a number of the total amount of up- or down-regulated proteins at 24 h. The analysis indicated
that ionizing radiation caused impairment of mitochondrial structure and function
114
changes affected the proteins involved in oxidative phosphorylation such as NADH dehydrogenase, cytochrome c, and succinate
dehydrogenase.
Notes
1. To avoid keratin contamination, always wear gloves and cover
the hair during handling samples and gels.
2. To minimize proteolysis of the protein lysates, it is recommended to use protease inhibitors and keep the samples on ice.
3. During precipitation, do not overdry the pellet after removal
of the acetone as this might lead to difficulty in resolubilization of the pellet in the resuspension buffer.
4. If you have problems in determining the protein concentration of the lysate, it may be due to some component in the lysis
buffer. In this case try another protein quantification method.
5. If the protein yield after resolubilization is low, test other solubilization buffers as the result may differ with the types of tissues used. For more information, see the protocols mentioned
in ref. 6.
6. Use ultrapure proteomics grade reagents to prepare buffers for
staining and digestion.
7. For peptide or protein digestion, a ratio of between 1:100 and
1:20 (w/w) of trypsin to substrate is recommended.
8. HPLC setting and proteomics quantification should be handled by trained researchers.
9. To achieve a comprehensive and statically significant interpretation, use at least five biological and technical replicates if
they are available.
10. In order to evaluate the technical variability of mass spectrometry runs of label-free peptide quantifications of FFPE samples, analyze the aliquot of one control sample by repetitive
LC-MS/MS runs at least three replicates.
11. In order to evaluate the degree of the reversion of the crosslinking, the variable 30 Da methylol modification caused by
the cross-linking [7, 22, 23] was set for lysine, histidine, and
arginine residues during searching against the database [4].
Acknowledgements
This work was supported by a grant from the European
Community's Seventh Framework Programme (EURATOM)
contract n232628 (STORE).
115
References
1. Tapio S, Atkinson MJ (2008) Molecular information obtained from radiobiological tissue
archives: achievements of the past and visions
of the future. Radiat Environ Biophys 47(2):
183187
2. Tapio S, Hornhardt S, Gomolka M et al
(2010) Use of proteomics in radiobiological
research: current state of the art. Radiat
Environ Biophys 49(1):14
3. Tapio S, Schofield PN, Adelmann C et al
(2008) Progress in updating the European
radiobiology archives. Int J Radiat Biol
84(11):930936
4. Azimzadeh O, Barjaktarovic Z, Aubele M et al
(2010) Formalin-fixed paraffin-embedded
(FFPE) proteome analysis using gel-free and
gel-based proteomics. J Proteome Res 9(9):
47104720
5. Giusti L, Lucacchini A (2013) Proteomic studies of formalin-fixed paraffin-embedded tissues. Expert Rev Proteomics 10(2):165177
6. Steiner C, Ducret A, Tille JC et al (2014)
Applications of mass spectrometry for quantitative protein analysis in formalin-fixed paraffin-embedded tissues. Proteomics 14(45):
441451
7. Metz B, Kersten GF, Hoogerhout P et al
(2004) Identification of formaldehydeinduced modifications in proteins: reactions
with model peptides. J Biol Chem 279(8):
62356243
8. Magdeldin S, Yamamoto T (2012) Toward
deciphering proteomes of formalin-fixed
paraffin-embedded (FFPE) tissues. Proteomics
12(7):10451058
9. Hatakeyama K, Wakabayashi-Nakao K, Aoki Y
et al (2012) Novel protein extraction approach
using micro-sized chamber for evaluation of
proteins eluted from formalin-fixed paraffinembedded tissue sections. Proteome Sci 10:19
10. Shi SR, Taylor CR, Fowler CB et al (2013)
Complete solubilization of formalin-fixed,
paraffin-embedded tissue may improve proteomic studies. Proteomics Clin Appl 7(34):
264272
11. Xiao Z, Li G, Chen Y et al (2010) Quantitative
proteomic analysis of formalin-fixed and
paraffin-embedded nasopharyngeal carcinoma
using iTRAQ labeling, two-dimensional liquid
chromatography, and tandem mass spectrometry. J Histochem Cytochem 58(6):517527
12. Ono A, Kumai T, Koizumi H et al (2009)
Overexpression of heat shock protein 27 in
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
squamous cell carcinoma of the uterine cervix:

a proteomic analysis using archival formalinfixed, paraffin-embedded tissues. Hum Pathol
40(1):4149
Jain MR, Liu T, Hu J et al (2008) Quantitative
proteomic analysis of formalin fixed paraffin
embedded oral HPV lesions from HIV
patients. Open Proteomics J 1:4045
Donadio E, Giusti L, Cetani F et al (2011)
Evaluation
of
formalin-fixed
paraffinembedded tissues in the proteomic analysis of
parathyroid glands. Proteome Sci 9(1):29
Ostasiewicz P, Zielinska DF, Mann M et al
(2010) Proteome, phosphoproteome, and
N-glycoproteome are quantitatively preserved
in formalin-fixed paraffin-embedded tissue and
analyzable by high-resolution mass spectrometry. J Proteome Res 9(7):36883700
Azimzadeh O, Scherthan H, Yentrapalli R et al
(2012) Label-free protein profiling of formalinfixed paraffin-embedded (FFPE) heart tissue
reveals immediate mitochondrial impairment
after ionising radiation. J Proteomics 75(8):
23842395
Hauck SM, Dietter J, Kramer RL et al (2010)
Deciphering membrane-associated molecular
processes in target tissue of autoimmune uveitis by label-free quantitative mass spectrometry.
Mol Cell Proteomics 9(10):22922305
Laemmli UK (1970) Cleavage of structural
proteins during the assembly of the head of bacteriophage T4. Nature 227(5259):680685
Thomas PD, Kejariwal A, Guo N et al (2006)
Applications for protein sequence-function
evolution data: mRNA/protein expression
analysis and coding SNP scoring tools. Nucleic
Acids Res 34(Web Server issue):645650
da Huang W, Sherman BT, Lempicki RA
(2009) Systematic and integrative analysis of
large gene lists using DAVID bioinformatics
resources. Nat Protoc 4(1):4457
Ashburner M, Ball CA, Blake JA et al (2000)
Gene ontology: tool for the unification of biology. The Gene Ontology Consortium. Nat
Genet 25(1):2529
Fowler CB, Cunningham RE, O'Leary TJ et al
(2007) 'Tissue surrogates' as a model for
archival formalin-fixed paraffin-embedded tissues. Lab Invest 87(8):836846
Nirmalan NJ, Harnden P, Selby PJ et al (2008)
Mining the archival formalin-fixed paraffinembedded tissue proteome: opportunities and
challenges. Mol Biosyst 4(7):712720
Chapter 11
Full-Length Protein Extraction Protocols and Gel-Based
Downstream Applications in Formalin-Fixed Tissue
Proteomics
Alessandro Tanca, Sergio Uzzau, and Maria Filippa Addis
Abstract
Archival formalin-fixed, paraffin-embedded (FFPE) tissue repositories and their associated clinical information can represent a valuable resource for tissue proteomics. In order to make these tissues available for
protein biomarker discovery and validation studies, dedicated sample preparation procedures overcoming
the intermolecular cross-links introduced by formalin need to be implemented. This chapter describes a
full-length protein extraction protocol optimized for downstream gel-based proteomics applications.
Using the procedures detailed here, SDS-PAGE, western immunoblotting, GeLC-MS/MS, 2D-PAGE,
and 2D-DIGE can be carried out on FFPE tissues. Technical tips, critical aspects, and drawbacks of the
method are presented and discussed.
Key words FFPE, Paraffin-embedded, Biobanks, Archival tissues, GeLC-MS/MS, Proteomics,
2D-PAGE, DIGE, Electrophoresis, Immunoblotting, Mass spectrometry
Introduction
Tissue biopsies are subjected to formalin fixation and paraffin
embedding (FFPE) in order to enable their microscopic examination in the context of routine diagnostic procedures. Once the
diagnostic process is completed, the residual tissues are stored in
long-term repositories. Clearly, these archives possess a huge
potential for retrospective tissue proteomics studies, thanks to their
matched datasets on the patient clinical records including diagnosis, disease progression, and response to therapy. In addition, when
considering the restrictions due to significant ethical issues, logistic
complexity, or poor availability of rare disease samples, that are
connected with the recovery of sufficient numbers of high-quality,
well-characterized fresh tissues, archived FFPE tissues become an
especially valuable resource [1, 2].
117
118
Alessandro Tanca et al.
Nevertheless, the formalin fixation chemistry introduces stable

intermolecular cross-links within the tissue. As a consequence,
until less than a decade ago it was practically impossible to access
proteomic information of FFPE tissues. In the latest years, and
with varying degrees of success, several research groups have
worked at the development of methods aimed to solve this problem, leading to the devisal of strategies that fall into two main
approaches: digestion of the whole cross-linked protein matrix into
peptides, or extraction of full-length proteins by the combined
action of physical and chemical factors [35]. Typically, direct
tryptic digestion strategies generate products that are ideally suitable to liquid chromatography-tandem mass spectrometry (LCMS/MS) analysis, followed by various bioinformatic pipelines for
data parsing and interpretation (shotgun proteomics) [6, 7].
Notwithstanding the elevated performances of shotgun proteomics
in terms of information depth and data robustness, the ability to
obtain full-length proteins can still offer its own advantages, by
enabling the application of a wider range of downstream analytical
approaches, including separation by gel electrophoresis [810].
Each electrophoresis-based method bears its own specific advantages, but all share the added informational value of preserving
information on molecular weight or charge of the protein. Among
1D-based approaches, the potential of making FFPE extracts available to SDS-PAGE and western immunoblotting is clearly evident
and unquestioned. Despite the outstanding technical progresses
made in the proteomics field, western immunoblotting still remains
the gold standard for large-scale protein expression validation, due
to its technical reliability combined with the relative ease of execution and the low cost of instrumentation and reagents. SDS-PAGE
separation of FFPE extracts can also be followed by band cutting,
in-gel digestion and MS/MS identification (GeLC-MS/MS), in
an alternative to the shotgun proteomics approach. The advantages of GeLC-MS/MS are the ability to process extracts directly
without sample cleanup, the direct visualization of the approximate amount and quality of proteins being processed, the maintenance of molecular weight information, as well as the ability to
carry out data analysis by using merged information or only information on molecular weight areas of interest [11].
In addition to 1D separations, 2D-PAGE approaches, including the 2D-DIGE technology, can also be implemented on FFPE
tissue extracts upon removal of non-compatible extraction reagents.
Notwithstanding the tremendous rise and potential of shotgun
proteomics approaches, 2D-PAGE is still one of the workhorses of
differential proteomics, being accessible, relatively easy to implement, and not strictly dependent from the use of high-performance
mass spectrometry instrumentation or data processing capabilities.
This widespread technique can provide information on the global
patterns of protein expression, enabling the quantitative and
FFPE Tissue Protein Extraction and Analysis
119
qualitative study of all proteins in relation to each other without

requiring their previous identification, with the added advantage of
visualizing proteoforms of different charges or molecular masses
[12]. On the other hand, 2D-PAGE has well-known limitations,
including poor reproducibility, low sensitivity, and narrow linear
dynamic ranges, but these can be overcome by implementation of
the 2D-DIGE, that entails the use of fluorescent stains, powerful
imaging instrumentation, and ad-hoc experimental design
approaches, providing a considerable increase in reproducibility,
sensitivity, robustness, and linear dynamic range [13, 14].
The protocol described in this chapter enables extraction of
full-length proteins from FFPE tissues building on the antigen
retrieval concept, described in a seminal work by Shi et al., and
elaborated in several variants by different research groups [15, 16].
Here, protein extraction relies on exposure of fixed proteins to
combined physical agitation, heat, detergents and reducing agents,
in a buffered environment (Fig. 1a). The extracts generated by this
protocol are readily suitable to protein quantitation and SDSPAGE separation, followed either by transfer onto nitrocellulose
membranes for probing with antibodies, or to in-gel digestion
and MS identification. In order to enable 2D-PAGE separation,
the ionic detergents used for extraction are exchanged with
other reagents suitable to isoelectric focusing of proteins.
Fig. 1 Schematic workflow of the protocol. (a) Protein extraction steps. (b) Downstream gel-based
applications
120
Once cleared by these interfering substances, proteins can also be

subjected to cyanine labeling before separation and undergo the
more reliable and robust 2D DIGE workflow (Fig. 1b).
The successful implementation of this protocol enables the
insertion of FFPE tissue extracts into already established proteomic
analysis pipelines. However, it should always be kept in mind that,
although extraction and analysis procedures have now reached satisfactory quality levels, FFPE patterns are still not totally comparable to those generated using fresh-frozen tissues. Slight
differences do in fact remain concerning high molecular weight
and basic proteins. Furthermore, issues such as length of fixation
time and tissue cellularity can impact on quality and complexity of
the protein pattern, as discussed and detailed previously in several
dedicated works [1719]. In conclusion, when high-quality, fresh
tissues are not readily available or practical, the application of the
described protocols opens the valuable possibility to exploit the
FFPE tissue archive resource for proteomics studies.
Materials
All protocols require the use of a microcentrifuge and of 1.5 ml
microcentrifuge tubes. All reagents may be purchased from SigmaAldrich, if not otherwise specified.
2.1 Tissue
Deparaffinization
and Rehydration
1. Formalin-fixed paraffin-embedded tissue sections (preferably

510 sections, 310 m thick, placed into safe-lock microcentrifuge tubes) (see Notes 1 and 2).
2. Xylene (see Note 3).
3. Ethanol (absolute, 96 and 70 %).
2.2 Protein
Extraction
and Quantification
1. Extraction buffer: 2 % (w/v) sodium dodecyl sulfate (SDS),

200 mM dithiothreitol (DTT), 20 mM TrisHCl, pH 8.8
(see Notes 4 and 5).
2. Thermomixer (specifications: temperature up to 100 C, mixing speed up to 500 rpm; e.g. Thermomixer Comfort from
Eppendorf).
3. Detergent and reducing agent compatible protein quantitation
kit (e.g. EZQ protein quantitation kit from Life Technologies)
(see Note 6).
2.3
SDS-PAGE
1. Laemmli sample buffer: 2 % (w/v) SDS, 10 % (v/v) glycerol,

5 % (v/v) beta-mercaptoethanol, 62.5 mM TrisHCl, pH 6.8,
bromophenol blue (as much as needed).
2. Thermoblock (specifications: temperature up to 100 C; e.g.
Thermomixer Comfort from Eppendorf).
121
3. Precast mini-format polyacrylamide gel (e.g. 420 %

Mini-PROTEAN TGX Gel from Bio-Rad) (see Note 7).
4. Apparatus for mini-format polyacrylamide gel electrophoresis
(e.g. Mini-PROTEAN Tetra Cell from Bio-Rad).
5. Power supply (e.g. PowerPac Basic Power Supply from
Bio-Rad).
6. Coomassie-based staining solution
SafeStain from Life Technologies).
(e.g.
SimplyBlue
7. Microwave oven.
2.4
Western Blotting
1. Apparatus for wet electrophoretic protein transfer (e.g. Mini

Trans-Blot Electrophoretic Transfer Cell from Bio-Rad)
(see Note 8).
2. Whatman 3MM cellulose chromatography paper.
3. Nitrocellulose membrane (e.g. Hybond ECL membrane
from GE Healthcare) (see Note 9).
4. Transfer buffer: 25 mM Tris, 192 mM glycine, 20 % methanol
(see Note 10).
5. Stir bar and magnetic stirrer.
6. Freezer pack.
7. High-current power supply (e.g. PowerPac HC HighCurrent Power Supply from Bio-Rad).
8. Ponceau S solution (see Note 11).
9. Bovine Serum Albumin (BSA), lyophilized powder.
10. Phosphate buffered saline (PBS): 137 mM NaCl, 2.7 mM
KCl, 10 mM Na2HPO4, 2 mM KH2PO4, pH 7.4.
11. PBS-T solution: 0.05 % Tween-20 in PBS.
12. Blocking solution: PBS-T plus 3 % BSA.
13. Primary antibody solution: primary antibody diluted in PBS-T
plus 1 % BSA.
14. Secondary antibody solution: peroxidase-conjugated secondary antibody diluted in PBS-T plus 1 % BSA.
15. Chemiluminescent peroxidase substrate.
16. Imaging system suitable for chemiluminescent signal acquisition (e.g. Molecular Imager VersaDoc MP imaging system
from Bio-Rad).
2.5 2-D DIGE

and 2-D PAGE
1. 2-D Clean-Up Kit (GE Healthcare).

2. TUC buffer: 7 M urea, 2 M thiourea, 4 % (w/v) 3-[(3Cholamidopropyl)dimethylammonio]-1- propanesulfonate
(CHAPS), 10 mM TrisHCl, pH 8.8.
3. CyDye DIGE Fluors (GE Healthcare).
122
4. N,N-Dimethylformamide (DMF) anhydrous, >99 % (see

Note 12).
5. Immobilized pH gradient (IPG) 24 cm strips (e.g. Immobiline
DryStrip from GE Healthcare) (see Note 13).
6. Carrier ampholytes (e.g. IPG buffer from GE Healthcare).
7. 2-D reducing buffer: 8 M urea, 130 mM DTT, 4 % (w/v)
CHAPS, 2 % (v/v) IPG Buffer.
8. Rehydration buffer: 8 M Urea, 13 mM DTT, 4 % (w/v)
CHAPS, 1 % (v/v) IPG Buffer.
9. Rehydration tray for IPG strips (e.g. IPG box, along with a
reswelling tray, from GE Healthcare) (see Note 14).
10. Isoelectric focusing (IEF) apparatus (e.g. Ettan IPGphor 3
from GE Healthcare) with IPG manifold.
11. Paper wicks (two per strip).
12. Twenty-four centimeter gel casting system (e.g. Ettan
DALTtwelve Gel Caster from GE Healthcare).
13. Gradient former (e.g. Model 495 Gradient former from
Bio-Rad).
14. Light solution for gradient gel casting: 10 % (v/v)
Acrylamide/Bis-acrylamide (29:1) solution, 0.375 M Tris
HCl, pH 8.8, 0.05 % (w/v) ammonium persulfate (APS),
0.025 % (w/v) N,N,N,N-Tetramethylethylenediamine
(TEMED).
15. Heavy solution for gradient gel casting: 18 % (v/v)
Acrylamide/Bis-acrylamide (29:1) solution, 0.375 M Tris
HCl, pH 8.8, 10 % (v/v) glycerol, 0.05 % (w/v) APS, 0.025 %
(w/v) TEMED.
16. Displacing solution: 50 % (v/v) glycerol, 0.375 M TrisHCl,
pH 8.8, bromophenol blue (as much as needed).
17. Water-saturated n-butanol: 70 % (v/v) n-butanol.
18. Equilibrating buffer: 6 M urea, 2 % (w/v) SDS, 30 % (v/v)
glycerol, 50 mM TrisHCl, pH 8.8.
19. Equilibrating-reducing buffer: equilibrating buffer with 2 %
(w/v) DTT.
20. Equilibrating-alkylating buffer: equilibrating buffer with 2.5 %
(w/v) iodoacetamide (IAM).
21. Equilibration tray.
22. Agarose solution: 0.4 % (low melting point) agarose, 0.375 M
TrisHCl, pH 8.8, bromophenol blue (as much as needed).
23. Second dimension apparatus (e.g. Ettan DALTtwelve system from GE Healthcare).
24. Running buffer: 25 mM Tris, 192 mM glycine, 0.1 % SDS.
123
25. Laser scanner suitable for fluorescent sample imaging (e.g.

Typhoon Trio + laser scanner from GE Healthcare).
26. Software for differential analysis of 2D-DIGE images (e.g.
DeCyder 2-D Differential Analysis Software, version 7.0 or
higher, from GE Healthcare).
27. Fixing solution: 40 % methanol, 10 % acetic acid (see Note 15).
28. Colloidal Coomassie dye (e.g. HPE Coomassie Staining
Kit from SERVA).
29. Flatbed scanner suitable for densitometric analysis of 2D-gels
(e.g. ImageScanner III from GE Healthcare).
2.6 In Gel Trypsin
Digestion
1. Clean scalpel.
2. Acetonitrile (ACN).
3. 50 mM ammonium bicarbonate (ABC) (see Note 16).
4. Reducing solution: 10 mM DTT, 50 mM ABC (see Note 5).
5. Alkylating solution: 55 mM IAM, 50 mM ABC.
6. Trypsin, proteomics grade.
7. Air circulation thermostat.
8. 20 % trifluoroacetic acid (TFA).
9. Vacuum centrifuge (e.g. Concentrator plus from Eppendorf).
10. 0.2 % formic acid (FA).
Methods
3.1 Tissue
Deparaffinization
and Rehydration
1. Add 1 ml xylene per each tube containing the FFPE tissue

sections.
2. Vortex 10 s and incubate 5 min at RT.
3. Centrifuge 3 min at 16,000 g at RT.
4. Remove supernatant (see Note 17).
5. Repeat steps 14 twice.
6. Add 1 ml 100 % ethanol per tube.
7. Repeat steps 24.
9. Repeat steps 24.
11. Repeat steps 24.
12. Spin down, remove all remaining supernatant and let the tube
open in the hood for 5 min to dry the pellet (see Note 18).
124
3.2 Protein
Extraction
and Quantification
1. Add the extraction buffer to the tissue pellet (usually 100 l

buffer per 10 mg tissue) and gently mix with the pipette tip
(without pipetting or vortexing) (see Note 19).
2. Incubate at 99 C for 20 min at 500 rpm in a thermomixer.
3. Incubate at 80 C for 2 h at 500 rpm in a thermomixer (see
Note 20).
4. Centrifuge for 10 min at 16,000 g, collect the supernatant
and transfer it to a clean tube (see Note 21).
5. Quantify the protein extract with a detergent and reducing
agent compatible quantitation kit and store it at 20 or 80 C
(see Note 6).
3.3 Western
Immunoblotting
1. Mix the protein extract(s) (usual load 110 g per well) with
an equal volume of Laemmli sample buffer (see Note 22).
2. Incubate the sample(s) for 5 min at 95 C in a thermoblock,
then spin down briefly.
3. Assemble the apparatus for mini-format polyacrylamide gel
electrophoresis, add an adequate volume of running buffer
and load the sample(s).
4. Connect the gel electrophoresis apparatus to the power supply
and run according to the manufacturers instructions (approx.
30 min at 200 V for Bio-Rad TGX gels).
5. Equilibrate sponges, 3MM paper sheets, polyacrylamide gel
(containing the separated protein sample) and nitrocellulose
membrane by soaking them for a few seconds in transfer
buffer.
6. Assemble a sandwich composed by: sponge, two 3MM
paper sheets, polyacrylamide gel, nitrocellulose membrane,
two 3MM paper sheets, sponge (from the black cathode to the
red anode).
7. Put the sandwich into the transfer apparatus, along with a prefrozen freezer pack and a stir bar.
8. Position the transfer apparatus on a magnetic stirrer and fill it
up with transfer buffer.
9. Connect the transfer apparatus to the power supply and run at
250 mA for (at least) 1 h (see Note 23).
10. Stain the membrane with Ponceau staining for a few seconds
and destain with deionized water until the background is
almost completely white (if necessary, acquire the image with
a scanner) (see Note 11).
11. Block the membrane for 1 h with blocking solution (see Note 24).
12. Remove the blocking solution, pour an adequate volume of
PBS-T and wash the membrane with rapid agitation for 5 min.
125
13. Remove PBS-T, add the primary antibody solution and incubate
with gentle agitation for 1 h (after incubation, collect the primary antibody solution if planning to re-use it) (see Note 24).
14. Wash the membrane with an adequate volume of PBS-T with
rapid agitation for 5 min.
15. Repeat step 14 twice.
16. Remove PBS-T, add the secondary antibody solution and
incubate with gentle agitation for 1 h (after incubation, remove
the secondary antibody solution).
17. Repeat step 14 five times.
18. Wash the membrane with an adequate volume of PBS (alone,
without Tween-20) with rapid agitation for 5 min.
19. Remove excess PBS from the membrane (without allowing it
to dry out completely).
20. Put the membrane on a glass covered by a plastic film, cover it
with an adequate volume of chemiluminescent substrate and
incubate for 5 min.
21. Remove excess substrate from the membrane (without allowing it to dry out completely) and put the membrane between
two tracing paper sheets.
22. Acquire the chemiluminescent signal with a suitable imaging
system (see Note 25).
3.4
GeLC-MS/MS
1. Separate protein extract(s) by SDS-PAGE, according to steps

14 of Subheading 3.3, except the total sample load (usually
2030 g per lane).
2. Stain the gel with a Coomassie-based dye (e.g. three sequential 1 min microwave incubations in deionized water followed
by a 45 s microwave incubation with the staining solution if
using the SimplyBlue SafeStain staining).
3. Put the stained gel on a clean glass.
4. Fractionate the entire lane(s) in 1530 slices using a clean scalpel, and transfer each slice into a clean tube (see Note 26).
5. Shrink gel slices by adding ACN (up to completely cover the
gel piece) and incubating for 10 min at RT (gel pieces become
opaque and stick together).
6. Completely remove ACN, add the reducing solution up to
completely cover gel pieces and incubate for 45 min at 56 C
(see Note 5).
7. Chill tubes to RT and remove the remaining solution.
8. Add ACN up to completely cover gel pieces and incubate for
10 min at RT.
126
9. Completely remove ACN, add the alkylating solution up to

completely cover gel pieces and incubate for 30 min at RT in
the dark.
10. Remove the remaining solution, add ACN up to completely
cover gel pieces and incubate for 10 min.
11. Completely remove ACN.
12. If gel pieces are not completely destained (blue is still visible)
add 50 mM ABC and incubate at RT for 10 min, then shrink
with ACN for 10 min (repeat sequentially until gel pieces are
destained); otherwise, skip to the following step (see Note 27).
13. Dissolve trypsin in ABC in order to reach a 10 ng/l working
solution.
14. Add 10 l of trypsin working solution to the dry gel pieces and
leave them in an ice bucket or at 4 C for 60120 min.
15. Remove excess of trypsin solution, then add ABC up to completely cover gel pieces.
16. Place tubes with gel pieces into an air circulation thermostat
and incubate overnight at 37 C.
17. Chill tubes to RT, spin down and transfer each ABC supernatant to a new tube.
18. Add ACN up to completely cover gel pieces and incubate for
10 min.
19. Collect the ACN supernatant and add it to the tube containing the ABC supernatant belonging to the same gel slice
(see Note 28).
20. Acidify by adding into each tube 10 l of 20 % TFA per 100 l
of peptide solution.
21. Dry the solution in a vacuum centrifuge.
22. Resuspend the pellet in 0.2 % FA and store at 20 or 80 C
until mass spectrometry (MS) analysis.
3.5 Gradient Gel
Casting for 2D-DIGE
1. Fill the gel caster by alternating cassettes with separator sheets,

then tighten all the screws evenly.
2. Place the gradient former on a magnetic stir plate and add a
stir bar to the mixing chamber labeled light (both stopcocks
and the black lever should be in the closed position).
3. Prepare the heavy and light acrylamide solutions, except
APS and TEMED, and the displacing solution.
4. Immediately prior to pouring, add TEMED and APS to both
solutions, mix evenly, then gently pour the appropriate solutions into the gradient chambers (see Note 29).
5. Turn on the stirring bar in the mixing chamber to a steady
speed and maintain this same speed throughout casting.
127
6. Start casting the gels by opening the stopcock valve to the

multi-casting chamber as well as the valve closer to the gradient former (the black lever on the gradient former is still closed
at this time) (see Note 30).
7. Allow the light monomer solution to enter the multi-casting
chamber until the level of solution in the mixing chamber is
equal to the level of the heavy solution in the reservoir
chamber, then open the black lever on the gradient former
and begin mixing the solutions and creating the gradient.
8. When the acrylamide solution in the gradient chamber has almost
finished, close the stopcock on the gradient former and add to
the reservoir chamber about 200 ml of displacing solution.
9. Open the stopcock on the gradient former and allow the displacing solution to enter the multi-casting chamber until the
acrylamide solution has reached the desired level at the top.
10. Close the stopcock on the multi-casting chamber, immediately
pipette water-saturated n-butanol onto each gel to overlay, and
allow gels to polymerize at RT for at least 12 h (see Note 31).
3.6
2D-DIGE
The following protocol has been optimized using the GE

Healthcare materials and apparatus suitable for 2D-DIGE analysis.
Alternative reagents and instruments might also be used, but
refinements to the protocol might be therefore necessary.
1. Precipitate the protein extract(s) to be analyzed by 2D-DIGE
it with the 2-D Clean-Up Kit, following the manufacturers
instructions (see Note 32).
2. Resuspend the protein pellet with the TUC buffer, for a final
concentration of about 5 g/l (see Note 33).
3. Add 1.5 volumes of DMF to 1 volume of CyDye stock solution, to make a 400 pmol/l CyDye working solution.
4. Pipette a volume of protein sample equivalent to 50 g into a
clean tube, then add 1 l of CyDye working solution, mix by
pipetting and leave on ice for 30 min in the dark.
5. Add 1 l of 10 mM lysine to stop the reaction, mix by pipetting, and leave for 10 min on ice in the dark.
6. Add an equal volume of 2-D reducing buffer to each sample
and leave on ice for 10 min (see Note 34).
7. Mix the labeled protein samples that are going to be separated
on the same gel.
8. Add an adequate volume of rehydration buffer up to 450 l in
total (for 24 cm strips).
9. Pipette the sample evenly over a slot of the IPG box (see
Note 35).
128
10. Carefully pull off the cover film from the IPG strip gel and
place the strip into the slot, gel-side down, so that the rehydration solution can distribute evenly under the strip.
11. Gently close the lid of the IPGbox and allow the IPG strip gels
to rehydrate at RT for 624 h (see Note 36).
12. Transfer the strips from the IPGbox to the IPGphor manifold,
so that strips are faced up under the cover fluid with the anodic
end (+) pointing at the anode of the IPGphor.
13. Add 150 l deionized water to each paper wick (two per strip),
and position the wicks on each end of the IPG strips so that
one end of the wick overlaps the end of the IPG strip gel.
14. With the electrode cams in the open position, put the electrode assembly on top of all the wicks (the electrode must be
in contact with the wick), then close the lid and start the
appropriate IEF program.
15. Just before the end of the IEF program, prepare the
equilibration-reduction solution by adding 20 mg DTT per
ml of equilibration buffer.
16. Stop the IEF, drain the excess of mineral oil and briefly rinse
out the IPG strips with deionized water (see Note 37).
17. Position each IPG strip in a slot of the equilibration tray, add
an adequate volume of equilibration-reduction solution up to
completely cover the strip and incubate for 15 min with gentle
agitation.
18. Turn the pump valve of the Ettan DALTtwelve to circulate,
then fill the tank to the 7.5 l line with 1 running buffer.
19. On the control unit, adjust the temperature to the desired setting (25 C) and turn the pump on.
20. Prepare the equilibration-alkylation solution by adding 25 mg
IAM per milliliter of equilibration buffer.
21. Remove DTT solution from the IPG strips, add an adequate
volume of equilibration-alkylation solution up to completely
cover the strip and incubate for 15 min with gentle agitation.
22. Carefully unload the cassettes containing the polymerized gels
from the gel caster and rinse the top surface and the glass
plates with deionized water to remove residual n-butanol and
acrylamide.
23. Using forceps, remove the equilibrated IPG strips from the
equilibration solution and rinse them with fresh SDS running
buffer.
24. Slowly pipette agarose solution on the top of each gel, then
position the strip centered on the glass plate, and push against
the plastic backing of the IPG strip (with forceps or a plastic
support) to slide the strip between the two glass plates until it
is into contact with the surface of the slab gel.
129
25. Carefully slide the gels into the tank and fill the upper buffer
chamber with 2.5 l of 2 running buffer (up to the min/max
fluid line) (see Note 38).
26. Close the lid, set the running conditions according to the
desired run length (for instance, 1 W per gel overnight) and
start the run.
27. When the run has finished, unload the gel cassettes from the
electrophoresis chamber, rinse them with deionized water and
proceed with image acquisition using Typhoon laser scanner.
28. Upload the image files into the DeCyder software and carry
out differential analysis according to the software manual.
3.7 Preparation
of a Preparative
Gel, Excision,
and Digestion
of Differential Spots
1. Collect a volume of the protein sample obtained in step 2,

Subheading 3.6, corresponding to at least 300 g of samples
(preferably 500700 g).
2. Perform 2D-PAGE analysis according to steps 826 of
Subheading 3.6.
3. When the run has finished, unload the cassette from the electrophoresis chamber, open it and carefully put the gel into a
clean container.
4. Fix the gel with fixing solution for 3060 min, stain overnight
with a colloidal Coomassie dye, destain with deionized water
and proceed with image acquisition using a flatbed scanner
(see Note 39).
5. Excise from the preparative gel all the differential spots identified upon DeCyder analysis.
6. Shrink gel slices by adding ACN (up to completely cover the
gel piece) and incubating for 10 min at RT.
7. Perform in gel digestion and peptide recovery according to
steps 1122 of Subheading 3.4.
Notes
1. FFPE tissue blocks should be carefully selected and evaluated
by an expert pathologist in order to assure the presence of a
minimum percentage of cells of the desired tissue type (for
instance, >90 % of neoplastic tissue), and consecutive sections
should be obtained for proteomic analysis. In general, five
5-m-thick tissue sections should provide sufficient material
for a gel-based proteomic analysis. Furthermore, especially for
precious tumor samples, an additional step of manual dissection or laser microdissection can be favorably added to enrich
in cancer cells and keep the amount of adjacent connective and
normal tissue to the minimum.
130
2. Safe-lock tubes should be used, since a loose locking may

cause the tube to uncork during the high temperature incubation steps, with a consequent loss of biological material.
3. Xylene may cause harmful effects by eye and skin contact as well
as by inhalation, ingestion or aspiration, spanning from minor
irritation to headache, nausea and vomiting. It has therefore to
be used with care, wearing gloves and under a chemical hood.
As an alternative for tissue deparaffinization, a mineral oil mixture (heated at 95 C) can be used with similar results.
4. Several different extraction buffers have been described in the
literature as suitable for extraction of full-length proteins from
FFPE tissue samples. SDS and DTT are both widely recognized as essential to ensure a proper extraction and solubilization of proteins, although their concentration may vary (e.g.
4 % SDS is well documented). The extraction buffer can also
comprise additional reagents, such as octyl glucoside, glycerol
and polyethylene glycol [4].
5. All solutions containing DTT should be prepared fresh shortly
before use. Alternatively, stock solutions without DTT may be
prepared, and an appropriate amount of DTT powder may be
added just prior to use.
6. The use of a protein quantitation method being compatible
with detergents (SDS) and reducing agents (DTT) is mandatory. The fluorescence-, membrane-based EZQ system is the
most used in our lab with FFPE protein extracts for three reasons: (a) it requires a very low volume of sample for quantification (1 l); (b) it provides a satisfactory level of linearity and
sensitivity even for solutions with low protein concentrations;
(c) the washing steps enable an efficient removal of most
detergent and reducing agent. These characteristics fit well
with the very low size of many FFPE tissue samples, and
therefore with the low amount and/or concentration of the
corresponding protein extracts. Anyway, comparable or even
better quantification performance can be achieved using alternative detergent and reductant compatible methods.
7. It has been demonstrated that FFPE protein extracts contain a
lower amount of high molecular weight (MW) proteins compared to fresh-frozen tissues; a general depletion in the upper
part of 1D- or 2D-PAGE pattern can be therefore observed in
most cases. According to this, in order to enhance resolution
in the mediumlow MW part of the pattern, an acrylamide
percentage higher than that used for other samples can be conveniently chosen for gel-based separation (e.g. 420 % for precast gradient gels and 1018 % for home-made, large format
gradient gels; 14 % fixed-percentage gels can also be used).
8. Alternatively, semi-dry or dry protein transfer systems could
be used with similar results.
131
9. Alternatively, a PVDF membrane can be employed. In this

case, preliminary activation of PVDF membrane with methanol is needed.
10. Up to 0.1 % SDS may be added to the transfer buffer to maximize transfer of hydrophobic proteins.
11. Membrane staining with Ponceau is aimed at: (a) verifying if
bubbles have formed which hamper a complete transfer of the
proteins from certain regions of the gel; (b) checking if the
total load of the different samples is actually comparable; (c)
acquire an image of the total protein pattern which can be
overlapped to the final antibody signal as a control.
12. DMF can have carcinogen and teratogen effects, and thus
needs to be used with great care, wearing gloves and under a
chemical hood. It is also very important that a high grade,
anhydrous DMF is used to ensure an efficient labeling. Use a
DMF bottle less than 3 months old from day of opening and
prevent it from being exposed to air or water.
13. Using a large strip format (24 cm) is advised to reach a high
resolution and a proper separation of protein spots.
14. The IPGbox kit allows the passive rehydration to be performed
in the dark and, more importantly, without the need for covering the strips with mineral oil.
15. Some research groups have replaced methanol (more toxic)
with ethanol (safer, but slightly less efficient).
16. The ABC solution should be filtered before use and stored at
4 C to avoid contaminations.
17. Take particular care when removing supernatant in order not
to throw away tissue portions. If extracting proteins from
filamentous or colloidal tissues, increasing centrifugation time
may help pellet the tissue more efficiently.
18. Tissue pellets may also be vacuum dried, in order to maximize
absorption and penetration of the extraction buffer in the next
step. However, special attention to be paid to the vacuum centrifugation time to avoid excessive drying (or even burning)
which would make it difficult or impossible to resuspend the
tissue in the extraction buffer.
19. Usually gently mixing with a tip (a movement approximately
like stirring coffee with a spoon) is enough to adequately
ensure contact between buffer solution and tissue fragments.
Strictly avoid vortexing or hard pipetting, because of the presence of SDS in the buffer. Alternatively, a tissue disruption
step (e.g. performed using a steel bead and a mechanical
homogenizer) may allow a deeper and more even penetration
of the extraction buffer within the tissue.
20. At this stage, an overnight storage at 80 C may help improve
lysis and protein release from the tissue.
132
21. If a clear separation between the residual pellet and the protein
containing supernatant does not occur, collect the greatest
possible volume of supernatant without touching the tissue,
then centrifuge again the remaining wet pellet for at least
15 min at 16,000 g to squeeze it. Finally, collect the new
supernatant and merge it with the one obtained previously.
22. The protein load should be settled depending on the following factors: (a) concentration/abundance of the target antigen
within the protein mixture; (b) specificity of the antibody; (c)
sensitivity of the chemiluminescent substrate and of the signal
detection system. For a mediumlow abundance antigen, a
monoclonal antibody, the Sigma chemiluminescent peroxidase substrate and the VersaDoc imager, 1 g of total protein
should be enough.
23. When interested in high MW proteins, the transfer time may
be extended to 1.52 h.
24. Incubations may be extended up to overnight based on sample or antibody characteristics. For overnight incubations, preincubate the membrane at RT for approx. 30 min, then keep
the membrane at 4 C during the night, and finally equilibrate
the membrane the day after at RT for other 30 min.
25. If using the VersaDoc imaging system, a 5 min exposure is
usually enough to detect a visible signal.
26. The number of slices in which the lane is cut (i.e. the degree
of sample fractionation) is usually proportional to the number
of protein identifications achieved through MS analysis.
Alternatively, the number of protein identifications can be also
increased by increasing the length of the LC separation prior
to MS (a detailed description of the LC-MS/MS analysis
methods is outside the scope of this chapter).
27. Alternatively, samples can be destained using a solution containing 40 % (v/v) ethanol and 75 mM ABC and/or incubating the samples at 56 C and 500 rpm in a thermomixer until
the blue dye is no longer visible.
28. If the gel pieces are not completely shrunk and opaque after
collecting ACN, add fresh ACN up to cover the gel pieces,
incubate for 10 min, collect the new ACN supernatant and
add it to the supernatant mixture belonging to the same gel
slice.
29. In other words, the solution with the lower acrylamide concentration is to be poured in the mixing chamber labeled
light, and the solution with the higher acrylamide concentration in the reservoir chamber labeled heavy. Note that
the level of the solutions in the two chambers will NOT be
equal at this time.
133
30. Do not allow any air bubbles to enter the plates when casting
gradient gels, since bubbles hamper a correct formation of the
gradient.
31. The multi-casting chamber should be stored at RT possibly
overnight to ensure the complete acrylamide polymerization.
In fact, acrylamide polymerization continues in a silent way
even after gels start to visually appear as polymerized.
Afterwards, unused gels can be stored between wet papers at
4 C for up to 2 weeks.
32. Protein precipitation is performed to remove SDS and make
protein solution suitable for IEF separation. Alternatively,
detergent removal spin columns, dialysis membranes, or
MW-cut-off spin filters can be used to eliminate SDS from the
protein extract.
33. At this stage, sample pH should be between 8 and 9 to ensure
maximum labeling efficiency.
34. After this step, cyanine-labeled samples can be stored for at
least 3 months at 80 C in the dark.
35. Ensure that no bubbles are trapped within the strips.
36. Overnight rehydration is recommended to allow high MW
proteins to enter the gel.
37. Focused strips can be stored at 80 C for a few weeks.
38. Lubricate glasses with running buffer to prevent the rubber
tubing from sticking to the cassettes.
39. Alternatively, gels can be stained with fluorescent dyes (e.g.
Sypro Ruby) and their image can be acquired using an imager
suitable for fluorescent signal detection.
References
1. Klopfleisch R, Weiss AT, Gruber AD (2011)
Excavation of a buried treasure DNA,
mRNA, miRNA and protein analysis in formalin fixed, paraffin embedded tissues. Histol
Histopathol 26(6):797810
2. Fowler CB, O'Leary TJ, Mason JT (2013)
Toward improving the proteomic analysis of
formalin-fixed, paraffin-embedded tissue.
Expert Rev Proteomics 10(4):389400
3. Blonder J, Veenstra TD (2009) Clinical proteomic applications of formalin-fixed paraffinembedded tissues. Clin Lab Med 29(1):
101113
4. Tanca A, Pagnozzi D, Addis MF (2012)
Setting proteins free: progresses and achievements in proteomics of formalin-fixed,
paraffin-embedded tissues. Proteomics Clin
Appl 6(12):721
5. Maes E, Broeckx V, Mertens I, Sagaert X,

Prenen H, Landuyt B, Schoofs L (2013)
Analysis of the formalin-fixed paraffinembedded tissue proteome: pitfalls, challenges, and future prospectives. Amino Acids
45(2):205218
6. Stewart NA, Veenstra TD (2008) Sample
preparation for mass spectrometry analysis of
formalin-fixed paraffin-embedded tissue: proteomic analysis of formalin-fixed tissue.
Methods Mol Biol 425:131138
7. Krizman DB, Burrows J (2013) Use of
formalin-fixed, paraffin-embedded tissue for
proteomic biomarker discovery. Methods Mol
Biol 1002:8592
8. Addis MF, Tanca A, Pagnozzi D, Crobu S,
Fanciulli G, Cossu-Rocca P, Uzzau S (2009)
Generation of high-quality protein extracts
134
9.
10.
11.
12.
13.
14.

from formalin-fixed, paraffin-embedded tissues. Proteomics 9(15):38153823
Addis MF, Tanca A, Pagnozzi D, Rocca S,
Uzzau S (2009) 2-D PAGE and MS analysis of
proteins from formalin-fixed, paraffin-embedded
tissues. Proteomics 9(18):43294339
Tanca A, Pagnozzi D, Falchi G, Tonelli R,
Rocca S, Roggio T, Uzzau S, Addis MF (2011)
Application of 2-D DIGE to formalin-fixed,
paraffin-embedded tissues. Proteomics 11(5):
10051011
Tanca A, Pagnozzi D, Burrai GP, Polinas M,
Uzzau S, Antuofermo E, Addis MF (2012)
Comparability of differential proteomics data
generated from paired archival fresh-frozen and
formalin-fixed samples by GeLC-MS/MS and
spectral counting. J Proteomics 77:561576
Chevalier F (2010) Highlights on the capacities
of Gel-based proteomics. Proteome Sci 8:23
Viswanathan S, Unlu M, Minden JS (2006)
Two-dimensional difference gel electrophoresis. Nat Protoc 1(3):13511358
Tanca A, Pisanu S, Biosa G, Pagnozzi D,
Antuofermo E, Burrai GP, Canzonieri V,
Cossu-Rocca P, De Re V, Eccher A, Fanciulli
G, Rocca S, Uzzau S, Addis MF (2013)
15.
16.
17.
18.
19.
Application of 2D-DIGE to formalin-fixed diseased tissue samples from hospital repositories:

results from four case studies. Proteomics Clin
Appl 7(34):252263
Shi SR, Taylor CR, Fowler CB, Mason JT
(2013) Complete solubilization of formalinfixed, paraffin-embedded tissue may improve
proteomic studies. Proteomics Clin Appl
7(34):264272
Shi S-R, Shi Y, Taylor CR (2011) Antigen
retrieval immunohistochemistry. J Histochem
Cytochem 59(1):1332
Tanca A, Pagnozzi D, Falchi G, Biosa G, Rocca
S, Foddai G, Uzzau S, Addis MF (2011) Impact
of fixation time on GeLC-MS/MS proteomic
profiling of formalin-fixed, paraffin-embedded
tissues. J Proteomics 74(7):10151021
Thompson SM, Craven RA, Nirmalan NJ,
Harnden P, Selby PJ, Banks RE (2013) Impact
of pre-analytical factors on the proteomic analysis of formalin-fixed paraffin-embedded tissue. Proteomics Clin Appl 7(34):241251
Magdeldin S, Yamamoto T (2012) Toward
deciphering proteomes of formalin-fixed
paraffin-embedded (FFPE) tissues. Proteomics
12(7):10451058
Chapter 12
Enrichment of Low-Abundant Protein Targets
by Immunoprecipitation Upstream of Mass Spectrometry
Barbara Kaboord, Suzanne Smith, Bhavin Patel, and Scott Meier
Abstract
Immunoprecipitation (IP) is commonly used upstream of mass spectrometry (MS) as an enrichment tool for
low-abundant protein targets. However, several aspects of the classical IP procedure such as nonspecific
protein binding to the isolation matrix, detergents or high salt concentrations in wash and elution buffers,
and antibody chain contamination in elution fractions render it incompatible with downstream mass spectrometry analysis. Here, we discuss two IP workflows that are designed to minimize or eliminate these contaminants: the first employs biotinylated antibodies and streptavidin magnetic beads while the second method
utilizes a traditional antibody that is oriented and cross-linked to Protein AG magnetic beads. Both modified
magnetic supports have low background binding and both antibody immobilization strategies significantly
reduce or eliminate antibody heavy and light chain contamination in the eluent, minimizing potential ion
suppression effects and thereby maximizing detection of target antigens and interacting proteins.
Key words Immunoprecipitation, Co-immunoprecipitation, Mass spectrometry, Proteinprotein
interactions, Biotin, Streptavidin, Magnetic beads, Protein AG, In-solution digestion
Introduction
Immunoprecipitation is widely used to enrich specific protein targets from complex samples for the purpose of evaluating differential
protein expression [14], probing for post-translational modifications [512], or identifying interacting proteins [1316]. An antibody is used to bind an antigen in cell lysates, serum, or tissue
homogenates and the resulting immune complexes are captured on
a solid phase support such as a chromatography resin, magnetic
bead, plastic plate, or membrane. Mass spectrometry is increasingly
becoming the detection methodology of choice for IPs as the technology has the capability to be used for discovery experiments to
identify protein interactors [14, 1619], semi-quantitatively in conjunction with SILAC or TMT-labeling [2025], or quantitatively in
selective reaction monitoring (SRM) assays [2630] (Fig. 1).
135
136
Barbara Kaboord et al.
Sample
- Cell Culture
- Plasma/Serum
- Tissue
SDS-PAGE
Western blot
Immunoprecipitation
In-Gel/In-Solution Digestion
- Reduction/Alkylation
- Digestion
- Peptide cleanup
AQUA Peptide Spiking
Discovery MS
(LC-MS/MS)
Protein/PTMs
Identification,
Peptide Selection
Targeted MS
(LC-SRM/MS)
Absolute Quantitation
Fig. 1 Experimental workflow for IP to MS applications. Low-abundant proteins in a complex sample matrix typically need to be enriched in order to detect or quantify them. Protein targets are immunoprecipitated, processed
by in-gel or in-solution digestion, and analyzed by LC-MS/MS to identify proteins in the IP/Co-IP eluent, probe
for post-translational modifications, or select candidate peptides. Furthermore, heavy isotope-labeled peptide
standards can be added to the sample for targeted LC-SRM/MS for absolute quantitation. Gel electrophoresis
followed by silver stain or Western blot is still typically used as a validation component to this workflow
While the best antibody for IP enrichment is highly specific for

its target antigen, other abundant proteins from the sample milieu
can potentially bind nonspecifically to the antibody, the antigen, or
the solid support to which the antibody is immobilized. Buffers
containing salt and/or detergent are typically used to eliminate
these nonspecific protein binders, however, low-affinity and transient interactions are at risk of being washed away during those
steps. In addition, high salt and detergent are not compatible with
downstream mass spectrometry analysis. The use of protein crosslinking reagents is a viable strategy to preserve low-affinity complexes [17, 19, 31] during high stringency washes but creates an
additional level of complexity during the peptide/protein identification process and still results in high salt or detergent carry-over
in the final sample. In addition to abundant protein contaminants
from the original sample, antibody chains leach off the supports
during the IP elution step and become significant contaminants as
well. The presence of large quantities of any type of contaminating
protein can suppress ionization of the low-abundant target of interest, preventing its detection in the mass spectrometer. The choice
of IP solid support, the antibody attachment strategy, and the IP
protocol itself are critical areas to optimize in order to minimize
nonspecific protein binding and maximize antigen capture (Fig. 2).
IP-MS
137
Fig. 2 Evaluation of resin types and IP strategies using an EGFR immunoprecipitation model system. Direct IP
(antibody pre-immobilized on the solid support followed by incubation with sample) was compared for three
supports (Polyacrylamide (P), Agarose (A) and Magnetic (M)) and four different chemistries of directly coupled
antibody. Additionally, biotinylated antibody and two immobilized streptavidin resins were used in an Indirect
IP method (antibody + sample preincubation before binding to solid support). (a) Capture efficiency was determined by Western blot. (b) EGFR sequence coverage and background proteins were determined by LC-MS/MS
after IP elutions and trypsin digestion. IP using streptavidin magnetic beads resulted in fewer background
proteins identified and higher EGFR sequence coverage
Presented here are two IP strategies that minimize both the

amount of nonspecific protein binding to the immunoaffinity
matrix and the amount of antibody that leaches into the eluted
target. The first involves use of a biotinylated antibody and streptavidin magnetic beads. The biotin-labeling process results in multiple biotin modifications per antibody molecule, thus providing
multiple contacts or binding points to the streptavidin beads.
Because the biotinstreptavidin interaction is extremely high affinity (KD = 1015 M) and highly resistant to acid elution, the quantity
of antibody chains found in the eluent is significantly less than a
classical IP with Protein A or G (Fig. 3). Tactically, the use of biotinylated antibodies in IPs is advantageous since biotin labeling is
compatible with any antibody species and subtype and utilizes a
common streptavidin support matrix. This simplifies the overall
workflow and makes the biotinstreptavidin IP approach amenable
to multiplexing (e.g. immunoprecipitating and quantifying multiple targets within a cellular signaling pathway). In addition, the
biotin/streptavidin approach is preferred with clinical samples such
as serum or other biological fluid which normally contains
endogenous antibodies that would interfere with binding to the
classical Protein AG immunoprecipitation support.
138
AG
+ - + - Direct
- + - + Indirect
Heavy
chain
PP2A
Light
chain
b
Streptavidin
Magnec
Beads
MS results
% Sequence
Coverage
Protein AG
Magnec Beads
IP/Co-IP targets
PP2A, catalytic
subunit (PPP2CA
and PPP2CB)
PP2A, regulatory
subunit (PPP2R1A
and PPP2R1B)
PP4, catalytic
subunit (PPP4C)
Indirect IP
Indirect IP
Direct IP
No Crosslink
Direct IP
With
Crosslink
57%
69%
58%
42%
14%
11%
5%
11%
42%
42%
35%
21%
Fig. 3 Comparison of different IP methods for MS application. (a) Antibody heavy and light chain contamination
is significantly reduced in the Protein AG cross-link and the biotinylated antibody/streptavidin methods compared to the classical Protein AG IP methodology. (b) Indirect and direct IP methods were used to evaluate
enrichment of PP2A, catalytic subunit (PPP2CA/PPP2CB) from 0.5 mg A431 cell lysate using Streptavidin
magnetic beads and Protein AG magnetic beads. IP elutions were processed by in-solution digestion method
and analyzed by LC-MS/MS to assess sequence coverage and identify isoform-specific peptides. MS analysis
of IP methods using streptavidin and Protein AG magnetic bead supports allowed for identification of two PP2A
catalytic subunits (PPP2CA and B), two PP2A regulatory subunits (PPP2R1A and B) and PP4 catalytic subunit
(PPP4C). Protein AG cross-link IP method resulted in slight decrease in PPP2CA recovery as indicated by
Western blot and by the percent sequence coverage obtained in the mass spectrometry analysis
The second IP strategy is an oriented, cross-linked affinity

method [25, 3234] that uses Protein AG magnetic beads to capture the antibody similar to a classical IP however, before the antigen is introduced, the antibody is covalently cross-linked to the
Protein AG using a homobifunctional cross-linker such as
DSS. This cross-linking virtually eliminates the presence of IgG
heavy and light chain contamination in the eluent (Fig. 3). The
oriented, cross-linked approach is a good option if the antibody
contains a carrier protein as the binding to Protein AG serves as a
purification step before the cross-linking event. Because Protein
IP-MS
139
Table 1
Detection and quantification of low-abundant targets by MS
Detected by LC-MS/MS
Quantified by LC-SRM/MS
Target
Cell line
Neat
Enriched-IP
Neat
Enriched-IP
EGFR
A431
HEK293
+
+
+
+
AKT1
A431
HEK293
+
+
+
+
AKT2
A431
HEK293
+
+
+
+
AKT3
A431
HEK293
+
+
N/A
N/A
N/A
N/A
PTEN
A431
HEK293
+
+
+
+
PIK3CA
A431
PIK3R2
A431
KRAS
A431
N/A
N/A
STAT1
A431
N/A
N/A
PPP2AC
A431
N/A
N/A
N/A not available
AG combines the binding characteristics of both Protein A and

Protein G, this IP approach also works with a variety of antibody
species and subtypes. Classical IP without cross-linking of antibody
to Protein AG can always be performed if the presence of antibody
chains in the eluent is not a concern (e.g. if performing targeted
detection such as SRM assays).
IP enrichment is essential to identifying low abundance targets
in a complex biological milieu. Many cell signaling proteins and
biomarkers are low copy number and are not detected in crude
lysates or biological fluids due to the presence of a multitude of
other proteins competing for ionization even after LC fractionation. However, with upstream IP enrichment, low-abundant signaling proteins such as EGFR, AKT1, AKT2, PTEN, PIK3CA,
PIK3R2, KRAS, and STAT1 can be detected and successfully identified (Table 1). While other biochemical methods of affinity purification upstream of mass spectrometry (AP-MS) exist [35, 36],
immunoaffinity purification is used most often for specific enrichment of endogenous protein(s) and is the focus of the protocols
described in this chapter.
140
Materials
1. Pierce Streptavidin Magnetic Beads (Thermo Fisher Scientific,
Rockford, IL). Supplied as 10 mg/ml; binding capacity
~55 g biotinylated rabbit IgG per mg of beads.
2. N-hydroxysuccinimide-PEG4-biotin
(NHS-PEG4-biotin),
supplied as No-Weigh Format (8 2 mg in microtubes;
MW = 589) (Thermo Fisher Scientific): To make a 20 mM
solution, add 170 l DMF to one 2 mg reagent tube by puncturing the foil seal with the pipette tip. Pipette up and down
to mix. Use immediately (see Note 1).
3. Phosphate-buffered saline (PBS): 0.1 M sodium phosphate,
0.15 M NaCl, pH 7.5.
4. Zeba desalting column (0.5 ml; 7K cut-off) (Thermo Fisher
Scientific).
5. IP Lysis Buffer (Thermo Fisher Scientific): 25 mM TrisHCl,
pH 7.4, 150 mM NaCl, 1 mM EDTA, 1 % NP-40 and 5 %
glycerol.
6. Binding/Wash Buffer: 25 mM ammonium bicarbonate,
pH 8.0.
7. LC-MS grade water.
8. Elution Buffer: 0.5 % formic acid (LC-MS grade), 30 % acetonitrile (LC-MS grade).
9. Pierce Protein AG Magnetic Beads (Thermo Fisher Scientific).
Supplied as 10 mg/ml; binding capacity 5585 g rabbit IgG
per mg magnetic particles.
10. Disuccinimidyl suberate solution (DSS) (Thermo Fisher
Scientific): Dissolve 2 mg DSS in 150 l DMSO or DMF
immediately before use. Once reconstituted, the DSS must be
used immediately. DSS is moisture sensitive and must be
stored desiccated.
11. Modified PBS: PBS containing 0.05 % NP-40 (see Note 2).
12. 0.1 M glycine-HCl, pH 2.0.
13. Denaturation buffer: 6 M Urea, 50 mM TrisHCl, pH 8.0 (see
Note 3). Add 0.36 g urea (sequencing grade) in 0.5 ml
50 mM TrisHCl, pH 8.0, to dissolve then bring volume to
1 ml. Do not warm up to dissolve. Prepare fresh for each use.
14. 0.5 M Bond-Breaker TCEP Solution (Thermo Fisher
Scientific): Prepare a 1:50 working dilution by adding 10 l
0.5 M TCEP to 490 l 50 mM TrisHCl, pH 8.0.
15. 0.5 M Iodoacetimide: Dissolve 37 mg iodoacetamide in 400 l
MS-grade water. Make fresh and cover with foil to protect
from light.
IP-MS
141
16. Pierce Trypsin Protease (MS grade): Dissolve 20 g trypsin in

200 l 50 mM acetic acid to prepare a 0.1 g/l stock trypsin
solution. Store aliquots at 80 C. Prepare (just before use)
10 ng/l trypsin working solution by adding 10 l 0.1 g/l
trypsin stock to 90 l 50 mM TrisHCl, pH 8.0.
17. Pierce C18 Spin Tips (Thermo Fisher Scientific). Each tip
binds up to 10 g (20 l) of total peptide.
18. Trifluoroacetic acid (TFA) (LC-MS grade).
19. Low protein-binding microcentrifuge tubes (Axygen or
Eppendorf).
20. Magnetic stand (e.g. DynaMag-2 Magnet from Life
Technologies).
21. Thermomixer.
22. Speed vac.
23. Thermo Scientific Orbitrap XL mass spectrometer or
Thermo Scientific Orbitrap Fusion Tribrid mass spectrometer equipped with Thermo Scientific Proteome
Discoverer 1.4 software, Scaffold 4.0 (Proteome Software),
and SEQUESTHT search engine (University of Washington).
Methods
3.1 Biotinylation
of Antibodies
and MagneticStreptavidin
Immunoprecipitation
The most common approach to biotin-label antibodies is to use

N-hydroxysucciminimide (NHS) esters. At pH 79, NHS-biotin
esters will react specifically with the primary amines in lysine side
chains and the N-terminus, creating a stable amide bond. However,
the number of biotins incorporated per antibody should be controlled to prevent inactivation of primary amines that may be
required for antigen recognition and/or aggregation and precipitation due to over-biotinylation. Utilizing biotinylation reagents
with polyethylene glycol (PEG) spacer arms imparts greater water
solubility to biotin-labeled antibodies, helping to prevent aggregation. Using the appropriate molar excess of NHS ester over antibody helps titrate the number of biotin labels introduced. In the
following protocol, a 40-fold molar excess of biotin was found to
result in approximately 212 biotin moieties per antibody, a quantity that was found to maximize binding of an antibody on streptavidin beads, preserve antibody function, and minimize leaching of
antibody into the eluent due to multiple contact points with the
support.
1. Calculate the millimoles of NHS-PEG4-Biotin to add to the
labeling reaction to give a 40-fold molar excess over antibody:
mg antibody
mmol antibody 40 mmol biotin
mmol biotin.
mg antibody
mmol antibody
142
2. Calculate the microliters of 20 mM NHS-PEG4-Biotin to add

to each labeling reaction:
mmol biotin
589 mg
170 l
l biotin solution.
mmol biotin 2 mg
3. Mix 50200 g antibody in PBS (see Note 4) with a 40-fold

molar excess of NHS-PEG4-biotin in a total reaction volume
of 100 l. Pipette gently to mix.
4. Incubate the biotin labeling reaction at room temperature for
30 min. Remove excess non-reacted and hydrolyzed biotin
reagent by desalting spin column.
5. Centrifuge a 0.5 ml desalting spin column at 1,500 g for
1 min to remove the storage solution.
6. Equilibrate the desalting column by washing three times with
300 l PBS.
7. Place desalting column in a fresh collection tube and add the
reaction gently to the top of the resin bed.
8. Centrifuge at 1,500 g for 2 min. Biotinylated antibody is
present in the flow-through and is now in PBS (see Note 5).
9. Prepare cell or tissue lysates in IP Lysis Buffer to a final concentration of 1 mg/ml (see Note 6). Sample size per IP reaction is typically 500 l (500 g). For serum or plasma samples,
dilute 20 l (~1 mg) of serum or plasma in 480 l IP Lysis
Buffer.
10. For optimal results, pre-clear 500 l of the lysate sample by
incubating with 25 l of streptavidin magnetic beads for 1 h at
room temperature (optional) (see Note 7).
11. Prepare the immune complex by incubating above lysate with
5 g of antibody overnight at 4 C with continuous end-overend rotation (see Note 8).
12. Dispense 25 l streptavidin magnetic beads (for each IP reaction) into a low-binding microcentrifuge tube, collect the
beads on a magnetic stand for 1 min and remove the storage
solution. Wash the beads twice with IP Lysis Buffer, mixing
gently each time to fully suspend the beads. Collect the beads
between each wash by placing the tubes on a magnetic stand
for 1 min. Discard the supernatants (wash volumes).
13. Add the immune complex prepared in step 11 to the washed
streptavidin-magnetic beads and incubate at room temperature for 1 h with end-over-end rotation.
14. Collect the beads on a magnetic stand for 1 min.
15. Gently remove the supernatant (flow-through or non-bound)
and save for analysis if desired.
IP-MS
143
16. Wash the beads with 3 500 l 25 mM Ammonium bicarbonate, pH 8, by gently vortexing the beads to resuspend during
each wash step (see Note 9).
17. Wash the beads with 2 500 l of LC-MS grade water, by
gently vortexing the beads to resuspend during each wash step
(see Note 10).
18. Elute bound antigen by adding 100 l elution buffer (0.5 %
formic acid, 30 % acetonitrile), and incubating for 5 min
at room temperature with periodic gentle vortexing
(see Note 11).
19. Collect the beads on a magnetic stand and transfer the elution
to a fresh microcentrifuge tube.
20. Dry the elution in a speed vac.
21. Proceed to mass spec sample prep (Subheading 3.3) and/or
resuspend in SDS-PAGE sample buffer for Western blot analysis/validation.
3.2 Cross-linked
IP Method
An alternative strategy to minimize antibody leaching would be to

physically cross-link the antibody to the affinity support either
directly to a chemically-activated support or to an immobilized
Protein AG bead. The cross-linked Protein AG approach results in
an oriented antibody to maximize antigen capture and is the preferred strategy when the antibody contains a carrier protein such as
BSA or gelatin. A common homobifunctional cross-linker such as
disuccinimidyl suberate (DSS) reacts with amine groups on the
antibody and those on the Protein AG support. As with the use of
NHS-biotin esters, DSS must be used at the appropriate concentration to minimize inactivation of essential amines in the antigen
recognition site that would lead to loss of antibody function. Small
decreases in antigen binding are commonly observed with the
cross-linked IP method, but this loss of function may not be fully
recognized in practice as the immobilized antibody beads are generally still in excess over the target antigen [32, 33] (see Note 12).
1. Vortex the bottle of Pierce Protein AG Magnetic Beads to
obtain a homogeneous suspension. Dispense 25 l of magnetic bead suspension into a low-binding microcentrifuge
tube (see Note 13). Place tube on a magnetic stand for 1 min
to collect the beads. Remove and discard the storage
solution.
2. Wash the beads twice with 500 l Modified PBS, mixing gently each time to fully suspend the beads. Collect the beads
between each wash by placing the tubes on a magnetic stand
for 1 min. Discard the supernatants (wash volumes).
3. Dilute antibody in Modified PBS to a final concentration of
5 g per 100 l (see Note 14).
144
4. Add 100 l of diluted antibody solution to the washed magnetic

beads and incubate at room temperature for 15 min, gently
vortexing the beads every 510 min during incubation to
ensure that the beads stay in suspension.
5. Collect the beads with a magnetic stand. Remove and discard
the supernatant.
6. Wash the bound antibody-bead complexes twice with 300 l
Modified PBS. For each wash, thoroughly mix to suspend the
beads, collect on a magnetic stand for 1 min, then remove and
discard each wash volume.
7. Resuspend the beads in 46 l PBS. Add 4 l of 0.25 mM DSS
in DMSO or DMF (20 M final concentration; 10X molar
excess DSS over immobilized Protein AG) (see Note 15).
8. Incubate the cross-linker with antibody-Protein AG beads for
30 min at room temperature, vortexing gently every 1015 min
during the incubation to ensure an even bead suspension.
9. Collect the beads with a magnetic stand. Remove the
supernatant.
10. Quench the cross-linking reaction and remove non-crosslinked antibody by adding 100 l 0.1 M glycine buffer, pH 2
to the beads and gently mixing for 5 min at room temperature
(see Note 16). Collect the beads with a magnetic stand and
remove the supernatant.
11. Perform two additional 100 l washes with glycine buffer.
12. Equilibrate the cross-linked beads by washing twice with
200 l cold IP Lysis/Wash Buffer each wash.
13. Prepare cell or tissue lysates in IP Lysis Buffer to a final concentration of 1 mg/ml (see Note 6). Sample size per IP reaction is typically 500 l (500 g). For serum or plasma samples,
dilute 20 l (~1 mg) of serum or plasma in 480 l IP Lysis
Buffer.
14. Add 500 l lysate to the cross-linked magnetic beads and incubate for 1 h at room temperature with end-over-end mixing.
Alternatively, vortex the beads every 1015 min during the
incubation to ensure that the beads stay in suspension.
15. Collect the beads on a magnetic stand for 1 min.
16. Gently remove the supernatant (flow-through or non-bound)
and save for analysis if desired.
17. Wash the beads with 3 500 l 25 mM Ammonium bicarbonate, pH 8, by gently vortexing the beads to resuspend during
each wash step (see Note 9).
18. Wash the beads with 2 500 l of LC-MS grade water, by
gently vortexing the beads to resuspend during each wash step
(see Note 10).
IP-MS
145
19. Elute bound antigen by adding 100 l elution buffer (0.5 %

formic acid/30 % acetonitrile), and incubating for 5 min at room
temperature with periodic gentle vortexing (see Note 11).
20. Collect the beads on a magnetic stand and transfer the elution
to a fresh low-binding microcentrifuge tube.
21. Dry the elution in a speed vac.
22. Proceed to mass spec sample prep (Subheading 3.3) and/or
resuspend in SDS-PAGE sample buffer for Western blot analysis/
validation.
3.3 Mass Spec
Sample Prep and Data
Analysis
The proteins present in the IP elution fractions are typically denatured, reduced, alkylated, and digested into peptides prior to detection by mass spectrometry, because peptides fractionate, ionize,
and fragment more efficiently than intact proteins. In addition, the
resulting peptide mass spectra are easier to interpret during the protein
identification process. Because the protein amount in the IP eluent
is small, an in-solution digestion workflow is preferred as it minimizes sample loss, is fast and is amenable to high-throughput and
automation. Samples are first denatured in a strong chaotrope such
as urea to facilitate complete digestion. Disulfide bonds are reduced
with tris(2-carboxyethyl) phosphine (TCEP) or dithiothreitol
(DTT) and bond reformation is prevented by alkylating free sulfhydryl groups with iodoacetamide. The fully denatured proteins are
then digested with trypsin, although other proteases and protease
combinations can be used (e.g. chymotrypsin, Glu-C, Lys-C). The
resulting peptides are applied to C-18 tips or columns to remove
salts and buffers and to concentrate the sample prior to analysis by
nano or capillary reversed-phase liquid chromatography (LC), electrospray ionization, and tandem mass spectrometry (MS/MS).
The peptides are identified by sequence database searching using
the search algorithm SEQUEST or Mascot. The significance of
each peptide and protein identification is estimated using the software tools, Thermo Scientific Proteome Discoverer 1.4 and
Scaffold 4.0 software to assess percent sequence coverage, spectral
counts, and post-translational modification (PTMs).
1. Suspend dried samples in 10 l denaturation buffer (see Note 17).
2. Add 10 l of 10 mM TCEP to each sample and incubate at
37 C for 30 min. Centrifuge briefly to collect condensate to
bottom of tube.
3. To each 20 l sample of reduced protein add 0.83 l of 0.5 M
iodoacetamide solution (final concentration 20 mM). Incubate
at room temperature for 30 min in the dark (cover sample
tubes with foil OR place samples in a drawer).
4. Add 45 l of 50 mM TrisHCl, pH 8.0 to dilute urea concentration to <1 M.
146
5. Add 3 l of 10 ng/l trypsin to each sample (30 ng trypsin per

sample). Briefly vortex and digest overnight at 37 C in a thermomixer at 500 rpm. Centrifuge briefly to collect condensate
to bottom of tube.
6. Acidify samples by adding 2 l of 10 % TFA (pH <3) and vortex briefly.
7. Centrifuge at 15,000 g for 2 min to pellet the insoluble
sample.
8. Insert Pierce C18 spin tip into spin adapter seated in a labeled
1.5 ml low-binding microcentrifuge tube (one for each sample).
9. Wet each C18 tip by adding 20 l of 80 % acetonitrile/0.1 %
TFA and centrifuge at 1,000 g for 1 min. Repeat one additional time.
10. Equilibrate tip by adding 20 l of 0.1 % TFA and centrifuge at
1,000 g for 1 min. Repeat one additional time.
11. Apply 68 l of each digested sample to their corresponding
C18 tip and collect flow-through in the new labeled 1.5 ml
low-binding microcentrifuge tube by centrifuging at 1,000 g
for 3 min (see Note 18).
12. Reapply samples to their designated C18 tips and spin again at
1,000 g for 3 min.
13. Wash each C18 tip by adding 20 l of 0.1 % TFA at 1,000 g
for 1 min. Repeat one additional time.
14. Elute the samples into new 1.5 ml low-binding microcentrifuge tubes using 2 20 l of 80 % acetonitrile/0.1 % TFA and
centrifuging at 1,000 g for 1 min.
15. Speed vac samples to near dryness and resuspend samples with
30 l of 5 % acetonitrile/0.1 % formic acid before MS Analysis.
16. Take out 15 l of sample into autosampler vial and perform
LC-MS/MS analysis on an Orbitrap XL mass spectrometer or
equivalent.
Notes
1. The NHS-ester moiety readily hydrolyzes and becomes nonreactive. Do not prepare solutions for storage. Discard any
unused reconstituted reagent. Suspending the NHS-ester in a
dry solvent such as DMF or DMSO instead of water will slow
the hydrolysis rate, but storage should be limited to <1 month
at 20 C.
2. A very low percentage of detergent facilitates magnetic bead
collection and handling. NP-40 can be substituted with
Tween-20 if desired.
IP-MS
147
3. For in-solution digestion methods, detergents cannot be used

in the denaturation process since they will interfere in subsequent MS analysis. Instead of 6 M urea, other common denaturants that are used for in-solution digestion include 6 M
guanidine-HCl, 10 % sodium deoxycholate and 50 %
2,2,2-Trifluoroethanol (TFE). In order for trypsin to work
most efficiently, the concentration of the denaturants need to
be at a compatible final concentration (<1 M urea, <0.5 M
guanidine-HCl, <1 % sodium deoxycholate, and <5 % TFE).
4. For the biotinylation reaction, the antibody must be in an
amine-free buffer (such as PBS; avoid Tris or glycine buffers)
with no carrier protein (BSA or gelatin) present. Proteins or
other amine-containing molecules will interfere with aminereactive biotin labeling reagents. For optimal results, use an
affinity-purified antibody. If the antibody is not in an aminefree buffer such as PBS, perform buffer exchange by dialysis or
desalting column. Remove carrier proteins from commerciallyavailable antibodies by Protein AG purification or using a
commercially available antibody purification kit such as Pierce
Antibody Clean-up Kit (Thermo Fisher Scientific).
5. If desired, the extent of antibody biotinylation can be determined using a commercially-available kit such as the Fluorescence
Biotin Quantitation Kit from Thermo Fisher Scientific.
6. IP Lysis Buffer has been tested on representative cell types
including, but not limited to, HeLa, Jurkat, A431, A549,
MOPC, NIH 3T3, and U2OS. Typically, 106 HeLa cells yield
~10 mg of cell pellet and ~3 g/l (or 300 g) total protein
when lysed with 100 l of IP Lysis Buffer. IP Lysis Buffer is
compatible with the Pierce BCA Protein Assay (Thermo
Fisher Scientific). To minimize protein degradation and preserve phosphorylation modifications during lysate preparation, use protease and phosphatase inhibitors (e.g. Thermo
Scientific Halt Protease and Phosphatase Inhibitor
Cocktail) in the IP Lysis Buffer.
7. Pre-clearing cell lysates or tissue homogenates by incubating
with streptavidin beads in the absence of antibody helps remove
certain biotin-containing enzymes such as carboxylases and
decarboxylases that would normally bind directly to the streptavidin support independent of the presence of antibody, contributing to background.
8. The indirect method (binding antibody to antigen overnight
to form the immune complex followed by incubation with
streptavidin beads in a subsequent step) is the preferred IP
method when using low affinity antibodies or enriching for
very low abundant targets. Alternatively, the antibody-bead
complex can be assembled first and then incubated with the
148
protein sample (direct method). The direct method is faster,

but formation of the antibodyantigen complex is somewhat
less efficient due to potential mixing inefficiencies and the
need to abbreviate the incubation time of magnetic beads with
lysates to minimize nonspecific protein binding to the solid
support that would normally occur with an overnight incubation. The direct method works well with medium-abundant
targets and is more amenable to automation.
9. Typical IP wash buffers contain detergent or salt to wash away
weak, nonspecifically-bound proteins. However, salts and
detergents are not compatible with downstream mass spec
analysis. Ammonium bicarbonate is used in place of these wash
buffers because it is volatile, and thus removed, during subsequent speed vac steps.
10. Water washes are performed to eliminate any residual detergent from the IP Lysis buffer used to prepare the cell or tissue
lysate and to remove any residual buffering capacity before the
acid elution step. LC-MS grade water has low UV absorbance
to ensure maximum sensitivity at all detection wavelengths.
11. The elution buffer formulation described is fairly stringent and
highly effective (boiling the magnetic beads in a denaturing
SDS-PAGE loading buffer shows that all of the antigen was
efficiently eluted with the 0.5 % formic acid/30 % acetonitrile). However, very slight contamination by antibody chains
was observed in the elution due to a small degree of antibody
denaturation or a weakening of the biotinstreptavidin interaction. Some antibodyantigen interactions can be efficiently
disrupted with less acetonitrile (510 %) with virtually no antibody chain contamination, but elution efficiency is antibody
antigen dependent.
12. If the antibodyantigen interaction is known to be very low
affinity or if the antigen is in very low abundance, the IP
efficiency of the oriented, cross-linked approach may be significantly compromised not only by the DSS cross-linking but
also by the short (1 h) antibody-bead incubation with lysate
sample utilized in the direct IP method. If the binding function of the antibody cannot be restored by titrating the DSS to
a lower concentration (see Note 15), then either a longer
(overnight at 4 C) antibody-bead + lysate incubation may be
necessary (but may result in higher background). Alternatively,
a classical IP with an antibody + lysate overnight incubation at
4 C followed by 1 h incubation with Protein AG magnetic
beads and no cross-linking may be the best option. This will
result in considerable antibody chain contamination in the
elution but the target protein of interest still may be able to be
detected if it ionizes well.
IP-MS
149
13. The protocol is written for the binding and cross-linking of

beads and antibody for a single IP reaction (25 l magnetic
bead slurry/reaction). However, this procedure can be scaled
up to prepare a larger batch of cross-linked antibody-beads
which can be dispensed into individual IP reactions (replicates
and controls) before incubation with sample. Scale up antibody, DSS, and wash quantities proportionately.
14. Although serum may be used, the antibody that is specific for the
antigen of interest may comprise only 12 % of the total IgG in
the serum sample and will result in low antigen recovery.
15. If it is known that epitope recognition is mediated by lysine
residues in the antibodys antigen recognition site, it may be
necessary to titrate the amount of DSS used to minimize antibody inactivation. This is accomplished by performing a twofold dilution series of DSS during the antibody immobilization
reaction. Because DSS is a hydrophobic molecule, a microprecipitate may form when it is added to an aqueous medium,
which results in a cloudy appearance. The reaction will still
proceed efficiently and may disappear during conjugation.
16. Glycine at pH 2 will elute IgG that is not covalently coupled
to the immobilized Protein AG. Most polyclonal and most
monoclonal antibodies can tolerate low pH conditions for
short durations. If the antibody immobilized is intolerant of
low pH conditions, wash the beads with a high salt, neutral
pH buffer to remove uncoupled antibody instead.
17. If urea is used for denaturing proteins, the sample should not
be boiled before reducing. Urea can carbamylate proteins at
high temperatures. Incubation at 37 C for 30 min is sufficient
for denaturing the proteins.
18. For binding to C18 reversed-phase sorbents, a sample must be
free of excess organic solvents such as acetonitrile (ACN) or
methanol. For optimal flow and peptide recovery, do not completely dry the C18 tips after equilibration by excessive centrifugation times or speed.
References
1. Humpert ML, Tzouros M, Thelen S et al
(2012) Complementary methods provide evidence for the expression of CXCR7 on human
B cells. Proteomics 12:19381948
2. Sanders RL, Spector DH (2010) Human cytomegalovirus IE2 86 and IE2 40 proteins differentially regulate UL84 protein expression
posttranscriptionally in the absence of other
viral gene products. J Virol 84:51585170
3. Soul J, Alme M, Myrum C et al (2012)
Balancing Arc synthesis, mRNA decay, and
proteasomal degradation: maximal protein
expression triggered by rapid eye movement

sleep-like bursts of muscarinic cholinergic
receptor stimulation. J Biol Chem 287:
2235422366
4. Tanioka T, Tamura Y, Fukaya M, Shinozaki S
et al (2011) Inducible nitric-oxide synthase
and nitric oxide donor decrease insulin receptor substrate-2 protein expression by promoting proteasome-dependent degradation in
pancreatic -cells: involvement of glycogen
synthase kinase-3. J Biol Chem 286(33):
2938829396
150
5. Grnborg M, Kristiansen TZ, Stensballe A

et al (2002) A mass spectrometry-based proteomic approach for identification of serine/
threonine-phosphorylated proteins by enrichment with phospho-specific antibodies. Mol
Cell Proteomics 1(7):517527
6. Stokes MP, Farnsworth CL, Moritz A et al
(2012) PTMScan Direct: identification and
quantification of peptides from critical signaling proteins by immunoaffinity enrichment
coupled with LC-MS/MS. Mol Cell
7. Ghafari M, Hger H, Falsafi K et al (2012) Mass
spectrometrical identification of hippocampal
NMDA receptor subunits NR1, NR2A-D and
five novel phosphorylation sites on NR2A and
NR2B. J Proteome Res 11:18911896
8. Li Y, Silva JC, Skinner ME, Lombard DB
(2013) Mass spectrometry-based detection of
protein acetylation. Methods Mol Biol 1077:
81104
9. Guo A, Gu H, Zhou J et al (2014)
Immunoaffinity enrichment and mass spectrometry analysis of protein methylation. Mol
10. Jochmann R, Pfannstiel J, Chudasama P et al
(2013) O-GlcNAc transferase inhibits KSHV
propagation and modifies replication relevant
viral proteins as detected by systematic
O-GlcNAcylation analysis. Glycobiology 10:
11141130
11. Hsiao HH, Meulmeester E, Urlaub H (2012)
Identification of endogenous SUMO1 accepter
sites by mass spectrometry. Methods Mol Biol
893:431441
12. Schwertman P, Bezstarosti K, Laffeber C et al
(2013) An immunoaffinity purification method
for the proteomic analysis of ubiquitinated
protein complexes. Anal Biochem 440:
227236
13. Sun X-X, Challagundla KB, Dai M-S (2012)
Positive regulation of p53 stability and activity
by the deubiquitinating enzyme Otubain 1.
EMBO J 31:576592
14. Borch J, Roepstorff P, Mller-Jensen J
(2011) Nanodisc-based co-immunoprecipitation for mass spectrometric identification of
membrane-interacting proteins. Mol Cell
Proteomics 10(7):O110.006775. doi:10.
1074/mcpO110.006775
15. Penn JKM, Graham P, Deshpande G et al
(2008) Functioning of the Drosophila WilmsTumor-1-associated protein homolog, F1(2)d,
in sex-lethal-dependent alternative splicing.
Genetics 178:737748
16. Turriziani B, Garcia-Munoz A, Pilkington R
et al (2014) On-bead digestion in conjunction
with data-dependent mass spectrometry: a
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
shortcut to quantitative and dynamic interaction proteomics. Biology 3:320332

Vasilescu J, Guo X, Kast J (2004) Identification
of protein-protein interactions using in vivo
cross-linking
and
mass
spectrometry.
Mitchell L, Huard S, Cotrut M et al (2013)
mChIP-KAT-MS, a method to map protein
interactions and acetylation sites for lysine
acetyltransferases. Proc Natl Acad Sci U S A
110(17):E1641E1650
Huang BX, Kim H-Y (2013) Effective identification of Akt interacting proteins by two-step
chemical crosslinking, co-immunoprecipitation
and mass spectrometry. PLoS One 8(4):e61430.
doi:10.1371/journal.pone.0061430
Trinkle-Mulcahy L, Boulon S, Lam YW et al
(2008) Identifying specific protein interaction
partners using quantitative mass spectrometry
and bead proteomes. J Cell Biol 183:223239
Zhong J, Kim M-S, Chaerkady R et al (2012)
TSLP signaling network revealed by
SILAC-based phosphoproteomics. Mol Cell
Proteomics 11(6):M112.017764. doi:10.1074/
mcp.M112.017764
Culver BP, Savas JN, Park SK et al (2012)
Proteomic analysis of wild-type and mutant
Huntingtin-associated proteins in mouse
brains identifies unique interactions and
involvement in protein synthesis. J Biol Chem
287:2159921614
Zhang X, Lavoie G, Fort L et al (2012) Gab2
phosphorylation by RSK inhibits Shp2 recruitment and cell motility. Mol Cell Biol
33:16571670
Vogt A, Fuerholzner B, Kinkl N et al (2013)
Isotope coded protein labeling coupled immunoprecipitation (ICPL-IP): a novel approach for
quantitative protein complex analysis from native
tissue. Mol Cell Proteomics 12(5):13951406
Boulon S, Ahmad Y, Trinkle-Mulcahy L et al
(2010) Establishment of a protein frequency
library and its application in the reliable identification of specific protein interaction partners.
Wang Q, Chaerkady R, Wu J et al (2011)
Mutant proteins as cancer-specific biomarkers.
Proc Natl Acad Sci U S A 108:24442449
Lee AYH, Yates NA, Ichetovkin M et al (2012)
Measurement of fractional synthetic rates of
multiple protein analytes by triple quadrupole
mass spectrometry. Clin Chem 58:619627
Tong J, Taylor P, Peterman SM et al (2009)
Epidermal growth factor receptor phosphorylation sites Ser991 and Tyr998 are implicated in
the regulation of receptor endocytosis and
phosphorylations at Ser1039 and Thr1041. Mol
IP-MS
29. Wu C, Wei W, Li C et al (2012) Delicate analysis of post-translational modifications on
Dishevelled 3. J Proteome Res 11(7):
38293837
30. Rardin MJ, Held JM, Gibson BW (2013)
Targeted quantitation of acetylated lysine peptides by selected reaction monitoring mass
spectrometry. Methods Mol Biol 1077:
121131
31. Herzog F, Kahraman A, Boehringer D et al
(2012) Structural probing of a protein phosphatase 2A network by chemical cross-linking
and mass spectrometry. Science 337:
13481352
32. Kaboord B, Perr M (2007) Isolation of proteins and protein complexes by immunoprecipitation. In: Posch A (ed) Methods in
molecular biology, vol. 424. 2D PAGE: sample
33.
34.
35.
36.
151
preparation and fractionation. Humana,

Totowa, NJ, pp 379384
Qoronfleh MW, Ren L, Emery D et al (2003)
Use of immunomatrix methods to improve
protein-protein
interaction
detection.
J Biomed Biotechnol 5:291298
Kan A, Mohamedali A, Tan SH et al (2013)
An improved method for the detection and
enrichment of low-abundant membrane and
lipid raft-residing proteins. J Proteomics 79:
299304
Rigaut G, Shevchenko A, Rutz B et al (1999)
A generic protein purification method for protein complex characterization and proteome
exploration. Nat Biotechnol 17:10301032
Gingras A-C, Gstaiger M, Raught B, Aebersold
R (2007) Analysis of protein complexes using
mass spectrometry. Nat Rev 8:645654
Chapter 13
Principles of Protein Labeling Techniques
Christian Obermaier, Anja Griebel, and Reiner Westermeier
Abstract
Protein labeling methods prior to separation and analysis have become indispensable approaches for proteomic
profiling. Basically, three different types of tags are employed: stable isotopes, mass tags, and fluorophores.
While proteins labeled with stable isotopes and mass tags are measured and differentiated by mass
spectrometry, fluorescent labels are detected with fluorescence imagers. The major purposes for protein
labeling are monitoring of biological processes, reliable quantification of compounds and specific detection
of protein modifications and isoforms in multiplexed samples, enhancement of detection sensitivity, and
simplification of detection workflows. Proteins can be labeled during cell growth by incorporation of
amino acids containing different isotopes, or in biological fluids, cells or tissue samples by attaching specific
groups to the -amino group of lysine, the N-terminus, or the cysteine residues. The principles and the
modifications of the different labeling approaches on the protein level are described; benefits and shortcomings of the methods are discussed.
Key words Protein labeling, Quantification, Protein detection, Stable isotopes, Mass tags, Fluorophores, Difference gel electrophoresis, Mass spectrometry, Fluorescence imager
Introduction
Proteomics profiling employs various methods for identification,
monitoring of structural changes, and quantification of expression
levels of proteins in complex mixtures. The most commonly applied
techniques are the separation by gel electrophoresis and chromatography, and further investigation by image analysis, Western
blotting, and mass spectrometry. Lately, a comprehensive overview
on these tools has been published for the example of environmental
microbiology [1].
The different proteomic profiling workflows possess various
methodological shortcomings, which could partly be prevented or
at least reduced by pre-labeling proteins prior to the analysis.
Furthermore, labeling of specific protein moieties offers the possibility of studying some dedicated protein modifications. In the
following we try to give a small overview on methodical issues,
153
154
Christian Obermaier et al.
where labeling of proteins can improve the quality of the results

and enable methodical developments.
The conventional approach for protein detection in gel electrophoresis is staining the proteins in the gels subsequently to the separation. A number of post-staining methods exist using organic and
fluorescent dyes, and silver staining [2]. While the latter shows the
highest sensitivity of detection, it has a very limited capacity for
quantification. Labeling of proteins prior to the separation with an
appropriate and bright fluorescent dye could increase sensitivity and
quantification features at the same time. Furthermore this would
enable direct detection after separation, and save time and work.
Proteomics samples are highly complex: biological fluids, cell
lysates, and tissue extracts contain thousands of different proteins
and peptides with very high structural diversities. In the original
profiling approach this challenge is met as follows: With twodimensional gel electrophoresis the protein mixture is individualized into several hundred to a few thousand fraction spots, of
which each contains one or a few different proteins. After punching out the spots of interest and digesting the proteins with trypsin
inside the gel plug, the resulting peptides are eluted and subjected
to mass spectrometry analysis. It is still an undisputed fact, that
two-dimensional electrophoresis affords the highest resolution of
all protein separation methods. The technology using immobilized
pH gradients for isoelectric focusing in the first and high-resolution
SDS polyacrylamide gel electrophoresis in the second dimension
has meanwhile been further developed to a robust method with a
high level of reproducibility [3]. However, there exist still little
experimental variations, which cause partial protein losses [4]. The
incorporation of an internal control into the samples during separation would equalize such effects and improve the quantification
data. This can, however, only be realized by labeling the proteins
of different samples with different fluorescent tags and working
according to a multiplexing concept.
Redox proteomics has gained great importance: Oxido-redox
protein modifications, especially of cysteines, link redox metabolism to biological structure and function [5]. One of the strategies would be labeling cysteine thiol residues of differently
oxidized samples with different fluorescent dyes to compare and
determine the oxido-redox status of biological material by using
2D electrophoresis.
In the complementary gel-free workflow the complete initial
protein mixture is digested with endopeptidases. This means that
thousands of protein fractions are converted into an even more
complex mixture: each protein is cleaved into about 30 peptides on
an average. In most cases the workflow is performed in an automated mode; reversed-phase liquid chromatography of the peptides
on-line connected to electrospray ionization mass spectrometry.
Due to this enormous heterogeneity of the peptide mixture the
Protein Labeling Techniques
155
contents of the chromatography peaks are very complex, resulting

in undersampling in the mass spectrometer: During the chromatography separation the mass spectrometer has only a limited time
window to produce the related peptide mass spectra, which are
therefore incomplete. In practice, repeated runs of the same sample
show varying peak spectra; reproducibility is compromised. It is
therefore very complicated to find differences in protein expression
between different biological samples: Several technical replicates
would be needed to cope with adequate statistical requirements.
Furthermore, the gel-free quantification procedures based on mass
spectrometry peaks of peptides suffer from the limitation that the
peak intensities of the peptides do not correlate with their abundances [1]. However, because the mass spectrometry peak intensities of identical peptides correlate with their relative abundances in
different samples [6], labeling of the sample proteins with different
stable isotopes or mass tags would facilitate the qualitative and quantitative assessment of relative protein abundances in a multiplexing
approach. These quantification issues can also be solved by labeling
the peptides after protein digestion using non-isobaric isotope
coded amino-reactive groups or isobaric mass tags, which allow
quantification during a tandem MS step.
Thus protein or peptide labeling is indispensable for protein
quantification.
The following review describes the principles of different labeling
techniques which are applied on intact proteins, not on the digestion
products, the peptides.
Materials
Generally, the use of high-quality electrophoresis/proteomic-grade
chemicals is paramount to achieving successful experiments.
Fluorescent dyes for pre-labeling are available from a number of
specialist suppliers and some fine chemicals companies. Also the
stable isotope labeling kits and the mass tag kits are marketed by
several fine chemical and mass spectrometry companies.
Methods
3.1 Labeling with

Fluorescent Dyes Prior
to Electrophoretic
Separations
The easiest accessible targets for covalently attaching a fluorescent

group to a protein are the -amino groups of the lysines and the
N-terminus. According to the UniProt database lysine represents
6 % of amino acids in the average proteins, whichstatistically
means that lysines are present in almost every protein. Coupling of
a reagent via the N-hydroxy-succinimide (NHS) ester to primary
amines is a robust and well established procedure. Furthermore,
it is not necessary to derivatize the proteins prior to the reaction.
156
In order to ensure accessibility to the -amino groups of the lysines,

ideally the protein samples are denatured with high molar chaotropes
like urea or a combination of urea and thiourea.
Labeling of proteins prior to electrophoresis with a fluorogenic
amino reactive label has several advantages over post-staining: no
washing and staining procedures are needed, the image can be
recorded without fixing of the proteins, thus the gels can be further processed with Western blotting. In this way the results are
obtained much faster; the method is more cost and environment
friendly and easier, no buffers and organic solvents for staining and
washing are required.
nl et al. [7] have applied the concept of pre-labeling the
proteins in their original method difference gel electrophoresis
(DIGE), which has become one of the most successful methods in
comparative proteomics. In the DIGE technique, cyanine-NHS
CyDyes with different excitation and emission wavelengths are
covalently bound to the protein lysine moieties for multiplex analysis with two-dimensional electrophoresis. After the different protein samples have been pre-labeled with Cy2, Cy3, or Cy5 they are
combined together, and the mixture is submitted to isoelectric
focusing followed by SDS polyacrylamide gel electrophoresis. The
three dyes have been selected to have sufficiently different fluorescent properties to prevent cross-talking during imaging. For
perfect co-migration of the same proteins to the identical gel positions the dyes are charge-matched and have similar molecular
weights. For labeling the lysine moieties with cyanine dyes minimal labeling is applied: Only a small portion, about 5 %, of the
proteins in the sample are labeled by limiting the amount of dye
per protein, which is about 400 pM dye per 50 g of protein. This
prevents the proteins from becoming too hydrophobic, and insures
that only singly labeled proteins will be detected. In this minimal
labeling concept, the amount of multiple labeled proteins is statistically below the detection level. The resulting spot pattern is the
same like the pattern obtained from non-labeled proteins. The
limit of detection of a single protein is about 25 pg. This means
that the sensitivity is similar to silver staining. Downstream analyses
with mass spectrometry or Western blotting are not affected,
because 95 % of the separated proteins are un-labeled, hence not
modified.
Imaging is performed by exciting the fluorophore at the optimum wavelength with laser, LED or white light with excitation
filter and collecting the light with higher wavelength emitted from
the fluorophore using an emission filter with narrow band pass.
The patterns created from the different samples can be displayed
on the computer screen with false color representation. By superimposing these images with an image editing program a preview
of the result is obtained, which shows up- and down-regulations
of protein expression levels and serves as a control for sample
157
preparation and labeling, as well as appropriate co-migration of

the proteins. For accurate qualitative and quantitative evaluation
dedicated 2D software has to be employed. Advanced software
packages perform the image analyses almost automatically, for
instance by image warping from different gels. The principle of
the DIGE method is displayed in Fig. 1a, b (from ref. 8).
Because up to three samples are run in the identical matrix,
gel-to gel-variations between the sample patterns are excluded,
and the individual spots are co-detected across the samples for the
calculation of protein abundance ratios. However, the most powerful feature of DIGE is the possibility of the incorporation of an
internal control. Therefore, a pooled internal standard is created
by mixing aliquots taken from each sample of the experiment. This
sample pool is labeled with one of the fluorophores, mostly Cy2,
and co-electrophoresed with each sample. Thus every protein in
the population appears on each gel. Inherent gel-to-gel variations
between gels are compensated. Gel-to-gel matching becomes
much easier than in conventional gels, because the patterns of the
internal standard need to be matched, which are resulting from the
identical sample mixture. For quantitative comparison, the spot
volume ratios of the samples are calculated related to the internal
standard spot volume. This approach makes it possible to compare
protein abundances across different gels with high accuracy.
Appropriate sample cleanup and the adherence to exact labeling
conditions are very important prerequisites to achieve optimum
results (see Note 1). The principle of the DIGE method is displayed in Fig. 1a, b. Meanwhile a quadruplex kit with an additional
dye molecule for the detection close to infrared light is available,
which allows comparison of up to four different samples within
one 2D gel.
Alternatively, the fluorescent dyes can be attached to the thiol
group of the cysteine via a maleimide reactive group. Cysteines are
less abundant than lysine, which allows saturation labeling without
the risk of creating protein modifications with too high hydrophobicity. This offers a very high sensitivity of detection. Cysteine
labeling dyes are charge-neutral, and they have also similar molecular weights. Because all available cysteines will receive a label, the
proteins containing many cysteines will become multiply labeled:
migrate markedly slower in the SDS gel than unlabeled proteins,
and their light emission will be disproportionaly high. Proteins
without any cysteine cannot be detected. Thus the spot patterns
are different from those achieved with unlabeled or lysine-labeled
proteins. The efforts are higher than in lysine labeling: Prior to
labeling the disulfide bridges of the proteins must be cleaved with
a reductant like Triscarboxyethylphosphine (TCEP) or dithiothreitol (DTT) (see Note 2). For cysteine labeling only two dyes are
available: Cy3 and Cy5. The internal standard is usually labeled
with Cy3, the sample with Cy5. But the detection sensitivity of this
158
Fig. 1 (a) Schematic drawing of the DIGE 2-D electrophoresis method principle employing protein labeling with
fluorophores. The sample proteins are pre-labeled with Cy3 and Cy5. An internal standard is created by applying a mixture of all samples labeled with the third dye Cy2. (b) False color representation of the images of
samples and an internal standard, which have been co-separated in the same gel by DIGE. The images have
been acquired with a Typhoon multifluorescent laser scanner (GE Healthcare LifeSciences). (A) Cy2 channel:
pooled standard composed of the two samples; (B) Cy3 channel: mouse liver proteins, control; (C) Cy5 channel:
mouse liver proteins, treated; (D) overlay of all three images; the arrows point to a protein spot which is upregulated in the treated sample; (E) three-dimensional view of this spot in the Cy3 channel (left) and Cy5
channel (right). First dimension: Isoelectric focusing in IPG pH 310, 18 cm; second dimension: SDS polyacrylamide gel electrophoresis in a 12.5 % T gel. From ref. 8
159
method is very high, which makes it particularly useful for scarce

samples or detection of low abundant proteins [9].
Furthermore, cysteine labeling offers a new opportunity for
the specific detection of redox-modified proteins. The procedure
has lately been optimized [10]: Free protein thiols in sample A can
be directly labeled with Cy3 maleimide. However, since some protein thiols may have become oxidized and will no more react with
the Cy3 maleimide, the selective labeling of the oxidized subproteome (sample B), in particular its reversible modified thiols, which
can become reduced with a reductant such as DTT, might be also
of great interest. Thus, in order to maintain the overall spot pattern the specific labeling of sample A requires the exclusion of any
non-reacted labeling reagent included in sample B and vice versa.
Therefore, subsequently to the labeling of sample A, excessive label
is removed via gel filtration, which excludes the labeling of components of sample B, the none labeled protein thiols, and their
irreversible alkylation with the Cys-interacting compound (CinC).
In contrast, for sample B, free protein thiols are first irreversibly
alkylated with the Cys-interacting compound (CinC), the excessive
alkylating reagent removed via gel filtration, the eluent (eluted/
recovered sample) treated with a reductant like DTT, and subsequently labeled with the Cy3 maleimide.
The individual samples (A or B) are combined with the internal
pooled standard (saturated labeled sample) and submitted to a 2D
electrophoresis. When the false-color images, where Cy 3 (sample A)
is represented in green and Cy3 (sample B) in red, are overlaid,
green spots indicate redox-modified proteins, yellow spots show proteins containing occluded thiols. By using CinC instead of the formerly
employed N-ethylmaleimide the pattern matching is improved,
because the molecular sizes of the blocking reagent is more similar
to the sizes of the dye-maleimides. Irreversibly oxidized proteins
can be visualized by post-staining of the gel.
These differential multiplexing methods have initiated great
acceptance of protein labeling with fluorophores prior to the separation [11]. Their features improve the statistical basis for relative
quantitative measurements of protein expression levels considerably. These advantages have justified the relatively high investments
for dyes and fluorescent imagers.
The vast majority of gels are still post-stained, because of the
high price level of the labeling reagents and the investments for fluorescent imagers. Lately new types of fluorescent dye labels for
proteins have been introduced [12], which are less expensive and
can also be applied in presence of carrier ampholytes and reductants
(see Note 3).
However, it should be noted that gel-based separation
techniques have some limitations: very large and very small, very
hydrophobic proteinslike membrane proteinsand those with
very high isoelectric points are not included, LC/MS and CE/MS
possess superior sensitivity of detection features.
160
3.2 Protein Labeling

with Stable Isotopes
Nonradioactive isotope labeling is a very useful approach for

comparing samples by mass spectrometry analysis, because the
mass differences introduced into the sample components by different isotope compositions can be easily detected, while the chemical
properties of the molecules are not altered. Thus all the commonly
used physico-chemical sample preparation techniques can be
applied for reduction of the complexity of the sample.
Such an approach was first time utilized for proteins by Gysi
et al. [13], introducing isotope-coded affinity tags (ICAT): the
protein cysteine residues are chemically labeled with a reagent
comprised of a thiol-reactive group (iodoacetamide), biotin, and a
linker containing either eight hydrogen or eight deuterium atoms.
Typically the control sample is labeled with the light and the test
sample with the heavy tag and combined in a 1:1 ratio. The mixture is digested with an endopeptidase, thebiotin-containing
labeled peptides are purified by affinity chromatography on a
streptavidin column, and submitted to reversed-phase HPLC
coupled with mass spectrometry. Peptide pairs originating from
differential samples are identified by a mass difference of 8 Da,
their abundances are compared for relative quantification. Protein
identification is performed by MS/MS analysis of the respective
peptides. In order to overcome isotopic shift effects caused by the
deuterium interaction with reversed phase chromatography media,
and improve MS/MS spectra, the tags were later modified by using
13
C labels [14] and a cleavable biotin moiety [15]. Figure 2 shows
a schematic representation of the ICAT workflow. A great advantage of ICAT is the very efficient reduction of sample complexity,
because it isolates the few cysteine containing peptides. This makes
it very useful for automated LC/MS systems. However, because
not all proteins and peptides contain cysteines, the ICAT method
does unfortunately not result in global labeling; this leads to
reduced sequence coverage in mass spectrometry analysis and missing values.
An isotopic label can also be introduced metabolically by salts
and amino acids containing different compositions of deuterium,
13
C or 15N during cell growth. The mostly applied in vivo labeling
method is SILAC (Stable Isotope Labeling with Amino acids in
Cell culture) which has been introduced by Ong et al. [16].
Whereas the growth medium for one cell population contains normal amino acids, the second cell population is fed with a growth
medium containing amino acids with heavier isotopes. A number
of SILAC variations exist, differing by the choice of amino acids
and isotope combinations. After a couple of cell divisions, the
labeled amino acids are completely incorporated into the cellular
proteins. The samples are mixed and can be directly digested with
a protease, or previously separated by electrophoresis or chromatography for reduction of complexity. The peptide mixture is then
analyzed by LC-MS, relative protein amounts are quantified by
161
Fig. 2 Schematic drawing of the principle of ICAT. The different samples are reduced with a thiol reagent and
labeled at the free SH groups of their cysteines with the ICAT reagents (light and heavy) via the iodoacetamide
group (IAA). After tryptic digestion of the sample proteins the labeled (biotin-containing) peptides are purified
with a Streptavidin affinity column and submitted to LC/MS for relative quantification and protein identification
comparison of the abundance of respective peptide doublets.

The proteins are identified by MS/MS analysis of the peptides. The
SILAC principle is shown in Fig. 3.
By adding the labeled amino acids to the growth medium for
only a short period of time, also differences in de novo protein
turn-over can be studied. This method is called Pulsed SILAC
[17]. SILAC generally allows global labeling on the protein level,
which is a great advantage of the method. SILAC can only be used
for labeling growing cells. Meanwhile SILAC has already been
162
Fig. 3 Schematic drawing of the principle of SILAC. The cells are grown under different conditions in media
containing different isotope compositions of hydrogen, carbon, and nitrogen. The samples are then mixed
together. The duplex mixture can be fractionated by any protein separation method. After tryptic digestion of
the protein fractions, the peptides are submitted to LC/MS for relative quantification using the defined mass
shifts of the related peptide peaks, and protein identification
applied to labeling multicellular model organisms, such as mouse,

Caenorhabiditis elegans, Drosophila melanogaster, and Arabidopsis
thaliana.
In order to overcome limitations of the approaches described
above, Schmidt et al. have developed Isotope-Coded Protein
Labeling (ICPL), which is based on the well-established reaction
of N-nicotinoyloxy-succinimide (Nic-NHS) with the -amino
group of lysine residues and the available N-termini of intact proteins [18]. The ICPL technology is available with up to four labels,
which contain different combinations of hydrogen/deuterium
and 12C and 13C atoms, and create mass shifts of 0, 4, 6 and 10 Da.
In order to facilitate access to all free -amino groups of the lysines,
163
the samples are denatured, reduced and alkylated prior to the labeling reaction. The tagged samples are mixed in equal amounts and
separated by electrophoretic or chromatographic techniques.
Because all lysines are blocked by the ICPL labels, protein cleavage
is ideally performed by double digest with trypsin and endoproteinase GluC. This measure ensures improved sequence coverage.
The use of Nic-NHS as a label enhances the ionization of the
labeled peptides, and hence the sensitivity of mass spectrometry,
which allows also the detection of low abundant proteins. Figure 4
shows a schematic drawing of the ICPL principle. The method is
compatible with both MALDI and electrospray ionization mass
spectrometry, for quantification with ICPL MS/MS is not required.
Fig. 4 Schematic drawing of the principle of ICPL. The protein samples are labeled with the ICPL reagents
containing different isotope compositions of hydrogen and carbon at the -amino groups of the lysines. The
samples are then mixed together. The multiplex mixture can be fractionated by any protein separation method.
After tryptic digestion of the protein fractions, the peptides are submitted to LC/MS for relative quantification
using the defined mass shifts of the related peptide peaks, and protein identification
164
The evaluation software selects only the peptide peak sets with the
defined mass shifts of 0, 4, 6, and 10 showing differing expression
levels. The ratios of peak intensities of the peptide pairs, triplets or
quadruplets are used to determine the related protein abundances
originating from the different samples. Analyzing quadruplets
instead of triplets or doublets considerably reduces the number of
false-positive multiplets, because randomly occurring sets with the
four defined mass differences are considerably less frequent than
sets with only two or three defined mass shifts (see also Note 4).
ICPL has the advantage, that it yields high sequence coverage, also
when the complexity of the labeled samples has been reduced with
high-resolution separation techniques. This feature is very important for adequate detection of post-translational modifications and
protein isoforms. It offers multiplex analysis of up to four samples,
and it can be applied on any species, cell lysates, or tissue types.
Notes
1. For optimum results, a sample cleanup based on protein precipitation prior to labeling is recommended. It is essential
for labeling, that the sample solution has a pH value between
8 and 9. In order to avoid any interference with the labeling
procedure, the sample solution must not contain any primary
amines, like carrier ampholytes, and reductant, like dithiothreitol.
These reagents are added after labeling has been completed.
2. Saturation labeling is performed for one hour at a pH value of
8.0 and a temperature of 37 C. Like with minimal labeling,
the sample must not contain any carrier ampholytes and reductant. Also here these compounds are added prior to application
on the first dimension.
3. SERVA HPE Lightning Red is compatible with all additives
typically used for sample solubilization and protein extraction,
including carrier ampholytes and reductants like dithiothreitol
(DTT) and dithioerythreitol (DTE). For efficient labeling the
pH value of the sample must be 8 or higher.
4. The ICPL software cannot identify incomplete quadruplets,
which sometimes occur due to the complete absence of a protein in a sample as a result of a complete down-regulation. It is
therefore advised to prepare an internal standard, which is
composed of equal amounts of all samples to be analyzed. This
mixed sample is divided into four aliquots, which receive the
four labels before they are combined again. The internal
standard is treated in the same way like the samples. This will
create equally abundant quadruplets, which are used in the
software data base for the detection and quantification of
incomplete quadruplets.
165
References
1. Whlbrand L, Trautwein K, Rabus R (2013)
Proteomic tools for environmental microbiologya roadmap from sample preparation to
protein identification and quantification.
2. Miller I, Crawford J, Gianazza E (2006)
Protein stains for proteomic applications:
which, when, why? Proteomics 6:53855408
3. Moche M, Albrecht D, Maa S et al (2013)
The new horizon in 2D electrophoresis new
technology to increase resolution and sensitivity. Electrophoresis 34:15101518
4. Zhou S, Bailey MJ, Dunn MJ et al (2005) A
quantitative investigation into the losses of proteins at different stages of a two-dimensional
gel electrophoresis procedure. Proteomics 5:
27392747
5. Go Y-M, Jones DP (2013) The redox proteome. J Biol Chem 288:2651226520
6. Silva JC, Denny R, Dorschel CA et al (2005)
Quantitative proteomic analysis by accurate
mass retention time pairs. Anal Chem 77:
21872200
7. nl M, Morgan ME, Minden JS (1997)
Difference gel electrophoresis: a single gel
method for detecting changes in protein
extracts. Electrophoresis 18:20712077
8. Westermeier R, Scheibe B (2008) Difference
gel electrophoresis based on Lys/Cys tagging.
In: Posch A (ed) Sample preparation and fractionation for 2-D PAGE/proteomics. Methods
in molecular biology, vol 424. Humana,
New York, pp 7385
9. Berendt FJ, Frhlich T, Bolbrinker P et al
(2009) Highly sensitive saturation labeling
reveals changes in abundance of cell cycleassociated proteins and redox enzyme variants
during oocyte maturation in vitro. Proteomics
9:550564
10. Personal communication from Dr. Rudolf

Lichtenfels, Martin Luther University HalleWittenberg, Institute of Medical Immunology,
Halle (Saale), Germany
11. Cramer R, Westermeier R (eds) (2012)
Difference gel electrophoresis (DIGE): methods and protocols, vol 854. Springer GmbH,
Heidelberg
12. Griebel A, Obermaier C, Westermeier R et al
(2013) Simplification and Improvement of
Protein Detection in two-dimensional Electrophoresis Gels with SERVA HPE Lightning
Red. Arch Physiol Biochem 119:9499
13. Gygi SP, Rist B, Gerber SA et al (1999)
Quantitative analysis of complex protein mixtures using isotope-coded affinity tags. Nat
Biotechnol 17:994999
14. Hansen KC, Schmitt-Ulms G, Chalkley RJ et al
(2003) Mass spectrometric analysis of protein
mixtures at low levels using cleavable
13C-isotope-coded affinity tag and multidimensional chromatography. Mol Cell Proteomics
2:299314
15. Li J, Steen H, Gygi SP (2003) Protein profiling
with cleavable isotope-coded affinity tag
(cICAT) reagents: the yeast salinity stress
response. Mol Cell Proteomics 2:11981204
16. Ong SE, Blagoev B, Kratchmarova I et al
(2002) Stable isotope labeling by amino acids
in cell culture, SILAC, as a simple and accurate
approach to expression proteomics. Mol Cell
Proteomics 1:376386
17. Schwanhusser B, Gossen M, Dittmar G et al
(2009) Global analysis of cellular protein translation by pulsed SILAC. Proteomics 9:205209
18. Schmidt A, Kellermann J, Lottspeich F (2005)
A novel strategy for quantitative proteomics
using isotope-coded protein labels. Proteomics
5:415
Chapter 14
Isolation of Extracellular Vesicles for Proteomic Profiling
Abstract
Extracellular vesicles are nano-sized lipid bilayer vesicles released from most cells, including archaea, bacteria,
and eukaryotic cells. These membrane vesicles play multiple roles in cell-to-cell communication, including
immune modulation, angiogenesis, and transformation of cells by transferring genetic material and functional proteins. They contain specific subsets of proteins, DNA, RNA, and lipids that represent their cellular status. Furthermore, extracellular vesicles are enriched in cell type- or disease-specific vesicular
proteins, especially plasma membrane proteins, which have pathophysiological functions; these vesicular
proteins are considered novel diagnostic biomarkers as well as therapeutic targets. To profile the proteome,
various purification methods of extracellular vesicles have been developed, but density gradient ultracentrifugation is considered the most promising. In this chapter, we describe the isolation of extracellular
vesicles derived from SW480 cells and the preparation of tryptic peptides for mass-spectrometry-based
proteomic analysis.
Key words Exosomes, Ectosomes, Microvesicles, Proteomics, Iodixanol, Ultracentrifugation,
Methanol/chloroform precipitation, Biomarker
Introduction
Nearly all living cells on Earth release vesicles into the extracellular
space. Extracellular vesicles are spherical, lipid-bilayered structures
containing various proteins, nucleic acids, and lipids [13].
Although they have long been considered cellular dust or artifacts
from dead cells, recent research has revealed that extracellular vesicles play various physiological and pathological roles, such as
immune modulation, angiogenesis, and transformation of cells by
transferring genetic material and functional proteins [1, 2]. It is
known that cells release two types of extracellular vesicle: exosomes
and ectosomes [1, 2]. Exosomes, 40100 nm in diameter, are
released via the fusion of multivesicular bodies with the plasma
membrane, followed by the discharge of intraluminal vesicles from
the multivesicular bodies into the extracellular space. Ectosomes,
also known as microvesicles, are 1001,000 nm in diameter and
167
168
shed directly from the plasma membrane. Despite the obvious

difference in their biogenesis mechanisms, discrete separation of
exosomes and ectosomes is quite difficult because of the few physical differences between these two types of extracellular vesicle.
Therefore, we describe the isolation of extracellular vesicles, including both exosomes and ectosomes, in this chapter.
Extracellular vesicles have unique characteristics, such as size,
density, enriched molecules, and electrokinetic potential [46].
Therefore, based on these physical properties, various isolation
methods, including ultracentrifugation, immunoaffinity capture, flow
field-flow fractionation, and gel filtration, have been established for
mass-spectrometry-based proteomic analyses of extracellular vesicles from cell culture media and various biological fluids, including
plasma, urine, and ascites [711]. Among the various purification
methods, ultracentrifugation-based isolation has been applied
widely because of its high yield, simplicity, and accessibility [3].
However, this method can cause misidentification of non-vesicular
proteins, because ultracentrifugation at 100,000 g co-sediments
non-vesicular contaminants, including non-specifically bound
serum-abundant proteins on extracellular vesicles, apoptotic vesicles, protein aggregates, and high-molecular-weight protein
complexes [2]. These non-vesicular contaminants cause an increase
in the dynamic range of the proteome, reducing the number of
vesicular proteins identified by mass-spectrometry-based proteomic analysis [2]. To remove these contaminants, density gradient ultracentrifugation based on sucrose or iodixanol is one of the
best methods [12]. When compared with sucrose, iodixanol is
more chemically stable and less viscous. Thus, we prefer iodixanol
density gradient centrifugation to the more conventional sucrose
density gradient centrifugation [12].
Intraluminal proteins contained within extracellular vesicles
are protected from denaturing agents (e.g., urea, thiourea, and
guanidine hydrochloride) and proteases (e.g., trypsin) because of
the lipid bilayer structure, which makes the vesicles impermeable
to these reagents. Therefore, to identify the complete vesicular
proteome, it is necessary to lyse the extracellular vesicles. Detergent
(e.g., SDS, Triton X-100) can effectively lyse vesicles, but is difficult to remove from the sample. Detergent also causes deleterious
effects, including compromised peptide separation in reversedphase liquid chromatography and high noise in mass spectrometry.
To overcome these problems, we employ methanol/chloroform
precipitation, which effectively removes lipids and precipitates proteins [13]. In this chapter, we describe the purification of extracellular vesicles derived from SW480 cells, a human colorectal cancer
cell line, using iodixanol buoyant density gradient centrifugation,
delipidation using methanol/chloroform precipitation, and protein digestion using trypsin for proteomic profiling.
Proteomics of Extracellular Vesicles
169
Materials
Prepare all solutions using double-distilled water, to attain a sensitivity of 18 M cm at 25 C, and analytical grade reagents unless
otherwise indicated. Prepare and store all reagents at 4 C. All buffers should be passed through a 0.45 m filter. It is necessary to use
sterile equipment if the final use of the extracellular vesicles will
require sterility (e.g., in vitro cell-based assays or in vivo animal
experiments). In addition, purified extracellular vesicles should be
free of lipopolysaccharide contamination, especially for
immunology-related experiments.
2.1 Cell Culture,

Pre-clearing,
and Concentration
of Conditioned
Medium
1. SW480 cell line (ATCC CCL-228).

2. Cell culture medium I: RPMI 1640 medium supplemented
with 10 % fetal bovine serum and 1 Antibiotic-Antimycotic
solution.
3. Cell culture medium II: RPMI 1640 medium supplemented
with 1 Antibiotic-Antimycotic solution.
4. Phosphate-buffered saline (PBS): 155.17 mM NaCl, 1.54 mM
KH2PO4, 2.71 mM Na2HPO4, pH 7.2.
5. High-speed centrifuge.
6. QuixStand benchtop system (#56-4107-77) (GE Healthcare,
Buckinghamshire, UK).
7. 100 kDa hollow fiber cartridge (#56-4101-72) (GE Healthcare,
Buckinghamshire, UK).
2.2 Sucrose Cushion

and Iodixanol Buoyant
Density Gradient
Ultracentrifugation
1. 2.0 M sucrose buffer: 2.0 M sucrose, 150 mM NaCl, 20 mM

HEPES, pH 7.4.
2. 0.8 M sucrose buffer: 0.8 M sucrose, 150 mM NaCl, 20 mM
HEPES, pH 7.4.
3. HEPES-buffered saline (HBS): 150 mM NaCl, 20 mM
HEPES, pH 7.4.
4. Ultracentrifuge: e.g. Beckman Optima LE-80K (Beckman
Instruments, Palo Alto, CA, USA).
5. OptiPrep density gradient medium: 60 % iodixanol in water.
6. 50 % (w/v) iodixanol buffer: Mix 5 volume of OptiPrep density gradient medium with 1 volume of the following diluents:
0.25 M Sucrose, 0.9 M NaCl, 120 mM HEPES, pH 7.4.
7. 20 % (w/v) iodixanol buffer: Mix 2 volume 50 % (w/v) iodixanol buffer with 3 volume of the following diluent: 0.25 M
Sucrose, 150 mM NaCl, 20 mM HEPES, pH 7.4.
8. 5 % (w/v) iodixanol buffer: Mix 1 volume of 50 % (w/v) iodixanol buffer with 9 volume of the following diluent: 0.25 M
Sucrose, 150 mM NaCl, 20 mM HEPES, pH 7.4.
170
9. Spectrophotometer: e.g. SPECTROstarNano (BMG LABTECH

GmbH, Offenburg, Germany).
10. 96-well plates.
2.3 Methanol/
Chloroform
Precipitation
1. 1.5 mL and 2.0 microcentrifuge tubes.

2. Microcentrifuge.
3. Water (HPLC grade).
4. Methanol (p.A.).
5. Chloroform (p.A.).
2.4 In-Solution
Protein Digestion
1. Protein denaturation buffer: 6 M urea, 40 mM NH4HCO3 in

water. Prepare immediately prior to use.
2. Protein digestion buffer: 40 mM NH4HCO3 in water. Prepare
immediately prior to use.
3. Reduction solution: 500 mM tris(2-carboxyethyl)phosphine in
40 mM NH4HCO3. Prepare immediately prior to use.
4. Alkylation solution: 500 mM iodoacetamide in 40 mM
NH4HCO3. Prepare immediately prior to use and protect from
light.
5. Protease solution: 250 g/mL sequencing grade modified
trypsin.
2.5 Desalting Using

C18 Spin Column
1. C18 spin column (#89873) (Pierce, Rockford, IL, USA).

2. Activation solution: 50 % acetonitrile.
3. Equilibration solution: 0.1 % trifluoroacetic acid.
4. Elution solution: 0.1 % formic acid in 70 % acetonitrile.
Methods
3.1 Cell Culture,

Pre-clearing,
and Concentration
of Conditioned
Medium
1. Grow SW480 cells in cell culture medium I until they reach

8090 % confluency for adherent cells (see Note 1).
2. Wash cells twice with PBS, corresponding to 1.25-fold the volume of culture medium (see Note 2).
3. Incubate the cells in cell culture medium II for 24 h (see Note 3).
4. Collect the conditioned medium with a pipet and transfer to
centrifuge tubes. Centrifuge at 500 g for 10 min, and quickly
pour off the conditioned medium into clean centrifuge tubes
(see Note 4).
5. Centrifuge the tubes at 2,000 g for 15 min, and quickly pour
off the conditioned medium into clean centrifuge tubes.
Repeat this step (see Note 5).
171
6. Transfer the conditioned medium to the feed reservoir in a

QuixStand benchtop system with a 100-kDa hollow fiber
cartridge and concentrate to ca. 70 mL.
3.2 Sucrose Cushion
Ultracentrifugation
and Iodixanol Buoyant
Density Gradient
Ultracentrifugation
1. Add the concentrated conditioned medium to each of two

SW32 Ti ultracentrifuge tubes (ca. 35 mL per tube) over 1.0 mL
of 0.8 M and 0.5 mL of 2.0 M sucrose cushion buffers.
2. Conduct ultracentrifugation at 100,000 g for 2 h. The conversion of rpm to g-force is presented in Table 1.
3. Collect the 2 mL interface between the 0.8 and 2.0 M sucrose
cushions from two ultracentrifuge tubes and make the volume
up to 10 mL with HBS.
4. Add the diluted sample to a SW41 Ti ultracentrifuge tube over
0.35 mL of 0.8 M and 0.15 mL of 2.0 M sucrose cushion
buffers.
5. Conduct ultracentrifugation at 100,000 g for 2 h.
6. Collect the 0.5 mL interface between the 0.8 and 2.0 M
sucrose cushions and mix the sample with 1.42 mL HBS and
2.88 mL 50 % iodixanol buffer, resulting in 30 % iodixanol
buffer (w/v).
7. Place the sample at the bottom of a SW41 Ti ultracentrifuge
tube and overlay 3 mL of 20 % (w/v) and 2.5 mL of 5 % (w/v)
iodixanol buffer (see Note 6).
9. Collect each 1 mL fraction (n = 10) from the top of the iodixanol gradient and determine the protein concentration using a
Bradford assay (see Notes 7 and 8).
Table 1
Rotor information for purification of extracellular vesicles
Rotor
Rotor type
Tubes
Number of rotor
cavities nominal
volume of tube
SW32 Ti
Swinging bucket
Ultra-Clear
6 38.5 mL
28,500 rpm
(100,000 g)
SW41 Ti
Swinging bucket
Ultra-Clear
6 13.2 mL
28,500 rpm
(100,000 g)
40,200 rpm
(200,000 g)
Type 90 Ti
Fixed angle
Polycarbonate
8 10.4 mL
49,200 rpm
(150,000 g)
Conversion of
rpm to g-force
172
10. Add the extracellular-vesicle-enriched fraction to Type 90 Ti

ultracentrifuge tubes and make the volume up to 10 mL
with HBS.
12. Resuspend the pelleted extracellular vesicles with 100 L HBS.
3.3 Determination
of Fraction Density
1. Transfer 100 L of each of the density gradient fractions into

wells of a 96-well plate (Table 2).
2. Transfer 100 L of each of the density gradient fractions from
the control tube into wells of a 96-well plate (see Note 6).
3. Measure the absorbance at 340 nm of the solutions in each
well using spectrophotometer.
4. Make a standard curve using the absorbance values of the gradient fractions, and determine the density of each gradient
fraction (see Fig. 1a).
3.4 Methanol/
Chloroform
Precipitation
and In-Solution
Digestion
of Extracellular
Vesicles for Proteomic
Profiling (See Note 9)
3.4.1 Methanol/
Chloroform Precipitation
1. Prepare 100 L (total volume) of extracellular vesicles in a

1.5 mL microcentrifuge tube, corresponding to 20 g of
protein as determined by a Bradford assay (see Note 10). Use
HBS as necessary to make up the final volume of 100 L.
2. Add 400 L methanol, and mix thoroughly by vortexing.
3. Conduct centrifugation at 9,000 g for 5 min.
4. Add 100 L chloroform, and mix thoroughly by vortexing.
6. Add 300 L chloroform, and mix thoroughly by vortexing.
Table 2
Densities of iodixanol solutions
50 % Iodixanol buffer (L)
Diluent (L)a
% (w/v) Iodixanol
(g/mL)
1.031
100
1.059
12
88
10
1.078
20
80
14
1.098
28
72
16
1.107
32
68
20
1.127
40
60
24
1.146
48
52
30
1.175
60
40
34
1.194
68
32
40
1.223
80
20
Diluent: 0.25 M Sucrose, 150 mM NaCl, 20 mM HEPES, pH 7.4
a top
(kDa)
100
75
63
Iodixanol density gradient

2
Iodixanol density gradient

bottom
10
top
(kDa)
48
173
bottom
1
10
35
48
CD63
25
CD81
35
20
Density
(g/mL)
Density
(g/mL)
Fig. 1 Purification of SW480-derived extracellular vesicles. (a) Fractions of iodixanol buoyant density gradients
were analyzed by Western blotting. CD63 and CD81, marker proteins of extracellular vesicles, were detected
in the third fraction from the top. (b) Transmission electron microscopy indicated that purified extracellular
vesicles were small, closed vesicles ranging from 50 to 200 nm in diameter. Scale bar, 200 nm

8. Carefully remove the upper phase without disturbing the white
layer of precipitated proteins between the upper and lower phases.
9. Add 300 L methanol, and mix carefully by tapping.
11. Remove supernatant completely.
12. Briefly centrifuge and discard any residual methanol at the bottom of the 1.5 mL microcentrifuge tube.
13. Leave the 1.5 mL microcentrifuge tube open on the workbench for 5 min to allow the methanol and chloroform to
evaporate completely (see Note 11).
3.4.2 In-Solution
Digestion of Extracellular
Vesicles
1. Resuspend the precipitated proteins with 100 L protein denaturation buffer and vortex thoroughly to completely dissolve
the proteins.
2. Add 1 L of reduction solution and incubate for 1 h at room
temperature.
3. Add 5 L of alkylation solution and incubate for 30 min at
room temperature in the dark.
4. Add 900 L protein digestion buffer and 4 L of protease
solution (the ratio of trypsin to protein should be 1:20).
5. Incubate for 16 h at 37 C in a shaking incubator.
6. Vacuum dry or lyophilize the sample.
3.4.3 Desalting Using

C18 Spin Column
(See Note 12)
1. Centrifuge the C18 spin column at 1,500 g for 1 min to settle

the resin, and then place the C18 spin column into a 2.0 mL
microcentrifuge tube.
2. Add 200 L activation solution to the C18 spin column, centrifuge at 1,500 g for 1 min, and discard flow-through. Repeat
this step.
174
3. Add 200 L equilibration solution to the C18 spin column,

centrifuge at 1,500 g for 1 min, and discard flow-through.
Repeat this step.
4. Resuspend the sample with 150 L equilibration solution and
vortex thoroughly to completely dissolve the sample.
5. Place the C18 spin column into a new 2.0 mL microcentrifuge
tube and load the sample on top of the resin. Incubate for
10 min and centrifuge at 1,500 g for 1 min.
6. Load the flow-through on top of the resin and centrifuge at
1,500 g for 1 min.
7. Place the C18 spin column into a new 2.0 mL microcentrifuge
tube. Add 200 L equilibration solution to the C18 spin column and centrifuge at 1,500 g for 1 min. Discard flowthrough. Repeat this step twice.
8. Place C18 spin column in a new 2.0 mL microcentrifuge tube.
Add 100 L elution buffer to the top of the resin. Incubate for
10 min and centrifuge at 1,500 g for 1 min. Repeat this step.
9. Transfer a total of 200 L of the eluted sample to a 1.5 mL
microcentrifuge tube.
10. Vacuum dry or lyophilize the eluted sample.
11. For reversed-phase liquid chromatography coupled with mass
spectrometry, add 20 L of 0.1 % formic acid in water, vortex,
and centrifuge.
Notes
1. To acquire the appropriate amount of isolated extracellular
vesicles (>50 g based on protein amount) for proteomics and
other assays, it is necessary to use as many cells as possible
(i.e., a minimum of 50, 150 mm culture plates, corresponding
to ca. 1 109 cells). In addition, it is better to purify extracellular vesicles from large volumes of conditioned medium,
because the loss of extracellular vesicles is decreased in large
starting volumes.
2. Serum contains lots of extracellular proteins at high concentrations as well as serum-derived extracellular vesicles, which contaminate cultured cells and the culture plate. Without careful
washing, these components are co-purified during isolation of
extracellular vesicles. To entirely remove these contaminants,
use PBS corresponding to 1.25-fold the volume of culture
medium to wash the cells (e.g., 25 mL PBS and 20 mL culture
medium in a 150 mm culture plate).
3. Cell culture with serum-containing media can cause the copurification of serum-abundant extracellular proteins and
175
serum-derived extracellular vesicles, which can lead to the

identification of serum-derived components at the expense of
vesicular proteins because of the increased dynamic range of
the sample. To overcome this problem, extracellular vesicledepleted serum-containing media or serum-free media are
used for proteomics analysis of extracellular vesicles. Despite
the use of extracellular vesicle-depleted serum-containing
media, however, plenty of serum-abundant extracellular proteins remain non-specifically bound to the purified extracellular vesicles. Therefore, we recommend the use of serum-free
media alongside checking the viability of cells using an appropriate assay (e.g., annexin V staining or the PARP apoptosis
assay). Cell viability of 95 % is generally acceptable for proteomic analysis of extracellular vesicles. Cell culture in serumfree medium for >24 h can cause cell death and thus, generate
apoptotic vesicles, membrane fragments, and other cellular
debris.
4. Low-speed centrifugation will pellet dead cells, large membrane fragments, and cellular debris. It is recommended to use
fixed-angle rotors for these steps. If swinging bucket rotors are
used, the supernatant should be transferred to other centrifuge
tubes using a pipet, leaving about 0.5-cm supernatant above
the pellet.
5. After conditioned medium is collected, it is strongly recommended that purification of extracellular vesicles is conducted
as soon as possible. Alternatively, it is possible to store conditioned medium at 80 C after quick-freezing using liquid
nitrogen, but this likely leads to breakdown of the extracellular
vesicles.
6. To determine the density of iodixanol density gradient fractions, it is necessary that control SW41 Ti ultracentrifuge tubes
(containing the same iodixanol gradient without sample) are
co-centrifuged with the sample. Iodixanol can form self-generating gradients during ultracentrifugation and thus, the densities of iodixanol density gradient fractions are changed after
ultracentrifugation (see Fig. 1a).
7. The extracellular vesicle-enriched fraction is determined by
Western blotting for vesicular marker proteins, such as CD63
and CD81. Generally, extracellular vesicles are localized in the
third fraction (density, ~1.096 g/mL) from the top (see
Fig. 1a). To determine the enrichment of extracellular vesicles, the enriched fraction should be analyzed by electron
microscopy, which is the gold standard method for visualizing
extracellular vesicles (see Fig. 1b). Moreover, dynamic light
scattering analysis and nanoparticle tracking analysis are used
to determine the size distribution and particle concentration
of extracellular vesicles (see Fig. 2a, b). Dynamic light scattering
176
b
Concentration
20
10
0
10
(1 x 106 particles/mL)
Number (%)
a 30
4
3
2
1
0
100
1000
Size (nm)
10000
10
100
1000
10000
Size (nm)
Fig. 2 Size distribution and particle concentration of purified SW480-derived extracellular vesicles. (a) The size
distribution of extracellular vesicles was measured by dynamic light scattering and indicated that the average
diameter is 176.70 7.35 nm. (b) Nanoparticle tracking analysis showed that the average diameter of purified
extracellular vesicles is 172.41 16.42 nm, with a particle concentration of 7.56 0.49 109 particles/g
analysis and nanoparticle tracking analysis have determined

average diameters of SW480-derived extracellular vesicles of
176.70 7.35 nm and 172.41 16.42 nm, respectively.
Nanoparticle tracking analysis also showed that the particle
concentration is 7.56 0.49 109 particles/g of total
protein.
8. Extracellular vesicles can be stored in the density gradient fraction at 80 C as 100 L aliquots after quick-freezing using
liquid nitrogen. Avoid repeated freezing and thawing.
9. Because extracellular vesicles have lipid bilayers, denaturing
reagents, such as urea, thiourea, and guanidine hydrochloride,
cannot access intraluminal proteins and trypsin cannot digest
them. Methanol/chloroform precipitation can effectively
remove lipids as well as precipitate proteins without salts. With
this method, it is possible to precipitate ca. 2 g protein, but it
is recommended that >10 g of vesicular protein be used in
methanol/chloroform precipitation for high yield.
10. About 20 g vesicular protein is enough to identify several
hundred proteins without peptide fractionation (e.g.,
ion-exchange chromatography, isoelectric focusing) using
reversed-phase liquid chromatography coupled with mass
spectrometry. To identify more proteins using peptide fractionation, appropriate amounts of vesicular proteins should be
prepared.
11. Drying of the precipitated proteins for a long period results in
difficulty in complete re-solubilization in protein denaturation
buffer.
12. C18 spin columns can bind up to 30 g of total peptide.
177
Acknowledgement
We thank Gyeongyun Go for helping in isolation and characterization
of extracellular vesicles and Jaewook Lee for analysis of the extracellular vesicles in transmission electron microscope. This work was
supported by the National Research Foundation of Korea (NRF)
grant funded by the Korea government (MEST) (No. 2014023004
and No. 2012R1A1A2042534).
References
1. Choi DS, Kim DK, Kim YK, Gho YS (2013)
Proteomics, transcriptomics and lipidomics of
exosomes and ectosomes. Proteomics 13:
15541571
2. Choi DS, Kim DK, Kim YK, Gho YS (2014)
Proteomics of extracellular vesicles: exosomes
and ectosomes. Mass Spectrom Rev (in press).
doi:10.1002/mas.21420
3. Kim DK, Kang B, Kim OY, Choi DS, Lee J,
Kim SR, Go G, Yoon YJ, Kim JH, Jang SC,
Park KS, Choi EJ, Kim KP, Desiderio DM, Kim
YK, Lotvall JO, Hwang D, Gho YS (2013)
EVpedia: an integrated database of highthroughput data for systemic analyses of extracellular vesicles. J Extracell Vesicles 2:20384
4. Thery C, Amigorena S, Raposo G, Clayton A
(2006) Isolation and characterization of exosomes from cell culture supernatants and biological fluids. Curr Protoc Cell Biol Chapter 3,
Unit 3:22
5. Muralidharan-Chari V, Clancy J, Plou C,
Romao M, Chavrier P, Raposo G, DSouzaSchorey C (2009) ARF6-regulated shedding of
tumor cell-derived plasma membrane microvesicles. Curr Biol 19:18751885
6. Kato K, Kobayashi M, Hanamura N, Akagi T,
Kosaka N, Ochiya T, Ichik IT (2013)
Electrokinetic evaluation of individual exosomes by on-chip microcapillary electrophoresis with laser dark-field microscopy. Jpn J Appl
Phys 52:06GK10
7. Choi DS, Park JO, Jang SC, Yoon YJ, Jung JW,
Choi DY, Kim JW, Kang JS, Park J, Hwang D,
Lee KH, Park SH, Kim YK, Desiderio DM,
Kim KP, Gho YS (2011) Proteomic analysis of
8.
9.
10.
11.
12.
13.
microvesicles derived from human colorectal

cancer ascites. Proteomics 11:27452751
Mathivanan S, Lim JW, Tauro BJ, Ji H,
Moritz RL, Simpson RJ (2010) Proteomics
analysis of A33 immunoaffinity-purified exosomes released from the human colon tumor
cell line LIM1215 reveals a tissue-specific
protein signature. Mol Cell Proteomics 9:
197208
Kang D, Oh S, Ahn SM, Lee BH, Moon MH
(2008) Proteomic analysis of exosomes from
human neural stem cells by flow field-flow fractionation and nanoflow liquid chromatographytandem mass spectrometry. J Proteome Res
7:34753480
Dean WL, Lee MJ, Cummins TD, Schultz DJ,
Powell DW (2009) Proteomic and functional
characterisation of platelet microparticle size
classes. Thromb Haemost 102:711718
Looze C, Yui D, Leung L, Ingham M, Kaler
M, Yao X, Wu WW, Shen RF, Daniels MP,
Levine SJ (2009) Proteomic profiling of human
plasma exosomes identifies PPARgamma as an
exosome-associated protein. Biochem Biophys
Res Commun 378:433438
Hong BS, Cho JH, Kim H, Choi EJ, Rho S,
Kim J, Kim JH, Choi DS, Kim YK, Hwang D,
Gho YS (2009) Colorectal cancer cell-derived
microvesicles are enriched in cell cycle-related
mRNAs that promote proliferation of endothelial cells. BMC Genomics 10:556
Wessel D, Flugge UI (1984) A method for the
quantitative recovery of protein in dilute solution in the presence of detergents and lipids.
Chapter 15
A Protocol forExosome Isolation andCharacterization:
Evaluation ofUltracentrifugation, Density-Gradient
Separation, andImmunoaffinity Capture Methods
DavidW.Greening, RongXu, HongJi, BowJ.Tauro,
andRichardJ.Simpson
Abstract
Exosomes are 40150nm extracellular vesicles that are released from a multitude of cell types, and perform
diverse cellular functions including intercellular communication, antigen presentation, and transfer of
tumorigenic proteins, mRNA and miRNA.Exosomes are important regulators of the cellular niche, and
their altered characteristics in many diseases, such as cancer, suggest their importance for diagnostic and
therapeutic applications, and as drug delivery vehicles. Exosomes have been purified from biological fluids
and invitro cell cultures using a variety of strategies and techniques. In this chapter, we reveal the protocol
and key insights into the isolation, purification and characterization of exosomes, distinct from shed
microvesicles and apoptotic blebs. Using the colorectal cancer cell line LIM1863 as a cell model, a comprehensive evaluation of exosome isolation methods including ultracentrifugation (UC-Exos), OptiPrep
density-based separation (DG-Exos), and immunoaffinity capture using anti-EpCAM-coated magnetic
beads (IAC-Exos) were examined. All exosome isolation methodologies contained 40150nm vesicles
based on electron microscopy, and positive for exosome markers (Alix, TSG101, HSP70) based on immunoblotting. This protocol employed a proteomic profiling approach to characterize the protein composition of exosomes, and label-free spectral counting to evaluate the effectiveness of each method in exosome
isolation. Based on the number of MS/MS spectra identified for exosome markers and proteins associated
with their biogenesis, trafficking, and release, IAC-Exos was shown to be the most effective method to
isolate exosomes. However, the use of density-based separation (DG-Exos) provides significant advantages
for exosome isolation when the use of immunoaffinity capture is limited (due to antibody availability and
suitability of exosome markers).
Keywords Exosome, Protocol, Isolation, Purification, EpCAM, Immunoaffinity capture, Density,
Proteomics, Extracellular vesicle, Shed microvesicles
1 Introduction
Almost all cells release different types of membrane microvesicles
and nanovesicles, which have a variety of important physiological
effects. Recently, evidence has emerged that cells naturally release
extracellular vesicles (EVs), which convey critical information in
DOI10.1007/978-1-4939-2550-6_15, Springer Science+Business Media New York 2015
179
180
David W. Greening et al.
the form of small RNAs, mRNAs, bioactive lipids, and proteins

[13]. The protein content of different types of EVs mostly reflects
that of the parent cells and vesicles are enriched in certain molecules, including components associated with adhesion, membrane
trafficking, cytoskeleton molecules, heat shock proteins, cytoplasmic enzymes, signal transduction proteins, cytokines, chemokines,
proteinases, and cell-specific antigens [49]. Exosomes are a discrete population of small (40150nm diameter) membranous
vesicles that are released into the extracellular space from multivesicular bodies (MVBs) by most cell types [4]. Since their discovery
over 30 years ago [10, 11], it has become clear that exosomes
contribute to many aspects of physiology and disease, and for their
important role in intercellular communication [6, 1215]. Exoso
mes are important regulators of the cellular niche, and their alte
red characteristics in many diseases, such as cancer, suggest their
importance for diagnostic and therapeutic applications, and as
drug delivery vehicles [16]. Despite recent advances in our understanding of exosome biology [13, 17], much of this information
has been obtained from heterogeneous or impure exosome preparations, which have confounded interpretation of findings. For example, it is well known that eukaryotic cells release many membranous
particle types into the microenvironment, these include exosomes,
exosome-like microparticles, shedding microvesicles (sMVs) [6] or
large oncosomes [18], apoptotic blebs (ABs) and gesicles [19].
Hence, there is an urgent need to better define exosome preparations
so that information obtained at both protein and RNA levels can be
appropriately interpreted with respect to unambiguous biological
function. Likewise, it is important to accurately define homogeneous
exosome populations before embarking on large-scale production for
the purpose of detailed biochemical analyses and/or preparation of
clinical-grade reagents.
Several strategies have been used for exosome isolation including
ultracentrifugation, density gradient separation, and immunoaffinity
capture. Our group recently performed a proteomic analysis evaluating the ability of each of these techniques (namely ultracentrifugation
(UC-Exos), OptiPrep density gradient centrifugation (DG-Exos),
and immunoisolation using EpCAM (CD326) antibodies coupled
to magnetic beads (IAC-Exos)) to enrich for exosome markers and
proteins involved in exosome biogenesis, trafficking, and release
from LIM1863 cells (Figs.1 and 2) [5]. Although exosomes prepared using all three isolation strategies contained 40150nm
vesicles positive for exosome markers Alix, TSG101, and HSP70
(Fig.3ac), GeLC-MS/MS combined with label-free spectral counting revealed that immunoaffinity capture (IAC) enriched for exosome and exosome-associated proteins by at least twofold more than
the other two methods studied. To assess the three purification strategies we monitored the enrichment of several protein classes that
have been inextricably associated with exosome biogenesis and/or
Evaluation of Exosome Purification Methods
181
Culture medium / supernatant

Differential centrifugation
480 x g, 5 min
2,000 x g, 10 min
Supernatant
(sMV)
Shed microvesicles (sMV)
preparation
preparation
Concentrated culture
medium
ion
medium preparat
preparation
Membrane filtration
(0.1 m pore size)
Ultracentrifugation
(10,000 x g, 30 min)
Centrifugal membrane concentration

(100K NMWL)
Concentrated culture medium

(CCM, 1.5 mL)
Supernatant
(soluble factors and crude
exosomes)
PBS wash
Ultracentrifugation
(10,000 x g, 30 min)
sMVs
Ultracentrifugation
(100,000 x g, 1 h)
Ultracentrifugation
(100,000 x g, 1 h)
Crude Exosomes
Fig. 1 Crude exosome and shed microvesicle isolation. The experimental workflow used for crude exosome
and shed microvesicle isolation. Cell media/supernatant or biofluids are processed by differential centrifugation to remove intact cells and cell debris. The resulting supernatant can be processed immediately or stored
(4C for up to 1 week, or 20C). Two different methods for isolation of crude exosomes are presented in this
chapter and include: (1) concentrated culture medium (CCM) preparation, or (2) the isolation of shed microvesicles (sMVs) approach. For the CCM preparation (Subheading3.1), the supernatant is filtered using a membrane 0.1m filter unit and concentrated to ~1.5ml using centrifugal ultrafiltration (100K nominal molecular
weight limit (NMWL) filters). The resulting CCM is further ultracentrifuged at 100,000g to obtain the crude
exosome pellet. For isolation of sMVs (Subheadings3.43.5), the culture medium is centrifuged at 10,000g
for 30min, whereby the supernatant is ultracentrifuged at 100,000g to obtain the crude exosome pellet
(see Note 9)
functionendosomal sorting complex required for transport

(ESCRT) and their associated proteins, Rab GTPases, tetraspanins,
proteins implicated in intracellular trafficking, as well as proteins that
may be involved in exosome internalization in a recipient cell
(Table1, Fig.4). To enable this comparative enrichment assessment,
we employed a proteomic label-free peptide spectral count strategy
that entails summating the number of significant peptide MS/MS
spectra for each individual protein, and normalizing them with
respect to the total number of spectra identified in the given sample.
The normalized ratios can then be compared between samples to
estimate enrichment. Our findings indicate that IAC enabled the
preparation of highly enriched exosomes compared to differential
centrifugation and density gradient isolation methods. However, the
182
Fig. 2 Strategies for exosome purification. In order to evaluate different strategies for exosome purification and
characterization, we utilized a human cell line model (LIM1863). (a) A confocal microscopy cross-section
through a LIM1863 organoid highlights concentrated syntaxin 3 staining at the apical ring/lumen (red), and
A33 staining at the basolateral cell periphery (green). Scale 20m. (b) Exosomes were isolated from CCM
derived from human colon cancer LIM1863 cells (1.5mg protein) (Subheading3.1) using the following strategies: ultrafiltration combined with ultracentrifugation at 100,000g (UC-Exos, 375g, Subheadings3.1 and
3.4); OptiPrep density gradient centrifugation (DG-Exos, 75g, Subheading3.6); and EpCAM immunoaffinity capture (IAC-Exos, 195g, Subheading3.7)
use of density-based separation (DG-Exos) provides significant

advantages for exosome isolation when the use of immunoaffinity
capture is limited (due to availability/suitability of exosome markers). We further provide a protocol for the isolation of shed microvesicles (sMVs) [6]. Exosomes are quite distinct from sMVs
(heterogeneous 2001,000nm diameter vesicles) being shed from
the PM into the extracellular space upon cellular activation by various stimuli [20]. Recently, we have isolated sMVs and two exosome
populations (immunoaffinity isolated A33-exosomes and EpCAMexosomes) and profiled miRNA signature from the three distinct EV
subtypes [21].
2 Materials
2.1 Cell Culture
and Concentrated
Culture Medium
(CCM) Preparation
1. Human LIM1863 colorectal cancer cells (see Note 1).

2. Cell culture medium A: 5% (v/v) fetal calf serum (FCS), 0.1%
(v/v) Insulin-Transferrin-Selenium (ITS), 1mg/ml hydrocortisone in RPMI-1640 medium.
183
Fig. 3 Exosome characterization. UC-Exos (a), DG-Exos (b), and IAC-Exos (c) were characterized by Western
blotting and electron microscopy. For Western blotting (Subheading3.8), each exosome preparation (10g) is
separated by 1D-SDS-PAGE, electrotransferred, and probed with exosome markers Alix, HSP70, and TSG101.
Exosomes from each purification strategy were negatively stained using uranyl acetate and viewed by electron
microscopy (Subheading3.9). The scale bar represents 100nm. Exosome isolation using OptiPrep density
gradient separation demonstrates each of the 12 isolated protein fractions (1.011.30g/ml) stained for protein
quantitation with SYPRO Ruby (d) and immunoblot analysis (10g) (e) revealing the exosomal marker Alix
predominantly detected in fraction 1.09g/ml (exosome-containing fraction)
Exosome
biogenesis
ESCRT
Accessory
ESCRT-II
ESCRT-III
ESCRT-I
Category
TSG101
VPS28
VPS37B
FAM125A
Protein name
Tumor susceptibility gene 101

Vacuolar protein sorting 28 homolog
Vacuolar protein sorting 37 homolog B
Family with sequence similarity 125,
member A
89853 FAM125B
member B
51571 FAM49B
member B
84313 VPS25/EAP20 Vacuolar protein sorting 25 homolog
27243 VPS2A/
Chromatin modifying protein 2A
CHMP2A
25978 VPS2B/
Chromatin modifying protein 2B
CHMP2B
128866 VPS32B/
CHMP4B
92421 VPS32C/
Chromatin modifying protein 4C
CHMP4C
51652 VPS24/
Vacuolar protein sorting 24 homolog
CHMP3
10015 ALIX/
Programmed cell death 6 interacting
PDCD6IP
protein
9525
VPS4B/SKD1B Vacuolar protein sorting 4 homolog B
5119
VPS46A/
Chromatin modifying protein 1A
CHMP1A
57132 VPS46B/
CHMP1B
51510 VPS60/
Chromatin modifying protein 5
CHMP5
9798
KIAA0174
KIAA0174
7251
51160
79720
93343
Gene ID Gene symbol
1.7
3.2
7.1
5.6
24.3
1.2
1.2
19.4
1.2
4.1
1.2
1.2
2.5
2.3
3.1
1.3
4.4
1.2
1.3
1.2
4.4
13.3
25.5
15.4
1.2
4.8
11.7
7
7.8
4
4.4
4.1
1.3
4.8
1.3
6.7
8.6
2.5
3.9
3.3
2.5
4.8
7.2
5.3
5.3
2.8
1.5
1.6
2.7
28.8
6
1.2
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Exo
Vesicle
RSC (DG/UC)a RSC (IAC/UC)b RSC (IAC/DG)c Cartad pediae
Table 1
Relative quantification of proteins identified in UC-Exos, DG-Exos, and IAC-Exos using label-free-based spectral counting
MFI2
NT5E
XPNPEP2
4241
4907
7512
Syntenin
Trafficking
GTPase
and release
5878
5868
5869
7879
8766
9230
376267
57403
11021
382
27111
6386
RAB5C
RAB5A
RAB5B
RAB7A
RAB11A
RAB11B
RAB15
RAB22A
RAB35
ARF6
SDCBP2
SDCBP
CD9
CD63
CD81
CD82
CD151
TSPAN1
TSPAN14
TSPAN15
TSPAN3
TSPAN5
TSPAN6
TSPAN8
PLSCR1
PLSCR3
CEACAM5
1048
928
967
975
3732
977
10103
81619
23555
10099
10098
7105
7103
Membrane
5359
Architecture 57048
CD59
966
GPI-anchor
Tetraspanin
Gene ID Gene symbol
Category
RAB5C
RAB5A
RAB5B
RAB7A
RAB11A
RAB11B
RAB15
RAB22A
RAB35
ADP-ribosylation factor 6
Syndecan binding protein (syntenin) 2
Syndecan binding protein (syntenin)
CD59 molecule, complement

regulatory protein
Carcinoembryonic antigen-related cell
adhesion molecule 5
Antigen p97
5-Nucleotidase, ecto (CD73)
X-Prolyl aminopeptidase
(aminopeptidase P) 2
CD9 molecule
CD63 molecule
CD81 molecule
CD82 molecule
CD151 molecule
Tetraspanin 1
Tetraspanin 14
Tetraspanin 15
Tetraspanin 3
Tetraspanin 5
Tetraspanin 6
Tetraspanin 8
Phospholipid scramblase 1
Phospholipid scramblase 3
Protein name
1.5
1.2
1.2
1
5.1
6.8
3.5
1.2
1.6
4.1
1.8
1.5
1.5
1.2
1.4
10.9
1.2
1.2
1.8
2.2
4.8
1.2
4.1
2.2
1.2
1.2
1.2
4.8
1.6
17.6
1.2
2
3.2
3.2
2
2.7
11.4
6.3
4.4
2.3
5.2
2.7
5.3
2.1
25.8
10.7
3
14.9
11.8
10.8
4.4
33.7
4
51.7
6.3
2.1
2.1
2.8
13
2.2
1.8
6.7
1.3
3.9
3.9
2
13.8
1.7
1.8
5.3
1.4
21.3
4.8
3.5
3.1
31.2
7.6
32.1
18
14.2
6.1
2
7
4.8
12.5
2.8
1.7
1.7
3.4
2.7
1.4
10
8.1
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
(continued)
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Exo
Vesicle
VAMP3
9341
HLA-A
HLA-C
3105
3107
649853 HLA-A29.1
CD44 molecule (Indian blood group)

Transferrin receptor (p90, CD71)
Nicastrin
Fibrinogen alpha chain
Fibrinogen beta chain
Fibrinogen gamma chain
Mucin 13, cell surface associated
Prostaglandin F2 receptor negative
regulator
Major histocompatibility complex
class I HLA-A29.1
Major histocompatibility complex,
class I, A
Major histocompatibility complex,
class I, C
Syndecan 1
Syndecan 4
Vesicle-associated membrane protein 2
(synaptobrevin 2)
Vesicle-associated membrane protein 3
(cellubrevin)
Protein name
Relative spectral count ratio (Rsc) for proteins identified in DG-Exos, compared with UC-Exos
Relative spectral count ratio (Rsc) for proteins identified in IAC-Exos, compared with UC-Exos
c
Relative spectral count ratio (Rsc) for proteins identified in IAC-Exos, compared with DG-Exos
d
Presence of proteins in the exosome database ExoCarta (http://www.exocarta.org)
e
Presence of proteins in the extracellular vesicle database Vesiclepedia (http://microvesicles.org)
MHC
component
CD44
TFRC
NCSTN
FGA
FGB
FGG
MUC13
PTGFRN
SDC1
SDC4
VAMP2
6382
6385
6844
Syndecan
SNARE
Gene ID Gene symbol
Category
Recognition Internalization 960

and uptake
motif
7037
Protein
23385
binding
2243
domain
2244
2266
56667
5738
Table 1
(continued)
1.6
3.7
2.2
1.8
2.3
1.2
1.2
1.2
1.2
1.4
3
3.4
6
1.2
2.1
4.2
2.5
3.4
10.6
9.4
2.4
1.3
2.4
9.9
1.8
1.2
4.3
2
1.3
1.6
1.9
4.5
1.5
12.8
11.4
2.9
1.5
1.7
3.3
3.4
4
26
2.5
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Exo
Vesicle
187
Centrifugation
728
248
122
93
265
171
86
98
Immunoaffinity
627
50.0
Normalised spectral count ratio (Nsc)
Density
571
40.0
30.0
20.0
10.0
TSG101
8.0
CD9
CD63
CD81
20.0
CD82
CD151
TSPAN1
15.0
TSPAN3
TSPAN5
10.0
TSPAN6
TSPAN8
TSPAN14
5.0
0.0
VPS28
VPS37B
FAM125A
6.0
TSPAN15
55.0
FAM49B
50.0
VPS25/EAP20
45.0
VPS4B/SKD1B
2.0
VPS46A/CHMP1A
VPS46B/CHMP1B
VPS60/CHMP5
0.0
UC-Exos
DG-Exos
IAC-Exos
5.0
4.0
3.0
2.0
1.0
0.0
UC-Exos
DG-Exos
IAC-Exos
RAB5A
RAB5B
RAB5C
RAB7A
RAB11A
RAB11B
RAB15
RAB22A
RAB35
IAC-Exos
35.0
VPS24/CHMP3
DG-Exos
40.0
VPS32B/CHMP4B
VPS32C/CHMP4C
UC-Exos
60.0
FAM125B
VPS2B/CHMP2B
4.0
Normalised spectral
count ratio (Nsc)
25.0
Alix
VPS2A/CHMP2A
30.0
30.0
25.0
SDCBP
SDCBP2
20.0
SDC1
SDC4
VAMP2
15.0
VAMP3
CD44
TFRC
10.0
NCSTN
FGA
FGB
5.0
FGG
MUC13
PTGFRN
0.0
UC-Exos
DG-Exos
IAC-Exos
Fig. 4 Semi-quantitative normalized spectral count ratios of selected exosome proteins. For proteomic profiling
of the three different methods for exosome purification, the distribution of identified proteins is shown in a
three-way Venn diagram depicting number of proteins commonly observed in all three datasets (265), and
proteins that are unique to each purification strategy (a). The relative abundance of a protein (containing at
least two significant peptides) within a sample was estimated using semi-quantitative normalized spectral
count ratios (Nsc, Eq.1). For each individual protein, significant peptide MS/MS spectra were summated, and
normalized by the total number of significant peptide MS/MS spectra identified in the sample. The ratio serves
an indicator of protein abundance, i.e., the higher the ratio, the more abundant the protein within the sample.
Protein categories of interest included proteins associated with endosomal sorting complex required for transport (ESCRT) components and cargo selection/sorting in exosomes (b), Rab GTPases involved in docking and/
or fusion of MVBs with the plasma membrane (c), tetraspanins, important for EV formation (d), and other
proteins attributed to exosomal trafficking and recipient cell internalization (e)
188
Table 2
Ultracentrifuge, rotor, and consumable information for exosome and shed microvesicle purification
Speed ( g) Rotor
Exosome preparation
Large scale processing 100,000
OptiPrep
100,000
Small scale (washing) 100,000
SW45Ti fixed angle

Polycarbonate Bottle #355655
SW40Ti swinging-bucket Thinwall Polypropylene #331374
TLA-55 fixed angle
Microcentrifuge Polypropylene
Tube #357448
Shed microvesicle preparation

Large scale processing 10,000
JLA 16.25 fixed angle
Small scale (washing)
TLA-55 fixed angle
10,000
Polyallomer tubes or polycarbonate

bottles (Beckman Colter)
Polypropylene Wide Mouth Bottles

#356011
Microcentrifuge Polypropylene
Tube #357448
3. Cell culture medium B: 0.6% (v/v) ITS in RPMI-1640

medium.
4. Tissue culture flasks.
5. 50ml polypropylene centrifuge tubes.
6. Refrigerated centrifuge.
7. Polyallomer tubes or polycarbonate bottles, appropriate for
the ultracentrifuge rotor (see Table2, Note 2).
8. Amicon Ultra-15, Ultracel 3K nominal molecular weight
limit (NMWL) centrifugal filter devices (Merck-Millipore)
(see Note 3).
9. Amicon Ultra-15, Ultracel 100K NMWL centrifugal filter
devices (Merck-Millipore).
10. VacuCap 60 (0.1m) Supor Membrane filters (Pall Life
Sciences).
11. Phosphate-buffered saline (PBS).
2.2 Whole Cell
Lysate (WCL)
Preparation
1. 7080% confluent cells (see Notes 1 and 3).

2. Ice cold PBS.
3. SDS sample buffer: 4% (w/v) sodium dodecyl sulfate (SDS),
20% (v/v) glycerol, 0.01% (v/v) bromophenol blue, 125mM
TrisHCl, pH6.8.
4. 1.5ml ultracentrifuge tubes, polypropylene (Beckman Coulter).
5. TLA-55 rotor (Beckman Coulter).
6. Optima MAX-MP Tabletop Ultracentrifuge (Beckman Coulter).
2.3 Protein
Quantitation
2.3.1 Bradford Assay
189
1. BSA standard (0.011mg/ml).

Prepare a set of standard dilutions with BSA, starting with
500g/ml, performing twofold dilutions in PBS to 4g/ml
(seven dilutions). Dispense 100l aliquots of the diluted BSA
standard curve solutions into microcentrifuge tubes and store
up to 6 months at 20C.
2. Bradford colorimetric assay kit (Bio-Rad).
3. Bradford concentrate solution (Bio-Rad).
4. PBS.
5. CCM, or biofluid
6. 0.5ml microcentrifuge tubes.
7. Flat-bottom 96-well plates.
8. Microplate reader with 590nm filter.
2.3.2 Direct Detect

Spectrometer
1. Direct Detect Spectrometer (Merck-Millipore).

2. Direct Detect Assay-free cards (package of 50).
3. Direct Detect Spectrometer Quick Start Card.
2.3.3 Gel-Based
Densitometry Method
1. SDS sample buffer: 4% (w/v) SDS, 20% (v/v) glycerol, 0.01%

(v/v) bromophenol blue, 250mM TrisHCl, pH6.8.
2. CCM, biofluid, or WCL.
3. NuPAGE 1-mm 10- or 12-well 412% (w/v) Bis-Tris Precast
gels (Life Technologies).
4. NuPAGE 1 MES running buffer.
5. XCell Surelock gel tank.
6. BenchMark Protein Ladder standard of known protein concentration (1.7g/l) (#10747-012, Life Technologies).
7. SYPRO Ruby (Life Technologies).
8. SYPRO Ruby fix solution: 40% (v/v) methanol, 10% (v/v)
acetic acid in water.
9. SYPRO Ruby destaining solution: 10% (v/v) methanol, 6%
(v/v) acetic acid.
10. Orbital shaker.
11. Typhoon 9410 variable mode imager, green (532nm) excitation laser, and 610BP30 emission filter (Molecular Dynamics).
12. ImageQuant software (Molecular Dynamics) or suitable densitometry-based analysis software.
2.4 Shed
Microvesicle
(sMV) Isolation
1. CM or biofluid sample (500l, 1.5mg protein).

2. Ultracentrifuge and matched rotor (Optima TM XPN, Beckman
Coulter).
3. Sterile/filtered PBS.
190
4. JLA 16.25 rotor (large scale preparation)Polypropylene

Wide Mouth Bottles (#356011, Beckman Coulter).
5. TLA-55 fixed angle (small scale/washing)Microcentrifuge
Polypropylene Tube (#357448, Beckman Coulter).
2.5 Ultracentrifu
gation Exosome
(UC-Exo) Isolation
1. CCM (<0.1m) (500l, 1.5mg protein).
2.6 OptiPrep
Density Gradient
Exosome (DG-Exo)
Isolation
1. CCM (<0.1m) (500l, 1.5mg protein).
2. SW40Ti swinging (large scale)Thinwall Polypropylene tubes

(#331374, Beckman Coulter).
3. Ultracentrifuge and matched rotor (Optima TM XPN,
Beckman Coulter).
4. Stock OptiPrep solution: 60% (w/v) aqueous iodixanol.
Prepare discontinuous iodixanol gradient: 40% (w/v), 20%
(w/v), 10% (w/v) and 5% (w/v) solutions of iodixanol by
diluting stock with 0.25M sucrose, 10mM TrisHCl, pH7.5.
2.7 EpCAM
Immunoaffinity
Capture Exosome
(IAC-Exo) Isolation
1. CCM (<0.1m) (500l, 1.5mg protein).

2. EpCAM (CD326) magnetic microbeads, 100l (Miltenyi
Biotec).
3. LS Microcolumn.
4. Solid support magnet (SSM).
5. IAC Rinsing Solution: MACS BSA Stock Solution diluted
1:20 with autoMACS Rinsing Solution (Miltenyi Biotec).
8. IAC Elution buffer: 0.2M glycine, TrisHCl, pH2.8.
9. SDS sample buffer.
2.8 Western Blot

Analysis
1. iBlot Dry Blotting System and transfer membranes (Invitrogen).

2. Tris-buffered saline (TBS): 50mM TrisHCl, pH7.4, 150mM
NaCl.
3. TTBS: TBS with 0.05 (w/v) Tween 20.
4. 5% (w/v) skim milk powder in TTBS.
5. Mouse anti-TSG101 (BD Biosciences): 1:500in TTBS.
191
6. Mouse anti-HSP70 (BD Biosciences): 1:1,000in TTBS.

7. Mouse anti-Alix (Cell Signaling Technology): 1:1,000in TTBS.
8. IRDye 800 goat anti-mouse IgG (LI-COR Biosciences):
1:15,000in TTBS.
9. Orbital shaker.
10. Odyssey Infrared Imaging System, v3.0 (LI-COR Biosciences).
2.9 Electron
Microscopy (EM)
1. Exosome preparations (~2g protein).
2.9.1 Transmission EM
3. Fixing solution: 1% (v/v) glutaraldehyde.
4. Formvar coated 200mesh copper grids (ProSciTech).
5. 1% (w/v) aqueous uranyl acetate (ProSciTech).
6. Tecnai F30 electron microscope.
2.9.2 CryoEM
1. Exosome preparations (~2g protein).

2. Aurion Protein-G gold 10nm (ProSciTech).
4. Glow-discharged C-flat holey carbon grids (ProSciTech).
5. Liquid ethane.
6. Liquid nitrogen.
7. Gatan cryoholder (Gatan Inc.)
8. Tecnai G2 F30 electron microscope.
2.10 GeLC-MS/MS
1. Imperial Protein Stain (Thermo Fisher Scientific).

2. GridCutter (The Gel Company).
3. Protein LoBind Tubes1.5ml microcentrifuge tube (Eppendorf)
or Protein LoBind Plates (Eppendorf).
4. 100mM ammonium bicarbonate: 0.4g NH4HCO3 in 50ml
water.
5. 50mM ammonium bicarbonate/acetonitrile (1:1v/v).
6. 50mM ammonium bicarbonate in water.
7. 10mM dithiothreitol (DTT) in 100mM ammonium
bicarbonate.
8. 50mM iodoacetamide (IAA) in 100mM ammonium
bicarbonate.
9. Trypsin buffer: 10mM ammonium bicarbonate, 10% (v/v)
acetonitrile.
10. Trypsin solution: dissolve 20g trypsin in 1.5ml of trypsin buffer and keep on ice. The concentration of trypsin is 13ng/l.
11. 5% (v/v) formic acid in water.
192
12. Extraction buffer: mix 0.25ml 5% (v/v) formic acid, 0.25ml

water and 0.5ml acetonitrile.
13. Thermomixer temperature range up to 56C.
14. Thermostat oven at 37C.
15. Sonicator.
16. Vacuum centrifuge (lyophilizer).
17. STAGE-Tip/desalting columnremove small disks [2, 3] of
C18 Empore filter using a 22G flat-tipped syringe and ejecting
disks into P200 pipette tips. Ensure that the disk is securely
wedged in the bottom of the tip. Condition the columns (wet
membrane) for each sample by using extraction buffer.
18. MS sample vials with snap lid (Thermo Fisher Scientific).
3 Methods
In this methodology, two different approaches for isolation of
crude exosomes are presented: (a) concentrated culture medium
(CCM) preparation, and (b) the isolation of shed microvesicles
(sMVs) approach (Fig.1).
3.1 Cell Suspension
Culture andCCM
Preparation
1. Human colon carcinoma LIM1863 cell organoids [22] are

cultured in cell culture medium A with 10% CO2 at 37C
2. Wash LIM1863 cell organoids (~2109 cells) four times with
30ml of serum-free RPMI-1640 media (or sterile PBS can be
used) and culture for 24h in 750ml cell culture medium B
(see Notes 3 and 5). Cell viability should be assessed in the
presence of culture medium B (see Note 6).
3. ~750ml of culture medium (CM) is collected into 50ml polypropylene tubes and centrifuged at 4C (480g for 5min
followed by 2,000g for 10min) to remove intact cells and cell
debris (see Note 7). Note to retain the supernatant media and
leave approximately half a centimetre of liquid above the pellet.
Falcon tubes should not be reused at this stage (see Note 8). This
supernatant contains both soluble and membrane vesicle components (see Note 9).
4. CM is filtered using a VacuCap 60 filter unit fitted with a
0.1m Supor Membrane.
5. CM concentrated to ~1.5ml using an Amicon Ultra-15,
Ultracel centrifugal filter device with a 100K NMWL (3,000g
until 95% of media has been filtered). Several centrifugal filter
devices can be used for the same sample to increase efficiently
of concentration, before combining retentate into one centrifugal filter for final concentration (see Note 10).
193
6. CCM storage: Short-term on ice (within 3 days), long-term up

to 6 months 20C.
3.2 Whole Cell
Lysate (WCL)
Preparation
1. Confluent cells are washed twice with ice cold PBS and lysed
with 2ml of SDS sample buffer for 10min on ice. Cells are
counted with a hemacytometer.
2. To remove DNA and insoluble cellular debris, lysates are
centrifuged at 435,000g for 30min (TLA-100.2 rotor),
with supernatants collected and stored at 80C (up to 6
months).
3.3 Protein
Quantitation
(See Note 11)
3.3.1 Bradford Assay
(BCA)
1. Thaw a set of BSA standards, and store up to 2 weeks at 4C.

2. For Bradford assay of CCM: CCM should be placed on ice.
3. In a flat-bottom 96-well plate, load 10l of PBS in the first
well (blank) and 10l of each standard BSA dilution in the
eight following wells.
4. Load 6l of PBS to each well for CCM samples.
3.3.2 Direct Detect

Spectrometer
1. Place the Assay-free card(s) on the spotting tray. Select an

appropriate buffer blank, preferably the same solution in which
the protein has been prepared.
2. Pipette 2l of the buffer solution to the default blank position
on the card (position 1).
3. Pipette 2 l of sample to sample positions 2 through 4 as
needed. Dry and insert card into reader.
4. Sample measurement/calibration.
3.3.3 Protein Staining

Densitometry
1. 5l sample aliquots (CCM/WCL) are solubilized in SDS sample buffer (10l), heated 95C for 5min (closed cap), briefly
centrifuged, and loaded on NuPAGE 412% (w/v) Bis-Tris
Precast gels.
2. 5l BenchMark Protein Ladder (1.7g/l) is loaded into
a separate well on NuPAGE 412% (w/v) Bis-Tris Precast
gels for quantitation. Note that a blank lane should be retained
for background subtraction.
3. Electrophoresis is performed at 150V for 1h in NuPAGE
1 MES running buffer.
4. After power off, the gel is removed from the tank and fixed in
50ml fixing solution for 30min on an orbital shaker and
stained with 30ml SYPRO Ruby for 40min, followed by
destaining in SYPRO destaining solution for 1h.
5. The gel is imaged on a Typhoon 9410 variable mode imager,
using a green (532nm) excitation laser and a 610BP30 emission filter at 100m resolution.
194
6. Densitometry quantitation is performed using ImageQuant

software to determine protein concentration relative to a
BenchMark Protein Ladder standard of known protein concentration (1.7g/l), and normalized based on background
subtraction.
3.4 Shed
Microvesicle (sMV)
Isolation (See Note 12)
1. For isolation of sMVs, the cell culture medium (CM) is centrifuged at 4C (480g for 5min followed by 2,000g for
10min) to remove intact cells and cell debris. Separate tubes
should be used following each centrifugation stage.
2. The CM is further centrifuged at 10,000g for 30min to
remove sMVs (Table2, see Notes 2 and 13).
3. sMVs are resuspended in 1ml of sterile/filtered PBS, before
further centrifugation at 10,000g for 30min to obtain the
washed sMV pellet (see Note 14).
4. sMV pellet can be resuspended in either SDS sample buffer
(for PAGE analysis) or sterile/filtered PBS (EM imaging or
functional studies).
3.5 Ultracentrifuga
tion Exosome (UC-Exo)
Isolation
1. CCM (500l, 1.5mg protein) or CM-depleted with sMVs is

centrifuged at 100,000g (TLA-45 fixed angle) for 1h at
4C (Table2, see Notes 2 and 15).
2. Exosome pellet is resuspended in 1ml sterile/filtered PBS and
re-centrifuged (100,000g, 1h) to obtain UC-Exos.
3. UC-Exos (~375g protein) are resuspended in 50l PBS and
either used immediately or stored at 80C (see Note 16).
3.6 OptiPrep
Density Gradient
Exosome (DG-Exo)
Isolation
1. The CCM (500l, 1.5mg protein) is overlaid on top of the

discontinuous iodixanol gradient. The gradient is generated by
diluting a stock solution of OptiPrep (60% (w/v)) with
0.25M sucrose, 10mM TrisHCl, pH7.5 (40% (w/v), 20%
(w/v), 10% (w/v) and 5% (w/v) solutions of iodixanol).
These solutions are made immediately prior to centrifugation,
and allow up to 30min mixing time for each solution. The
gradient was formed by adding 3ml of 40% iodixanol solution, followed by careful layering of 3ml each of 20% and
10% solutions, and 2ml of the 5% solution (see Note 17).
2. This combined OptiPrep density solution is centrifuged at
100,000g (SW40Ti swinging bucket) for 18h at 4C
(Table2, see Notes 2 and 15).
3. After centrifugation, 12 individual 1ml gradient fractions are
collected manually (with increasing densitytop-bottom collection). A 1ml pipet, and isolation from the meniscus is used
for this careful procedure (see Note 18).
195
4. Fractions are diluted with 2ml PBS and centrifuged at

100,000g for 3h at 4C followed by washing with 1ml
PBS, and re-suspension in 50l PBS.
5. To determine density of each fraction, a control OptiPrep
gradient containing 500l of 0.25M sucrose, 10mM Tris
HCl, pH7.5 should also be run in parallel. Fractions from the
control should be collected as described in Subheading3.6,
step 3, serially diluted 1:10,000 with distilled water, and the
iodixanol concentration determined by absorbance at 244nm
using a molar extinction coefficient of 3,201g1cm1 [23].
3.7 EpCAM
Immunoaffinity
Capture Exosome
(IAC-Exo) Isolation
(See Note 19)
1. CCM (500l, 1.5mg protein) is incubated with EpCAM-

microbeads (100l) for 4h at 4C (see Note 20).
2. 3ml LS Microcolumn placed in a solid support magnet (SSM)
and rinsed three times with Rinsing Solution (MACS BSA
Stock Solution diluted 1:20 with autoMACS Rinsing
Solution).
3. Exosome-bound microbeads are pipetted into the column and
washed three times with 1ml Rinsing Solution.
4. The LS column is removed from the SSM and exosome-bound
microbeads recovered by rinsing the column at RT with
31ml Rinsing Solution.
5. Exosome-bound microbeads are washed twice with 1ml PBS,
placed in ultracentrifuge microfuge vials, and centrifuged at
100,000g for 1h at 4C (Table2).
6. The supernatant is removed and IAC-Exos (yield ~195g)
eluted from the microbeads with either 100l of 0.2M glycine, TrisHCl, pH2.8 for EM imaging, or lysed with 100l
of SDS sample buffer for PAGE analysis (see Note 16).
3.8 Western Blot

Analysis (See Note 21)
1. Samples (~10g protein) are lysed in SDS sample buffer with

50mM DTT and heated for 5min at 95C (closed cap,
although cap can be opened for evaporation to reduce load
volume).
2. Electrophoresis is performed on lysed samples at constant
150V for 1h.
3. Following electrophoresis, proteins are electro-transferred
onto nitrocellulose membranes using iBlot Dry Blotting
System.
4. Membranes are blocked with 5% (w/v) skim milk powder in
Tris-buffered saline with 0.05% (v/v) Tween-20 (TTBS) for
1h at RT.Care should be taken not to touch and disrupt the
membrane (only the immediate corners/edges).
5. Membranes are probed with primary antibodies (anti-TSG101,
anti-HSP70, anti-Alix) for 1h in TTBS followed by incubation
196
with secondary antibody, IRDye 800 goat anti-mouse IgG for

1h in darkness.
6. All antibody incubations are carried out using gentle orbital
shaking at RT.
7. Western blots are washed three times in TTBS for 10min after
each incubation step and subsequently visualized using the
Odyssey Infrared Imaging System (800nm).
3.9 Electron
Microscopy (EM)
(See Notes 22 and23)
1. Exosome preparations (~2g protein) are fixed in 1% (v/v)

glutaraldehyde, layered onto formvar-coated 200mesh copper
grids, and allowed to dry at RT.
3.9.1 Transmission EM
2. Grids are washed twice with water for 5min, and stained with
1% (w/v) uranyl acetate in distilled water for 10min.
3. Imaging is performed at an acceleration voltage of 200kV
using a Gatan UltraScan 1000 (2k2k) CCD (charge-coupled
device) camera coupled to a Tecnai F30 (FEI, Netherlands)
electron microscope.
4. Typically 1020 fields of view are obtained.
3.9.2 CryoEM
1. Exosome preparations (~2g protein) mixed with Aurion

Protein-G gold 10nm at 1:3 ratio. This will allow vesicle diameters to be determined and the possibility for tomographic data
collection.
2. Samples are transferred onto glow-discharged C-flat holey carbon grids and excess liquid blotted. Grids are immediately
plunge-frozen in liquid ethane.
3. Grids mounted in a Gatan cryoholder in liquid nitrogen.
4. Images are acquired at 300kV using a Tecnai G2 F30in low
dose mode.
5. Typically 1020 fields of view are obtained.
3.10 Extraction,
Reduction, Alkylation
1. Exosome samples (20g) lysed in SDS sample buffer, and proteins separated by SDS-PAGE and visualized by Imperial
Protein Stain.
2. Individual gel lanes are placed on a clean glass plate and cut
into equal slices (202mm) using a GridCutter or clean scalpel and individual gel slices subjected to in-gel reduction,
alkylation, and trypsinization (see Note 24).
3. For each band prepare a 1.5ml microcentrifuge tube with
500l of 50mM ammonium bicarbonate/acetonitrile (1/1).
4. Cut excised bands (spots) into cubes (ca. 11mm if you used
1mm thick gel) and transfer gel pieces carefully into low
protein-binding microcentrifuge tubes.
197
5. For sample reduction, microcentrifuge tubes are centrifuged

briefly, heated using a thermomixer at 56C (700rpm for
15min), and the solution discarded. 200l acetonitrile is
added and gel pieces should shrink and take an opaque-white
color. Remove acetonitrile and let air-dry for 5min in thermomixer at 56C.Add 50l fresh DTT solution and incubate at
56C for 30min.
6. For sample alkylation set thermomixer to 22C and remove
DTT solution completely. Immediately add 70l IAA solution, incubate for 20min in thermomixer at 22C (700rpm)
covered by aluminum foil.
7. The IAA solution is then removed, 300l acetonitrile added
for 2min, then removed. Add 100l of 50mM ammonium
bicarbonate/acetonitrile (1/1) and incubate for 30min in
thermomixer with light mixing at RT.The sample should be
lightly centrifuged and supernatant removed.
8. 300 l acetonitrile is added to the samples, removed completely, and air dried for 5min.
3.11 Tryptic
Digestion and STAGETip Desalting
1. For tryptic digestion, add enough trypsin to cover the dry gel
pieces (typically, 5060l depending on gel volume). Store all
samples immediately on ice. After 30min, check if all solution
is absorbed and add more trypsin, if necessary. Gel pieces
should be completely covered with trypsin.
2. Leave gel pieces for another 30min to saturate with trypsin
and add 20l of 50mM ammonium bicarbonate to cover the
gel pieces.
3. Place tubes with gel pieces into the thermostat oven and incubate samples overnight at 37C.
4. Withdraw supernatants to low protein binding tubes, use a
pipette with fine gel loader tip (re-use these tips for all following peptide collection steps).
5. Add extraction buffer (100l) to each tube and sonicate for
15min. Collect supernatants into the corresponding tube.
Repeat.
6. Dry down samples (covered with Vacufilm with four pin holes)
in vacuum centrifuge for 15min.
7. For STAGE-Tip desalting, prepare as many desalting columns
as necessary by punching out small disks [2, 3] of C18 Empore
filter using a 22G flat-tipped syringe and ejecting the disks
into P200 pipette tips. Ensure that the disks are securely
wedged in the bottom of the tip. Careful preparation of these
STAGE-Tip devices will ensure effective filtration.
8. Condition columns by forcing methanol through (50l) and
check whether the STAGE-Tips are leaky.
198
9. Remove any remaining organic solvent in the column by

forcing buffer A through the disk (2) (40l).
10. Adjust pH of peptide sample to pH<2.5 using 2% (v/v) TFA.
11. Force the acidified peptide sample through the C18-StageTip
column.
12. Wash the column with buffer A.Elute the peptides from the
C18 material using 2030l buffer B.Elute directly into a
microfuge tube or auto-sampler plate. Repeat elution.
13. Carefully dry samples in the speed-vac without heating, until
all acetonitrile has evaporated (~23l final volume). Note to
not completely overdry/dehydrate peptide sample due to
issues with sample loss and resolubilization.
14. Mix the sample (1:1) with sample buffer up to 8l.
15. Withdraw to MS-specific vial for analysis (typically 3l loaded
representing ~3.54g sample). To determine peptide concentration a spectrophotometer analysis can be obtained based
on 215nm absorbance, and comparison with known
standard.
16. Short-term storage at 4C (within 2 weeks), or long-term at
80C (up to 18 months).
3.12 MS/MS
Analysis
1. RP-HPLC is performed on a nanoAcquity (C18) 1500.15-

mm i.d. reversed-phase UPLC column (Waters), using an
Agilent 1200 HPLC, coupled online to an LTQ-Orbitrap
mass spectrometer equipped with a nanoelectrospray ion
source [24].
2. RP-HPLC column is developed with a linear 60min gradient
with a flow rate of 0.8l/min at 45C from 0100% solvent
B where solvent A was 0.1% (v/v) aqueous formic acid and
solvent B was 0.1% (v/v) aqueous formic acid/60% (v/v)
acetonitrile.
3. Survey MS scans are acquired with the resolution set to a value
of 30,000. Real-time recalibration is performed using a background ion from ambient air in the C-trap [25].
4. Up to five of selected target ions are dynamically excluded for
3min.
3.13 Database
Searching andProtein
Identification
1. Parameters used to generate peak lists, using Extract-MSn as

part of Bioworks 3.3.1, include: minimum mass 700; maximum
mass 5,000; grouping tolerance 0Da; intermediate scans 200;
minimum group count 1; 10 peaks minimum and TIC of 100.
2. Peak lists for each LC-MS/MS run are merged into a single
MGF file for Mascot searches. Automatic charge state recognition is used due to the high resolution survey scan (30,000).
199
3. LC-MS/MS spectra are searched against human protein

sequence database (e.g., human RefSeq [26] protein database,
38,791 sequences) using Mascot (Matrix Science, UK).
4. Searching parameters include: fixed modification (carboxymethylation of cysteine; +58Da), variable modifications (oxidation of methionine; +16Da), three missed tryptic cleavages,
20ppm peptide mass tolerance and 0.8Da fragment ion mass
tolerance.
5. An MS/MS spectrum is deemed significant if its Mascot ion
score is greater than its identity score. The Mascot ion score
for an MS/MS match is based on the calculated probability (P)
that the observed match between the experimentally observed
peptide and the database sequence is a random event. The
reported score is 10Log(P). Mascot derives an identity score
for each peptide and protein; a score above that value indicates
that the identification assigned has a user-defined probability
(default setting of 97%) of being correct [27].
6. Significant protein identifications contained at least two unique
peptide identifications. The false-discovery rate (derived from
corresponding decoy database search [28]) should be defined
as being <1%for typical studies we define as <0.3% for exosome preparations.
7. Gene Ontology (GO) annotation is retrieved from the Human
Protein Reference Database (HPRD) [29].
8. Protein sequence analysis of exosome proteins is performed to
identify significantly enriched domains/motifs using SMART
[30].
9. The transmembrane (TM) domain prediction tool TMHMM
[31] is used to predict the presence of TM domains in the protein sequences that were identified by MS.
10. Proteomic data from previous exosome studies can be downloaded from ExoCarta (http://www.exocarta.org), an exosome database containing all exosome proteins identified from
multiple organisms. In addition, the extracellular vesicle database resources Vesiclepedia (http://microvesicles.org) and
EVPedia (http://evpedia.info) can be utilized (see Note 25).
3.14 Label-Free
Spectral Counting
(See Note 26)
1. The relative abundance of a protein within a sample is estimated using semi-quantitative normalized spectral count ratios
(Nsc). For each individual protein, significant peptide MS/MS
spectra (i.e., ion score greater than identity score) are summated, and normalized by the total number of significant MS/
MS spectra identified in the sample (Eq.1)
N sc = (n + f ) / (t - n + f ) ,
(1)
200
where n is the number of significant peptide spectral counts for

each protein in the sample, t is the total number of significant
MS/MS spectral counts identified in the sample, and f is the
correction factor set to 1.25 [5]. A correction factor is required
to allow an Nsc value to be calculated when n=0.
2. To compare relative protein abundance between samples the
ratio of normalized spectral counts (Rsc) is estimated (Eq.2),
as previously described [32] based on studies by [33, 34].

Rsc = (nB + f
) (t A - n A + f ) / (n A + f ) (t B - nB + f ) ,
(2)
where n is the significant protein spectral count, t is the total

number of significant MS/MS spectra in the sample, f is a correction factor set to 1.25, and A and B are the samples being
compared.
4 Notes
1. Cells should be grown to reach 7080% confluency for adherent cells, or 6070% of their maximum concentration for cells
grown in suspension. The advantage of utilizing as many cells
as possible ensures a concentrated CM and enrichment of EVs.
When washing adherent cells (typically 10ml), the CM should
be removed and replaced with exosome-production medium
(Cell culture medium B, typically 20ml).
2. Resource for rotor speed and conversion:
https://www.beckmancoulter.com/wsrportal/wsr/researchand-discovery/products-and- services/centrifugation/rotors/
index.htm?t=3
Resource for tubes and adapters:
https://www.beckmancoulter.com/wsrportal/wsr/researchand-discovery/products-and-services/centrifugation/tubesand-adapters/index.htm
3. To isolate CM from adherent cells, the cell culture medium
should be collected under sterile conditions and transferred to
50ml polypropylene centrifuge tubes. CM should be centrifuged at 480g for 5min (the supernatant transferred using a
pipette to a separate 50ml polypropylene centrifuge tubes)
and 2,000g for 10min. The supernatant is collected using a
pipette and stored at 4C.To isolate CM from cell suspensions, the cell culture medium should be collected under sterile
conditions and transferred to 50ml polypropylene centrifuge
tubes. CM should be initially centrifuged at 5min at 200
400g, 4C, before following the protocol for adherent cells
(480g, 5min and 2,000g, 10min).
201
4. For large-scale exosome and microvesicle preparation, we have

employed the use of the two-compartment bioreactor CELLine
tank system to achieving continuous culture and high EV
yields (in mg quantities). This is attained by separating the bioreactor into a medium compartment and a cell compartment.
CELLine does not require any specific adaptation of cell culture techniques or media composition and is suitable for applications based on serum-supplemented or serum free cultures.
We have utilized both CELLine classic (CELLine CL 1000,
#90005, INTEGRA Biosciences) and CELLine adhere
(CELLine AD 1000, #90025) for suspension or anchorage-
dependent cells to be grown in the bioreactor and extensive
EV yields generated. By combining high product and cell concentration with recurring culture medium collection, large
amounts (mg quantities) of highly concentrated secreted proteins and EVs are routinely obtained in CELLine systems [21].
We would recommend washing the cell compartment (every
second day, 3-washes in media) to remove poorly adherent,
nonviable cells/dead cells/debris before addition of fresh FCS
supplemented medium to both compartments. It has been
reported that CELLine culture system for exosome production from tumor cell cultures significantly increased exosome
yield ~12 fold than conventional flasks to ~10.1g/ml [35].
In addition, morphology, phenotype and function of these
exosomes was confirmed, and shown to be identical to traditional flask culture methods.
5. Standard growth medium for most cells in culture require fetal
bovine serum (FBS) as a growth supplement to DMEM/
RPMI.FBS is derived from bovine (cow) serum and contains
a high abundance of cow-derived exosome vesicles. These exosomes can interfere or cause significant protein background
issues when studying exosomes secreted from cells of interest
in standard culture conditions. For exosome production, we
typically modify the growth medium either serum-free media
(SFM) or a combination of SFM and Insulin Transferrin-
Selenium (ITS, 0.20.6% (w/v)) (i.e., culture medium B) to
ensure cells remain viable during exosome production. Any
changes to cell growth medium supplements should be evaluated over the time-course of exosome-production (typically
2448h). Further methods to modify the cell growth medium
(although still the possibility of bovine secreted proteins)
include, 1% (w/v) bovine serum albumin (BSA) added instead
of FBS, and depletion of FCS-containing exosomes based on
ultracentrifugation (100,000g, 18h), or filtration/immunodepletion (CD63 positive, exosome-sized vesicles removed, no
measureable bovine microRNAs). This exosome-depleted FBS
growth supplement (Exo-FBS) is commercially available
(EXO-FBS-250A-1, System Biosciences).
202
6. To monitor cell viability in the presence of exosome-production

medium (culture medium B) a Trypan Blue dye assay should
be performed. Dilute cell sample in Trypan Blue dye by preparing a 1:1 dilution of the cell suspension using a 0.4% (w/v)
Trypan Blue solution. Trypan Blue should be sterile filtered
(0.22 m) in order to remove particles in the solution that
would disturb counting. Carefully and continuously fill the
hemocytometer chamber and incubate for 1min at RT.Count
cells under the microscope in four 11mm squares of one
chamber and determine the average number of cells per square
(all hemocytometers consist of two chambers, each is divided
into nine 1mm2 squares). For an accurate determination, the
total number of cells overlying one 1mm2 should be between
2050 cells/square. If the cell density is >200 cells/square, the
cell suspension should be diluted accordingly. Nonviable cells
will stain blue, viable cells will be unstained, and this difference
should be used to calculate cell viability.
7. For sample and exosome sterility, it is only necessary to use
sterile conditions if the final use of exosomes is going to require
sterility (e.g., functional invivo or invitro assays). If only biochemical analyses will be performed (e.g., immunoblot, proteomics, RNA), very clean, but not necessarily sterile, tubes are
required. The use of filtered and sterile PBS is often used to
ensure no contaminants are included during preparation and
solubilization. If sterility is required, sterile centrifuge and
ultracentrifuge tubes must be used, with centrifuge tube holders and all related steps must be performed in an enclosed tissue culture hood. To sterilize ultracentrifuge tubes, wash the
clean tubes and their lids briefly in 70% (v/v) ethanol, rinse
twice in sterile PBS, and remove PBS.All rotor lids and caps,
centrifugal ultrafilter units should also be washed with 70%
ethanol (depending on compatibility with filter membranes,
check with manufacturer requirements).
8. All centrifugations for sample preparation and exosome isolation and purification should be performed at 4C.Tubes
should be weighed with sample to ensure weight distribution
is equal during centrifugation.
9. Given that there are a diverse membrane vesicles of endosomal
and plasma membrane origin (exosomes, microvesicles/sMVs,
apoptotic bodies, etc.) throughout the extracellular environment [3], it is important to define which vesicle type researchers intend to isolate. In the context of exosomes, we have
provided two different methods for the isolation of crude exosomes, including filtration-based/CCM, and isolation of sMVs
using ultracentrifugation (Fig.1). The selection of which
method for exosome isolation is dependent on several factors
including:
203
(a) Availability of large-scale instruments and centrifuges (due to

the concentration of exosomes in culture medium), large culture volumes (>5001,000ml) are often utilized in order to
generate exosome yields for characterization/functional studies. This introduces issues with the capacity of centrifugation,
availability (rotors, instrumentation), and time. If researchers
are limited by large-scale ultracentrifugation, then the CCM
preparation approach would be suggested.
(b) Exosome yield (typically exosome yields are much greater
(>2530%, data not shown) using the sMV approach due to
the absence of membrane filters (0.1m) which can perturb
exosome isolation), however this approach requires the use of
ultracentrifugation (often large-scale use) see point (a) above.
(c) Requirement to isolate sMVs for analysis (using the CCM
preparation vesicles >0.1m are not retained and therefore
restricting their subsequent analysis). If researchers are interested in the isolation of sMVs, then this fraction can be easily
isolated from culture medium in the process of obtaining the
crude exosome pellet.
10. A wide range of centrifugal filters are commercially available
for concentrating and filtrating protein solutions, removing
small solutes, and/or buffer exchanging. Vertical or angular
membrane configurations reduce concentration polarization
(membrane fouling) and allow high flow rates for optimal solvent passage even with high proteinaceous solutions [24, 36].
Note that low-protein binding filter membranes should be utilized (i.e., Amicon Ultra-15 which utilize a Ultracel regenerated cellulose membrane and compatible with 70% (v/v)
ethanol-based sterilization).
11. For protein quantitation we typically employ different methods based on sample quantity and buffer compatibility, including the Bradford assay (10l sample), protein densitometry
(5 l sample), or Direct Detect Spectrometer (2l sample).
For samples employing detergents, or other reagents which
will affect absorbance we would recommend using the protein
densitometry method. The protein densitometry method is
reproducible, has a linear quantitation range over three orders
of magnitude [37], and is compatible with GeLC-MS/MS
[6, 9, 38]. Please note that SYPRO Ruby fluorescent protein
stain and a variable mode imager, using a green (532nm) excitation laser and a 610BP30 emission filter are required for the
protein densitometry method. The Direct Detect Spectrometer
(DDHW00010-WW, Merck Millipore) is based on infrared
quantification, where samples are applied directly to a cardbased hydrophilic polytetrafluoroethylene membrane, measuring amide bonds in protein chains. Protein concentrations
204
from 0.2 to 5mg/ml can accurately be determined. Buffers

containing more than 5% (w/v) glycerol, SDS, and/or Tween
surfactant should be avoided due to the excessive time required
to achieve dryness. It is important to perform protein quantification on exosome/sMV/WCL samples prior to immunoblotting, or proteomic profiling.
12. For the isolation of sMVs, a direct ultracentrifugation approach
is employed. sMVs are distinct heterogeneous 2001,000nm
diameter vesicles being shed from the plasma membrane into
the extracellular space upon cellular activation by various stimuli [20]. Following 10,000g centrifugation, the sMV pellet is
resuspended in 1ml of PBS and 10,000g centrifugation performed to obtain the washed sMV fraction (resuspended in
100l PBS). The isolation and proteomic characterization of
sMVs has been described previously [6].
13. For ultracentrifugation, note to mark each ultracentrifuge tube
and orient the tube in the rotor with the mark facing up. The
mark should be used a reference for location of a pellet following centrifugation. For swinging-bucket rotors, the pellet is at
the bottom of the tube. For fixed-angle rotors, the pellet is on
the side of the tube near the base.
14. For supernatant removal following ultracentrifugation, for
fixed-angle rotors, pour the supernatant rather than use a
pipet. For swinging-bucket rotors, remove supernatant with
pipet and leave 2mm of supernatant above the pellet.
15. The ultracentrifuge tubes are thin walled and will collapse if
improperly filled. All tubes should be at least 80% capacity to
ensure the tube is not flexible. Tube specifications should be
checked with the manufacturer prior to use. Note to add additional PBS if required. Ensure that the rotor is properly seated
within the centrifuge.
16. The stability of exosomes has been examined during storage at
20, 4, and 37C [39]. The size of the exosomes was shown
to decrease at 4C (34 days) and 37C (from 2 days), indicating a possible structural change or degradation. Multiple
freezing to 20C and thawing did not affect exosome size
based on nanoparticle tracking analysis. In our studies, exosome degradation has been monitored at 4C (within 72h)
and 37C (within 24h) (data not shown). Recommend to
store exosomes at 4C on ice for short-term use (within 3
days), and for long periods at either 20C/80C or lyophilization (over 12 months at 4C). Exosome/sMV samples
should be stored in small (50100l) aliquots to avoid repeated
freezing and thawing.
17. Exosomes have been shown to float at densities ranging from
1.091.15g/ml on continuous sucrose or iodixanol density
205
gradient following centrifugation [5, 8, 9]. OptiPrep velocity

gradients have low viscosity, isoosmotic gradients that provide
rapid and efficient separation of extracellular vesicles.
OptiPrep velocity gradients have been shown to efficiently
separate exosomes from HIV-1 particles [40].
18. Following OptiPrep density gradient, isolated protein fractions (in this case 12 fractions) can be assessed for density
(e.g., range 1.011.30g/ml), and protein yield by staining
with SYPRO Ruby, and assessing expression of exosomal
markers using immunoblot analysis. As an example, OptiPrep
density gradient fractions (10g) revealed exosomal marker
Alix predominantly detected in fraction 1.11g/ml (fraction 7)
(Fig.3d/e).
19. The availability and suitability of exosome markers for immunoaffinity capture is dependent on their target specificity and
application. Currently, several markers for exosomes have been
used for immunoaffinity capture including Glycoprotein A33+
exosomes derived from human epithelial colon cancer cells
[41], EpCAM+ exosomes derived from human epithelial colon
cancer cells [5], sequential Glycoprotein A33+ and EpCAM+
exosomes from human epithelial colon cancer cells to reveal
specific subpopulations of exosomes [6], MHC II+ exosomes
derived from dendritic cells [42], HER2+ exosomes derived
from BT-474 breast cancer cells [43], and CD45+ exosomes
derived from Jurkat and SupT1/CCR5 cells [44].
Commercially, Dynabeads-based CD63-specific reagent
(#10606D, Life Technologies) and streptavidin reagent with
choice of a biotinylated antibody to purify a specific vesicle
population based on a surface antigen (#10608D, Life
Technologies) are available. The use of density-based separation (DG-Exos, Fig.2b) provides significant advantages for
exosome isolation when the use of immunoaffinity capture is
limited (due to availability/suitability of exosome markers).

20. CD326 (EpCAM)
+
exosomes are bound with EpCAM
MicroBeads, with the CM loaded onto a MACS Column
which is placed in the magnetic field of a MACS Separator
(SSM). The magnetically labeled EpCAM+ exosomes are
retained within the column. The EpCAM-vesicles are parsed
through; this fraction is depleted of EpCAM+ exosomes.
After removing the column from the magnetic field, the
magnetically retained EpCAM+ exosomes are eluted as the
positively selected exosome fraction. For this application,
100 l of CD326 (EpCAM) MicroBeads were used for
500ml CM (1.5mg CCM) generated from 2109 cells.
Working on ice may require increased incubation times.
Higher temperatures and/or longer incubation times may
lead to nonspecific vesicle binding.
206
21. To characterize extracellular vesicles as exosomes, it is important

to demonstrate the expression of common exosomal proteins
using immunoblotting. The commonly used markers are: Alix
(PDCD6IP, programmed cell death 6 interacting protein),
TSG101 (tumor susceptibility gene 101), CD63 (tetraspanin
CD63), CD81 (tetraspanin CD81), and Hsp70 (heat shock
protein 70).
22. Exosomes that have been purified should be used to obtain
high-quality electron micrographs without storage at either
20 or 80C.Note that EM should be performed within 1
week of exosome purification.
23. We provide two methods (TEM and cryoEM) for electron
microscopy exosome sample preparation. For TEM analysis,
exosomes are fixed using glutaraldehyde and negatively stained
using uranyl acetate. This approach typically has a rapid sample
preparation approach (<20min) although provides exosome
images of low resolution. Significant shortcomings being that
the sample preparation steps and imaging techniques require
dehydration, chemical fixation and/or staining of the biological specimens. This has the potential to under-represent the
diameter of vesicles due to dehydration and chemical fixation.
In contrast, cryoEM does not use staining or chemical fixation
procedures and samples are directly applied onto an EM grid,
vitrified and visualized. Vitrification is a cryo-fixation method
to preserve biological specimens to near-atomic resolution
[45], while water is transformed into a glass-like state without
formation of ice crystals. CryoEM of vitrified whole cells or
EVs enables observation of biological structures in a nearnative state [46, 47]. Further, cryoEM allows tomographic
data collection and the ability for spatial visualization of more
complex structures. Therefore, cryoEM has the capacity to
reveal highly textured EVs and exosomes.
24. In this chapter, although individual gel lanes were cut into
equal slices (202mm), more conventional high resolution
mass spectrometers can utilize more complex, unfractionated
(or limited fractionation) samples. For this reason we typically
employ a short range gel (<5mm, or 6min at 150V) and use
a clean scalpel to cut the single individual gel slice for reduction, alkylation, and tryptic digestion.

25. Databases for extracellular vesicle (Vesiclepedia: http://
microvesicles.org) (EVPedia: http://evpedia.info) and exosomes (ExoCarta: http://www.exocarta.org) are manually
curated databases of proteins, RNA, and lipids. These databases
are an excellent resource for investigators, however several key
limitations have been discussed about their utility. The primary
issue is terminologies used in naming EVs, and then categorizing specific isolated EVs into a standardized nomenclature.
207
Inaddition, various studies employ differences in their purification and characterization, which inevitably introduces and
potentially confounds interpretation of biochemical data.

26. In this chapter, label-free proteomic quantification was
employed based on spectral counting (SpC). SpC counts the
number of spectra identified for a given peptide in different
biological samples and then integrates the results for all measured peptides of the protein(s) that are quantified [48]. SpC
has become a commonly used approach for measuring protein
abundance in label-free shotgun proteomics.
Acknowledgements
This work was supported, in part, by the National Health &
Medical Research Council of Australia (program grant #487922
(RJS), project grant #1057741 RJS).
Conflict of interest: The authors declare no conflict of interest.
References
1. El Andaloussi S, Mager I, Breakefield XO etal
(2013) Extracellular vesicles: biology and
emerging therapeutic opportunities. Nat Rev
Drug Discov 12:347357
2. Rak J, Guha A (2012) Extracellular vesicles
vehicles that spread cancer genes. Bioessays
34:489497
3. Raposo G, Stoorvogel W (2013) Extracellular
vesicles: exosomes, microvesicles, and friends.
J Cell Biol 200:373383
4. Simpson RJ, Lim JW, Moritz RL etal (2009)
Exosomes: proteomic insights and diagnostic
potential. Expert Rev Proteomics 6:267283
5. Tauro BJ, Greening DW, Mathias RA etal
(2012) Comparison of ultracentrifugation,
density gradient separation, and immunoaffinity capture methods for isolating human colon
cancer cell line LIM1863-derived exosomes.
Methods 56:293304
6. Tauro BJ, Greening DW, Mathias RA etal
(2013) Two distinct populations of exosomes
are released from LIM1863 colon carcinoma
cell-derived organoids. Mol Cell Proteomics
12:587598
7. Ji H, Erfani N, Tauro BJ etal (2008) Difference
gel electrophoresis analysis of Ras-transformed
fibroblast cell-derived exosomes. Electro
phoresis 29:26602671
8. Ji H, Greening DW, Barnes TW etal (2013)
Proteome profiling of exosomes derived from
human primary and metastatic colorectal
c ancer cells reveal differential expression of key

metastatic factors and signal transduction components. Proteomics 13:16721686
9. Tauro BJ, Mathias RA, Greening DW etal
(2013) Oncogenic H-ras reprograms Madin-
Darby canine kidney (MDCK) cell-derived
exosomal proteins following epithelial-
mesenchymal transition. Mol Cell Proteomics
12:21482159
10. Harding C, Stahl P (1983) Transferrin recycling in reticulocytes: pH and iron are important determinants of ligand binding and
processing. Biochem Biophys Res Commun
113:650658
11. Pan BT, Johnstone RM (1983) Fate of the
transferrin receptor during maturation of sheep
reticulocytes invitro: selective externalization
of the receptor. Cell 33:967978
12. Valencia K, Luis-Ravelo D, Bovy N etal (2014)
miRNA cargo within exosome-like vesicle
transfer influences metastatic bone colonization. Mol Oncol 8:689703
13. Record M, Carayon K, Poirot M etal (2014)
Exosomes as new vesicular lipid transporters
involved in cell-cell communication and various pathophysiologies. Biochim Biophys Acta
1841:108120
14. Mahaweni NM, Kaijen-Lambers ME, Dekkers
J etal (2013) Tumour-derived exosomes as
antigen delivery carriers in dendritic cell-based
immunotherapy for malignant mesothelioma.
208
J Extracell Vesicles 2, doi:10.3402/jev.

v2i0.22492
15. El Andaloussi S, Lakhal S, Mager I etal (2013)
Exosomes for targeted siRNA delivery across
biological barriers. Adv Drug Deliv Rev
65:391397
16. Johnsen KB, Gudbergsson JM, Skov MN etal
(2014) A comprehensive overview of exosomes
as drug delivery vehicles endogenous nanocarriers for targeted cancer therapy. Biochim
17. Vlassov AV, Magdaleno S, Setterquist R etal
(2012) Exosomes: current knowledge of their
composition, biological functions, and diagnostic and therapeutic potentials. Biochim
18. Morello M, Minciacchi VR, de Candia P etal
(2013) Large oncosomes mediate intercellular
transfer of functional microRNA.Cell Cycle
12:35263536
19. Mangeot PE, Dollet S, Girard M etal (2011)
Protein transfer into human cells by VSV-G-
induced nanovesicles. Mol Ther 19:
16561666
20. Cocucci E, Racchetti G, Meldolesi J (2009)
Shedding microvesicles: artefacts no more.
Trends Cell Biol 19:4351
21. Ji H, Chen M, Greening DW, He W, Rai A,
Zhang W, Simpson RJ (2014) Deep sequencing of RNA from three different extracellular
vesicle (EV) subtypes released from the human
LIM1863 colon cancer cell line uncovers distinct miRNA-enrichment signatures. PLoS
One. 9(10):e110314. PMID: 25330373

22. Whitehead RH, Jones JK, Gabriel A etal
(1987) A new colon carcinoma cell line
(LIM1863) that grows as organoids with spontaneous differentiation into crypt-like structures invitro. Cancer Res 47:26832689

23. Schroder M, Schafer R, Friedl P (1997)
Spectrophotometric determination of iodixanol in subcellular fractions of mammalian cells.
24. Greening DW, Simpson RJ (2010) A centrifugal ultrafiltration strategy for isolating the
low-molecular weight (<or=25K) component
of human plasma proteome. J Proteomics
73:637648
25. Olsen JV, de Godoy LM, Li G etal (2005)
Parts per million mass accuracy on an Orbitrap
mass spectrometer via lock mass injection into
a C-trap. Mol Cell Proteomics 4:20102021
26. Pruitt KD, Tatusova T, Maglott DR (2007)
NCBI reference sequences (RefSeq): a curated
non-redundant sequence database of genomes,
transcripts and proteins. Nucleic Acids Res
35(Database Issue):6165
27. Perkins DN, Pappin DJ, Creasy DM etal (1999)

Probability-based protein identification by
searching sequence databases using mass spectrometry data. Electrophoresis 20:35513567
28. Greening DW, Glenister KM, Kapp EA etal
(2008) Comparison of human platelet
membrane-
cytoskeletal proteins with the
plasma proteome: towards understanding the
platelet-plasma nexus. Proteomics Clin Appl
2:6377
29. Mishra GR, Suresh M, Kumaran K etal (2006)
Human protein reference database 2006
update. Nucleic Acids Res 34(Database Issue):
411414
30. Letunic I, Doerks T, Bork P (2009) SMART 6:
recent updates and new developments. Nucleic
Acids Res 37(Database Issue):229232

31. Sonnhammer EL, von Heijne G, Krogh A
(1998) A hidden Markov model for predicting
transmembrane helices in protein sequences.
Proc Int Conf Intell Syst Mol Biol 6:175182

32. Chen YS, Mathias RA, Mathivanan S etal
(2011) Proteomics profiling of Madin-Darby
canine kidney plasma membranes reveals Wnt-5a
involvement during oncogenic H-Ras/TGFbeta-mediated epithelial-mesenchymal transition. Mol Cell Proteomics 10:M110.001131
33. Beissbarth T, Hyde L, Smyth GK etal (2004)
Statistical modeling of sequencing errors in
SAGE libraries. Bioinformatics 20(Suppl 1):
i31i39
34. Old WM, Meyer-Arendt K, Aveline-Wolf L etal
(2005) Comparison of label-free methods for
quantifying human proteins by shotgun proteomics. Mol Cell Proteomics 4:14871502
35. Mitchell JP, Court J, Mason MD etal (2008)
Increased exosome production from tumour
cell cultures using the Integra CELLine Culture
System. J Immunol Methods 335:98105

36. Greening DW, Simpson RJ (2011) Low-
molecular weight plasma proteome analysis
using centrifugal ultrafiltration. Methods Mol
Biol 728:109124
37. Steinberg TH, Lauber WM, Berggren K etal
(2000) Fluorescence detection of proteins in
sodium dodecyl sulfate-polyacrylamide gels
using environmentally benign, nonfixative,
saline solution. Electrophoresis 21:497508
38. White IR, Pickford R, Wood J etal (2004) A
statistical comparison of silver and SYPRO
Ruby staining for proteomic analysis.

39. Sokolova V, Ludwig AK, Hornung S etal
(2011) Characterisation of exosomes derived
from human cells by nanoparticle tracking
analysis and scanning electron microscopy.
Colloids Surf B Biointerfaces 87:146150
209
tain CD45: no evidence for a distinct exosome/

40. Cantin R, Diou J, Belanger D etal (2008)
HIV-1 budding pathway. Retrovirology 5:64
Discrimination between exosomes and HIV-1:
purification of both vesicles from cell-free 45. Grigorieff N, Harrison SC (2011) Near-atomic
supernatants. J Immunol Methods 338:2130
resolution reconstructions of icosahedral
viruses from electron cryo-microscopy. Curr
41. Mathivanan S, Lim JW, Tauro BJ etal (2010)
Opin Struct Biol 21:265273
Proteomics analysis of A33 immunoaffinity-
purified exosomes released from the human 46. Koning RI, Koster AJ (2009) Cryo-electron
colon tumor cell line LIM1215 reveals a
tomography in biology and medicine. Ann
tissue-
specific protein signature. Mol Cell
Anat 191:427445
Proteomics 9:197208
47. Tatischeff I, Larquet E, Falcon-Perez JM

42. Clayton A, Mitchell JP, Court J etal (2008)
etal (2012) Fast characterisation of cellHuman tumor-derived exosomes down-
derived extracellular vesicles by nanoparticles
modulate NKG2D expression. J Immunol
tracking analysis, cryo-electron microscopy,
180:72497258
and Raman tweezers microspectroscopy. J
43. Koga K, Matsumoto K, Akiyoshi T etal (2005)
Extracell Vesicles 1, doi:110.3402/jev.v1i0.
Purification, characterization and biological
19179
significance of tumor-derived exosomes. 48. Asara JM, Christofk HR, Freimark LM etal
Anticancer Res 25:37033707
(2008) A label-free quantification method by
44. Coren LV, Shatzer T, Ott DE (2008) CD45
MS/MS TIC compared to SILAC and spectral
immunoaffinity depletion of vesicles from
counting in a proteomics screen. Proteomics
Jurkat T cells demonstrates that exosomes con8:994999
Chapter 16
Chloroplast Isolation and Affinity Chromatography
for Enrichment of Low-Abundant Proteins in Complex
Proteomes
Roman G. Bayer, Simon Stael, and Markus Teige
Abstract
Detailed knowledge of the proteome is crucial to advance the biological sciences. Low-abundant proteins
are of particular interest to many biologists as they include, for example those proteins involved in signal
transduction. Recent technological advances resulted in a tremendous increase in protein identification
sensitivity by mass spectrometry (MS). However, the dynamic range in protein abundance still forms a
fundamental problem that limits the detection of low-abundant proteins in complex proteomes. These
proteins will typically escape detection in shotgun MS experiments due to the presence of other proteins
at an abundance several-fold higher in order of magnitude. Therefore, specific enrichment strategies are
required to overcome this technical limitation of MS-based protein discovery. We have searched for novel
signal transduction proteins, more specifically kinases and calcium-binding proteins, and here we describe
different approaches for enrichment of these low-abundant proteins from isolated chloroplasts from pea
and Arabidopsis for subsequent proteomic analysis by MS. These approaches could be extended to include
other signal transduction proteins and target different organelles.
Key words Chloroplast isolation, Affinity chromatography, Mass spectrometry, Proteomics,
Organellar proteome, ATP-binding protein, Calcium-binding protein
Introduction
Organellar proteomics has become a strong focus in (plant) biology over the last years due to several aspects: Organelles perform
key activities for cellular metabolism and provide the basis for
growth and production of valuable compounds [1]. Moreover, the
use of isolated organelles as starting material does already significantly reduce the complexity of the sample as compared to total
cell extracts. However, in the case of leaves the difference between
isolated chloroplasts and total leaves becomes rather small in terms
of their overall protein composition, thus calling for additional
enrichment strategies for the identification of low-abundant
211
212
Roman G. Bayer et al.
proteins such as signaling components. With the notion that

organelles are increasingly being seen as important signal transduction hubs in the cell, for example in retrograde signaling, oxidative
stress signaling, hormone signaling and defense response [24],
this latter aspect becomes more and more important.
The study of chloroplasts, as a central organelle to these pathways and specific to plants, has strong implications for agricultural
research and consequently, chloroplast proteomics from a number
of different plant species has gained a lot of attention during the
last decade [58]. However, despite all efforts, the knowledge of
regulatory components in chloroplasts such as protein kinases is
still very limited [9], underpinning the need for application of specifically targeted enrichment strategies [10]. This is even more
important as it turned out that particularly for those components
the application of prediction algorithms, which are based on the
genome sequence, is strongly biased [10].
Traditionally, highly pure and intact chloroplasts used for functional studies and protein import experiments were obtained from
Spinacia oleracea or Pisum sativum. Since their genomes have not
been sequenced, chloroplast isolation protocols were later on also
adapted for Arabidopsis thaliana. Our chloroplast isolation protocol from Arabidopsis is based on the method from Kunst [11],
which yields reasonable amounts of intact chloroplasts with stromal content sufficient for downstream purification steps and/or
analyses. To circumvent the labor-intensive isolation of Arabidopsis
chloroplasts for mass spectrometry-based proteomic approaches,
the isolation of chloroplasts from pea provides a useful alternative,
since the release of an expressed sequence tag (EST) database,
which could be used for protein identification [12]. Therefore, we
describe both methods here.
The chloroplast proteome can further be divided into several
sub-compartments: (1) the soluble part, the chloroplast stroma;
(2) the thylakoid membrane, and (3) the chloroplast envelope
membrane. As for the latter two recently proteomics protocols
have been published [13, 14], we focus here exclusively on the
stromal proteins, which become easily accessible after osmotic disruption of isolated chloroplasts. A general flow-scheme of our
experimental strategy is displayed in Fig. 1. After extraction of stromal proteins we perform a size exclusion chromatography to
remove the highly abundant Rubisco protein complex (~540 kDa)
which generally presents the biggest problem for MS-based identification of low-abundant proteins. In addition, also ribosomes are
removed during this step. Subsequently, after size exclusion chromatography, all fractions of interest can be pooled and either
directly be subjected to mass spectrometry or to downstream
purification approaches (i.e., affinity chromatography) in order to
further reduce the sample complexity.
213
Fig. 1 Scheme of the work-flow and typical profile of a size exclusion chromatography (SEC) of stroma extracts
from isolated chloroplasts on a Superdex S200 gel filtration column. (a) Scheme of the work-flow to stroma
extract. (b) Scheme of a typical elution profile of stromal proteins on a Superdex S-200 size exclusion column.
X-axis shows ml of eluting sample. Y-axis shows OD280 indicating protein content. Fractions eluting after the
large peak of Ribulose 1,5 Bisphosphate Carboxylase/Oxygenase (RuBisCO) are pooled and subjected to: MS
mass spectrometry, IEX ion exchange chromatography, AC affinity chromatography, HIC hydrophobic interaction chromatography; (c) SDS PAGE of 1.44 ml fractions collected from 73 ml to 96 ml of a Supdex S200 16/60
gel filtration column. T total stromal protein extract loaded
Materials
2.1 Plant Material

(See Note 1)
1. Leaves from Arabidopsis: Arabidopsis plants are grown for

~68 weeks under short day conditions (photoperiod of 16 h
dark/8 h light at 100150 mol/m2s; 22 C +/5; humidity
60 % +/20 %) on soil.
214
2. Leaves from Pea: Pea seedlings are grown for 89 days under
long day conditions (photoperiod of 8 h dark/16 h light at
~70 mol/m2s; 21 C +/5; humidity 7090 %) on vermiculite.
2.2 Chloroplast
Isolation from Pea
(Pisum sativum)
1. P-ISO buffer: 330 mM sorbitol, 20 mM MOPS, 13 mM Tris

(see Note 2), 0.1 % BSA, 3 mM MgCl2 (store at 4 C <1 week).
2. P-WASH buffer: 330 mM sorbitol, 50 mM HEPES-KOH,
pH 7.6, 3 mM MgCl2 (store at 20 C).
3. 80 % Percoll solution: 330 mM sorbitol, 50 mM HEPESKOH, pH 7.6, 80 % Percoll (GE healthcare life sciences, v/v)
(store at 20 C).
4. 40 % Percoll solution: 330 mM sorbitol, 50 mM HEPESKOH, pH 7.6, 40 % Percoll (GE healthcare life sciences, v/v)
(store at 20 C).
5. Percoll step gradient: 7 ml of 80 % Percoll solution are placed
carefully below 12 ml 40 % Percoll with a 10 ml pipette in a
round-bottom tube.
6. Miracloth (Merck).
7. Waring blender (see Note 3).
2.3 Chloroplast
Isolation from
Arabidopsis
1. Homogenization buffer (HB buffer): 450 mM sorbitol,

20 mM Tricine-KOH, pH 8.4, 10 mM EDTA, 5 mM
NaHCO3, 0.1 % BSA, 10 mM isoascorbate, 1 mM reduced
glutathione (see Note 4).
2. Re-suspension buffer (RB buffer): 300 mM sorbitol, 20 mM
Tricine-KOH, pH 7.6, 5 mM MgCl2, 2.5 mM EDTA (store
at 4 C).
3. Percoll continuous gradient: Per gradient mix 15 ml Percoll
with 15 ml 2 RB buffer (600 mM sorbitol, 40 mM TricineKOH, pH 7.6, 10 mM MgCl2, 5 mM EDTA, store at 4 C
<1 week) in centrifugation tubes and centrifuge for 30 min at
~53,000 g in a swing-out rotor (brake slow) in an ultracentrifuge to form a continuous gradient.
4. Miracloth (Merck).
5. Waring blender (see Note 3).
2.4 Solutions
for Chloroplast Lysis
1. Osmotic lysis buffer: 10 mM Tricine-KOH, pH 8.0, 10 mM

MgCl2, 1 mM DTT and protease inhibitor cocktail (Complete
Mini EDTAfree, Roche).
2. THY buffer: 25 mM Tricine-KOH, pH 8.0, 10 mM MgCl2,
10 mM isoascorbate, 2 mM -mercaptoethanol (store at
20 C).
2.5 Gel Filtration

Superdex 200 (S200)
1. S200-A buffer: 50 mM TrisHCl, pH 7.8, 50 mM NaCl,

10 mM MgCl2. Filter buffer through a 0.22 m membrane
and degas before use on chromatography (e.g., FPLC).
215
2. PD-10 desalting columns (GE Healthcare).

3. Centriprep Centrifugal Filter Units (molecular weight cut-off:
10 kDa, Millipore).
4. S200 gel filtration column (GE Healthcare).
2.6 Gel Filtration
Superdex 75 (S75)
1. S200-A buffer: 50 mM TrisHCl, pH 7.8, 50 mM NaCl,

10 mM MgCl2. Filter buffer through a 0.22 m membrane
and degas before use on chromatography (e.g., FPLC).
10 kDa, Millipore).
3. S75 gel filtration column (GE Healthcare).
2.7 Ion Exchange

Chromatography
MonoQ/MonoS
1. MQ-A buffer for MonoQ: 20 mM TrisHCl, pH 8.0. Filter

the buffer through a 0.22 m membrane and degas.
2. MQ-B buffer for MonoQ: 20 mM TrisHCl, pH 8.0, 1 M
NaCl. Filter the buffer through a 0.22 m membrane and degas.
3. MS-A buffer for MonoS: 20 mM MESNaOH (pH 6.0); filter
buffer through a 0.22 m membrane and degas.
4. MS-B buffer for MonoS: 20 mM MESNaOH, pH 6.0, 1 M
NaCl. Filter the buffer through a 0.22 m membrane and degas.
5. MonoQ column for anion exchange chromatography (GE
Healthcare).
6. MonoS column for cation exchange chromatography (GE
Healthcare).
10 kDa, Millipore).
2.8 Hydrophobic
Interaction
Chromatography
(HIC)On
Phenyl-Superose
1. PS-A buffer: 50 mM sodium phosphate buffer, pH 7.0. Filter

the buffer through a 0.22 m membrane and degas.
2. PS-B buffer: 50 mM sodium phosphate buffer, pH 7.0 (prepared according to Lab FAQS from Roche Applied Sciences),
1.5 M (NH4)2SO4. Filter the buffer through a 0.22 m membrane and degas.
10 kDa, Millipore).
5. Phenyl-Superose column (GE Healthcare).
2.9 ATP/Purvalanol
B Affinity
Chromatography
1. ATP buffer: buffer S200-A (see Subheading 2.5, item 1) +

100 mM NaCl, 0.05 % NP-40.
2. PurB buffer: buffer S200-A (see Subheading 2.5, item 1) +
350 mM NaCl, 0.5 % Triton X-100.
216
3. C10-linked Aminophenyl-ATP-Sepharose (Jena Bioscience,

Jena, Germany, see Note 14): Pour affinity Sepharoses (500 l
of slurry) into disposable polystyrene columns (Thermo
Scientific) and always run by gravity flow at room temperature.
4. The preparation of Purvalanol B (PurB) affinity Sepharose is
described in Wissing et al. [15].
2.10 Heat Treatment
of Isolated
Chloroplasts and
Protein Extraction
1. Chloroplast lysis buffer for subsequent heat treatment: 20 mM

DTT, 0.1 % Triton X-100, protease inhibitor cocktail
(Complete Mini EDTA-free, Roche).
2.11 Eu3+-IDA
Column Affinity
Chromatography
1. Equilibration buffer: 100 mM TrisHCl, pH 7.5, 2 M NaCl,

200 mM CaCl2.
2. IDA column-loading buffer: 100 mM TrisHCl, pH 7.5, 3 M
NaCl, 200 mM CaCl2.
3. Sulfate buffer: 600 mM Na2SO4, 100 mM TrisHCl, pH 7.5,
2 M NaCl.
4. Malonate buffer: 40 mM malonate, 600 mM Na2SO4, 100 mM
TrisHCl, pH 7.5, 2 M NaCl.
5. Citrate buffer: 0.2 M phosphate buffer, pH 7.5, 3 M NaCl,
200 mM citrate.
6. Fill 1 ml of IDASepharose (Thermo Scientific) in a disposable polystyrene column and wash with 5 ml of 100 mM
EDTA (pH 7.0), followed by 10 ml of double-distilled water.
Load the column with 5 CV of 50 mM EuCl3 (Alfa Aesar,
USA), wash with 25 ml double-distilled water, and equilibrate
with 10 ml of equilibration buffer.
Methods (See Note 5)
3.1 Chloroplast
Isolation from Pea
(Pisum sativum)
(See Note 6)
The chloroplast isolation protocol for pea is adapted from the

method published by Schleiff [16].
1. Cut about 100120 g leaves from the shoots and homogenize
in about 150 ml P-ISO buffer in a Waring blender using three
pulses: lowhighlow for 3 s each (see Notes 3 and 7).
2. Filter the homogenate through four layers of Miracloth into
four 50 ml round-bottom tubes and centrifuge for 2 min at
2,800 g (brake on).
3. Gently re-suspend the pellets by pipetting in 1 ml P-WASH
buffer (see Note 8), load this on top of two Percoll step gradients and centrifuge for 5 min at 8,000 g (brake off) using
a swing-out rotor (for example Sorvall HB-4) (see Note 9).
217
4. Remove broken chloroplasts on top of the 40 % Percoll layer

with a vacuum pump together with excess 40 % Percoll layer
but leave some of the 40 % Percoll on the 4080 % interphase,
not to expose intact chloroplasts to air, while recovering them.
5. Recover intact chloroplasts after centrifugation from the
4080 % interphase and transfer them into two 50 ml tubes.
6. Wash the isolated chloroplasts with ~30 ml P-WASH buffer
and precipitate by centrifugation for 2 min at 2,800 g
(brake on).
For use of the chloroplasts in protein import experiments, the
washing step should be repeated. Finally, re-suspend the pellets in
~500 l P-WASH buffer; pool, and either use directly for further
analysis (see Notes 4, 5 and 8) or freeze in liquid nitrogen for storage at 80 C (see Note 11). Do not forget to take a sample for
chlorophyll content estimation (see Subheading 3.3).
3.2 Chloroplast
Isolation from
Arabidopsis thaliana
(See Note 6)
The experimental procedure for the isolation of chloroplasts from

Arabidopsis was adapted from the protocol of Kunst [11]. Prepare
Percoll gradients prior to the harvesting of leaves!
1. Cut ~90 g of leaves and homogenize them in 500 ml of homogenization buffer (HB buffer) using a Waring blender and three
pulses: lowlowhigh, 23 s each (see Notes 3 and 7).
2. Filter the homogenate through four layers of Miracloth
into one GS3 tube and centrifuge for 5 min at 1,519 g
(brake on).
3. Gently re-suspend the pellet in 20 ml 1 re-suspension buffer
(RB buffer) using a fine paintbrush and distribute the chloroplast suspension to the four preformed Percoll gradients
(see Notes 8 and 9), 5 ml on top of each, using a pipette.
4. Centrifuge for 6 min at ~10,700 g (brake slow), recover
intact chloroplasts (lower green band of each gradient) and
transfer them into two 50 ml tubes (see Note 9). Take care not
to touch the pellet of the tube containing starch granules.
5. Wash the chloroplasts with 1 RB buffer by centrifugation for
3 min at 1,700 g (brake on) and discard the supernatant.
Re-suspend each pellet in ~300 l 1 RB buffer, pool the isolated
chloroplasts, and immediately freeze them in liquid nitrogen
and store at 80 C (see Note 10). Do not forget to take a
sample for chlorophyll content estimation (see Subheading 3.3).
3.3 Estimation of the

Chlorophyll Content of
Isolated Chloroplasts
According to Arnon [17]
1. Mix 5 l isolated chloroplast suspension with 5 ml 80 % acetone and centrifugate for 2 min at 3,000 g. Subsequently,
measure the OD645 and OD663 of the supernatant (using glass
or quartz cuvettes) and determine chlorophyll concentration
using the following formula (see Note 11):
218
(OD645 20.2 + OD663 8.02) 1,000 = g/ml chlorophyll in

the sample
If the yield is too low see Notes 12 and 13.
3.4 Chloroplast
Stroma Extraction
1. Incubate chloroplasts (amount equivalent to ~20 mg of chlorophyll) in ~1.5 volumes of hypotonic osmotic lysis buffer for
5 min on ice to achieve osmolytic bursting of the envelopes.
2. Centrifuge for 6 min at 12,000 g at 4 C and transfer the
supernatant containing the stromal content to a 50 ml tube.
3. Re-suspend the chloroplast pellet in ~5 ml osmotic lysis buffer
and repeat the extraction step once.
4. Pool the stromal protein extracts and keep on ice until further
treatment.
The thylakoid pellet can be re-suspended in ~5 ml THY buffer
and stored at 80 C for future usage.
3.5 Gel Filtration

Superdex 200 (S200)
1. Prior to the size exclusion chromatography (SEC), exchange

the buffer of the stromal protein extracts to buffer S200-A on
PD-10 Desalting columns according to the manufacturers
instructions.
2. Concentrate the protein solution to a volume of ~500 l using
Centriprep Centrifugal Filter Units.
3. Clear the extract by centrifugation for 10 min at 16,100 g
and 4 C, and apply the supernatant to an S200 gel filtration
column.
4. Perform size exclusion chromatography on an FPLC system at
a flow rate of 0.8 ml/min using the S200-A as running buffer,
fractionate the eluate and store at 4 C until further treatment.
The protein content of the eluting fractions can be determined,
for example by photometry (e.g., measuring OD280) or by SDSPAGE after TCA precipitation (using a standard protocol).
Functional assays such as in vitro protein kinase assays can also be
easily performed on the protein fractions.
3.6 Gel Filtration

Superdex 75 (S75)
1. Pool the fractions of interest from the S200 gel filtration and
concentrate them to a volume of ~500 l using a Centriprep
Centrifugal Filter Unit (NMWL: 10 kDa). After clarification
by centrifugation for 3 min at 16,100 g and 4 C apply the
supernatant to a S75 gel filtration column and perform size
exclusion chromatography on an FPLC system at a flow rate of
0.4 ml/min using solution S200-A as running buffer.
Fractionate the eluate and conduct protein content measurements and/or functional assays as desired. Store protein samples at 4 C until further treatment.
3.7 Ion Exchange

Chromatography
MonoQ/MonoS
219
1. Concentrate protein samples obtained from size exclusion

chromatography to a volume of ~500 l using a Centriprep
Centrifugal Filter Unit.
2. Centrifuge for 10 min at 16,100 g at 4 C and apply the
supernatant to a MonoS or MonoQ column.
3. Perform ion exchange chromatography on an FPLC system at
a flow rate of 2 ml/min.
4. Elute proteins by applying a linear gradient over five column
volumes (CV) of buffer B from 0 to 100 % (MQ-B for MonoQ
and MS-B for MonoS column, respectively).
5. Fractionate the eluate and conduct protein content measurements and/or functional assays as desired. Store protein samples at 4 C until further treatment.
3.8 Hydrophobic
Interaction
Chromatography
(HIC)On
Phenyl-Superose
1. Buffer exchange the protein extract of interest to PS-B buffer

using PD-10 Desalting columns according to the manufacturers instructions.
2. Concentrate the sample to a volume of ~500 l using a
Centriprep Centrifugal Filter Unit, clear by centrifugation for
10 min at 16,100 g at 4 C, and apply the supernatant to a
Phenyl-Superose column.
3. Perform hydrophobic interaction chromatography on an
FPLC system at a flow rate of 2 ml/min and elute proteins by
applying a gradient of PS-B and PS-A buffer. Fractionate the
eluate and conduct protein content measurements and/or
functional assays as desired. Store protein samples at 4 C until
further treatment.
3.9 ATP/Purvalanol
B Affinity
Chromatography
1. Equilibrate ATP or PurB column with ten column volumes

(CV) of ATP buffer or PurB buffer, respectively.
2. Pool the protein extracts of interest (e.g., eluates from size
exclusion or ion exchange chromatography; amount equivalent up to ~1.5 mg protein) and adjust the ionic content to
ATP buffer or PurB buffer by dilution with an appropriate
stock solution, and apply the sample to the column.
3. Wash the column with 20 CV of ATP buffer or PurB buffer
and elute bound proteins with 6 CV of 0.5 % SDS.
4. Fractionate the eluate and analyze, for example by SDS-PAGE
after TCA precipitation.
Regions of interest on the protein gel (e.g., after Coomassie or
silver staining) can subsequently be excised and subjected to MS
analysis.
220
3.10 Heat Treatment

of Isolated
Chloroplasts and
Protein Extraction
Some small proteins like calmodulins are particularly resistant to

heat treatment, which allows a rapid and easy enrichment from
complex protein mixtures, because larger proteins will typically
denature upon heating and precipitate after centrifugation.
1. Lyse 3 ml isolated chloroplasts (containing ~4 mg/ml chlorophyll) by addition of 7 ml of chloroplast lysis buffer and incubate for 10 min on ice.
2. Divide the chloroplast suspension into 1 ml aliquots and heat
rapidly to 75 C for 5 min on a heat block, and immediately
cool on ice.
3. Pellet the heat-denatured proteins and thylakoid membranes
by centrifugation at 20,000 g for 10 min.
4. Clear the supernatant further by another centrifugation for
30 min at 100,000 g at 4 C.
3.11 Eu3+-IDA
Column Affinity
Chromatography
Ca2+ binding proteins can be enriched by affinity chromatography

on immobilized Eu3+ ions using an iminodiacetic acid (IDA) column. This method has previously been described by Chaga et al.
[18], and was adapted to chloroplast proteins. Protein obtained
from chloroplast stroma SEC or from heat-treated isolated chloroplasts was loaded on the column.
1. For heat-treated samples, re-buffer the sample to IDA columnloading buffer, using PD-10 desalting columns. For SEC fractionated stroma extracts, add salts to the extracts to match the
composition of the IDA column-loading buffer.
2. After sample loading, wash the column with 10 ml equilibration
buffer, 5 ml of sulfate buffer, and 2.5 ml of malonate buffer.
3. Elute proteins with citrate buffer. Finally, strip the column
with 100 mM EDTA.
For further use, exchange the buffers of the citrate eluate and
EDTA-strip to 50 mM TrisHCl pH 7.5 on a PD-10 column and
precipitate proteins by addition of four volumes of ice-cold acetone. Proteins can now be subjected to SDS-PAGE, visualized by
silver staining and bands of interest can be excised for subsequent
MS analysis.
Notes
1. Starch granules in the chloroplast will contribute to breaking
of chloroplasts during handling. Therefore prepare chloroplasts in the morning, when starch levels in the chloroplast are
lowest.
2. No titration necessary. Optimal pH is reached by mixing the
two buffer components (MOPS and Tris).
221
3. To gain better yields of intact chloroplasts and to allow a

shorter and more gentle homogenization the blades of the
Waring blender should be as sharp a possible (can usually be
detached for sharpening).
4. Add isoascorbate and glutathione always freshly prior to use
and readjust pH to 8.4 with KOH.
5. Isolation of chloroplasts requires a certain amount of practice
in order to achieve reproducible results and an optimal yield in
terms of intact organelles. However, based on empirical experience there is a natural variation and sometimes the yield of
intact chloroplasts is much lower than expected although the
plants have been grown under the same conditions and the
same technical procedures have been followed.
6. Carry out all steps at 4 C (in a cooling chamber or on ice).
7. If the homogenization appears to be a problem, younger plant
material should be used.
8. When pipetting chloroplast suspensions always use CUT TIPS
in order to minimize mechanical damage to the organelles.
9. One typical error that occurs frequently is that no sorbitol is
added in the Percoll solutions. This will result in osmotic burst
of the chloroplasts.
10. For easy storage and handling, chloroplast suspensions are
pipetted drop by drop directly in liquid nitrogen and afterwards collected in Eppendorf tubes. In this way only part of a
tube can be used, while the rest is kept frozen at 80 C.
11. Expected chloroplast yields at the volumes mentioned in these
protocols are in the range of 12 mg chlorophyll for Arabidopsis
and 710 mg for pea.
12. The stromal content of Arabidopsis chloroplasts is frequently
lost during the isolation procedure [19]. This becomes visible
in a comparison of the total protein content of isolated chloroplasts from different species by SDS-PAGE as, for example,
shown in Fig. 2 for pea and Arabidopsis and could present a
serious problem for proteomic approaches. Therefore, it is
crucial to check the amount of stromal proteins by SDS-PAGE
before further usage when using Arabidopsis leaves as source
for chloroplast isolation.
13. To spot at which step problems might occur in cases where the
yield of intact chloroplasts is too low, just check a drop of the
chloroplast suspension from the different steps by light microscopy (intact chloroplast can easily be recognized by application of phase contrast).
222
Fig. 2 Comparison of the protein content of isolated chloroplasts from different

species. Ps, Pisum sativum. At, Arabidopsis thaliana. Stroma and thylakoid proteins extracted from isolated chloroplasts analyzed by SDS-PAGE. The relative
amount of the large subunit of RuBisCO (52 kDa) can be used as a measure of
chloroplast intactness
Acknowledgements
Work in the authors lab is supported by grants from the Austrian
Science Fund (FWF) to MT (P 23435-B12 and P 25359-B21) and
the Marie Curie Initial training network (ITN) CALIPSO (GA
ITN 2013-607607) from the European Union.
References
1. Rolland N, Curien G, Finazzi G, Kuntz M,
Marechal E, Matringe M et al (2012) The biosynthetic capacities of the plastids and integration between cytoplasmic and chloroplast
processes. Ann Rev Genet 46:233264
2. Jarvis P, Lopez-Juez E (2013) Biogenesis and
homeostasis of chloroplasts and other plastids.
Nat Rev Mol Cell Biol 14:787802
3. Shapiguzov A, Vainonen JP, Wrzaczek M,
Kangasjarvi J (2012) ROS-talkhow the apoplast, the chloroplast, and the nucleus get the
message through. Front Plant Sci 3:292
4. Nomura H, Komori T, Uemura S, Kanda Y,
Shimotani K, Nakai K et al (2012) Chloroplastmediated activation of plant immune signalling in Arabidopsis. Nature Commun 3:926
5. Kleffmann T, Russenberger D, von Zychlinski

A, Christopher W, Sjolander K, Gruissem W
et al (2004) The Arabidopsis thaliana chloroplast proteome reveals pathway abundance and
novel protein functions. Curr Biol 14:354362
6. Zybailov B, Rutschow H, Friso G, Rudella A,
Emanuelsson O, Sun Q et al (2008) Sorting
signals, N-terminal modifications and abundance of the chloroplast proteome. PLoS One
3:e1994
7. Ferro M, Brugiere S, Salvi D, SeigneurinBerny D, Court M, Moyet L et al (2010) AT_
CHLORO, a comprehensive chloroplast
proteome database with subplastidial localization and curated information on envelope proteins. Mol Cell Proteomics 9:10631084

8. Huang M, Friso G, Nishimura K, Qu X,
Olinares PD, Majeran W et al (2013)
Construction of plastid reference proteomes for
maize and Arabidopsis and evaluation of their
orthologous relationships; the concept of
orthoproteomics. J Proteome Res 12:491504
9. Bayer RG, Stael S, Rocha AG, Mair A,
Vothknecht UC, Teige M (2012) Chloroplastlocalized protein kinases: a step forward
towards a complete inventory. J Exp Bot
63:17131723
10. Bayer RG, Stael S, Csaszar E, Teige M (2011)
Mining the soluble chloroplast proteome by affinity chromatography. Proteomics 11:12871299
11. Kunst L (1998) Preparation of physiologically
active chloroplasts from Arabidopsis. Methods
Mol Biol 82:4348
12. Brautigam A, Shrestha RP, Whitten D,
Wilkerson CG, Carr KM, Froehlich JE et al
(2008) Low-coverage massively parallel pyrosequencing of cDNAs enables proteomics in
non-model species: comparison of a speciesspecific database generated by pyrosequencing
with databases from related species for proteome analysis of pea chloroplast envelopes.
J Biotechnol 136:4453
13. Simm S, Papasotiriou DG, Ibrahim M,
Leisegang MS, Muller B, Schorge T et al
(2013) Defining the core proteome of the
14.
15.
16.
17.
18.
19.
223
chloroplast envelope membranes. Front Plant

Sci 4:11
Tomizioli M, Lazar C, Brugiere S, Burger T,
Salvi D, Gatto L et al (2014) Deciphering
thylakoid sub-compartments using a mass
spectrometry-based approach. Mol Cell
Wissing J, Jansch L, Nimtz M, Dieterich G,
Hornberger R, Keri G et al (2007) Proteomics
analysis of protein kinases by target classselective prefractionation and tandem mass
spectrometry. Mol Cell Proteomics 6:
537547
Schleiff E, Soll J, Kuchler M, Kuhlbrandt W,
Harrer R (2003) Characterization of the translocon of the outer envelope of chloroplasts.
J Cell Biol 160:541551
Arnon DI (1949) Copper enzymes in isolated
chloroplasts. Polyphenoloxidase in beta vulgaris. Plant Physiol 24:115
Chaga GS, Ersson B, Porath JO (1996)
Isolation of calcium-binding proteins on selective adsorbents. Application to purification of
bovine calmodulin. J Chromatogr A 732:
261269
Halliwell B (1978) The chloroplast at work. A
review of modern developments in our understanding of chloroplast metabolism. Prog
Biophys Mol Biol 33:154
Chapter 17
Depletion ofRuBisCO Protein Using theProtamine
Sulfate Precipitation Method
RaviGupta andSunTaeKim
Abstract
Ribulose-1,5-bisphosphate carboxylase/oxygenase (RuBisCO) is a major high-abundant protein (HAP) in
the plant leaves which hinders analysis of low-abundant proteins (LAP). In this chapter, we describe a
highly simple RuBisCO depletion method using protamine sulfate (PS). Addition of 0.1% PS is sufficient
to precipitate the RuBisCO from the leaf extracts of diverse plants including monocots and dicots. Our
results of S
DS-PAGE, Western blotting, and two-dimensional gel electrophoresis showed that both large
and small subunits of RuBisCO were precipitated in the pellet fractions, while LAPs were enriched in the
supernatant fraction after PS precipitation.
Key words RuBisCO, Protamine sulfate, High-abundant proteins, Low-abundant proteins, Two-
dimensional gel electrophoresis
1 Introduction
In the last two decades, the focus of research has been shifted from
genomics to proteomics due to the fact that proteins are the real
executors of the gene expression and thus can provide a better
picture of functional aspects of the cells [1]. As plants perceive
signals from the environment through leaves, one of the major
goals of the scientific community is to dissect the components of
signaling cascades which are initiated in response to stresses or
other environmental cues like photoperiodism, etc, by analyzing
the leaf proteome. However, the major bottleneck for analysis of
leaf proteome is presence of a high-abundant protein (HAP)
RuBisCO, which mask the expression and identification of lowabundant proteins (LAP) and thus hinders their analysis [2]. Mass
spectrometry-based identification of the leaf proteins often results
in the repeated identification of large and small subunits of
RuBisCO, resulting in a noteworthy loss of time and money [3].
RuBisCO is present in the order of 105107 molecules per cell
and therefore, mask the LAPs like phosphatase, kinase, t ranscriptional
225
226
factors and other regulatory proteins whose concentrations are less

than 10100 molecules per cell [4]. As LAPs are important component of signaling, it is very important to analyze these in order to
understand the physiology of plant cells. Therefore, several methods using calcium/phytate [5], polyethylene glycol [6, 7], DTT
[8], RuBisCO IgY affinity [9, 10], etc., have been proposed to
deplete RuBisCO for the enrichment of LAPs. However, most of
these methods are laborious and time consuming. Furthermore,
most of these methods are comprised of multiple washing steps
which often results in protein loss. Recently, we reported a protamine sulfate-based preferential precipitation of RuBisCO from
plant leaves which results in significant enrichment of LAPs [11].
This novel method is simple, rapid, cost-effective, and universal
(i.e., can be used for both monocots and dicots).
In this chapter, we describe PS protocol in details so that it can
be followed up by other researchers.
2 Materials
2.1 Protein
Extraction, RuBisCO
Precipitation,
andProtein
Quantitation
Components
1. Pre-chilled pestle and mortar: Take a set of pestle and mortar

and pre-chill it using liquid nitrogen.
2. Tris-Mg-NP-40 buffer: 0.5M TrisHCl, pH8.3, 20mM
MgCl2 and 2% NP-40. Take about 800mL of double-distilled
water and add 60.57g of Tris and 1.904g of MgCl2. Dissolve
them using magnetic stirrer. After complete dissolution of Tris
and MgCl2, add 20mL of NP-40 (see Note 1). Mix it again
using magnetic stirrer. Set pH of the solution to 8.3 using 6M
HCl (see Note 2). Make final volume 1L and autoclave. NP-40
may precipitate after autoclave therefore, re-dissolve the solution using magnetic stirrer after autoclaving and store at 4C.
3. Water-saturated phenol: Melt phenol at 60C.Add a pinch of
8-Hydroxyqunoline, which is an antioxidant reagent. Add
equal volume of double-distilled water and stirred using magnetic stirrer for 30min. Allow separation of upper aqueous
phase and lower phenol phase. After separation, remove upper
aqueous phase and replace with fresh double-distilled water and
again stirred for another 30min. Repeat this process until the
pH of phenol reaches to 6.57.0 (usually it takes 1 day). Store
water-saturated phenol at 4C in dark bottle (see Note 3).
4. 1% Protamine sulfate solution: Dissolve 0.1g of protamine
sulfate in 10mL double-distilled water. Keep it on a rocker to
completely dissolve (see Note 4).
5. 12.5% TCA/acetone: Dissolve 12.5g of TCA in 1L of 100%
acetone. Store at 20C in a dark bottle.
6. 80% Acetone: Take 800mL of 100% acetone and add 200mL
of double-distilled water. Mix it briefly and store at 20C.
7. Protein assay: 2D-Quant kit (GE Healthcare).
RuBisCO Depletion Using Protamine Sulfate
2.2 SDS-PAGE
Components
227
1. 30% Acrylamide solution: Take 30g of acrylamide and 0.8g of

bis-acrylamide (see Note 5) and dissolve in 60mL of double-
distilled water in an amber beaker (or normal beaker covered
with aluminum foil). After complete dissolution of the components, make final volume to 100mL.Filter the solution using
an appropriate filter (0.45m pore size) and store in an amber
bottle at 4C.
2. Separating gel buffer: 1.5M TrisHCl, pH8.8. Take 181.7g
Tris base and dissolve it in 800mL of double-distilled water.
Adjust the pH of the solution to 8.8 using 6M HCl (see Note 2).
Make final volume to 1L.Store at 4C.
3. Stacking gel buffer: 0.5M TrisHCl, pH6.8. Dissolve 60.6g
Tris base in 800mL double-distilled water and set the pH to
6.8. Adjust final volume to 1L and store at 4C.
4. 10% Sodium dodecyl sulfate (SDS): Dissolve 10g of SDS in
100mL of double-distilled water and store at RT.
5. 10% Ammonium sulfate: Take 0.1g of ammonium sulfate and
dissolve in 1mL of double-distilled water. Use freshly prepared
ammonium sulfate solution.
6. N,N,N,N-tetramethyl-ethylenediamine (TEMED).
7. SDS-PAGE loading buffer: 50mM TrisHCl, pH6.8, 2% SDS,
10% glycerol, 4% -mercaptoethanol, 0.02% bromophenol
blue. Take 1mL of 0.5M TrisHCl, pH6.8, and dissolve 0.2g
SDS.Add 1mL glycerol and pinch of bromophenol blue. Make
final volume to 9.6mL and store at RT.Add 40L of
-mercaptoethanol in 1mL of loading buffer just prior to use.
8. Running buffer: 25mM Tris, 250mM glycine, 0.1%
SDS.Dissolve 3.02g Tris base and 18.8g glycine in 800mL
of double-distilled water. Add 10mL of 10% SDS and make
the final volume to 1L.Store at RT.
9. Instant Blue stain (Expedeon, UK).
10. Colloidal Coomassie Brilliant Blue (CBB) stain: Take 500mL
of double-distilled water in 2L glass beaker and add 36mL
O-phosphoric acid. Add 170g ammonium sulfate and 1g CBB
G-250 to this solution. Dissolve completely on a magnetic stirrer. Make final volume of the solution to 660mL.Store it in a
dark bottle at room temperature. Add 340mL of methanol
prior to use.
11. Destaining solution: 30% methanol. Add 300mL methanol in
700mL of double-distilled water.
2.3 Immunoblotting
Components
1. 10Tris Buffered Saline (TBS): Take 500mL of double-

distilled water in 2L glass beaker and add 24.2g Tris and 80g
sodium chloride. Dissolve it completely and set pH7.6 using
HCl. Make final volume to 1L.Store it at 4 C.Dilute it to
1 prior to use.
228
2. TBST: 0.1% Tween-20in TBS.Take 100mL of TBS solution

and add 100L of Tween-20 to it. Stir on a magnetic stirrer
for complete dissolution of Tween-20.
3. Transfer buffer for protein transfer: Take 14.4g of glycine and
0.8g of Tris. Dissolve it in 800mL of double-distilled water.
Add 200mL of HPLC grade methanol.
4. Blocking Solution: 5% skim milk solution in TBS.Take 5g of
skim milk powder and dissolve in 90mL of TBS solution.
Make final volume to 100mL.Always use fresh blocking
solution.
5. Polyvinylidene fluoride (PVDF) membrane.
6. Ponceau-S stain.
7. Primary antibodies: Take 1L of primary antibodies of rice
APX1, APXb, Cu/Zn SOD, MnSOD, DHAR, SalT and dilute
in 20mL of TBS (1:20,000 dilution).
8. Secondary antibodies: Take 1L of secondary antibodies and
dilute in 20mL of TBS (1:20,000 dilution).
9. SuperSignal West Pico kit (Pierce, USA).
3 Methods
3.1 Extraction
ofTotal Proteins
andDifferential
Precipitation
ofRuBisCO
1. Take 1g of fresh healthy green leaves, grown in growth c hamber,

and grind it in the pre-chilled pestle and mortar using liquid
nitrogen.
2. Transfer the fine powder to a 50mL round bottom tube and
add 10mL of Mg-NP-40 buffer and vortex vigorously.
3. Sonicate the homogenate for 12min to allow breaking of the
cell walls and then centrifuge at 12,000g for 10min at 4C.
Transfer the supernatant containing total soluble proteins into
a new tube.
4. Take out 1mL of the total protein extract and add chilled
12.5% TCA-acetone to precipitate the proteins. Incubate at
20C for 1h.
5. To the rest of the supernatant fraction, add PS solution to a
final concentration of 0.1% (add 1mL of 1% PS solution to
9mL protein solution). Allow precipitation of RuBisCO by
incubating it on ice for 30min and then centrifuge at 12,000g
for 5 min at 4 C to separate supernatant and pellet.
6. Transfer the supernatant (PSS) into a fresh tube, add four
volumes of chilled 12.5% TCA-acetone and incubate at
20C for 1h.
7. Dissolve the pellet (PSP) in same volume (~10mL) of
Mg-NP-40 buffer, vortex vigorously and sonicate for 12min
229
Grind it in liquid nitrogen

Add 10 mL of Tris-Mg/NP-40 buffer and sonicate
Centrifuge at 12,000 g for 10 min at 4 C
Take out the supernatant in fresh falcon tube
Take 1 mL of this supernatant as total

proteins (T)
Add 1 mL 1% PS solution to 9 mL of
supernatant and incubate on ice for 30 min
Centrifuge at 12,000 g for 10 min at 4 C
Take out the supernatant in fresh

falcon tube (PSS)
Dissolve pellet in 10 mL of Tris-Mg/

NP-40 buffer and sonicate (PSP)
Add 4 volumes of 12.5 % TCA-acetone and incubate at 20 C for 1 hr

Collect supernatant and wash with 80% acetone three times
Flowchart of precipitation of RuBisCO using 0.1% PS
Take 1 gm fresh leaves
Store pellet in 80% acetone until use
Fig. 1 Workflow of the differential precipitation of RuBisCO by Protamine sulfate
to completely dissolve the pellet (see Note 6). Add four

volumes of chilled 12.5% TCA-acetone and incubate at 20C
for 1h.
8. After incubation, centrifuge the total, PSS and PSP fractions at
12,000g for 10min at 4C.Discard the supernatant and wash
the pellets of each fraction (total, PSS, and PSP) with 80%
acetone thrice. Sonicate the pellet each time for complete
suspension of pellet. This will enhance the results of washing.

9. Store the final pellets in 1.2mL of 80% acetone at 20C
until use (Fig.1).
3.2 Protein
Quantification
andSDS-PAGE
Analysis
1. Take out 100L of each sample in fresh Eppendorf tubes and

centrifuge at 12,000g for 5min at 4C to precipitate the
proteins.
2. Air dry the pellets for 510min or vacuum dry for 23min.
3. Dissolve the pellets either in any low ionic strength of buffer of
choice (usually 50mM TrisHCl, pH7.5, is sufficient) or
directly in 1 SDS-PAGE loading buffer.
4. Quantify the proteins in each fraction either using 2-DE quant
kit (GE Healthcare) or by any other commercial kits. If the
proteins are dissolved in SDS-PAGE loading buffer, use 2-D
Quant kit (GE Healthcare) for protein quantification following
manufacturers protocol.
230
Rice
M
Arabidopsis
Rice
T
170
130
100
70
P
APx1
APxb
55
40
Cu/Zn SOD
35
DHAR
25
Mn SOD
15
SalT
RubL
10
pH 4
T
pH 7 pH 4
S
pH 7 pH 4
P
pH 7
408
516
139
Rice
Arabidopsis
590
626
283
Fig. 2 Effect of PS on RuBisCO precipitation and corresponding enrichment of LAPs. (a) After 0.1% PS precipitation, equal amount of total (T), supernatant (S) and pellet (P) proteins of rice and Arabidopsis were loaded on
the 12% SDS-PAGE.LAPs were enriched in the supernatant fractions while RuBisCO was enriched in the pellet
fractions. (b) Western blot analysis of rice proteins to show enrichment of LAPs in the supernatant fraction and
RuBisCO depletion in the pellet fraction. 2-DE gel profiles of rice (c) and Arabidopsis proteins (d) after 0.1%
PS precipitation confirming enrichment of LAPs in supernatant and precipitation of RuBisCO in pellet fractions.
Total number of spots identified in 2-DE gels are mentioned at the bottom of the corresponding gels
5. Cast 12% SDS-PAGE gel according to the Laemelli [12] and

separate equal amount (~20g/lane) of total, PSS, and PSP
proteins (see Note 7).
6. After completion of the gel, stain it with Instant Blue stain and
analyze the results. RuBisCO preferentially precipitates in the
pellet fraction, while LAPs enrich in the supernatant fraction
after 0.1% PS precipitation as can be seen in rice and Arabidopsis
(see Fig.2a).
3.3 Western Blotting
and Two-Dimensional
Gel Electrophoresis
1. After 0.1% PS precipitation, differential precipitation of

RuBisCO in pellet fraction and enrichment of LAPs in supernatant fraction can be further confirmed using Western blotting
and two-dimensional gel electrophoresis (2-DE).
231
2. For Western blotting, resolve the proteins on 12% SDS-PAGE

and then transfer the proteins on a PVDF membrane at 120V
for 1:30h at 4C (see Note 8).
3. Stain the blot using Ponceau-S stain to check transfer of proteins
on the PVDF membrane (see Note 9).
4. Block the membrane using blocking solution for 2h at 37C
(see Note 10).
5. Incubate the membrane with primary antibodies (rice APX1,
APXb, Cu/Zn SOD, MnSOD, DHAR, and SalT) for 2h with
constant shaking. After three washings in TBST, add secondary antibodies and incubate for 12h (for details of Western
blotting experiments, please refer ref. [13]).
6. Wash again with TBST and develop the blot using SuperSignal
West Pico kit according to the manufacturers protocol. After
developing the blot, wash it using double-distilled water and
scan it. APX1, APXb, Cu/Zn SOD, MnSOD, DHAR, and
SalT are enriched in the PSS fraction, while large subunit of
RuBisCO is precipitated in the PSP fraction (see Fig.2b).
7. For 2-DE, dissolve 500g protein of total, PSS, and PSP fractions in lysis buffer and load it on 24cm IPG strips pH47.
After isoelectric focusing, equilibrate the IPG strips and then
resolve the proteins on 12% SDS-PAGE for second dimensional
separation [for details of Western blotting and 2-DE experiments, please refer [11, 13]]. After completion of the gels,
stain the gel using colloidal-CBB stain overnight with constant
shaking. Next day, destain the gels using 30% methanol. 2-DE
gels of rice and Arabidopsis fractions clearly showed enrichment of LAPs in PSS fractions and precipitation of RuBisCO
in PSP fractions (see Fig.2c, d). Thus, based on these results, it
can be concluded that this PS precipitation method is simple,
rapid, cost-effective and can be applied for both monocots
(rice) and dicots (Arabidopsis).
4 Notes
1. Addition of NP-40in the protein extraction buffer solubilizes
the membranes which results in the extraction of membrane
bound proteins also and thus increases the protein yield. In
addition, NP-40 enhances the protein solubility.
2. Initially concentrated HCl can be used to bring the pH of Tris
near 8. After that, diluted HCl should be used to bring the
required pH.While handling HCl, always wear gloves.
3. As phenol is highly volatile and caustic, perform all steps under
Fume Hood only. Always wear gloves and mask while handling
phenol. Avoid contact of phenol to the skin.
232
4. Always use fresh PS solution. It is better to first prepare PS

solution before start of the experiment.
5. Time of sonication can be increased if the pellet is not dissolved
in 12min.
6. Acrylamide is neurotoxic; therefore, always wear gloves and
mask while weighing and preparing acrylamide.
7. Generally 20g of protein is loaded per lane for the CBB staining. However, if the protein amount is less, more sensitive
staining like silver staining can be performed.
8. Nitrocellulose (NC) membrane can also be used for Western
blotting; however, the binding capacity of NC membrane is
less than that of PVDF membrane.
9. Alternatively, a pre-stained protein ladder can be loaded to
check the transfer proteins on NC-membrane.
10. Time of blocking can be varied from 1h to overnight. Usually
1h of blocking is sufficient to prevent the nonspecific binding
of antibodies to the membrane.
Acknowledgement
This work was supported by a grant from the Next-Generation
BioGreen 21 Program (SSAC, grant#: PJ011070032015), Rural
Development Administration (RDA), Republic of Korea.
References
1. Agrawal GK, Rakwal R (2011) Rice proteomics:
a move towards expanded proteome coverage
to comparative and functional proteomics
uncovers the mysteries of rice and plant biology.
2. Rose JKC, Bashir S, Giovannoni JJ, Jahn MJ,
Saravnan RS (2004) Tackling the plant proteome: practical approaches, hurdles and
experimental tools. Plant J 39:715733
3. Schulze WX, Usadel B (2010) Quantitation in
mass-spectrometry-based proteomics. Annu
Rev Plant Biol 61:491516
4. Malcevschi A, Marmiroli N (2012) Plant protein analysis. In: Heazlewood J (ed) Proteomic
applications in biology. InTech Manhattan,
NewYork, ISBN: 978-953-307-613-3
5. Krishnan HB, Natarajan SS (2009) A rapid
method for depletion of Rubisco from soybean
(Glycine max) leaf for proteomic analysis of
lower abundance proteins. Phytochemistry
70:19581964
6. Kim ST, Cho KS, Jang YS, Kang KY (2001)

Two-dimensional electrophoretic analysis of
rice proteins by polyethylene glycol fractionation for protein arrays. Electrophoresis 22:
21032109
7. Xi J, Wang X, Li S, Zhou X, Yue L, Fan J, Hao
D (2006) Polyethylene glycol fractionation
improved detection of low-abundant proteins by
two dimensional electrophoresis analysis of plant
proteome. Phytochemistry 67:23412348
8. Cho JH, Hwang H, Cho MH, Kwon YK, Jeon
JS, Bhoo SE, Hahn TR etal (2008) The effect
of DTT in protein preparations for proteomic
analysis: removal of a highly abundant plant
enzyme, ribulose bisphosphate carboxylase/
oxygenase. J Plant Biol 51:297301
9. Cellar NA, Kuppannan K, Langhorst ML, Ni
W, Xu P, Young SA (2008) Cross species applicability of abundant protein depletion columns
for ribulose-1,5-bisphosphate carboxylase/
oxygenase. J Chromatogr B 861:2939

10. Agrawal GK, Jwa N-S, Rakwal R (2009) Rice
proteomics: ending phase I and the beginning
of phase II.Proteomics 9:935963
11. Kim YJ, Lee HM, Wang Y, Wu J, Kim SG,
Kang KY, Park KH, Kim YC, Choi IS, Agrawal
GK, Rakwal R, Kim ST (2013) Depletion of
abundant plant RuBisCO protein using the
protamine sulfate precipitation method.
233

12. Laemelli UK (1970) Cleavage of structural
proteins during the assembly of the head of
bacteriophage T4. Nature 227:680685
13. Kim ST, Cho KS, Yu S, Kim SG, Hong JC,
Han CD, Bae DW, Nam MH, Kang KY (2003)
Proteomic analysis of differentially expressed
proteins induced by rice blast fungus and elicitor in suspension-cultured rice cells. Proteomics
3:23682378
Chapter 18
Step-by-Step Preparation of Proteins for Mass
Spectrometric Analysis
Abstract
Nowadays, identification of proteins from biological samples by mass spectrometry is widely used. In principle
there are two scenarios. Proteins are pre-fractionated in some way, e.g. by gel electrophoresis or are
analyzed as complex mixture (shot gun). Shot gun proteomics became recently more popular, because of
technological developments on the mass spectrometer side which allows now the identification of several
thousand proteins from complex biological matrix. However, in many cases it is still useful to separate
proteins first in a gel. But not only mass spectrometer technology made progress. This is also true for the
sample preparation. Recently, protocols and techniques were developed which make the analysis of starting
material in the low microgram range possible and also simplify the whole procedure. Detailed protocols
will be described allowing also inexperienced beginners to get good results.
Key words Shotgun proteomics, In-solution digest, In-gel digest, FASP, STAGE-tip, OASIS, Mass
spectrometry, Sample preparation
Introduction
Mass spectrometry became a valuable tool in biological analysis of
proteins. During the recent years, the sensitivity of the instruments
improved dramatically and this development is still ongoing. We
are close being able to analyze whole proteomes of low microgram
range in a single shot (shotgun analysis). The sample preparation
becomes more and more important. It should be quick, relatively
easy to perform, reproducible and no or minimal loss of sample.
Also recently protein sample preparation for mass spectrometry has
made very much progress toward handling of small samples and
easiness. Multistep procedures are usually not suited for starting
material in the low microgram range. The loss of material during
sample preparation is substantial. Therefore, one-step procedures
are preferred. Inexperienced scientific staff is facing the problem
of old protocols in the literature, incomplete handling procedures
or protocols are scattered throughout several publications making
235
236
it not easy getting a working protocol. The lack of important

information or little tricks is the most often reason for irreproducible experiments. Due to limitation in words or pages of publications mostly the experimental part is shortened in a way that it is
impossible to reproduce the procedure without prior knowledge.
Very often the protocols are written by people using the procedures
on a daily basis and handling of samples and techniques used are so
obvious for them that little but important information is simply
forgotten. In this chapter, we will disclose all details and following
the protocols will produce good results instantly. We are using
these protocols ourself for daily work, but also in training courses
for scientist and scientific staff.
Separation of proteins in gels is still widely used therefore we
start with a manual high-throughput method for in-gel digestion
[1, 2] which will work for 1D gel bands as well as for 2D gel spots
picked by a robot ( 1.5 mm) directly into the OASIS HLB
Elution Plate [3]. The investment in the whole OASIS set up is
only a few percent compared to the investment in a full automatic
robot and is in the range of 510 chromatographic columns used
for the mass spectrometric analysis.
Materials
Prepare all solutions and buffers using ultrapure water of 18 M cm
at 25 C, use LC-MS grade solvent and reagents of highest purity.
Follow all waste disposal regulations when disposing waste
materials.
2.1 In-Gel Digestion

by OASIS HLB
Elution Plate
for Coomassie Gels
1. 100 mM ammonium bicarbonate (ABC), pH 8.0, in water.

2. Acetonitrile (ACN).
3. 0.1 % formic acid (FA) for LC-MS sample preparation. For
MALDI-MS (see Note 1).
4. Reduction solution: 10 mM dithiothreitol (DTT) in 100 mM
ABC, pH 8.0.
5. Alkylation solution: 55 mM iodoacetamide (IAA) in 100 mM
ABC, pH 8.0.
6. Digestion solution: 7.4 ng/L trypsin (MS grade) in 50 mM
ABC, pH 8.0.
7. Extraction solution MALDI: acetonitrile:water:trifluoroacetic
acid (TFA) = 49.95 : 49.95 : 0.1 %.
8. Extraction solution LC-MS: acetonitrile:water:FA = 47.5 :
47.5 : 5 %.
9. OASISHLB Elution Plate, 30 m (Waters, USA).
10. Positive pressure-96 stand (Waters, USA).
Sample Preparation for Mass Spectrometry
237
2.2 Destaining [3]

of Mass Spectrometry
Compatible [4, 5]
Silver Gel
1. 100 mM ABC, pH 8.0, in water.
2.3 Filter-Aided
Sample Preparation
(FASP) [6, 7]
1. Lysis Buffer: 2 % SDS and 100 mM DTT in 100 mM Tris

HCl, pH 7.6. Add DDT always just prior use.
2. Acetonitrile.
3. 30 mM potassium ferricyanide K3[Fe(CN)6] in water.
4. 100 mM sodium thiosulfate in water.
2. Buffer B: 8 M urea, 100 mM TrisHCl, pH 8.0.

3. Buffer C: Add 10 mg of iodoacetamide to 1 mL of buffer B.
4. Buffer D: 20 mM TrisHCl, pH 8.5, 10 % ACN.
5. 1 g/L Trypsin (MS grade) in 50 mM acetic acid.
6. 25 mM TrisHCl, pH 8.5.
7. Diagenode Bioruptor plus (Diagenode, Belgium).
8. Nanosep 30 k Omega Centrifugal Device (Pall, USA).
9. Electric mortar (VWR, Germany).
10. NanoDrop 2000 (Thermo Scientific, USA).
2.4 One-Step
In-Solution Protein
Solubilization,
and In-Solution
Digest [8]
1. Guanidine buffer: 6 M Guanidine hydrochloride (GdmCL),

10 mM Tris-(2-carboxyethyl)phosphine hydrochloride (TCEP),
40 mM 2-chloroaceteamide (CAA), 100 mM TrisHCl,
pH 8.5 (see Note 2).
3. Buffer A: 20 mM TrisHCl, pH 8.5, 10 % ACN.
4. Bandelin Sonoplus (Bandelin, Germany).
2.5 One-Step
Guanidine Method
for Large-Scale
Analysis
1. Guanidine buffer: 6 M GdmCL, 10 mM TCEP, 40 mM CAA,

100 mM TrisHCl, pH 8.5 (see Note 2).
3. Buffer A: 20 mM TrisHCl, pH 8.5, 10 % ACN.
5. Electric mortar (VWR, Germany).
6. Nanosep 30 k Omega Centrifugal Device (Pall, USA).
2.6 C. elegans
Sample Preparation
1. Guanidine buffer: 6 M GdmCL, 10 mM TCEP, 40 mM CAA,

100 mM TrisHCl, pH 8.5 (see Note 2).
3. Buffer A: 20 mM Tris, pH 8.5, 10 % ACN.
238
2.7 STAGE Tip

Peptide Cleaning
and Desalting [9]
1. Solution A: 100 % MeOH.

2. Solution B: 0.1 % FA in 80 % ACN.
3. Solution C: 0.1 % FA.
4. EmporeTM C18-SD solid phase extraction disk for desalting
(#66871-U) (3 M, USA) with a capacity of 150 g.
2.8 Pre-fractionating
with Styrene
Divinyl Benzene
(SDB-RPS) [8]
1. SDB: 47 mm Styrene Divinyl Benzene (SDB-RPS) (3 M,USA).

2. Buffer 1: 0.1 % TFA.
3. Buffer 2: 0.2 % TFA.
4. Buffer 3: 100 mM ammonium formate, 40 % ACN, 0.5 % formic acid.
5. Buffer 4: 150 mM ammonium formate, 40 % ACN, 0.5 % formic acid.
6. Buffer 5: 5 % ammonium hydroxide in 80 % ACN.
7. Buffer 6: 0.1 % formic acid.
Methods

by OASIS MTP
Put the excised-gel spots/bands directly into the OASIS HLB

Elution Plate. The OASIS platform allows the use of an eightchannel pipette to add solutions, and with the help of the positive
pressure-96 stand the solutions are then simultaneously removed.
Thus, one can make a manual digest of 96 samples in the same
time and of the same quality that one could typically digest only 12
samples manually using standard 0.5 mL tubes. The in-gel digestion protocol for protein spots contains six steps: excision of spots/
bands, destaining, reducing, alkylation, digestion, and peptide
extraction.
3.1.1 Excision of Spots/

Bands from Gels
1. Use spot cutter to excise the gel plugs from the 2D-gel directly
into the OASIS MTP well. Recommended cutting tip size is
1.5 mm in diameter. Alternatively cut a band by scalpel directly
from the gel, place it at the upper rim of an OASIS MTP well
and cut it in 1 mm3 pieces. Then push them down into the well
by a pipette tip.
2. Remove the liquid from the excision process into the waste
plate (24 10 mL MTP plate) using the positive pressure-96
stand (PPS) (if not otherwise stated 15 PSI of nitrogen pressure was always used).
3. Add 70 L acetonitrile (ACN) and push the gel pieces down
into the solution if necessary.
239
4. Wait for 10 to 15 min until the gel plugs are fully dehydrated
(they will become small and white) (see Note 3).
5. Using the PPS, remove all liquid into waste plate.
3.1.2 Coomassie Gel
Destaining
1. Rehydrate gel pieces in 100 L of 100 mM ABC, pH 8.0.

2. After 10 min add an equal volume acetonitrile and wait for
1015 min until the gel pieces are destained (see Note 4).
3. Remove all liquid using the PPS and add 100 L acetonitrile
for 10 min to dehydrate the gel pieces.
4. Remove all liquid using the PPS.
3.1.3 Silver Gel

Destaining [10]
1. Rehydrate gel pieces with 50 L potassium ferricyanide and

incubate for 10 min at 37 C.
2. Then add 50 L sodium thiosulfate and incubate for 20 min at
37 C.
3. After 15 min remove the liquid.
4. Then add 50 L water and after 10 min 50 L acetonitrile.
5. Remove all the liquid after 15 min.
6. Add 70 L ABC and wait for 10 min.
7. Remove all the liquid after 15 min.
8. Add acetonitrile and wait for 1015 min (see Note 3).
9. Remove acetonitrile by PPS.
3.1.4 Reduction
1. Rehydrate gel pieces in 100 L reduction solution and incubate for 1 h at room temperature to reduce the cysteine bridges
in the protein (see Note 5).
2. Remove excess liquid using the PPS.
3. Incubate the gel pieces with 100 L ACN and wait for
1015 min until the gel pieces have dehydrated (see Note 3).
4. Remove all liquid with the PPS.
3.1.5 Alkylation
1. Swell gel pieces with 100 L alkylation solution and incubate

for 20 min at room temperature in the dark.
2. Remove alkylation solution with the PPS and wash gel pieces
with 100 L ABC for 15 min.
3. Remove ABC with the PPS, add 100 L ACN and wait 10 min
for the gel pieces to dehydrate and shrink.
3.1.6 Application
of Trypsin
1. Remove all liquid with the PPS, then rehydrate the gel pieces
in 33 L digestion solution (contains trypsin) at room
temperature.
2. After 20 min remove remaining buffer with the PPS.
240
3. Add 50 l of 50 mM ABC, but prepared without trypsin, to

cover the gel pieces and keep them wet during enzymatic
digestion.
4. Leave samples covered at room temperature overnight or if
available in an incubator at 37 C.
3.1.7 Extraction
and Cleaning/Desalting
of Peptides
1. With the PPS remove the digest solution into the waste plate
(peptides will bind to the HLB column).
2. Wash the OASIS HLB column with 70 L 0.1 % FA.
3. Exchange the waste plate for the collection plate (0.5 mL/
well) (see Note 6).
4. Add 50 L of extraction solution to each well of the Oasis plate
and wait 20 min, then remove the extraction solution into collection plate with the PPS.
5. Add 50 L of acetonitrile to each well and wait 10 min, then
remove the extraction solution into collection plate with the
PPS.
6. Completely dry eluted peptides in the collection plate using a
vacuum centrifuge at 45 C for 60 min.
7. Freeze the dried extracts (peptides) at 20 C or 80 C (for
storage >1 month).
8. Re-dissolve the peptides in 7 L 0.1 % formic acid for ESI-MS
(see Note 7).
3.2 Filter-Aided
Sample Preparation
(FASP) [6, 7]
The filter-aided sample preparation belongs to the newer methods

and is useful for starting material quantities above 50 g. Particular
attention is drawn to possible filter leakage of the membrane and
the total loss of sample during the procedure (see Note 8).
1. 1.5 mL Eppendorf tube containing 50 Drosophila melanogaster heads (about 150 g protein).
2. Heat 1 mL of lysis buffer to 95 C (2 min) and then add
100 L to the heads.
3. With an electric mortar homogenize the sample for 2 min.
4. Heat the sample for 3 min at 95 C.
5. For further homogenization a Diagenode Bioruptor plus is
used. Set 15 cycles each 30 s sonication, 30 s break, position:
high performance.
6. Centrifuge sample at 16,000 g for 10 min.
7. Add 200 L of buffer B to a Nanosep 30 k Omega Centrifugal
Device followed by 50 L (about 75 g protein) of the supernatant of the homogenized sample.
8. Centrifuge at 12,000 g for 15 min.
241
9. Collect the flow through (in case filter was leaking) and add
250 L of buffer B to the residue in the filter.
11. Add again 250 L buffer B and centrifuge at 12,000 g for
15 min.
12. Alkylation step: add 300 L of buffer C, mix with pipette
carefully and wait 20 min at room temperature in the dark
(see Note 9).
14. Add 250 L buffer B and centrifuge again (12,000 g for
15 min).
15. Add 200 L buffer D, centrifuge again and repeat this step
once more.
16. Then add 50 L buffer D to the sample followed by 3 g of
Trypsin and mix carefully by pipetting up and down.
17. Fill a plastic sample box partly (two rows) with water and the
digest solution containing closed filter devices are placed in an
empty space. With closed lid place it into an incubator at 37 C
overnight.
18. Following the digest add 200 L buffer D and centrifuge at
10,000 g for 15 min.
19. Place the flow through in a vacuum concentrator for 30 min at
45 C. Around 100 L should be left.
20. Take 2 L of the digested sample for a NanoDrop OD 260/280
measurement [11] with water as reference (see Note 10).
3.3 One-Step
In-Solution Protein
Solubilization,
and In-Solution
Digest [8]
This procedure is particular useful for shot gun proteomics of samples like a single fly head (~3 g protein) as described here.
1. Load one single Drosophila melanogaster head to the bottom
of a 0.5 mL centrifugation tube (Eppendorf tube).
2. Add 5 L of Guanidine buffer (see Note 2).
3. Heat sample for 10 min at 95 C.
4. For the homogenization of the single head use a Bandelin
Sonoplus Ultrasonic device with a 1.5 mm Tip. Apply a total of
1.0 kJ energy at 50 % amplitude, which takes about 1 min. At
the same time the head is also mechanically crushed with the
tip of the US stick which acts like a pistil.
5. The homogenate is centrifuged to the bottom of the tube with
a minitable centrifuge at 1,500 g and heated again for 10 min
at 95 C.
6. Again homogenize for 30 s at 50 % amplitude by the Bandelin
Ultrasonic device.
242
7. Centrifuge the homogenate to the bottom of the tube again

at 1,500 g.
8. Now 50 L of buffer A are added and mixed by pipette
(see Note 11).
9. Add 1 L Trypsin (stock: 1 g/L) and incubate at 37 C over
night at 800 rpm in an Eppendorf Thermomixer.
10. Next day centrifuge the tube at 15,000 g for 15 min at room
temperature.
11. Take 2 L of the supernatant and measure the peptide concentration by NanoDrop OD 260/280 as described in Subheading 3.2
and proceed with the STAGE tip procedure in Subheading 3.6.
3.4 One Step
Guanidine Method
for Large Scale
Analysis
For larger samples from 100 g to several mg of protein the onestep procedure in Subheading 3.3 needs to be modified.
1. To ten whole Drosophila melanogaster bodies (about 300 g)
add 100 L of Guanidine buffer.
2. Mechanically homogenize it with an electric mortar in a
1.5 mL centrifugation tube.
3. Boil the mixture for 5 min at 95 C in an Eppendorf Thermo
mixer at 600 rpm.
4. Homogenize further by sonication (five cycles 30 s, 30 s, position high) in a BioruptorTM plus in ice cold water.
5. The homogenate is centrifuged to the bottom of the tube with
a mini table centrifuge at 1,500 g.
6. Boil again for 5 min at 95 C in a Eppendorf Thermo mixer at
600 rpm.
7. Homogenize again with five cycles 30 s, 30 s, position high in
a BioruptorTM plus in ice cold water.
8. Dilute with 900 mL buffer A.
9. Centrifuge for 10 min at 15,000 g.
10. Place 100 L of supernatant on an Omega 30 kDa filter device.
11. Add 5 g Trypsin at a concentration of 1 g/L in 50 mM
acetic acid and incubated overnight at 37 C as described in
Subheading 3.2.
12. Next day centrifuge the tube at 15,000 g for 15 min at room
temperature.
13. Take 2 L of the flow through and measure the protein concentration with NanoDrop OD 260/280 as described and
proceed with STAGE tip procedure Subheading 3.6.
3.5 Preparation
of C. elegans Sample
1. Add 100 L of Guanidine buffer to about 30 L of wet worm

pellet (see Note 12).
2. Heat the sample in a Thermomixer at 95 C for 10 min at 600 rpm.
243
3. Homogenize with Bioruptor plus at: 30 s sonication, 30 s off,

ten cycles, position: high performance in an ice bath.
4. Heat again in the Thermomixer at 95 C for 10 min at 600 rpm.
5. Homogenize again with Bioruptor plus: 30 s sonication, 30 s
off, ten cycles, position: high performance.
7. Keep the supernatant.
8. Take 4 L of the protein solution, dilute 20 times with buffer
A to reduce the concentration of Guanidine below 0.6 M.
9. Check the concentration of the protein solution with NanoDrop.
10. Digest the diluted protein solution at 37 C overnight.
Trypsin:protein = 1:50 (see Note 13).
11. Centrifuge the digest at 15,000 g for 10 min and retain the
supernatant.
12. For STAGE tip cleaning, see Subheading 3.6.
3.6 STAGE Tip
Peptide Cleaning
and Desalting [9]
For the STAGE tip procedure a 2 mL centrifuge tube (Eppendorf

tube) with a custom-made adapter (Fig. 1) (see Note 14) to hold
the tip or alternatively a TOMY STAGE tip centrifuge (Sonation)
is used. STAGE tips can be produced with different capacities.
With three layers of 3 M EmporeTM C18-SD solid phase extraction disk and 1 mm inner diameter of the syringe for producing the
tip one has about 10 g binding capacity. For an inner diameter of
1.5 mm or 2.0 mm the binding capacities are about 30 g or 50 g,
respectively. For larger quantities of protein up to 150 g are
commercial 4 mm (1 mL) EmporeTM C18-SD solid phase extraction cartridges from 3 M available. Recently, 3 M Empore Styrene
Fig. 1 Centrifuge adapters for 200 L tips and its assembly with a 2 mL tube
244
Fig. 2 Spring-loaded syringe for quick preparation of large batches of STAGE tips
Divinyl Benzene (SDB-RPS or SDB XC) Extraction disks with

slightly different selectivities were found to be useful also for prefractionating samples [8].
The STAGE tip is produced by punching a plug out of a disk
made of three layers of the EmporeTM C18-SD solid phase
extraction disk with a syringe needle. By a wire with a little bit
smaller diameter than the needles inner diameter the plug is
removed from the backside and directly placed and tightly stuffed
into the narrow end of a 200 L pipette tip (Eppendorf). One can
also make a spring-loaded syringe that helps preparing larger
quantities of STAGE tips in short times (Fig. 2) [9].
1. Wash the tips with 200 L of solution A, centrifuge at 3,000 g
for 2 min to remove the liquid.
2. Wet the tips with 200 L of solution B, centrifuge at 3,000 g
for 2 min to remove the liquid.
3. Equilibrate the tips with 200 L of solution C, centrifuge at
3,000 g for 2 min to remove the liquid.
4. First add 100200 L of solution C to the tip, then add 10 g
sample for a 1 mm Empore plug (see Note 15), centrifuge at
3,000 g for 2 min to remove the liquid.
5. To wash salts out add 100 L of solution C and centrifuge at
3,000 g for 2 min.
6. Add 60 L of solution B to elute the peptides. Plug a 10 mL
syringe (see Note 16) with an adapter made from a BD
MicrolanceTM 3 (Figure 3) into the wide end of the tip and
push the liquid into a 0.5 mL Eppendorf tube (see Note 17).
7. Concentrate the samples with a vacuum concentrator at 30 C
for about 25 min till dryness.
8. Add 20 L 0.1 % FA and take 2 L to measure the concentration with the NanoDrop OD 260/280 against water as
reference.
245
Fig. 3 Syringe with orange adapter for the 200 L tip for elution of peptides
3.7 Pre-fractionation
of Protein Digests
with Styrene
Divinyl Benzene
(SDB-RPS) [8]
If a maximum of proteins should be identified an easy to perform

pre-fractionation can follow the STAGE tip cleaning. The SDB
membrane containing tips are made in the same way as the STAGE
tips with a 1.5 mm inner diameter syringe. Also three layers of the
SDB membrane are used.
1. The digest cleaned with STAGE Tip C18-SD
Subheading 3.6) is re-suspended with 80 l of buffer 1.
(see
2. Equilibrate the SDB tips with 200 L of buffer 2.

3. Load 80 L of the digest to the SDB tips and remove the liquid by centrifugation at 3,000 g for 2 min or use the STAGE
tip centrifuge as described above (Subheading 3.6).
4. Add 200 L buffer 2 for washing and centrifuge at 3,000 g
for 2 min.
5. Eluted the peptides with 80 L of buffer 3, buffer 4 and buffer
5 sequentially, and collect eluates separately in 0.5 mL
Eppendorf tubes.
6. Dry the three fractions in a vacuum centrifuge at 45 C for
about 25 min.
7. Re-suspend the fractions with 20 L buffer 6.
8. Check the concentration with NanoDrop.
Notes
1. 0.1 % trifluoroacetic acid in the case of MALDI preparation.
2. It is important to produce the Guanidine/TCEP/CAA/Tris
buffer always fresh, because the components react with each other
over time. First dissolve 5.73 g of GdmCl in 10 mL of Milli-Q
water, then weight the other chemicals into a separate container
and dissolve them with 3,000 L of the GdmCl solution completely and mix it with the rest of the GdmCl solution.
3. It happens that acetonitrile runs through the HLB column
without pressure. In this case just add acetonitrile again after
5 min.
246
4. If gel pieces are not destained, repeat the steps. More than
twice destaining will not improve the result.
5. The reaction can be shortened to 30 min by incubating at 37 C.
6. The PPS stand requires a certain height of the collection plate.
We usually use a stack of three 0.5 mL/96-well collection
plates, where the lower ones are just spacers.
7. For MALDI-MS use 5 L 0.1 % TFA.
8. Some filters may even leak from beginning. To test that, add
200 L buffer B to a couple of filters and centrifuge at 12,000 g
for 5 min. Compare the amount of liquid left in the filter.
Discard the filters with no or significantly less liquid. It is not
unusual that during the procedure the filter starts leaking too.
9. DTT for reduction is in the lysis buffer.
10. This is to calculate how much one needs to take for STAGE tip
peptide desalting and cleaning for MS analysis. Three layers of
membrane have a capacity of 10 g.
11. Dilute at least 1:10, otherwise trypsin digest will not work.
12. 100 L of wet worm pellet corresponds to ~1.8 mg peptides.
13. For SILAC samples with only Lysine label use LysC.
14. Also a hole in the lid of a 2 mL Eppendorf tube with the right
diameter is fine.
15. The amount is calculated from the NanoDrop OD 260/280
measurement.
16. Before you connect the syringe pull the plunger out to the
maximum volume.
17. To squeeze the extraction solution out needs some force.
Hold the tip and press it tightly against the syringe to avoid it
falling off.
References
1. Eckerskorn C, Lottspeich F (1989) Internal
amino acid sequence-analysis of proteins separated by gel-electrophoresis after tryptic digestion in polyacrylamide matrix. Chromatographia
28:9294
2. Rosenfeld J, Capdevielle J, Guillemot JC,
Ferrara P (1992) In-gel digestion of proteins
for internal sequence analysis after one- or twodimensional gel electrophoresis. Anal Biochem
203:173179
3. Franz T, Li X (2012) The OASIS HLB Elution
plate as a one-step platform for manual highthroughput in-gel digestion of proteins and
peptide desalting. Proteomics 12:24872492
4. Kerenyi L, Gallyas F (1973) ber Probleme

der quantitiven Auswertung der mit physikalischer Entwicklung versilberten Agarelektrophoretogramme. Clin Chim Acta 47:425436
5. Chevallet M, Luche S, Diemer H, Strub JM,
Van Dorsselaer A, Rabilloud T (2008) Sweet
silver: a formaldehyde free silver staining using
aldoses as developing agents, with enhanced
compatibility
with
mass
spectrometry.
6. Manza LL, Stamer SL, Ham AJ, Codreanu SG,
Liebler DC (2005) Sample preparation and
digestion for proteomic analyses using spin filters. Proteomics 5:17421745

7. Wisniewski RJ, Zougman A, Nagaraj N, Mann
M (2009) Universal sample preparation method
for proteome analysis. Nat Methods 6:359362
8. Kulak NA, Pichler G, Paron I, Nagaraj N,
Mann M (2014) Minimal, encapsulated
proteomic-sample processing applied to copynumber estimation in eukaryotic cells. Nat
Methods 11:319324
9. Rappsilber J, Ishihama Y, Mann M (2003) Stop
and go extraction tips for matrix-assisted laser
247
desorption/ionization, nanoelectrospray, and

LC/MS sample pretreatment in proteomics.
Anal Chem 75:663670
10. Li X, Franz T (2014) Up to date sample preparation of proteins for mass spectrometric analysis.
Arch Physiol Biochem 14
11. Desjardins P, Hansen JB, Allen M (2009)
Microvolume protein concentration determination using the NanoDrop 2000c spectrophotometer. J Vis Exp 33:1610
Chapter 19
Identification of Protein N-Termini Using TMPP
or Dimethyl Labeling and Mass Spectrometry
Jingjing Deng, Guoan Zhang, Fang-Ke Huang, and Thomas A. Neubert
Abstract
Determination of a proteins N-terminal sequence can be important for the characterization of protein
processing. To increase the confidence of protein N-terminal identification, chemical derivatization of the
N-terminal amine group by (N-Succinimidyloxycarbonylmethyl)tris(2,4,6-trimethoxyphenyl)phosphonium bromide (TMPP) or dimethyl labeling followed by mass spectrometric analysis is commonly performed. Using this approach, proteins can be separated by SDS-PAGE, and the protein N-terminus of
interest is labeled by TMPP or dimethyl in-gel before tryptic digestion and LC-MS analysis. The N-terminus
of a protein can thus be easily identified because the N-terminal tryptic peptides are preferentially labeled.
Peptides with N-terminal derivatization produce a better fragmentation pattern during tandem mass spectrometric analysis, which significantly facilitates sequencing of these peptides.
Key words TMPP labeling, Dimethyl labeling, N-terminal sequence analysis, Mass spectrometry
Introduction
Identification of the N-terminal sequence of an intact or cleaved
protein is crucial for its biochemical and structural characterization
[1]. In conventional shotgun proteomic analysis, it is difficult to
identify protein N-termini due to infrequent detection of protein
N-terminal peptides. Several methods have been developed based
on the N-terminal labeling to overcome this limitation, including
TMPP and dimethyl labeling [2, 3].
The TMPP labeling approach is straightforward and has been
successfully applied to different proteins [4]. Two characteristics of
this labeling reagent promote the sensitivity of the method: (1) TMPP
labeling introduces a permanent positive charge resulting in an
enhanced ionization efficiency and thus a better detection of lowabundance peptides; (2) the hydrophobic TMPP group shifts the
retention time of TMPP-derivatized peptides in reversed-phase
chromatography toward a less complex part of the chromatogram,
increasing the sensitivity of detection, especially for short N-terminal
249
250
Jingjing Deng et al.

SDS-PAGE
N
Reduction
and alkylation
In-gel digestion
C
Intensity
TMPP or
dimethyl labeling
m/z
Data analysis
LC-MS/MS
Fig. 1 Flow chart showing the process of protein separation by SDS-PAGE, N-terminal labeling, in-gel digestion, and LC-MS/MS
peptides that otherwise would not be retained on the column [3, 5].
In addition, TMPP is fully compatible with all standard detergents,
chaotropic agents, and reducing conditions used for protein extraction in proteomics [5], which makes TMPP labeling a commonly
used method for protein N-terminal sequencing.
An alternative method based on chemical labeling is dimethylation, which labels peptide N-termini and -amino groups of lysine
with water-soluble formaldehyde via reductive methylation [68].
In MS/MS analysis, this labeling strategy provides a signal enhancement for the a1 and yn 1 ions, which are not detectable from most
of the nonderivatized fragments [8]. Because of its simplicity, as
well as low cost, dimethyl labeling is another promising strategy for
protein N-termini identification.
Due to the relatively simple experimental design (classical 1D
SDS-PAGE, a single chemical derivatization step performed at the
protein level in-gel, followed by protein digestion and LC-MS/
MS, as illustrated in Fig. 1), the two chemical derivation approaches
can be readily used to analyze both purified proteins and highly
complex protein mixtures [5]. Here, N-terminal identification
using TMPP and dimethyl labeling of purified protein hemoglobin
and BSA is described.
Materials
Prepare all solutions using HPLC grade solvents unless indicated
otherwise. Prepare and store all reagents at room temperature
unless indicated otherwise. Diligently follow all waste disposal regulations when disposing waste materials.
Labeling Methods for Identifying Protein N-termini
2.1
SDS-PAGE
251
1. Laemmli sample buffer (2): 65.8 mM TrisHCl, pH 6.8,

2.1 % SDS, 26.3 % (w/v) glycerol, 0.01 % bromophenol blue
(see Note 1).
2. 415 % precast polyacrylamide gel (see Note 2). Store at 4 C.
3. SDS-PAGE running buffer: 25 mM Tris, 192 mM glycine,
0.1 % SDS. For a 10 solution, dissolve 30.0 g of Tris base,
144.0 g of glycine, and 10.0 g of SDS in 1,000 ml of H2O
(see Note 3). Store the running buffer at room temperature
and dilute to 1 with ultrapure water before use.
4. Protein gel staining solution: 45 % methanol, 10 % glacial acetic acid, 45 % water, and 3 g/l Coomassie Brilliant Blue R250.
2.2 Reduction
and Alkylation
of Excised Protein
Bands
1. TEAB buffer: 0.1 M TEAB, pH 8.0. Dilute 1 M triethyl

ammonium bicarbonate with water.
2. Acetonitile (ACN 100 %).
3. Reducing solution I: 10 mM DTT. Dissolve 0.15 g DTT in
1 ml water to make 1 M DTT solution, store in 20 C. Dilute
1 M DTT with TEAB buffer to make 10 mM DTT.
4. Alkylation solution: 55 mM solution. Dissolve 10.172 mg
iodoacetamide in 1 ml TEAB buffer (see Note 4).
5. Destaining solution: 50 % ACN in TEAB buffer.
2.3 N-Terminal
Protein Labeling
and Protein Digestion
1. N-terminal labeling solution I: 100 mM succinimidyloxycarbonylmethyl tris(2,4,6-trimethoxyphenyl)phosphonium bromide (TMPP-Ac-OSu). Dissolve 100 mg TMPP in 1.3 ml
ACN (see Note 5).
2. Reducing solution II: 260 mM cyanoborohydride. Dissolve
16.3 mg sodium cyanoborohydride in 1 ml water (see Note 6).
3. Acetate buffer: 100 mM sodium acetate, pH 5. Mix 14.8 ml
0.2 M acetic acid solution (11.5 ml acetic acid in 1 l H2O) and
35.2 ml 0.2 M sodium acetate solution (16.4 g sodium acetate
in 1 l H2O), and fill up to 100 ml with water.
4. N-terminal labeling solution II: 4 % formaldehyde. Dilute a
30 % formaldehyde solution with water.
5. HEPES buffer: 0.1 M HEPES, pH 8.2. Dissolve 23.8 mg
HEPES in 1 ml water. Adjust pH value to 8.2 with sodium
hydroxide.
2.4 Mass
Spectrometric
Analysis
1. C18 pipette tips.

2. Self-packed columns (20 cm in length): 3 m Reprosil-Pur
C18-AQ beads (Dr. Maisch HPLC GmbH), 360/75 PicoFrit
column (360 m OD 75 m ID, 10 m Tip ID).
3. Easy-nLC 1000 Nano HPLC and Q Exactive Mass Spectrometer (Thermo Fisher Scientific).
252
4. Mobile phases in HPLC: buffer A, 0.1 % FA in water; buffer B,

0.1 % FA in ACN.
5. Data analysis software: Mascot Distiller 2.5.1.
Methods
Carry out all procedures at room temperature unless otherwise
specified.
3.1
SDS-PAGE
1. Dilute the protein samples (0.1, 0.3, 1, 3, 10 pmol) with water.

Mix the diluted samples with the 2 Laemmli sample buffer
(1:1). Heat protein samples at 95 C for 5 min. Spin the tubes
for a few seconds and load protein samples to the 415 % gel
along with protein standard. Electrophorese at 15 mA until
the sample has entered the separation gel and then continue at
25 mA until the dye front has reached the bottom of the gel
(see Note 7).
2. Following electrophoresis, open the gel cassette with the lever
and gently remove the gel from the cassette and transfer to a
clean container. Rinse the gel with water for 10 min to remove
any particulate matter, and repeat twice (see Note 8).
3. Stain the gel with approximately 1520 ml Coomassie Blue
Staining solution (see Note 9).
4. Replace the staining buffer with water and wash the gel on an
orbital shaker for 3060 min until the background color is
weak.
5. Use a clean scalpel to excise the gel band of interest, cut the gel
band into small pieces and transfer the gel band into a clean
0.5 ml Eppendorf tube (see Notes 10 and 11).
3.2 Reduction
and Alkylation
of Excised Protein
Bands
1. Add 200 l 50 % ACN in TEAB buffer and agitate (600

1,000 rpm) in an Eppendorf thermomixer for 20 min. Discard
the liquid. Add 200 l TEAB and agitate (6001,000 rpm) for
20 min. Discard the liquid (see Note 12). Repeat until the gel
band is completely destained.
2. Add 50 l 100 % ACN and agitate (6001,000 rpm) for
10 min. Discard the liquid and dry the gel in a vacuum centrifuge for 5 min.
3. Add 50 l freshly prepared 10 mM DTT solution to cover the
gel pieces and agitate (6001,000 rpm) for 45 min at 56 C.
Cool to room temperature, and discard the liquid.
4. Add 100 l freshly prepared 55 mM iodoacetamide to the
tube. React for 30 min in the dark (see Note 13).
5. Spin briefly and discard the liquid.
253
6. Add 100 l 50 % ACN in TEAB buffer and agitate (600

1,000 rpm) for 10 min to wash away the reagents. Discard the
liquid. Add 100 l 100 % ACN and agitate for 10 min. Spin
briefly and discard the liquid. Repeat. Dry the gel piece in a
vacuum centrifuge for 5 min.
3.3 N-Terminal
Labeling and Protein
Digestion
1. For TMPP labeling: add 10 l TMPP solution in ACN to dry

gel piece. Then add 40 l 100 mM HEPES buffer and incubate overnight (see Note 14).
For dimethyl labeling: add 100 l 100 mM sodium acetate, 1 l 4 % formaldehyde and 1 l 260 mM sodium cyanoborohydride to the dehydrated gel pieces. Incubate the gel
pieces for 20 min while mixing using a bench top test tube
mixer (see Note 15).
2. Discard the liquid. Add 100 l water and agitate for 10 min.
Spin briefly and discard the liquid. Add 100 l ACN and agitate for 10 min. Spin briefly and discard the liquid. Repeat
washing alternately with water and ACN for 34 times until
excess labeling reagent is removed.
3. Digestion. Add freshly prepared working solution of Promega
trypsin (0.20.3 g Trypsin in 20 l TEAB buffer), incubate
on ice for 30 min. Add 40 l TEAB buffer and leave on agitator (300 rpm) overnight at 37 C.
4. Peptide extraction. Add 50 l 5 % formic acid (FA), agitate for
15 min, collect the liquid; then add 50 l acetonitrile, agitate
for 15 min, collect the liquid to the same tube.
5. Repeat step 4. Concentrate collected sample to dryness in a
vacuum centrifuge. Dissolve peptides with 10 l 0.1 % FA.
6. Desalt peptides with a C18 ZipTip (following the protocol
from Millipore), and concentrate sample to dryness in a speed
vac. Dissolve peptides in 10 l 0.1 % FA.
7. Transfer solution to an injection vial for LC-MS/MS
(see Note 16).
3.4 Mass
Spectrometric
Analysis
1. A nanoHPLC system was used for separation of the protein

digests. Buffer A and buffer B were used as mobile phases for
gradient separation. Protein digests were automatically loaded
onto a C18 reversed-phase column (column temperature
50 C). Due to the increased retention time of TMPP-labeled
N-terminal peptides [5], elution was achieved with a gradient
of 030 % B over 40 min, 3090 % B over 40 min, and 90 % B
for 10 min. For dimethyl labeling (no significant retention
time shift is caused by dimethyl labeling of peptides), elution
was achieved with a gradient of 030 % B over 80 min, 3090 %
B over 1 min, and 90 % B for 9 min.
254
2. The Q-Exactive mass spectrometer was operated in datadependent mode using a top ten method. Full MS scans were
acquired in the Orbitrap mass analyzer over a range of 300
1,650 m/z with resolution 70,000 (m/z 200). The target
value was 3.00E + 06. The ten most intense peaks with charge
state 2 were fragmented in the HCD collision cell with normalized collision energy of 27 %, and tandem mass spectra
were acquired in the Orbitrap mass analyzer with resolution
17,500 (m/z 200). The target value was 1.00E + 06. The ion
selection threshold was 1.70E + 04 counts, and the maximum
allowed ion accumulation times were 20 ms for full MS scans
and 60 ms for MS/MS [9].
3. Peak lists were created and searched by Mascot Distiller.
Software setting: semiTrypsin with one missed cleavage; carbamidomethyl (C) as fixed modification; oxidation (M) as variable modification. For TMPP labeling, a mass addition of
572.1811(C29H33O10P) (N-terminus, K and Y) must be
included as a variable modification. For dimethyl labeling, a
mass addition of 28.0313 (C2H4) (N-terminus, K, P and R) as
a variable modification must be included. Peptide tolerance
was set to 10 ppm for peptides and MS/MS tolerance was
0.03 Da (see Note 17).
4. MS/MS spectra of TMPP-labeled peptide VLSPADK
(Hemoglobin A) and TMPP-labeled peptide DTHKSEIAHR
(BSA) are shown in Fig. 2. As illustrated in Fig. 3, dimethyllabeled peptide VLSPADK or DTHKSEIAHR identified the
same N-terminal sequences of Hemoglobin A and BSA respectively. We also found dimethyl labeling to be more sensitive
than TMPP labeling (Detection limit: 0.1 pmol BSA for
dimethyl; 3 pmol BSA for TMPP).
Notes
1. Add 50 l 2-mercaptoethanol to 950 l Laemmli sample buffer. Protein sample is mixed 1:1 (v/v) with the sample buffer.
2. Gel format: 8.6 6.7 cm (W L), 10-well comb for 50 l
samples.
3. SDS is a detergent. Handle it with care and take measures to
prevent inhalation of airborne SDS. It is best to weigh SDS in
a fume hood. Check the pH of the 10 solution. The pH of
the buffer should be between 8.1 and 8.5. Do not adjust pH
with acid or base.
4. Iodoacetamide powder is stored at 2 to 8 C and protected
from light. Iodoacetamide solution is made freshly before use.
255

y4 y3 y2 y1
TMPP-Ac - V L S P A D K
a5
b1b2b3b4 b6
y1
147.11
100
b3
872.37
[M+2H]2+:
651.3027 m/z
y2
262.14
90
a5
1012.47
Relative Abundance
80
b6
1155.49
70
60
50
40
30
y4
430.23
20
b2
785.34
b1
672.26
y3
333.18
10
b4
969.43
0
100
200
300
400
500
600
700
m/z
800
900
1000
1100
1200
1300
y6 y4 y3 y2 y1
TMPP-Ac - D T H K S E I A H R
a2 a3
b1 b2 b3 b4 b6
b1
688.21
b
100
90
[M+4H]4+:
442.1996 m/z
Relative Abundance
80
y3
383.21
70
60
50
40
30
y2
312.18
y1
175.12
a2
761.26
y4
++
496.30 b4
527.71
20
10
b6++
635.75
y6
712.37
b3
926.32
b2
789.261
a3
898.33
b4
1054.41
200
300
400
500
600
m/z
700
800
900
1000
1100
Fig. 2 Detailed characterization of TMPP-labeled N-terminal peptides of Hemoglobin A and BSA. (a) MS/MS
spectrum of the doubly charged N-terminal peptide VLSPADK of Hemoglobin A. Peptide sequence is shown at
the top of the spectrum, with annotation of the matched ions. (b) MS/MS spectrum of the quadruply charged
N-terminal peptide DTHKSEIAHR of BSA. The fragmentation of the derivatized peptide produces a- and b-type
ion series in addition to y-type ions
256

y6 y5 y4 y3 y2 y1
Dimethyl - V L S P A D K
a1
100.11
a1
100
90
[M+2H]2+:
379.2248 m/z
Relative Abundance
80
70
60
50
y4
430.23
40
y6
630.34
30
20
y1
147.11
10
0
100
150
200
y5
517.26
y3
333.18
y2
226.12
250
300
350
400
450
500
550
600
650
m/z
y9
y7 y6 y5 y4 y3 y2 y1
Dimethyl - D T H K S E I A H R
b
100
a1
a1
116.07
90
[M+3H]3+:
407.8804 m/z
80
Relative Abundance
70
60
y3
383.21
50
40
y1
175.12
30
20
y2
312.18
y6
712.37
y4
496.30
y9++
539.79
10
y7
840.46
y5
625.34
0
100
200
300
400
500
600
700
800
900
1000
m/z
Fig. 3 Detailed characterization of dimethyl-labeled N-terminal peptides of Hemoglobin A and BSA. (a) MS/MS
spectrum of the doubly charged N-terminal peptide VLSPADK of Hemoglobin A. (b) MS/MS spectrum of the
triply charged N-terminal peptide DTHKSEIAHR of BSA. Unlike TMPP labeling, the dimethyl labeling strategy
enhances a1 and yn 1 ions
5. TMPP solution is made freshly to ensure high labeling

efficiency.
6. Cyanoborohydride solutions should be made freshly to ensure
high labeling efficiency [10].
257
7. Pull the green tape gently to remove it from the bottom of the
cassette and remove the comb by pulling upward in one
smooth motion to keep the lane straight.
8. Do not touch the gel with ungloved hands and always use a
new or acid-cleaned dish to avoid contamination.
9. Make sure the staining buffer covers the gel completely.
10. Be sure to use extremely clean surfaces and new scalpels. This
should be done in a flow hood to minimize the possibility of
contamination by dust, hair, flakes of skin, or other forms of dirt.
11. Use polypropylene tubes and low-retention tips to minimize
protein loss by adsorption to tube walls.
12. An increase of temperature (e.g. to 50 C) promotes the
destaining process.
13. Iodoacetamide is unstable and light-sensitive. Prepare iodoacetamide solutions immediately before use and perform alkylation in the dark. Excess reaction time will cause other
functional groups to be labeled.
14. For TMPP labeling, do not use ammonium bicarbonate or Tris
buffer that can decrease the labeling efficiency.
15. For dimethyl labeling, during sample preparation, no buffers and
solutions containing primary amines (such as ammonium bicarbonate and Tris) should be used that can decrease the labeling
efficiency, as formaldehyde can react with these reagents [10].
16. Before loading peptide samples for LC-MS/MS, avoid sample
contamination by keratin, polymers, and detergents. Try to work
as cleanly as possible, especially before tryptic digestion, because
contamination with other proteins can prevent identification of
the protein of interest. The most frequent contaminants are BSA
and human keratin. The keratin comes from dust, small hairs,
and fingerprints. Even a small hair contains overwhelming
amounts of keratin compared to the amount of your sample.
17. Selective N-terminal TMPP derivatization can be achieved by
keeping the solution at pH 8.2, exploiting the weaker basicity
of the N-terminal amine relative to the -amino group of the
lysine side chain [5]. In order to minimize derivatization of
tyrosine residues, 0.1 M hydroxylamine can be added to the
solution to quench the derivatizing reagent 1 h after TMPP
labeling reaction begins [5, 11].
Acknowledgement
This work was supported by NIH Shared Instrumentation Grant
S10RR027990 and NINDS grant P30 NS050276 to T.A.N. We
thank the ABRF Protein Sequencing Research Group (2013) for
helpful protocols, reagents and advice.
258
References
1. Speicher KD, Gorman N, Speicher DW (2009)
N-terminal sequence analysis of proteins and
peptides. Curr Protoc Protein Sci. 57:11.10:
11.10.111.10.31.
2. Gallien S, Perrodou E, Carapito C et al (2009)
Ortho-proteogenomics: multiple proteomes
investigation through orthology and a new
MS-based protocol. Genome Res 19:128135
3. Armengaud J (2009) A perfect genome annotation is within reach with the proteomics and
genomics alliance. Curr Opin Microbiol 12:
292300
4. Huang ZH, Wu J, Roth KD et al (1997) A
picomole-scale method for charge derivatization of peptides for sequence analysis by mass
spectrometry. Anal Chem 69:137144
5. Bertaccini D, Vaca S, Carapito C et al (2013)
An improved stable isotope N-terminal labeling approach with light/heavy TMPP to
automate proteogenomics data validation:
dN-TOP. J Proteome Res 12:30633070
6. Hsu JL, Huang SY, Shiea JT et al (2005)
Beyond quantitative proteomics: signal
enhancement of the a1 ion as a mass tag for
7.
8.
9.
10.
11.
peptide sequencing using dimethyl labeling.

Hsu JL, Huang SY, Chen SH (2006) Dimethyl
multiplexed labeling combined with microcolumn separation and MS analysis for time
course study in proteomics. Electrophoresis
27:36523660
Hsu JL, Huang SY, Chow NH et al (2003)
Stable-isotope dimethyl labeling for quantitative proteomics. Anal Chem 75:68436852
Sun L, Zhu G, Dovichi NJ (2013) Comparison
of the LTQ-Orbitrap Velos and the Q-Exactive
for proteomic analysis of 11000 ng RAW
264.7 cell lysate digests. Rapid Commun Mass
Spectrom 27:157162
Boersema PJ, Raijmakers R, Lemeer S et al
(2009) Multiplex peptide stable isotope
dimethyl labeling for quantitative proteomics.
Nat Protoc 4:484494
Boersema PJ, Aye TT, van Veen TA et al (2008)
Triplex protein quantification based on stable
isotope labeling by peptide dimethylation
applied to cell and tissue lysates. Proteomics
8:46244632
Chapter 20
Optimization ofCell Lysis andProtein Digestion
Protocols forProtein Analysis by LC-MS/MS
DominicWinter, AlirezaDehghani, andHannoSteen
Abstract
Mass spectrometry-based proteomics is the method of choice for analyzing (sub-) proteomes from virtually
any source; with cells grown in culture being the most frequently investigated samples. Using HeLa cells,
we describe a strategy for the optimization of protocols for whole proteome analysis. We cover cell lysis,
protein precipitation and digestion, comparing the results obtained using different parameters and offering
various possibilities for the optimization of a proteomic workflow which can be used for virtually any type
of animal cell grown in culture.
Key words Mass spectrometry, Cell lysis, Protein digestion, Proteomics, Method optimization
1 Introduction
Mass spectrometry-based proteomics has become the major tool
for the identification, characterization, and quantification of proteins. The commonly used strategy, the so-called bottom up
strategy [1], is based on the proteolytic cleavage of the proteins
to be analyzed followed by analysis of the resulting peptides by
mass spectrometry. The extraction of proteins from sample material and the proteolytic digestion are strongly influencing the
number of proteins identified and hence the performance of the
analysis [2, 3]. Dependent on the type of sample analyzed and
the person performing the experiment, buffers, protocols and
techniques vary strongly (e.g. [47]) making it difficult for the
inexperienced operator to choose a particular method. It is often
advisable, if a sufficient amount of material is available, to test
different conditions for sample preparation in order to determine
the optimal parameters for each sample. To achieve optimal
results, several steps of the protocol can be optimized. These
include: harvesting and lysis of cells, extraction and subsequent
proteolytic digestion of proteins and, if in-gel digestions are carried
259
260
Dominic Winter et al.
out, extraction of peptides from the gel pieces. To this end, we

optimized a protocol for the processing of HeLa S3 cells, the
arguably most commonly used cell line. We tested three different
buffers in combination with two methods for mechanical disruption for cell lysis and two methods for precipitation of proteins.
Using the optimal parameters from these experiments, we optimized the extraction of peptides from the in-gel-digested samples
in order to allow for an efficient recovery of peptides from the gel
pieces. Finally, we generated whole proteome samples which were
digested using nine different strategies: six employing in-solution
digestion and three in-gel digestion protocols.
2 Materials
Prepare all solutions (except of PBS and trypsin/EDTA solution)
using HPLC grade water.
2.1 Cell Harvesting
1. Phosphate Buffered Saline (PBS, 1).

2. Trypsin/EDTA solution: 0.25% trypsin, 1mM EDTA in
Dulbeccos modified Eagle medium (DMEM).
3. Liquid nitrogen or dry ice.
2.2 Cell Lysis
1. RIPA buffer: 25mM TrisHCl, pH8, 150mM NaCl, 1%

NP-40, 1% SDS, 1mM EGTA, 1 Protease Inhibitor Cocktail
(see Notes 13).
2. GlyNP40 buffer: 50mM HEPES, pH7.6, 150mM KCl,
1mM MgCl2, 10% glycerol, 0.5% NP-40, 1mM EGTA, 1
Protease Inhibitor Cocktail (see Notes 13).
3. GCL (Guanidine HCl-based) buffer: 6M Guanidine-HCL,
0.1M TEAB (Tetraethylammonium bicarbonate), 1%
Triton X-100, 1mM EGTA, 1 Protease Inhibitor Cocktail
(see Notes 13).
4. Single use 3ml plastic syringes and 26.5G needles.
5. Probe sonicator.
2.3 Protein
Precipitation
1. TCA-solution: 100% (w/v) Trichloroacetic acid, for preparing

5ml of 100% TCA solution add 5g of TCA to 2.27ml of
HPLC-grade water (see Note 4).
2. Acetone.
3. Chloroform/Methanol solution: mix two parts Chloroform
and one part Methanol (see Note 5).
4. 1ml glass syringe.
5. Methanol.
Optimization of MS Sample Preparation

andPeptide Extraction
261
1. LDS Sample Buffer (4).

2. Reduction solution: 1M Dithiothreitol (DTT) (see Note 6).
3. Alkylation solution 1: 40% Acrylamide (see Note 7).
4. SDS-PAGE gel.
5. Coomassie blue.
6. Scalpel.
7. TEAB solution: 100mM TEAB.
8. Gel destaining solution: 30% ACN (acetonitrile), 70mM TEAB.
9. Trypsin solution (0.1g/l): Resuspend trypsin using HPLC
water on ice (in order to minimize auto-proteolysis) and vortex the vial to resuspend the enzyme (see Note 8).
10. Peptide extraction solutions (see Note 4).
(a) Organic extraction solution 1 (OES1): 1% AA (acetic
acid), 50% ACN.
(b) Organic extraction solution 2 (OES2): 1% FA (formic
acid), 50% ACN.
(c) Organic extraction solution 3 (OES3): 1% AA, 1% FA,
50% ACN.
(d) Organic extraction solution 4 (OES4): 0.1% TFA (trifluoroacetic acid), 50% ACN.
(e) Organic extraction solution 5 (OES5): 0.1% HFBA (heptafluorobutyric acid), 50% ACN.
(f) Organic extraction solution 6 (OES6): 0.1M TEAB, 50%
ACN.
(g) Organic extraction solution 7 (OES7): 1% FA in ACN.
(h) Aqueous extraction solution 1 (AES1): 1% AA.
(i) Aqueous extraction solution 2 (AES2): 1% FA.
(j) Aqueous extraction solution 3 (AES3): 1% AA, 1% FA.
(k) Aqueous extraction solution 4 (AES4): 0.1% TFA.
(l) Aqueous extraction solution 5 (AES5): 0.1% HFBA.
(m) Aqueous extraction solution 6 (AES6): 0.1M TEAB.
(n) Peptide resuspension solution: 5% ACN, 5% FA.
2.5 Digestion
ofWhole Cell Lysates
1. TEAB solution (TS): 100mM TEAB.

2. Urea solution (US): 8M urea in 100mM TEAB (see Note 9).
3. RapiGest solution (RS): 0.1% RapiGest in 100mM TEAB.
4. RapiGest/Urea solution (RUS): Mix two parts of RapiGest
solution with one part of Urea solution.
5. ProteaseMax solution (PS): 0.2% ProteaseMax in 100mM
TEAB.
262
6. ProteaseMax/Urea solution (PUS): Mix 2 parts of ProteaseMax

solution with 1.5 parts of Urea solution.
7. ACN-TEAB solution (ATS): 80% ACN, 20% 100mM TEAB.
8. Reduction solution: 1M Dithiothreitol (DTT) (see Note 6).
9. Alkylation solution 2: 1M Iodoacetamide (IAA) (see Note 10).
10. ProteaseMax in-gel solution (PGS): 0.025% ProteaseMax in
100mM TEAB.
11. Trypsin solution (see Subheading2.4).
12. Coomassie Blue.
13. Scalpel.
14. Gel destaining solution: 30% ACN, 70mM TEAB.
15. Peptide extraction solutions (see Subheading2.4).
3 Methods
It is very important to avoid introduction of keratin and other
contaminants such as polyethylene glycol (PEG) during the experimental steps. The keratin proteins may lead to abundant peptide
signals occluding peptides of the sample proteins to be identified,
while other contaminants with surfactant-like properties can significantly affect the ionization of the analytes of interest. Therefore
new plastic ware, pre-stacked pipet tips, LC-MS grade solvents and
high purity chemicals should be used. A laboratory coat and fresh
gloves should be worn and they should be kept clean. Frequent
changes of gloves further reduce the risk of contamination. It
should especially be avoided to bring any surface in contact with
potentially contaminated surfaces, skin, clothing and/or hair that
is, or will be, in direct contact with the sample. In order to minimize sample loss, maximum recovery/low binding 1.5ml tubes
are recommended for all steps. For the optimization of a proteomic
workflow, several parts of the protocol can be altered. An overview
for a possible strategy is depicted in Fig.1.
3.1 Cell Harvesting
Cells should always be washed with PBS or similar solutions (e.g.

Hanks Balanced Salt Solution, HBSS) before being harvested in
order to remove proteins and other contaminants found in the
culture medium.
1. Aspirate the cell culture medium and wash the cells with 10ml
PBS.
2. Add trypsin/EDTA solution (1ml for 10cm plates, 2.5ml for
15cm plates) (see Note 11) to the cells and incubate for 5min.
Subsequently add 10ml of ice cold 1 PBS to the plate, detach
cells by trituration and transfer the suspension into a 15ml
tube. Centrifuge at 4C for 5min at 500g, aspirate the PBS
263
a
HELA S3 Cells
Lysis Buffers:
RIPA
Mechanical Cell disrupon:
Protein Precipitaon:
GlyNP40
GCl
Syringe
Sonicator
C/M
TCA
b
Samples prepared with Opmized Lysis and Extracon Procedure
In Soluon
Digeson
Urea
RapiGest
RapiGest+Urea
ProteaseMax
ProteaseMax+Urea
Acetonitrile
In Gel Digeson
No Additives
RapiGest
ProteaseMax
In Gel Digeson
Pepde
Extracon
FA / AA
TFA/HFBA
ACN
Fig. 1 Experimental design for the workflow optimization of the proteomic analysis of HeLa S3 cells. The optimization steps include (a) cell lysis buffers, mechanical cell disruption and protein precipitation as well as (b) protein digestion in
solution and in SDS PAGE gels and peptide extraction from in-gel digests (reproduced from [3] with permission from Wiley)
without disturbing the pellet and wash cells once more with
10ml ice cold PBS followed by centrifugation at 4C for
5min at 500g. If the samples are not to be processed directly,
flash-freeze cells using liquid nitrogen or dry ice and store
at 80C.
3.2 Cell Lysis
The following steps should be carried out on ice.

1. Add the lysis buffer (RIPA buffer, GlyNP40 buffer or GCL
buffer) to the cell pellet with a pellet volume to buffer volume
ratio of 1:10 and transfer the sample to a 1.5ml tube.
(a) For syringe lysis, the cell suspension has to be passed five
times through a 26.5G (or higher) needle in order to disrupt cell walls and shear the DNA.It is important that the
needles chosen are long enough to reach the bottom of
264
the 1.5ml tube and that no air is entering the syringe

while lyzing the cells, as otherwise foaming can occur
resulting in reduced lysis efficiency and sample loss. A
fresh syringe and needle should be used for each sample.
(b) For cell lysis using a sonicator, the tip of the sonicator has
to be submerged in the cell suspension and a sufficient
amount of energy has to be introduced to lyze the cells
and shear the DNA.For small sample volumes of up to
500l usually a low amplitude is sufficient. If a microtip is
used (tip diameter of 1mm), it is possible to increase the
amplitude up to 90%, dependent on the sonicator.
Formation of aerosol and/or foam should be avoided by
adjusting the amplitude accordingly as otherwise sample
loss can occur. Sonicate the samples for 30s and put them
on ice for a minimum of 30s after each cycle to prevent
overheating of the samples. Perform at least three cycles of
sonication and incubation on ice (see Note 12).
2. Centrifuge the cell lysate at 4C for 30min at 20,000g.
Transfer the clear supernatant to a new 1.5ml tube, avoid the
pellet (cell debris) at the bottom of the tube and (if present)
the fluffy white layer on top of the liquid (lipids). The cell
lysate can be stored at 80C for months to years.
3. At this point a protein assay can be performed (e.g. Lowry,
BCA or Bradford assay [8]) to determine the protein content
of the lysate. Sample dilutions of 1:10 and 1:20 are usually sufficient to generate values which are in the linear range of the
assay using a standard BSA dilution series (0.0251mg/ml)
(see Note 13).
Based on our experience, the efficiency of the cell lysis with
sonication is greatly dependent on the tip diameter of the sonicator, the energy employed for lysis, the sample volume and the buffer composition. For the results presented in Fig.2 a tip with a
diameter of 5mm was used. Using RIPA or NP40 buffer resulted
in foam/aerosol generation and concomitant sample loss. Only the
GCL buffer, most likely because of its higher viscosity due to the
high salt content, did not lead to sample loss. In contrast to the
sonication-based lysis, the syringe-based lysis was more effective
due to smaller losses.
We also evaluated the use of a sonication tips with a smaller
diameter (data not shown). Using such microtip resulted in even
better lysis than the syringe-based lysis as judged by the size of the
resulting pellet after centrifugation. When compared to the use of
the 5mm sonication tips, the losses were minimal when using the
sonication microtip. Thus, we advise to use, if available, a microtip
or otherwise the syringe-based cell lysis.

RIPA
GlyNP40
GCl
RIPA
TCA C/M
TCA C/M
TCA C/M
TCA C/M
Syringe, 26.5 G needle

5 passages
GlyNP40
TCA C/M
265
GCl
TCA C/M
Sonicator, 40% amplitude

5 x 30s
Fig. 2 Coomassie stained SDS-PAGE gels of HeLa S3 protein extracts obtained using combinations of different
lysis buffers (RIPA buffer, GlyNP40 buffer, GCL buffer), lysis methods (syringe and sonicator), and precipitation
methods (TCA and C/M). TCA Trichloroacetic acid precipitation, C/M: chloroform/methanol precipitation (reproduced from [3] with permission from Wiley)
3.3 Protein
Precipitation
It is usually beneficial to purify proteins from the cell lysate as the

detergents and salts of the lysis buffer can interfere with the subsequent proteomics workflow. If SDS-PAGE is to be performed, this
step can also be omitted.
1. Protein Precipitation: Use the clear supernatant of the cell
lysate for protein precipitation with one of the following
methods:
(a) TCA Precipitation [9]: Add TCA solution to the sample
to a final concentration of 12.5% (v/v) and vortex for
10s. Incubate the sample on ice for 30min and centrifuge
at 4C for 30min at 20,000g. Remove the supernatant
without disturbing the pellet and discard the liquid. Add
1ml ice cold acetone to wash the pellet. Centrifuge at
4C for 30min at 20,000g and repeat the washing step.
Air-dry the pellet at room temperature.
(b) Chloroform/Methanol Precipitation [10]: Add ice cold
chloroform/methanol solution to the sample with a ratio
5:1 (five parts cold chloroform/methanol solution to one
part sample). Vortex for 20s and centrifuge at 4C for
30min at 20,000g. After centrifugation, three phases
are visible: the lower phase is the organic chloroform phase
and the upper phase is the aqueous methanol phase. The
proteins form a solid third phase at the interface of the two
liquid phases. The liquid phases have to be removed for
266
which a glass syringe works best. When removing the liquid

phases, great care should be taken not to touch the protein disk with the tip of the syringe since it tends to stick
to the tip and may be lost. The lower (chloroform) phase
should be removed first. Usually, it works best if the 1.5ml
tube is held at an angle of approximately 45; and one follows the wall of the 1.5ml tube with the needle of the
glass syringe, moving past the protein disk at the upper
side of the tube, slightly bending the needle. While removing the chloroform phase, the protein disk usually attaches
to the wall of the 1.5ml tube making it easy to remove
most of the remaining liquid. Add 1ml of ice cold methanol to wash the disk comprising the precipitated proteins.
Centrifuge the samples at 4C for 30min at 20,000g
and remove the supernatant. Air-dry the pellet at room
TCA protein precipitation was compatible with all three buffers
tested. In contrast, C/M precipitation of samples containing GCL
buffer was not possible, as addition of chloroform/methanol to
the GCL buffer resulted in the precipitation of large amounts of
guanidinium chloride. For protein extraction, the combination of
the GlyNP40 or RIPA buffer with syringe lysis and C/M precipitation delivered the best results (see Fig.2). We decided to continue
with the combination of GlyNP40 buffer with syringe lysis and
C/M precipitation, as the SDS contained in the RIPA buffer may
not have been completely removed during the precipitation step.
For the evaluation of the efficacy of the cell lysis and/or protein
precipitation, SDS-PAGE followed by Coomassie or silver staining
can be used (for instructions see Subheading3.4, step 14).
If samples are supposed to be used for whole proteome analysis, the
protein pellet can also be used directly for in-solution digestion.
For the subsequent mass spectrometric analysis of the generated
cell lysates, the proteins have to be proteolytically digested. There
are two major approaches: in-solution digestion [4] and in-gel
digestion [11]. In case of in-gel digestion, generated peptides have
to be extracted from the gel pieces for which several protocols exist.
In order to identify the most efficient way of extracting peptides, we
compared different protocols using cell lysates generated with the
most efficient method for cell lysis and protein precipitation.
andPeptide Extraction
1. Prepare aliquots of 25100g of protein for each sample. Add

4 LDS sample buffer to each sample to a final concentration
of 1 and incubate them at 95C for 10min in order to denature proteins (see Note 15).
2. Reduce disulfide bonds by adding reduction solution to a final
concentration of 5mM DTT and incubation at 56C for
45min at 800rpm in a Thermomixer (see Note 15).
267
3. Alkylate thiol groups by adding alkylation solution 1 to a final concentration of 1% acrylamide and incubating for 30min at room
temperature. Perform SDS-PAGE electrophoresis (see Notes 16
and 17).
4. Stain the gel using Coomassie blue staining solution. The
staining and destaining steps should each be performed for at
least 2h, preferably overnight (see Notes 18 and 19).
5. For optimization of peptide extraction, choose one region of
the sample lane (e.g. the region between 50 and 80kDa) and
excise the same region from all lanes using a clean scalpel.
Move the big gel piece to a clean surface (e.g. a single use plastic Petri dish), cut the gel bands into small pieces (~1mm3
cubes) and put them in 1.5ml tubes (see Note 20).
6. Add 500l of gel destaining solution and incubate at 25C for
30min at 800rpm in a Thermomixer. If the gel pieces are not
all detached from the bottom of the 1.5ml tube increase the
rpm value or the volume of destaining solution.
7. Discard the liquid and repeat step 6 as long as it takes to
remove the Coomassie from the gel pieces (usually 12 repeats
are sufficient).
8. Discard the liquid, add 500l of 100% ACN and incubate the
samples at 25C for 15min at 800rpm in a Thermomixer.
The gel pieces should shrink and turn white during this step.
9. Discard the liquid and vacuum centrifuge the gel pieces until
they are completely dry. The lids of the 1.5ml tubes have to be
open during vacuum centrifugation.
10. Add 10l of a 1:1 mixture of trypsin solution and 0.1M TEAB
to each sample tube (0.5g trypsin per sample) and incubate
until the gel pieces have taken up all the liquid.
11. Add 50l of 0.1M TEAB and incubate for 30min at 37C.If
the samples are not completely covered by liquid, add as much
0.1M TEAB as needed to cover the samples completely.
12. Incubate the samples overnight at 37C.
13. Transfer the supernatant to a new 1.5ml tube. Do not discard
this solution as the liquid supernatant will contain most of the
peptides to be analyzed.
14. Choose one combination of extraction buffers for each sample
(see Table1). Add 100l of solution A to the 1.5ml tubes containing the gel pieces and incubate at 800rpm, 25C for 15min
in a Thermomixer and transfer the supernatant to the 1.5ml
tubes from step 13. Repeat this step using solutions B and C and
combine all extracts for each individual sample. If 100l are not
sufficient to cover all gel pieces during peptide extraction increase
the volume accordingly until all gel pieces are covered.
15. Vacuum centrifuge the 1.5ml tubes with the pooled supernatants until dryness.
268
Table 1
Combinations of solutions for extracting peptides from gel pieces
Solution A
Solution B
Solution C
Combination 1
OES1
AES 1
100% ACN
Combination 2
OES 2
AES 2
100% ACN
Combination 3
OES 3
AES 3
100% ACN
Combination 4
OES 4
AES 4
100% ACN
Combination 5
OES 5
AES 5
100% ACN
Combination 6
OES 6
AES 6
100% ACN
Combination 7
AES 2
OES 2
OES 7
Combination 8
OES 4
AES 6
100% ACN
Combination 9
OES 1
AES 6
100% ACN
Combination 10
OES 2
AES 6
100% ACN
OES organic extraction solution, AES aqueous extraction solution, ACN acetonitrile
16. Resuspend the samples by adding 20l peptide resuspension

solution.
17. Proceed to MS analysis (see Note 21).
In our hands, the best results for the extraction of peptides
from gel pieces were obtained with combination 8 (see Fig.3a).
3.5 Digestion
ofWhole Cell Lysates
3.5.1 In-Solution
Digestion
Whole cell lysates can either be digested in-solution or in-gel. For

performing in-gel digestion, the sample should be allowed to electrophorese ~1cm into the gel in order to keep the gel volume in a
range which allows handling in a single 1.5ml tube.
1. Prepare protein pellets containing 100g of protein.
2. Resuspend each protein pellet in one of the following solutions: 30l of US, 30l of RS, 45l of RUS, 20l of PS, 35l
of PUS or 60l of ATS.
3. Incubate the samples for 45min at 37C at 800rpm in a
Thermomixer for solubilization of the pellet.
4. Add reduction solution (final concentration 5mM DTT) for
reducing disulfide bonds and incubate for 1h at 56C.
5. Add alkylation solution 2 (final concentration 15mM IAA) to
each sample for alkylation of thiol groups. Incubate the samples for 45min at room temperature in dark.
6. Add 1g of trypsin to each sample (enzyme to protein ratio is
1:100) and adjust the sample volume to 100l using 100mM
TEAB (see Note 22).
269
a
buffer 1
1% AA 50% ACN, 1% AA, 100% ACN
buffer 2
1% FA 50% ACN, 1% FA, 100% ACN
buffer 3
1% AA and 1% FA 50% ACN, 1% AA and 1% FA, 100% ACN
buffer 4
0.1% TFA 50% ACN, 0.1% TFA, 100% ACN,
buffer 5
0.1% HFBA 50% ACN, 0.1% HFBA, 100% ACN,
124
128
buffer 8
0.1% TFA 50% ACN, 0.1M TEAB, 100% ACN

1% AA 50% ACN, 0.1M TEAB, 100% ACN
819
549
607
131
200
solution urea
400
600
800
2749
565
TM
solution urea and RapiGest
2156
444
solution ProteaseMaxTM
2604
524
solution urea and ProteaseMaxTM
2283
470
solution 80% ACN
proteins
2315
360
2716
306
PreteaseMax
2338
370
peptides
794
198
TM
1000
1782
385
solution RapiGestTM
gel
614
111
RapiGestTM
637
160
buffer 10 1% FA 50% ACN, 0.1M TEAB, 100% ACN
gel
776
141
1% FA, 1% FA 50% ACN, 1% formic acid 99% ACN,
no additives
674
139
0.1M TEAB 50% ACN, 0.1M TEAB, 100% ACN,
gel
protiens
569
127
buffer 7
peptides
593
125
buffer 6
buffer 9
602
125
500
1000
1500
2000
2500
3000
Fig. 3 Comparison of different buffers for protein digestion (a) in the gel and (b) in solution as well as for the
extraction of peptides from in-gel digests (reproduced from [3] with permission from Wiley)
7. Incubate at 37C overnight without shaking.

8. RapiGest has to be removed from the respective samples.
Therefore add 5l 100% TFA (final concentration 5%) to the
sample containing RapiGest [12], incubate for 45min at 37C
at 800rpm in a Thermomixer, centrifuge for 30min at 20,000g
and transfer the supernatants to new tubes (see Note 23).
9. Proceed to MS analysis.
3.5.2 In-Gel Digestion
1. Prepare protein pellets of 100g of protein, add 1 LDS buffer to the samples and incubate 10min at 95C.
2. Reduce disulfide bonds by adding DTT to a final concentration of 5mM and incubation for 1h at 56C.
3. Alkylate thiol groups by adding alkylation solution 2 to a final
concentration of 15mM IAA and incubation for 45min at
room temperature in the dark.
4. Perform the electrophoresis and let the sample migrate 1cm
into the gel. Optional: Stain the gel using e.g. Coomassie.
270
5. Excise the entire gel area containing the proteins, cut the gel to
pieces and prepare the gel pieces as described in Subheading3.3.
6. Add 1g of trypsin in a sufficient amount of solution (either
TS, RS or PGS) to cover all gel pieces.
7. Incubate the samples overnight at 37C without shaking.
8. Transfer the supernatants to new 1.5ml tubes and perform
peptide extraction following the optimized protocol obtained
from the experiments in Subheading3.3.
9. In order be able to compare the samples to the in-solution
digestion protocols, adjust the sample volume to similar values
as obtained in the in-solution digestion by use of a vacuum
centrifuge. Reduce the ACN content below 5% as this will
otherwise interfere with the binding of the peptides to the
reversed phase material.
10. For the sample containing RapiGest, reduce the sample volume to approximately 100l (take care not to dry down the
sample completely) and add 5l of 100% TFA (final concentration 5%), incubate at 37C at 800rpm for 45min in a
Thermomixer, centrifuge at 20,000g for 30min and transfer
the supernatant to a new tube.
11. Proceed to MS analysis (see Note 21).
In our hands, the highest number of peptide and protein identifications were achieved by in-solution digestion in the presence of
RapiGest (see Fig.3b).
4 Notes
1. One tablet of Protease Inhibitor Cocktail is sufficient for 10 or
50ml of buffer. A 100 stock can be prepared by adding 1
tablet to 100l or 500l HPLC-grade water, respectively, and
kept at -20C for up to 1 year.
2. NP-40 and Triton X-100 are highly viscous detergents. In
order to pipet correct volumes, they should be handled at
room temperature or higher temperatures, pipet tips should be
cut to generate a larger opening and pipetting should be done
slowly. After adding the viscous detergents the solution should
not be vortexed to avoid formation of foam.
3. HEPES, TRIS/HCl, EGTA, KCl, MgCl2, Guanidine-HCl,
NaCl, and SDS can be prepared as stock solutions and kept at
room temperature for several months. TEAB usually ships as
1M stock solution and should be stored at 4C.
4. Highly concentrated acids (TCA, TFA, HFBA, AA, and FA)
cause severe skin burns and eye damage. They should be
handled in the fume hood, safety glasses and gloves should
be worn at all time.
271
5. Chloroform and methanol are toxic and cause irritation to

skin. The stock bottles should be opened and the solution prepared in the fume hood. The C/M solution can be stored at
room temperature in appropriate gas tight glass bottles.
6. In order to prepare a 1M stock solution of DTT, dissolve
0.155g of DTT powder in HPLC-grade water and bring the
volume to 1ml. The solution can be stored at 20C for up
to 1 year.
7. For preparing a 40% acrylamide solution dissolve 0.4g of
acrylamide in HPLC-grade water and bring the volume to
1ml. This solution can be stored at 20C for up to 1 year.
Acrylamide is toxic as powder and in solution; the powder
should be handled in a fume hood.
8. It is advisable to divide the solution in aliquots of 1g each
(10l) in order to prevent multiple freeze/thaw cycles. The
trypsin solution can be stored at 20C for up to 1 year. If no
ice is available, it is advisable to resuspend the trypsin in 0.1%
acetic or formic acid or in solutions provided by some manufacturers for prevention of auto-proteolysis.
9. For making 1ml of a 8M urea solution use 479.8mg of urea
and add 635.3l of 0.1M TEAB.The urea solution should
always be prepared fresh in order to minimize artificial
carbamylation.
10. Dissolve 0.184g of iodoacetamide powder in HPLC-grade
water and bring the volume to 1ml to prepare a 1M iodoacetamide stock solution. This solution can be kept at 20C for
up to 1 year. Exposure of iodoacetamide to light should be
limited as much as possible in order to minimize light-induced
decomposition.
11. The use of trypsin for harvesting of cells leads to a loss of the
extracellular domains of cell surface proteins. If these domains
are to be covered in the analysis, cells can also be harvested
using enzyme-free solutions (e.g. Cellstripper by Cellgro) or
by placing the culture dish on ice and directly adding a small
volume of lysis buffer (13ml) followed by scraping of the
plate with a cell scraper. As it is not possible to wash the cells
after scraping them, it is advisable to wash them at least two
times with 10ml of ice cold PBS on the plate before adding
the lysis buffer.
12. For optimal results, the tip of the sonicator should be kept in
the center of the 1.5ml tube and 13mm above the bottom.
Clean the sonicator tip, e.g. by using 70% ethanol and precision wipes (Kimtech Science), before using it for the next sample to minimize carry over. The tip of the sonicator should not
be touched while the sonicator is running to avoid injuries.
272
13. The compatibility of the protein assay with the components of

the lysis buffer has to be verified before performing the assay.
14. It is not absolutely necessary to remove all of the liquid at this
step if doing so may result in contact of the syringe needle with
the protein disk as this may result in its loss. If the protein disk
is still sticking to the wall of the 1.5ml tube after adding the
methanol, vortex the sample vigorously until the disk detaches.
15. Reduction of proteins can also be done simultaneously with
protein denaturation by adding DTT to the LDS sample buffer
before incubating at 95C.The acrylamide can be added
directly after the sample cooled down to room temperature. If
it is not intended to perform any mass spectrometry experiments with the SDS-PAGE gel, the alkylation step can be
omitted. It is also possible to reduce and alkylate the samples
after running the gel as part of the digestion protocol.
16. Precast SDS-PAGE gels decrease the risk of contaminations
but are not a necessity. If the SDS-PAGE gels are self-made,
special care should be taken to work as clean as possible to
avoid contaminations by keratins.
17. Gels or gel pieces can be stored at 4C in water for several
weeks. If longer storage is planned, 0.020.05% sodium azide
should be added to prevent bacterial or fungal growth.
18. It is possible to use other staining methods as, e.g. silver staining if these are compatible with the mass spectrometric
analysis.
19. If multiple lanes are to be cut in the same way (e.g. five conditions) make the horizontal cuts for all lanes first, this makes it
easier to make reproducible cuts between different samples.
Then make vertical cuts and transfer gel slices to the lid of the
plastic dish for cutting it to cubes. Do not macerate the gel as
small gel pieces can easily be transferred into, e.g. the autosampler vial resulting in clogged transfer lines.
20. For removing buffers it is practical to be able to pipet a big
volume while having a narrow pipet tip in order to prevent the
loss of small gel pieces. This can be achieved by combining a
small (10l) pipet tip with a big (1ml) pipet tip. The same
pipet tip can be used for removing the destaining solution
from the gel pieces of different samples before adding the protease. After proteolytic digestion, separate tips have to be used.
21. If salt-sensitive analysis approaches are planned (e.g. MALDI)
samples have to be desalted by using e.g. STAGE tips [13].
22. TEAB was used in these experiments as it is compatible with
primary amine labeling protocols (as employed in TMT, iTRAQ
or dimethylation [14]), however, several common buffers like
273
(NH4HCO3, tris (hydroxymethyl) aminomethane (TRIS) or

4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES))
will work as well as long as the pH is correct. Alternatively to
DTT, e.g. tris (2-carboxyethyl) phosphine (TCEP) can be used
for reduction of disulfide bonds and acrylamide or chloracetamide can be used instead of iodoacetamide for alkylation of
thiol groups.
23. When adding TFA to the RapiGest containing samples, it can
happen that gas bubbles are forming. The RapiGest pellet
which is present after centrifugation is very loose, great care
should be taken while recovering the supernatant.
Acknowledgements
This work was supported by the German Academic Exchange
Service (DAAD) and the 7th Framework Program of the European
Union.
References
1. Bogdanov B, Smith RD (2005) Proteomics
by FTICR mass spectrometry: top down and
bottom up. Mass Spectrom Rev 24:
168200
2. Winter D, Hung CW, Jaskolla TW etal (2012)
Enzyme-cleavable tandem peptides for quantitative studies in MS-based proteomics. Proteomics
12:34703474
3. Winter D, Steen H (2011) Optimization of cell
lysis and protein digestion protocols for the
analysis of HeLa S3 cells by LC-MS/MS.
4. Singh S, Winter D, Kirchner M etal (2014)
Co-regulation proteomics reveals substrates
and mechanisms of APC/C-dependent degradation. EMBO J 33:385399
5. Villen J, Gygi SP (2008) The SCX/IMAC
enrichment approach for global phosphorylation analysis by mass spectrometry. Nat Protoc
3:16301638
6. Wisniewski JR, Zougman A, Nagaraj N etal
(2009) Universal sample preparation method
for proteome analysis. Nat Methods 6:
359362
7. Zahedi RP, Lewandrowski U, Wiesner J etal
(2008) Phosphoproteome of resting human
platelets. J Proteome Res 7:526534
8. Sapan CV, Lundblad RL, Price NC (1999)

Colorimetric protein assay techniques.
Biotechnol Appl Biochem 29:99108
9. Paulo JA, Lee LS, Wu B etal (2010) Optimized
sample preparation of endoscopic collected
pancreatic fluid for SDS-PAGE analysis.

10.
Jiang L, He L, Fountoulakis M (2004)
Comparison of protein precipitation methods
for sample preparation prior to proteomic
analysis. J Chromatogr A 1023:317320
11. Shevchenko A, Thomas H, Havlis J etal (2006)
In-gel digestion for mass spectrometric characterization of proteins and proteomes. Nat
Protoc 1:28562860
12. Yu YQ, Gilar M, Lee PJ etal (2003) Enzyme-
friendly, mass spectrometry-compatible surfactant for in-solution enzymatic digestion of
proteins. Anal Chem 75:60236028
13. Rappsilber J, Ishihama Y, Mann M (2003) Stop
and go extraction tips for matrix-assisted laser
desorption/ionization, nanoelectrospray, and
LC/MS sample pretreatment in proteomics.
Anal Chem 75:663670
14. Bantscheff M, Schirle M, Sweetman G etal
(2007) Quantitative mass spectrometry in proteomics: a critical review. Anal Bioanal Chem
389:10171031
Chapter 21
Comprehensive Protocol to Simultaneously
Study Protein Phosphorylation, Acetylation,
and N-Linked Sialylated Glycosylation
Marcella Nunes Melo-Braga*, Mara Ibez-Vea*,
Martin Rssel Larsen, and Katarzyna Kulej
Abstract
Post-translational modifications (PTMs) such as phosphorylation, acetylation, and glycosylation are an
essential regulatory mechanism of protein function and they are associated with a range of biological processes. Since most PTMs alter the molecular mass of a protein, mass spectrometry (MS) is the ideal analytical
tool for studying various PTMs. However, PTMs are generally present in substoichiometric levels and
therefore their unmodified counterpart often suppresses their signal in MS. Consequently, PTM analysis by
MS is a challenging task requiring highly specialized and sensitive enrichment methods. Currently, several
methods have been implemented for PTM enrichment and each of them has its drawbacks and advantages
as they differ in selectivity and specificity toward specific protein modifications. Unfortunately, for most of
the more than 300 known modifications we have none or poor tools for selective enrichment.
Here, we describe a comprehensive workflow to simultaneously study phosphorylation, acetylation,
and N-linked sialylated glycosylation from the same biological sample. The protocol involves an initial
titanium dioxide (TiO2) step to enrich for phosphopeptides and sialylated N-linked glycopeptides followed
by glycan release and post-fractionation using sequential elution from immobilized metal affinity chromatography (SIMAC) to separate mono-phosphorylated and deglycosylated peptides from multiphosphorylated ones. The IMAC flow-through and acidic elution is subsequently subjected to a next
round of TiO2 enrichment for further separation of mono-phosphopeptides from deglycosylated peptides.
In addition, the acetylated peptides present in the first TiO2 flow-through are enriched by immunoprecipitation (IP). Finally, the samples are fractionated by hydrophilic interaction liquid chromatography (HILIC)
to reduce sample complexity and increase the coverage during LC-MS/MS analysis. This allows the analysis of multiple types of modifications from the same highly complex biological sample without decreasing
the quality of each individual PTM study.
Key words Protein post-translational modification (PTM) enrichment, Phosphorylation, Acetylation,
Sialic acid (SA)-glycosylation, Immunoprecipitation (IP), TiSH Comprising of titanium dioxide
(TiO2), Sequential elution from immobilized metal affinity chromatography (SIMAC) and hydrophilic interaction liquid chromatography (HILIC), Liquid chromatography coupled with tandem
mass spectrometry (LC-MS/MS)
*Authors contributed equally to this chapter.
275
276
Marcella Nunes Melo-Braga et al.
Introduction
Proteins are frequently modified by post-translational modifications
(PTMs) such as phosphorylation, acetylation, and glycosylation.
PTMs regulate protein structure, function and lifetime, modulating
their activity in dynamic cells. Aberrations of protein regulation by
modifications lead to the development of various disorders and diseases. Despite the pivotal role of PTMs in a range of biological
processes, PTM analysis by mass spectrometry (MS) is a challenging task since modified peptides are present in substoichiometric
amount in comparison to their unmodified counterpart in proteolytic digests from cells or tissues. However, recent advances in
PTM enrichment methods combined with the new generation of
MS instrumentation have tremendously contributed to large-scale
studies of PTMs.
Nowadays, several techniques exist for phosphopeptide
enrichment prior to MS analysis. The most extensively used methods involve metal ions for the binding of negatively charged phosphopeptides, i.e. immobilized metal affinity chromatography
(IMAC) [1, 2], and metal oxide affinity chromatography such as
titanium dioxide (TiO2) [35]. However, studies comparing various phosphopeptide enrichment methods showed that each
method allows the isolation of distinct subset of phosphopeptides
whereas none of the methods are able to provide a whole phosphoproteome, even though they partially overlap [6]. On the
other hand, this is highly dependent on the individuals performing the analysis and on the numerous protocols that exist for the
phosphopeptide enrichment. The possibility to combine the
strength of different enrichment methods by serial performance
has substantially improved the phosphoproteomic field. For
instance, the combination of IMAC and TiO2 chromatography,
known as Sequential elution from IMAC (SIMAC), has been
employed to separate mono-phosphorylated peptides from multiply phosphorylated peptides in large-scale studies [7]. IMAC has
a stronger selectivity for multi-phosphopeptides than for monophosphopeptides, leading to a better characterization of the first
ones. Following these enrichment methods, a new multi-dimensional large-scale phosphopeptide enrichment strategy, known as
the TiSH protocol was proposed to study phosphorylation from
low amount of starting material (300 g). The TiSH strategy
includes an initial TiO2 enrichment followed by SIMAC and
hydrophilic interaction liquid chromatography (HILIC) fractionation to decrease the complexity of mono-phosphopeptide fraction [8]. The TiSH approach lead to several advances: (1) the
TiO2-based pre-enrichment step improves the enrichment specificity compared to a setup utilizing only SIMAC; (2) TiO2 is more
robust than IMAC being compatible with several chemical reagents
[9], and (3) HILIC fractionation of the mono-phosphorylated
Strategy for Simultaneously Enrichment of PTMs
277
peptides increase the phosphopeptide coverage from complex

mixtures. This increases throughput and allows for performing
large-scale phosphoproteomics from low amounts of cells or biological tissues.
Despite the high selectivity of TiO2 toward phosphopeptides,
it was later shown that TiO2 also has a high selectivity toward
N-linked sialylated glycopeptides presumably due to the high negative charge of these molecules [10]. Although there are several
methods to enrich glycopeptides such as lectins [11] and hydrazine
affinity purification [12], only TiO2 has the capability to selectively
isolate sialylated glycopeptides [13]. Due to these additional properties, TiO2 has been successfully applied to simultaneously enrich
phosphopeptides and N-linked sialylated-glycopeptides [1416].
Conversely to protein phosphorylation and glycosylation
there is only one enrichment strategy available for lysine acetylation, which is acetyl-lysine immunoprecipitation (IP) [17]. Acetyllysine IP has allowed the identification of hundreds to thousands
of lysine acetylation sites in different organisms such as bacteria
[18], plant [19], parasite [20], and human [21]. In addition, the
IP of lysine acetylation has been combined with strong cation
exchange (SCX) to simultaneously study acetylation and phosphorylation [22], as well as with TiO2 to simultaneously study
phosphorylation, sialylated N-glycosylation and lysine acetylation
in mouse brains [15, 23]. However, the relative low amount of
lysine acetylation sites from biological material compared to phosphorylation sites suggests that lysine-acetylation is a rather low
abundant PTM compared to phosphorylation. This means that
the enrichment of peptides carrying lysine acetylation is highly
dependent on the amount of starting material.
The simultaneous study of different PTMs allows the discovery of potential interplays between them, which are known to coregulate a range of biological processes [24]. Therefore, a number
of strategies for serial enrichments have been reported in order to
study different PTMs from the same sample (e.g. [24]).
Here, a modified protocol of the TiSH approach, to simultaneously study phosphopeptides, sialylated glycopeptides, and lysine
acetylated peptides, is presented. Using this strategy, a comprehensive overview of proteomes and their selected PTMomes can be
achieved from relatively low amount of biological material.
Materials
All the solutions and buffers should be prepared with Milli-Q water
(UHQ), analytical grade reagents, and highest purity chemicals.
Organic solutions should be prepared fresh or stored for no more
than 2 weeks before performing the protocol to avoid changes in
buffers composition.
278
2.1 Model Sample

Preparation
1. Cell line: HeLa (human cervix epithelial adenocarcinoma) cells.

2. Culture medium: Dulbeccos Modified Eagles Medium
(DMEM, GlutaMAX supplement, Gibco, Life Technologies)
supplemented with 10 % fetal bovine serum (FBS) (Gibco
Life Technologies) and 1 % penicillin-streptomycin (Gibco
Life Technologies).
3. Nunclon 140 20 mm cell culture dishes (Thermo Scientific).
4. Nunc Cell Scrapers (Thermo Scientific).
5. Washing buffer 1: Hanks Balanced Salt Solution (HBSS; no
calcium, no magnesium, Gibco Life Technologies).
6. Washing buffer 2: HBSS containing complete EDTA-free protease inhibitor cocktail (Roche Diagnostics, Mannheim,
Germany) and phosphatase inhibitor cocktail (PhosSTOP,
Roche Diagnostics, Mannheim, Germany).
2.2 Lysis, Reduction,

Alkylation,
and Digestion
of Model Sample
1. Lysis and reduction: 6 M urea, 2 M thiourea, 10 mM dithiotreitol (DTT), phosphatase inhibitors, 0.1 mM sodium pervanadate, and 0.04 AU of Lys-C (lysyl endopeptidase, WAKO
Pure Chemical Industries, Ltd., Osaka, Japan) (see Note 1).
2. Alkylation: 20 mM iodoacetamide in 20 mM triethylammonium bicarbonate (TEAB).
3. Digestion: 2 % trypsin (Sigma porcine trypsin, St. Louis, MO, US)
(see Note 2).
4. Stop digestion: 100 % formic acid.
2.3 Desalting/
Concentration
of the Peptide Mixture
on Reversed-Phase
(RP) Columns
1. POROS Oligo R3 reversed-phase material (Applied Biosystems,

Foster City, CA, US) or Oasis HLB cartridges (Waters, Milford,
MA, US) (see Note 3).
2. p200 pipette tips.
3. 3 M Empore C8 disk (3 M Bioanalytical Technologies, St.
Paul, MN, US).
4. Syringe for HPLC loading (P/N 038250, N25/500-LC PKT
5, SGE, Ringwood, Victoria, Australia) to create a small plug
of C18 membrane material.
5. Plastic syringe of 1 and 5 mL (4606051V, B. Braun Medical
Inc., US) to create the pressure of the micro-columns.
6. RP loading buffer: 0.1 % trifluoroacetic acid (TFA).
7. RP elution buffer: 60 % ACN, 0.1 % TFA.
2.4 Titanium Dioxide

(TiO2) Batch Mode
Enrichment
of Phosphopeptides
and Sialylated
N-Glycopeptides
1. Titanium dioxide (TiO2) beads (Titansphere, 5 m, GL

Sciences Inc., Japan).
2. Low-binding Eppendorf tubes 1.5 mL (Sorenson BioScience
Inc., Utah, US).
3. TiO2 loading buffer: 80 % ACN, 5 % TFA, and 1 M glycolic
acid.
279
4. TiO2 washing buffer 1: 80 % ACN, 1 % TFA.

5. TiO2 washing buffer 2: 10 % ACN, 0.1 % TFA.
6. TiO2 elution buffer: 1.5 % ammonium hydroxide (pH 11.3),
always prepare fresh solution.
2.5 Deglycosylation
of N-Linked
Glycopeptides
1. Deglycosylation buffer: 20 mM TEAB, GlykoSialidase A

(ProZyme, Inc., San Leandro, CA, US1 unit), PNGase F
glycerol free (New England Biolabs Inc., Ipswich, MA,
US500,000 U/mL).
2.6 Enrichment
of Multiphosphorylated
Peptides by Sequential
Elution from IMAC
Beads (SIMAC)
1. Iron-coated PHOS-select metal chelate beads (Sigma,

Missouri, US).
2. GELoader tips 200 L (Alpha Laboratories, Hampshire, UK).
3. SIMAC loading buffer: 50 % ACN, 0.1 % TFA.
4. SIMAC elution buffer 1: 20 % ACN, 1 % TFA.
5. SIMAC elution buffer 2: 1.5 % ammonium hydroxide
(pH 11.3), always prepare fresh solution.
6. Sample acidification: 100 % formic acid, 10 % TFA.
2.7 Separation
of MonoPhosphopeptides
from Deglycosylated
Peptides by Second
TiO2 Enrichment
1. TiO2 loading buffer: 70 % ACN, 2 % TFA.
2.8 Acetyl-Lysine
Immunoprecipitation
(IP)
1. Acetyl lysine antibody, agarose (ImmuneChem Pharmaceuticals

Inc., Burnaby, British Columbia, Canada) against epsilon
amino group of lysine residues, K-Ac.
2. TiO2 washing buffer: 50 % ACN, 0.1 % TFA.

3. TiO2 elution buffer: 1.5 % ammonium hydroxide (pH 11.3),
always prepare fresh solution.
4. Sample acidification: 100 % formic acid, 10 % TFA.
2. IP loading buffer: 50 mM MOPS, 10 mM sodium phosphate,

50 mM sodium chloride, pH 8.0.
3. IP elution buffer: 1 % TFA.
2.9 Hydrophilic
Interaction Liquid
Chromatography
(HILIC) Fractionation
1. Sample dissolution buffer: 90 % ACN, 0.1 % TFA (see Note 4).

2. MicroHPLC-Agilent 1200 Series (Agilent, Santa Clara, CA, US).
3. In-house packed 15 cm 0.3 mm TSKGel Amide 80 3 m column (Tosoh Bioscience, Stuttgart, Germany).
4. Solvent A: 0.1 % TFA.
5. Solvent B: 90 % ACN, 0.1 % TFA.
2.10 nanoLC-MS/MS
Analysis
1. Sample dissolution buffer: 0.1 % formic acid.

2. In-house packed 2 cm 100 m Reprosil-Pur C18-AQ 3 m
pre-column connected to 17 cm 75 m Reprosil-Pur C18-AQ
3 m analytical column (Dr. Maisch GmbH, Germany).
280
3. Easy-nLC system (Thermo Scientific, Bremen, Germany).

4. Solvent A: 0.1 % formic acid.
5. Solvent B: 95 % ACN, 0.1 % formic acid.
6. LTQ Orbitrap Velos system (Thermo Scientific, Bremen,
Germany).
2.11
Data Analysis
1. Data processing and search software: Proteome Discoverer

v1.4 software (ThermoFisher Scientific), MASCOT (v2.3,
Matrix Science Ltd, London, UK) and Sequest HT search
engines.
2. Database: Swiss-Prot human v3.53 database.
Methods
The complete workflow for the multiple PTM enrichment strategy
described in this chapter is illustrated in Fig. 1. The workflow consists of an initial simultaneous enrichment of phosphopeptides and
N-linked sialylated glycopeptides using the first TiO2 step. The
flow-through (FT) from this step is used to enrich acetylated peptides by IP. The K-Ac antibodies capture acetylated peptides,
whereas the non-modified peptides are present in the IP-FT. After
enzymatically release of N-linked glycan structures from the
sialylated N-linked glycopeptides in the TiO2 eluate, the multiphosphopeptides are separated from the mono-phosphopeptides
and deglycosylated peptides by SIMAC. Three fractions are
obtained from IMAC: (I) IMAC-FT, (II) an acidic elution fraction
(pH 2.3), and (III) a basic elution fraction (pH 11.3). The FT
and acidic elution that contains deglycosylated peptides and monophosphopeptides are combined and subjected to a second milder
TiO2 chromatographic step to separate both types of modified
peptides. Finally, in order to decrease the complexity of monophosphopeptides, deglycosylated peptides and non-modified peptides, HILIC fractionation is performed prior LC-MS/MS analysis.
Here, a total of 500 g of whole HeLa cell lysate was used for a
qualitative analysis to illustrate the average outcome of the protocol; although, the protocol can be optimized to different amounts
(lower or higher) and applied also for quantitative studies.
3.1 Model Sample

Preparation
1. Grow HeLa cells in DMEM medium until 90 % confluence

(~1 107 cells as starting material).
2. Remove DMEM medium and wash cell monolayer twice with
5 mL washing buffer 1 pre-warmed to 37 C.
3. Add 5 mL ice-cold wash buffer 2 and harvest cells by scraping
them off the plate.
281
Fig. 1 Enrichment workflow. An initial titanium dioxide (TiO2) step is performed to enrich phosphopeptides and
sialylated N-linked glycopeptides followed by glycan release. Then, SIMAC is performed to separate monophosphorylated and deglycosylated peptides from multi-phosphorylated peptides. The IMAC-FT together with
the IMAC acidic elution is subjected to a second round of TiO2 enrichment to separate mono-phosphopeptides
from deglycosylated peptides. In addition, the first TiO2-FT is used to enrich acetylated peptides by IP. Finally,
the samples are fractionated by HILIC prior to LC-MS/MS analysis
4. Collect cells into 15 mL falcon tube and pellet them by 5 min

centrifugation, 250 g at 4 C.
5. Remove supernatant and snap-froze the cell pellet in liquid
nitrogen.
6. Store the cell pellet at 80 C until further sample preparation.
282
3.2 Lysis, Reduction,

Alkylation,
and Digestion
of Model Sample
1. Add 50 L lysis buffer to the HeLa pellet (lysis buffer volume

depends on the pellet size; here we used 1 107 cells).
Vortex well and incubate for 2 h at room temperature (RT)
2. After incubation, dilute the sample ten times with 20 mM
TEAB, pH 7.5 and sonicate on ice. In order to alkylate the
sample, 20 mM iodoacetamide is added and the sample is incubated for 20 min in the dark at RT.
3. Digest the sample using trypsin (enzyme to substrate ratio
1:50) overnight (1216 h) at RT.
4. After incubation, quench the reaction with formic acid to a
final concentration of 5 % and leave for 5 min at RT. Then,
centrifuge for 15 min at 14,000 g in order to pellet lipids.
5. Transfer the supernatant to another low-binding Eppendorf
tube.
3.3 Desalting/
Concentration
on HLB/RP Cartridge
(See Notes 7 and 8)
1. Add 0.1 % TFA to the tryptic peptide sample to achieve

approximately 1 mL of final volume and adjust the pH to
below 2.0.
2. Activate the HLB cartridge with 1 mL of 100 % methanol followed by 1 mL of 100 % ACN.
3. Equilibrate the cartridge twice with 2 mL of 0.1 % TFA.
4. Load the sample onto the HLB cartridge slowly and collect the
FT.
5. Pass again the FT through the same HLB cartridge.
6. Wash the cartridge twice with 2 mL of 0.1 % TFA.
7. Elute the peptides in a new low-binding Eppendorf tube with
1 mL of 60 % ACN/0.1 % TFA and lyophilize the sample.
8. Reconstitute the sample in 0.1 % formic acid.
9. Take an aliquot for amino acid analysis (AAA) to determine
peptide concentration (see Note 9).
3.4 TiO2 Batch Mode

Enrichment
of Phosphopeptides
and Sialylated
N-Glycopeptides
1. Transfer 3 mg of TiO2 beads (for 500 g of starting material)

to a new low-binding Eppendorf tube (see Note 10).
2. Dilute the peptide sample at least ten times in TiO2 loading
buffer (v/v) and add the TiO2 beads (see Note 11). Alternatively,
adjust the sample volume to achieve the proper loading buffer
concentration such as for 100 L sample add 50 L water,
50 L of 100 % TFA, 800 L of 100 % ACN and 76 mg of
glycolic acid.
3. Incubate the sample on the shaker for 10 min at RT. Then,
centrifuge to pellet the beads using a table centrifuge for 15 s
(see Note 12).
283
4. Transfer the supernatant to another low-binding Eppendorf

tube containing half of the amount of the TiO2 beads as used
in the first round. Repeat the incubation as described above to
increase the yield of peptides modified by phosphorylation and
sialylated N-glycosylation (see Note 13).
5. Save the supernatant (TiO2FT) for further enrichment of
acetylated and non-modified peptides (see Subheading 3.8).
6. Pool the TiO2 beads from the two incubations using 100 L of
TiO2 loading buffer and transfer to a new low-binding
Eppendorf tube (see Note 14).
7. Vortex the sample for 10 s and then centrifuge to pellet the
beads. The supernatant is removed and pooled together with
the TiO2FT.
8. Add 70100 L of TiO2 washing buffer 1, vortex for 15 s and
centrifuge to pellet the beads. This step is performed to
remove the contaminating hydrophobic non-modified peptides.
Repeat this step using 50100 L of TiO2 washing buffer 2 in
order to remove the hydrophilic non-modified peptides and
the neutral glycopeptides (see Note 15).
9. Dry the beads for 10 min in the vacuum centrifuge or on the
table (see Note 16).
10. Elute the phosphopeptides and sialylated N-glycopeptides by
adding 100150 L of TiO2 elution buffer. Vortex and incubate the solution in the shaker for 10 min to allow an efficient
elution.
11. Centrifuge the solution for 1 min and pass the supernatant
over a filter (C8 stage tip) into a new low-binding Eppendorf
tube to avoid the presence of TiO2 beads in the solution. Wash
the TiO2 beads with 30 L of elution buffer for 5 min, pass
over the C8 filter and pool with the other TiO2 eluate. To
recover any peptide that bound to the C8 filter, add 5 L of
30 % ACN to the C8 stage tip and pass this solution through
and collect with the TiO2 eluates.
12. Lyophilize the eluted peptides. Any remnant of ammonia can
interfere with the subsequent steps.
13. Lyophilize also the TiO2FT to perform the lysine acetylation
enrichment (see Subheadings 3.8 and 3.9).
3.5 Deglycosylation
of N-Linked
Glycopeptides
1. Redissolve the lyophilized peptides from the TiO2 enrichment

in 40 L of 20 mM TEAB, pH 7.5 and add 1 L of PNGase F
and 0.5 L of Sialidase A.
2. Incubate at 37 C overnight.
3. To quench the reaction, add 1 L of 10 % TFA.
284
3.6 Enrichment
of Multiphosphorylated
Peptides by SIMAC
1. Slowly, dilute the acidified deglycosylated peptide solution

with 200 L of SIMAC loading buffer (see Note 17). Adjust
the pH to 1.61.8 using 10 % TFA solution.
2. As in the TiO2 enrichment, the amount of IMAC beads per
sample amount is a crucial factor to avoid non-specific binding.
The optimal quantity is 6080 L beads per 300 g of starting
material [8].
3. Transfer 130 L of IMAC beads (for 500 g of starting material) to a low-binding Eppendorf tube and wash twice with
200 L of IMAC loading buffer by mixing and table centrifugation (see Note 18).
4. Add the peptide solution to the IMAC beads and incubate for
30 min at RT under slow rotation shaking.
5. Centrifuge the sample for 15 s at 14,000 g and transfer half of
the supernatant to a new low-binding Eppendorf tube called
SIMACFT.
6. Squeeze the tip of a 200 L GELoader tip to retain the IMAC
beads.
7. Transfer the remaining solution with IMAC beads to the constricted 200 L GELoader tip. Use a syringe to press the liquid
through into the SIMACFT tube and pack the IMAC
micro-column.
8. Wash the IMAC column with 50 L of loading buffer and collect it in the SIMACFT tube.
9. Elute the mono-phosphorylated peptides from the IMAC column with 70 L of SIMAC elution buffer 1. The eluate is collected together with the SIMACFT from the above steps 5,
7, and 8. This step should be performed slowly (around 1
drop/s) to allow mono-phosphorylated peptides to fall off the
IMAC beads whereas multiphosphorylated peptides still are
retained on the IMAC resin.
10. Elute the multi-phosphorylated peptides with 80 L of SIMAC
elution buffer 2 directly down into a pre-equilibrated p200
stage tip blocked with a C8 filter plug and packed with R3 material (12 cm of R3) (see Subheading 3.10). It is important to
acidify the eluate with 8 L of 100 % formic acid and 2 L of
10 % TFA prior to the contact with R3 material (see Note 19).
11. Elute the multi-phosphorylated peptides after washing the column (see Subheading 3.10) and lyophilize the sample.
3.7 Separation
of MonoPhosphopeptides
from Deglycosylated
Peptides by Second
TiO2 Enrichment
1. Dilute the SIMACFT at least ten times with second TiO2

loading buffer or adjust the loading buffer concentration by
adding 100 % ACN and 100 % TFA to achieve a final concentration of 70 % ACN/2 % TFA. Add the loading buffer or the
100 % ACN slowly to avoid peptide precipitation.
285
2. Add the same amount of TiO2 used previously in Subheading 3.4

(the same TiO2 beads can be used if they are washed extensively
with pH 11.0). Vortex and incubate for 10 min on a shaker.
3. Centrifuge to pellet the beads and transfer the supernatant to
a new low-binding Eppendorf tube containing half of the
amount of TiO2 beads. Then, incubate for 10 min as described
in Subheading 3.4.
4. Centrifuge to pellet the beads and save the supernatant that
contains the deglycosylated peptides.
5. Pool the TiO2 pellets using 100 L of second TiO2 washing
buffer, vortex and centrifuge. Remove the supernatant and
combine it with the deglycosylated peptides.
6. Dry the beads for 10 min in the vacuum centrifuge or on the
table (see Note 16).
7. Add 100 L of TiO2 elution buffer, vortex and incubate for
10 min on a shaker to elute the mono-phosphopeptides.
8. Centrifuge the solution for 1 min and pass the supernatant
over a C8 stage filter repeating the item 10 from Subheading 3.6
but using two R3 micro-columns of 1.5 cm each (the number
of R3 micro-columns depends on the amount of starting material) (see Note 20). This step is performed sequentially being
the solution passed through the first R3 and collected directly
on top of the second R3 micro-column. Wash and elute the
mono-phosphopeptides from both columns as described in
Subheading 3.10. Pool the eluates in the same tube. Lyophilize
the mono-phosphopeptides.
9. Lyophilize the second TiO2FT that contains the deglycosylated peptides. Subsequently, reconstitute the sample in 80 L
of 0.1 % TFA prior to R3 micro-purification as described in
Subheading 3.10.
3.8 Desalting/
Concentration
of the TiO2FT on RP
Columns Prior
to Acetyl-Lysine IP
1. The FT from the first TiO2 is used to enrich acetylated peptides

by immunoprecipitation.
3.9
1. Reconstitute the above dried peptide sample in 200 L of IP

loading buffer and adjust the pH to 8.0 with sodium hydroxide or hydrochloric acid, respectively.
Acetyl-Lysine IP
2. Dissolve the sample in 1 mL of 0.1 % TFA.

3. Purify the sample by a HLB RP cartridge as described in
Subheading 3.3 in order to remove glycolic acid present in the
TiO2 loading buffer.
2. For 100200 g of sample use 20 L of slurry of anti-Lys

acetylated antibody immobilized on agarose beads.
3. Transfer 67 L of anti-K-Ac antibody beads (for 500 g of
starting material) to a new low-binding Eppendorf tube and
286
wash twice with 200 L of IP loading buffer by mixing and

table centrifugation (see Note 21).
4. Incubate the reconstituted sample with the antibody for 12 h
under rotation at 4 C.
5. Pellet the antibody beads using a table centrifuge and transfer
the supernatant to a new low-binding Eppendorf tube. The
supernatant (K-AcFT) contains the non-modified peptides.
6. Wash the anti-K-Ac-antibody beads four times with 200 L of
IP loading buffer as mentioned above. Pool all the supernatants together into the K-AcFT tube.
7. Wash twice the antibody beads with 200 L of water as
described above.
8. Elute the K-Ac peptides with 200 L of elution buffer and
subsequently purify the eluate as described in Subheading 3.10.
Lyophilize the purified K-Ac peptides.
9. Lyophilize the K-AcFT that contains the non-modified
peptides to be used for subsequent proteome analysis.
3.10 Desalting/
Concentration
on RP Columns
1. Prepare a micro-column using p200 tip by plugging the constricted end with C8 material filter. Pack 2 cm column with R3
material slurry in 100 % ACN by applying air pressure using a
syringe.
2. Equilibrate the column with 50100 L of 0.1 % TFA.
3. Load the acidified sample (pH < 2.0) into the column and
make a gently air pressure. It is important to perform this step
slowly. In case the sample is dried, add 80 L of 0.1 % TFA to
reconstitute the sample prior to load. Collect the FT.
4. Pass again the FT through the micro-column to increase the
binding of peptides.
5. Wash the column with 100 L of 0.1 % TFA.
6. Elute the peptides into a new low-binding Eppendorf tube
with 100 L of 60 % ACN/0.1 % TFA.
7. Lyophilize the sample in a vacuum centrifuge.
3.11 HILIC
Fractionation of MonoPhosphopeptides,
Deglycosylated
Peptides and Nonmodified Peptides
1. Reconstitute the sample in 45 L of HILIC solvent B by adding first 0.45 L of 10 % TFA and then 4.05 L of milli-Q
water. Vortex and finally add 40.5 L of 100 % ACN slowly to
avoid precipitation. This step should be performed immediately prior to fractionation since the acetonitrile evaporates fast
altering the efficiency of the method.
2. Inject 40 L onto the microHPLC HILIC column.
3. Fractionate the peptide mixture using the gradient described
in Table 1.
4. Collect the fractions according to Table 2.
287
Table 1
HILIC gradient
Time (min)
Concentration of Sol. B
(90 % ACN, 0.1 %TFA)
Flow (L/min)
00:00
100
12
08:60
100
12
09:00
95
35:00
60
39:00
42:00
46:00
100
48:00
100
The gradient used for fractionation of all obtained modified and unmodified peptide
samples by HILIC microHPLC, except for the multi-phosphorylated peptides
Table 2
Timeline of HILIC fraction collector
Time (min)
Trigger mode
Time slices (min)
00:00
Time-based
10
10:00
Time-based
14:00*
Time-based
36:00**
Time-based
48:00
off
The timeline used to collect all peptide samples beside the multi-phosphorylated peptides. In addition, alteration to *20:00 and **40:00 was done for the sample containing
mono-phosphorylated peptides
5. Pool the fractions according to the UV absorbance intensity

( = 210 nm) and chromatographic profile (Fig. 2). The number of analyzed fractions depends on the complexity of the
sample and the accessibility to LC-MS/MS instruments.
6. Lyophilize the fractions.
3.12 nanoLC-MS/MS
1. Reconstitute the dried samples (multi-phosphopeptides, acetylated peptides as well as the HILIC fractions from monophosphopeptides, deglycosylated peptides and non-modified
peptides) in 0.1 % formic acid.
2. Inject 5 L of sample onto an in-house packed Reprosil-Pur
C18-AQ (3 m; Dr. Maisch GmbH, Germany) pre-column
288
Fig. 2 HILIC chromatogram. The figure illustrates the HILIC chromatogram of non-modified peptides in order to
show how fractions were pooled in the present work. Asterisk denote individual fractions
(2 cm 100 m) connected to an analytical column

(17 cm 75 m) using an Easy-nLC system (Thermo Scientific,
Bremen, Germany).
3. Run the samples at a flow of 250 nL/min, using 95 % ACN
(B) and water (A) both containing 0.1 % formic acid as mobile
phase.
4. Depending on the sample, the gradient was 034 % solvent B
in 60, 90 or 120 min, 34100 % solvent B in 5 min, and 8 min
at 100 % solvent B.
5. MS analysis was performed in an LTQ Orbitrap Velos system
(Thermo Scientific, Bremen, Germany) with data-dependent
acquisition for CID MS/MS analysis of the ten most intense
ions except for phosphopeptides where MSA MS/MS analysis
was used.
3.13
Data Analysis
Data processing and search were performed using Proteome

Discoverer v1.4 (ThermoFisher Scientific). Data were searched
against the Swiss-Prot human v3.53 database using an in-house
MASCOT server (v2.3, Matrix Science Ltd, London, UK) and the
Sequest HT. Trypsin was selected as digestion enzyme and two
missed cleavages were allowed. Database searches were performed
with the following parameters: precursor mass tolerance of 10 ppm,
product ion mass tolerance of 0.8 Da, and cysteine carbamidomethylation as fixed modification. Searches were also conducted
with the following variable modifications: methionine oxidation;
289
serine, threonine, tyrosine phosphorylation; asparagine and glutamine deamidation. Protein grouping was performed, in order to
avoid presence of different proteins identified by non-unique peptides. Only peptides with up to q-value of 0.01 (Percolator),
Mascot and Sequest HT rank 1, Mascot and Sequest HT search
engine 1, a Sequest HT Cn of 0.1, a cut-off value of Mascot score
22 and a cut-off value of XCorr score greater than 1.5, 2, 2.5 and
3 for charge states of +1, +2, +3, and +4, respectively, were considered for further analysis.
3.14
Results
Figure 3 highlights the number of modified peptides identified

specifically by the analysis of the dedicated enrichment step
described in this chapter. The four different steps achieved various
specificity and sensitivity regarding enrichment of modified peptides. Mono- and multi-phosphorylated peptides were efficiently
enriched from the elution of the IMAC and second TiO2 chromatography achieving the identification of thousands of phosphopeptides. Hundreds of sialylated N-glycopeptides were enriched from
the FT of the second TiO2 chromatography. Finally, acetylated
peptides were enriched from the elution of the K-Ac IP. The relative low number of identified acetylated lysines is due to the lower
abundance of this PTM in comparison with phosphorylation or
N-linked sialylated glycosylation in biological material.
Furthermore, a high number of peptides contain more than one
differently modified site. Therefore, a significant part of acetylated
sites might co-purify with phosphopeptides during the first TiO2
enrichment step, causing the reduced number of acetylated peptides identified in K-Ac IP.
Fig. 3 Results obtained from the analysis of phosphopeptides, sialylated N-linked glycopeptides and acetylated
peptides using the optimized workflow. The number of phosphorylated peptides (mono- and multiply phosphorylated peptides), sialylated N-linked glycopeptides and acetylated peptides identified from 500 g of HeLa
cell Lys-C/tryptic digestion using the combination of TiO2, SIMAC, and K-Ac IP enrichment methods
290
Notes
1. Protease inhibitor cocktail can be added if needed.
2. The trypsin is purified by benzamidine sepharose affinity to
guarantee the same proteolytic activity for the entire batch,
as well as to significantly reduce the level of autodigestion
products.
3. The choice between POROS Oligo R3 or Oasis HLB cartridge
depends on the quantity of material. For peptide samples with
quantity higher or equal to 500 g, the Oasis HLB cartridge is
often a choice.
4. To avoid peptide precipitation add to the lyophilized sample
0.45 L of 10 % TFA and then 4.05 L of milli-Q water. Vortex
and finally add 40.5 L of 100 % ACN. The final concentration
of dissolution buffer should be 90 % ACN/0.1 % TFA.
5. Any protein extraction, denaturation, and digestion method is
compatible with the present simultaneous PTMs enrichment
strategy.
6. Mix by vortexing and pipette cell pellet up and down after adding lysis buffer. Cell lysate will form a very viscous solution due
to the presence of nuclear DNA.
7. POROS Oligo R3 micro-columns or Oasis HLB cartridges
desalting step is required if the buffer solutions are not compatible with labeling methods such as iTRAQ, TMT, dimethyl
labeling; or prior to determination of protein concentration by
AAA analysis.
8. TiO2 enrichment might be performed directly after the protein
digestion step since TiO2 chromatography is compatible with
most of the commonly used buffer solutions [9].
9. A labeling strategy such as iTRAQ, TMT or dimethyl labeling
can be introduced in this step for quantitative studies.
Remember to check the compatibility of buffer solutions with
the specific labeling reagents.
10. To reduce nonspecific binding to the TiO2 resin, it is important
to adjust the amount of TiO2 beads to the amount of sample.
The optimal quantity is 0.6 mg TiO2 beads per 100 g peptide
solution [8].
11. In case the sample was lyophilized before, it is crucial to reconstitute the sample in a small volume of 0.1 % TFA. Moreover,
the TiO2 loading buffer should be added slowly to avoid peptide precipitation.
12. Centrifugation time is not critic.
13. This step can be done up to three times, depending on the
sample complexity.
291
14. This step is performed to avoid contamination with non-modified

peptides that may bind to the Eppendorf tube surface.
15. The washing buffer 2 is used to remove the peptides that bind
in the HILIC mode to TiO2. The FT of washing buffer 2 contains neutral glycopeptides which can be analyzed as a separate
PTM fraction.
16. It is very important to dry the TiO2 beads in the vacuum centrifuge. In case vacuum centrifuge is not available, it is necessary to check and adjust the pH to 11.3 in the next step.
17. Prior adding 200 L of SIMAC loading buffer, the sample can
be lyophilized and then reconstituted in a small volume of
0.1 % TFA.
18. Avoid vortexing the IMAC beads in high speed.
19. The direct purification is performed to avoid loss of multiphosphopeptides that can bind to the tube surface. However,
multi-phosphopeptides can be transferred to another lowbinding Eppendorf tube, following with adjustment of the pH
eluate below 2.0 prior to desalting on RP column.
20. This step uses two POROS Oligo R3 micro-columns to ensure
a good recovery of phosphopeptides.
21. The anti-K-Ac antibody beads stock solution and all solutions
containing K-Ac antibody beads should be kept on ice
constantly.
22. Different LC-MS/MS set-up can be used. However, the results
will differ from those obtained using the present protocol.
References
1. Andersson L, Porath J (1986) Isolation of
phosphoproteins by immobilized metal (Fe3+)
affinity chromatography. Anal Biochem 154:
250254
2. Neville DC, Rozanas CR, Price EM, Gruis DB,
Verkman AS, Townsend RR (1997) Evidence
for phosphorylation of serine 753 in CFTR
using a novel metal-ion affinity resin and
matrix-assisted laser desorption mass spectrometry. Protein Sci 6:24362445
3. Pinkse MW, Uitto PM, Hilhorst MJ et al
(2004) Selective isolation at the femtomole
level of phosphopeptides from proteolytic
digests using 2D-NanoLC-ESI-MS/MS and
titanium oxide precolumns. Anal Chem 76:
39353943
4. Kuroda I, Shintani Y, Motokawa M, Abe S,
Furuno M (2004) Phosphopeptide-selective
column-switching RP-HPLC with a titania
precolumn. Anal Sci 20:13131319
5. Larsen MR, Thingholm TE, Jensen ON et al
(2005) Highly selective enrichment of phos-
6.
7.
8.
9.
phorylated peptides from peptide mixtures

using titanium dioxide microcolumns. Mol
Cell Proteomics 4:873886
Bodenmiller B, Mueller LN, Mueller M et al
(2007) Reproducible isolation of distinct, overlapping segments of the phosphoproteome.
Nat Methods 4:231237
Thingholm TE, Jensen ON, Robinson PJ,
Larsen MR (2008) SIMAC (sequential elution
from IMAC), a phosphoproteomics strategy
for the rapid separation of monophosphorylated from multiply phosphorylated peptides.
Mol Cell Proteomics 7:661671
Engholm-Keller K, Birck P, Storling J et al
(2012) TiSH: a robust and sensitive global
phosphoproteomics strategy employing a
combination of TiO2, SIMAC, and HILIC.
Jensen SS, Larsen MR (2007) Evaluation of the
impact of some experimental procedures on different phosphopeptide enrichment techniques.
Rapid Commun Mass Spectrom 21:36353645
292
10. Larsen MR, Jensen SS, Jakobsen LA, Heegaard

NH (2007) Exploring the sialiome using titanium dioxide chromatography and mass spectrometry. Mol Cell Proteomics 6:17781787
11. Bunkenborg J, Pilch BJ, Podtelejnikov AV,
Wisniewski JR (2004) Screening for
N-glycosylated proteins by liquid chromatography mass spectrometry. Proteomics 4:454465
12. Zhang H, Li XJ, Martin DB, Aebersold R
(2003) Identification and quantification of
N-linked glycoproteins using hydrazide chemistry, stable isotope labeling and mass spectrometry. Nat Biotechnol 21:660666
13. Palmisano G, Lendal SE, Engholm-Keller K
et al (2010) Selective enrichment of sialic acidcontaining glycopeptides using titanium dioxide
chromatography with analysis by HILIC and
mass spectrometry. Nat Protoc 5:19741982
14. Palmisano G, Parker BL, Engholm-Keller K
et al (2012) A novel method for the simultaneous enrichment, identification and quantification of phosphopeptides and sialylated
glycopeptides applied to a temporal profile of
mouse brain development. Mol Cell Proteomics
11:1191202, M112.017509 [pii] 10.1074/
mcp.M112.017509
15. Edwards AV, Schwammle V, Larsen MR (2014)
Neuronal process structure and growth proteins
are targets of heavy PTM regulation during
brain development. J Proteomics 101:7787
16. Melo-Braga MN, Schulz M, Liu Q et al (2013)
Comprehensive quantitative comparison of the
membrane proteome, phosphoproteome and
sialiome of human embryonic and neural stem
cells. Mol Cell Proteomics 13:311328
17. Kim SC, Sprung R, Chen Y et al (2006)

Substrate and functional diversity of lysine acetylation revealed by a proteomics survey. Mol
Cell 23:607618
18. Zhang J, Sprung R, Pei J et al (2009) Lysine
acetylation is a highly abundant and evolutionarily conserved modification in Escherichia
coli. Mol Cell Proteomics 8:215225
19. Melo-Braga MN, Verano-Braga T, Leon IR
et al (2012) Modulation of protein phosphorylation, N-glycosylation and Lys-acetylation in
grape (Vitis vinifera) mesocarp and exocarp
owing to Lobesia botrana infection. Mol Cell
20. Jeffers V, Sullivan WJ Jr (2012) Lysine acetylation is widespread on proteins of diverse function and localization in the protozoan parasite
Toxoplasma gondii. Eukaryot Cell 11:735742
21. Choudhary C, Kumar C, Gnad F et al (2009)
Lysine acetylation targets protein complexes
and co-regulates major cellular functions.
Science 325:834840
22. van Noort V, Seebacher J, Bader S et al (2012)
Cross-talk between phosphorylation and lysine
acetylation in a genome-reduced bacterium.
Mol Syst Biol 8:571. doi:10.1038/msb.2012.4
23. Edwards AV, Edwards GJ, Schwammle V et al
(2014) Spatial and temporal effects in protein
post-translational modification distributions in
the developing mouse brain. J Proteome Res
13:260267
24. Mertins P, Qiao JW, Patel J et al (2013)
Integrated proteomic analysis of posttranslational modifications by serial enrichment. Nat Methods 10:634637
Chapter 22
Protein Profiling and Phosphoprotein Analysis
by Isoelectric Focusing
Abstract
Protein profiling enables the qualitative characterization of a proteome of interest. Phosphorylation is
a post-translational modification with regulatory functions in a plethora of cell processes. We present
an experimental workflow for simultaneous analysis of the proteome and phosphoproteome with no
additional enrichment for phosphoproteins/phosphopeptides. Our approach is based on isoelectric
focusing (IEF) which allows the separation of peptide mixtures on an immobilized pH gradient (IPG)
according to their isoelectric point. Due to the negative charge of the phosphogroup, most of the
phosphopeptides migrate toward acidic pH values. Peptides and phosphopeptides are then identified
by mass spectrometry (MS) and phosphopeptide spectra are manually checked for the assignment of
phosphorylation sites. Here, we apply this methodology to investigate synaptosome extracts from
whole mouse brain. IEF-based peptide separation is an efficient method for peptide and phosphopeptide identification.
Key words Isoelectric focusing, Immobilized pH gradient, Mass spectrometry, Peptides,
Phosphopeptides, Proteome profiling, Phosphoproteome, Synaptosomes, Mouse brain
Abbreviations
CID
IEF
IPG
LC-ESI-MS/MS
MS
pI
S
T
Y
Collision-induced dissociation
Isoelectric focusing
Immobilized pH gradient
Liquid chromatography-electrospray ionization-tandem mass spectrometry
Mass spectrometry
Isoelectric point
Serine
Threonine
Tyrosine
293
294
Introduction
Mass spectrometry (MS)-based proteome profiling provides a
protein map for the biological material under investigation, facilitating functional and quantitative analyses. Phosphorylation is a
post-translational modification that occurs in approximately onethird of all proteins and is involved in many biological processes in
health and disease [1].
For proteome and phosphoproteome analysis, effective fractionation methods prior to MS are crucial to ensure high proteome
coverage through peptide and phosphopeptide identification.
Here, we describe how isoelectric focusing (IEF), which has been
traditionally used to separate proteins in the first dimension of
2D-gel electrophoresis, can be used as an effective fractionation
method both for peptides and phosphopeptides. Peptide mixtures
are loaded onto an immobilized pH gradient (IPG) strip and upon
application of electric current, peptides are separated according to
their isoelectric point (pI). The unique characteristic of IEF compared to other fractionation methods is the ability to simultaneously
analyze peptides and phosphopeptides. This is due to the addition
of a negatively charged phosphogroup to a peptide sequence which
results in a decrease of its pI. As a consequence, phosphopeptides
migrate toward the acidic part of the IPG strip [2].
MS analysis of the whole and the acidic part of the IPG strip
allows the investigation of the proteome and the phosphoproteome, respectively, with no additional step for phosphopeptide
enrichment. The pI focusing position of each peptide on the IPG
strip after IEF can be used to confirm subsequent MS-based peptide identifications and/or validate the presence of post-translational
modifications [35]. When applied to brain tissue, IEF has been
shown to result in increased proteome coverage compared to
1D-SDS gel electrophoresis [6]. In this chapter, we provide a
detailed guide to proteome and phosphoproteome analysis by
IEF. As study material we use synaptosomes, which are artificially
isolated synapses [7], extracted from whole mouse brain. We chose
brain synaptosomes for two reasons; (i) their proteome is highly
complex (ii) phosphorylation plays a critical role in the regulation
of synaptic function and neurotransmission [8]. In this protocol,
we describe the following steps: (1) synaptosome enrichment from
brain tissue and synaptosome enrichment quality control by
Western blot, (2) sample preparation for IEF and IEF, (3) MS, (4)
MS raw data analysis for peptide and phosphopeptide identification
and assignment of phosphorylation sites. The experimental workflow is shown in Fig. 1. Mouse synaptosome analysis by IEF
resulted in the identification of up to 3,000 proteins and 118 phosphoproteins [6].
IEF Peptide and Phosphopeptide Analysis
295
Brain tissue
Synaptosome enrichment
quality control
Synaptosome extraction
Reduction &
carboxymethylation
In-gel digestion
IEF
3.5
4.5
Peptide extraction
Hexane oil extraction
Extracted/cleaned peptides
LC ESI-MS/MS
Database search
Peptide/Phosphopeptide identification
Proteome profiling
Phosphoproteome profiling
Manual MS spectra validation
Phosphosite assignment
Fig. 1 Experimental set-up of IEF-based analysis of proteome and phosphoproteome from mouse brain
synaptosomes
296
Materials
2.1 Synaptosome
Enrichment and
Western Blot for
Enrichment Quality
Control
2.1.1 Synaptosome
Enrichment
1. Homogenization equipment: Branson sonifier 250 (G.

Heinemann, Schwaebisch Gmuend, Germany).
2. Centrifugation equipment: 5804R centrifuge (Eppendorf) and
L8-70M ultracentrifuge (Beckman Coulter).
3. Complete protease cocktail inhibitor tablets (Roche Diagnostics).
4. Phosphatase inhibitor cocktail II and III (Sigma-Aldrich).
5. 0.32 M sucrose: 2.74 g sucrose in 25 ml distilled water.
8. Buffer A: 0.32 M sucrose, 4 mM HEPES, complete protease
cocktail inhibitor tablets (added according to the manufacturers instructions), pH = 7.4. Adjust pH by adding 1 M NaOH.
2.1.2 Western Blot
1. Electrophoretic equipment: Mini-PROTEAN Tetra Cell

(Bio-Rad).
2. Bradforf protein assay dye reagent concentrate (Biorad).
3. Immobilon PVDF membrane (Millipore).
4. ECL Plus reagent kit (GE Healthcare).
5. TBS-T buffer: 137 nM NaCl, 20 mM TrisHCl, pH = 8.0,
0.05 % Tween 20.
6. Blocking buffer: 5 % carnation instant non-fat dry milk in
TBS-T buffer.
2.2
IEF
1. IEF equipment: Protean IEF Cell (Bio-Rad).

2. Lyophilization equipment: Savant Speed Vac plus SC210A
concentrator (Thermo Fisher Scientific).
3. 200 mM ammonium bicarbonate, 8 M urea, pH = 8.5.
4. Ammonium bicarbonate.
5. 10 mM dithiothreitol.
6. 50 mM iodoacetamide.
7. Lys-C (Wako).
8. Trypsin (Promega).
9. IPG strips (pH = 3.54.5, 18 cm) (GE Healthcare) (see Note 1).
10. Hexane.
11. Mineral oil.
12. OMIX tips (Varian).
13. Vivaspin 5 kDa cartridge (Vivascience).
2.3
ESI-LC-MS/MS
297
1. Nanoflow HPLC-2D system (Eksigent) including a trap column Zorbax SB300, 5 mm 0.3 mm, packing material 5 m
RP-C18 (Agilent Technologies).
2. LTQ Orbitrap XL mass spectrometer (Thermo Fisher).
3. Nano electrospray ionization source (Thermo Fisher).
4. Picofrit self-packed column 75 m 15 cm i.d., 10 m tip
(New Objective), packing material 3 m RP-C18.
5. Acetonitrile.
6. Formic acid.
2.4 MS Data
Analysis Software
1. LC-ESI-MS/MS data acquisition software: Xcalibur v. 2.07

(Thermo Fisher).
2. Nano-LC controlling software: LC-Eksigent v.2.09 (Eksigent).
3. Raw MS and MS/MS spectra processing software: BioWorks
3.3.1 (Thermo Fisher).
4. Protein database search engine: BioWorks 3.3.1 and SEQUEST
v.28 software (Thermo Fisher).
Methods
3.1 Synaptosome
Enrichment
and Western Blot
for Enrichment Quality
Control
3.1.1 Synaptosome
Enrichment
Synaptosome enrichment is performed according to [9] with slight

modifications. For optimal results, perfused, snap frozen mouse
brain tissue is recommended (see Note 2).
1. Tissue (mg) is homogenized in 10 volumes (l) of Buffer A by
1215 up and down strokes. Phosphatase inhibitors II and III
are added to Buffer A at a 1:100 v/v ratio each.
2. Homogenates are centrifuged at 1,000 g for 10 min, 4 C
(S1: supernatant 1; P1: pellet 1 containing nuclei and cell
debris).
3. S1 is transferred to a different tube and P1 is resuspended in 10
volumes (l) of Buffer A (phosphatase inhibitors optional) and
centrifuged at 1,000 g for 10 min, 4 C (S2: supernatant 2;
P2: pellet 2). S1 and S2 fractions are combined into fraction S.
4. S is centrifuged at 17,000 g for 55 min, 4 C (S3: supernatant
3; P3: pellet 3 containing membrane fraction with intact
synaptosomes).
5. P3 is resuspended in 0.32 M sucrose and laid on top of a discontinuous sucrose density gradient (1 ml 0.32 M/1 ml
0.8 M/1 ml 1.2 M sucrose) (phosphatase inhibitors optional).
6. Sucrose gradient with P3 is ultracentrifuged at 100,000 g for
2 h, 4 C.
298
7. Synaptosomal fraction is extracted from the 0.8 M/1.2 M

sucrose interphase (see Note 3), diluted at a 1:1 v/v ratio in
distilled water and ultracentrifuged at 164,000 g for 60 min,
4 C (S4, P4).
8. P4 is dissolved in 20 l distilled water and stored at 20 C
(see Note 4).
3.1.2 Western Blot
1. Protein content is estimated by Bradford protein assay.

2. An aliquot of each synaptosomal fraction is loaded onto a
12.5 % SDS polyacrylamide gel.
3. Electrotransfer of protein extracts onto an Immobilon PVDF
membrane is performed for 1 h at 100 V. Membranes are
blocked overnight with blocking buffer, followed by incubation with the selected primary antibody for 1.52 h at room
temperature, washing twice with TBS-T and 1 h incubation at
room temperature with the corresponding secondary antibody
(see Note 5).
4. Immune complexes are detected by ECL Plus reagent kit.
3.2
IEF
1. Synaptosomal proteins in distilled water are dissolved to a final

concentration of 200 mM ammonium bicarbonate, 8 M urea,
pH 8.5 (final volume of 100 l) (see Note 6).
2. Sample is reduced with 10 mM dithiothreitol for 1 h at 37 C
and then alkylated in the dark by 50 mM iodoacetamide for
30 min at room temperature.
3. Urea concentration is reduced to 2 M by using a Vivaspin
5 kDa cartridge according to the manufacturers instructions. The sample is washed three times with 200 mM
ammonium bicarbonate, pH 8.5, and concentrated to a
final sample volume of 100 l. The sample is diluted by the
addition of 100 l distilled water in a total volume of 200 l
(see Note 7).
4. The sample is digested with 24 g endoproteinase Lys-C overnight at room temperature followed by overnight digestion
with 24 g trypsin at 37 C (see Note 8).
5. Distilled water and urea are added to a final volume of 300 l
and a concentration of 2.5 M, respectively.
6. The sample is loaded on the IPG strip, left to be absorbed for
1 h at room temperature and then covered with mineral oil.
7. The rehydration runs for 12 h at constant voltage (50 V).
8. After rehydration is completed, electrode wicks wet with
deionized water are placed at the basic and acidic ends of the
IPG strip.
9. IEF is performed following the voltage/current steps according to the manufacturers instructions for the IPG strip used
(see Note 9).
299
10. After IEF completion, the IPG strip is removed from the tray
and mineral oil is drained by vertically holding the IPG strip.
11. The IPG strip is cut in 4 mm pieces. Each gel piece is transferred in a sample tube prefilled with 100 l 5 % formic acid.
12. Peptides from each gel piece are extracted with 50 l 5 % formic acid three times (15 min vortexing and 15 min sonication)
and the three supernatants are combined. To remove the residual mineral oil from the combined supernatant, the peptide
extracts are overlaid with hexane, mixed briefly and the upper
part of the organic solvent is discarded. The aqueous phase is
kept and the oil extraction with hexane (50 l) is repeated
three times. After oil extraction, the aqueous phase is kept and
lyophilized (see Note 10).
13. The resulting pellet is oil extracted with 10 l hexane and left
to dry overnight. The following day, pellets are dissolved in
25 l 5 % formic acid, desalted with OMIX tips according to
the manufacturers instructions and lyophilized (see Note 11).
3.3
ESI-LC-MS/MS
1. Lyophilized peptides from each IPG strip fraction are dissolved

in 6 l 1 % formic acid.
2. LC-ESI-MS/MS analysis is performed with a nanoflow
HPLC-2D system coupled online to an LTQ-Orbitrap mass
spectrometer via a nano ESI. The mass spectrometer is operated in the positive ion mode data-dependent scan acquisition
using Xcalibur. Full scans are recorded in the Orbitrap mass
analyzer (resolution-FWHM 60000) at a mass/charge (m/z)
range of 3801,600 in profile mode. The MS/MS analysis of
the five most intense peptide ions per scan is recorded in the
LTQ mass analyzer in centroid mode (top five method).
Additional MS conditions: spray voltage 1.92.1 kV; normalized collision energy 35 %; dynamic exclusion 120 s; activation
q = 0.25 and activation time 30 ms. Detailed MS conditions are
provided in [6].
3. From each IPG strip fraction, 3 l are loaded onto an in-house
packed column and analyzed with a 2.5 h gradient (washing
with 0.1 % formic acid for 20 min and elution with 95 % acetonitrile/0.1 % formic acid from 2 to 45 % at a flow rate of
200 nl/min).
3.4
Data Analysis
3.4.1 Proteome Profiling
1. MS raw files are searched against a decoy mouse database utilizing BioWorks and SEQUEST. Search parameters are as follows: peptide mass tolerance, 20 ppm; fragment ion mass
tolerance, 1 Da; enzyme, trypsin; missed cleavage sites, up to
two; only tryptic peptides allowed; static modification, cysteine
carboxyamidomethylation; variable modification, methionine
oxidation.
300
2. Protein and peptide identifications are filtered using the

following parameters: Delta CN value (cross correlation normalized) is 0.08; X correlation values versus ion charge state
1.9 (+1), 2.7 (+2), 3.5 (+3). Minimum distinct peptides per
protein: 2.
3.4.2 Phosphoprotein
Analysis
1. For phosphoprotein identification, steps 1 and 2 of

Subheading 3.4.1 are performed by including in the search
parameters serine (S), threonine (T) and tyrosine (Y) phosphorylation as variable modifications (see Note 12).
2. The algorithm used for database search provides a probability
score which is indicative of the reliability of the phosphopeptide identification.
3. Besides the probability score, MS/MS spectra should be manually interrogated to confirm the identification of a phosphorylation site in a peptide sequence. Phosphopeptide hits
resulting from the protein database search are manually
checked for the 98, 49, or 33 amu mass-shifted peaks for single, double, or triple charged ions, respectively. These m/z signals result from the neutral loss of phosphoric acid (H3PO4)
from the precursor and respective b- and y-type fragment ions
during the collision-induced dissociation (CID)-MS/MS fragmentation process. Due to the low stoichiometry of phosphorylation, spectra with a 98, 49, or 33 amu mass-shift loss signal
of the precursor ion that is greater than 0.4 % of the base peak
are considered valid phosphopeptide hits (see Note 13).
4. In case of multiple potential phosphorylation sites for a given
peptide, phosphorylation site assignment depends exclusively
on the observation of neutral loss of H3PO4 (98 Da) in the
sequence of b- or y-type ions in one of S, T, or Y residues that
could be phosphorylated (see Note 14). An example of manual
phosphosite assignment is shown in Fig. 2.
Notes
1. There are different options for IPG lengths and pH ranges
according to the application of interest. For high amounts of
starting material to be loaded, longer IPG strips are more
appropriate. For phosphorylation studies, IPG strips with a pH
range toward more acidic values are recommended [2, 10].
2. A minimum amount of approx. 30 mg brain tissue is required
to obtain a synaptosomal fraction. When studying brain regions
with lower tissue weight, pooling is necessary to study the synaptosomal proteome.
3. The crude synaptosome fraction at the interface of the
1.2 M/0.8 M sucrose gradient appears as a white cloud.
301

y*6
EGDGSATTDAAPATSPK
680.39
100
95
90
85
80
Relative Abundance
75
70
65
60
55
50
45
40
30
25
20
313.12
15
10
5
y6-98
y3-98
y*3
y*
b15-98
y*10
y*7
35
y9*
751.33
y*11
1038.55
937.44
y5*
583.24
* +2
b*11
15
976.37
706.76
411.12
1245.71
y11-98
1041.50
1314.58
b14
1139.57
y12
*
y13
y*14
b*15
1354.58
1412.68
y*15
1210.71 1297.56
1470.11
244.17
300
400
500
600
700
800
900
1000
1100
1200
1300
1400
1500
1600
m/z
Fig. 2 ESI-MSMS spectrum of Neuromodulin phosphorylated peptide EDGSATTDAAPATSPK. The single- and
double-charged b- and y-fragment ions identified by SEQUEST algorithm are shown. The phosphorylation site
at S15 is unambiguously determined by the loss of the phosphoric acid (H3PO4) from the y3 ion (y3-98) and by
the absence of the phosphoric acid loss peak from the b14 ion fragment. Phosphorylation at S15 is ascertained
by multiple fragment ions, e.g. y3-98, y6-98, y11-98, b15-98. The phosphorylation site is underlined in the peptide sequence. b- and y-fragment ions are highlighted in blue and red, respectively. Peaks marked by asterisk
(*) denote phosphorylated fragments
Absorption of the interphase can be performed by a syringe

and is the step that introduces the highest variability in the
synaptosome enrichment protocol.
4. The sample solubility can be increased by adding some drops
of 1 M NaOH.
5. For synaptosome enrichment quality control, specific antibodies
for pre- and post- synaptic proteins can be used for assessing
the enrichment of both pre- and post-synaptic fractions. As a
negative control, a nuclear-specific antibody can be used
(nuclei should not be present in the synaptosomal fraction).
As positive controls commercially available synaptosome preparations can be used. For optimal monitoring of the enrichment
procedure aliquots of all centrifugation steps can be loaded
and analyzed. For a representative example see [6].
6. The protein amount loaded on the IPG strip depends on its
length and pI range as well as sample type. We recommend
following the guidelines of the IPG strip manufacturer. For
302
synaptosome protein extracts we load 300 and 200 g protein

onto IPG strips, 17 and 11 cm (pH 3.54.5), respectively.
7. Urea concentration should be decreased in order to avoid the
inhibition of trypsin activity. The maximum total volume that
can be loaded on the IPG strip is provided in the guidelines of
the IPG strip manufacturer. The dilution volume should be
accordingly calculated.
8. The use of Lys-C and trypsin enhances the number of proteolytic peptides with Lys and Arg at the C-terminus for highly
complex protein mixtures. A combination of different restriction enzymes can be used according to sample and application
requirements.
9. The advantage of the automated program is that after IEF
completion the electric current is kept at a constant value of
50 A, thus maintaining the resolution of the pI-based peptide
separation.
10. The peptide extraction by formic acid is very efficient due to
the low percentage and thin layer of polyacrylamide in the IPG
strip. The removal of the mineral oil from the peptide extracts
is crucial to obtain high-quality MS data and avoid polymers.
Hexane extraction does not have a negative impact on peptide
extraction efficacy [2].
11. Desalting/filtering prior to mass spectrometry analysis is to
our experience required so as to eliminate salts and other ionsuppressor compounds which reduce the efficiency of the ionization process and result in low intensive peaks, compromising
the peptide/protein identification.
12. Under normal conditions, phosphorylation generally occurs
more often on S and T residues. Phosphorylation on Y residues
accounts for less than 1 % of the O-phosphorylated residues in
a protein sequence.
13. As in this protocol no phosphoprotein enrichment step precedes phosphopeptide analysis, the number of identified phosphopeptides is lower compared to phosphoproteome-targeted
analyses due to the low stoichiometry of phosphorylation in a
peptide mixture. Strategies exclusively directed to phosphopeptide analysis may include a phosphoprotein enrichment
step prior to IEF, as previously described [11].
14. It should be noted that the unambiguous assignment of phosphorylation sites in phosphopeptides with multiple potential
phosphorylation sites can be hindered by gas-phase phosphate
group rearrangement reactions occurring under CID-MS/MS
conditions [12].
303
Acknowledgments
This work is funded by the Max Planck Society. M.D.F. is supported by a grant from the Deutsche Forschungsgemeinschaft (FI
1895/1-1). We thank Chris Turck for useful comments.
References
1. Cohen P (2001) The role of protein phosphorylation in human health and disease. The Sir
Hans Krebs Medal Lecture. Eur J Biochem
268:50015010
2. Maccarrone G, Kolb N, Teplytska L, Birg I,
Zollinger R, Holsboer F, Turck CW (2006)
Phosphopeptide enrichment by IEF. Electrophoresis 27:45854595
3. Cargile BJ, Bundy JL, Freeman TW, Stephenson
JL Jr (2004) Gel based isoelectric focusing of
peptides and the utility of isoelectric point in protein identification. J Proteome Res 3:112119
4. Uwaje NC, Mueller NS, Maccarrone G, Turck
CW (2007) Interrogation of MS/MS search
data with an pI Filter algorithm to increase
protein identification success. Electrophoresis
28:18671874
5. Xie H, Bandhakavi S, Roe MR, Griffin TJ
(2007) Preparative peptide isoelectric focusing
as a tool for improving the identification of
lysine-acetylated peptides from complex mixtures. J Proteome Res 6:20192026
6. Filiou MD, Bisle B, Reckow S, Teplytska L,
Maccarrone G, Turck CW (2010) Profiling
of mouse synaptosome proteome and phosphoproteome by IEF. Electrophoresis 31:
12941301
7. Schrimpf SP, Meskenaite V, Brunner E,
Rutishauser D, Walther P, Eng J, Aebersold R,
8.
9.
10.
11.
12.
Sonderegger P (2005) Proteomic analysis of

synaptosomes using isotope-coded affinity
tags and mass spectrometry. Proteomics 5:
25312541
Smart TG (1997) Regulation of excitatory and
inhibitory neurotransmitter-gated ion channels
by protein phosphorylation. Curr Opin
Neurobiol 7:358367
Gray EG, Whittaker VP (1962) The isolation
of nerve endings from brain: an electronmicroscopic study of cell fragments derived by
homogenization and centrifugation. J Anat 96:
7988
Cargile BJ, Talley DL, Stephenson JL Jr
(2004) Immobilized pH gradients as a first
dimension in shotgun proteomics and analysis
of the accuracy of pI predictability of peptides.
Beranova-Giorgianni S, Desiderio DM,
Giorgianni F (2009) Phosphoproteome
analysis by in-gel isoelectric focusing and
tandem mass spectrometry. Methods Mol
Biol 519:383396
Palumbo AM, Reid GE (2008) Evaluation of
gas-phase rearrangement and competing fragmentation reactions on protein phosphorylation site assignment using collision induced
dissociation-MS/MS and MS3. Anal Chem
80:97359747
Chapter 23
Principles and Examples of Gel-Based Approaches
for Phosphoprotein Analysis
Abstract
Methods for analyzing the phosphorylation status of proteins are essential to investigate in detail key
cellular processes, including signal transduction and cell metabolism. The transience of this post-translational
modification and the generally low abundance of phosphoproteins require specific enrichment and/or
detection steps prior to analysis. Here, we describe three gel-based approaches for the analysis of differentially expressed phosphoproteins. These approaches comprise (1) the sequential fluorescence staining of
two-dimensional (2-D) gels using Pro-Q Diamond and SYPRO Ruby dyes to visualize and quantify
phosphoproteins in total cellular lysates as well as (2) affinity enrichment of phosphoproteins in conjunction with sequential fluorescence staining of the 2-D gels and (3) affinity enrichment of proteins prior to
pre-electrophoretic fluorescence labeling and 2-D gel electrophoresis.
Key words Phosphoproteins, 2-D gel electrophoresis, IMAC, Pro-Q Diamond, SYPRO Ruby, 2-D
fluorescence difference gel electrophoresis
Introduction
The reversible modification of proteins, in particular of serine, threonine,
and tyrosine residues, with one or more phosphate groups is a key
regulator of a multitude of cellular processes including growth and
differentiation [1]. Moreover, protein phosphorylation/dephosphorylation driven by specific kinases/phosphatases is involved in the
information transmission within a cell and is part of the communication system of cells with their environment [2]. Thus, the unambiguous characterization of the phosphorylation status of proteins in
response to a certain stimuli or environmental changes is crucial to
gain a better knowledge of cell signaling processes and in addition of
vital biological events. Various methods are available to characterize
and identify these post-translational modified proteins. Beside gelfree, mass spectrometry-based approaches that enable the study
of the modified amino acids on the peptide as well as on the protein level [3], different approaches based on two-dimensional gel
305
306
electrophoresis (2-D GE) have been successfully applied [2, 46].

This method has a high resolution capacity and enables the direct
visualization of phosphorylation changes of proteins. The attachment
of a phosphate group to a protein or the removal of it alters the net
charge of this protein and thus, leads to a shift in migration at the
isoelectric point level [2]. This shift can then be easily monitored by
2-D GE.
However, main difficulties of studying phosphoproteins arise
from the low stoichiometry of phosphorylation as well as from
their low abundance [7]. To overcome these problems, specific
detection strategies have been developed, e.g. fluorescence-based
gel stains, such as Pro-Q Diamond [8]. This non-covalent fluorescence stain selectively targets phosphoproteins in gels and the
signal intensity of each spot correlates with the number of phosphate groups attached to either serine, threonine or tyrosine residues of the specific protein [9].
To circumvent the limitations of the low abundance of phosphoproteins, various enrichment methods have been introduced as
well. One of these techniques is based on the use of antibodies
against specific phosphoamino acids for immunoprecipitation of
proteins [10]. Another frequently used method utilizes immobilized metal ion affinity chromatography (IMAC) [11]. Using
IMAC proteins that carry a phosphate group on any amino acid
are selectively bound by the metal ion affinity chromatography
resin [12] and thus, can further be separated from unphosphorylated proteins.
In addition to the ability to detect and identify these highly
biological relevant proteins, the study of the changes of the phosphorylation status of proteins is essential. For differential expression analysis, the applied approaches comprise often a sequence of
several detection and/or fractionation methods. The sequential
use of Pro-Q Diamond with a total fluorescence stain, such as
SYPRO Ruby, enables the measurement of the relative phosphorylation level by normalizing the phospho-specific fluorescence signal to the total fluorescence signal of each individual spot [13].
Consequently, this allows the assessment of the altered phosphorylation status of the individual protein [13]. Alternatively, twodimensional fluorescence difference gel electrophoresis (2-D
DIGE) in combination with Pro-Q Diamond [14] or preceded by
a prefractionation step has frequently been applied to detect
changes in the phosphorylation level of protein [6, 15].
As it is obvious, various gel-based approaches have been established and applied to analyze phosphoproteins. In this chapter, we
describe in detail the workflows of three different 2-D gel-based
approaches that enable both detection and quantitation of phosphoproteins (see Fig. 1). Beside 2-D separation of total cell lysates
combined with sequential staining, in particular Pro-Q Diamond
and SYPRO Ruby, affinity enrichment of proteins was used in
Gel-Based Phosphoprotein Analysis
307
Fig. 1 Overview of different gel-based approaches to analyse phosphoproteins. In the first approach (light
grey ), total cell lysates are directly separated by 2-D GE and the gels are sequentially post-stained using a
phosphospecific stain, visualizing the phosphoproteins content followed by a total fluorescence stain. The
second approach (grey ) consists of an enrichment step to purify the phosphoproteins using affinity chromatography and subsequent separation by 2-D GE and sequential post-staining. In the third described approach
(dark grey ), affinity-enriched proteins are labeled with fluorescence dyes for differential analysis prior to gel
electrophoresis. The sample preparation steps include sample clean up and the use of a protein concentration
assay. Copyright Wiley-VCH Verlag GmbH & Co. KGaA. Modified with permission from [16]
conjunction with either sequential staining of the gels (see Fig. 2)

or pre-electrophoretic fluorescence labeling of the purified phosphoproteins prior to 2-D GE (see Fig. 3). Finally, the proteins were
silver stained to enable manual spot picking for subsequent mass
spectrometric analysis. In a recent study, the applicability of these
approaches to analyze scarce samples, e.g. obtained from ex vivo
experiments, was compared [16]. This study showed that the
approach comprising affinity enrichment and fluorescence tagging
of the proteins was most suitable for the analysis of small amounts
of phosphoproteins and subsequent mass spectrometry-based
Fig. 2 Affinity chromatography of proteins combined with sequential post-staining

of 2-D gels to detect and quantitate phosphoproteins. A total of 62 g IMAC purified proteins were subjected to 2-D GE and stained with Pro-Q Diamond, followed
by SYPRO Ruby. Finally, the proteins were visualized with silver nitrate. Copyright
Wiley-VCH Verlag GmbH & Co. KGaA. Reproduced with permission from [16]
309
Fig. 3 Affinity chromatography of phosphoproteins combined with pre-electrophoretic fluorescence labeling.

30 g of IMAC-enriched phosphoproteins were labeled with Cy5 dye prior to 2-D GE. The fluorescent dye
tagged proteins were visualized by scanning using a Typhoon Trio scanner (GE Healthcare) (upper panel ). After
image acquisition the gel was stained with silver nitrate as described in Subheading 3.5.3 (lower panel ).
Copyright Wiley-VCH Verlag GmbH & Co. KGaA. Reproduced with permission from [16]
identification (see Fig. 3) [16]. Moreover, tagging of the IMACenriched proteins with fluorescent dyes can further be used for
quantitation by 2-D DIGE. To sum up, one should choose an
adjusted approach to analyze the phosphoproteome depending on
the available biological samples.
310
Materials
Prepare all buffers and solutions using ultrapure water (e.g.
obtained from a Milli-Q system, Millipore, Bedford, MA; MHQ
water, 18 M cm conductivity at 25 C). Store all solutions at
4 C, unless otherwise stated.
2.1 Cell Lysis

for Phosphoprotein
Analysis
1. Equipment: 1.5 mL microcentrifuge tubes, 1 mL single-use

syringe with disposable cannulas (0.55 25 mm, size 17;
B. Braun Melsungen; Melsungen, Hesse Germany), optical
microscope, vortex/mixer, cuvettes/spectrophotometer,
temperature-controlled centrifuge for 1.5 mL tubes (e.g.
Centrifuge 5810R, Eppendorf, Hamburg, Germany).
2. CHAPS solution: 10 % CHAPS (w/v) in MHQ water. Weigh
1 g CHAPS and transfer it to a 15 mL tube. Add water to a
volume of 10 mL and mix until dissolved.
3. Cell lysis buffer: Lysis/Binding/Wash buffer (included in the
Phosphoprotein Enrichment Kit, see Subheading 2.2), 1
EDTA-free protease inhibitor cocktail (Halt protease inhibitor single-use cocktail, Thermo Fisher), 1 phosphatase inhibitor cocktail (Halt phosphatase inhibitor cocktail, Thermo
Fisher), 0.25 % CHAPS (v/v) (see Note 1). To prepare 1 mL
of the buffer, mix 25 L 10 % CHAPS solution, 10 L 100
protease inhibitor cocktail and 10 L 100 phosphatase inhibitor cocktail and make up to 1 mL with Lysis/Binding/Wash
buffer (see Note 2).
4. Bicinchoninic acid protein assay Kit (BCA assay) (Thermo
Fisher) (see Note 3).
2.2 Affinity
Enrichment Using
IMAC Columns
1. Equipment: 1.5 mL microcentrifuge tubes, 15 and 50 mL

tubes, vortex/mixer, temperature-controlled centrifuge with
swinging-bucket or fixed-angle rotor, rated for 2,000 g
(e.g. Beckmann Coulter, Brea, CA, USA).
2. Phosphoprotein Enrichment Kit: #90003, Thermo Fisher
Scientific, Rockford, IL, USA.
3. CHAPS solution: 10 % CHAPS (w/v) in MHQ water. Weigh
1 g CHAPS and transfer it to a 15 mL tube. Add water to a
volume of 10 mL and mix until dissolved.
4. Binding/Wash buffer: Lysis/Binding/Wash buffer (included
in the Kit), 0.25 % CHAPS (v/v). Add 625 L 10 % CHAPS
solution to 25 mL with Lysis/Binding/Wash buffer.
5. Elution Buffer: elution buffer (included in the Kit), 0.25 %
CHAPS (v/v). Make up 125 L 10 % CHAPS solution to
5 mL with elution buffer (see Note 4).
2.3 Sample
Preparation (Protein
Precipitation
and Protein Labeling)
311
1. Equipment: 1.5 mL microcentrifuge tubes, vortex/mixer,

temperature-controlled centrifuge for 1.5 mL tubes.
2. 2-D Clean-up Kit (GE Healthcare, Uppsala, Sweden)
(see Note 5).
3. (Optional) Fluorescence labeling.
(a) 0.3 M TrisHCl solution, pH 8.5: 0.3 M Tris in MHQ
water. Dissolve 3.63 g Tris in 60 mL water. Adjust the pH
to 8.5 with HCl and fill up to 100 mL with water. Store at
room temperature.
(b) Labeling buffer: 4 M urea, 30 mM TrisHCl (pH 8.5),
1 % NP-40 (v/v). Weigh 4.8 g urea and mix with 2 mL
0.3 M TrisHCl (pH 8.5) and 2 mL 10 % NP-40 detergent solution and fill up to 20 mL with water. Aliquot and
store at 20 C.
(c) Cydye DIGE fluor minimal dyes (GE Healthcare)
(see Note 6).
(d) Dimethylformamide (DMF).
(e) Stop solution: 10 mM L-Lysine in MHQ water. Dissolve
0.0015 g L-Lysine in 1 mL water. Aliquot and store at
20 C.
2.4 Two-Dimensional
Gel Electrophoresis
(2-D GE)
1. Equipment: Isoelectric focusing apparatus compatible with

IPG strips (e.g. Protean IEF cell, Bio-Rad or similar), disposable rehydration trays (e.g. Bio-Rad), vortex/mixer; centrifuge
for 1.5 mL tubes, power supply, SDS-PAGE equipment for
second dimension (e.g. Criterion cell, Bio-Rad or similar),
shaking platform.
2. IPG (immobilized pH gradient) buffer (GE Healthcare):
pH 310 nonlinear.
3. Dithiothreitol (DTT) solution: 1 M DTT in MHQ water.
Weigh 1.54 g DTT and fill up to 10 mL with water. Aliquot
and store at 20 C.
4. 1.5 M TrisHCl solution, pH 8.8: 1.5 M Tris in MHQ water.
Dissolve 18.17 g Tris in 50 mL water. Adjust the pH to 8.8
with HCl and fill up to 100 mL with water.
5. Rehydration buffer: 7 M urea, 2 M thiourea, 4 % CHAPS (w/v),
20 mM DTT, 0.2 % IPG buffer (v/v). Weigh 8.41 g urea, 3.04 g
thiourea and 0.80 g CHAPS and fill up to 20 mL with water.
Aliquot and store at 20 C. Add 1 M DTT solution, IPG buffer and a trace of bromophenol blue just prior to use.
6. IPG strips for isoelectric focusing (IEF): e.g. ReadyStrip IPG
strips (Bio-Rad, Hercules, CA, USA), pH 47 or pH 310
nonlinear.
312
7. Equilibration buffer: 7 M urea, 2 M thiourea, 2 % SDS (w/v),

50 mM TrisHCl (pH 8.8), 30 % glycerol (85 % w/w). To make
200 mL, dissolve 84.10 g urea, 30.45 g thiourea and 4 g SDS in
50 mL water. Add 6.67 mL 1.5 M TrisHCl (pH 8.8) and
70.59 mL glycerol. Fill up to 200 mL with water (see Note 7).
Aliquot and store at 20 C.
8. Reduction equilibration buffer: 30 mM DTT in equilibration
buffer. Weigh 0.025 g DTT and make up to 5 mL with equilibration buffer. This solution should be prepared freshly just
before use.
9. Alkylation equilibration buffer: 180 mM iodoacetamide in
equilibration buffer. Weigh 0.165 g iodoacetamide, fill up to
5 mL with equilibration buffer and add a trace of bromophenol
blue. This solution should be prepared freshly just before use.
10. Precast gels for SDS-PAGE (e.g. Criterion precast gels 10.5
14 % TrisHCl, Bio-Rad).
11. Molecular weight marker: PeppermintStick Phosphoprotein
Molecular Weight Standards (Invitrogen, Carlsbad, CA,
USA). Store at 20 C.
12. Sealing solution: 1 % agarose (w/v) in MHQ water. Weigh 1 g
agarose and make up to 100 mL with water. Microwave the
solution to dissolve the agarose. Store at room temperature.
13. SDS-PAGE electrophoresis running buffer: 0.025 M Tris,
0.192 M glycine, 0.1 % SDS (w/v). Weigh 3.03 g Tris, 14.41 g
glycine and 1 g SDS and make up to 1 L with water (see Note 8).
Store at room temperature.
2.5 Sequential
Gel Staining
2.5.1 Pro-Q Diamond
Staining
Store all components at room temperature unless stated otherwise.
1. Equipment: gel trays (clean with 70 % ethanol before use),

shaking platform.
2. Pro-Q Diamond Phosphoprotein gel stain (Invitrogen).
3. Fixing solution: 10 % acetic acid (v/v), 50 % ethanol (v/v). For
1 L fix solution mix 400 mL water with 100 mL acetic acid and
add 500 mL ethanol.
4. Destain solution: 50 mM sodium acetate solution (pH 4), 20 %
acetonitrile (v/v). Prepare 1 M sodium acetate solution by dissolving 136 g sodium acetate trihydrate in 600 mL water and
adjust the pH to 4 with acetic acid. Make up to 1 L with water.
For 1 L destain solution mix 50 mL 1 M sodium acetate solution (pH 4) with 750 mL water and add 200 mL acetonitrile.
2.5.2 SYPRO Ruby

Staining

shaking platform.
2. SYPRO Ruby Protein gel stain (Invitrogen).
313
3. Wash solution: 7 % acetic acid (v/v), 10 % ethanol (v/v). For

1 L wash solution mix 830 mL water with 70 mL acetic acid
and add 100 mL ethanol.
2.5.3 Silver Staining
(According to [17])

shaking platform.
2. Sensitizing solution: 0.02 % (w/v) sodium thiosulfate in MHQ
water. Weigh 0.2 g sodium thiosulfate and add water up to 1 L.
3. Silver nitrate solution: 0.2 % (w/v) silver nitrate, 0.02 % (v/v)
formaldehyde (37 %). Weigh 2 g silver nitrate and make up to
1 L with water. Finally, add 200 l formaldehyde (37 %).
Protect the silver nitrate solution from light.
4. Development solution: 3 % (w/v) sodium carbonate, 0.05 %
(v/v) formaldehyde (37 %), 0.0005 % (w/v) sodium thiosulfate. Weigh 30 g sodium carbonate and 5 mg sodium thiosulfate and transfer to an empty flask. Swirl and add water up to
1 L. Finally, add 500 L formaldehyde (37 %).
5. Stop solution: 5 % acetic acid (v/v) in MHQ water. For 1 L
stop solution add 50 mL acetic acid to 950 mL water.
6. Storage solution: 1 % acetic acid (v/v) in MHQ water. For 1 L
store solution add 10 mL acetic acid to 990 mL water.
2.6 Image
Acquisition
and Analysis
1. Equipment: Fluorescence image capture device (e.g.

VersaDoc 4000MP Imaging System (Bio-Rad) or Typhoon
Trio instrument (GE Healthcare)), Image software (e.g.
PDQuest advanced version 8.0.1, Bio-Rad).
Methods
3.1 Cell Lysis

for Phosphoprotein
Analysis
All steps should be performed on ice.

1. Add 500 L cell lysis buffer per 50 mg wet weight pellet of washed
cells (see Note 9). Disperse the cell pellet as following: pipette up
and down at least ten times, vortex for approximately 30 s and
homogenize by passage through a 17-G needle ten times.
2. Leave the cells on ice for 15 min.
3. Check the cell lysis process using a light microscope. In case of
insufficient lysis add more buffer and prolong the incubation
on ice.
4. Remove insoluble material by centrifugation at 10,000 g for
20 min at 4 C. Transfer the supernatant to a new tube.
5. Store the cell lysate at 80 C for later use or continue with the
next step.
6. Determine the protein concentration of the samples using the
BCA assay according to the manufacturers protocol (see Notes
10 and 11).
314
3.2 Affinity
Enrichment Using
IMAC Columns
The protocol is based on the manufacturers protocol but has been

further adapted for low sample amounts. A second enrichment
step is included to increase the phosphoprotein yield.
1. Adjust 1 mg total protein to 0.25 mg/mL with the Binding/
Wash buffer. Store on ice.
2. Upend the phosphoprotein enrichment column three times.
3. Remove the bottom part of the column and loosen the cap
(see Note 12).
4. Place the column in a 50 mL tube and centrifuge at 1,000 g
for 1 min at 4 C. Discard the flow through.
5. Pipette 5 mL Binding/Wash buffer onto the column.
6. Place the column again in the 50 mL conical tube and centrifuge at 1,000 g for 1 min at 4 C. Discard the flow through.
7. Cap the bottom part of the column and add the diluted lysate
of step 1. Screw the cap and upend three times.
8. Incubate the column on a platform rocker for 30 min at 4 C.
9. Remove the bottom plug and place the column in a new 50 mL
conical tube. Centrifuge the column at 1,000 g for 1 min at
4 C. The flow through is stored on ice.
10. Wash the column with 5 mL Binding/Wash buffer as described
in steps 56. Repeat twice for a total of three wash steps.
11. Cap the bottom of the column and transfer the column to a
new 50 mL conical tube. Add 1 mL elution buffer and incubate at room temperature for 3 min. Within these 3 min agitate the column three times.
12. Remove the plug, transfer the column to a new 50 mL conical
tube and centrifuge at 1,000 g for 1 min at 4 C. Repeat this
step four times.
13. Pool the eluted fractions (total volume 5 mL).
14. Use the flow through of step 9 and repeat steps 213 with a
new enrichment column, except that the flow through obtained
in step 9 is discarded.
15. Pipette the eluates from both columns separately into the
upper chamber of the iCON concentrator centrifugal devices.
16. Centrifuge at 3,000 g for approximately 45 min at 4 C in
a swinging-bucket rotor until the eluate of the first column
has a volume between 120 and 150 L and the eluate of the
second used column has a volume between 50 and 80 L
(see Note 13).
17. Pool the concentrated eluates and store at 80 C for later use
or continue with Subheading 3.3.
3.3 Sample
Preparation (Protein
Precipitation
and Protein Labeling)
315
1. Use the 2-D Clean-up Kit according to the manufacturers

protocol (see Note 14).
2. Redissolve your sample in an appropriate buffer (e.g. labeling
buffer, pH 8.5, as described in Subheading 2.3).
3. Determine the protein concentration of your redissolved
sample.
4. (Optional) Pre-electrophoretic fluorescence labeling.
(a) Check the pH of your sample with a pH indicator strip by
pipetting 0.5 L of your sample onto the indicator strip
(see Note 15).
(b) Reconstitute the fluorescent dyes in DMF to generate a
working solution according to the manufacturers protocol (see Note 16).
(c) Add 4 pmol/g protein of reconstituted CyDye minimal
dye to your sample. Vortex for 5 s and centrifuge at maximum speed for 10 s.
(d) Put the samples on ice for 30 min in the dark.
(e) Add 1 L stop solution to terminate the labeling process.
Vortex for 5 s and centrifuge at maximum speed for 10 s.
(f ) Put the sample on ice for 10 min in the dark.
(g) The samples can be stored at 80 C or can be further
subjected to 2-D gel electrophoresis (2-D GE).
3.4 Two-Dimensional
Gel Electrophoresis
(2-D GE)
1. Mix your samples with the appropriate amount of rehydration

buffer (see Note 17).
2. Leave the samples for 10 min at room temperature.
3. Centrifuge the sample at maximum speed for 2 min at room
temperature to remove any insoluble materials and bubbles.
4. Pipette the samples into the designated lanes of the IPG focusing tray without introducing any bubbles.
5. Peel apart the IPG strip without touching the gel side (starting
from the acidic end to the other end). Place each IPG strip
with the gel side down into the designated sample lane of the
tray (see Note 18).
6. Cover each strip with 2 mL mineral oil to prevent the dehydration of the gel. Put the lid onto the tray.
7. Perform passive rehydration for 12 h at 20 C and focus the
proteins as recommended for each strip type by the manufacturer (see Note 19).
8. Immediately remove the strips from the focusing tray by holding one end of the strip with forceps. Allow the excessive mineral oil to drip off on an absorbent paper.
316
9. At this point the strips can either be stored in, for example, a
15 mL tube or a disposable rehydration tray at 80 C or
immediately continue with step 10.
10. Put the strips with the gel side up into a disposable rehydration
tray and cover each single strip with 2 mL reduction equilibration buffer. Incubate at room temperature for 19 min with
gentle agitation (see Note 20).
11. Immediately remove the strips and put it into a new disposable
rehydration tray. Cover each strip with 2 mL alkylation equilibration buffer. Incubate at room temperature for 19 min with
gentle agitation.
12. During step 11, melt the 1 % agarose solution (sealing solution) and store at 70 C till use.
13. Remove the strips from the rehydration tray by holding one
side of the strip with forceps and carefully align each strip onto
the top of a second-dimension SDS polyacrylamide gel (for
example: CriterionTM, Bio Rad) without any air bubbles
between the strip and the gel.
14. Load the molecular weight standard into the designated well
(see Note 21).
15. Quickly overlay the strip with 12 mL sealing solution. Avoid
introducing air bubbles. Let the sealing solution solidify for
approximately 2 min.
16. Separate the proteins in 1 SDS running buffer.
17. Open the gel plates with the use of a spatula. Carefully transfer
the gels into a cleaned container, add some water to prevent
the gels from drying out and immediately continue either with
steps 18 or 19.
18. In case of pre-electrophoretic labeling, visualize the fluorescence images of the gels, as stated in Subheading 3.6, step 1.
19. In case of sequential post staining, handle the gels as described
in Subheading 3.5 (see Note 22).
3.5 Sequential Gel
Staining
3.5.1 Pro-Q Diamond
Staining
Carry out all steps at room temperature unless otherwise specified,

according to the manufacturers protocol (see Note 23). Use a
shaking platform.
1. Soak the 2D gels in fixation solution for 30 min. Discard the
fixation solution.
2. Incubate the gels in fixation solution overnight at 4 C. Discard
the fixation solution.
3. Wash the gels three times for 15 min with water. Change the
water in-between.
317
4. Add the Pro-Q Diamond solution and incubate for 2 h in the

dark. Discard the staining solution.
5. Destain the gels 30 min with the destain solution in the dark.
Repeat this step twice.
6. Wash the gels four times for 15 min with water in the dark.
7. Continue with image acquisition of the gels (as stated in
Subheading 3.6, step 1).
3.5.2 SYPRO Ruby
Staining
Carry out all steps at room temperature unless otherwise specified,

according to the manufacturers protocol (see Note 23). Use a
shaking platform.
1. Following image acquisition put the gels into new containers
and add the SYPRO Ruby solution. Incubate overnight in the
dark. Remove the SYPRO Ruby solution.
2. Rinse the gels with wash solution for 30 min in the dark.
Discard the wash solution.
3. Rinse the gels with water for 15 min twice in the dark. Discard
the washing solutions.
4. Continue with image acquisition (as stated in Subheading 3.6,
step 1).
3.5.3 Silver Staining
1. After image acquisition, proceed with the silver staining protocol to visualize the proteins (according to [17]).
2. Put the gels in a new tray and wash them in water for 30 min.
3. Incubate the gels for 1 min in sensitizing solution. Remove the
sensitizing solution.
4. Wash three times for 20 s with water. Remove the water after
each washing step (see Note 24).
5. Add the silver nitrate solution and incubate the gel for 20 min
at 4 C in the dark.
6. Repeat step 3.
7. Add development solution and remove it when the solution
turns slightly turbid. Add fresh development solution and
incubate the gel until it is sufficiently stained. Discard the
solution.
8. Wash with water for 20 s. Discard the water.
9. Incubate the gels in stop solution for 5 min. Remove the stop
solution.
10. Wash the gels for 10 min with water. Repeat this step twice.
Discard the water after each washing step.
11. The gels can now be stored in storage solution at 4 C.
318
3.6 Image
Acquisition
and Analysis
1. Visualize the fluorescence images using an appropriate instrument

(e.g. Typhoon Trio scanner, GE Healthcare or VersaDoc MP
4000 Imaging System, Bio-Rad) at an excitation/emission
wavelength corresponding to the used fluorescent stain/dyes
(see Note 25). Images can further be displayed and analyzed
by appropriate software (e.g. PDQuest advanced version 8.0.1,
Bio Rad).
2. Remove the gels from the imaging systems and continue with
the silver staining protocol (see Subheading 3.5.3) to visualize
the proteins for subsequent manual spot picking.
Notes
1. The use of phosphatase/protease inhibitor cocktails is essential
to prevent the degradation of proteins and the removal of
phosphate groups from proteins.
2. The volume of the lysis buffer is dependent on the wet weight
cell pellet. Prepare the lysis buffer volume according to the
weight of each cell pellet as well as to the number of samples.
Add protease and phosphatase inhibitors to the lysis buffer
freshly before use.
3. Other protein concentration assays can be used as well.
However, no single available protein assay is compatible with
all buffer components. Therefore, select an appropriate assay
method that is most compatible with your samples and the
buffer components (e.g. BCA Protein AssayReducing Agent
Compatible, Thermo Fisher).
4. For each column load you need approximately 25 mL Lysis/
Binding/Wash buffer containing 0.25 % CHAPS as well as
5 mL elution buffer containing 0.25 % CHAPS. Prepare the
volume according to the number of samples.
5. Other systems can be used as well, such as ReadyPrep 2-D
Cleanup Kit (Bio Rad).
6. Other commercial available fluorescent dyes can be used as
well, e.g. Refraction-2D G-dyes (NH DyeAGNOSTICS,
Halle, Saxony-Anhalt, Germany).
7. Prepare the equilibration buffer at least 3 h prior to use, considering that it takes the buffer components a while to dissolve
completely. To avoid carbamylation, do not heat urea containing buffers above 37 C. Furthermore, add only small amounts
of water at once, to prevent that the total volume of 200 mL
will not be exceeded.
8. It is more convenient to prepare a 10 running buffer: 0.25 M
Tris, 1.92 M glycine and 1 % SDS (w/v). Weigh 30.3 g Tris,
319
144 g glycine and 10 g SDS and make up to 1 L with water.

Dilute 100 mL of the 10 running buffer in 900 mL water.
9. Use a phosphate-free wash buffer and include EDTA-free
protease and phosphatase inhibitors. Other cell lysis buffers
(e.g. RIPA buffer; Radioimmunoprecipitation assay buffer)
supplemented with protease and phosphatase inhibitors can be
used as well. If performing affinity chromatography, check the
compatibility of the lysis buffer with the columns prior to use.
10. For highest accuracy use the same dilution buffer for standard
samples as for experimental samples.
11. Since the amount of sample is limited, we reduced the volume
for the test tube procedure. If performing the test tube procedure, please make sure that your photometer is usable for less
volumes.
12. To ensure an optimal flow through the column during the centrifugation step always loosen the cap of the column before the
centrifugation step.
13. In general, 2,5004,500 g can be set using a swinging bucket
rotor and 5,0007,000 g can be set for a fixed-angle rotor.
Accordingly, the centrifugation time has to be adapted. Always
check the sample volume during the centrifugation process.
14. The buffer used to redissolve the precipitated protein pellet
should be compatible with the next experimental steps including
protein concentration assay, fluorescence labeling, and 2-D GE.
15. Phosphoproteins are known to be acidic and in case the pH of
your sample solution is below 8 add 2-D compatible labeling
buffer pH 9.5.
16. The working solution is not stable and should be used
immediately.
17. The final volume depends on the used IPG strip size (cm) (e.g.
for 11 cm strips the final volume is 200 L). Follow the guidelines provided by the respective manufacturer.
18. Remove any air bubbles between the sample solution and the
strip by carefully tapping the IPG strip or remove the strip and
place it again, from one side to the other gradually.
19. The following focusing steps are used for 11 cm strips with
broad pH gradients (pH 47 or pH 310 nonlinear): Passive
rehydration is followed by four steps of pre-isoelectric focusing
with rapid increase up to 100 V for 30 min, 200 V for 30 min,
500 V for 30 min and 1,000 V for 30 min. Isoelectric focusing
is performed with linear increase up to 4,000 V over 1 h, followed by a rapid increase to 6,000 V. The focusing step is
stopped when a total of 23,000 Vh is reached.
320
20. Approximately 2 mL equilibration buffer are required if the

strips are equilibrated in a disposable rehydration tray (11 cm
strips). If the strips are in a 15 mL tube, then 10 mL equilibration buffer are required per strip (11 cm strips).
21. The PeppermintStick phosphoprotein molecular weight standard contains two phosphorylated (ovalbumin and bovine
-casein of 45.0 and 23.6 kDa, respectively) and four nonphosphorylated proteins (-galactosidase, BSA, avidin, and
lysozyme of 116.25, 66.2, 18.0, and 14.4 kDa, respectively).
This marker serves as control for the Pro-Q Diamond staining
process.
22. Each gel tray for the different post staining solutions has to be
carefully cleaned prior to use.
23. The required volume of each solution is dependent on the size
of the used tray. Use a container similar in size of the gel to
save the expensive staining solutions and add enough solution
volume to fully immerse the gels. Use the Pro-Q Diamond
stain solution only once.
24. The incubation time for the washing steps must be strictly
adhered to.
25. Images can be obtained with any fluorescence capture device.
Cy5 has an excitation/emission maxima of 649/670 nm.
Cy2 and Cy3 exhibit an excitation/emission maxima of
489/506 nm and 550/570 nm, respectively. Pro-Q Diamond
dye absorbs maximally at 555 nm and emits maximally at
580 nm. SYPRO Ruby dye absorbs maximally in the ultraviolet region and has also an excitation maximum at 450 nm.
Recommended filter settings for different instruments are
provided by the respective manufacturers.
Acknowledgements
This work was funded by Life Science Calls, Niedersterreichische
Forschungs-und BildungsgmbH (NFB).
References
1. Machida M, Kosako H, Shirakabe K et al
(2007) Purification of phosphoproteins by
immobilized metal affinity chromatography
and its application to phosphoproteome analysis. FEBS J 274:15761587
2. Raggiaschi R, Lorenzetto C, Diodato E et al
(2006) Detection of phosphorylation patterns
in rat cortical neurons by combining phosphatase treatment and DIGE technology.
Proteomics 6:748756
3. Artemenko KA, Bergstrom Lind S, Elfineh L et al

(2011) Optimization of immunoaffinity enrichment and detection: toward a comprehensive
characterization of the phosphotyrosine proteome
of K562 cells by liquid chromatography-mass
spectrometry. Analyst 136:19711978
4. Morandell S, Stasyk T, Grosstessner-Hain K
et al (2006) Phosphoproteomics strategies for
the functional analysis of signal transduction.

5. Kotlo K, Johnson KR, Grillon JM et al (2012)
Phosphoprotein abundance changes in hypertensive cardiac remodeling. J Proteomics 77:113
6. Eustace AJ, Dowling P, Henry M et al (2011)
2D-DIGE analysis of phospho-enriched fractions from dasatinib-treated melanoma cell
lines. J Proteomics 74:490501
7. Delom F, Chevet E (2006) Phosphoprotein
analysis: from proteins to proteomes. Proteome
Sci 4:15
8. Schulenberg B, Aggeler R, Beechem JM et al
(2003) Analysis of steady-state protein phosphorylation in mitochondria using a novel fluorescent phosphosensor dye. J Biol Chem
278:2725127255
9. Steinberg TH, Agnew BJ, Gee KR et al (2003)
Global quantitative phosphoprotein analysis
using multiplexed proteomics technology.
10. Bergstrom Lind S, Artemenko KA, Elfineh L
et al (2011) Toward a comprehensive characterization of the phosphotyrosine proteome.
Cell Signal 23:13871395
11. Andersson L, Porath J (1986) Isolation of phosphoproteins by immobilized metal (Fe3+) affinity
chromatography. Anal Biochem 154:250254
321
12. Talvas J, Obled A, Sayd T et al (2008) Phosphoproteomic approach to identify new targets of
leucine deprivation in muscle cells. Anal
Biochem 381:148150
13. Chen A, McEwen ML, Sun S et al (2010)
Proteomic and phosphoproteomic analyses of
the soluble fraction following acute spinal cord
contusion in rats. J Neurotrauma 27:263274
14. Stasyk T, Morandell S, Bakry R et al (2005)
Quantitative detection of phosphoproteins by
combination of two-dimensional difference gel
electrophoresis and phosphospecific fluorescent staining. Electrophoresis 26:28502854
15. Tang W, Deng Z, Oses-Prieto JA et al (2008)
Proteomics studies of brassinosteroid signal
transduction using prefractionation and twodimensional DIGE. Mol Cell Proteomics 7:
728738
16. Steinberger B, Besenfelder U, Brem G et al
(2013) Comparison of gel-based phosphoproteomic approaches to analyse scarce oviductal
epithelial cell samples. Proteomics 13:1216
17. Blum H, Beier H, Gross HJ (1987) Improved
silver staining of plant proteins, RNA and
DNA in polyacrylamide gels. Electrophoresis
8:9399
Chapter 24
Neutral Phosphate-Affinity SDS-PAGE System
for Profiling of Protein Phosphorylation
Emiko Kinoshita-Kikuta, Eiji Kinoshita, and Tohru Koike
Abstract
In this chapter, we describe a standard protocol for phosphate-affinity SDS-PAGE that uses a dizinc(II)
complex of the phosphate-binding molecule Phos-tag in conjunction with a neutral-pH gel system (Zn2+
Phos-tag SDS-PAGE) to detect shifts in the mobilities of phosphoproteins. A previous protocol for affinity
electrophoresis that uses polyacrylamide-bound Mn2+-Phos-tag and Laemmlis buffer system under conditions of alkaline pH has limitations in separating certain phosphoproteins. The current protocol provides
major improvements in separation and detection of various phosphorylated protein species. We here
introduce two neutral-pH gel systems buffered with BisTrisHCl and TrisAcOH, respectively, for Zn2+
Phos-tag SDS-PAGE, and we also discuss their characteristics on the basis of comparative studies on phosphorylation profiling of proteins with a wide range of molecular masses. Each analytical procedure, from the
beginning of gel preparation to the end of electrophoresis, requires 2.55 h with either buffer system.
Key words Phos-tag, Affinity electrophoresis, Protein phosphorylation, Phosphoproteomics, Neutral
SDS-PAGE
Introduction
1.1 General
Background
Phosphorylation of proteins is one of the most common forms of

post-translational modification (PTM) that occurs in cell signaling
in biological species, and it affects several key properties of proteins
involved in numerous cellular events [1]. In living cells, many proteins are continuously and dynamically phosphorylated and
dephosphorylated at specific amino acid residues under the regulation of complex signaling networks [2, 3], resulting in the production of a variety of phosphoproteins with various states of
phosphorylation. This reversible PTM is catalyzed by the opposing
activities of large families of protein kinase and phosphatase
enzymes. For example, the human genome encodes more than
500 protein kinases [4] and about 150 protein phosphatases [5].
These numbers reflect the importance and the complexity of protein phosphorylation. Abnormal phosphorylation resulting from an
323
324
Emiko Kinoshita-Kikuta et al.
imbalance in the enzymatic reactions of kinases and phosphatases

has been implicated in a wide range of human diseases, including
cancers [6] and neurodegenerative disorders [7]. Therefore, effective analytical strategies for quantitative and qualitative monitoring
of alterations in the phosphorylation states of certain proteins are
essential tools for studies on the proteome, particularly in relation
to the elucidation of the molecular origins of diseases and the
rational molecular design of drugs.
In recent years, the large-scale identification of phosphoproteins and their sites of phosphorylation has been made possible
through dramatic advances in mass spectrometry (MS)-based
methods of shotgun proteomics, coupled with improvements in
MS instrumentation and the development of better methods for
the enrichment of phosphopeptides through enzymatic digestion
[8]. MS-based techniques also provide several strategies for the
quantitative analysis of phosphorylation, particularly the quantitative determination of levels of phosphorylation of individual residues in peptide digests. Under certain biological conditions, even
a single protein can show heterogeneity in its phosphorylation status. In many cases, the protein can be reversibly phosphorylated at
several residues, resulting in the simultaneous presence of multiple
forms of the protein that are phosphorylated at different sites and
with different stoichiometry. These distinct phosphorylated species
can act either independently or synergistically, and their presence is
closely associated with specific cellular events [9]. To understand
the function of a given protein in detail, it is therefore important to
be able to determine its overall phosphorylation status, as well as
the states of phosphorylation of individual sites within the protein.
Because MS-based strategies permit the analysis of digested peptides rather than intact proteins, these techniques are not ideally
suited to the identification of the many individual phosphorylated
species that can be formed from a particular protein or its alternatively spliced isoforms (splice variants) (see Fig. 1c).
Two-dimensional gel electrophoresis (2-DE), involving isoelectric focusing (IEF) and sodium dodecyl sulfatepolyacrylamide
gel electrophoresis (SDS-PAGE), is a classical and powerful analytical strategy that can be used to separate and analyze complex
proteomic samples, such as cell or tissue lysates, by means of differences in the electrical charges (pI) and the apparent molecular
masses of the constituent proteins [8]. In contrast to MS-based
strategies, 2-DE permits the study of intact proteins, and the technique can be used to analyze the distinct protein species produced
by various PTMs such as phosphorylation, glycosylation, sulfation, or partial proteolysis. The various modified species appear as
individual distinct migration spots along the horizontal and/or
vertical axes on the 2-DE gel. By combining 2-DE with metabolic
labeling with a radioactive [32P]orthophosphate isotope [10] or
with gel staining by a Pro-Q Diamond phosphoprotein-specific
fluorescent dye [11], the stoichiometry of phosphorylation can be
Neutral Phos-Tag SDS-PAGE System
325
Fig. 1 Phosphate-affinity SDS-PAGE for the detection of shifts in mobilities of phosphoprotein species.
(a) Structure of acrylamide-bound Phos-tag ligand and scheme for reversible capture of a phosphomonoester
dianion (ROPO32) by polyacrylamide-bound Phos-tag. (b) Schematic representation of the principle of
phosphate-affinity SDS-PAGE. (c) Flow schemes summarizing strategies for the analysis of phosphorylated
proteins and peptides by phosphate-affinity 2-DE analysis and by MS analysis, respectively
readily determined from the intensities of the spots corresponding

to individual phosphorylated species formed from a particular
protein. The site of phosphorylation of the separated target can be
confirmed by means of Western blotting analysis using the appropriate site-specific anti-phosphoprotein antibody or by an MS-based
technique after in-gel digestion. However, it is difficult to separate
phosphorylated species that have similar pI values (i.e., proteins
that have the same numbers of phosphate groups, but in which
the phosphate groups are present at different phosphorylation
sites within the molecule) by means of the 2-DE procedure.
Therefore, the conventional analytical strategy has limitations in
respect of detailed monitoring of the phosphorylation status of
individual proteins.
1.2 Phos-tag
Technology
as a Novel Strategy for
the Analysis of Protein
Phosphorylation
We have developed the Phos-tag (phosphate-binding tag) technology

as a novel strategy for the analysis of protein phosphorylation (Phostag Consortium, http://www.phos-tag.com/english/index.html).
The Phos-tag technology utilizes an alkoxide-bridged dinuclear
metal complex of Phos-tag {1,3-bis[bis(pyridin-2-ylmethyl)amino]
propan-2-olato dizinc(II) complex} that selectively captures
326
phosphomonoester dianions (ROPO32) in aqueous solution

under conditions of neutral pH [12]. To date, technologies based
on derivatives of Phos-tag have contributed extensively in the
development of a range of practical procedures for phosphoproteomic research. Among the methods that have been developed
are immobilized zinc(II) ion-affinity chromatography for the
enrichment of phosphoproteins [1316] and phosphopeptides
[1520], surface plasmon resonance techniques for the detection
of phosphopeptides [2123], a wide variety of lab-on-a-chip techniques that use various array formats for high-throughput profiling
of protein phosphorylation [2427], and Western blotting techniques for the detection of phosphoproteins on protein-blotting
membranes [2629]. We have also shown that a novel type of
phosphate-affinity SDS-PAGE using the Phos-tag molecule (Phostag SDS-PAGE) can be used to detect shifts in the mobilities of
phosphoproteins in comparison with those of their nonphosphorylated counterparts [29, 30].
1.3 Existing Protocol
for Phos-tag SDSPAGE (Mn2+Phos-tag
SDS-PAGE)
with PolyacrylamideBound Mn2+Phos-tag
and Laemmlis Buffer
System
We synthesized a Phos-tag ligand with a pendent acrylamide moiety

(Fig. 1a) and we used this as a novel additive (comonomer) in a
separating gel for use in Laemmlis method [31]. The analytical
technique based on this Phos-tag gel is now widely used in lifescience laboratories for the separation and detection of phosphoproteins. The principle underlying our analytical strategy involves
monitoring changes in the electrophoretic mobilities of phosphorylated proteins that result from reversible trapping of their phosphate moieties by Phos-tag molecules immobilized in the gel; these
changes are independent of the nature of the individual phosphorylated amino acid residues (Fig. 1b). At first, we tried to use
polyacrylamide-bound Zn2+Phos-tag in SDS-PAGE. However,
the expected phosphate-selective shift in mobility was not observed
under the usual SDS-PAGE conditions adopted in Laemmlis procedure. It is likely that the conditions that exist during the electrophoresis, such as an abundance of SDS anions and alkaline buffers,
are not conducive to selective trapping of phosphate by the zinc(II)
complex. We therefore examined whether complexes of other metals (Mn2+, Cu2+, Co2+, Fe2+, and Ni2+) might function as phosphatetrapping molecules under the conditions present during
SDS-PAGE. We found that a polyacrylamide-bound manganese(II)
analog (Mn2+Phos-tag) is indeed capable of acting as a phosphateaffinity site in the Laemmli buffer system. In a separating gel
containing copolymerized acrylamide-bound Mn2+Phos-tag, the
degree of migration of a phosphoprotein is less than that of its
nonphosphorylated counterpart, because the tag molecules trap
the phosphoprotein reversibly during electrophoresis (see Fig. 1b).
On this basis, we developed a novel type of phosphate-affinity
SDS-PAGE for the separation of phosphoproteins from their corresponding nonphosphorylated analogs. We named this technique
327
Mn2+Phos-tag SDS-PAGE. By subsequent general colorimetric gel

staining or Western blotting with a phosphorylation-independent
antibody, it is possible to detect both phosphorylated and nonphosphorylated proteins simultaneously. This permits temporal
changes in the ratio of phosphorylated to nonphosphorylated
forms produced by the reactions of certain protein kinases and
phosphatases to be determined quantitatively by means of Mn2+
Phos-tag SDS-PAGE without the need for any special apparatus,
radioisotopes, or chemical labels. In addition, this affinity electrophoresis technique has the following major advantages.
1. Various phosphorylated forms of a single protein can be
detected as multiple migration bands on the gel [30].
2. Phosphoprotein species that contain identical numbers of phosphate groups but in which the phosphate groups are present at
different locations within the molecule can be separated [32].
3. Unstable His- and Asp-phosphorylated proteins involved in a
two-component signaling system can be detected simultaneously during their phosphotransfer reactions [32].
4. A phosphate-affinity 2-DE procedure coupled with other electrophoresis methods permits more-detailed analyses of the
phosphorylated congeners of a particular protein and its splice
variants (see Fig. 1c) [33, 34].
This technique is therefore appropriate for quantitative and
qualitative analyses of temporal and spatial variations of proteins
with different phosphorylation states in vitro or in vivo without the
need to fragment the phosphorylated proteins into peptides with
concomitant loss of significant information regarding the molecular weights of these protein species [3539].
Since it was first reported in 2006 [29], Mn2+Phos-tag SDSPAGE has been used by many research groups to determine the
phosphorylation status of various proteins [40]. Because of the
simplicity of the procedure, acrylamide-bound Phos-tag studies
can be performed at any laboratory that is equipped with the apparatus and reagents for normal SDS-PAGE. However, there remain
some problems in the separation of certain phosphoproteins by
Mn2+Phos-tag SDS-PAGE. For example, it was reported that no
up-shifted band of tau protein, a classical hallmark of Alzheimers
disease, was observed following treatment by a certain tyrosine
kinase [32], although the phosphorylation of a tyrosine moiety was
later confirmed by immunoblotting with an anti-phosphotyrosine
antibody. A similarly unfortunate result has been reported in the
analysis of the phosphorylation status of a human cardiac troponin
T [41], a diagnostic marker for myocardial infarction. Furthermore,
the Mn2+Phos-tag SDS-PAGE procedure does not permit the
detection of any shift in the mobility of pepsin, a standard phosphoprotein, in comparison with that of its dephosphorylated form.
328
Mn2+Phos-tag SDS-PAGE, therefore, has some limitations in the

analysis of protein phosphorylation. An additional disadvantage is
that Mn2+Phos-tag gels, which are cast in an alkaline buffer, are
unstable during long-term storage, and tend to break down into
acrylic acids after 12 months [42], causing changes in the pore
sizes of the gels that result in poor resolution in subsequent SDSPAGE studies.
1.4 An Improved
Protocol for Zn2+
Phos-tag SDS-PAGE
for Advanced Protein
Phosphorylation
Profiling
To overcome the limitations described above, we developed an

improved Phos-tag SDS-PAGE method for advanced phosphorylation profiling of proteins, and we named this new method Zn2+
Phos-tag SDS-PAGE [42]. Marked improvements in the separation
and detection of phosphoprotein species are achieved by using a
dizinc(II) complex of acrylamide-bound Phos-tag (see Fig. 1a) in
conjunction with a neutral-pH gel system buffered with 2-[bis(2hydroxyethyl)amino]-2-(hydroxymethyl)propane-1,3-diol (Bis
Tris) and hydrochloric acid (BisTrisHCl). The utility of the
newly modified method has been demonstrated by visualization of
novel up-shifted bands of typical phosphoproteins, such as pepsin,
recombinant human tau protein treated in vitro with various tyrosine kinases, or the intracellular signaling molecule -catenin.
Other research groups have also demonstrated the utility of this
system for advanced protein phosphorylation profiling [43, 44].
Furthermore, we have demonstrated that Zn2+Phos-tag gels cast
in a neutral buffer are much more stable during 6 months of storage than are the corresponding Mn2+Phos-tag gels. Moreover,
storage of the Zn2+Phos-tag gel does not require any special
knowhow or equipment. The gel, together with the casting glass
plates, is merely wrapped in a Saran wrap to protect it from drying
and then stored at room temperature under normal laboratory illumination until required. Alternatively, the gels can be stored at
4 C in a refrigerator. The improved shelf life of the gels makes the
Zn2+Phos-tag SDS-PAGE method particularly appealing for laboratory practice. In summary, we have developed a simple, convenient, and reliable protocol for phosphate-affinity SDS-PAGE that
uses a gel system that can be prepared and stored in house.
1.5 Alternative
Protocol for Zn2+
Phos-tag SDS-PAGE
for the Analysis
of Large
Phosphoproteins
with Molecular
Masses in Excess
of 200 kDa
The separation and detection of high-molecular-mass proteins

generally requires extremely porous polyacrylamide gels that contain less than 5 % (w/v) of polyacrylamide. We have previously
reported a procedure for Mn2+Phos-tag SDS-PAGE in which
highly porous gels strengthened by homogeneous addition of
0.5 % (w/v) agarose are used for the analysis of large phosphoproteins with molecular masses in excess of 200 kDa [37, 38].
However, in the case of the Zn2+Phos-tag SDS-PAGE procedure
using a neutral-pH gel system buffered with BisTrisHCl described
in Subheading 1.4, we found that problems occur in the preparation of highly porous gels, in that all the protein bands appear as
329
smeared images because the BisTris-buffered gels have inadequate

sieving properties when the concentration of polyacrylamide is less
than 4 % (w/v) [45]. Because BisTris is a weakly basic amine
(pKa = 6.5 at 20 C) that is partially protonated at neutral pH values, the free amine species (R3N) is converted into an amine radical
cation (R3N+) that can act as a radical quencher during polymerization of acrylamide in the presence of ammonium persulfate
[46]. Therefore, the abundant BisTris molecules interfere with
the formation of the polyacrylamide gel matrix, particularly at very
low concentrations of acrylamide.
To overcome this problem, we developed an alternative procedure for the analysis of large phosphoproteins by Zn2+Phos-tag
SDS-PAGE that uses a neutral-pH gel system buffered with Tris
and acetic acid (TrisAcOH) [45]. The primary amine Tris
(pKa = 8.2 at 20 C) is almost fully protonated at neutral pH values,
and the resulting primary ammonium species (RNH3+) should
cause less inhibition of the polymerization of acrylamide. In fact,
the use of the TrisAcOH buffer system (pH 7.0) permits the production of a separating gel that shows sieving properties at polyacrylamide concentrations of less than 4 % (w/v). The utility of our
newly adapted TrisAcOH system has been demonstrated by visualization of a novel up-shifted band of a high-molecular-mass protein, ataxia telangiectasia-mutated kinase (ATM; 350 kDa), on a
Zn2+Phos-tag gel strengthened with 0.5 % (w/v) agarose.
Therefore, depending on the target proteins, appropriate
selection of one of our two current procedures for Zn2+Phos-tag
SDS-PAGE using systems buffered with BisTrisHCl or Tris
AcOH, respectively, can provide a broad coverage and can give
detailed information for large numbers of phosphoproteins with a
wide range of molecular weights. The improvements in the detection of shifts in the mobilities of the proteins are the result of an
increase in the affinity of the phosphorylated targets for the
dizinc(II) complex of Phos-tag Acrylamide under conditions of
neutral pH.
1.6 Applications
of the Current
Zn2+Phos-tag
SDS-PAGE Protocol
Below, we describe a useful protocol that provides improvements

in the detection of phosphoproteins that the previous Mn2+Phostag SDS-PAGE protocol did not adequately detect. The resolving
power of the current Zn2+Phos-tag SDS-PAGE methodology
permits the determination of the precise phosphorylation status of
many phosphoproteins. For example, by applying the current
protocol to a typical standard phosphoprotein (commercially
available ovalbumin derived from chicken egg white), we succeeded in characterizing four distinct forms of the protein with
different serine phosphorylation states [47]; these states arise
from differences in the phosphorylation status of the Ser-68 and
Ser-344 residues. Quantitative analysis showed that the protein
consists of approximately 65 % of the diphosphorylated species,
330
30 % of the species monophosphorylated at Ser-68, 4 % of the

species monophosphorylated at Ser-344, and 1 % of the nonphosphorylated species. The pair of species with a single phosphate
group, one at Ser-68 and the other at Ser-344, appeared as two
different migration bands on the Zn2+Phos-tag SDS-PAGE gel.
The older Mn2+Phos-tag SDS-PAGE protocol did not permit this
complete profiling of the phosphorylation status of ovalbumin.
We have characterized the two neutral-pH gel systems, Bis
TrisHCl and TrisAcOH, described above, by reporting comparative studies on the separation of a wide range of proteins with
molecular masses from 10 to 350 kDa [45]. For 10200 kDa cellular proteins, the BisTrisHCl system shows a higher resolving
power than the TrisAcOH system in the 2-D fluorescencedifference gel electrophoresis (2-D DIGE) analysis of certain phosphoproteins such as histone H3 (15 kDa) or elongation factor 2
(95 kDa). Large differences were also observed in the onedimensional (1-D) migration patterns of phosphorylated species of
extracellular signal-regulated kinases 1 and 2 (ERK1/2,
44/42 kDa). These differences arose from changes in the phosphorylation status of the Thr-202 and Tyr-204 residues in the two
buffer systems at the same concentration of Zn2+Phos-tag. The
two buffer systems therefore show differences in their separation
power and related features in profiling of the phosphorylation
states of 10200 kDa proteins. In contrast, shifts in the mobilities
of multiply phosphorylated forms of the high-molecular mass protein ATM (350 kDa) could only be detected by using the Tris
AcOH system with a low-concentration polyacrylamide gel
strengthened with agarose. We have therefore identified protocols
based on the two Phos-tag SDS-PAGE systems that can be used
under neutral pH conditions, the choice of the particular system
being dependent on the nature of the target proteins, which can
have wide ranges of molecular weights.
1.7 Limitations
of the Current
Zn2+Phos-tag
SDS-PAGE Protocol
As in the earlier protocol for Mn2+Phos-tag SDS-PAGE [37, 38],

the agarose-containing gel in the TrisAcOH system for the separation analysis of large phosphoproteins frequently undergoes partial
melting as the result of resistive heating effects during electrophoresis and electroblotting. In the electroblotting procedure, in particular, melting of the gel causes it to adhere to the protein-blotting
poly(vinylidene difluoride) (PVDF) membrane, hampering its
removal from the membrane. To circumvent this problem, it is
necessary to adapt the protocol to conditions of a relatively lower
constant voltage of 3.5 V/cm of distance between the two electrodes in the electroblotting unit, and to extend the time for the
wet-tank electroblotting procedure to 16 h (overnight). With
regard to the detection of protein phosphorylation in complex celllysate samples, the ability to identify up-shifted phosphorylated
331
forms of a given protein by using the current protocol is entirely

dependent upon its immunoreactivity with the corresponding
antibody, as in the previous protocol. To determine the sites of
phosphorylation of up-shifted phosphorylated species, it is necessary to adopt a downstream procedure, such as immunoblotting
using high-quality site-specific anti-phosphoprotein antibodies or
MS analysis after in-gel digestion. If used in conjunction with highquality site-specific anti-phosphoprotein antibodies, the gel-based
electrophoretic separation procedure is capable of mapping in
great depth the sites with a low abundance of phosphorylation on
each of the phosphoprotein species that are formed in cell signaling. Furthermore, when used in conjunction with advanced
MS-based methods, the procedure permits the identification of
novel phosphoprotein species, providing a greater understanding
of the detailed properties of particular proteins involved in specific
cellular events (see Fig. 1c). However, this gel-based strategy
might have a disadvantage because of its low throughput. In addition, the current Zn2+Phos-tag SDS-PAGE protocol is sometimes
incapable of showing a shift in the mobility of a phosphoprotein
containing an imidazole-phosphate moiety, as produced by autophosphorylation of a certain bacterial histidine kinase under the
experimental conditions [48]. The fact that the zinc(II) complex
of Phos-tag does not bind readily with a phosphorylated imidazole
group under the conditions of neutral SDS-PAGE is a limitation
of the current protocol. If such an unfortunate case occurs, we
recommend that the experiments are repeated using Mn2+Phos-tag
SDS-PAGE; this protocol, which uses an alkaline gel buffer, should
solve the problem.
Materials (See Note 1)
2.1 Sample
Preparation
1. Standard phosphoproteins: Bovine milk -casein, bovine milk

-casein, chicken egg ovalbumin, porcine pepsin.
2.1.1 Dephosphorylation
of Standard
Phosphoproteins
2. Alkaline phosphatase (AP): Bovine intestinal mucosa AP.

3. Dephosphorylation reaction buffer: 50 mM TrisHCl, pH 9.0,
1.0 mM MgCl2 (see Note 2).
4. Lysis buffer (3) for stopping the dephosphorylation reaction:
195 mM TrisHCl, pH 6.8, 3.0 % (w/v) SDS, 15 % (v/v)
2-sulfanylethanol (see Note 3), 30 % (v/v) glycerol, 0.10 %
(w/v) bromophenol blue (BPB). Store at 20 C.
2.1.2 Preparation
of Cell Lysate
1. Culture medium for HeLa cells: Dulbeccos modified Eagle

medium (DMEM), 10 % (v/v) fetal bovine serum (FBS),
100 units/mL penicillin, 100 g/mL streptomycin.
2. Stimulation solution A: 2.0 M actinomycin D.
332
3. Culture medium for A431 cells: RPMI1640, 10 % (v/v) FBS,

100 units/mL penicillin, 100 g/mL streptomycin.
4. Stimulation solution B: 250 ng/mL epidermal growth factor
(EGF).
5. Tris-buffered saline (TBS): 10 mM TrisHCl, pH 7.5, 0.10 M
NaCl.
6. Lysis buffer (1): 65 mM TrisHCl, pH 6.8, 1.0 % (w/v)
SDS, 5.0 % (v/v) 2-sulfanylethanol, 10 % (v/v) glycerol,
0.03 % (w/v) BPB. Store at 20 C.
2.2 Gel Casting
for Electrophoresis
2.2.1 Mn2+Phos-tag
SDS-PAGE Using
the Laemmlis Buffer
System
1. Phos-tag Acrylamide solution: 5.0 mM acrylamide-bound

Phos-tag ligand (Phos-tag Acrylamide AAL-107), 3.0 % (v/v)
MeOH (see Note 4) in distilled water. Store at room temperature in the dark.
2. Manganese(II) chloride solution: 10 mM MnCl24H2O in distilled water (see Note 5). Store at room temperature.
3. Acrylamide/bis solution (30 %): 30 % (w/v) solution of a 29:1
mixture of acrylamide and N,N-methylenebisacrylamide in
water (see Note 6). Store at room temperature in the dark.
4. Separating gel buffer (4): 1.5 M TrisHCl, pH 8.8, 0.40 %
(w/v) SDS. Store at room temperature.
5. Stacking gel buffer (4): 0.5 M TrisHCl, pH 6.8, 0.40 %
(w/v) SDS. Store at room temperature.
6. Ammonium persulfate (APS) (see Note 7) solution and N,N,
N,N-tetramethylethane-1,2-diamine (TEMED) (see Note 8):
10 % (w/v) APS in distilled water. Regarding TEMED, the
original solution is used as received from the supplier.
7. Electrophoresis running buffer (see Note 9): 25 mM Tris,
192 mM glycine, 0.10 % (w/v) SDS. Store at room
temperature.
2.2.2 Zn2+Phos-tag
SDS-PAGE Using a Bis
TrisHCl Buffer System
1. Phos-tag Acrylamide solution: See Subheading 2.2.1, item 1.

2. Zinc(II) chloride solution: 10 mM ZnCl2 in distilled water (see
Note 10). Store at room temperature.
3. Acrylamide/bis solution (30 %): See Subheading 2.2.1, item 3.
4. Gel buffer (2.8) (see Note 11): 1.0 M BisTrisHCl, pH 6.8.
Store at room temperature.
5. APS solution and TEMED: See Subheading 2.2.1, item 6.
100 mM 3-morpholin-4-ylpropane-1-sulfonic acid (MOPS),
0.10 % (w/v) SDS. Store at room temperature. Sodium bisulfite is dissolved to a concentration of 5.0 mM in the buffer
solution immediately before use (see Note 13).

2.2.3 Zn2+Phos-tag
SDS-PAGE Using
a TrisAcOH Buffer System
333
1. Phos-tag Acrylamide solution: See Subheading 2.2.1, item 1.

2. Zinc(II) chloride solution: See Subheading 2.2.2, item 2.
3. Acrylamide/bis solution (30 %): See Subheading 2.2.1, item 3.
4. Gel buffer (5) (see Note 14): 1.0 M TrisAcOH buffer,
pH 7.0 (see Note 15). Store at room temperature.
5. Agarose solution (see Note 16): 1.5 % (w/v) SeaKem Gold
agarose (Lonza Group, Basel, Switzerland).
6. APS solution and TEMED: See Subheading 2.2.1, item 6.
50 mM N-[2-hydroxy-1,1-bis(hydroxymethyl)ethyl]glycine
(tricine), 0.10 % (w/v) SDS in distilled water. Store at room
temperature. Dissolve sodium bisulfite in the buffer solution to a concentration of 5.0 mM immediately before use
(see Note 13).
2.3
Gel Staining
1. Fix solution before staining gel: 10 % (v/v) MeOH, 7.0 %

(v/v) AcOH in distilled water. Store at room temperature.
2. Coomassie Brilliant Blue G-250 (CBB) stain solution: 0.25 %
(w/v) CBB, 40 % (v/v) MeOH, 10 % (v/v) AcOH in distilled
water. Store at room temperature.
3. Washing solution after staining gel: 40 % (v/v) MeOH, 10 %
(v/v) AcOH in distilled water. Store at room temperature.
2.4
Electroblotting
1. Blotting buffer: 25 mM Tris, 192 mM glycine, 10 % (v/v)

MeOH for the wet-tank method (see Note 18). Store at room
temperature.
2. Ethylenediaminetetraacetic acid disodium salt (EDTA2Na)
solution: 0.50 M EDTAsodium hydroxide (NaOH), pH 8
(see Note 19). Store at room temperature.
3. Blotting buffer containing 1 mM EDTA: 100 mL of blotting
buffer (see Subheading 2.4, item 1), 0.20 mL of EDTA solution (see Subheading 2.4, item 2).
4. TBS-T solution: 10 mM of TrisHCl, pH 7.5, 0.10 M NaCl,
0.10 % (v/v) poly(oxyethylene) sorbitan monolaurate (Tween
20). Store at room temperature.
2.5 Equipment
for Electrophoresis
and Blotting
1. SDS-PAGE equipment: Atto model AE-6500 mini-slab gel

system (gels: 1 mm thick, 9 cm wide, and 9 cm long) (Tokyo,
Japan). The setup can be readily adapted to other formats,
including large-type gels.
2. Electroblotting equipment: Nihon Eido model NA-1511C
electroblotting wet-tank unit (Tokyo, Japan). The setup can
be readily adapted to other formats but not to a semi-dry one
(see Note 18).
334
Methods
3.1 Sample
Preparation
3.1.1 Preparation
of Standard
Phosphoproteins
Dephosphorylated by AP
3.1.2 Preparation
of a Lysate from HeLa Cells
1. Add each 50 g of standard phosphoprotein and 3.3 units

(0.8 ng) of AP to 200 L of dephosphorylation reaction buffer.
2. Incubate solutions at 37 C for 5 min or 12 h.
3. Stop the reactions by adding 100 L of lysis buffer (3).
1. Grow HeLa cells in culture medium under a humidified atmosphere of 5 % CO2 and 95 % air at 37 C.
2. Treat the cells (~107 cells) with 2.0 M actinomycin D for 2 h.
3. Wash the cells twice with TBS.
4. Lyse the cultures in 0.50 mL of lysis buffer (1).
5. Sonicate briefly the lysate samples, boil them for 5 min, and
store at 20 C.
6. Subject an aliquot (10 L) of the resulting sample solution to
Phos-tag SDS-PAGE.
3.1.3 Preparation
of a Lysate from A431
Cells
1. Grow A431 cells in culture medium under a humidified atmosphere of 5 % CO2 and 95 % air at 37 C.
2. Treat the cells (~107 cells) with 250 ng/mL of EGF for 0 (no
treatment), 2, 5, 10, 30, 60, 90, or 120 min.
3. Wash each culture twice with TBS.
4. Lyse the cultures in 0.50 mL of lysis buffer (1).
5. Sonicate briefly the lysate samples, boil them for 5 min, and
store at 20 C.
6. Subject an aliquot (10 L) of the resulting solution to Phostag SDS-PAGE.
3.2
Electrophoresis
3.2.1 Preparation
of the Zn2+Phos-tag
SDS-PAGE Gel
(See Note 20)
1. Clean the glass plates for casting the gels. It is vital that the
glass plates (1 mm thick, 9 cm wide, and 9 cm long for minislab gels) are washed thoroughly with a rinsable detergent, and
that they are subsequently thoroughly rinsed with distilled
water before the gels are cast.
2. Select an appropriate gel type for Zn2+Phos-tag SDSPAGE. For the analysis of 10200 kDa phosphoproteins, a gel
buffered with either BisTrisHCl or TrisAcOH gel can be
used. Cast this gel according to the procedures described in
steps 310 (below). For the analysis of large phosphoproteins
with molecular masses of more than 200 kDa, a TrisAcOH
gel should be used. Cast this gel according to the procedures
described in steps 1116 below.
3. Prepare the separating gel solution for the analysis of 10200 kDa
phosphoproteins. As a typical example of the Zn2+Phos-tag gel
335
buffered with BisTrisHCl (8 % w/v polyacrylamide, 100 M

Mn2+Phos-tag), a separating gel solution (~7 mL) is prepared
by mixing 1.866 mL of acrylamide/bis solution, 2.5 mL of gel
buffer (2.8), 140 L of Phos-tag Acrylamide solution, 140 L
of zinc(II) chloride solution (2 equivalents on Phos-tag), 7 L
of TEMED, and 2.312 mL of distilled water in a 50-mL centrifuge tube. As a typical example of the Zn2+Phos-tag gel
buffered with TrisAcOH (8 % w/v polyacrylamide, 100 M
Mn2+Phos-tag), a separating gel solution (~7 mL) is prepared
by mixing 1.866 mL of acrylamide/bis solution, 1.4 mL of gel
buffer (5), 140 L of Phos-tag Acrylamide solution, 140 L
of zinc(II) chloride solution (2 equivalents on Phos-tag), 7 L
of TEMED, and 3.412 mL of distilled water in a 50-mL centrifuge tube (see Note 21).
4. Add 35 L of APS solution and mix gently.
5. Transfer the separating gel solution to the gap between the
glass plates, pour a 50 % (v/v) aqueous solution of propan2-ol on top of the separating gel solution, and allow the gel
solution to polymerize for about 20 min at room temperature
(see Note 22; for troubleshooting advice, see Table 1).
6. Prepare the stacking gel solution. In a similar manner to the
preparation of the separating gel solution, prepare a typical
stacking gel solution (~2 mL, 4 % w/v polyacrylamide) buffered with BisTrisHCl by mixing 267 L of acrylamide/bis
solution, 715 L of gel buffer (2.8), 2 L of TEMED, and
1,006 L of distilled water in a 50-mL centrifuge tube. As a
typical example of the stacking gel solution buffered with Tris
AcOH (~2 mL, 4 % w/v polyacrylamide), prepare the gel solution by mixing 267 L of acrylamide/bis solution, 400 L of
gel buffer (5), 2 L of TEMED, and 1,321 L of distilled
water in a 50-mL centrifuge tube.
7. Discard the 50 % (v/v) propan-2-ol solution on the separating
gel and wipe the top surface of the separating gel with a filter
paper to remove any residual liquid.
8. Add 10 L of APS solution to the stacking gel solution, mix
gently, and then pour the solution onto the separating gel.
9. Insert a sample-well comb and allow the gel solution to polymerize for about 20 min (see Note 23) (for troubleshooting
advice, see Table 1).
10. Carefully remove the sample-well comb from the stacking gel,
and assemble the gel plate and electrophoresis apparatus.
After step 10, go to Subheading 3.2.2.
11. Prepare the separating gel solution for the analysis of large
phosphoproteins with molecular masses in excess of 200 kDa.
It is not necessary to make a stacking gel in this case. As a typical example of the Zn2+Phos-tag gel buffered with TrisAcOH
Problem
The acrylamide does not

polymerize within 20 min or so
The separating gel solution

hardens before it is poured into
the gel-casting system
When the clip and spacer are

removed, the polymerized gel
shrinks or slips down between
the glass plates
When the agarosepolyacrylamide

composite gel is electroblotted,
the gel partially sticks to the
PVDF membrane and is
difficult to separate from the
membrane
Step
Steps 5 and 9 in
Subheading 3.2.1
Step 13 in
Subheading 3.2.1
Step 16 in
Subheading 3.2.1
Step 7 in
Subheading 3.4
Table 1
Troubleshooting advice
The agarosepolyacrylamide
composite gel is partially melted
during electroblotting because of
the generation of heat
The polyacrylamide gel is not

sufficiently strengthened by agarose
Agarose gelling
Deterioration of TEMED or APS
Possible reasons
Do not set the voltage to more than 3.5 V/

cm. Even if an electroblotting unit with a
cooling apparatus is used, the use of
higher voltages in this step is not
recommended (see Note 34)
Make sure that the final agarose

concentration is 0.5 % (w/v) as
recommended
After the agarose is melted in the microwave

oven, the solution should be mixed and
poured into the casting system as quickly
as possible. If necessary, preheat the
casting system and plastic pipette tips in an
oven at 4045 C
Use fresh TEMED and APS. APS should be

dissolved in distilled water just before use
Solution
336
1. Decrease the amounts of the smallmolecule substances by dialysis filtration.

(Do not apply commercially available
prestained molecular-weight protein
markers, to avoid distortion of bands)
2. Shear DNA into smaller fragment by brief
periods of sonication, or add benzonase
nuclease to the sample solution
3. Centrifuge the sample at 14,000 g for
10 min before applying the supernatant
to the Phos-tag gel
4. Purify the sample by using a
chromatography resin, such as Whatman
CDR (cell debris remover; catalogue
number 4025050). Spin column
chromatography permits the purification
of small volumes of the sample solution
(10200 L) without significant losses
Add ZnCl2 or Zn(NO3)2 to the sample
solutions to a final concentration of
12 mM or, when using Phos-tag, increase
the concentration of ZnCl2 or Zn(NO3)2
in the separating gel up to 200 M
1. Contamination by small-molecule
substances, such as inorganic salts
or surfactants
2. Contamination by viscous
genomic DNA
3. Contamination by cell debris or
insoluble biomaterials
4. Impurities, such as lipids or small
molecules derived from cells,
interfere with the interaction
between the Zn2+Phos-tag and
phosphoprotein
The samples contain metal-chelating

agents, such as EDTA or ethylene
glycol-bis(2-aminoethyl ether)N,N,N,N-tetraacetic acid (EGTA)
When the crude samples, such as

whole cell lysates, are applied
to the gel, the banding pattern
is badly distorted or smeared
The degree of migration of the

target protein is different in
each lane, and the banding
pattern is distorted
Step 8 in
Subheading 3.4
Solution
Possible reasons
Problem
Step

337
338
(~8.5 mL, 3 % w/v polyacrylamide) containing 0.5 % (w/v)

agarose, a separating gel solution (~8.5 mL) is prepared by
mixing 0.85 mL of acrylamide/bis solution, 1.7 mL of gel
buffer (5), 34 L of Phos-tag Acrylamide solution, 34 L of
zinc(II) chloride solution (2 equivalents on Phos-tag), 9 L of
TEMED, and 2.977 mL of distilled water in a 50-mL centrifuge tube (see Note 24 and Fig. 2).
12. Prepare the 1.5 % (w/v) agarose solution. Suspend SeaKem
Gold agarose (0.75 g) in 50 mL of distilled water in a 100-mL
flask and melt the agarose solution completely by using a
microwave oven (see Note 25).
13. Add 2.833 mL of the hot agarose solution (>90 C) to the
separating gel solution and mix gently (for troubleshooting
advice, see Table 1).
14. Add 43 L of APS solution and mix gently.
15. Transfer the warm gel solution (which should be above the
gelling temperature of the agarose) to the gap between the
glass plates, and insert a sample-well comb and allow the acrylamide to polymerize for about 20 min.
16. Carefully remove the comb from the gel, and assemble the gel
plate and electrophoresis apparatus (for troubleshooting
advice, see Table 1) (see Note 26).
Fig. 2 Schematic representation of the relationship between the degree of migration (Rf value) and the molecular weight for myosin heavy chain (205 kDa) and
ATM (350 kDa) in 2.54.0 % (w/v) polyacrylamide slab gels. The Rf value of myosin heavy chain was determined by means of CBB gel staining. The Rf value of
ATM was determined by Western blotting analysis of HeLa cell lysate using the
anti-ATM antibody. The Rf value of 1.0 is defined as the position of the BPB dye

3.2.2 Procedure
for Electrophoresis
(See Note 27)
339
1. Fill the electrode chambers with the running buffer for

electrophoresis. Take care not to introduce bubbles onto the
bottom surface of the gel set (see Note 28).
2. Load the protein samples mixed with the sample-loading dye
solution into the wells (see Note 29).
3. Attach the leads to the power supply. When you are using gel
buffer systems without agarose, run the gels under constantcurrent conditions of 40 mA/gel at room temperature until
the BPB dye reaches the bottom of the separating gel. With a
mini-slab gel (1 mm thick, 9 cm wide, and 9 cm long), the
time required for complete electrophoresis is about 1.5 h for
the BisTrisHCl gel buffer system and 2 h for the TrisAcOH
system. For the TrisAcOH system containing agarose, run the
gels under constant-current conditions of 15 mA/gel at room
temperature until the BPB dye reaches the bottom of the separating gel. The time required for complete electrophoresis is
about 4 h when the mini-slab gel is used (see Note 30).
3.3
Gel Staining
1. When the electrophoresis run is complete, remove the gel from

the apparatus and fix it using the fix solution.
2. The fixed gel is incubated in a CBB stain solution to visualize
the protein bands.
3. The stained gel is washed in the washing solution until the
background is clear.
3.4
Electroblotting
1. When the run is complete, remove the gel from the apparatus
and soak it in the blotting buffer containing 1 mM EDTA for
30 min (see Note 31).
2. Soak the gel in the blotting buffer without EDTA for a further
1030 min.
3. Prepare the poly(vinylidene difluoride) (PVDF) membrane
(e.g., Fluorotrans W, Nippon Pall, Tokyo, Japan) by cutting it
to the same size as the gel and soak it for 30 s in 100 %
MeOH. Then, incubate the membrane in the blotting buffer
for more than 30 min.
4. Prepare four pieces of blotting paper (e.g., 3MM; Whatman,
Maidstone, UK) by cutting them to the same size as the gel.
5. To form a blotting sandwich on the electroblotting screen
attached to the electroblotting equipment, assemble the gel,
PVDF membrane, and 3MM paper as follows. Soak the blotting sponge attached to the electroblotting equipment in the
blotting buffer and place it on the electroblotting screen. Then
soak two pieces of 3MM paper in the blotting buffer and place
them on the sponge; these are followed sequentially by the gel
and PVDF membrane. Avoid incorporating air between the
various layers. Then place two more sheets of 3MM paper and
340
one sponge soaked in the blotting buffer on the membrane

and close the electroblotting screen. Insert the electroblotting
screen into the chamber unit of the electroblotting equipment
and fill up with the blotting buffer (see Note 32).
6. Gently add SDS solution to the blotting buffer of the chamber
unit to a final concentration of 0.1 % (w/v) (see Note 33).
7. Attach the leads to the power supply. When you are using gel
buffer systems without agarose, run the gel under constantvoltage conditions. With a Nippon Eido model NA-1511C
electroblotting wet-tank unit, a constant-voltage of 30 V (4 V/
cm) for 16 h (overnight) without cooling is optimal. When an
electroblotting unit with a cooling system is used, the equipment can be operated at higher voltages (2022 V/cm) for
1 h. For the TrisAcOH system with agarose, run the gel under
constant-voltage conditions of 3.5 V/cm for 16 h (overnight)
(see Note 34) (for troubleshooting advice, see Table 1).
8. After blotting, soak the PVDF membrane in TBS-T solution,
and then perform the immunoblotting analysis (for troubleshooting advice, see Table 1).
3.5
Typical Results
3.5.1 Separation and

Detection of Standard
Phosphoproteins
[42, 45, 47]
The current Zn2+Phos-tag SDS-PAGE produced better separations of phosphorylated species of various proteins than did the
previous Mn2+Phos-tag SDS-PAGE. Comparative studies were
performed by using several molecular-weight markers (1497 kDa)
and four standard phosphoproteins: -casein (24 kDa), -casein
(24 kDa), ovalbumin (45 kDa), and pepsin (35 kDa) (Fig. 3).
Differences in the degrees of migration and banding patterns
between the phosphorylated forms of the proteins and their
dephosphorylated counterparts, obtained by treatment with AP,
Fig. 3 (continued) Zn2+Phos-tag (right-hand panel). The figures in parts (a) and (b) are reproduced from ref. [45]
with the permission of the publisher, Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim, Germany. (c) 10 % (w/v)
Polyacrylamide gel containing 0 M (left-hand panel) or 100 M polyacrylamide-bound Mn2+Phos-tag (righthand panel). The figures in part (c) are reproduced from ref. [42] with permission of the publisher, Wiley-VCH
Verlag GmbH & Co. KGaA, Weinheim, Germany. Lanes 111 correspond to (1) -casein (1.0 g), (2) -casein
treated with AP for 5 min (1.5 g), (3) -casein treated with AP for 12 h (1.0 g), (4) -casein (1.0 g), (5)
-casein treated with AP for 5 min (1.5 g), (6) -casein treated with AP for 12 h (1.0 g), (7) molecular-weight
markers (APRO marker low range; APRO Science, catalogue number SP-0110) [from the top: phosphorylase b
(97 kDa), BSA (66 kDa), ovalbumin (45 kDa), carbonic anhydrase (29 kDa), trypsin inhibitor (21 kDa), and lysozyme (14 kDa)], (8) ovalbumin (1.0 g), (9) ovalbumin treated with AP for 12 h (1.0 g), (10) pepsin (1.0 g),
and (11) pepsin treated with AP for 12 h (1.0 g). The Rf value of 1.0 is defined as the position of the BPB dye.
The gels were stained with CBB dye. Apparent molecular weights of marker bands are indicated on the right-hand
side of the left panel. By using the Zn2+Phos-tag gel buffered with BisTrisHCl, three distinct phosphorylated
species of ovalbumin could be separated, as indicated by arrowheads (see lane 8 in part (b); from the top:
species phosphorylated at both Ser-68 and Ser-344, species monophosphorylated at Ser-68, and species
monophosphorylated at Ser-344, respectively) [47]
Fig. 3 Comparison of mobilities of standard phosphoproteins in Zn2+Phos-tag gels buffered with TrisAcOH
and BisTrisHCl and in Mn2+Phos-tag gel. (a) TrisAcOH system on 12 % (w/v) polyacrylamide gel containing 0 M (left-hand panel) or 100 M polyacrylamide-bound Zn2+Phos-tag (right-hand panel). (b) BisTris
HCl system on 8.0 % (w/v) polyacrylamide gel containing 0 M (left-hand panel) or 100 M polyacrylamide-bound
342
were examined for Zn2+Phos-tag gels buffered with TrisAcOH

(Fig. 3a) or with BisTrisHCl (Fig. 3b) and for Mn2+Phos-tag
gel (Fig. 3c). In normal SDS-PAGE gels (left-hand panels; no
Phos-tag), no significant difference was observed in the electrophoresis banding images of any of the samples in the three buffer
systems under our experimental conditions. The Phos-tag gels
(right-hand panels, 100 M) were prepared to contain identical
concentrations of polyacrylamide to that present in the corresponding normal gel. In all the Phos-tag gels, phosphorylated
-casein, -casein, and ovalbumin migrated less than did their
completely dephosphorylated counterparts. The partially dephosphorylated -casein (lane 2) and -casein (lane 5) produced multiple bands. The banding images of -casein and -casein were
similar for all three buffer systems, whereas some differences were
observed in the banding image of ovalbumin, where three distinct
up-shifted bands were clearly detectable in the Zn2+Phos-tag gel
buffered with BisTrisHCl (see arrowheads in lane 8 of Fig. 3b).
Furthermore, under our experimental conditions, the migrated
positions of ovalbumin (45 kDa) and BSA (66 kDa) in the
molecular-weight markers (lane 7) were inverted on the Zn2+
Phos-tag gel buffered with BisTrisHCl. We did not observe this
inversion on the Zn2+Phos-tag gel buffered with TrisAcOH or
Mn2+Phos-tag gel, even at lower concentrations of polyacrylamide. In the banding image of pepsin, a shift in the mobility of
the phosphorylated species (lane 10) compared with that of its
dephosphorylated counterpart (lane 11) was detectable in the both
Zn2+Phos-tag gels, but not in the Mn2+Phos-tag gel.
3.5.2 Separation
and Detection of a HighMolecular-Mass
Phosphorylated ATM [45]
The TrisAcOH buffer system in Zn2+Phos-tag SDS-PAGE had

some advantages in permitting the casting of highly porous polyacrylamide gels containing less than 4 % (w/v) of polyacrylamide
for the separation and analysis of large phosphoproteins with
molecular masses of more than 200 kDa. We demonstrated the
detection of a novel phosphorylated form of ATM (350 kDa) by
using this system. ATM is a DNA damage signaling-related protein
that becomes autophosphorylated at its Ser-367, Ser-1893, and
Ser-1981 residues in response to DNA damage, and subsequently
initiates kinase activity toward other molecules [49]. Recently,
autophosphorylation at Ser-2996 has also been reported [50]. The
various phosphorylation states of ATM therefore form key components of the signaling pathway. A set of lysates was prepared from
HeLa cells after treating the cells (107 cells) with 0 M (control) or
2 M actinomycin D (as a DNA-damaging reagent) for 2 h. The
induced culture and the control culture were each washed twice
with a TBS solution and then lysed in 0.50 mL of a 1 sampleloading dye. The two lysates were first analyzed by Zn2+Phos-tag
SDS-PAGE on a 3.0 % (w/v) polyacrylamide gel strengthened with
0.5 % (w/v) agarose containing 20 M of polyacrylamide-bound
343
Zn2+Phos-tag in conjunction with the TrisAcOH buffer system

(Fig. 4a, left-hand panel). Subsequent immunoblotting with a
phosphorylation-independent anti-ATM antibody produced a single band in the case of the control sample (left lane: ) and three
additional up-shifted bands in the case of the treated sample (right
lane: +). Next, the two lysates were analyzed by using the older
Mn2+Phos-tag SDS-PAGE method with the same concentration
of agarosepolyacrylamide composite gel containing 20 M of
polyacrylamide-bound Mn2+Phos-tag in conjunction with
Laemmlis buffer system (Fig. 4a, right-hand panel). Only two upshifted bands corresponding to phosphorylated forms of ATM
were detected in this case. Therefore, the current Zn2+Phos-tag
SDS-PAGE protocol using the TrisAcOH buffer (pH 7.0) permits a more-detailed detection of shifts in the mobility of phosphorylated species of ATM in response to DNA damage than does
the previous Mn2+Phos-tag SDS-PAGE. In a similar separation
analysis using the BisTrisHCl system, we obtained smeared electrophoresis signals as a result of the inadequate sieving property of
the dilute acrylamide gels (Fig. 4b), as mentioned in Subheading 1.5.
3.5.3 Comparison
of the Mobilities
of Phosphorylated ERK1/2
in the Two Neutral-pH Gel
Systems [45]
In an analysis of EGF-dependent phosphorylation of ERK1 (44 kDa)

and ERK2 (42 kDa), we demonstrated the presence of differences
in the distribution profiling of phosphorylated congeners of ERK1
and ERK2 between the two neutral buffer systems. Whole lysates
of untreated A431 cells (107 cells) and EGF-treated A431 cells
(250 ng/mL, 5 min) were subjected to Zn2+Phos-tag SDS-PAGE
in the presence of BisTrisHCl or TrisAcOH buffer, followed by
immunoblotting analysis using a phosphorylation-independent
anti-ERK1/2 antibody. Three distinct up-shifted bands were
detected in the samples of EGF-treated cell lysate on both types of
1-D Phos-tag gel (Fig. 5a). 2-DE coupling with normal Laemmlis
SDS-PAGE as the first dimension revealed that each up-shifted
band detected on the 1-D gels contained two variants each of
ERK1 and ERK2.
Next, we examined the time-course of changes in the EGFinduced phosphorylation levels of ERK1 and ERK2 by using Zn2+
Phos-tag SDS-PAGE with BisTrisHCl or TrisAcOH buffer.
The gels were analyzed by Western blotting with an anti-phosphosite-specific antibody against both pT202 and pY204 (anti-pT202/Y204
antibody). Subsequently, the same blots were probed again with
the phosphorylation-independent anti-ERK antibody (total ERK,
Fig. 5b). In the BisTrisHCl buffer system (upper panels), the top
band showed a cross activity with the anti-pT202/Y204 antibody, indicating that this band corresponds to an active form of ERK. The level
of the active form of ERK1/2 increased during 210 min and then
decreased for 30 min. The second band from the top appeared after
2 min and then slowly disappeared during 120 min. The third band
showed similar dynamics of phosphorylation to active ERK1/2.
344
Fig. 4 Separation analysis of phosphorylated ATM in TrisAcOH buffered Zn2+

Phos-tag gel strengthened by agarose. (a) The cells were treated with 0 M () or
2 M actinomycin D (+) for 2 h. The lysates were subjected to Phos-tag SDS-PAGE
on 3.0 % (w/v) polyacrylamide gel strengthened with 0.5 % (w/v) SeaKem Gold
agarose containing 20 M Phos-tag, followed by immunoblotting with a phosphorylation-independent anti-ATM mouse monoclonal antibody (clone 2C1; Santa Cruz
Biotechnology, Santa Cruz, CA, USA). The TrisAcOH gel buffer and TrisTricine
running buffer were used in Zn2+Phos-tag SDS-PAGE (left panel), and Laemmlis
gel buffer and running buffer were used in Mn2+Phos-tag SDS-PAGE (right panel).
Each lane contained 10 g proteins. (b) A typical electrophoresis image using
a BisTris-buffered gel at a concentration of 4 % (w/v) of polyacrylamide in
Zn2+Phos-tag SDS-PAGE. HeLa whole lysate (20 g proteins) was subjected
to normal SDS-PAGE [10 % (w/v) polyacrylamide gel] as the first dimension.
345
In the TrisAcOH buffer system (lower panels), on the other hand,

the second band from the top showed a cross activity with the antipT202/Y204 antibody, indicating that the band corresponds to active
forms of ERK1/2. From the time-dependent dynamics of phosphorylation of each band, we deduced that the top band and third
band in the TrisAcOH system were equivalent to the second and
third bands, respectively, in the BisTrisHCl system. Thus, a difference in the distribution profiling of phosphorylated species of
ERK1/2 between the two systems was demonstrated. The different results obtained with the two buffer systems suggest that the
affinity of Zn2+Phos-tag to phosphoproteins might be affected
not only by the pH of the gel, but also by other factors, such as the
nature of the gel buffer or the presence of leading or trailing ions
during electrophoresis.
The two up-shifted bands other than the active form were
identified as monophosphorylated species by immunoblotting with
two anti-phospho-site-specific antibodies against pT202 and pY204,
respectively. The third band that was detected in both buffer systems corresponds to a species that was monophosphorylated at the
Thr-202 residue, and the others correspond to the species monophosphorylated at the Tyr-204 residue (Fig. 5c, left and center
panels in both buffer systems). Because MEK is a kinase with a dual
specificity that simultaneously phosphorylates the Thr-202 and
Tyr-204 residues of ERK in a processive manner [51], we hypothesized that the monophosphorylated species might be generated
during dephosphorylation by certain activities of protein phosphatase toward ERK1/2. Several protein phosphatases, including Tyrspecific, Ser/Thr-specific, and dual-specific phosphatases, have
been reported to act as ERK phosphatases [52]. We therefore analyzed the EGF-dependent phosphorylation of ERK1/2 in A431
cells in the presence of a Ser/Thr-specific phosphatase inhibitor
(calyculin A) or a Tyr-specific phosphatase inhibitor (pervanadate)
(Fig. 5c, right-hand panels in both buffer systems). In the sample of
the lysate of the EGF-stimulated cells (250 ng/mL, 5 min) pretreated
with calyculin A (100 nM, 30 min), the active form (a diphosphorylated species) and a species that was monophosphorylated at
Thr-202 were observed in both buffer systems. In the sample of
Fig. 4 (continued) The separated sample lane was cut, and then the part corresponding to 80250 kDa was subjected to Zn2+Phos-tag SDS-PAGE [4 % (w/v)
polyacrylamide, 50 M Zn2+Phos-tag] buffered with BisTrisHCl as the second dimension. The Phos-tag gel was strengthened with 0.5 % (w/v) SeaKem
Gold agarose. All the spots appeared as smeared images after gel staining,
because the BisTris-buffered gel did not have adequate sieving properties at
concentration of less than 4 % (w/v) polyacrylamide. The Rf value of 1.0 is
defined as the position of the BPB dye. Figures are reproduced from ref. [45] with
permission of the publisher
Fig. 5 Comparison of mobility of phosphorylated ERK between the BisTrisHCl buffer system and TrisAcOH
buffer system. (a) A431 whole lysate (, 10 g proteins) and lysate from EGF-stimulated cells (+, 10 g proteins) were subjected to Zn2+Phos-tag SDS-PAGE [8.0 % (w/v) polyacrylamide and 25 M Zn2+Phos-tag]
using the BisTrisHCl buffer system or TrisAcOH buffer system. The gels were analyzed by Western blotting with
347
lysate from cells pretreated with pervanadate (1.0 mM, 30 min),

on the other hand, the active form and a species that was monophosphorylated at Tyr-204 were observed. When the cells were
stimulated with EGF in the presence of the MEK-specific inhibitor
PD98059 (100 M, 60 min), no phosphorylated species were
detected. These results show that the three up-shifted bands
observed in both buffer systems are MEK-dependent phosphorylated species in the EGF-signaling pathway; one is the active form
diphosphorylated at both Thr-202 and Tyr-204 residues by MEK,
and the others are partially dephosphorylated forms produced
during inactivation by certain ERK phosphatases. Therefore, we
found that our two buffer systems could be used to obtain an overview of the kinase/phosphatase-dependent dynamics of various
phosphorylation states of ERK1/2 in the EGF-signaling pathway.
Notes
1. All reagents and solvents used are purchased at the highest
commercial quality available and used without further purification. All aqueous solutions are prepared by using deionized
and distilled water.
2. Hydrochloric acid (HCl) is dangerously irritating to the skin,
eyes, and mucous membranes. When handing this chemical,
work in a chemical fume hood and wear gloves, eye protection,
and a mask.
Fig. 5 (continued) a phosphorylation-independent anti-ERK antibody (Millipore, Bedford, MA). The lysate from
the EGF-stimulated cells (10 g proteins) was analyzed by 2-DE, consisting of normal SDS-PAGE [8.0 % (w/v)
polyacrylamide] as the first dimension and Zn2+Phos-tag SDS-PAGE [8.0 % (w/v) polyacrylamide, 25 M
Zn2+Phos-tag] as the second dimension. The Rf value of 1.0 is defined as the position of the BPB dye. (b) The
time-course of phosphorylation of ERK after stimulation with EGF (250 ng/mL) was analyzed by Zn2+Phos-tag
SDS-PAGE [8.0 % (w/v) polyacrylamide and 25 M Zn2+Phos-tag] using the BisTrisHCl buffer system or the
TrisAcOH buffer system. Each lane contains 10 g of proteins. The gels were analyzed by Western blotting
with the dual-phosphorylation-specific anti-pT202/Y204 antibody (Cell Signaling Technology, Danvers, MA, USA),
and then the same blot was reprobed with the phosphorylation-independent anti-ERK antibody. (c) A431 whole
lysate (control, 10 g proteins) and lysate from EGF-stimulated cells (10 g proteins, 250 ng/mL of EGF for
5 min) were subjected to Zn2+Phos-tag SDS-PAGE [8.0 % (w/v) polyacrylamide and 25 M Zn2+Phos-tag]
using the BisTrisHCl buffer system or TrisAcOH buffer system, followed by immunoblotting with the
phospho-site-specific anti-pT202 antibody or anti-pY204 antibody (left and center panels, respectively, in both
systems) (Signalway Antibody, Pearland, TX, USA). Lysates from EGF-stimulated cells (10 g proteins, 250 ng/
mL of EGF for 5 min) pretreated with calyculin A (100 nM, 30 min) (Cell Signaling Technology), pervanadate [a
mixture of 1.0 mM sodium orthovanadate (Sigma-Aldrich) and 3.0 mM H2O2 (Nacalai Tesque), 30 min], or
PD98059 (100 M, 60 min) (Calbiochem, La Jolla, CA, USA) were similarly analyzed with the phosphorylationindependent anti-ERK antibody (right panels in the both systems). Figures are reproduced from ref. [45] with
permission of the publisher
348
3. 2-Sulfanylethanol is toxic by inhalation, ingestion, and skin

contact. When handling this chemical, work in a chemical
fume hood, wear gloves and a mask, and use a pipetting aid. If
prepared without 2-sulfanylethanol, the buffer solution can be
stored at room temperature. In this case, 2-sulfanylethanol
should be dissolved in the solution to a concentration of 15 %
(v/v) immediately before use.
4. The oily product, acrylamide-pendent Phos-tag ligand (10 mg),
is placed in a plastic tube and completely dissolved in MeOH
(0.10 mL). The solution is diluted with distilled water
(3.2 mL) by pipetting. MeOH is an inhalation toxin that
causes depression of the central nervous system; when handling this chemical, work in a chemical fume hood, wear
gloves, and use a pipetting aid. The Phos-tag solution is stable
for at least 6 months.
5. Do not use any other salts such as Mn(NO3)2 or Mn(OAc)2.
The solution of MnCl2 is stable for at least 6 months.
6. Because acrylamide monomer is a neurotoxin and a suspected
human carcinogen and teratogen, take care to avoid exposure
to this substance. When weighing powdered acrylamide, work
in a chemical fume hood and wear gloves, eye protection, and
a mask. Furthermore, acrylamide is unstable and can polymerize violently on heating to its melting point (84.5 C). It is
incompatible with acids, bases, oxidizing agents, reducing
agents, iron and its salts, copper, aluminum, brass, and freeradical initiators.
7. Avoid skin and eye contact, ingestion, and inhalation.
Prolonged exposure may result in skin burns and ulcerations.
Inhalation can cause respiratory irritation. The APS solution
should be prepared immediately before use.
8. TEMED is stored in a desiccator at room temperature. Buy
small bottles as its quality may degrade once the container is
opened, resulting in gels taking longer to polymerize. So after
opening it, use it up as soon as possible. The commercially
available solution is used as received.
9. Do not adjust the pH with acid or base.
10. Zinc(II) nitrate solution [10 mM Zn(NO3)26H2O (e.g.,
Nacalai Tesque, catalogue number 36932-32) in distilled
water] is a suitable substitute. Because ZnCl2 and Zn(NO3)2
are deliquescent salts, the solutions should be prepared by
using fresh products from newly opened bottles. Aqueous
solutions of ZnCl2 or Zn(NO3)2 are stable for at least 6 months.
11. This buffer is used for the separating and stacking gels.
13. Sulfite ion (SO32) is a reducing reagent that decreases O2 levels
in the electrophoresis running buffer solution and inhibits the
349
oxidation of reduced proteins in the gel. Buffer solution in

which sodium bisulfite has been dissolved should be promptly
used for electrophoresis. Do not store buffer solution containing sodium bisulfite.
14. This buffer is used for the separating and stacking gels.
15. Acetic acid (AcOH) is dangerously irritating to the skin, eyes,
and mucous membranes. When handing this chemical, work in
a chemical fume hood and wear gloves, eye protection, and a
mask. Keep away from heat and flame.
16. When analyzing for high-molecular-mass phosphoproteins,
agarose can optionally be used. We have confirmed in the
Zn2+Phos-tag SDS-PAGE methodology that some commercially available agaroses with a gel strength of more than
1,000 g/cm2 at 1.5 % (w/v) {e.g., Agarose LO3 TAKARA
[>2,200 g/cm2 at 1.5 % (w/v)] purchased from Takara Bio,
Agarose KANTO [9001,400 g/cm2 at 1.5 % (w/v)] purchased from Kanto Chemical, Agarose KANTO ME [1,400
1,700 g/cm2 at 1.5 % (w/v)] purchased from Kanto Chemical,
or Agarose KANTO LE [1,2001,500 g/cm2 at 1.5 % (w/v)]
purchased from Kanto Chemical} are all suitable substitutes for
SeaKem Gold agarose with a gel strength of >3,500 g/cm2 at
1.5 % (w/v). However, NuSieve GTG [>500 g/cm2 at 4 %
(w/v)] from Lonza is not suitable as a substitute for addition
to Zn2+Phos-tag SDS-PAGE gel buffered with TrisAcOH to
a final concentration of 0.5 % (w/v). NuSieve 3:1 [>1,400 g/
cm2 at 4 % (w/v)] from Lonza, when added to the SDS-PAGE
gel to a final concentration of 0.5 % (w/v), does not produce
sufficiently strong gels for subsequent handling during
electroblotting.
18. Although the semi-dry method is generally the most efficient
method for protein blotting in terms of time and consumption
of buffer reagents, it is not suitable for electroblotting from the
Phos-tag SDS-PAGE gel. The transfer efficiency of proteins is
lower in the semi-dry method than that in the wet-tank
method. Do not adjust the pH with acid or base.
19. During adjustment of the pH with NaOH, EDTA dissolves in
the distilled water. NaOH is dangerously irritating to the skin
and eyes. When handing this chemical, wear gloves and eye
protection.
20. Except for the buffer system, the procedure for Mn2+Phostag SDS-PAGE is almost identical. For details of the procedure,
see ref. [38]. The procedure of gel preparation from step 1 to
step 16 in Subheading 3.2.1 requires 1 h.
21. Because the optimal percentage of polyacrylamide [e.g.,
515 % (w/v)] depends on the molecular weight of the target
350
protein, an appropriate value should be determined for each

target. In Zn2+Phos-tag SDS-PAGE, the Rf values of both
phosphorylated and nonphosphorylated proteins are generally
smaller than are those in normal SDS-PAGE. The optimal concentration of Zn2+Phos-tag (e.g., 5150 M) to achieve sufficient separation between the phosphorylated and
nonphosphorylated proteins should be determined. It is recommended that, in the procedure using the mini-slab gel, tests
should be conducted by using low concentrations of Zn2+
Phos-tag (525 M) for complex samples that contain various
phosphorylated and nonphosphorylated proteins, such as cell
lysates.
22. After step 5, the polymerized separating gel should be allowed
to stand at room temperature for 16 h (overnight). It is necessary to introduce the layer of 50 % (v/v) aqueous propan-2-ol
solution on top of the polymerized separating gel to prevent
the gel from drying out.
23. After step 9, Zn2+Phos-tag gels cast in BisTrisHCl or Tris
AcOH neutral buffer will be stable for at least 6 months.
However, the gel, casting glass plates, and sample-well comb
should be wrapped in Saran wrap to prevent drying out of the
gel. It is possible to store the wrapped gel at room temperature
under normal laboratory illumination until required. Recently,
precast Zn2+Phos-tag SDS-PAGE gels have become commercially available (SuperSep Phos-tag; Wako Pure Chemical
Industries, Ltd.) [53].
24. The optimal concentration (w/v) of polyacrylamide (e.g., 2.5
4.0 %) depends on the molecular weight of the target protein.
Refer to Fig. 2, which shows typical examples of the relationships
between the degree of migration (Rf value) and the molecular
weight of two proteins, myosin heavy chain (205 kDa) and
ATM (350 kDa), in 2.54.0 % (w/v) polyacrylamide mini-slab
gels containing 0.5 % (w/v) agarose buffered with TrisAcOH
without Zn2+Phos-tag. The optimal concentration of Zn2+
Phos-tag (e.g., 5100 M) to achieve sufficient separation
between the phosphorylated and nonphosphorylated proteins
should be determined as described in step 3 (see Note 21). It
is also recommended that when agarosepolyacrylamide composite mini-slab gels are used with complex samples of lysate,
tests should be conducted using low concentrations of Zn2+
Phos-tag (525 M).
25. To prevent changes in the concentration of the agarose solution as a result of boiling and evaporative loss of water, any
volume lost after complete melting of the agarose solution
should be made up with distilled water.
26. Agarosepolyacrylamide composite gel cast in TrisAcOH
neutral buffer are also stable for at least 6 months after step 16.
351
The gel can be stored until required by adopting the procedure

described above in Note 23.
27. The procedure of electrophoresis from step 1 to step 3 in
Subheading 3.2.2 requires 1.54 h.
28. If any bubbles are observed, they should be carefully and completely removed.
29. Various contaminants (e.g., EDTA, inorganic salts, or surfactants) in the sample protein solutions can cause disruption
(waving and/or tailing) of the electrophoresis bands in Zn2+
Phos-tag SDS-PAGE. To minimize this disruption, it is recommended that the samples should be desalted before
loading if there are large differences in the concentrations of
contaminants in the various samples. Furthermore, to avoid
distortion of protein bands, do not use commercially available prestained molecular-weight protein markers in Zn2+
Phos-tag SDS-PAGE.
30. Agarose-containing gels frequently display partial melting during electrophoresis because of the development of heat. To
avoid this problem with the agarosepolyacrylamide composite gel, it is necessary to use conditions of a relatively lower
constant voltage (15 mA/gel) for a longer time of about 4 h in
the procedure for the TrisAcOH system.
31. The presence of Zn2+Phos-tag in the gel causes inefficient
electroblotting. This can be ameliorated by treatment with
EDTA, which chelates the zinc(II) ions.
32. The use of wet-tank equipment is strongly recommended for
optimal efficiency of protein transfer from the Zn2+Phos-tag
SDS-PAGE gel. The efficiency of transfer of proteins from the
Zn2+Phos-tag gel is much higher in the wet-tank method than
in the semi-dry method.
33. The addition of SDS is recommended for optimal efficiency of
protein transfer from the Zn2+Phos-tag SDS-PAGE gel. The
optimal percentage of SDS [e.g., 0.050.2 % (w/v)] to achieve
adequate electrotransfer should be determined.
34. Do not set the voltage above 3.5 V/cm for the TrisAcOH
system with agarose. Higher voltages during this step are not
recommended, even if an electroblotting unit with a cooling
apparatus is used.
Acknowledgments
This work was supported in part by KAKENHI Grants (24590050,
25293005, 25560417, 25117718, and 26460036) and a research
grant from the Takeda Science Foundation.
352
References
1. Hunter T (2000) Signaling: 2000 and beyond.
Cell 100:113127
2. Olsen JV, Blagoev B, Gnad F et al (2006)
Global, in vivo, and site-specific phosphorylation dynamics in signaling networks. Cell 127:
635648
3. Ubersax JA, Ferrell JE Jr (2007) Mechanisms
of specificity in protein phosphorylation. Nat
Rev Mol Cell Biol 8:530541
4. Manning G, Whyte DB, Martinez R et al
(2002) The protein kinase complement of the
human genome. Science 298:19121934
5. Kerk D, Templeton G, Moorhead GB (2008)
Evolutionary radiation pattern of novel protein
phosphatases revealed by analysis of protein
data from the completely sequenced genomes
of humans, green algae, and higher plants.
Plant Physiol 146:351367
6. Brognard J, Hunter T (2011) Protein kinase
signaling networks in cancer. Curr Opin Genet
Dev 21:411
7. Martin L, Latypova X, Terro F (2011) Posttranslational modifications of tau protein:
implications
for
Alzheimers
disease.
Neurochem Int 58:458471
8. Kosako H, Nagano K (2011) Quantitative
phosphoproteomics strategies for understanding protein kinase-mediated signal transduction
pathways. Expert Rev Proteomics 8:8194
9. Schmelzle K, White FM (2006) Phosphoproteomic approaches to elucidate cellular signaling networks. Curr Opin Biotechnol 17:
406414
10. Stasyk T, Dubrovska A, Lomnytska M et al
(2005) Phosphoproteome profiling of transforming growth factor (TGF)- signaling:
abrogation of TGF1-dependent phosphorylation of transcription factor-II-I (TFII-I)
enhances cooperation of TFII-I and Smad3 in
transcription. Mol Biol Cell 16:47654780
11. Agrawal GK, Thelen JJ (2006) Large scale
identification and quantitative profiling of
phosphoproteins expressed during seed filling
in oilseed rape. Mol Cell Proteomics 5:
20442059
12. Kinoshita E, Takahashi M, Takeda H et al
(2004) Recognition of phosphate monoester
dianion by an alkoxide-bridged dinuclear
zinc(II) complex. Dalton Trans 11891193
13. Kinoshita-Kikuta E, Kinoshita E, Yamada A
et al (2006) Enrichment of phosphorylated proteins from cell lysate using a novel phosphateaffinity chromatography at physiological pH.
14. Kinoshita-Kikuta E, Kinoshita E, Koike T

(2009) Phos-tag beads as an immunoblotting
enhancer for selective detection of phosphoproteins in cell lysates. Anal Biochem 389:
8385
15. Kinoshita E, Yamada A, Takeda H et al (2005)
Novel immobilized zinc(II) affinity chromatography for phosphopeptides and phosphorylated proteins. J Sep Sci 28:155162
16. Kinoshita-Kikuta E, Yamada A, Inoue C et al
(2010) A novel phosphate-affinity bead with
immobilized Phos-tag for separation and
enrichment of phosphopeptides and phosphoproteins. J Integr OMICS 1:157169
17. Nabetani T, Kim YJ, Watanabe M et al (2009)
Improved method of phosphopeptides enrichment using biphasic phosphate-binding tag/
C18 tip for versatile analysis of phosphorylation dynamics. Proteomics 9:55255533
18. Kikuchi J, Iwafune Y, Akiyama T et al (2010)
Co- and post-translational modifications of
the 26S proteasome in yeast. Proteomics 10:
27692779
19. Oyama M, Nagashima T, Suzuki T et al (2011)
Integrated quantitative analysis of the phosphoproteome and transcriptome in tamoxifenresistant breast cancer. J Biol Chem 286:
818829
20. Tsunehiro M, Meki Y, Matsuoka K et al (2013)
A Phos-tag-based magnetic-bead method for
rapid and selective separation of phosphorylated biomolecules. J Chromatogr B Anal
Technol Biomed Life Sci 925:8694
21. Inamori K, Kyo M, Nishiya Y et al (2005)
Detection and quantification of on-chip phosphorylated peptides by surface plasmon resonance imaging techniques using a phosphate
capture molecule. Anal Chem 77:39793985
22. Inamori K, Kyo M, Matsukawa K et al (2008)
Optimal surface chemistry for peptide immobilization in on-chip phosphorylation analysis.
Anal Chem 80:643650
23. Inamori K, Kyo M, Matsukawa K et al (2009)
Establishment of screening system toward discovery of kinase inhibitors using label-free onchip phosphorylation assays. Biosystems 97:
179185
24. Kinoshita E, Kinoshita-Kikuta E, Koike T
(2013) Sandwich assay for phosphorylation of
protein multiplexes by using antibodies and
Phos-tag. Anal Biochem 438:104106
25. Ikeda H, Kamimoto J, Yamamoto T et al (2013)
A peptide microarray fabricated on a nonfouling phosphatidylcholine-polymer-coated
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
surface for a high-fidelity analysis of a cellular

kinome. Curr Med Chem 20:44194425
Kinoshita E, Kinoshita-Kikuta E, Sugiyama Y
et al (2012) Highly sensitive detection of protein
phosphorylation by using improved Phos-tag
Biotin. Proteomics 12:932937
Kinoshita E, Kinoshita-Kikuta E, Koike T
(2013) Phos-tag-based microarray techniques
advance phosphoproteomics. J Proteomics
Bioinform S6:008
Nakanishi T, Ando E, Furuta M et al (2007)
Identification on membrane and characterization of phosphoproteins using an alkoxidebridged dinuclear metal complex as a phosphate
binding tag molecule. J Biomol Tech 18:
278286
Kinoshita E, Kinoshita-Kikuta E, Takiyama K
et al (2006) Phosphate-binding tag, a new tool
to visualize phosphorylated proteins. Mol Cell
Proteomics 5:749757
Kinoshita-Kikuta E, Aoki Y, Kinoshita E et al
(2007) Label-free kinase profiling using phosphate affinity polyacrylamide gel electrophoresis. Mol Cell Proteomics 6:356366
Laemmli UK (1970) Cleavage of structural
Kinoshita E, Kinoshita-Kikuta E, Matsubara M
et al (2008) Separation of phosphoprotein isotypes having the same number of phosphate
groups using phosphate-affinity SDS-PAGE.
Kinoshita E, Kinoshita-Kikuta E, Matsubara M
et al (2009) Two-dimensional phosphateaffinity gel electrophoresis for the analysis of
phosphoprotein isotypes. Electrophoresis 30:
550559
Kimura Y, Nagata K, Suzuki N et al (2010)
Characterization of multiple alternative
forms of heterogeneous nuclear ribonucleoprotein K by phosphate-affinity electrophoresis. Proteomics 10:38843895
Yamada S, Nakamura H, Kinoshita E et al
(2007) Separation of a phosphorylated histidine protein using phosphate affinity polyacrylamide gel electrophoresis. Anal Biochem 360:
160162
Ishiai M, Kitao H, Smogorzewska A et al
(2008) FANCI phosphorylation functions as a
molecular switch to turn on the Fanconi anemia
pathway. Nat Struct Mol Biol 15:11381146
Kinoshita E, Kinoshita-Kikuta E, Ujihara H
et al (2009) Mobility shift detection of phosphorylation on large proteins using a Phos-tag
SDS-PAGE gel strengthened with agarose.
(2009) Separation and detection of large
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
353
phosphoproteins using Phos-tag SDS-PAGE.

Nat Protoc 4:15131521
Sugiyama M, Sugiyama Y, Hatano N et al
(2010) The DNA-binding activity of mouse
DNA methyltransferase 1 is regulated by
phosphorylation with casein kinase 1/.
Biochem J 427:489497
(2009) Phosphate-affinity gel electrophoresis
using a Phos-tag molecule for phosphoproteome
study. Curr Proteomics 6:104121
Messer AE, Gallon CE, McKenna WJ et al
(2009) The use of phosphate-affinity SDSPAGE to measure the cardiac troponin I phosphorylation site distribution in human heart
muscle. Proteomics Clin Appl 3:13711382
Kinoshita E, Kinoshita-Kikuta E (2011)
Improved Phos-tag SDS-PAGE under neutral
pH conditions for advanced protein phosphorylation profiling. Proteomics 11:319323
Aguilar HN, Tracey CN, Tsang SC et al (2011)
Phos-tag-based analysis of myosin regulatory
light chain phosphorylation in human uterine
myocytes. PLoS One 6:e20903
Lee J, Kim JC, Lee SE et al (2012) Signal
transducer and activator of transcription 3
(STAT3) protein suppresses adenoma-tocarcinoma transition in Apcmin/+ mice via regulation of Snail-1 (SNAI) protein stability.
J Biol Chem 287:1818218189
(2012) Phos-tag SDS-PAGE systems for phosphorylation profiling of proteins with a wide
range of molecular masses under neutral pH
conditions. Proteomics 12:192202
Sepulveda P, Binner JGP (2001) Persulfateamine initiation systems for gelcasting of
ceramic foams. Chem Mater 13:40654070
Kinoshita-Kikuta E, Kinoshita E, Koike T
(2012) Separation and identification of four
distinct serine-phosphorylation states of ovalbumin by Phos-tag affinity electrophoresis.
Kinoshita E, Kinoshita-Kikuta E, Shiba A et al
(2014) Profiling of protein thiophosphorylation by Phos-tag affinity electrophoresis: evaluation of adenosine 5-O-(3-thiotriphosphate)
as a phosphoryl donor in protein kinase reactions. Proteomics 14:668679
Kozlov SV, Graham ME, Peng C et al (2006)
Involvement of novel autophosphorylation sites
in ATM activation. EMBO J 25:35043514
Kozlov SV, Graham ME, Jakob B et al (2011)
Autophosphorylation and ATM activation:
additional sites add to the complexity. J Biol
Chem 286:91079119
Aoki K, Yamada M, Kunida K et al (2011)
Processive phosphorylation of ERK MAP
354
kinase in mammalian cells. Proc Natl Acad Sci

U S A 108:1267512680
52. Stephen MK (2000) Protein phosphatases and
the regulation of mitogen-activated protein
kinase signalling. Curr Opin Cell Biol 12:
186192
53. Kinoshita-Kikuta E, Kinoshita E, Koike T

(2012) A laborsaving, timesaving, and more
reliable strategy for separation of lowmolecular-mass phosphoproteins in Phos-tag
affinity electrophoresis. Int J Chem (Mumbai,
India) 4(5):18
Chapter 25
Enrichment and Identification of Bacterial Glycopeptides
by Mass Spectrometry
Abstract
Large-scale analysis of protein N- and O-linked glycosylation by mass spectrometry has traditionally been
performed in eukaryotes by parallel approaches aimed at elucidating glycan structures (glycomics) and
their formerly glycosylated peptides (glycoproteomics) without reference to their intact state. Such analyses depend heavily on commercial glycosidases (e.g. protein N-glycosidase F) that can remove glycans
from the peptide backbone for separate analyses. Bacterial glycosylation has only recently been identified
as a widespread phenomenon. In many cases however, unique bacterial sugars preclude enzymatic removal,
therefore ultimately requiring a site-specific approach for intact glycopeptide analysis. Here, we describe
protocols for the enrichment of bacterial glycopeptides using zwitterionichydrophilic interaction liquid
chromatography (ZIC-HILIC) and their analysis using liquid chromatography coupled to electrospray
ionization tandem mass spectrometry (LC-MS/MS) with collision-induced dissociation (CID) and higher
energy collisional dissociation (HCD) fragmentation for glycan structure elucidation and glycopeptide
identification.
Key words Glycoprotein, Hydrophilic interaction liquid chromatography (HILIC), Mass spectrometry,
Membrane proteins, Proteomics
Introduction
1.1 Protein
Glycosylation
in Bacteria
Protein N- and O-linked glycosylation is increasingly being recognized

as a common post-translational modification in pathogenic bacteria
[1, 2]. Although the functional context for bacterial glycosylation is
largely unknown, there is mounting evidence that glycosylation is
required for wild-type levels of fitness and deletion of genes encoding glycan biosynthesis enzymes or glycosyltransferases (GTases)
results in marked changes in virulence and host colonization [36].
There is thus a substantial need to identify the protein targets of
glycosylation, and hence a requirement for robust methods that
enable qualitative and quantitative analyses of glycosylated substrates and site localization, as well as the determination of glycan
structures and/or composition. In eukaryotes, such approaches
355
356
generally rely on methods to chemically or enzymatically isolate

glycans from their attachment site, allowing parallel, yet independent, analysis of the glycome and glycoproteome [7, 8]. Such
approaches are however, generally incompatible with prokaryotic
glycosylation due to the inherent diversity of bacterial glycans,
which predominantly utilize a varied repertoire of unique monosaccharides, either as sole modifications or as part of more complex
polysaccharides [1, 2]. Additionally, in many cases these unique
monosaccharides inhibit typical analytical approaches (e.g. the
asparagine-linked bacillosamine in the Campylobacter jejuni
N-linked glycan is resistant to cleavage by protein N-glycosidase F
[PNGase F]).
1.1.1 Diversity
of Bacterial Glycans
and Glycosylation Systems
Over the last decade there have been a number of newly characterized general glycosylation systems identified in a wide range of
bacterial species [1, 2, 5, 6, 9, 10], which have facilitated the
elucidation of glycan structures and their protein targets.
Glycosylation should not be regarded as a single discrete modification but rather as a class of several possible types, broadly clustered into N- and O-linkages (although rarer linkages such as
C-mannosylation also occur), that can occur in both gram-negative and -positive organisms and are characterized by the enzymatic formation of a chemical linkage between a carbohydrate
chain or monomer and protein substrates by the actions of oligosaccharyltransferases (OSTases) or GTases. Addition of glycans to
protein can also be via en bloc or sequential transfer. For en bloc
transfer, the glycan is typically assembled on a membrane-bound
lipid carrier (e.g. the C. jejuni N-linked and Neisseria meningitidis O-linked systems) and added upon completion to synthesized
proteins within the periplasmic space; while sequential transfer
implies the addition of a single sugar to protein potentially followed by additional monosaccharides (e.g. Haemophilus influenzae high molecular weight adhesin 1 (HMW1) glycosylation and
C. jejuni flagellin glycosylation [11, 12]).
The OSTases and GTases, which appear to be present across
bacterial genera, typically target protein substrates at similar consensus motifs (e.g. the D/E-X-N-X-S/T [X Pro] of N-linked
glycosylation in C. jejuni), or in disordered protein regions (e.g.
the low complexity or repeat regions targeted by ligase-like
O-linked systems). Glycosylation events can involve the attachment of multiple carbohydrates, ranging from single hexose or
N-acetylhexosamine sugars or bacterial-specific carbohydrates such
as pseudaminic acid or other nonulosonic acid derivatives [2, 11,
13], to complex glycan structures [2, 5, 6, 10]. Further complexity
is generated by the addition of non-carbohydrate moieties such as
methylated aspartic acid, phosphate, and phosphoethanolamine
MS Identification of Bacterial Glycopeptides
357
(pEtN) to multiple glycan structures [14, 15]. Finally, glycosylation

site occupation may also be more complex than initially thought,
with increasing evidence of glycan micro-heterogeneity, leading to
the same site being occupied by multiple unique glycans, as well as
site-specific competition with other modifications such as direct
protein pEtN or phosphocholine (pCh) addition [16].
1.1.2 Proteomic
Approaches to Bacterial
Glycoprotein Identification
Methods to detect glycosylation events in prokaryotes are limited

and generally rely on prior knowledge of the glycan composition
(and hence known mass). Initial exploratory approaches included
glycan staining, lectin blotting or affinity purification [17], and
targeting using substrate probes combined with gel-based separations and identification by mass spectrometry (MS). Such
approaches are generally clouded by false positives (e.g. nonglycosylated co-purifying proteins in lectin affinity purifications)
and the reliance on gel technologies overlooks glycosylated proteins that are hydrophobic or contain multiple transmembrane
regions, as well as high molecular mass proteins. For optimal
glycoprotein identification and site elucidation, a high-throughput strategy compatible with complex mixtures and gel-free
separation was developed [18]. Here, we describe the use of
zwitterionichydrophilic interaction liquid chromatography
(ZIC-HILIC) for the enrichment of bacterial glycopeptides based
on their hydrophilicity and coupled to the use of liquid chromatographytandem mass spectrometry (LC-MS/MS) by collisioninduced dissociation (CID) fragmentation for the confirmation of
glycan structure and higher energy collisional dissociation (HCD)
fragmentation for elucidation of glycopeptide sequence. From our
experience, larger glycans allow more efficient enrichment, with
the N-linked glycans of Campylobacter species [14, 18] and multimers of Acinetobacter species [19] easily enriched from ~100 g
of peptide material. Despite this, smaller glycans can also be effectively enriched by this approach (e.g. trisaccharides from
Burkholderia cepacia [6]). It should be noted that in addition to
trypsin other proteases can be used to generate different glycopeptide mixtures and thus enable greater coverage of the glycoproteome [18]. The use of multiple MS/MS fragmentation
approaches within a single LC-MS/MS analysis also provides a
streamlined analysis of glycosylated substrates (Orbitrap/Tribrid
instruments are thus an ideal platform for these analyses). As with
other proteomic approaches, this technique when coupled with
labeling techniques such as dimethylation, SILAC or iTRAQ, also
enables quantitative analysis of bacterial glycosylation. The development of this robust and selective approach to identify glycoproteins has been essential in expanding our understanding of
glycosylation within several bacterial species.
358
Materials
2.1 Generation
of Bacterial Cells
1. Bacterial strain of interest.

2. Growth medium for bacterial strain of interest.
3. Glycopeptide enrichment control, such as bovine fetuin or for
more complex mixtures use C. jejuni (e.g. strain NCTC11168
or 81-176) or Acinetobacter baumannii (e.g. strain 17978 or
19606).
4. Phosphate-buffered saline (PBS), pre-chilled at 4 C before
use.
2.2 Preparation
of Protein Samples
1. Trypsin-compatible buffer solution, e.g. 40 mM ammonium

bicarbonate (ABC), pH 8.0. Pre-chilled at 4 C before use.
2.2.1 Whole Cell Lysates
2. Benzonase nuclease (250 units/L, 90 %), store at 4 C.

3. Protease inhibitor cocktail (e.g. cOmplete Mini, Roche Diagnostics).
4. Probe Sonicator with micro-tip probe (e.g. Branson Ultrasonics).
2.2.2 MembraneAssociated Proteins
1. Trypsin-compatible buffer solution, pH 8.0. Pre-chilled at

4 C before use.
2. Benzonase nuclease (250 units/L, 90 %) Sigma-Aldrich,
Store at 4 C.
3. 0.1 M sodium carbonate, pre-chilled at 4 C before use.
4. Protease inhibitor cocktail (e.g. cOmplete Mini, Roche
Diagnostics).
5. Probe Sonicator with micro-tip probe (e.g. Branson Ultrasonics).
2.3 Trypsin Digestion

of Proteins
1. 6 M urea, 2 M thiourea, prepare fresh prior to use (see Note 1).

2. 100 mM ABC. Prepare fresh, ensure pH ~8.0 prior to use.
3. 1 M dithiothreitol (DTT) in 100 mM ABC, pH 8.0. Prepare
fresh and store at 4 C.
4. Iodoacetamide (100 mM in 100 mM ABC pH 8.0). Prepare
fresh and stored in the dark at 4 C.
5. 0.1 g/L Lys-C (Wako Pure Chemical Industries) in Milli-Q
water. Store as single use aliquots at 80 C.
6. 0.1 g/L Trypsin (porcine sequencing grade) in 100 mM
ABC, pH 8.0. Store as single use aliquots at 80 C.
2.4 Desalting
of Peptides
1. For large amounts (>200 g) of peptide material use Oasis

hydrophiliclipophilic balance (HLB) solid-phase extraction
(SPE) cartridges (Waters Corporation).
2. For small amounts (<200 g) of peptide material use STOP
and GO tips (STAGE) constructed according to the published
method [20].
359
3. Acetonitrile (MeCN, CHROMASOLV gradient grade for

HPLC 99.9 % purity, Sigma Aldrich).
4. Buffer A*: 0.1 % trifluoroacetic acid (TFA, CHROMASOLV
for HPLC 99.0 % purity, Sigma Aldrich) in Milli-Q water
(see Note 2).
5. Buffer B*: 0.1 % TFA in 80 % MeCN.
6. Vacuum centrifuge.
2.5 Construction
of ZIC-HILIC STAGE
Tips
1. 20 gauge, Kel-F Hub NDL blunt needle (Hamilton Company).

2. Plunger assembly N, RN, LT, LTN 1701/10 L (Hamilton
Company).
3. ZIC-HILIC material (ZIC-HILIC, 10 m, 200 , SeQuant).
4. C8 material (Empore Octyl C8 extraction disk, 3 M) (SigmaAldrich) (see Note 3).
5. P10 pipette tip.
6. 10 mL syringe.
2.6 ZIC-HILIC
Enrichment
of Glycopeptides
1. ZIC-HILIC buffer A: 0.1 % formic acid (FA, ~98 % purity) in

Milli-Q water. Prepare fresh on day of the enrichment.
2. ZIC-HILIC buffer B: 5 % FA in 80 % MeCN. Prepare fresh on
day of the enrichment (see Note 4).
3. ZIC-HILIC buffer C: 95 % MeCN. Prepare fresh on day of
the enrichment.
4. Vacuum centrifuge.
2.7
LC-MS/MS
1. Mass spectrometer capable of resonance-based and beam-type

collision-induced dissociation (referred to as CID and HCD
fragmentation, respectively) e.g. LTQ Orbitrap Velos/Elite or
Fusion instruments (Thermo Scientific) (see Note 5).
2. Nanoflow HPLC system (e.g. Agilent 1100/1200 series or
Thermo Scientific EasyLC systems).
3. 96-well autosampler plate.
4. Pre/Trapping column: 20 mm, 100 m inner diameter (i.d.),
360 m outer diameter (o.d.) fused silica (Polymicro
Technologies), packed with Aqua C18 reversed phase (RP)
material (5 m, Phenomenex).
5. Analytical column: 300 mm, 75 m i.d. 360 m o.d fused silica
packed with Reprosil-Pur C18-AQ RP material (3 m, Dr.
Maisch).
6. Conductive/coated fused silica emitters, 20 m (i.d.); 360 m
(o.d.), either pulled in-house or purchased from New
Objective.
7. Buffer A: 0.1 % FA in Milli-Q water.
360
8. Buffer A*: 0.1 % TFA in Milli-Q water.

9. Buffer B: 0.1 % FA in 80 % MeCN.
2.8
Data Analysis
1. MASCOT (Matrix Science) search engine for peptide identification (proprietary software).
2. Expert Annotation website (http://www.biochem.mpg.de/
mann/tools/).
3. Xcalibur MS data viewer (Thermo Scientific).
4. GPMAW 8.2 (Lighthouse Data; proprietary software).
5. Proteome Discoverer (Thermo Scientific; proprietary software).
6. Microsoft Excel (Microsoft; proprietary software).
3
3.1
Methods
Bacterial Growth
1. Grow bacterial strain of interest to the required phase or density and harvest cells by centrifugation at 2,000 g for 10 min
at 4 C.
2. Wash cell pellets two times with ice-cold PBS and collect cells
by centrifugation at 2,000 g for 10 min at 4 C.
3. Freeze dry cells and store at 80 C for long-term storage. To
ensure adequate material, >100 mg of freeze-dried cells should
be generated.
3.2 Preparation
of Protein Samples
3.2.1 Whole Cell Lysates
1. Weigh freeze-dried cells (for gram-negative bacteria use

1020 mg).
2. Resuspend cells in 1 mL ice-cold trypsin-compatible buffering
solution with protease inhibitor cocktail. Vortex to disperse
cells in solution. Keep samples on ice.
3. Add 250 U of Benzonase nuclease (see Note 6).
4. Lyse cells using a tip-probe sonicator with five rounds of 30 s
with 1 min on ice between each round (see Note 7).
5. Remove cellular debris by centrifugation at 20,000 g for
30 min at 4 C.
6. Collect supernatant, quantify protein amount and aliquot
1 mg protein lysate stocks.
7. Snap freeze and freeze dry aliquots, store at 80 C for longterm storage.
3.2.2 MembraneAssociated Proteins
1. Weigh freeze-dried cells (for gram-negative bacteria use

4080 mg).
2. Complete steps 25 as in Subheading 3.2.1.
361
3. Resuspend cellular debris in 1 mL ice-cold trypsin-compatible

buffering solution and sonicate samples as in Subheading 3.2.1,
step 4. Repeat five times (see Note 8).
4. Pool the supernatants, add 6 volumes of ice-cold 0.1 M sodium
carbonate and stir mixture gently for 1 h at 4 C.
5. Collect precipitated membranes by centrifugation at 35,000 g
for 1 h at 4 C.
6. Remove and discard the supernatant and wash the membrane
pellet vigorously with 20 mL of ice-cold trypsin-compatible
buffering solution. Repeat step 5. Repeat steps 6 and 5.
7. Resuspend the membrane pellet in 10 mL ice-cold trypsincompatible buffering solution by tip-probe sonication, quantify protein amount and aliquot 1 mg stocks.
8. Snap freeze and freeze dry aliquots, store at 80 C for longterm storage.
3.3 Trypsin Digestion
of Proteins
1. Resuspend 1 mg dried protein lysate in 200 L of 6 M urea,

2 M thiourea and vortex.
2. Add 2 L of 1 M DTT (final concentration 10 mM), vortex
and allow sample to incubate at room temperature for 1 h in
the dark.
3. Add 50 L of 100 mM iodoacetamide (20 mM final concentration), mix and incubate in the dark for 1 h.
4. Quench excess IAA by addition of 2 L 1 M DTT, incubate for
20 min in the dark.
5. Add 5 g Lys-C (1/200 w/w) and incubate samples for 4 h at
25 C.
6. Dilute sample 1:5 by addition of 1,100 L 100 mM ABC (final
volume 1,400 L) and incubate for 24 h with 40 g trypsin at
room temp in the dark (1/25 w/w) (see Note 9).
3.4 Desalting
of Peptides (See Note 10)
1. Acidify digests with 30 L acetic acid per mL of sample, vortex

to mix.
2. Centrifuge at 10,000 g for 10 min to remove insoluble
material.
3. Prepare SPE columns by washing with 1 mL 100 % MeCN for
Oasis HLB SPE cartridges or 50 L for STAGE tips.
4. Wash SPE column with two rounds of 1 mL Buffer B* for
5. Wash SPE column with two rounds of 1 mL Buffer A* for
6. Load acidified sample onto prepared SPE column.
7. Wash SPE column with two rounds of 1 mL Buffer A* for
362
Fig. 1 Construction of ZIC-HILIC STAGE tips. (a) Excision of C8 disk, using a Kel-F Hub NDL blunt needle generating a 2 mm C8 disk; (b) Insertion of C8 disk into P10 tip using the Plunger assembly enables the C8 disk to be
lodged at the end of the P10 tip; (c) Packing of ZIC-HILIC material onto the C8 disk [I) shows an unpacked tip;
II) a tip with only the C8 disk; and III) the final ZIC-HILIC tip to be used for glycopeptide enrichment containing
0.5 cm of packed ZIC-HILIC material]
8. Elute peptides into an Eppendorf tube with 1 mL Buffer B*

for Oasis HLB SPE Cartridges or 50 L for STAGE tips.
9. Dry sample by vacuum centrifugation.
3.5 Construction
of ZIC-HILIC STAGE
Tips
1. Using a 20-Gauge, Kel-F Hub NDL blunt needle, excise a C8

disk (Fig. 1a).
2. Eject the C8 disk into a P10 tip using the Plunger assembly
(Fig. 1b).
3. Resuspend the ZIC-HILIC resin in Milli-Q water and load
resin on C8 disk.
4. Pack ZIC-HILIC resin to a height of 0.5 cm using pressure
from a 10 mL syringe (Fig. 1c).
3.6 ZIC-HILIC
Enrichment
of Glycopeptides
1. Resuspend peptides in ZIC-HILIC buffer B, 200 g of peptide to 100 L of ZIC-HILIC buffer B and vortex. Store at
4 C.
2. Prepare ZIC-HILIC column for glycopeptide binding, wash
with four washes of 50 L ZIC-HILIC buffer A, four washes
of 50 L ZIC-HILIC buffer C and four washes of 50 L ZICHILIC buffer B (see Note 11).
3. Load peptides onto ZIC-HILIC STAGE tip column and wash
ten times with 50 L ZIC-HILIC buffer B (see Note 12).
4. Elute glycopeptides with three washes of ZIC-HILIC buffer A.
5. Dry eluted glycopeptides by vacuum centrifugation and store
at 20 C prior to analysis.
3.7
LC-MS/MS
1. For optimal glycopeptide analysis, utilize a nanoflow LC setup

directly coupled, via an electrospray (ESI) ion source, to a mass
spectrometer (see Note 13).
363
2. Resuspend the dried samples from Subheading 3.6 in 12 L

buffer A* and transfer to a 96-well autosampler plate.
3. Inject 5 L of the glycopeptide sample onto the nanoflow LC
system.
4. Load samples onto the pre-/trapping column at 5 L/min for
10 min with buffer A.
5. Separate glycopeptides using the analytical column by altering
the gradient from 100 % buffer A to 40 % buffer B over
148 min, followed by 40 % buffer B to 80 % buffer B over
12 min. Wash the column for 10 min with 100 % buffer B to
remove additional material retained on the column and to condition it for the next injection by washing for a further 10 min
with 100 % buffer A. Use a constant flow of 250 nL/min for
all steps and infuse the elution into the mass spectrometer with
a coated fused-silica emitter.
6. Operate the mass spectrometer (e.g. LTQ Orbitrap Velos/
Elite or Fusion) in data-dependent acquisition mode, automatically switching between MS and MS/MS selecting the 5 most
intense ions. Acquire the full-scan MS scan in the Orbital trap
at a resolution of R = 60,000 (at 400 m/z) between 350 and
1,800 m/z (see Note 14). Set the automatic gain control
(AGC) target to 1,000,000 ions and a maximum injection
time of 50 ms. Isolate the precursor ions of interest with an ion
selection width of 2.0 m/z and perform CID fragmentation
followed by HCD fragmentation on the same ion (see Note 15).
CID fragmentation (normalized collision energy 35) is accomplished within the linear ion trap with an AGC target of 2 104
and maximum fill time of 100 ms. HCD fragmentation (normalized collision energy 40) is carried out with an AGC of
2 105, maximum fill time of 250 ms, resolution set to 7,500
and mass window 2002,000 m/z. Dynamic exclusion should
be enabled with the exclusion window reflecting the width and
time of observed chromatography peaks.
3.8
Data Analysis
1. Prior to analysis of data files, the presence of MS/MS scans

representing glycopeptides can be assessed by performing
extracted ion chromatograms of a diagnostic carbohydrate ion,
such as the oxonium 204.086 m/z ion of N-acetylhexosamines
(Fig. 2a).
2. Generate MASCOT generic files (.mgf) for each .RAW file
with Proteome Discoverer (see Note 16).
3. Search the .mgf files against an appropriate database and decoy
database using the MASCOT engine through Proteome
Discoverer. Set fixed modifications to carbamidomethylation
and variable modifications to methionine oxidation and deamidation of asparagine/glutamine. Allow two missed cleavages,
Fig. 2 Data analysis of ZIC-HILIC enriched glycopeptides. (a) Extracted ion chromatogram of the diagnostic oxonium ion 204.07204.09 m/z, the glycopeptide
979.46704 m/z, charge state +3, is highlighted in red; (b) CID fragmentation of the
glycopeptide 979.46704 m/z containing the glycan HexNAc-Hex2-HexNAc-258;
365
a precursor mass tolerance of 10 ppm and an MS/MS fragment mass tolerance of 0.05 Da for HCD scans and 0.6 Da
for CID scans.
4. Export MS/MS scan events that do not result in peptide identification to Microsoft Excel.
5. Use the mgf graph module in GPMAW 8.2 to identify possible glycopeptides in this list that contain the diagnostic oxonium 204.086 m/z ion (see Note 17).
6. Manually inspect the CID and HCD scan events of each potential glycopeptide using Xcalibur to confirm that these scans of
the same peptide are dissimilar (Fig. 2b, c).
7. For HCD scan events, the deglycosylated peptide and the peptide plus one carbohydrate are typically the most abundant
ions and can be used to determine a putative deglycosylated
mass of the glycopeptide (Fig. 2d).
8. To facilitate a glycopeptide assignment from an HCD scan,
ions below the mass of the predicted deglycosylated peptide
are extracted within Xcalibur using the Spectrum List function.
Ions with a deconvoluted mass above that of the deglycosylated peptide and those corresponding to known carbohydrate
oxonium ions are removed and the resulting peak list used to
generate a .mgf file that can be searched with MASCOT, using
the same parameters defined for the initial search (step 3).
9. To further validate glycopeptide matches, HCD scan events
are annotated using the Expert Annotation tool (Fig. 2e).
Notes
1. Avoid heating urea/thiourea solution to decrease the risk of
protein carbamylation [21].
2. Decant TFA or use glass pipette, avoid the introduction of
plastic tips into the stock solution.
3. An alterative method for the construction of ZIC-HILIC columns using constricted GELoader tips in place of C8 support
has also been reported, for instructions see [22].
4. Alterative ion pairing agents can be used and have been shown
to augment the efficiency of HILIC enrichment [23]. With
certain samples these agents can be more effective than FA and
should be considered during optimization.
Fig. 2 (continued) (c) HCD fragmentation of the glycopeptide 979.46704 m/z

demonstrating the deglycosylated peptide 301AKPASTPAVK310; (d) Annotation of the
deglycosylated peptide 301AKPASTPAVK310 using the Expert Annotation software
366
5. For complete assignment of glycopeptides (absolute site localization independent of any sequence motif constraints), CID,
HCD and electron transfer dissociation (ETD) approaches can
be used provided the instrumentation supports these fragmentation techniques. Alternatively, if only the peptide sequence
and confirmation of glycosylation are required, beam-type
fragmentation on Q-Tof or Q-Exactive instrumentation may
be used.
6. An alterative to Benzonase nuclease is Cryonase (Clontech
Laboratories Inc., Takara Bio Group), which provides enhanced
nuclease digestion at 4 C.
7. Ensure the sample remains chilled during sonication and is at
4 C before beginning another round of sonication.
8. With each round of sonication the amount of extracted protein
should decrease, for most gram-negative bacteria we have
examined 5 rounds results in effective solubilization of membrane material.
9. Alternative enzymes may also be useful for generation of glycopeptides of suitable length conducive to their analysis
(e.g. GluC, AspN). Importantly, due to the steric hindrance
provided by larger glycans, non-specific proteinases such as
thermolysin, pepsin and proteinase-K can also be used to
generate peptides of adequate length to enable glycopeptide
identification [24]. If these enzymes are used, the corresponding digestion buffer may also require alteration to
ensure maximum efficiency.
10. The presence of salts and urea/thiourea severely compromises
the enrichment of glycopeptides using ZIC-HILIC. For optimal results ensure the effective clean-up of peptide samples.
11. Do not allow the ZIC-HILIC resin to become dry during the
glycopeptide enrichment process, as this appears to severely
compromise the final glycopeptide yield.
12. For complex peptide lysates, 10 washes with ZIC-HILIC buffer B is effective for the removal of the majority of nonglycosylated peptides, but for less complex samples such as
single proteins, 3 washes are typically sufficient. The number of
wash steps should be varied to assess the effect on the enrichment of glycopeptides for any new sample type.
13. A microflow LC system can also be used but it should be noted
that the use of high flow rates would lead to a loss of sensitivity.
14. As the addition of a glycan increases the mass of a peptide but
does not increase the observed charge state, to increase the
chances of selecting a glycopeptide the lower threshold of the
MS1 mass window can be increased to 500 m/z, thus
5001,800 m/z.
367
15. Alternative combinations of fragmentation may be used, such as

ETD to gain information about the site of glycosylation. In
addition to these fragmentation approaches, dynamic datadependent methods can also be used, such as using the presence
of diagnostic ions within HCD scans to trigger ETD [25, 26].
16. An alterative to Proteome Discoverer for the generation of .mgf
files is MSconvert (ProteoWizard), an open source and freely available software package http://proteowizard.sourceforge.net/.
17. As N-acetylhexosamines are common units within glycopeptides, the use of the diagnostic oxonium 204.086 m/z ion is
appropriate but it should be noted that some glycans might
not contain N-acetylhexosamines. In these cases, information
about the glycan composition can be used to determine a more
appropriate diagnostic ion.
Acknowledgments
This work was supported by an Australian Research Council
Discovery Project Grant to S.J.C. (ARC DP 110103573). N.E.S.
is supported by a National Health and Medical Research Council
of Australia (NHMRC) Overseas (Biomedical) Early Career
Fellowship (APP1037373) and a Michael Smith Foundation for
Health Research Trainee Postdoctoral Fellowship (award # 5363).
References
1. Iwashkiw JA, Vozza NF, Kinsella RL, Feldman
MF (2013) Pour some sugar on it: the expanding world of bacterial protein O-linked glycosylation. Mol Microbiol 89:1428
2. Nothaft H, Szymanski CM (2010) Protein
glycosylation in bacteria: sweeter than ever.
Nat Rev Microbiol 8:765778
3. Szymanski CM, Burr DH, Guerry P (2002)
Campylobacter protein glycosylation affects
host cell interactions. Infect Immun 70:
22422244
4. Howard SL, Jagannathan A, Soo EC, Hui JP,
Aubry AJ, Ahmed I, Karlyshev A, Kelly JF,
Jones MA, Stevens MP, Logan SM, Wren BW
(2009) Campylobacter jejuni glycosylation
island important in cell charge, legionaminic
acid biosynthesis, and colonization of chickens. Infect Immun 77:25442556
5. Iwashkiw JA, Seper A, Weber BS, Scott NE,
Vinogradov E, Stratilo C, Reiz B, Cordwell SJ,
Whittal R, Schild S, Feldman MF (2012)
Identification of a general O-linked protein glycosylation system in Acinetobacter baumannii
6.
7.
8.
9.
and its role in virulence and biofilm formation.

PLoS Pathog 8:e1002758
Lithgow KV, Scott NE, Iwashkiw JA, Thomson
ELS, Foster LJ, Feldman MF, Dennis JJ
(2014) A general protein O-glycosylation system within the Burkholderia cepacia complex
is involved in motility and virulence. Mol
Microbiol 92:116137
Zielinska DF, Gnad F, Wisniewski JR, Mann
M (2010) Precision mapping of an in vivo
N-glycoproteome reveals rigid topological and
sequence constraints. Cell 141:897907
Zhang H, Li X, Martin DB, Aebersold R
(2003) Identification and quantification of
N-linked glycoproteins using hydrazide chemistry, stable isotope labeling and mass spectrometry. Nat Biotechnol 21:660666
Fredriksen L, Moen A, Adzhubei AA, Mathiesen
G, Eijsink VG, Egge-Jacobsen W (2013)
Lactobacillus plantarum WCFS1 O-linked protein glycosylation: an extended spectrum of target
proteins and modification sites detected by mass
spectrometry. Glycobiology 23:14391451
368
10. Nothaft H, Scott NE, Vinogradov E, Liu X,

Hu R, Beadle B, Fodor C, Miller WG, Li J,
Cordwell SJ, Szymanski CM (2012) Diversity
in the protein N-glycosylation pathways within
the Campylobacter genus. Mol Cell Proteomics
11:12031219
11. Thibault P, Logan SM, Kelly JF, Brisson JR,
Ewing CP, Trust TJ, Guerry P (2001)
Identification of the carbohydrate moieties and
glycosylation motifs in Campylobacter jejuni flagellin. J Biol Chem 276:3486234870
12. Grass S, Lichti CF, Townsend RR, Gross J, St
Geme JW 3rd (2010) The Haemophilus influenzae HMW1C protein is a glycosyltransferase
that transfers hexose residues to asparagine
sites in the HMW1 adhesin. PLoS Pathog
6:e1000919
13. Verma A, Schirm M, Arora SK, Thibault P,
Logan SM, Ramphal R (2006) Glycosylation
of b-Type flagellin of Pseudomonas aeruginosa:
structural and genetic basis. J Bacteriol 188:
43954403
14. Scott NE, Nothaft H, Edwards AV, Labbate
M, Djordjevic SP, Larsen MR, Szymanski CM,
Cordwell SJ (2012) Modification of the
Campylobacter jejuni N-linked glycan by
EptC-mediated addition of phosphoethanolamine. J Biol Chem 287:2938429396
15. Twine SM, Reid CW, Aubry A, McMullin DR,
Fulton KM, Austin J, Logan SM (2009) Motility
and flagellar glycosylation in Clostridium difficile. J Bacteriol 191:70507062
16. Gault J, Malosse C, Dumenil G, ChamotRooke J (2013) A combined mass spectrometry
strategy for complete posttranslational modification mapping of Neisseria meningitidis major
pilin. J Mass Spectrom 48:11991206
17. Young NM, Brisson JR, Kelly J, Watson DC,
Tessier L, Lanthier PH, Jarrell HC, Cadotte
N, St Michael F, Aberg E, Szymanski CM
(2002) Structure of the N-linked glycan present on multiple glycoproteins in the Gramnegative bacterium, Campylobacter jejuni. J
Biol Chem 277:4253042539
18. Scott NE, Parker BL, Connolly AM, Paulech J,
Edwards AV, Crossett B, Falconer L, Kolarich
D, Djordjevic SP, Hjrup P, Packer NH, Larsen
MR, Cordwell SJ (2011) Simultaneous glycanpeptide characterization using hydrophilic
interaction chromatography and parallel frag-
19.
20.
21.
22.
23.
24.
25.
26.
mentation by CID, higher energy collisional

dissociation, and electron transfer dissociation
MS applied to the N-linked glycoproteome of
Campylobacter jejuni. Mol Cell Proteomics
10:M000031MCP000201
Scott NE, Kinsella RL, Edwards AV, Larsen
MR, Dutta SM, Saba J, Foster LJ, Feldman MF
(2014) Diversity within the O-linked protein
glycosylation systems of Acinetobacter species.
Rappsilber J, Mann M, Ishihama Y (2007)
Protocol for micro-purification, enrichment,
pre-fractionation and storage of peptides for
proteomics using StageTips. Nat Protoc
2:18961906
Kollipara L, Zahedi RP (2013) Protein carbamylation: in vivo modification or in vitro artefact? Proteomics 13:941944
Gobom J, Nordhoff E, Mirgorodskaya E,
Ekman R, Roepstorff P (1999) Sample purification and preparation technique based on
nano-scale reversed-phase columns for the
sensitive analysis of complex peptide mixtures
by matrix-assisted laser desorption/ionization mass spectrometry. J Mass Spectrom
34:105116
Mysling S, Palmisano G, Hjrup P, ThaysenAndersen M (2010) Utilizing ion-pairing
hydrophilic interaction chromatography solid
phase extraction for efficient glycopeptide
enrichment in glycoproteomics. Anal Chem
82:55985609
Larsen MR, Hojrup P, Roepstorff P (2005)
Characterization of gel-separated glycoproteins using two-step proteolytic digestion
combined with sequential microcolumns and
mass spectrometry. Mol Cell Proteomics
4:107119
Wu SW, Pu TH, Viner R, Khoo KH (2014)
Novel LC-MS(2) product dependent parallel
data acquisition function and data analysis workflow for sequencing and identification of intact
glycopeptides. Anal Chem 86:54785486
Saba J, Dutta S, Hemenway E, Viner R (2012)
Increasing the productivity of glycopeptides
analysis by using higher-energy collision
dissociation-accurate mass-product-dependent
electron transfer dissociation. Int J Proteomics
2012:560391
Chapter 26
In-Gel Peptide IEF Sample Preparation for LC/MS Analysis
Tom Berkelman, Sricharan Bandhakavi, and Aran Paulus
Abstract
The technique of proteolytically digesting a sample and identifying its protein components by liquid
chromatography followed by mass spectrometry (LC-MS) is a widely used analytical tool. Prior fractionation by isoelectric focusing (IEF) may be performed to increase the depth of proteome coverage. Here,
we describe a method for in-gel IEF separation of a proteolytic digest that utilizes commercially available
immobilized pH gradient (IPG) strips and a widely used IEF instrument.
Key words LC-MS, Isoelectric focusing, IEF, Proteomics, Off-gel, In-gel, Immobilized pH
gradient
Introduction
The separation of proteolytically digested samples by reversedphase liquid chromatography coupled to mass spectrometry with
electrospray ionization (LC-MS) is a powerful and widely used
strategy for proteomic analysis. This technique can detect thousands of individual peptides, and by referencing a database, identify
their proteins of origin. Termed bottom-up or shotgun proteomics, this is a powerful profiling method for proteins present in
a biological sample [1]. Additional depth of coverage can be
obtained by preceding LC-MS with an orthogonal upstream fractionation [2] generating fractions of lower complexity that are analyzed separately. Orthogonal separation techniques that are used for
this purpose include strong cation exchange chromatography (SCX)
[3], hydrophilic interaction chromatography (HILIC) [4], and isoelectric focusing (IEF) [58]. The depth of proteome coverage
obtainable with IEF can be higher than that obtained with chromatographic methods [9, 10] and IEF has become widely used for
this application. IEF has the additional benefit of providing a search
constraint for peptide identification, since the isoelectric point (pI)
of a peptide can be predicted directly from its sequence [1113].
IEF fractionation of sample digests is predominantly performed
369
370
Tom Berkelman et al.
using immobilized pH gradient (IPG) strips due to the reproducibility and versatility afforded by commercial pre-cast IPG gels covering various pH ranges. The separation may either be performed
in an off-gel mode or in an in-gel mode. In the off-gel
mode, the digest is electrophoretically fractionated through a series
of sample cups placed in contact with the IPG gel [8, 14]. The
separated peptides are recovered in solution directly from the sample cups. In the in-gel mode, the digest is loaded into an IPG gel
and separated within the gel matrix [10, 12, 13, 1517]. The separated peptides are eluted from excised sections of IPG gel. Whereas
the off-gel method offers a certain degree of convenience, in-gel
separations have several advantages. Peptide yields have been found
to be higher and the separation is more rapid [18]. Additionally,
in-gel instrumentation allows greater flexibility in the selection of
IPG strip length and pH range. There is also no restriction with
in-gel separations on the proportion of the length of IPG strip
represented by each fraction. An off-gel fraction, by necessity, represents a section of the gel corresponding to the footprint of a
single sample cup. The pH range of an in-gel separation is determined by the spacing of the cuts made. The number of fractions
generated can therefore be tailored more closely to the needs of
the experimenter. Since the distribution of peptide pI is variable
across the pH scale, it can be advantageous to cut the IPG strip
into sections of variable length such that each eluted fraction contains a similar number of peptides, thereby taking fullest advantage
of the downstream analysis methodology [19]. This strategy is not
possible with off-gel IEF. The following method describes in-gel
preparative IEF using the PROTEAN i12 IEF System, a widely
used instrument which controls voltage and current independently
to each IPG strip, allowing multiple pH ranges, or samples of differing conductivity to be run simultaneously.
Materials
Reagents and prepared buffer stocks are from Bio-Rad or SigmaAldrich except when indicated. Reagent quality should be
reagent grade or better. Water should be of the highest purity
possible. Measures should be taken to avoid keratin contamination (see Note 1).
2.1 Lysate
Preparation
(See Note 2)
1. Phosphate Buffered Saline (PBS): 150 mM NaCl, 10 mM

sodium phosphate, pH 7.8. Prepare by tenfold dilution of a
stock solution.
2. Lysis Buffer: 8 M urea, 50 mM TrisHCl, pH 8.0. Weigh
4.805 g of urea and bring to about 9 ml with water. Add 0.5 ml
1 M TrisHCl, pH 8.0. Vortex to dissolve the urea and adjust
the volume to 10 ml (see Note 1).
3. Protein assay kit (e.g. DC protein assay kit from Bio-Rad).
In-Gel Peptide IEF Fractionation
2.2 Reduction
and Alkylation
371
1. 1.2 M dithiothreitol (DTT): Weigh 185.1 mg of DTT and dissolve in sufficient water for 1 ml of solution. The solution may
be stored frozen in aliquots.
2. 600 mM iodoacetamide: Weigh 111.0 mg of iodoacetamide
and dissolve in sufficient water for 1 ml of solution.
2.3
Digestion
1. Lys-C (lysyl endopeptidase) (Wako Chemicals, USA).

2. Trypsin (sequencing grade modified trypsin) (Promega, USA).
3. 50 mM ammonium bicarbonate: Weigh 395 mg of ammonium
bicarbonate and dissolve in 100 ml water (see Note 1).
2.4 Desalting
and Concentration
of Peptide Digest
1. Empore C18 solid phase extraction cartridges.

2. 0.1 % TFA in water.
3. 0.1 % TFA in acetonitrile.
4. 0.1 % TFA, 70 % acetonitrile.
2.5 In-Gel IEF

Fractionation
1. 11 cm pH 310 ReadyStrip IPG strips (Bio-Rad).

2. 8 M urea, 0.2 % (w/v) Bio-Lyte 3-10: Weigh 4.805 g of urea
and bring to about 9 ml with water. Add 50 l Bio-Lyte,
pH 310 (Bio-Rad, supplied at 40 %). Bring volume to 10 ml
with water (see Note 1).
3. PROTEAN i12 IEF System, including 11 cm Focusing Tray
and 11 cm Rehydration/Equilibration Tray.
4. Mineral oil.
5. Electrode wicks for gel-side-up configuration.
2.6
Peptide Elution
1. Hexane.
2. Razor blades.
3. 0.1 % TFA, 35 % acetonitrile: Mix 6.5 ml of 0.1 % TFA in water
with 3.5 ml of 0.1 % TFA in acetonitrile.
4. 0.1 % TFA, 80 % acetonitrile: Mix 2 ml of 0.1 % TFA in water
with 8 ml of 0.1 % TFA in acetonitrile.
2.7 Solid Phase

Extraction Clean-up
of Peptide Fractions
1. Empore SDB-XC solid phase extraction disk.

2. Blunt-ended needle, 17 gauge.
3. 0.1 % TFA in water.
4. 0.1 % TFA in 80 % acetonitrile.
Methods
3.1 Lysate
Preparation
1. Grow HeLa cells in 10 cm plates to a density of approximately

3 106 cells per plate. After washing three times with cold
phosphate-buffered saline, flash freeze cell pellets and store at
80 C (see Note 2).
372
2. Resuspend the cell pellet in Lysis Buffer to a concentration of

approximately 37.5 107 cells/ml of Lysis Buffer. Sonicate
briefly and clarify by centrifugation (20,000 g, 5 min).
Determine the protein concentration by DC protein assay
using IgG as a standard. Protein concentration should be at
least 5 mg/ml.
3.2 Reduction
and Alkylation
1. Transfer a portion of the lysate containing 2 mg protein and

bring the volume to 400 l with additional Lysis Buffer (the
final protein concentration will be 5 mg/ml see Note 3).
2. Add 6.7 l 1.2 M DTT and mix (final concentration is 20 mM).
Incubate at 20 C for 1 h (see Note 4).
3. Add 81 l 600 mM iodoacetamide and mix (final concentration is 100 mM). Incubate at 20 C for 30 min.
3.3
Digestion
1. Resuspend two 20 l vials of Lys-C in a total of 512 l of Lysis

Buffer. Add the entire volume of Lys-C solution to the entire
volume of reduced and alkylated lysate. Incubate overnight at
20 C (see Note 5).
2. Resuspend a 20 l vial of trypsin in 3 ml 50 mM ammonium
bicarbonate. Add the entire volume of trypsin solution to the
Lys-C digest and incubate at 20 C for an additional 4 h.
3.4 Desalting
and Concentration
of Peptide Digest
1. Equilibrate four 3 ml Empore C18 solid phase extraction cartridges for de-salting. Place each cartridge in a 15 ml conical
centrifuge tube. Apply 1 ml of methanol to each column and
centrifuge at 1,500 g for 1 min in a fixed-angle rotor. Discard
the flow-through and repeat with 0.5 ml 0.1 % TFA, 70 % acetonitrile followed by 0.5 ml 0.1 % TFA in water.
2. Apply one-quarter of the digested sample to each column
(roughly 1 ml each) and centrifuge at 135 g for 10 min.
Discard the flow-through.
3. Wash each column three times with 0.5 ml 0.1 % TFA, centrifuging as in step 2 and discarding the flow-through after each
centrifugation.
4. Elute the peptides from each column with 0.5 ml 0.1 % TFA,
70 % acetonitrile, centrifuging as in step 2 and reserving the
eluates. Repeat the elution and pool all of the eluates.
5. Determine the absorbance at 280 nm of the pooled eluates,
using 0.1 % TFA, 70 % acetonitrile to blank the spectrophotometer. Estimate the peptide concentration and yield by
assuming that a 1 mg/ml peptide solution has an absorbance
at 280 nm of 1.1 (see Note 6).
6. Dry the peptides down by vacuum centrifugation.
3.5 In-Gel IEF

Fractionation
373
1. Resuspend the peptides in 8 M urea, 0.2 % (w/v) Bio-Lyte

3-10 to a concentration of 250 g/ml.
2. For each separation run, pipet 200 l of the peptide solution
along the center of the channel of the i12 rehydration/
equilibration tray. Take care to not introduce air bubbles when
expelling the solution (see Notes 710).
3. Using forceps, remove the cover sheet from each IPG strip, then
gently place the IPG strip gel-side down onto the solution in the
channel. Move the IPG strip back and forth slightly to ensure
that the solution is distributed along the length of the IPG strip.
Take care to not trap air bubbles beneath the IPG strip.
4. Overlay each IPG strip with 5 ml mineral oil. Apply the mineral oil to both ends of the channel and allow it to flow toward
the middle of the channel (see Note 11).
5. Cover the tray and leave it on a level bench overnight
(1218 h) for complete rehydration.
6. Using forceps, remove the IPG strips from the rehydration
tray, remove excess mineral oil, and place the rehydrated IPG
strips gel-side up in the channels of the focusing tray. Position
the positive (+) ends of the IPG strips against the positioning
stops in each channel.
7. Wet the gel-side up wicks (notched) with distilled or deionized
water and blot off excess water. Use two wicks per IPG strip:
place a wick at each end of each IPG strip. Position the electrode assemblies in the focusing tray and press down on the
green tabs to snap the electrode assemblies into place.
8. Place the focusing tray with the rehydrated IPG strips on the
Peltier platform and connect the electrodes to the instrument.
Overlay each IPG strip with 5 ml mineral oil.
9. Run IEF according to the following protocol: 500 V for
30 min, 1,000 V for 30 min, linear ramp to 8,000 V over 1 h,
8,000 V for 6 h. Maintain a current limit of 50 A throughout
(see Note 12).
3.6
Peptide Elution
1. Following electrophoresis, briefly rinse each IPG strip in a beaker of hexane (see Note 13).
2. Print out a cutting template consisting of 12 parallel 1 cm lines
placed 9 mm apart from one another (see Fig. 1).
3. Place a clean sheet of glass over the template. Place each IPG strip
gel-side up on the glass over the template as shown in Fig. 1. Cut
through the gel and backing over each line with a razor blade,
using a fresh blade for each cut, generating twelve 9 mm gel sections. Keep track of which section corresponds to which fraction.
4. Place each section in a 1.5 ml tube and discard the 5 mm pieces
from each end of the IPG strip (the sites of electrode contact).
374
Site of electrode
wick contact
1
pH 3
Site of electrode
wick contact
2
10
11
pH 10
Fig. 1 Use of template for cutting IPG strip to generate IEF fractions. Each IPG strip is placed on a clean sheet
of glass over a template comprising twelve parallel 1 cm lines placed 9 mm apart so that the ends of the IPG
gel extended 5 mm beyond the outermost lines. The IPG strip is cut over each line. Fraction numbers are
indicated
5. Add 150 l 0.1 % TFA in water to each tube. Incubate the tube
with the gel section and 0.1 % TFA for 1.5 h with gentle shaking. Remove the liquid and reserve (see Note 14).
6. Repeat step 5 using 0.1 % TFA, 35 % acetonitrile. Pool the
removed liquid with the eluted material from step 5.
7. Repeat step 6 using 0.1 % TFA, 70 % acetonitrile. Pool the
removed liquid with the eluted material from steps 5 and 6.
8. Dry the pooled material by vacuum centrifugation.
3.7 Solid Phase
Extraction Clean-up
of Peptide Fractions
1. Use a 17-gauge blunt-ended needle to punch small plugs of

material from an SDB-XC (styrene-divinylbenzene) solid phase
extraction disk (see Note 15).
2. Prepare extraction cartridges by packing the plugs of solid
phase extraction material into 200 l pipet tips, three per tip.
Push the plugs of material as far into the pipet tip as possible
using a fine wire or needle.
3. Place each packed tip in a 1.5 ml centrifuge tube through a
hole bored through the lid with a cork borer.
4. Equilibrate each cartridge with 90 l 0.1 % TFA in water. Pipet
90 l of solution into the cartridge and wash the liquid through
by centrifuging the cartridge and tube for 1 min at 1,000 g in
a fixed-angle rotor. Discard the flow-through liquid.
5. Dissolve each dry peptide fraction in 180 l 0.1 % TFA in
water.
6. Apply half of each sample (90 l) to each cartridge. Centrifuge
at 110 g for 10 min. Discard the flow-through liquid
(see Note 16).
7. Apply 90 l 0.1 % TFA in water to each cartridge. Centrifuge
as in step 6. Discard the flow-through liquid.
8. Elute the peptides with 90 l 0.1 % TFA in 80 % acetonitrile.
Apply 90 l of solution to each cartridge. Centrifuge as in step
6. Retain the eluted liquid.
9. Dry the eluted fractions by vacuum centrifugation. This material is now ready for analysis by LC-MS.
375
Notes
1. Clean technique is practiced throughout in order to minimize
keratin contamination. Solutions containing urea, iodoacetamide, ammonium bicarbonate or enzyme are prepared immediately before use.
2. The technique may be applied to any mammalian cell line cultured by any method.
3. The basic procedure may be scaled up or down, depending on
the number of IPG strips to be run.
4. Reduction and alkylation of cysteine thiols prior to digestion
and LC-MS enables more effective digestion and simplifies MS
analysis as all cysteine residues will be modified with an acetamido group. The sample is reduced with DTT and alkylated
with an excess of iodoacetamide.
5. A two-step digestion is employed utilizing Lys-C at an enzyme
to protein ratio of 1:50 followed by trypsin at an enzyme to
protein ratio of 1:100. Lys-C cleaves at lysine residues and is
compatible with 8 M urea and thus can be applied in undiluted lysis solution. Trypsin cleaves at both lysine and arginine residues and digests optimally only when the urea
concentration is 2 M or lower. The reaction is therefore
diluted prior to addition of trypsin. This procedure ensures
more complete digestion than trypsin alone. If desired, the
digest may be directly diluted and treated with trypsin, omitting the Lys-C digestion step, however digestion may not be
as efficient or complete.
6. The portion of the sample used to measure the absorbance at
280 nm should be added back to the pooled eluates following
the measurement. The cuvette used for the measurement
should therefore be thoroughly cleaned before use, or a disposable, UV transparent cuvette should be used.
7. Handle IPG strips from the ends using forceps.
8. Other pH ranges may be used.
9. Do not write on the IPG strip backing. The ink will be extracted
in subsequent steps and contaminate the sample.
10. The sample is applied to the entire length of the immobilized
pH gradient by rehydration into a dry IPG strip. As voltage is
applied, each peptide in the sample focuses to a narrow zone in
the pH gradient corresponding to its pI.
11. Mineral oil prevents evaporation of the sample solution.
12. Additional details of the sample application and IEF procedure
may be found in the user manual for the PROTEAN i12 IEF
System.
376
3500
Peptides unique to fraction
Peptide identifications
3000
Peptides found in other fractions
2500
2000
1500
1000
500
0
1
10
11
Fraction number
Fig. 2 Peptide distribution across the separation. Peptide fractions from a HeLa cell lysate were generated as
described. Each fraction was dissolved in 10 l 0.1 % formic acid, 2 % acetonitrile. Half of each fraction was
analyzed by LC-MS as described in [21]. Peptides found uniquely in a single fraction are indicated in a darker
shade. Peptides not restricted to a single fraction are indicated in a lighter shade (see Note 17)
13. The hexane rinse is to remove residual mineral oil prior to

peptide extraction.
14. Three sequential elutions are carried out with increasing
concentrations of acetonitrile. This procedure is based on
reference [20].
15. Commercial small-volume solid phase extraction cartridges
based on C-18-derivatized material may be used in place of the
cartridges described here.
16. The other half of each fraction (the remaining 90 l) may be
applied to another sold phase extraction cartridge if replicates
are desired.
17. Peptides are not evenly distributed across the pH gradient (see
Fig. 2). This has been noted in similar studies [8, 12, 13, 16].
18. In-gel fractionation was found to result in a higher number of
peptide identifications than off-gel fractionation performed
with the same pH gradient. Figure 3 shows the average pI of
peptides found in each fraction and Fig. 4 summarizes the distribution of LC-MS peptide identifications from the unfractionated digest and following two different IEF fractionation
methods. Among the possible reasons for this observation is
the fact peptide recovery from off-gel electrophoresis relies on
diffusion of the peptides out of the gel matrix during IEF.
Peptides with a tendency to remain within the gel matrix,
through ionic, hydrophobic or other interactions with the
377
10.00
9.00
Average pI
8.00
7.00
6.00
5.00
4.00
3.00
1
10
11
Fraction
Fig. 3 Average pI of peptides found in each fraction. The isoelectric point of each identified peptide was calculated from the peptide sequence using the compute pI/MW tool on the ExPasy server (www.expasy.org), which
is based on the method described in [22]. The average pI for each reaction is shown. Error bars show the
standard deviation within each fraction
Fig. 4 Distribution of LC-MS peptide identifications from the unfractionated digest and following two different
IEF fractionation methods. Identical quantities of digested HeLa cell lysate were fractionated as described here
(in-gel), and by off-gel IEF as described in [8] using an IPG strip of the same length and pH gradient. LC-MS
analysis was performed as described in [21]. An identical quantity of unfractionated digest was analyzed the
same way (Unfractionated). The Venn diagram shows the total number of peptide identifications generated by
each method, the number of identifications unique to each method, and the identifications in common among
the different methods (see Note 18)
378
immobilized buffers, would be less likely to be recovered in the

sample cups following separation. The method employed here
for elution of in-gel separated peptides is more stringent and
may be responsible for the higher peptide yield.
References
1. Zhang X, Fang A, Riley CP, Wang M, Regnier
FE, Buck C (2010) Multi-dimensional liquid
chromatography in proteomics. Anal Chim
Acta 664:101113
2. Motoyama
A,
Yates
JRI
(2008)
Multidimensional LC separations in shotgun
3. Peng J, Elias JE, Thoreen CC, Licklier LJ,
Gygi SP (2003) Evaluation of multidimensional chromatography coupled with tandem
mass spectrometry (LC/LCMS/MS) for
large-scale protein analysis: the yeast proteome.
4. McNulty DE, Annan RS (2008) Hydrophilicinteraction chromatography reduces the complexity of the phosphoproteome and improves
global phosphopeptide isolation and detection.
Mol Cell Proteomics 7:971980
5. Moritz RL, Ji H, Schtz F, Connolly LM,
Kapp EA, Speed TP, Simpson RJ (2004) A
proteome strategy for fractionating proteins
and peptides using continuous free-flow electrophoresis coupled off-line to reversed-phase
high-performance liquid chromatography.
Anal Chem 76:48114824
6. Cargile BJ, Sevinsky JR, Essader AS,
Stephenson JL Jr, Bundy JL (2005)
Immobilized pH gradient isoelectric focusing
as a first-dimension separation in shotgun proteomics. J Biomol Tech 16:181189
7. Malmstrm J, Lee H, Nesvizhskii AI,
Shteynberg D, Mohanty S, Brunner E, Ye M,
Weber G, Eckerskorn C, Aebersold R (2006)
Optimized peptide separation and identification for mass spectrometry based proteomics
via free-flow electrophoresis. J Proteome Res
5:22412249
8. Hrth P, Miller C, Preckel T, Wenz C (2006)
Efficient fractionation and improved protein
identification by peptide OFFGEL electrophoresis. Mol Cell Proteomics 5:19681974
9. Essader AS, Cargile BJ, Bundy JL, Stephenson
JL Jr (2005) A comparison of immobilized ph
gradient isoelectric focusing and strong-cationexchange chromatography as a first dimension
in shotgun proteomics. Proteomics 5:2434
10. Gan CS, Reardon KF, Wright PC (2005)
Comparison of protein and peptide prefractionation methods for the shotgun proteomic
11.
12.
13.
14.
15.
16.
17.
18.
analysis of Synechocystis sp. PCC 6803.

Xie H, Bandhakavi S, Griffin TJ (2005)
Evaluating preparative isoelectric focusing of
complex peptide mixtures for tandem mass
spectrometry-based proteomics: a case study in
profiling chromatin-enriched subcellular fractions in Saccharomyces cerevisiae. Anal Chem
77:31983207
Cargile BJ, Bundy JL, Freeman TW,
Stephenson JL Jr (2004) Gel based isoelectric
focusing of peptides and the utility of isoelectric point in protein identification. J Proteome
Res 3:112119
Krijgsveld J, Gauci S, Dormeyer W, Heck AJR
(2006) In-gel isoelectric focusing of peptides
as a tool for improved protein identification.
Geiser L, Dayon L, Vaezzadeh A, Hochstrasser
D (2011) Shotgun proteomics: a relative quantitative approach using off-gel electrophoresis and
LC-MS/MS. Methods Mol Biol 681:459472
Scherl A, Franois P, Charbonnier Y, Deshusses
JM, Koessler T, Huyghe A, Bento M, StahlZeng J, Fischer A, Masselot A, Vaezzadeh A,
Gall F, Renzoni A, Vaudaux P, Lew D,
Zimmermann-Ivol GC, Binz P-A, Sanchez
J-C, Hochstrasser DF, Schrenzel J (2006)
Exploring
glycopeptide-resistance
in
Staphylococcus aureus: a combined proteomics
and transcriptomics approach for the identification of resistance-related markers. BMC
Genomics 7:296311
Chick JM, Haynes PS, Molloy MP, Bjellqvist
B, Baker MS, Len ACL (2008) Characterization
of the rat liver membrane proteome using peptide immobilized pH gradient isoelectric focusing. J Proteome Res 7:10361045
Geiser L, Vaezzadeh A, Deshusses JMP (2011)
Shotgun proteomics: a qualitative approach
applying isoelectric focusing on immobilized
pH gradient and LC-MS/MS. Methods Mol
Biol 681:449458
Berkelman T, Bandhakavi S, Stone MD, HahnWindgassen A, Paulus A. Use of the
PROTEAN i12 IEF system for in-gel peptide fractionation prior to LC-MS and comparison with off-gel fractionation. Tech note
6140. Bio-Rad Laboratories

19. Vaezzadeh AR, Hernandez C, Vadas O,
Deshusses JJM, Lescuyer P, Lisacek F,
Hochstrasser DF (2008) pICarver: a software
tool and strategy for peptides isoelectric focusing. J Proteome Res 7:43364345
20. Rappsilber J, Ishihama Y, Mann M (2003)
Stop and go extraction tips for matrix-assisted
laser desorption/ionization, nanoelectrospray,
and LC/MS sample pretreatment in
21. Bandhakavi S, Stone MD, Onsongo G, Van
Riper SK, Griffin RJ (2009) A dynamic
379
range compression and three-dimensional

peptide fractionation analysis platform
expands proteome coverage and the diagnostic potential of whole saliva. J Proteome
Res 8:55905600
22. Bjellqvist B, Hughes GJ, Pasquali C, Paquet
N, Ravier F, Sanchez JC, Frutiger S,
Hochstrasser D (1993) The focusing positions of polypeptides in immobilized
pH gradients can be predicted from their
amino acid sequences. Electrophoresis 14:
10231031
Chapter 27
Western Blotting Using In-Gel Protein Labeling
as a Normalization Control: Stain-Free Technology
Abstract
Western blotting is a commonly used laboratory technique for semi-quantifying protein amounts. It is
important when quantifying protein expression to account for differences in the amount of total protein
loaded onto the gel using a loading control. Common loading controls include housekeeping proteins,
such as -actin or GAPDH, quantified by Western blot, or total protein, quantified using a stain such as
Coomassie Brilliant Blue or Ponceau S. A more recently developed method for total protein quantification
utilizes stain-free technology, which has a linear dynamic detection range and allows for protein detection
on both gels and membranes. Here, we describe the theory and use of stain-free gels for total protein
quantification and normalization of Western blots.
Key words Stain-free technology, Western blotting, Loading control, Data normalization,
Immunoblotting
Introduction
Western blotting is a highly valuable laboratory technique used to
quantify protein levels [1]. When carried out properly, Western
blots can be used to accurately semi-quantify amounts of specific
proteins in a complex sample; however, several considerations must
be taken into account to ensure the accuracy and quality of the blot
[2, 3]. One important consideration is the amount of protein in
each lane, which may vary due to errors in protein concentration
determination, loading volume, or differences in transfer efficiency
[4]. To account for these potential differences, the amount of the
target protein should be normalized to the amount of protein
present in each lane using a loading control.
A commonly used loading control is the expression of a housekeeping protein, such as -actin, glyceraldehyde 3-phosphate
dehydrogenase (GAPDH), or -tubulin, quantified by Western
blot. The rationale for using housekeeping proteins is that they are
constitutively expressed and necessary for maintaining cell viability;
381
382
therefore, their expression is not expected to change under different

experimental conditions. However, several publications have suggested that housekeeping proteins are unsatisfactory loading controls under certain conditions [57]. One concern with using
housekeeping proteins is that they are often overloaded. The target
protein and loading control must both be loaded within the linear
detection range for accurate quantification; however, housekeeping proteins are highly abundant and consequently often overloaded, especially when a low abundance target protein is being
detected, so that differences in protein amounts between lanes
cannot be accurately determined [4, 6]. It is difficult to determine
if housekeeping proteins are overloaded for the tissue or cell of
interest without generating standard curves to determine the linear
range of the housekeeping protein. Additionally, the expression of
housekeeping proteins has been shown to change under various
conditions, such as cell growth and differentiation, disease states,
and in response to certain stimuli [810].
An alternative to housekeeping proteins as loading controls is
total protein amount, which has the advantage that normalization
does not depend on expression of a single protein. Total protein
may be detected on the blotting membrane using stains such as
Ponceau S or Coomassie blue, or the more recently developed
stain-free method [11, 12]. In the stain-free method, the gel contains trihalo compounds that react with tryptophan residues. When
activated by UV light, the modified tryptophan residues produce
fluorescence which can be quantified to measure the relative amount
of total protein present [4]. Quantification of total protein using
stain-free technology has been verified as a reliable loading control
and has the advantage that it gives the researcher a convenient way
to verify that protein separation and transfer were successful before
proceeding with the Western blotting process [4, 5, 13, 14]. Here
we use the stain-free method to quantify total protein in rat heart
homogenates and normalize the amount of the proteasome subunit
Rpt1, detected by Western blot, demonstrating that the stain-free
method has a broad linear detection range, accurately reflects the
amount of protein in each lane, and serves as a reliable loading control for Western blotting.
Materials
All solutions should be prepared with ultrapure water (18 M)
and analytical grade reagents. All reagents are prepared and stored
at room temperature unless otherwise indicated. Sodium azide is
not added to any of the solutions and all waste disposal protocols
should be followed when discarding Western blotting waste.
Stain-Free Western Blotting
2.1 SDSPolyacrylamide Gel

Electrophoresis
(SDS-PAGE)
383
1. Laemmli sample buffer (2): 65.8 mM TrisHCl, pH 6.8,

2.1 % SDS, 26.3 % (w/v) glycerol, 0.01 % (w/v) bromophenol
blue with -mercaptoethanol (50 L for each 950 L of 2
Laemmli sample) (catalog # 161-0737, Bio-Rad, Hercules,
CA, USA) (see Note 1).
2. SDS-PAGE gels with stain-free capabilities: 420 % 18-well
TGX Stain-Free Precast Gel (catalog # 567-8094, Bio-Rad).
3. Protein standards: e.g. Spectra Multicolor Broad Range
Protein Ladder (catalog # 26634, Thermo Scientific, Waltham,
MA, USA) (see Note 2).
4. SDS-PAGE running buffer: 192 mM glycine, 0.1 % SDS,
25 mM Tris base. To prepare 1 L of 10 SDS-PAGE running
buffer, weigh out 30.2 g Tris base, 144 g glycine and 10 g
SDS. Mix to dissolve in water, and then fill to 1 L with water
(see Note 3).
2.2
Protein Transfer
1. Precut blotting membrane and filter paper: Trans-Blot Turbo

Nitrocellulose Transfer Pack, Midi format (catalog # 1704159, Bio-Rad) (see Note 4).
2. Blotting apparatus: Trans-Blot Turbo Transfer System (catalog
# 170-4155, Bio-Rad).
3. Tris-buffered saline (TBS): 150 mM NaCl, 10 mM TrisHCl,
pH 7.4.
4. TBST: TBS with 0.05 % (v/v) Tween-20 (see Note 5).
5. Blocking solution: 3 % bovine serum albumin (BSA) in
TBST. Weigh out the appropriate amount of BSA (1.5 g for
3 % BSA or 0.5 g for 1 % BSA) and transfer to a 50 mL conical
centrifuge tube. Fill to 50 mL with TBST and vortex to mix
completely (see Notes 6 and 7).
6. Antibody diluent solution: 1 % BSA in TBST (see Notes 6 and 7).
7. Substrate for enhanced chemiluminescence (ECL): Clarity
Western ECL Substrate (catalog #170-5060, Bio-Rad) which
contains Clarity Western peroxidase reagent and Clarity Western
luminol/enhancer that are mixed at a ratio of 1:1 (see Note 8).
8. Extra-large blot box (see Note 9).
9. Forceps suitable for handling membranes gently.
10. Roller for use with blotting apparatus (catalog # 165-1279,
Bio-Rad).
11. Kimwipes (Kimberly-Clark
(see Note 10).
Corporation,
Texas,
USA)
12. Imaging equipment compatible with stain-free technology:

ChemiDoc MP (catalog # 170-8280, Bio-Rad).
13. Computer with ImageLab software, version 5 (Bio-Rad)
(see Note 11).
384
2.3 Tissue Sample

for SDS-PAGE
and Antibodies
for Immunostaining
1. Rat heart lysate (4 mg/mL) in homogenization buffer:

150 mM NaCl, 5 mM MgCl2, 1 mM EDTA, 50 mM Tris
HCl, pH 7.5. To prepare 500 mL homogenization buffer, mix
15 mL 5 M NaCl, 2.5 mL 1 M MgCl2, 1 mL 0.5 M EDTA,
and 12.5 mL 2 M Tris base. Fill to 450 mL with water and
adjust the pH to 7.5 with HCl, then fill to 500 mL with water
and mix. Homogenization buffer is stored at 4 C and may be
used for up to 6 months.
2. Primary antibody solution: mouse anti-proteasome 19S Rpt1/
S7 subunit (MSS1-104) antibody (# BML-PW8825, Biomol
International, LP, Farmingdale, NY, USA) used at a dilution of
1:1,000 (v/v) in 1 % BSA/TBST.
3. Secondary antibody solution: peroxidase-conjugated rabbit
anti-mouse IgG antibody (# A9044, Sigma-Aldrich, St. Louis,
MO, USA) used at a concentration of 1:5,000 (v/v) in 1 %
BSA/TBST.
Methods
All procedures are carried out at room temperature unless otherwise noted.
3.1 Sample
Preparation
and SDS-PAGE
1. Homogenize approximately 100 mg of rat heart in 1 mL cold

homogenization buffer on ice using a dounce homogenizer
(chop the hearts with a razor blade, then homogenize with 25
strokes). Centrifuge the homogenate at 12,000 g for 30 min,
remove the supernatant, and dilute the supernatant to 4 mg/
mL protein concentration with homogenization buffer.
Homogenate may be stored at 80 C.
2. To prepare tissue samples for SDS-PAGE, combine equal
volumes of 4 mg/mL rat heart homogenate and 2 Laemmli
sample buffer (-mercaptoethanol added). Dilute a portion of
the 2 Laemmli sample buffer to 1 with water and dilute a
portion of heart homogenate to 1 mg/mL by combining one
part 2 mg/mL homogenate with one part 1 Laemmli sample
buffer. Boil homogenate at 95 C for 3 min. Spin down briefly
using a microcentrifuge and allow sample to cool to room
temperature.
3. Remove the comb and tape from the gel, place the gel in the
electrophoresis cell, and fill with SDS-PAGE running buffer to
the fill line. Load 3 L of the protein ladder in lanes 1, 7, and
13 and increasing amounts of heart homogenate (1050 g)
into the remaining lanes (Table 1) (see Notes 12 and 13). Run
at 150 V, 500 mA maximum current for 65 min or until the
dye front reaches the bottom of the gel.
4. Switch off the power supply and disconnect the electrical leads.
Remove the gel from the electrophoresis cell, rinse with a small
385
Table 1
Gel loading pattern
Lane
1 2
7 8
10 11 12 13 14 15 16 17 18
Sample amount (g)
10 20 30 40 50
10 20 30 40 50
10 20 30 40 50
Volume (L)
10 10 15 20 25
10 10 15 20 25
10 10 15 20 25
Concentration (g/L)
Fig. 1 Detection of 1050 g of rat heart homogenate in polyacrylamide gels

using the stain-free method. A Criterion stain-free gel was activated by exposing
the gel to 1 min of UV transillumination with the ChemiDoc MP. The activated gel
was then imaged using the ChemiDoc MP
amount of deionized water to remove buffer, and open the gel

cassette using the aluminum opening lever supplied with the
gels, so that the gel remains on one side of the cassette. Rinse
with deionized water again and remove the gel wells and the
agarose from the bottom of the gel using a razor blade.
5. In the activation process, tryptophan residues undergo a UVinduced reaction with trihalo compounds in the stain-free gel to
produce a fluorescence signal. Activate the gel by UV transillumination for 1 min using the Bio-Rad ChemiDoc MP Imaging
System (see Note 14). Image the gel with the stain-free gel setting and automatically optimized exposure time setting (Fig. 1).
If the gel is overexposed, decrease the exposure time manually.
386
3.2
Protein Transfer
1. Open the nitrocellulose transfer pack and place one-half of the

filter paper with the nitrocellulose membrane on top of it inside
the open Transblot Turbo cassette. Lay the gel on top of the
nitrocellulose, taking care to lay it evenly. Place the second
layer of filter paper on top of the gel and roll air bubbles out
with the roller. Place the lid on top and turn to closed. Put the
cassette in the TurboBlot and run with the Turbo setting for
one midi gel (the transfer will take 7 min).
2. Remove the membrane from the cassette and image using the
stain-free membrane setting with automatically optimized
exposure time (Fig. 2). Take care to image before the membrane dries, as a dry membrane reduces the quality of the
image. You may decide to cut the membrane into smaller sizes
at this point (see Notes 15 and 16).
3. After imaging, dry the membrane. This improves the attachment of proteins to the membrane.
3.3
Immunostaining
1. Place the membrane in a blotting container and rinse with

water or TBST, taking care not to pour directly onto the membrane, as this may disrupt the protein bound to the membrane.
Submerge the membrane in blocking solution so that the
entire surface of the membrane is covered and incubate with
shaking for 1 h. Pour off the blocking solution and rinse briefly
with TBST.
2. Incubate with shaking in mouse anti-proteasome 19S Rpt1/
S7 subunit primary antibody for 2 h (see Notes 17 and 18).
Pour off the primary antibody and rinse once briefly with
TBST, then three times for 5 min in TBST with shaking.
Fig. 2 Detection of total protein in rat heart homogenates on nitrocellulose membrane using the stain-free
method. (a) Detection of 1050 g total heart protein on nitrocellulose membrane using stain-free imaging. (b)
Quantification of total protein detected on the membrane versus amount of protein loaded onto the gel (n = 3
for each data point). Data are represented as mean standard error
387
3. Incubate with shaking in peroxidase-conjugated rabbit antimouse IgG secondary antibody for 1 h (see Note 19). Pour off
the secondary antibody and rinse briefly with TBST, then rinse
three times for 5 min in TBST with shaking.
4. To image, lift the membrane with forceps, taking care to touch
only the side of the membrane. Allow the TBST to drain off
the membrane so that it does not dilute the ECL substrate;
however, take care not to dry the membrane. The edge of the
membrane may be dabbed with a Kimwipe to remove excess
liquid. Place the membrane on the imaging surface, laying
from one edge to the other to exclude bubbles.
5. Combine equal parts of Clarity western peroxide reagent and
Clarity western luminol/enhancer reagent. Mix the solution
and pipette it onto the membrane (protein side up) so that
every part of the membrane is evenly covered with ECL substrate (see Note 20). Either proceed directly to the imaging
step (incubation in between addition of the ECL substrate and
imaging is not necessary) or incubate the ECL covered membrane for 2 min. If the ECL reagent is left on the membrane
for 2 min, then remove the excess ECL reagent and cover the
membrane with a thin sheet protector to prevent drying.
6. Image the blot with the Chemi Hi Sensitivity setting (Image
Lab software) and then adjust the exposure time and sensitivity/resolution settings as needed. If the target protein (Fig. 3)
Fig. 3 Quantification and normalization of Rpt1 levels in rat heart homogenates.

(a) Western blot of heart homogenate (1050 g) probing for the proteasome
subunit Rpt1. (b) Quantification of Rpt1 levels, not normalized and normalized to
total protein. Total protein was determined using the stain-free detection method
on the nitrocellulose membrane (n = 3 for each data point). Data are represented
as mean standard error
388
is detected in a short timeframe when using the Hi Sensitivity

setting, image with Chemi Hi Resolution and adjust the exposure time as needed. To compare the band of interest to the
molecular mass of the protein standard (protein ladder), take
a multichannel image using the Colorimetric setting (automatically optimized exposure time) and Chemi Hi Sensitivity
or Hi Resolution setting. This protocol will automatically
overlay the chemiluminescent image with the colorimetric
image, allowing you to determine the molecular weight of the
band of interest.
Notes
1. Unused sample buffer may be stored at 20 C for up to 3
months or at 80 C for more than 6 months.
2. Any suitable pre-stained protein standard may be used; however
it should be noted that some protein standards give very strong
signals when imaged using the stain-free method, likely due to a
high tryptophan content of certain standards or the dye used to
stain them. This may interfere with total protein quantification.
3. Do not adjust the pH with acid or base. Once diluted to a 1
buffer the pH of the buffer should be 8.38.6.
4. Polyvinylidene difluoride (PVDF) membranes may also be
used with stain-free gels. We noticed that when the total protein on the membrane is imaged there is a slightly higher background with PVDF membranes than nitrocellulose membranes.
However, if fluorescently labeled secondary antibodies are
used, low fluorescence PVDF membranes are recommended.
5. 10 TBS can be stored at room temperature for at least 6 months
(10 TBS will crystallize if stored at 4 C).
6. Unused BSA should be stored at 4 C and may be used for up
to approximately 1 week after preparation. If the primary
antibody is to be used again, it is recommended that freshly
prepared BSA is used, as the BSA will limit the amount of
time in which the antibody can be used.
7. We have found that blocking and diluting certain antibodies
with solutions of BSA in TBST gives lower background than
when milk in TBST is used, though certain antibodies work
better with milk. If the antibody being used gives a high level
of background, the protocol may be optimized by testing the
antibody with solutions of BSA and milk to see which gives
lower background and stronger signal.
8. We recommend storing a small volume of ECL substrate at
room temperature for regular use, and the remainder at 4 C
for long-term storage. Using cold ECL reagent decreases the
sensitivity of the blot.
389
9. Blotting boxes of different sizes are available from Denville

Scientific, Inc., South Plainfield, NJ, USA. The most commonly used blotting box size used for Criterion gels is 15.5 cm
(l) 10.5 cm (w) 3.5 cm (h).
10. Kimwipes are scientific cleaning wipes that are used when
leaving fibers or lint on a surface is unwanted.
11. ImageLab imaging software is free with the ChemiDoc MP
and some other Bio-Rad imagers. A convenient feature of the
ImageLab software is that users are allowed an unlimited number of installations on lab computers and free software upgrades
from Bio-Rad.
12. Loading significantly different volumes may cause samples to
run unevenly. To load a larger volume of protein ladder, the
protein ladder may be diluted with 1 Laemmli sample buffer.
We recommend mixing at a 1:1 ratio and loading 6 L onto
the gel. Some of the heart homogenate was diluted to 1 mg/
mL so that 10 L could be loaded for 10 g of total protein.
13. For scientists that do not routinely run SDS-PAGE gels, we
recommend using gel loading tips to prevent the sample from
spilling into neighboring wells. Place the tip at the bottom of
the well and slowly dispense the sample, raising the pipette tip
toward the top of the well as the sample is dispensed. Be careful
not to pipette air bubbles into the well, as it will disturb the
sample.
14. In our experience, longer activation times such as 2 or 5 min
give higher intensity but do not increase the linearity of the
total protein amount detected. This suggests that shorter activation times only modify a portion of tryptophan residues. If
the antibody being utilized detects tryptophan residues, it may
therefore be advisable to use a shorter activation time so that
the antibody is still able to bind to the unmodified epitopes.
The gel only needs to be activated once; if the gel is imaged a
second time, change activation time to None.
15. To cut the membrane at a specific molecular weight to use for
multiple blots, you may cut with a razor or scissors based on the
molecular weight markers. You may also wish to use a Ponceau
S stain as a guide to cut evenly across the width of the membrane. For this purpose, a short 510 s rinse with Ponceau S is
usually sufficient. After cutting the membrane, remove the
Ponceau S stain by washing multiple times with TBST.
16. After imaging, blots can be dried and stored at 4 C in a zip
lock bag for at least 1 week. If the membrane needs to be reimaged after the membrane is dried, wet the membrane with
TBST and image.
17. In general, incubation of antibodies in 1 % BSA/TBST or 1 %
milk/TBST gives stronger signal than antibodies in 35 % BSA
or milk.
390
18. Many primary antibodies may be used more than once. If you
wish to reuse the antibody, you may store it at 4 C for up to
approximately 1 week. The antibody should not be used if the
BSA or milk has gone bad.
19. The quality of the secondary antibody is important for obtaining a high-quality Western blot. If consistently poor results
are obtained, we recommend trying a different secondary
antibody.
20. The volume of ECL substrate required to evenly cover the
membrane will depend on the area of the membrane. For an
uncut midi membrane, 56 mL ECL substrate is sufficient. If
a clear plastic support is used to cover the membrane during
imaging, a smaller volume of ECL substrate can be used, as the
plastic support will spread the ECL substrate over the membrane. If a long exposure time (90 s or longer) is required, it is
recommended to use a plastic support to prevent the membrane from drying during imaging. The ECL reagent should
not be left on the membrane for longer than 5 min as this will
result in a shorter time frame for which the maximum signal is
obtained.
Acknowledgement
This work was supported by the National Institutes of Health
Grants HL080101 and HL096819.
References
1. Kurien BT, Scofield RH (2006) Western blotting. Methods 38:283293
2. Towbin H, Staehelin T, Gordon J (1979)
Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc Natl Acad Sci
U S A 76:43504354
3. Murphy RM, Lamb GD (2013) Important
considerations for protein analyses using antibody based techniques: down-sizing Western
blotting up-sizes outcomes. J Physiol 591:
58235831
4. Posch A, Kohn J, Oh K, Hammond M, Liu N
(2013) V3 stain-free workflow for a practical,
convenient, and reliable total protein loading
control in western blotting. J Vis Exp 82:
50948
5. Gilda JE, Gomes AV (2013) Stain-Free total

protein staining is a superior loading control to
beta-actin for Western blots. Anal Biochem
440:186188
6. Dittmer A, Dittmer J (2006) Beta-actin is not a
reliable loading control in Western blot analysis.
7. Eaton SL, Roche SL, Llavero Hurtado M,
Oldknow KJ, Farquharson C, Gillingwater
TH, Wishart TM (2013) Total protein analysis
as a reliable loading control for quantitative
fluorescent Western blotting. PLoS One
8:e72457
8. Ruan W, Lai M (2007) Actin, a reliable marker
of internal control? Clin Chim Acta 385:15
9. Schmittgen TD, Zakrajsek BA (2000) Effect
of experimental treatment on housekeeping

gene expression: validation by real-time, quantitative RT-PCR. J Biochem Biophys Methods
46:6981
10. Rodriguez-Mulero S, Montanya E (2005)
Selection of a suitable internal control gene for
expression studies in pancreatic islet grafts.
Transplantation 80:650652
11. Zeng L, Guo J, Xu HB, Huang R, Shao W,
Yang L, Wang M, Chen J, Xie P (2013) Direct
Blue 71 staining as a destaining-free alternative
loading control method for Western blotting.
12. Aldridge GM, Podrebarac DM, Greenough
WT, Weiler IJ (2008) The use of total protein
391
stains as loading controls: an alternative to

high-abundance single-protein controls in
semi-quantitative immunoblotting. J Neurosci
Methods 172:250254
13. Gurtler A, Kunz N, Gomolka M, Hornhardt S,
Friedl AA, McDonald K, Kohn JE, Posch A
(2013) Stain-Free technology as a normalization tool in Western blot analysis. Anal Biochem
433:105111
14. Colella AD, Chegenii N, Tea MN, Gibbins IL,
Williams KA, Chataway TK (2012) Comparison
of Stain-Free gels with traditional immunoblot
loading control methodology. Anal Biochem
430:108110
Chapter 28
2-D Western Blotting for Evaluation of Antibodies
Developed for Detection of Host Cell Protein
Tom Berkelman, Adriana Harbers, and Sricharan Bandhakavi
Abstract
Recombinant proteins generated for therapeutic use must be substantially free of residual host cell protein
(HCP). The presence of host cell protein (HCP) is usually assayed by ELISA using a polyclonal antibody
mixture raised against a population of proteins derived from the host cell background. This antibody
should recognize as high a proportion as possible of the potential HCPs in a given sample. A recommended method for evaluating the assay involves two-dimensional electrophoretic separation followed by
Western blotting.
We present here a method using commercial anti-HCP antibody and samples derived from Chinese
Hamster Ovary (CHO) cells. The 2-D electrophoresis procedure gives highly reproducible spot patterns
and entire procedure can be completed in less than 2 days. Software analysis enables the straightforward
generation of percent coverage values for the antibody when used to probe HCP-containing samples.
Key words HCP, Host cell protein, 2-D Electrophoresis, Western blotting, Anti-HCP
Introduction
Most therapeutic proteins are produced by cultured cells which
have been engineered through recombinant DNA technology to
produce the relevant protein. The protein is either produced in the
host cell or secreted into the medium. In the case of intracellular
protein production, the host cells are harvested and lysed, and the
protein of interest is purified from the lysate. In the case of secreted
protein, the protein of interest is purified from the culture medium
in which the cells have been grown. In either case, the starting
material for purification of the protein of interest contains other
proteins from the host cell line, and these so-called host cell proteins (HCP) are potential impurities in the final drug substance.
HCP impurities can induce an immune response in the patient, so
their presence needs to be minimized. Validation and quality
control testing of any therapeutic protein needs to include a test to
show that HCP contamination is at a minimal level.
393
394
HCP contaminants can have an undesirable effect even at very

low concentration and in general, the only sufficiently sensitive
assay methodology for HCP is immunochemical. The presence of
HCP is usually assayed by ELISA using a polyclonal antibody mixture that has been developed to have reactivity against a wide range
of potential HCPs. Development of the polyclonal antibody
requires optimization of the number of potential targets recognized, and the specificity of the assay needs to be fully validated
prior to adoption [13].
A recommended method for this validation involves twodimensional (2-D) electrophoretic separation followed by Western
blotting [47]. This technique displays the proteins in a representative HCP sample as an array of hundreds to thousands of individual
spots. Western blotting determines which of those proteins are recognized by the polyclonal antibody. This allows the calculation of a
match rate or coverage score corresponding to the ratio of the number of proteins recognized by the antibody to the total number of
proteins detectable in the sample (see Fig. 1). Although this method
allows a straightforward determination of the range of reactivity of
the antibody mixture, adoption of this methodology has been hindered by the perceived lack of sensitivity, lack of reproducibility and
overall laborious nature of the 2-D electrophoresis and blotting
workflow. Tools that simplify, standardize, and accelerate 2-D electrophoresis and blotting have become available. Advanced electrophoresis and blotting instrumentation, stable and reproducible
pre-cast gels, next-generation imaging and image analysis software
all allow this assay to be performed in a reproducible manner in a
relatively short time. We have undertaken to demonstrate this by
generating coverage scores for a commercially available anti-HCP
antibody against samples derived from CHO cells, a common cell
line used in the production of protein therapeutics. We present here
a procedure that can be completed in under 24 h. Two replicate 2-D
gels are run. One is stained directly with a sensitive fluorescent
stain, the other is transblotted onto PVDF membrane, stained posttransfer and compared to the other gel to verify transfer efficiency.
The blotted membrane is stained for total protein and subsequently
immunodetected with the anti-HCP antibody preparation to be
evaluated. This allows comparison of total protein and immunodetected protein patters from the same blot, simplifying automated
analysis, as the two images are in registry with each other. This demonstration experiment was conducted with a commercially available
anti-CHO HCP antibody.
Materials
Reagents and prepared buffer stocks are from Bio-Rad Laboratories
or Sigma-Aldrich except when indicated. Reagent quality should
be reagent grade or better. Water should be of the highest purity
possible.
2-D Western Blotting for Anti-HCP Evaluation
395
Fig. 1 Workflow for evaluation of antibodies developed for detection of host cell
proteins
396
2.1 Sample
Preparation
1. Protein-free growth medium for CHO cells.

2. Phosphate Buffered Saline (PBS): 150 mM NaCl, 10 mM
sodium phosphate, pH 7.8. Prepare by tenfold dilution of a
stock solution.
3. Centrifugal concentrator with 10,000 MWCO PES membrane
(Vivacell 70, Sartorius).
4. ReadyPrep 2-D Cleanup kit (Bio-Rad).
5. 8 M urea: Weigh 4.805 g of urea into a 15 ml tube. Bring
volume to 10 ml with water and fully dissolve (see Note 1).
This solution may be stored at 20 C or lower for up to a week
and used to prepare Sample solution 1 (see below under 2-D
Electrophoresis).
6. 7 M urea, 2 M thiourea: Weigh 4.204 g of urea and 1.522 g
thiourea into a 15 ml tube. Bring volume to 10 ml with water
and fully dissolve (see Note 1). This solution may be stored at
20 C or lower for up to a week and used to prepare Sample
solution 2 (see below under 2-D Electrophoresis).
7. Quick Start Bradford Protein Assay (Bio-Rad).
2.2 2-D
Electrophoresis,
First-Dimension IEF
1. 1 % bromophenol blue: Prepare 10 ml of 50 mM Tris base by

dissolving 60.6 ml of Tris base in 10 ml of water. Add 100 ng
of bromophenol blue and vortex until dissolved. Store at room
2. Sample solution 1: 8 M urea, 4 % (w/v) 3-[(3-Cholamidopropyl)
dimethylammonio]-1-propanesulfonate hydrate (CHAPS),
40 mM dithiothreitol (DTT), 0.2 % Bio-Lyte, pH 310,
0.001 % (w/v) bromophenol blue. To 5 ml of 8 M urea add
200 mg CHAPS, 30.9 mg DTT, 25 l Bio-Lyte 3/10 carrier ampholyte (supplied at 40 %), 5 l 1 % bromophenol
blue.
3. Sample solution 2: 7 M urea, 2 M thiourea, 4 % (w/v) CHAPS,
40 mM DTT, 0.2 % Bio-Lyte, pH 310, 0.001 % (w/v) bromophenol blue. To 5 ml of 7 M urea, 2 M thiourea add 200 mg
CHAPS, 30.9 mg DTT, 25 l Bio-Lyte 3/10 carrier ampholyte (supplied at 40 %), 5 l 1 % bromophenol blue.
4. Instrument for IPG-strip focusing: PROTEAN i12 IEF
System, including 11 cm Focusing Tray and 11 cm
Rehydration/Equilibration Tray.
5. 11 cm pH 310 NL ReadyStrip IPG strips (Bio-Rad, see
Note 3).
6. Mineral oil.
7. Electrode wicks for gel-side-up configuration.
2.3 2-D
Electrophoresis,
Equilibration
and SecondDimension SDS-PAGE
397
1. 1.5 M TrisHCl pH 8.8.

2. 10 Tris/Glycine/SDS electrophoresis buffer: 1.92 M glycine, 150 mM Tris base, 1 % (w/v) sodium dodecyl sulfate
(SDS). This buffer may be purchased as a prepared stock solution (e.g. Bio-Rad).
3. Equilibration buffer: 50 mM TrisHCl, pH 8.8, 6 M urea,
30 % (w/v) glycerol, 2 % (w/v) SDS, 0.001 % bromophenol
blue. Combine 6.7 ml 1.5 M TrisHCl, pH 8.8, 72.07 g urea,
60 g glycerol, 20 ml 20 % (w/v) SDS. Bring the volume to
almost 200 ml with water. Stir to dissolve (see Note 4). Correct
the volume to 200 ml. Add 200 l 1 % bromophenol blue (see
Subheading 2.2 and Note 2). This solution may be aliquoted
into plastic tubes and stored at 80 C.
4. Equilibration buffer with 1 % DTT: Dissolve 200 mg DTT in
20 ml Equilibration buffer (see Note 5).
5. Equilibration buffer with 2.5 % iodoacetamide: Dissolve 500 mg
iodoacetamide in 20 ml Equilibration buffer (see Note 5).
6. Any kD Criterion TGX precast gels with IPG well
(Bio-Rad).
7. Agarose sealing solution: Combine 5 ml 10 Tris/Glycine/
SDS electrophoresis buffer with 45 ml water. Add 250 mg agarose. Swirl to disperse. Heat in a microwave oven until the
agarose is completely dissolved. Add 50 l 1 % bromophenol
blue. This solution may be aliquoted into plastic tubes and
stored at 4 C (see Note 6).
8. Criterion Cell vertical midi-format electrophoresis cell.
2.4 Staining
and Blotting
1. Oriole Fluorescent Gel Stain (Bio-Rad).

2. Trans-Blot Turbo PVDF Midi Transfer Packs (Bio-Rad).
3. Trans-Blot Turbo transfer system (Bio-Rad).
4. SYPRO Ruby Protein Blot Stain (Bio-Rad or Life Technologies).
2.5 Immunodetection and Analysis
1. 10 Tris-buffered Saline (TBS): This buffer may be purchased

as a prepared stock solution (Bio-Rad, see Note 7).
2. 10 % Tween 20: This may be purchased as a prepared stock
solution (Bio-Rad).
3. Non-fat dry milk (Blotting Grade Blocker, Bio-Rad, see Note 8).
4. TTBS: 500 mM NaCl, 20 mM TrisHCl pH 7.5, 0.05 %
Tween 20. For each liter of buffer required, combine 895 ml
water, 100 ml 10 TBS, 5 ml 10 % Tween 20. This solution
should be used the same day it is made.
5. Blocking Solution: 5 % non-fat dry milk in TTBS. Dissolve 5 g
of non-fat dry milk in 100 ml TTBS.
398
6. Anti-CHO HCP antibody (Cygnus Technologies 3G-0016-PA,

see Note 9).
7. HRP-conjugated Bovine Anti-Goat antibody (Santa Cruz
Biotechnology, see Note 9).
8. Clean plastic wrap or sheet protector.
9. Clarity Western ECL Substrate (Bio-Rad).
10. ChemiDoc MP imager (Bio-Rad, see Note 10).
2.6 Software
Analysis
and Generation
of a Match Rate
for Assessment
of Antibody Coverage
1. Image Lab Software.

2. PDQuest 2-D Analysis Software.
Methods
3.1 Sample
Preparation
1. Grow CHO-K1 cells as adherent cultures in 225 ml flasks with

serum-free medium. When the culture reaches confluence, pour
off the culture supernatant (approximately 75 ml) and reserve.
Wash the flask with PBS. Apply 20 ml PBS and harvest the cells
by scraping. Transfer the cell suspension to a tared tube.
Centrifuge the cell suspension (180 g, 5 min). Discard the
supernatant and determine the weight of the cell pellet. Store
the cell pellet at 80 C (see Notes 11 and 12).
2. Apply the reserved cell culture supernatant to the centrifugal
concentrator. Centrifuge at 4 C, 1,000 g until the volume is
reduced to 5 ml or less. This may take up to several hours.
3. Process the concentrated material with the 2-D Cleanup Kit
according to the accompanying manual. Follow the instructions
for Processing Dilute Samples or Samples Greater than 100 l
(see Note 13).
4. Resuspend the final dried pellet in 250 l sample solution 1
(see Note 13). This is the secreted protein sample.
5. Suspend the cell pellet from step 1 in sample solution 2. Use
1 ml of sample solution 2 for each 50 mg of cell pellet. Seal the
tube and sonicate briefly in a bath-type sonicator. Centrifuge
for 5 min at 14,000 g.
6. Transfer the supernatant to another tube. This is the cell lysate
protein sample.
7. Estimate the protein concentration of both protein samples
using the Quick Start Bradford Protein Assay according to the
supplied instructions (see Note 14). The protein concentrations
should be at least 5 mg/ml. The material may be stored at
80 C until ready to perform the 2-D analysis.
3.2 2-D
Electrophoresis,
First-Dimension IEF
399
1. For each separation to be run, dilute a portion of each sample

containing 150 g of protein into 200 l sample solution. For
the secreted protein sample, use sample solution 1. For the
cell lysate protein sample, use sample solution 2 (see Notes 15
and 16). At least two separations should be run for each sample
2. For each separation run, pipet 200 l of solution along the
center of the channel of the PROTEAN i12 rehydration/
equilibration tray. Take care to not introduce air bubbles when
expelling the solution.
3. Using forceps, remove the cover sheet from each IPG strip,
then gently place the IPG strip gel-side down onto the solution
in the channel. Move the IPG strip back and forth slightly to
ensure that the solution is distributed along the length of the
IPG strip. Take care to not trap air bubbles beneath the IPG
strip (see Notes 19 and 20).
4. Overlay each IPG strip with 5 ml mineral oil. Apply the mineral
oil to both ends of the channel and allow it to flow toward the
middle of the channel (see Note 21).
5. Cover the tray and leave it on a level bench overnight (1218 h)
for complete rehydration.
6. Using forceps, remove the IPG strips from the rehydration
tray, remove excess mineral oil, and place the rehydrated IPG
strips gel-side up in the channels of the focusing tray. Position
the positive (+) ends of the IPG strips against the positioning
stops in each channel.
7. Wet the gel-side up wicks (notched) with distilled or deionized
water and blot off excess water. Use two wicks per IPG strip:
place a wick at each end of each IPG strip. Position the electrode assemblies in the focusing tray and press down on the
green tabs to snap the electrode assemblies into place.
8. Place the focusing tray with the rehydrated IPG strips on the
Peltier platform and connect the electrodes to the instrument.
Overlay each IPG strip with 5 ml mineral oil.
9. Run IEF according to the preset gradual protocol for
pH 310 NL IPG strips: 250 V for 20 min, gradual ramp to
8,000 V over 1 h, 8,000 V for 26,000 Vh. Maintain a current
limit of 50 A throughout. If not proceeding directly to the
next step, store the IPG strip at 80 C (see Note 22).
3.3 2-D
Electrophoresis,
Equilibration,
and SecondDimension SDS-PAGE
1. Prepare SDS-PAGE Running Buffer by tenfold dilution of 10

Tris/Glycine/SDS electrophoresis buffer. Prepare 1 l for each
Criterion Cell to be run (each cell runs two gels).
2. Melt sufficient Agarose Sealing Solution for the second dimensions to be run. Each gel requires 1 ml of solution. If the solution has been aliquoted into sealed tubes, it may be melted in
400
a 95 C heat block. Otherwise, it may be melted in a boiling

water bath (see Note 23).
3. Place one IPG strip gel-side up in each channel of a rehydration/
equilibration tray, and fill the channels with 4 ml equilibration
buffer with 1 % DTT.
4. Incubate with gentle agitation for 15 min, then decant.
5. Fill the channels with 4 ml equilibration buffer with 2.5 %
iodoacetamide and incubate again for 15 min.
6. After equilibration, remove the IPG strips and briefly rinse
with SDS-PAGE running buffer. This step rids the strip of
excess iodoacetamide and serves to lubricate the IPG strip for
placement on the second dimension.
7. Position each second-dimension gel cassette in a test tube rack
or suitable stand so that it is leaning slightly backwards from
vertical.
8. Place the equilibrated IPG strip (anodic side to the left) onto
the long plate with the plastic backing against the plate.
9. Slide the IPG strip between the plates using a spatula to push
against the plastic backing. Ensure that the plastic backing
remains fully in contact with the long plate and be careful not to
damage the gel with the spatula. Make sure that the IPG strip is
positioned directly on top of the second-dimension gel without
any bubbles in the interface between the two gel surfaces.
10. To secure the IPG strip in place, overlay it with molten agarose
solution. Avoid trapping air bubbles between the IPG strip and
second-dimension gel. Dislodge any bubbles by tapping the
plastic backing on top of the strip.
11. Stand the gel upright and allow the agarose to set for 510 min
before loading the gel into the electrophoresis cell.
12. Insert the gel cassette into the electrophoresis cell and fill the
buffer chambers with SDS-PAGE running buffer (see Note 24).
13. Connect the electrophoresis cell to a power supply. Run the
gels at 50 V for 10 min, then raise the voltage to 250 V (see
Note 25). Run until the bromophenol blue tracking dye
reaches the bottom of the gel (approximately 30 min).
3.4 Staining
and Blotting
1. There should be at least two gels run for each sample. One is
to be stained and the other is to be blotted.
2. Carefully open the cassette for the gel to be stained and use a
spatula to separate the agarose overlay, including the IPG strip,
from the polyacrylamide gel.
3. Carefully peel the gel from the cassette and place it in a tray
containing 100 ml Oriole Fluorescent Gel Stain. Place the lid
on the staining tray incubate with gentle staining or rocking
for 90 min (see Note 26).
401
4. For each gel to be blotted open a transfer pack and place the
anode stack in the middle of the base of the Trans-Blot Turbo
cassette. The blotting membrane should be at the top of the
stack.
5. Carefully open the cassette for the gel to be blotted and use a
spatula to separate the agarose overlay, including the IPG strip,
from the polyacrylamide gel. Peel the gel from the cassette and
carefully place it on the anode stack, covering the blotting
membrane. Use the roller to remove any air bubbles that may
be trapped between the gel and the membrane, or within the
transfer stack (see Note 27). Place the cathode stack on top of
the gel and roll once again to remove trapped air.
6. Place the lid on the cassette and lock it into place by turning
the green knob clockwise. Ensure that the locking pins fully
engage their locking slots.
7. Turn on the Trans-Blot Turbo instrument and slide the cassette into a cassette bay.
8. Select TURBO, then select 2 MINI OR 1 MIDI GEL, then
select RUN for the occupied cassette bay(s). The transfer will
take 7 min. At the end of the run, RUN COMPLETE will
appear on the screen.
9. Disassemble the cassette. Remove and discard the cathode
stack. Carefully lift the gel and place it in a tray containing
100 ml Oriole Fluorescent Gel Stain. Place the lid on the staining tray incubate with gentle staining or rocking for 90 min
(see Note 28).
10. Using forceps, remove the membrane and rinse it briefly with
distilled or de-ionized water (see Note 29).
11. Float each membrane, protein side down, in a plastic tray containing 7 % acetic acid, 10 % methanol.
12. Wash each membrane in distilled or deionized water four times
for 5 min each.
13. Stain each membrane for 15 min with 20 ml SYPRO Ruby
Protein Blot Stain (see Note 30).
14. Wash each membrane in distilled or deionized water two to
three times for 1 min.
15. Capture images of the stained blots with the ChemiDoc MP
imager. Use the imager settings for SYPRO Ruby and use the
longest possible exposure time that does not result in image
saturation.
16. When each gel (both the unblotted gel and the replicate blotted gel) has been in Oriole stain for 90 min, wash briefly in
distilled or deionized water, and capture a fluorescent image
with the ChemiDoc MP imager. Use the imager settings for
Oriole stain. For the unblotted gel, use the longest possible
402
exposure time that does not result in image saturation. Use the
same exposure time for the replicate blotted gel. Ensure that
the zoom setting for all images is the same as used for the
stained membrane.
3.5 Immunodetection
1. Place each blotted membrane, protein side up, in a plastic tray

slightly larger than the membrane. Apply 40 ml of Blocking
Solution. Incubate at room temperature for 1 h with gentle
shaking.
2. Dilute the Anti-CHO HCP antibody (primary antibody) 150fold into Blocking Solution. Prepare 25 ml for each membrane
to be processed.
3. Decant the Blocking Solution from the blotted membranes.
Replace with 25 ml of diluted primary antibody prepared in
step 2. Incubate overnight (at least 12 h) at 4 C with gentle
shaking (see Note 31).
4. Decant the antibody solution. Replace with 40 ml
TTBS. Incubate at room temperature for 5 min with gentle
shaking. Repeat this step four times for a total of five washes.
5. Dilute the HRP-conjugated Bovine Anti-Goat antibody
(secondary antibody) 3,000-fold into TTBS. Prepare 25 ml for
each membrane to be processed.
6. Apply 25 ml of diluted secondary antibody to each membrane.
Incubate at room temperature for 1 h with gentle shaking.
7. Wash each membrane with 40 ml TTBS as described in step 4.
Wash a total of six times, 5 min each.
8. Mix the Clarity kit components in a 1:1 ratio. Each blot
requires 12 ml of the mixed reagent (6 ml of each
component).
9. Place the membrane protein-side up on a clean piece of plastic
wrap or plastic sheet protector.
10. Pour 12 ml of mixed Clarity substrate solution over each membrane. Incubate for 5 min (see Note 32).
11. Lift the membrane from the substrate solution with forceps.
Allow excess substrate solution to drain. Place the membrane
in a plastic sheet protector or cover with plastic wrap.
12. Capture chemiluminescent images of the immunodetected blots
with the ChemiDoc MP imager. Image at the same zoom setting and resolution used in Subheading 3.4, step 16. This will
require creating a custom imager setting. Under Application
select Custom and insert the settings No filter, No illumination, and Binning 1 1 (see Note 33). Adjust the exposure
to give good image quality without saturation.
3.6 Software
Analysis
and Generation
of a Match Rate
for Assessment
of Antibody Coverage
403
1. For each sample, use Image Lab software (used to capture the
images on the ChemiDoc MP) to open the images for the
SYPRO Ruby-stained blot and the immunodetected blot.
Crop both images to the edge of the membrane. If necessary,
first rotate each image so that the edges of the membrane are
parallel with the frame of the image. This will ensure that the
stained blot image and the Western blot image are in registry
for analysis with PDQuest software. Export the images for
analysis. (File > Export > Export for Analysis). This creates
16 bit tiff images.
2. Launch PDQuest. Open both of the files exported from Image
Lab.
3. Crop the images to include only the area containing well
resolved protein spots. Use the Advanced Crop function to
ensure that the cropped images are in registry with each other.
Place the crosshair on a recognizable feature common to both
images (Edit > Image > Advanced Crop > Define Crop Area,
see Fig. 2).
4. Create a new experiment (File > New Experiment). Select
both of the cropped images. Move through the next set of
screens by clicking Next. This will select the default options.
5. Follow the instructions on the Spot Detection Parameter
Wizard. Increase the sensitivity to maximize the number of
spots identified (see Fig. 3).
6. Select Proceed. In the next dialog box, select the cropped
image of the stained gel as the Master gel. Select Add
unmatched spots from all gel images to the Master gel. This
will create an artificial image containing all of the detected spots
from both the stained blot and the immunodetected Western
blot (see Fig. 4).
7. Choose default values by clicking Next through the successive
dialog boxes. The software will calculate matches and display
the Master image, the blot images and an Experiment Summary
with match rates (see Fig. 5). Match Rate 2 for the Western
blot image is the ratio of the number of protein spots recognized by the antibody to the total number of proteins detectable in the sample (see Fig. 6).
8. Choose View > Spot > Show Crosshairs or press F5 to show
the detected spots (see Fig. 7).
9. Show matched spots by selecting elect Analyze > Analysis Set
Manager, then Create and Matching (see Fig. 8). The
resulting images will have identified spots marked with a colored + and matched spots marked with an x in a different
color. This will aid in the determination of whether or not a
spot identification or match is valid (see Fig. 9).
404
Fig. 2 Advanced Crop of the 2-D images
405
Fig. 3 Spot detection
10. Determine a more accurate match rate by deleting incorrectly

identified spots. These can often be visually identified near the
edges of the image. In this case shown in Fig. 10, a series of
spots along the top edge of the blot image probably do not
correspond to real proteins. Delete spurious spots by selecting
Edit > Spot > Remove Spot. The spots are erased by clicking
on the spot or drawing a box around a set of spots. This
operation should be performed both on the Raw 2-D image
and on the Master.
11. Select Edit > Experiment Summary to display the final corrected match rate (see Fig. 11).
12. If desired, the match may be further refined by manually adding and deleting spots. Spot matches may be systematically
verified using the Spot Review Tool (Analyze > Spot Review
see Fig. 12).
406
Fig. 4 Choose Master Gel
Fig. 5 Spot matching
Fig. 6 Match rate
Fig. 7 Detected spots
Fig. 8 Show matched spots
408
Fig. 9 Matched spots
Fig. 10 Erase misidentified spots
409
Fig. 11 Corrected match rate
Fig. 12 Spot review
Notes
1. Solutions containing urea are prepared immediately before use.
2. It is convenient to use a concentrated stock solution of
Bromophenol Blue, but this dye will not dissolve in unbuffered water. It is therefore prepared in Tris base solution.
3. Other pH ranges may be used. pH 310 NL (non-linear) was
chosen for this experiment because it covers most of the range
of protein pIs. This non-linear gradient is flatter in the region
of the gradient where most proteins are found, thereby spreading them out and allowing resolution of more individual
proteins.
4. Equilibration buffer is highly viscous and it may be difficult to
stir efficiently. Allow several hours for the urea to dissolve into
the liquid components.
5. The amount of Equilibration solution with DTT or iodoacetamide may be scaled up or down depending on the number of
IPG strips run. These solutions may be prepared from aliquoted
Equilibration buffer (see Subheading 2.3 step 3). Frozen tubes
of Equilibration buffer should be allowed to thaw slowly in a
beaker of water (do not heat).
410
6. This solution contains SDS and is highly prone to boiling over.

It should be microwaved at low power with constant
monitoring. When the solution starts to boil over, turn off the
power, remove from the microwave oven and swirl before
returning to the microwave oven. Repeat until the agarose is
fully dissolved. Aliquots of agarose sealing solution will solidify
and can be re-melted before use in a 95 C heat block or by
immersion in boiling water.
7. This stock solution gives 1 TBS with a composition of 20 mM
Tris-Cl pH 7.5, 500 mM NaCl.
8. Non-fat dry milk generally gives high-stringency, low background immunodetection. Alternative blocking reagents may
be substituted.
9. This primary detection antibody is from goat, necessitating the
use of an anti-goat secondary detection reagent.
10. This imaging instrument allows both fluorescent and chemiluminescent imaging at the same resolution with the same
camera. This simplifies alignment and matching of the total
protein (fluorescent) image and immunodetected (chemiluminescent) image.
11. This method could be applied to cell culture supernatant and
cellular lysate prepared from other cell types.
12. Alternative cell culture conditions may be employed, but cells
should be grown in protein-free, defined medium so that no
proteins that do not derive from CHO cells will be found in
the cell culture medium. Suspension culturing will result in a
higher concentration of secreted protein in the growth medium
and will require less concentration to produce a sample to be
analyzed by 2-D electrophoresis.
13. Concentrated cell culture supernatant contains interfering
substances that will prevent effective analysis by 2-D electrophoresis. The 2-D Cleanup Kit is a convenient means for
removing these substances and further concentrating the sample by selective precipitation. Other methods, such as acetone
precipitation [8], have been used successfully. The dry protein
pellet may be slow to dissolve. Brief sonication will accelerate
the process but do not allow the material to heat up.
14. Protein is most accurately quantitated in solution lacking
detergent or reductant, so concentrated protein samples are
first prepared without these additives. The detergent (CHAPS)
and reductant (DTT) are supplied by the solution in which the
sample is diluted prior to first-dimension SDS.
15. IEF may be conducted in either 8 M urea or in 7 M urea, 2 M
thiourea. The urea/thiourea mixture results in better protein
solubilization, particularly of hydrophobic proteins, but urea
411
alone generally results in a cleaner, higher resolution 2-D pattern [9]. In this experiment, the secreted proteins, which are
generally very soluble, are separated in a solution containing
urea alone. Cell lysate protein, which contains more hydrophobic proteins, is separated in a solution containing a urea/
thiourea mixture.
16. Each of the sample solution components has a specific role in
promoting high resolution IEF separation. Urea and thiourea
are protein denaturants that promote the complete unfolding
of polypeptide chains so that all ionizable groups are exposed
to the solution. CHAPS is a detergent that prevents protein
aggregation and promotes solubility. DTT is a thiol reductant
that breaks disulfide bonds within and between polypeptide
chains, and maintains proteins in a fully reduced state. Carrier
ampholytes enhance protein solubility by minimizing protein
aggregation due to chargecharge interactions. Bromophenol
blue is a tracking dye. Its inclusion is not necessary, but clearance of the dye from the IPG strip during electrophoresis provides a visual confirmation that electrical current is being
delivered to the IPG strip appropriately.
17. Each sample is run in replicate and separated on two seconddimension gels, one of which is stained and the other blotted.
18. Recombinant therapeutics can be expressed either as secreted
proteins recovered from the culture medium or as cellular proteins that are recovered from cell lysates. The HCPs found in
either case are expected to differ from each other. Cell culture
medium, however, is expected to contain some proteins that
are released from cells that lyse during growth and transfer.
This experiment analyzes reactivity of an antibody directed
against secreted HCP in both a secreted protein- and cell
lysate-derived samples (see Figs. 13 and 14).
19. Handle IPG strips from the ends using forceps.
20. The sample is applied to the entire length of the immobilized
pH gradient by rehydration into a dry IPG strip. As voltage is
applied, each peptide in the sample focuses to a narrow zone in
the pH gradient corresponding to its pI.
21. Mineral oil prevents evaporation of the sample solution.
22. Additional details of the sample application and IEF procedure
may be found in the user manual for the PROTEAN i12 IEF
System.
23. It will take at least 10 min for the agarose sealing solution to
melt. It is best to start this prior to equilibrating the IPG strips
for the second dimension.
24. Although size standards are not strictly required for this application, they may be applied to the gel at this stage if desired.
412
Fig. 13 CHO cell lysate probed with anti-CHO HCP: Stained gels, stained blot, and Western blot. CHO cell lysate
was separated in duplicate by 2-D electrophoresis. One gel was stained with Oriole Fluorescent gel stain (a).
The second gel was transferred to PVDF membrane and stained with Oriole following transfer (b). The PVDF
membrane was stained with SYPRO Ruby Blot Stain to visualize total protein (c). Following staining, the blot
was immunodetected with anti-CHO HCP and visualized by chemiluminescence (d)
Colored standards (e.g. Precision Plus Protein WesternC

Standard, Bio-Rad) can serve to verify electrophoresis and
transfer and to orient the blotting membrane and determine
which side contains transferred protein. Apply 5 l of standards
solution into the standards well using a syringe or gel loading
pipet tip.
25. The 10 min low voltage step is to allow gradual transfer of
the proteins from the IPG strip to the second-dimension gel.
This reduces vertical streaking and gives a higher resolution
second-dimension separation.
26. The fluorescent stain used is resistant to photobleaching and
shielding the staining reaction from normal room light should
not be necessary. If the gel is under bright light, or in sunlight,
the staining tray may be covered with aluminum foil.
27. The gel is delicate and easily torn. Handle the gel wearing
gloves and avoid touching the gel anywhere besides the edges
and corners. The gel is best held by the bottom corners. Place
the gel on the anode stack and membrane by draping the gel
over the membrane starting from the top of the gel. Once the
413
Fig. 14 CHO cell secreted protein probed with anti-CHO HCP: Stained gels, stained blot and Western blot. CHO
cell secreted protein was separated in duplicate by 2-D electrophoresis. One gel was stained with Oriole
Fluorescent gel stain (a). The second gel was transferred to PVDF membrane and stained with Oriole following
transfer (b). The PVDF membrane was stained with SYPRO Ruby Blot Stain to visualize total protein (c).
Following staining, the blot was immunodetected with anti-CHO HCP and visualized by chemiluminescence (d)
gel is in place on the membrane, it should not be lifted again

or moved.
28. The gel is stained following transfer to verify completeness of
transfer.
29. At this stage, the membrane may be allowed to dry and can be
stored for several weeks at room temperature between sheets
of blotting paper.
30. SYPRO Ruby Protein Blot Stain is a sensitive fluorescent stain
for proteins blotted onto PVDF or nitrocellulose membranes.
It is used in this application because it does not interfere with
subsequent immunodetection [10].
31. Alternatively, this step may be carried out for 1 h at room
temperature.
32. The substrate solution should form a puddle that covers the
blot. Ensure that the surface of the blot is completely covered
with substrate solution.
33. The default settings for capture of chemiluminescence images
take advantage of binning in order to shorten exposure times
414
at the expense of image resolution. In order to capture images

at sufficient resolution for analysis of 2-D gels, and to generate
images at the same resolution as the fluorescent images generated for total protein, binning is turned off, or set to 1 1. This
may require exposure times of several minutes.
References
1. Eaton LC (1995) Host cell contaminant
protein assay development for recombinant
biopharmaceuticals. J Chromatogr A 705:
105114
2. Dagoussat N, Haeuw J-F, Robillard V, Damien
F, Libon C, Corvaa N, Lawny F, Nguyen TN,
Bonnefoy J-Y, Beck A (2001) Development of
a quantitative assay for residual host cell proteins in a recombinant subunit vaccine against
human respiratory syncitial virus. J Immunol
Methods 251:151159
3. Wan M, Wang Y, Rabideau S, Moreadith R,
Schrimsher J, Conn G (2002) An enzymelinked immunosorbent assay for host cell protein contaminants in recombinant PEGylated
staphylokinase mutant SY161. J Pharm Biomed
Anal 28:953963
4. Savino E, Hu B, Sellers J, Sobjak A, Majewski
N, Fenton S, Yang T-Y (2011) Development of
an in-house, process-specific ELISA for detecting HCP in a therapeutic antibody, Part 1.
Bioprocess Int 9:3845
5. Wang X, Hunter AK, Mozier NM (2009) Host
cell proteins in biologics development: identification, quantitation and risk assessment.
Biotechnol Bioeng 103:446458
6. Zhu-Shimoni J, Yu C, Nishihara J, Wong RM,

Gunawan F, Lin M, Krawitz DC, Liu P,
Wandoval W, Vanderlaan M (2014) Host cell
protein testing by ELISA and the use of
orthogonal methods. Biotechnol Bioeng 111:
23672379
7. Tscheliessnig AL, Konrath J, Bates R,
Jungbauer A (2013) Host cell protein analysis
in therapeutic protein bioprocessingmethods
and applications. Biotechnol J 8:655670
8. Valente KN, Schaefer AK, Kempton HR,
Lenhoff AM, Lee KH (2014) Recovery of
Chinese hamster ovary host cell proteins for
proteomic analysis. Biotechnol J 9:8799
9. Rabilloud T, Adessi C, Giraudel A, Lunardi J
(1997) Improvement of the solubilization of
proteins in two-dimensional electrophoresis
with immobilized pH gradients. Electrophoresis
18:307316
10. Berggren K, Steinberg TH, Lauber WM,
Carroll JA, Lopez MF, Cernokalskaya E, Zieske
L, Diwu Z, Haugland RP, Patton WF (1999) A
luminescent ruthenium complex for ultrasensitive detection of proteins immobilized on
membrane supports. Anal Biochem 276:
129143
Chapter 29
Free Flow Electrophoresis for Separation of Native
Membrane Protein Complexes
Lutz Andreas Eichacker, Gerhard Weber, Ute Sukop-Kppel,
and Robert Wildgruber
Abstract
This chapter describes the technology of free flow electrophoresis (FFE) and protocols to separate membrane
protein complexes for proteome analysis. FFE is a highly versatile technology applied in the field of protein
analysis. It is superior to native PAGE due to its fast continuous processing of sample at high resolution.
Additionally, the dynamic separation range from ions, peptides, to proteins, protein complexes, up to
organelles, and whole cells makes it the method of choice in the analysis of proteins. FFE is carried out in an
aqueous medium without inducing any solid matrix, such as acrylamide, so that it simplifies the analysis of
protein complexes for the downstream analysis. Here, we describe the novel zone electrophoresis interval
method (IZE-FFE) for separation of protein complexes from the thylakoid membrane of Arabidopsis thaliana by charge only. Protein complexes isolated by IZE FFE were characterized according to molecular weight
by Blue Native PAGE and were proteins stained with coomassie.
Key words Free flow electrophoresis, Thylakoid membrane, Protein complexes, Blue Native PAGE
Abbreviations
BN
FFE
HPMC
iZE
PAGE
ZE
Blue Native PAGE

Free flow electrophoresis
Hydroxy-propyl-methylcellulose
Interval zone electrophoresis
Polyacrylamide gel electrophoresis
Zone electrophoresis
Introduction
The molecular mass ranges of thylakoid membrane protein
complexes that regulate biogenetic processes cover several orders
of magnitude. Isolation and characterization of protein complexes
and analysis by single particle electrotomography or structure
415
416
Lutz Andreas Eichacker et al.
determination by crystallization requires the biological material in

a liquid environment after separation. In the last years, native
PAGE has become an established laboratory standard for the
separation of protein complexes [14]. The system provides
excellent resolution, but it provides the isolated protein complexes
in a gel environment and high-resolution separation of complexes
takes 16 hours.
The problem of releasing the protein complexes from the gel
is solvable; however, the additional steps compromise complex
stability and require time before the molecules can be analyzed
further. In general, there are three problems with native gels that
can be overcome by using free-flow electrophoresis. Disassembly
of protein complexes at the liquidgel interface is avoided, an
upper limit for entering of complexes into the gel is circumvented,
and separation of the complexes is tremendously speeded up. Why
is it essential to overcome the gel-based limitations? The liquid/gel
junction confronts proteins with a deleterious molecular barrier for
entry into the molecular sieve network and the integrity of many
complexes that fulfill the entry size requirement is compromised.
For supercomplex analysis, the molecular mass entry limit of highresolution polyacrylamide gels according to molecular mass standards appears at about 1.5 MDa; although separation of assemblies
with a molecular mass of about 10 MDa has been claimed [5, 6].
Size exclusion limits an analysis of functional cell structures that
operate in supramolecular arrangements. Analysis of cellular complexity and characterization of the biochemical dynamics requires
to analyze complex arrangements over large molecular mass differences top down, e.g. from the cell to the organelle, to the membrane, and the polysomal/cotranslational level in one experimental
system. A barrier-free analysis strategy is therefore essential to analyze the dynamics of regulatory cell processes. Finally, many complexes are labile and disintegrate during prolonged separation. It is
therefore essential to keep the time between sample preparation
and analysis of the cell components as short as possible.
Free-flow electrophoresis offers a solution to overcome the
constraints of native PAGE. As the name implies, the technique is
gel-free. Using interval zone electrophoresis, separation of membrane protein complexes of interest has been achieved within about
6 min (Fig. 1). Separation is based on the native charge of the
complexes and resolution is achieved by mobility differences in
defined pH zones. The separation technology can be automated
and has been used for analytical and preparative tasks [68].
Separated protein fractions are pooled in microtiter plates and
hence are directly accessible in the liquid state [8]. Steps for analysis of the free-flow fractions, including native and SDS-PAGE,
spectroscopy, mass spectrometry, and single particle analysis have
successfully been conducted and optimized recently.
Free Flow Electrophoresis for Separation
417
Fig. 1 FFE-separated protein complexes visualized after native BN-PAGE. Thylakoid membranes corresponding
to 300 g Chl were solubilized in 300 L detergent mixture 1 and 2. Supernatant 2 was applied to Free Flow
Electrophoresis. The FFE fractions containing the solubilized protein complexes were then concentrated and
subsequently loaded onto native BN-PAGE. Chl-protein complexes and standard proteins (kDa) were stained
with colloidal Coomassie G250 and detected by white light scanning
Materials
2.1 Preparation
of Thylakoid
Membranes (Rosette
Leaves
from Arabidopsis
thaliana)
All buffers should be freshly prepared and stored at 4 C.

1. Extraction buffer: 25 mM Tricine-NaOH, pH 7.8, 330 mM
sorbitol, 1 mM Na-EDTA, 10 mM KCl, 0.15 % (w/v) BSA,
4 mM sodium ascorbate.
2. Lysis buffer: 10 mM Tricine-NaOH, pH 7.8, 5 mM MgCl2,
10 mM NaF.
3. Washing Buffer: 25 mM Tricine-NaOH, pH 7.8, 100 mM
sorbitol, 5 mM MgCl2, 10 mM KCl, 10 mM NaF.
4. Storage buffer (TMKGS): 10 % (v/v) glycerol, 25 mM TricineNaOH, pH 7.8, 100 mM sorbitol, 5 mM MgCl2, 10 mM KCl.
2.2 Solubilization
Buffers
for the Membranes
1. DDM solution: 195.84 mM (10 % w/v) n-dodecyl--Dmaltoside. Store at 20 C.

2. DIG solution: 81.34 mM (10 % w/v) digitonin (see Note 1).
Store at 20 C.
3. Solubilization solution 1: 16 mM Digitonin, 250 mM Sucrose,
50 % v/v TMKGS.
4. Solubilization solution 2: 8 mM Digitonin, 8 mM n-dodecyl-D-maltoside, 250 mM Sucrose, 50 % v/v TMKGS.
418
2.3
FFE Media
1. Anodic stabilization medium (inlet 1): 100 mM HCl, 50 mM

Formic acid, 50 mM Hydroxy-isobutyric acid (HIBA),
250 mM Sucrose, adjust with BisTris to pH 3.84.1.
2. Separation medium 1 (inlets 2, 3, 4): 10 mM HIBA, 250 mM
Sucrose, 0.1 % Digitonin, adjust with BisTris to pH 5.4.
3. Separation medium 2 (inlets 5, 6): 10 mM HIBA, 250 mM
4. Separation medium 3 (inlet 7): 10 mM HIBA, 250 mM
Sucrose, 5 mM NaCl, 0.1 % Digitonin, adjust with BisTris to
pH 7.0.
5. Separation medium 4 (inlet 8): 10 mM HIBA, 250 mM
6. Cathodic stabilization medium (inlet 9): 150 mM HIBA,
375 mM Imidazol, 250 mM Sucrose.
7. Counterflow medium: 250 mM Sucrose, 50 mM BisTris,
20
mM
N-(1,1-Dimethyl-2-hydroxyethyl)-3-amino-2hydroxypropanesulfonic acid (AMPSO), resulting in pH 8.0.
8. Anodic electrode medium: 100 mM sulfuric acid.
9. Cathodic electrode medium: 100 mM sodium hydroxide,
200 mM Glycine, resulting in pH 10.
10. FFE System: FFE advanced, FFE Service GmbH, Munich,
Germany.
11. Viva-Spin 100,000 MWCO tubes at 10 C (Viva-Products,
Inc., Littleton, MA).
2.4
BlueNative-PAGE
1. Thylakoid membranes from Arabidopsis thaliana: 1 g/L

total Chlorophyll, frozen in liquid N2. Stored at 80 C.
2. Cathode buffer: 50 mM Tricine, 50 mM BisTris, resulting in
pH 6.8, 0.02 % Coomassie G250. Assemble as 10 concentrate.
Store at 4 C (see Notes 2 and 3).
3. Anode buffer: 50 mM Tricine, 50 mM Bis-Tris, resulting in
pH 6.8. Assemble as 10 concentrate. Store at 4 C (see Note 3).
4. Electrophoresis gel: 312 % Bis-Tris-Gels, 100 100 mm.
5. Electrophoresis chamber: vertical slab gel electrophoresis unit
(e.g. mini cell from Novex).
6. Power supply.
2.5 Analysis
of Thylakoid
Membrane Complexes
1. Molecular weight standard: NativeMark (Invitrogen).

2. White light scanning: Epson Perfection V750, transmission
mode scanner.
3. Colloidal Coomassie stain: 0.12 % (w/v) Coomassie G250,
20 % (v/v) Methanol, 10 % (v/v) Phosphoric acid, 10 %
Ammonium sulfate [9].
419
Methods
The FFE system performs electrophoretic separations in patented
and well-defined separation solutions without the use of a solid
(e.g. polyacrylamide) matrix. The gel-free basis enables the highresolution separation of charged or chargeable entities like cell
organelles, protein complexes and proteins, or peptides on a fast,
preparative, and continuous basis. Fluid-phase separation is operated in the three modes zone electrophoresis (ZE), including the
novel iZE method, isoelectric focusing (IEF), and isotachophoresis.
The system enriches low-abundant proteins with high reproducibility. The sample is applied using a peristaltic pump and is induced
into a separation chamber consisting of two parallel plates. Under
laminar flow, the sample is transported within a thin (0.20.4 mm)
film of aqueous medium formed between the two plates. The plates
are bordered by two electrodes that generate a high-voltage electric
field perpendicular to the laminar flow. Charged particles like ions,
peptides, proteins, organelles, membrane fragments, or whole cells
that are deflected in this electric field are separated and seamlessly
fractionated into microtiter plates.
The iZE method described in this chapter delivers excellent
resolution for the separation of protein complexes and membrane
proteins. Compared to IEF separations, media require no addition
of polymers like Hydroxy-ethyl-propyl-cellulose (HPMC) to suppress electroendosmotic flow. This enables an easy concentration
of the separated samples by ultra-filtration and ensures highest
compatibility for any downstream technology like, e.g. native and
SDS-PAGE, HPLC, MS, and ELISA.
The protocol describes how protein assemblies from the thylakoid membrane of Arabidopsis thaliana can be prepared for iZEFFE separation. Separation by iZE-FFE is based on the charge of
the protein assemblies. The resolution of the iZE-FFE protocol is
visualized using blue native PAGE as a native PAGE approach. In
general, blue-native PAGE is intended to separate protein complexes in a state that reflects the physiologically relevant functional
state of the protein assembly [4]. It is therefore advisable to minimize protein complex degradation through sample preparation on
ice and fast electrophoresis at 4 C. Solubilization of thylakoid
membranes resulted in the formation of membrane assemblies of
chlorophyll-binding thylakoid membrane protein complexes that
were charge separated by iZE-FFE within 6 min. The composition
of the FFE fractions was analyzed in a BN-PAGE approach. Here,
one to several protein bands have been determined per FFE fraction. Unique protein compositions were identified per every
second FFE fraction that was loaded for BN-PAGE separation.
Especially, five distinctive maxima were determined after the FFE
fractionation (Fig. 1, fractions 35, 37, 49, 53 and 59). BN-PAGE
420
lanes and proteins bands were analyzed further by denaturing

SDS-PAGE and proteins and complexes were identified by mass
spectrometry. Data revealed that iZ-FFE was perfectly suited to
separate the highly complex array of membrane protein complexes
from the photosynthetic thylakoid membrane. Protein complexes
and subcomplexes were well detected upon Coomassie staining
and white light scanning (Fig. 1). Detailed characterization of protein complexes and subcomplexes is presently under investigation.
Separation of the protein complexes by native PAGE is based on a
sieving of the complexes by the acrylamide concentrationdependent pore size of the gel. The method presented here corresponds to a 312 % (v/v) linear gradient separating gel (Fig. 1).
The gradient gel facilitated a separation of molecular weight standards in the molecular weight range of 661,500 kDa (Fig. 1).
3.1 Arabidopsis
thaliana Sample
Preparation
1. Arabidopsis thaliana plants are grown on soil for 34 weeks in a

growth chamber with ambient white light of 100 mol/m2/s at
a light/dark cycle of 8 h-light/16 h-dark.
2. Homogenize leaves 4 4 s in 150 mL of ice-cold extraction
buffer.
3. Filter homogenized leaves through one layer of Miracloth and
collect through a funnel in 3 50 mL conical tubes.
4. Centrifuge for 3 min at 1,800 g. Collect the pellet and discard
the supernatant.
5. Gently resuspend the pellet in 5 mL extraction buffer using a
paintbrush. Wash the brush in 25 mL extraction buffer.
Redistribute into two conical tubes and fill to 50 mL with
extraction buffer.
6. Centrifuge for 3 min at 1,800 g. Collect the pellet and discard
the supernatant.
7. Gently resuspend each pellet in 5 mL of lysis buffer with a
paintbrush.
8. Combine resuspended material in one conical tube and fill to
50 mL with lysis buffer.
9. Divide sample into two Sorvall HB4 tubes and fill to 25 mL
with lysis buffer.
10. Incubate sample for 5 min in the dark on ice.
11. Centrifuge for 5 min at 6,000 g.
12. Supernatant = chloroplast stroma. Purify stroma from residual
thylakoids by centrifugation at 12,000 g for 30 min.
13. Pellet = chloroplast thylakoid membrane.
14. Gently resuspend thylakoid membrane pellets from both
tubes in 5 mL of wash buffer with a paintbrush and collect in
one tube.
421
15. Collect thylakoids by centrifugation for 5 min at 5,900 g.

16. Discard the supernatant and keep the pelleted thylakoids.
Resuspend thylakoids in 3 mL wash buffer using paintbrush.
17. Transfer into glass homogenizer and lever piston three times.
18. Transfer into 5 mL plastic tube.
19. Clean homogenizer with 2 mL of wash buffer and homogenize
2 and collect in 5 mL tube.
20. Measure the concentration of chlorophyll. Transfer 10 L thylakoid membrane extract into 990 L 80 % Acetone at 4 C.
Determine absorbance at 652 nm. Calculate the concentration
of Chl (g/L).
21. For storage of thylakoid membranes determine Chl concentration and dilute in storage buffer to a concentration of 1 g
Chl/L.
22. Freeze aliquots of 100 L in liquid N2 and store tubes at 80 C.
3.2 Solubilization
of Thylakoid
Membranes
from Arabidopsis
thaliana for Native FFE
1. Use thylakoid membranes corresponding to 300 g Chl.

2. Concentrate thylakoid membranes by centrifugation at
7,500 g and 4 C for 10 min and discard supernatant.
3. Solubilize in 300 L of detergent mixture 1 for 10 min at 10 C
(see Note 1).
4. Separate non-solubilized material
25,000 g and 10 C for 10 min.
by
centrifugation
at
5. Transfer the supernatant 1 containing the solubilized protein

complexes to a new micro tube.
6. Solubilize the pellet in 300 L of detergent mixture 2 for
10 min at 10 C.
7. Separate non-solubilized material
25,000 g and 10 C for 10 min.
by
centrifugation
at
8. Transfer the supernatant 2 containing the solubilized protein

complexes to a new micro tube.
9. Load supernatant of the different solubilizations onto FFE
(see Fig. 1 for separation of supernatant 2) (see Note 2).
3.3
FFE Method
A native separation buffer and the novel interval Zone

Electrophoresis protocol is used for protein complex separation.
FFE is conducted at 10 C using the following conditions:
The experiments run in a horizontal position of the separation
chamber using a 0.2 mm spacer. The voltage is adjusted to 1,600 V
which results in a current of ~85 mA. The residence time in the
separation chamber is approximately 4.5 min per interval. Fractions
are collected in polypropylene microtiter plates numbered from 1
(anode) through 96 (cathode).
422
1. Place the sample to the ice cooled sample holder on the FFE
system (see Notes 37).
2. Clean interior surface of chamber: move separation chamber to
an upright position and open the pairs of chamber clamps
simultaneously. Use the following sequence for cleaning
(H2Odist 18 n/20 S):water isopropanolpetroletherisopropanolwater (see Note 8).
3. Apply the spacer (0.2 mm) without covering the inlets, apply
the electrode membrane (soaked in Glycerol/Isopropanol,
mix 1:1) and the filter paper (0.3 mm), soaked in distilled
water onto the membrane.
4. Close separation chamber: lift the Plexiglas side of the chamber
to meet the mirrored side, first fasten central and adjacent
clamps loosely (release chamber clamps by two turns each).
Then fasten central and adjacent clamps tight, close top and
bottom clamps, place fractionation plate on top of fractionation housing, finally make sure that all inlets of media and
counterflow are free and are not covered by the spacer.
5. Fill separation chamber with degassed dist. water: open (lift)
valves at the top of the chamber to degas the chamber, fasten
wedge clamps (I1I9) of the media pump and place tubes in
reservoir filled with H2Odist, switch on medium pump on
control board. When the chamber is totally flushed and free of
air add counter flow by closing wedge clamps on pump.
6. Set up the Electrode Buffer, Stabilization Buffers and counter
flow buffer and separation buffers.
7. Connect anode (+) and cathode () electrolytes, connect plastic cover and start electrode pump.
8. Adjust the media flow rate to 120 mL/h and start the media
flow, wait for 15 min to flush the buffers in.
9. Adjust the voltage to 1,600 V, set the current to 150 mA.
10. Set the sample flow rate to 4,000 L/h.
11. Set the media pump flow rate to 16.4 digits.
12. Turn on media pump.
13. Turn on sample pump and flush the sample in for 70 s, then
switch the sample pump to backwards direction for 2 s and
turn off the sample pump.
14. Wait 20 additional seconds, and then turn on the high
voltage.
15. Wait 30 s, and then adjust the media pump to 3.6 units.
16. Wait 4 min, then turn the high voltage off and elute the
separated samples with a media pump speed of 28.8 units.
423
17. After 100 s the sample can be collected with a MTP or deep
well plate, therefore place the MTP in the fraction collection
device and collect for 120 s.
18. If more separated sample is needed, repeat the steps 817.
3.4 Concentration
of FFE Fractions
3.5 Electrophoresis
of Protein Complexes
from FFE Fractions
on BN PAGE
For native LN- and BN-PAGE analysis, a volume of 200 L of each

IZE fraction was concentrated by centrifugation using Viva-Spin
100,000 MWCO tubes at 10 C (Viva-Products, Inc., Littleton,
MA) according to the manufacturers protocol until a volume of
25 L was retained.
1. Prepare cathode-, anode- and gel-media for BN-PAGE-gels.
Store at 4 C.
2. Label the position of the wells on the outer cassette with a felt pen
and remove the comb carefully from the native gel (see Note 9).
3. Assemble cassette sandwich within electrophoretic apparatus.
4. Fill in cathode buffer into the upper buffer chamber.
5. Rinse the wells with cathode buffer (see Note 10).
6. Underlay the samples into the wells using a micro syringe or
disposable pipette tips (see Note 11).
7. Fill anode buffer in the lower buffer chamber until electrode is
immersed.
8. Place the gel chamber in a Styrofoam box, use 2 minus 20 C
cooling pads to maintain low temperature during separation.
9. Complete assembly of electrophoresis unit and connect to
power supply.
10. Set the power supply to limit voltage. Use 35 V constant.
Set mA and W to maximal values. Run overnight for highest
resolution. Apply power to electrophoretic set (see Note 12).
11. Stop electrophoresis when the front composed of Ponceau
S/chlorophyll has reached the bottom of the separating gel
(see Note 13).
12. For further analysis of protein complexes like absorbance or
mass spectroscopy measurements open sandwich and cut chlorophyll gel bands or stain proteins before further processing of
sample of interest.
Notes
1. Digitonin is dissolved by heating the solution to 100 C. After
heating, Digitonin can immediately be mixed with other detergents. The rest of the detergent mixture can be stored at
20 C. Always reheat solutions before usage.
424
2. Depending on the organism and membrane that should be

solubilized, different detergent concentrations and mixtures
need to be tested individually.
3. Turbidity of protein samples is an indication of insufficient
solubility and/or protein precipitation. Protein samples may
not be turbid; otherwise the resolution of the electrophoretic
separation will be poor, turbid protein samples have to be
cleared by centrifugation prior to the FFE separation.
4. The salt concentration of samples may not exceed 25 mM.
If the conductivity of the sample due to the salt concentration
is too high, it has to be desalted or diluted to reach <25 mM of
total salt concentration.
5. Generally, the sample should be as similar to the separation
media as possible in behalf of the chemical and physical properties like density, conductivity and viscosity.
6. Additionally, tolerable additives for sample preparations can be:
Urea up to 8 M; 0.11 % of detergents like CHAPS, CHAPSO,
Octyl--D-glucoside,
Digitonin,
Dodecyl--D-maltoside,
Triton-X-114 (IEF); up to 50 mM DTT.
7. Protein sample is usually applied on the cathodic inlet, only
few samples work better when applied at the middle or anodic
inlet; this can be checked for each sample.
8. The system has to be cleaned thoroughly prior use, usually the
separation chamber is wiped with water, isopropanol and
petrolether and again with isopropanol and water to make sure
any organic and anorganic contamination is cleaned.
9. Depending on the electrophoresis apparatus, the gel comb has
to be removed before or after assembly of the sandwich in the
electrophoresis apparatus.
10. Fill 50 mL syringe with cathode buffer and use needle with
diameter thinner than spacer thickness to rinse wells.
11. If a micro syringe is used, rinse with anode buffer before
applying a new sample. If wells remain unused, load detergent
mixture 1 or 2 to the empty well.
12. Depending on laboratory condition and available equipment,
the run of the native gel electrophoresis could be done in the
cold room. In order to control the running behavior, check
temperature and pH of cathode buffer in upper buffer
chamber.
13. Gel requires about 18 h to complete separation of the solubilized protein complexes (Invitrogen, 35 V constant, Gel
length = 100 mm).
425
References
1. Schgger H, Von Jagow G (1991) Blue native
electrophoresis for isolation of membrane protein complexes in enzymatically active form.
2. Reisinger V, Eichacker LA (2012) Native DIGE
of fluorescent plant protein complexes. In:
Cramer R, Westermeier R (eds) Difference gel
electrophoresis (DIGE). Humana Press,
New York, pp 343353
3. 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 27:27592781
4. Wittig I, Karas M, Schgger H (2007) High resolution clear native electrophoresis for in-gel functional assays and fluorescence studies of membrane
protein complexes. Mol Cell Proteomics 6:
12151225
5. Strecker V, Wumaier Z, Wittig I, Schgger H

(2010) Large pore gels to separate mega protein
complexes larger than 10 MDa by blue native
electrophoresis: isolation of putative respiratory
strings or patches. Proteomics 10:33793387
6. Justesen BH et al (2013) Isolation of monodisperse nanodisc-reconstituted membrane proteins
using free flow electrophoresis. Anal Chem
85:34973500
7. Canut H, Bauer J, Weber G (1999) Separation
of plant membranes by electromigration techniques. J Chromatogr B Biomed Sci Appl
722:121139
8. Wildgruber R, Weber G, Wise P, Grimm D, Bauer
J (2014) Free-flow electrophoresis in proteome
sample preparation. Proteomics 14:629636
9. Candiano G et al (2004) Blue silver: a very sensitive colloidal Coomassie G-250 staining for proteome analysis. Electrophoresis 25:13271333
Chapter 30
Three-Dimensional Electrophoresis for Quantitative
Profiling of Complex Proteomes
Sergio Mauro, Bertrand Colignon, Marc Dieu, Edouard Delaive,
and Martine Raes
Abstract
Quantitative 2D-gel-dependent proteomics became feasible with 2D fluorescence difference gel electrophoresis
(2D-DIGE), and this technique has gained wide acceptance because it has eliminated the gel to gel variations
and greatly facilitated the quantitative comparisons across gels for many different experimental conditions.
However, the co-migration of several proteins in the same spot is still a major limitation which detracts from
the accuracy of comparative quantification and prevents unambiguous post-translational modifications
(PTMs) detection.
A protocol based on traditional polyacrylamide gel IEF sample fractionation, and followed by two
consecutive SDS-PAGE electrophoreses alleviates co-migration limitations. The use of two different buffer
systems for SDS-PAGE is central to the proposed approach.
Key words 2D-DIGE, 3D-DIGE, Western blot, Co-migration, Complex proteomes
Introduction
The early work of Tiselius demonstrated the value of electrophoretic
fractionation as an analytical technique for separating proteins [1].
Since this pioneering work, the technique has seen many refinements resulting in increased resolving power.
The introduction of inert support materials such as paper,
agarose, and acrylamide greatly improved the resolving power by
eliminating the continuous mixing of the sample components associated with the current-driven convection.
Among the various separation media, polyacrylamide gel has
gained the most widespread use. In addition to preventing thermal
convection, the polyacrylamide matrix acts as a molecular size
selective sieve increasing the resolving power. Therefore, during
polyacrylamide gel electrophoresis (PAGE) protein separation is
based on a combination of charge, shape, and size.
427
428
Sergio Mauro et al.
Two additional major technical improvements turned this

veteran technique into a high-throughput quantitative proteomic
analytical method.
First, the introduction of the negatively charged sodium dodecyl
sulfate detergent (SDS) brought two unique advances [24]. SDS
binds to most proteins at a constant ratio of 1.4 g SDS per g protein
and forces globular proteins to adopt a rod-like conformation so
that the migration rate of the proteins during SDS-PAGE is inversely
proportional to their molecular weight. An additional advantage is
that the detergent action of SDS solubilizes insoluble proteins
including membrane proteins and disrupts supramolecular assemblies and this makes SDS PAGE a method of choice for resolving
complex mixtures of proteins.
Second, the discontinuous buffer system which was introduced
by Laemmli [5] is by far the most commonly used PAGE protocol
for protein separation. The advantage of the discontinuous buffer
system is that the resolving power is increased by first concentrating the proteins together in a small zone, before they are separated
as a function of their molecular weight. This is achieved by casting
the gels in two sections with different properties: a first (the stacking gel) and a second (the resolving gel). The stacking gel has large
pores so the proteins are not sieved but the pH is chosen so that
the cations and anions from the electrode and gel buffers migrate
as two separate fronts. This concentrates the proteins of the sample
in the very narrow zone between the fronts. When the proteins
reach the second, resolving, portion of the gel, the two ionic fronts
disappear and the proteins are separated, thanks to the SDS,
according to their molecular weight.
In 1975 OFarrel [6] introduced two-dimensional electrophoresis by combining isoelectric focusing (IEF) of proteins in the first
dimension and SDS-PAGE as the second separation dimension
(2-dimensional SDS PAGE, 2D-PAGE). This improved the utility
of gel electrophoresis for resolving complex mixtures of proteins
but still left the problem of gel to gel variability (due to the fact
that known and unknown have to be run on different gels).
Quantitative 2D-gel-dependent proteomics became feasible with
2D fluorescence difference gel electrophoresis (2D-DIGE), and this
technique has gained wide acceptance because it alleviates some of
the problems that have bedeviled the use of 2D-PAGE. 2D-DIGE
essentially differs from the traditional 2D-PAGE in that it relies on
the pre-electrophoretic labeling of samples with several fluorescent
dyes [7]. The use of internal standard, made by pooling the different
samples, provides a reference for each protein. This eliminated the
gel to gel variations and greatly facilitated the quantitative comparisons across gels for many different experimental conditions.
However, despite these significant improvements, there is still
a major limitation, i.e., the co-migration of several proteins in the
same spot. This prevents the unambiguous identification of proteins
3D Differential Electrophoresis
429
and detracts from the accuracy of comparative quantification, and

complicates the link between identification, quantification, and specific protein detection such as Western blot or specific staining. It also
prevents unambiguous PTMs detection [8].
Recent improvements in the performance of mass spectrometry have facilitated the development of high-precision quantitative proteomic using non-gel-based techniques such as ICATs and
iTRAQ. Nevertheless, extensive studies comparing gel-free and
in-gel proteomics strategies show that there are useful complementarities between the two [810]. Actually, in spite of some
shortcomings, 2D-PAGE, and in particular 2D-DIGE, remains a
powerful, rapid, and robust approach.
Here, we describe a simple and rapid 3D protocol (3D-DIGE)
which addresses the problem of co-migration interference. In contrast to previous studies [1113], our protocol has a wide range of
application in quantitative profiling of complex proteomes. It is based
on traditional polyacrylamide gel IEF sample fractionation. This is
followed by two consecutive SDS-PAGE electrophoreses. The use of
two different buffer systems for SDS-PAGE is central to our approach
which has been validated by comparing mass spectrometry (MS)
results for protein identification from 2D and 3D gels [14].
Materials
Unless indicated otherwise all solutions are prepared with Milli-Q
water (resistivity, 18.2 M cm at 25 C).
2.1 Sample
Preparation
and Isoelectric
Focusing
1. Immobilized pH gradient (IPG) strips, 24 cm, pH 47.

2. IPG buffer (pH 47).
3. IPG-strip rehydration buffer: 7 M Urea, 2 M Thiourea, 2 %
(w/v) CHAPS, 30 mM TrisHCl, pH 8.0. Aliquot into plastic
tubes and store at 20 C until use. Just before rehydrating
IPG-strips add 50 mM DTT.
4. Immobiline DryStrip Cover Fluid.
5. Anhydrous dimethylformamide (DMF) (99.8 % pure).
6. 2 mM TCEP (Tris(2-carboxyethyl)phosphine hydrochloride).
7. CyDye DIGE Fluor labeling Kit for Scarce Samples and
Preparative Gel labeling.
8. 1 M DTT.
9. Sample cups.
10. Paper electrode wicks.
11. IPG-strip Reswelling Tray.
12. IPGphor Manifold Ceramic Tray.
13. Ettan IPGphor3 Isoelectric Focusing System.
430
Sergio Mauro et al.
14. Sample rehydration buffer: 30 mM TrisHCl, pH 8.0, 7 M

Urea, 2 M Thiourea, 4 % (w/v) CHAPS.
15. Stopping buffer: 30 mM TrisHCl, pH 8.0, 7 M Urea, 2 M
Thiourea, 4 % (w/v) CHAPS, 260 mM DTT, 4 % IPG buffer.
2.2 2D- and
3D-Difference
Gel Electrophoresis
1. Precast polyacrylamide DIGE gels in a low fluorescent glass

cassette.
2. Ettan DALTsix Electrophoresis Unit.
3. DIGE Buffer Kit which consists of concentrated running buffers and sealing solution for attaching IPG-strips to the top of
the polyacrylamide gel.
4. Equilibration buffer: 6 M urea, 75 mM TrisHCl, pH 8.8,
30 % v/v glycerol, 2 % w/v SDS and 0.01 % w/v bromophenol
blue. 50 mM DTT or 50 mM IAA are added just prior to use.
5. NuPAGE Novex 412 % Bis-Tris Protein Gels, 1.0 mm,
15 well.
6. NuPAGE MES SDS running buffer (20).
7. NuPAGE Novex 7 % Tris-acetate protein gels, 1.0 mm
15 well.
8. NuPAGE Tris-acetate SDS running buffer (20).
9. XCell SureLock Mini-Cell.
10. NuPage LDS sample buffer.
11. 20 % SDS stock solution.
12. Prestained protein molecular weight marker.
2.3 2D- and 3DDifference Gel

Electrophoresis Image
Capture and Analysis
1. Typhoon 9400.
2. DeCyder 2-D Differential Analysis Software v6.5.
1. ECL Semi-Dry Blotter.

2.4 Western Blot
Analysis and MS
Protein Identification
2. Hybond LFP 0.2 PVDF membrane optimized for fluorescence

detection applications.
3. ECL blocking agent.
4. Primary antibody: SUMO1, small ubiquitin-like modifier
protein1.
5. Secondary antibody: Alexa Fluo Cy5647-AffiniPure F(ab)
2 fragment goat anti-rabbit IgG (H + L).
6. Nano-LC-ESI-MS/MS maXis Impact UHR-TOF (Bruker,
Bremen, Germany) coupled with a nanoLC UltiMate 3000
(Thermo Scientific).
431
Methods
3.1 Sample
Preparation
and Isoelectric
Focusing
1. Remove the CyDye DIGE Fluor from the 20 C freezer and

leave it to equilibrate at room temperature before opening.
3.1.1 Reconstitution
of CyDye Fluor
for Saturation DIGE-Based
3. For analytical 2D-DIGE gels, add 50 L DMF so as to reconstitute a 2 mM working dye solution (Cy3 and Cy5 are available).
2. For preparative 2D-DIGE gels, add 20 L DMF so as to reconstitute a 20 mM working dye solution (only Cy3 is available).
4. Vortex for 30 s.
5. Centrifuge for 30 s at 12,000 g.
6. If not used directly the solutions may be stored at 20 C up
to 3 months.
3.1.2 Protein Labeling

Using Saturation Dyes
Protein Labeling
for Analytic Gels
In order to eliminate variation between gels, treatment effects are

measured by reference to an internal standard. This standard is a
mix of an equal amount of all protein extractions. Thus for our
heat treatment experiment we have three sorts of sample: control,
treatment, and internal standard.
1. In an Eppendorf tube, add the sample volume corresponding
to 5 g protein (see Note 2).
2. Adjust the volume to 9 L with the sample rehydration buffer
(see Note 3).
3. Sonicate on ice (6 10-s pulses).
4. Add 1 L of a fresh 2 mM TCEP solution in water.
5. After vortexing, shortly centrifuge the sample in a microcentrifuge.
6. Incubate in the dark at 37 C for 1 h.
7. Add 2 L of the 2 mM CyDye DIGE Fluor.
8. Add CyDye DIGE Fluor Cy3 to label the pooled protein
extract (internal standard).
9. Add CyDye DIGE Fluor Cy5 to label the control or treated
protein extracts.
10. Mix each thoroughly and centrifuge in a microcentrifuge.
11. Incubate in the dark at 37 C for 30 min.
12. Stop the reaction by adding an equal volume of stopping
buffer.
13. Mix again thoroughly and centrifuge in a microcentrifuge.
14. Samples are ready for use or for conservation at 80 C until use.
432
Sergio Mauro et al.
Protein Labeling
for Preparative Gels
1. Add the sample volume corresponding to 100 g protein to an

Eppendorf tube (the final volume should not exceed 120 L).
2. Add 2 L of 20 mM TCEP.
3. Incubate at 37 C for 1 h.
4. Add 4 L of Dye for preparative gel (see Note 4).
5. Incubate 30 min at 37 C.
6. Stop the reaction by adding DTT (130 mM final), IPG buffer
47 (2 % final), and sample buffer up to max. 120 L.
7. Samples are ready for use or for conservation at 80 C
until use.
3.1.3 First Dimension

Isoelectric Focusing
IPG-Strip Rehydration
The protocol below describes the cup-loading approach that has

been used for both the analytical and preparative gels using 24 cm
long IPG-strips pH 47 for samples prepared as above.
1. Carefully level the reswelling tray so to ensure homogenous
rehydration of the strip.
2. Mix 450 L of the IPG-strip rehydration buffer with 8.9 L
of 1 M DTT stock solution.
3. Deliver 450 L of this mix into the center of one of the slots of
the re-swelling tray and remove any air bubble.
4. Remove the protective cover from the IPG-strip (see Note 5).
5. Position the IPG-strip with the gel side down and slide it back
and forth to remove air bubbles.
6. Overlay the IPG-strip with 3 mL of IPG cover fluid to limit
evaporation.
7. Allow the strip to rehydrate at room temperature for at least
16 h but do not exceed 24 h.
In-Cup Loading Sample

Application
1. Place the IPGphor Manifold onto the cooling plate of the

Ettan IPGphor3.
2. Remove the IPG-strip from the re-swelling tray with forceps.
3. Put the rehydrated IPG-gel strip into a slot of the IPGphor
Manifold, gel side up making sure that the acidic end of the
strip is facing toward the anode.
4. Moisten two paper electrode wicks with 150 L distilled water
and remove any excess of water by blotting with filter paper.
5. Apply these electrode wicks at the terminals of the IPG-strips.
6. Clip the moveable electrodes above the electrode wicks.
7. Position the moveable sample cup near the anode.
8. Ensure that there are no leaks in the sample cup, by filling the
sample cup with Immobiline DryStrip Cover Fluid.
433
9. Apply 108 mL of DryStrip Cover Fluid to cover the Manifold

slots.
10. For analytical gels, mix thoroughly 5 g of Cy3-labeled internal standard and 5 g of Cy5 experimental sample.
11. Load the mix in the sample cup.
12. For preparative gels load up to 120 L of the protein mix in
the sample cup.
13. Close the lid of the Ettan IPGphor3.
14. Program the ETTAN IPGphore3 as follows in order to focus
the proteins overnight:
(a) Step and hold 300 V/3 h.
(b) Gradient 600 V/3 h.
(c) Gradient 1,000 V/8 h.
(d) Gradient 8,000 V/3 h.
(e) 8,000 V until 50,000 Vh are reached.
15. After IEF completion, IPG-strips are either used immediately
or after removal of excess DryStrip Cover Fluid, stored at
80 C in a rigid container.
3.2 2D- and
3D-Difference Gel
Electrophoresis
1. Prepare 10 mL (fresh or just thawed from the freezer) of the

focused strips equilibration buffer containing 50 mM DTT
added prior to use.
3.2.1 Equilibration
of IPG-Strips
2. Prepare 10 mL (fresh or just thawed from the freezer) of the

focused strips equilibration buffer containing 50 mM IAA
added prior to use.
3. Equilibrate the IPG strip with gentle shaking for 10 min in
equilibration buffer containing DTT.
4. Equilibrate the IPG strip with gentle shaking for 10 min in
equilibration buffer containing IAA.
5. Rinse the IPG-strip with DIGE cathode buffer.
3.2.2 2D Difference
Gel Electrophoresis
1. Using a forceps and spatula place the IPG-strip on top of the

precast polyacrylamide gel cassette, ensuring close contact
between the strip and the polyacrylamide gel surface. Both analytical and preparative electrophoreses have been performed
using precast 12.5 % polyacrylamide 2D-DIGE gels.
2. Apply 4 L of the appropriate protein molecular weight markers blend to a 10 mm 5 mm piece of filter paper.
3. Place the filter paper next to the IPG strip.
4. Add 1 mL of hot agarose sealing solution over the IPG strip
and the filter paper.
5. Place the gel in Ettan DALTsix and fill any empty slots with
blank cassette inserts.
434
Sergio Mauro et al.
6. Wet the upper buffer chamber seals with some cathode buffer
and slide the chamber into place.
7. Fill the upper chamber with cathode buffer (approximately
1.2 L).
8. Fill the lower chamber with anode buffer up to the fill line.
9. Run the gels initially at constant 80 V for 1 h and then at
12 mA/gel overnight (1.5 W/gel) until the bromophenol
blue front reaches the end of the gel.
3.2.3 Two- and ThreeDimensional Gel
Electrophoresis Image
Capture and Analysis
3.2.4 Quantitative
3D-DIGE for Comparative
Proteomics Using
Analytical 2D-DIGE Gels
1. Capture the gels with a DIGE compatible imaging system.

2. Perform data analysis (differentially expressed protein spots)
with a DIGE compatible software package [1517].
3. Pick corresponding gel spots by hand or robot.
The 3D deconvolution of the 2D spots into their individual protein
components could be systematically achieved for any 2D spot.
However we have used it to study changes in Solanum tuberosum
leaf proteomes following a heat shock treatment and we have
shown that a robust gel image-based quantitative comparison of
protein expression levels could be performed on 3D gels [14].
The full 3D-DIGE protocol has two steps: analytical and
preparative (Fig. 1):
1. The analytical 3D-DIGE identifies and quantifies the differences between samples in the protein expression profile. This is
achieved by measuring the intensity of fluorescence of Cy3 and
Cy5 dyes in each spot which has been resolved in the second
dimension (2D-DIGE). The analytical 3D-DIGE step resolves
and quantifies the protein components present in each protein
spot identified in the 2D step (Fig. 2).
2. The preparative 3D-DIGE step was developed because 3D
spots derived from 3D analytical gels may not contain sufficient
material for downstream applications such as mass spectrometry and Western blotting analysis.
Our protocol begins with the 2D-DIGE procedure as described
above. Then the spot picking list is used to command a mechanical
spot picker and the excised samples are the starting point for a
third, or 3D-DIGE analysis.
The excised spots corresponding to control and heat shock are
delivered in a multiwell plate
1. Remove any liquid that has leached out from the excised gel
fragments.
2. Select a gel system as a function of the apparent molecular mass
(MM) of the excised spots, for MM 55 kDa load the excised
spot onto a NuPage Tris Acetate 7 % acrylamide gel; for MM
55 kDa use a Bis-Tris 412 % acrylamide MES buffer gel.
Fig. 1 The work flow for 3D-DIGE analysis for quantitative profiling of complex proteomes. The 3D-DIGE protocol
begins with the 2D-DIGE analytical gel analysis which generates the list of informative spots. These spots are
excised from a preparative 2D-DIGE gel and further electrophoresed in a different gel system. The resolved individual
components are analyzed in downstream applications such as mass spectrometry and Western blot analyses
436
Sergio Mauro et al.
Fig. 2 Analytical 3D-DIGE of total protein extracts of control and heat-treated leaves of Solanum tuberosum.
(a). Resolution of Solanum tuberosum leaves protein extracts by 2D-DIGE. Extracts from internal standard and
heat-treated leaves (42 C, 1 h) samples were differentially labeled with the Cy5 and Cy3 and separated by
two-dimensional electrophoresis on 24 cm (pH 47) IPG strips and 12.5 % polyacrylamide gels. Cy3-labeled
heat-treated sample and Cy5-labeled internal standard are shown. The arrow indicates the position of the spot
displayed in b. (b). Resolution of a spot selected from the picking list. The spots of interest (see arrow in a) from
control (C) and heat-treated leaves (T) were excised from two different analytical DIGE gels and reelectrophoresed in a NuPage Tris acetate 7 % gel. Each spot resolved into two, S1 and S2. A 3D view of the
fluorescence signal is shown. (c). Manual quantification of the changes in S1 and S2 abundance. The NuPage
Tris acetate 7 % gel was scanned using the Typhoon 9400 and the fluorescence intensity volumes were estimated with the DeCyder Image Analysis Software. After normalization, it appeared that S2 accounted for the
observed signal reduction. To confirm fluorescent signal preservation, we summed the volumes of all 3D spots
for each condition and compared them to the corresponding spot volumes on 2D gel. These results show the
relative conservation of total fluorescence and global variation (average volume ratio: 1.87 for 2D gel and
1.67 for 3D gel). The minor difference can be explained by a loss of material due to the fact that: (a) only the
cylinder surrounding the top of the peak of 2D spot was picked; (b) after the 3D run, quantification does not
include small smears into the spot volume estimation. An Anova test with p-value < 0.05 was realized with SAS
8.0 (SAS Institute Inc., Cary, North Carolina, USA). Error bars represent SD values from three replicates
437
3. The gel fragments corresponding to the same spot position in

the 2D-DIGE gels (control and treated) are positioned next to
each other.
4. Add 10 L of the following mix: 2.5 L of 4 NuPage LDS
sample buffer, 1 L 1 M SDS, 6.5 L water.
5. Using the XCell SureLock Mini-Cell, run the gel first at 50 V
constant for 1 h and then follow the manufacturers instructions (see Note 6).
6. Remove the gel from the plastic cassette.
7. Acquire the gel image using the Typhoon 9400.
8. Use the Decyder software for a quantitative analysis of the gel
image (fluorescence volumetric reading). The data set can be
analyzed using standard statistical software. Figure 2 shows the
results of such an analysis.
3.3 MS Protein
Identification
1. Run a preparative 2D-DIGE gel as described above (see Note 7).

2. Use the data from the 2D-DIGE analytical gel analysis to
locate spots of interest in the preparative 2D-DIGE gel.
3. Generate a new picking list to command the mechanical spot
picker.
4. Treat the excised spots as described in Subheading 3.2.4 up to
step 6.
5. Attach the gel to a cellophane backing (see Note 8).
6. Acquire the gel image using the Typhoon 9400.
7. Generate a new picking list to command the mechanical spot
picker.
8. Proceed with the MS analysis protocol.
3.4 Western
Blot Analysis
Our protocol was initially used to detect low abundant proteins

such as SUMO target proteins (see Fig. 3).
1. Run a preparative 2D-DIGE as described above.
2. Scan the 2D-DIGE gel (Typhoon 9400).
3. Load the gel onto the electro-blotting transfer apparatus
and transfer proteins to a low-fluorescent (PVDF) membrane
according to the manufacturers literature.
4. Scan the PVDF membrane (Typhoon 9400).
5. Block the membrane with ECL blocking agent.
6. Probe the PVDF membrane with primary antibody against
SUMO1 and then with the appropriate secondary antibody
coupled to Cy5.
7. Scan the PVDF membrane.
438
Sergio Mauro et al.
Fig. 3 3D-DIGE-based detection and identification of low-abundance proteins. 2D Western blot analysis of
control leaves protein extracts was performed with a polyclonal antibody against SUMO1 protein (Primary
antibody, dilution 1/800; secondary antibody, dilution 1/2,000). Over 70 SUMO-protein conjugate spots were
detected. One of these spots (a) was excised and re-electrophoresed in a Bis-Tris Glycine 412 % gel.
It resolved into three 3D-components a, b and c (b). A Western blot analysis using the anti-SUMO1 polyclonal
antibody identified spot c as a SUMO-protein conjugate (c)
8. Use the 2D-DIGE gel and PVDF images to create a picking list.
9. Run a new preparative 2D-DIGE gel and excise the spots according to the picking list.
10. Fractionate the protein content of these gel fragments as in
Subheading 3.2.4.
11. Scan the gel to visualize the individual proteic components.
12. Perform a new Western blot (steps 36) to detect SUMO
conjugate.
13. For the MS identification of the SUMO targets duplicate the
gel (go to step 9).
Notes
1. Our protocol requires in-gel protein detection without fixation. Although several methods that fulfill that requirement
are available [18, 19], we described a DIGE-based procedure
which combines high-sensitivity protein detection and differential protein expression analysis.
Two procedures for protein labeling are available for DIGE
application. In the minimal labeling procedure two protein
samples and the internal standard are separately labeled with
the spectrally resolvable cyanine dyes Cy2, Cy3 or Cy5 and
mixed before electrophoretic fractionation. The minimal labeling fluorophors contain a N-hydroxysuccinimidyl ester reactive
439
group that reacts with the -amino groups of lysine residues in

proteins. The coupling conditions have been optimized to
ensure that only a single lysine on each protein is coupled to
the CyDye. In the protocol for Scarce Samples, only two dyes
Cy3 and Cy5 are supplied with a thiol maleimide group that
will react with every cystein residue in a protein.
2. The leaf proteins was extracted as described in [20] and resuspended in the rehydration buffer. The protein concentration
should be 0.55 mg/mL. A maximum of 120 L containing
300 g protein can be loaded in the loading sample cup.
3. If necessary the pH of the protein extract should be adjusted at
pH 8.0 with 50 mM NaOH or 50 mM HCl.
4. The exact volume of the TCEP solution should be optimized
for each experiment to ensure complete reduction of the
cysteinecysteine disulfide bonds in proteins.
As for the case of TCEP, the exact volume of the CyDye
DIGE Fluor solution should be optimized for each experiment.
5. If the strip was stored at 80 C, allow the strip to equilibrate
at room temperature.
6. The inclusion of SDS and the initial migration at low constant
voltage prevent protein smears.
7. Mixing aliquots of individual samples could be useful to balance the amount of up- and down-regulated proteins.
8. We strongly recommend testing for the presence of impurities
that could interfere with protein identification.
References
1. Tiselius A (1937) Electrophoresis of serum
globulin. Biochem J 31:313317
2. Shapiro AL, Viuela E, Maizel JV Jr (1967)
Molecular weight estimation of polypeptide
chains by electrophoresis in SDS-polyacrylamide
gels. Biochem Biophys Res Commun 28:
815820
3. Weber K, Osborn M (1969) The reliability of
molecular weight determinations by dodecyl
sulfate-polyacrylamide gel electrophoresis. J Biol
Chem 244:44064412
4. Banker GA, Cotman CW (1972) Measurement
of free electrophoretic mobility and retardation
coefficient of protein-sodium dodecyl sulfate
complexes by gel electrophoresis. A method
to validate molecular weight estimates. J Biol
Chem 247:58565861
5. Laemmli UK (1970) Cleavage of structural
6. OFarrell PH (1975) High resolution twodimensional electrophoresis of proteins. J Biol

Chem 250:40074021
7. Unlu M, Morgan ME, Minden JS (1997)
Difference gel electrophoresis: a single gel
method for detecting changes in protein
extracts. Electrophoresis 18:20712077
8. Wu WW, Wang G, Baek SJ, Shen RF (2006)
Comparative study of three proteomic quantitative methods, DIGE, cICAT and ITRAQ,
using 2D gel- or LC-MALDI TOF/TOF.
9. Stevenson SE, Chu Y, Ozias-Akins P, Thelen JJ
(2009) Validation of gel-free, label-free quantitative proteomics approaches: applications for seed
allergen profiling. J Proteomics 72:555566
10. Baggerman G, Vierstraete E, De Loof A,
Schoofs L (2005) Gel-based versus gel-free
proteomics: a review. Comb Chem High
Throughput Screen 8:669677
440
Sergio Mauro et al.
11. Jiang D, Jarret HW, Haskins WE (2009)

Methods for proteomic analysis of transcription factors. J Chromatogr A 41:68816889
12. Vanfleteren JR (1989) Sequential twodimensional and acetic acid/urea/triton X-l00
gel electrophoresis of proteins. Anal Biochem
177:388391
13. Matanabe T, Jin Y (2010) Analysis of E. coli
soluble proteins by non-denaturing micro
2-DE/3-DE and MALDI-MS-PMF. Electrophoresis 31:27402748
14. Colignon B, Raes M, Dieu M, Delaive E, Mauro
S (2013) Evaluation of three-dimensional gel
electrophoresis to improve quantitative profiling of complex proteomes. Proteomics 13:
20772082
15. Kang Y, Techanukul T, Mantalaris A, Nagy JM
(2009) Comparison of three commercially
available DIGE analysis software packages:
minimal user intervention in gel-based proteomics. J Proteome Res 8:10771084
16. Wu Y, Zhang L (2011) Comparison of two
academic software packages for analyzing
17.
18.
19.
20.
two-dimensional gel images. J Bioinform

Comput Biol 9:775794
Raman B, Cheung A, Marten MR (2002)
Quantitative comparison and evaluation of
two commercially available, two-dimensional
electrophoresis image analysis software packages, Z3 and Melanie. Electrophoresis 23:
21942202
Joo WA, Speicher DW (2007) Protein detection in gels without fixation, Chapter 10. Curr
Protoc Protein Sci. doi:10.1002/0471140864.
ps1006s48
Steinberg TH, Lauber WM, Berggren K,
Kemper C, Yue S, Patton WF (2000) Fluorescence detection of proteins in sodium
dodecyl sulfate-polyacrylamide gels using environmentally benign, nonfixative, saline solution. Electrophoresis 21:497508
Islam N, Lonsdale M, Upadhyaya NM, Higgins
TJ, Hirano H, Akhurst R (2004) Protein extraction from mature rice leaves for two-dimensional
gel electrophoresis and its application in proteome analysis. Proteomics 4:19031908
Chapter 31
A Bead-Based Multiplex Sandwich Immunoassay
to Assess the Abundance and Posttranslational
Modification State of -Catenin
Nicola Groll, Cornelia Sommersdorf, Thomas O. Joos, and Oliver Poetz
Abstract
A system-wide analysis of cell signaling involves detecting and quantifying a range of proteins and their
posttranslational modification states in the same cellular sample. We propose a protocol for a miniaturized,
bead-based array and describe its efficiency in characterizing the different forms and functions of -catenin.
The protocol provides detailed instructions for cell culture and bead array assays that enable insights into
complex networks at the systems level.
Key words Bead-based array, Sandwich immunoassay, co-immunoprecipitations, Proteinprotein
interaction assay, -catenin, Wnt-pathway
Introduction
Protein arrays have become powerful tools to investigate the status
of signaling pathways in cells and tissues. Bead-based multiplexed
assays can be performed with hundreds of samples, thereby enabling
time-resolved pathway activities of stimulated or perturbed cells.
The data can be used to infer the structure of underlying signaling
networks. Protein array technology is well-suited for such investigations because it can simultaneously measure many different proteins
as well as entails minimal amounts of material [14].
Over the past decade, bead-based arrays have become
well-established in cell signaling research (Fig. 1). This technology
platform has been used to analyze signaling networks in a timeresolved manner. The polystyrene beads have an internal fluorescent color code that facilitates the measurement of up to 500
features. The surface coating includes paramagnetic material and
carboxyl groups that allow easy conjugation of proteins, peptides,
oligonucleotides, and carbohydrates. For a panel of sandwich
immunoassays, the beads are usually coated with different capture
441
442
Nicola Groll et al.
Assay type
Protein-protein interaction
assay
Sandwich immunoassay
Ba
it
Analyte
Detector antibody
Detection system
anti-non pSer33Ser37Thr41
anti-pSer33Ser37Thr41
anti-pSer45
anti-pSer552
anti-pSer675
anti-pan -catenin
phosphorylated and total
-catenin
PE
Ecad
Bead
7-9
Bead
1-6
Capture antibody/
bait protein
PE
PE
Assay set-up
Coip
GST-TCF4
GST-ICAT
GST-ECT
transcriptionally active
free -catenin
Bead
10
ms E-cadherin
E-cadherin-bound
-catenin
anti -catenin (c-term specific)
anti-mouse-PE conjugate
Fig. 1 The -catenin bead array panel employs antibodies and known interaction partners of -catenin to
study the phosphorylation status and its complexation status within the same multiplex assay system.
Differentially phosphorylated -catenin is measured with bead array-based sandwich immunoassays. By utilizing different phosphorylation state and sequence-specific antibodies and a single pan-specific antibody
serving as the capture molecule, it is possible to reliably discriminate and detect various phosphorylated
-catenin forms. The recombinantly expressed binding partners of -catenin, ECT (E-cadherin cytosolic tail),
ICAT (inhibitor of -catenin), and TCF4 (T-cell factor 4) are used as bait proteins in a proteinprotein interaction
assay for binding and detecting transcriptionally active -catenin. The miniaturized co-immunoprecipitation
(Coip) allows -catenin in complex with E-cadherin to be detected by fishing E-cadherin. In all assay set-ups,
the same antibody specific for the -catenin C-terminus is used as a detector
antibodies. Following incubation with a protein extract from tissue

or cell culture, the captured analytes are detected and quantified by
means of detection antibodies in a sandwich mode.
Since protein activity in signaling networks is often reflected by
transient phosphorylation, it is important to not only quantify the
total amount of a signaling protein but also its various molecular
forms. Antibodies specific to posttranslational modified sequences
are used to capture analytes, making the bead-based array platform
an appropriate tool for simultaneous monitoring the various types
of phosphorylation of a single protein [4, 5]. In our assay, different
states and functions of -catenin are monitored under various
treatment conditions in a time-dependent manner. -Catenin
plays a key role in the canonical Wnt-signaling pathway and in
cell-to-cell adhesion complexes. Its cellular functions, whether
transcription factor or cell adhesion molecule, are orchestrated by
Assessing the Abundance and Posttranslational
443
changes in concentration, phosphorylation and complexation associated with other factors such as cadherins. After activation of the
Wnt receptor, a free pool of -catenin accumulates in the cytoplasm, translocates into the nucleus and then functions as a transcriptional co-activator [6, 7].
The Wnt/-catenin signaling pathway plays an important role
in the regeneration of adult tissues and in embryonic development
[8]. Being a proto-oncogene, -catenin is a key factor in carcinogenesis. Mutations in the -catenin gene or in other genes involved
in -catenin regulation, such as APC or Axin, lead to hyperactivation of the Wnt signaling pathway and are implicated in the genesis
of common cancers, such as colorectal cancer and hepatocellular
carcinoma [911]. Wnt/-catenin signaling also impacts the epithelial mesenchymal transition (EMT), which is reportedly involved
in the invasion of tissue and formation of metastasis [12].
Here, we propose a protocol using three different assay setupssandwich immunoassays, proteinprotein interaction and
co-immunoprecipitationsin one panel to assess and quantify
-catenin, its phosphorylation and functional mode in a single
sample. The sandwich immunoassay allows the measurement of
total -catenin as well as the degree of phosphorylation at six different sites. For this purpose, one pan-specific -catenin antibody,
one de-phosphorylation and four phosphorylation-specific antibodies served as capture molecules. Thus discrimination and detection of non-phosphorylated, differentially phosphorylated and
total -catenin is carried out.
In miniaturized GST-pull down (GST pull down) assays,
cellular interaction partners of -catenin are used as baits to detect
-catenin in its free non-complexed form. The -catenin which
forms a complex with these baits is transcriptionally active, as previously described [4].
A capture antibody toward E-cadherin allows co-immunoprecipitation of the E-cadherin/-catenin complex [13]. This membrane complex reflects the integrity of the cell. Finally, all the analytes
are detected by an antibody which binds to the C-terminal region of
-catenin and a phycoerythrin-conjugated secondary antibody. This
enables a readout of the various assays and allows the different pools
and states of -catenin to be monitored at the same time.
Materials
2.1 Cell Stimulation

and Sample
Preparation
1. Cell culture plates from commercial suppliers (e.g. Greiner Bio

One, Frickenhausen, Germany).
2. Cells of interest: e.g. Hep 70.4 (CLS, Eppenheim, Germany).
3. Cell culture medium: 110 % w/v fetal bovine serum (FBS),
100 U/ml penicillin-streptomycin in Dulbeccos modified
Eagles medium (DMEM).
444
Nicola Groll et al.
4. Dimethylsulfoxide (DMSO).
5. Compound of interest in DMSO, e.g. proteasome inhibitor
MG132 (Calbiochem).
6. Phosphate-buffered saline (PBS).
7. Lysis buffer: 150 mM NaCl, 1 % w/v Triton X-100,
Phosphatase Inhibitor Cocktail II (Sigma Aldrich), Phosphatase
Inhibitor Cocktail III (Sigma Aldrich), Complete Protease
Inhibitor Cocktail (Roche Applied Science), 50 mM TrisHCl,
pH 7.4, 2,5 U/ml Benzonase Nuclease (Novagen).
8. BCA (bicinchoninic acid) protein assay reagent.
2.2 Generation
of Bead Array
2.2.1 Covalent Coupling
of Capture Antibodies
to Microspheres
1. 1.5 ml microcentrifuge tubes (STARLAB GmbH).

2. Carboxylated MagPlex Microspheres (Luminex Corp., Austin,
TX, USA): local distributors are listed on the Luminex web
page (www.luminexcorp.com).
3. Magnetic separator, DynaMag (Life technologies).
4. Activation buffer: 100 mM sodium phosphate (Na2HPO4),
pH 6.2.
5. 50 mg/ml 1-ethyl-3-[3-dimethylaminopropyl] carbodiimide
hydrochloride (EDC) in Activation buffer.
6. 50 mg/ml N-Hydroxysulfosuccinimide sodium salt (sulfo-NHS)
in DMSO.
7. Rotator Revolver Mixer for reaction vials and sample tubes
(Carl ROTH).
8. Coupling buffer: 50 mM 2-(N-morpholino) ethanesulfonic
acid (MES), pH 5.0.
9. Washing buffer: 0.1 % w/v Tween20 in phosphate-buffered
saline (PBS).
10. Storage buffer: 0.05 % w/v sodium azide in Roche Blocking
Reagent for ELISA (Roche Applied Science).
List of covalently coupled capture antibodies
1. Anti--catenin antibody (R&D Systems).
2. Anti-mouse E-cadherin antibody (R&D Systems).
3. Anti-glutathione S-transferase (GST) antibody (GE Healthcare).
4. Anti-rabbit IgG antibody (Jackson Immunoresearch).
2.2.2 Non-covalent
Coupling of Capture
Antibodies and Bait
Proteins to Microspheres
List of non-covalently coupled capture antibodies

1. Anti--catenin phosphorylated at Ser33/Ser37/Thr41 antibody
(Cell Signaling Technology).
2. Anti--catenin phosphorylated at Ser45 antibody (Cell
Signaling Technology).
445

5. Anti--catenin non-phosphorylated at Ser33/37/Thr41 antibody (Cell Signaling Technology).
List of non-covalently coupled bait proteins
1. Glutathione S-transferase E-cadherin cytosolic tail (GST-ECT).
2. Glutathione S-transferase T cell factor 4 (GST-TCF4).
3. Glutathione S-transferase -catenin-interacting protein 1
(GST-ICAT).
2.3 Sandwich
Immunoassay, CoImmunoprecipitation,
GST-Pull Down,
and Detection
1. KingFisher
Flex
(ThermoScientific).
Magnetic
Particle
Processors
2. KingFisher PCR-magnet head (ThermoScientific).

3. KingFisher Flex 96 tip comb for PCR magnet (ThermoScientific).
4. Plateshaker, e.g. Thermomixer comfort with microtitre plate
thermoblock (Eppendorf GmbH).
5. 96-well PCR plate.
6. Washing buffer: 0.1 % w/v Tween20 in phosphate-buffered
saline (PBS).
7. Assay buffer: 0.1 % w/v Tween20 Roche Blocking Reagent for
ELISA (Roche Applied Science).
8. Anti--catenin antibody (#610154, BD Biosciences).
9. Anti-mouse immunoglobulin G (IgG) antibody conjugated to
phycoerythrin (PE) (Jackson Immunoresearch).
10. Donkey serum and goat serum (Sigma-Aldrich).
11. Luminex FlexMap3D reader (alternatively LX100, LX200, or
MagPix): local distributors are listed on the Luminex website
(www.luminexcorp.com).
Methods
3.1 Cell Stimulation

and Sample
Preparation
1. Seed cells at the appropriate density in a 96-well tissue culture

plate (e.g. 1 104 Hep 70.4 in 100 l) and cultivate the cells
for 24 h or 70 % confluence (see Note 1).
3.1.1 Using 96-Well

Format Cell Culture
2. Replace the cultivation medium with medium lacking FBS and

starve cells for another 1216 h.
3. Reconstitute your chemical compound of interest (e.g. MG132,
proteasome inhibitor) in DMSO at a concentration of 1 mM.
4. Dilute the inhibitor stock to concentration [1 M] in 1 ml
tissue culture medium lacking FBS. For mock control, prepare
446
Nicola Groll et al.
serum-free tissue culture medium lacking inhibitors, but

including an equivalent volume of DMSO.
5. Incubate cells at 37 C and 5 % CO2 for 5 min24 h.
6. Prepare 100 l lysis buffer per cell culture well. Add protease
and phosphatase inhibitors and benzonase to the lysis buffer
immediately before use. Keep lysis buffer on ice (see Note 2).
7. Remove the medium using a multichannel pipette. Wash cells
twice using ice-cold PBS.
8. Remove all liquid and add 100 l ice-cold lysis buffer.
9. Incubate for 30 min at 4 C with gentle agitation on a rocking
platform.
10. Dislodge and solubilize all adherent cells by repeated pipetting.
Extracts should be clear and non-viscous.
11. Determine protein concentration using the BCA method.
12. Either proceed immediately with the bead-based sandwich
immunoassay protocol or store plates at 70 C until further use.
3.1.2 Using a 10 cm Cell
Culture Dish
1. Seed cells at the appropriate density in a 10 cm cell culture dish

(e g. 3 105 HT-29 in 10 ml) and cultivate the cells for 35
days to a confluence of 70 %.
2. Replace the growth medium with medium lacking FBS and
starve cells for another 1216 h.
3. Reconstitute your chemical compound of interest (e.g.
MG132, proteasome inhibitor) in DMSO at a concentration
of 1 mM.
4. Dilute the inhibitor stock to concentration of 1 M in 1 ml
tissue culture medium lacking FBS. For mock control, prepare
serum-free tissue culture medium lacking inhibitors, but
including an equivalent volume of DMSO.
5. Incubate cells at 37 C and 5 % CO2 over a consecutive period
of 5 min24 h.
6. Harvest all cells at the same time-point.
7. Prepare 1,000 l lysis buffer per cell culture well. Add protease
and phosphatase inhibitors and benzonase to the lysis buffer
immediately before use. Keep lysis buffer on ice.
8. Remove the medium and wash cells twice using ice-cold PBS.
9. Scrape adherent cells from the plate in 1 ml ice-cold PBS with
a rubber policeman.
10. Solubilize cells by repeated pipetting and transfer 1 ml in a
1.5 ml microcentrifuge tube.
11. Centrifuge at 5,000 g, 15 min and 4 C.
447
12. Remove supernatant and re-suspend the pellet in 1 ml lysis

buffer.
13. Incubate for 30 min at 4 C while gently mixing and tapping
several times.
14. Centrifuge at 10,000 rcf, 30 min and 4 C.
15. Transfer supernatant into a fresh tube.
16. Determine protein concentration using the BCA method.
17. Either proceed immediately with the bead-based sandwich
immunoassay protocol or store aliquots at 70 C until further use.
3.2 Generation
of Bead Array
3.2.1 Covalent Coupling

of Capture Antibodies
to Microspheres
To conjugate proteins directly on the magnetic beads, we used

EDC/NHS to generate succinimidyl ester deriving from the carboxyl functions of the microsphere surface. The chemical reaction
with protein amino groups results in the formation of amide bonds.
Buffer agents containing primary or secondary amino groups like
glycine or stabilizing agents like BSA should be avoided since
they compete with the antibodies for the activated carboxyl groups.
If the buffer formulation contains components carrying amino
groups, site-directed non-covalent strategies can be applied, such as
epitope tag or species-specific antibodies to immobilize recombinant
proteins and antibodies in a site-specific manner (see Notes 35).
In our study, we used GST-specific antibodies to immobilize
recombinant fusion proteins and rabbit-specific antibodies to immobilize the phosphorylation- and sequence-specific antibodies.
1. Sonicate selected beads stock and transfer 200 l (2.5 106
beads) of each bead stock solution (1.25 107 beads/ml) to
1.5 ml reaction tubes (polypropylene, clear).
2. Use a magnetic separator to collect the beads (see Note 6).
3. Remove the supernatant with a pipette and wash beads twice
with 200 l of activation buffer. Collect the beads by means of
the magnetic separator in between washes.
4. Re-suspend beads in 80 l of activation buffer.
5. Prepare NHS and EDC solutions immediately before use. Add
10 l of each solution to the bead suspension and incubate for
20 min at room temperature, in the dark, under rotation to
activate the beads.
6. Use the magnetic separator to collect the beads, discard the
activation solution.
7. Wash the beads three times with 500 l of coupling buffer.
Collect the beads by using the magnetic separator between
washes (see Note 6).
8. Dilute the antibodies to be immobilized with coupling buffer
to the following concentrations: anti--catenin and anti-E-
448
Nicola Groll et al.
cadherin antibodies to 80 g/ml; anti-rabbit IgG antibody to

250 g/ml.
9. Add 250 l of the diluted protein solution to the beads and
incubate at room temperature, in the dark, for at least 2 h
under rotation.
10. Place the tubes in a magnetic separator to collect the beads.
Remove the supernatant and discard.
11. Wash the beads three times with 500 l of washing buffer.
Collect the beads by using the magnetic separator between
washes (see Note 6).
12. Re-suspend the coupled beads in 100 l storage buffer.
13. Store beads in the dark at 4 C.
The bead recovery should be determined as follows:
1. Dilute the covalently coupled beads 1:500 in the assay buffer.
2. Transfer 100 l/well of diluted beads to a microtiter plate and
incubate the diluted beads for 30 min in a 96-well plate shaker,
rotating at 750 rpm at room temperature.
3. Measure the numbers of beads with a Luminex reader using
the following settings:
Sample size
50 l
Time-out
80 s
Total beads
10,000
Calculate the bead concentration as follows:

Beads per l = (number of beads/30 l) 500
3.2.2 Non-covalent
Coupling of Capture
Antibodies and Bait
Proteins to Microspheres
If it is not possible to couple proteins directly to the carboxy-modified

beads, due to unfavorable additives such as glycine, sodium azide,
BSA, carrier proteins or low protein concentration, it is recommended
to use a non-covalent immobilization strategy. In this case, it is advisable to use tag- or species-specific antibody-conjugated beads. Here,
we used this strategy to immobilize antibodies against phosphorylated
-catenin or GST-tagged bait proteins. To ensure the removal of all
unbound bait molecules that could possibly interfere with the sample
or other assay components, several washing steps are necessary.
1. Dilute GST-tagged bait protein (2.5 g of each protein) (see
Note 7) with 50 l PBS in separate 1.5 ml tubes (polypropylene, clear). Incubate in the presence of 105 beads coupled to
the antibody that recognizes GST for 6 h at room temperature,
under rotation.
2. Dilute phosphorylation-specific -catenin antibodies (250 ng)
with 50 l PBS in separate 1.5 ml tubes (polypropylene, clear).
449
Incubate in the presence of 105 beads coupled to the antibody

rabbit IgG in 50 l PBS for 4 h at room temperature, under
rotation.
3. Place the tubes in a magnetic separator to collect the beads.
Remove the supernatant and discard it.
4. Wash each bead population three times with 200 l PBS by
using a magnetic separator (see Note 6).
5. Re-suspend each bead population in 50 l bead assay buffer.
6. The beads cannot be stored and should be used directly in
the assay.
3.3 Sandwich
Immunoassay,
CoImmunoprecipitation,
GST-Pull Down,
and Detection
Before carrying out the assay procedure, the protein extracts need
to be normalized to protein concentration. Due to the cell type
and the quantity of the target protein (in this case: -catenin)
being proportional to total protein, it is advisable to determine
the optimal amounts in separate, preliminary experiments. We
suggest that a dilution series of cell culture lysates is made to identify a suitable range of lysate amounts for detecting the target
proteins. We recommend using 1025 g of total protein, as this
is usually appropriate for analyses (see Note 8).
Here we describe how the assay can be performed by using a
magnetic bead transfer system ([14] King Fisher, Thermo Fisher
Scientific) which enables semi-automated washing and incubation
steps. Of course, it is also possible to use a magnetic plate separator
for manual washing and incubation in a 96-well plate.
1. Prepare a bead-master mix by diluting each population of
coated beads in assay buffer in order to ensure that there are
2,000 beads for each population in 20 l of fully mixed bead
solution. Vortex the diluted beads to mix them.
2. Choose an appropriate protein amount of each sample, and
adjust each sample to the same protein concentration with
assay buffer at a volume of 40 l. Keep the samples on ice.
3. Combine 20 l bead mix with 40 l of diluted cell sample in
one well of the 96-well microplate (half-area, flat-bottomed
microplate, with a non-binding surface, is subsequently
referred to as the assay plate).
4. Use 40 l assay buffer as a reagent control sample.
5. Seal the plate with a microplate sealing film.
6. Incubate the samples containing the beads overnight in a
96-well plate shaker at 4 C and 750 rpm.
7. Prepare the secondary antibody solution at least 2 h before
assay processing to minimize the background signal. Goat and
donkey sera can be used to minimize the cross-reactivity of the
polyclonal antibodies. Dilute the PE-conjugated donkey antimouse IgG in assay buffer to a final concentration of 2.5 g/ml.
450
Nicola Groll et al.
Add donkey and goat sera to a final concentration of 1 % (v/v)

and incubate this solution for 2 h at 4 C in the dark.
8. Block 8 PCR plates with 100 l assay buffer per well for 10 min.
9. Remove buffer of the first plate and transfer the bead sample
mix from the assay plate to the sample plate.
10. Dilute the mouse anti--catenin antibody in assay buffer to a
final concentration of 1 g/ml. Remove buffer of the second
plate and add 50 l detection antibody solution per well
(detection antibody plate).
11. Remove buffer of the third plate and add 50 l secondary antibody per well (secondary antibody plate).
12. Add 100 l washing buffer per well in five remaining PCR
plates (four washing plates, one elution plate).
13. Assay processing is done by means of a semi-automated
platform (KingFisher 96 Thermo Fisher Scientific) with magnetic rods that move the sample particles during the various
steps. We used the associated KingFisher software (BindIt
Software 3.1 for KingFisher Instruments) to create a protocol
for magnetic particle-based methods (see Table 1 and Fig. 2 for
typical results).
14. Place all eight plates in the robotic system.
15. Place the elution plate containing the re-suspended beads in
the Luminex reader, using the following standard settings:
Doublet discriminator
7,50015,000
Sample volume
100 l
Minimal bead count per region
100
Table 1
Program of the semiautomatic, magnetic bead transfer system (King Fisher, Thermo Fisher Scientific)
Step
Plate
Assay step
Duration
Sample plate
Bead collection
10 s
Washing plate 1
Washing
1 min
Washing plate 2
Washing
1 min
Detection antibody plate
Incubation with detection antibody
60 min
Washing plate 3
Washing
1 min
Washing plate 4
Washing
1 min
Secondary antibody plate
Incubation with secondary antibody
45 min
Washing plate 2
Washing
1 min
Washing plate 4
Washing
1 min
10
Elution plate
Bead elution
20 s
1200
anti-total -catenin
1200
451
anti- pSer33/37/Thr41 -catenin
anti-nonpSer33/37/Thr41 -catenin
4500
MFI [AU]
MFI [AU]
600
600
0
0.1
10
3000
1500
300
300
PE
MFI [AU]
900
900
0.1
Time [h]
1
Time [h]
10
anti-pSer552 b-catenin
Bead
1-6
2000
0.1
1
Time [h]
anti-pSer675 -catenin
10
anti-pSer45
2000
450
1000
8000
300
150
500
MFI [AU]
1500
MFI [AU]
MFI [AU]
1500
0
0.1
1
Time [h]
10
500
0
0.1
1
Time [h]
0.1
10
60 GST-TCF4
GST-ICAT
1000
40
1
Time [h]
10
1
Time [h]
10
GST-ECT
N
PE
4000
40
MFI [AU]
Bead
7-9
MFI [AU]
30
6000
MFI [AU]
Ba
it
20
20
10
2000
0.1
1
time [h]
10
anti-ms E-cadherin
MFI [AU]
1
time [h]
10
0.1
20
PE
Bead
10
0
0.1
DMSO
MG132
Ecad
10
0.1
1
Time [h]
10
Fig. 2 Snapshots of different forms of -catenin monitored by a bead array assay. Hep 70.4 cells were treated
with the proteasome inhibitor MG132 and DMSO as a control, in intervals of 5, 30, 60, 180, 360, 720, and
1,440 min. The experiment was performed in triplicate. From each sample, a protein extract of 20 g was
analyzed by the -catenin bead array assay panel
Notes
1. Growth rate depends considerably on cell line and incubation
parameters.
2. The purpose of the assay described here is to examine a
molecular snapshot of a single protein. The conditions were
optimized, in particular for extracting the native -catenin and
its diverse forms as well as the intact E-cadherin complexes.
Generally, the extraction conditions, for instance, those including additives of the lysis buffer, can influence the native protein. It is thus necessary to optimize the appropriate detergent,
salt concentration and inhibitors if other proteins or pathways
shall be analyzed.
452
Nicola Groll et al.
3. It is important to ensure that reagents such as glycine, BSA,

etc. are not present in the capture antibody formulation as they
prevent or reduce capture antibody coupling efficiency.
4. In a bead-based sandwich immunoassay system, only a limited
number of proteins can be analyzed simultaneously (typically
<25). Due to cross-reactions, the increasing number of detection antibodies in the system usually results in higher background signals. It is therefore recommended that bead-based
assays are used for focused analysis with up to 25 analytes.
One of the most restricting factors is availability of antibodies required. Two antibodies are usually necessary for every
analyte: one for capture, the other for detection. They must
bind to two different epitopes of the targeted analyte.
5. As a rule, the assays are designed to analyze the native form of
proteins. Therefore, antibodies must be chosen according to
their ability to recognize native proteins.
6. Do not allow beads to dry. Vortex beads thoroughly during
each washing or mixing step. To ensure that all the beads are
collected, use the collector for 30 s; check whether the beads
are deposed on the wall and whether the liquid is clear.
7. GST-ECT, GST-TCF4, and GST-ICAT were expressed in
Escherichia coli and affinity-purified as described earlier [4].
8. Wide differences in results can occur, depending on the cell line
models. Here, the targeted protein -catenin can be mutated
and/or truncated. For example, HepG2 cells express a mutant
form of -catenin, which is transcriptionally active, because the
regulatory N-terminal site is deleted. High amounts of this
active form can be detected via the proteinprotein interaction
proposed here. Therefore it is advisable to use a positive control, to ensure that the assay functions accurately and that assay
results can be properly compared. We recommend that extracts
should only be taken from cell lines in which the pathway or the
targeted protein is known to be present and can be easily activated, e.g. HEK 293 [4]. An additional advantage of positive
controls is that they are helpful in identifying technical problems, e.g. batch-to-batch variations of reagents. For the same
reason, we recommend blank controls for detecting batch-tobatch variations of reagents.
References
1. Du J, Bernasconi P, Clauser KR, Mani DR,
Finn SP, Beroukhim R, Burns M, Julian B,
Peng XP, Hieronymus H, Maglathlin RL,
Lewis TA, Liau LM, Nghiemphu P,
Mellinghoff IK, Louis DN, Loda M, Carr
SA, Kung AL, Golub TR (2009) Bead-based
profiling of tyrosine kinase phosphorylation

identifies SRC as a potential target for glioblastoma therapy. Nat Biotechnol 27:
7783
2. Jones RB, Gordus A, Krall JA, MacBeath G
(2006) A quantitative protein interaction
3.
4.
5.
6.
7.
8.
network for the ErbB receptors using protein

microarrays. Nature 439:168174
Khan IH, Mendoza S, Rhyne P, Ziman M,
Tuscano J, Eisinger D, Kung HJ, Luciw PA
(2006) Multiplex analysis of intracellular signaling pathways in lymphoid cells by microbead suspension arrays. Mol Cell Proteomics
5:758768
Luckert K, Gotschel F, Sorger PK, Hecht A,
Joos TO, Potz O (2011) Snapshots of protein
dynamics and post-translational modifications
in one experiment{beta}-catenin and its functions. Mol Cell Proteomics 10(M110):007377
Luckert K, Gujral TS, Chan M, Joos TO,
Sorger PK, Macbeath G, Potz O (2012) A dual
array-based approach to assess the abundance
and posttranslational modification state of signaling proteins. Sci Signal 5:pl1
Gordon MD, Nusse R (2006) Wnt signaling:
multiple pathways, multiple receptors, and
multiple transcription factors. J Biol Chem
281:2242922433
Willert K, Jones KA (2006) Wnt signaling: is the
party in the nucleus? Genes Dev 20:13941404
Nusse R (1997) A versatile transcriptional effector of Wingless signaling. Cell 89:321323
453
9. Morin PJ, Sparks AB, Korinek V, Barker N,

Clevers H, Vogelstein B, Kinzler KW (1997)
Activation of beta-catenin-Tcf signaling in
colon cancer by mutations in beta-catenin or
APC. Science 275:17871790
10. de La Coste A, Romagnolo B, Billuart P,
Renard CA, Buendia MA, Soubrane O, Fabre
M, Chelly J, Beldjord C, Kahn A, Perret C
(1998) Somatic mutations of the beta-catenin
gene are frequent in mouse and human hepatocellular carcinomas. Proc Natl Acad Sci U S A
95:88478851
11. Polakis P (2000) Wnt signaling and cancer.
Genes Dev 14:18371851
12. Heuberger J, Birchmeier W (2010) Interplay
of cadherin-mediated cell adhesion and canonical Wnt signaling. Cold Spring Harb Perspect
Biol 2:a002915
13. Poetz O, Luckert K, Herget T, Joos TO (2009)
Microsphere-based co-immunoprecipitation in
multiplex. Anal Biochem 395(2):244248
14. Poetz O, Henzler T, Hartmann M, Kazmaier
C, Templin MF, Herget T, Joos TO (2010)
Sequential multiplex analyte capturing for
phosphoprotein profiling. Mol Cell Proteomics
9:24742481
Chapter 32
Identification of SUMO E3 Ligase-Specific Substrates
Using the HuProt Human Proteome Microarray
Eric Cox, Ijeoma Uzoma, Catherine Guzzo, Jun Seop Jeong,
Michael Matunis, Seth Blackshaw, and Heng Zhu
Abstract
The functional protein microarray is a powerful and versatile systems biology and proteomics tool that
allows the rapid activity profiling of thousands of proteins in parallel. We have recently developed a human
proteome array, the HuProt array, which includes ~80 % of all the full-length proteins of the human proteome. In one recent application of the HuProt array, we identified numerous SUMO E3 ligase-dependent
SUMOylation substrates. For many SUMO E3 ligases, only a small number of substrates have been identified and the target specificities of these ligases therefore remain poorly defined. In this protocol, we outline
a method we developed using the HuProt array to screen the human proteome to identify novel SUMO
E3 ligase substrates recognized by specific E3 ligases.
Key words Microarray, SUMO, Proteomics, E3 ligases, Posttranslational modifications
Introduction
The functional protein microarray is a powerful and versatile
systems biology and proteomics tool that allows the rapid activity
profiling of thousands of proteins in parallel. Applications of functional protein microarrays range from the identification of proteinbinding properties, to surveying targets of posttranslational
modifications, to uncovering novel enzymatic activities. Since the
development of the yeast proteome microarray over 10 years ago
[1], more recent work has seen the development of complete and
near-complete proteome arrays representing viruses, bacteria, and
plants [24]. However, most existing human protein microarrays
are comprised of only a minority of the human proteome [59].
We have recently developed a human proteome microarray, the
HuProt array, which includes nearly 20,000 full-length human
proteins [10]. The proteins used to generate this microarray
were expressed in yeast and purified under native conditions.
455
456
Eric Cox et al.
Expressing recombinant eukaryotic proteins in yeast improves the

likelihood that proteins will retain their biological activity relative
to prokaryotic and in vitro expression systems.
Numerous collaborations between our labs and others have so
far harnessed the power of the HuProt array to profile a wide range
of protein activities. The role of posttranslational modifications in
regulating enzymatic activity is one area of investigation particularly well suited for the HuProt array platform. A screen to define
the S-nitrosylated proteome revealed an important regulatory role
for this posttranslational modification in the control of ubiquitin
E3 ligase activity [11]. In other work, phosphorylation and glycosylation states of the protein kinase CK2 were shown to affect its
substrate specificity [12]. The HuProt array has also been used in
two separate studies to link novel proteinRNA interactions to
neurological disease, including an interaction between RNA splicing factors and a long noncoding RNA linked to schizophrenia
[13] and an interaction between multiple RNA binding proteins
and an expanded repeat-containing transcript implicated in amyotrophic lateral sclerosis [14]. Another ongoing project in our labs
is the generation of monospecific monoclonal antibodies whose
specificity can be quickly evaluated using the HuProt array [10].
The utility of the HuProt array further extends to exciting clinical
applications including the identification of novel biomarkers that
may be used as a diagnostic tool in primary biliary cirrhosis, an
autoimmune disease of the liver [15].
Protein SUMOylation is an essential posttranslational modification in most organisms, including yeast, C. elegans, Arabidopsis,
and mice [16]. The reversible SUMO-modification of target proteins involves an enzymatic cascade chemically similar to ubiquitylation, involving E1 activating, E2 conjugating, E3 ligating
enzymes and SUMO proteases. As the only classes of SUMOylation
enzymes for which multiple members have been identified, the
SUMO E3 ligases and the SUMO proteases have been proposed
to be the major factors determining substrate specificity. Recently,
we have conducted SUMOylation assays using the HuProt microarray to identify numerous previously uncharacterized SUMO E3
ligase-dependent substrates using a subset of human SUMO E3
ligases. While our study focused on some of the best characterized
SUMO E3 ligases, recently additional SUMO E3 ligases have been
described [1721] and it is likely that new SUMO E3 ligases await
discovery [22]. The methods that we describe here could be used
to identify substrates for these additional SUMO E3 ligases. For
most SUMO E3 ligases, only a limited number of substrates are
known. In this chapter, we will describe the on-chip SUMOylation
protocol that we have developed so that the reader may conduct
SUMOylation assays using the HuProt microarray with their
SUMO E3 ligase of interest.
Identifying SUMO Substrates Using the HuProt Microarray
2
2.1
457
Materials
Equipment
1. HuProt human proteome microarray (CDI Laboratories,

USA).
2. Bench-top centrifuge.
3. Four-well plate.
4. Humidity chamber (see Note 1).
5. Laboratory tissues (Kimwipes).
6. LifterSlip Coverslips for Microarray Slides (Thermo Scientific,
USA).
7. Microscope slide box (25-slide size).
8. Microarray analysis software, GenePix Pro 6.0 (MDS Analytical
Technologies, USA).
9. Orbital shaker.
10. GenePix 4000B Microarray Scanner (Molecular Devices,
USA).
2.2 Purification
of the SUMO
E1 Enzyme
1. E1 Binding Buffer: 20 mM TrisHCl, pH 8.0, 350 mM NaCl,

1 mM beta-mercaptoethanol, 10 mM imidazole.
2. E1 Wash Buffer: 20 mM TrisHCl, pH 8.0, 350 mM NaCl,
3. E1 Elution Buffer: 20 mM TrisHCl, pH 8.0, 350 mM NaCl,
2.3 Purification
of the SUMO
E2 Enzyme
1. E2 wash buffer: 1 mM DTT in PBS.
2.4 Common
Reagents
for Expression
and Purification
of SUMO Protein
and SUMOylation
Enzymes
1. General lysis buffer: 1 mg/ml lysozyme, 2 mM DTT, 1

Roche protease inhibitor cocktail (EDTA-free), 10 U/ml benzonase dissolved in PBS.
2. Enzyme dialysis buffer: 20 mM TrisHCl, pH 8.0, 75 mM
NaCl, 1 mM beta-mercaptoethanol.
3. PreScission protease (GE Healthcare).
4. 100 mM IPTG.
5. Glutathione sepharose (GE Healthcare).
6. Ni-NTA agarose (Life Technologies).
2.5 SUMO Antibody

Labeling
1. SUMO-1 affinity-purified mouse monoclonal antibody (21C7)

(#33-2400, Life Technologies).
2. DyLight 549 Antibody Labeling Kit (Pierce Biotechnology):
DyLight NHS ester, 0.67 M borate buffer, purification resin,
spin columns, microcentrifuge collection tubes.
458
Eric Cox et al.
2.6 On-Chip
SUMOylation Assay
1. SUMO blocking buffer: 2 % BSA, 0.05 % Tween-20 in TBS.

2. 2 SUMO conjugation buffer: 40 mM HEPES, pH 7.3,
200 mM NaCl, 20 mM MgCl2, 0.2 mM DTT.
3. SUMO reaction mix (for 200 l reaction): 100 l 2 SUMO
conjugation buffer, 0.252.5 M E1 enzyme, 0.077 M E2
enzyme, 20 nM E3 ligase (see Note 2).
4. TBST: 0.05 % Tween-20 in TBS.
5. 1 % SDS.
Methods
3.1 Purification
of SUMO Protein
1. Streak out colonies onto a LB + ampicillin agar plate from a

glycerol stock of GST-SUMO in pGEX6p.1 in BL21 cells.
2. Pick a single colony and inoculate into 5 ml LB with ampicillin
overnight at 37 C.
3. Dilute the 5 ml culture into 50 ml LB with ampicillin culture
overnight at 37 C.
4. Dilute 50 ml culture into 1 l LB with ampicillin until the
OD = 0.6, then drop the temperature to 20 C and induce with
1 mM IPTG and shake overnight at 20 C.
5. Freeze pellet at 80 C until ready to proceed with purification.
6. Thaw pellet at 37 C, resuspend with 25 ml general lysis buffer, rotate at RT for 15 min.
7. Centrifuge at 4 C for 30 min at 20,000 g to pellet insoluble
material.
8. During centrifugation prepare glutathione sepharose by washing 2 ml 50 % glutathione sepharose with 1 PBS in 50 ml
tube. Repeat twice.
9. Bind protein by mixing supernatant with glutathione sepharose, rotate at 4 C for 1 h or longer.
10. Wash sepharose by spinning down sepharose, discard supernatant, resuspend sepharose in 1 ml of 1 PBS. Transfer sepharose to column; wash with 12 ml of 1 PBS.
11. Cleave SUMO from GST tag by transferring sepharose bound
with GST-precission protease and GST-SUMO to the same
column. Parafilm column to prevent leakage. Incubate with
shaking at 4 C overnight.
12. Allow sepharose to settle. Remove supernatant containing
purified untagged SUMO. Apply 4 ml of 1 PBS and repeat.
13. Dialyze overnight against 2 l 1 PBS
14. Concentrate protein using a micron centrifugal filter (10 kDa
MWCO) by spinning 20 min at 500 g. Aliquot and freeze
with liquid nitrogen and store at 80 C.
3.2 Purification
of SUMO E1 Enzyme
459
1. Streak out colonies onto an LB + ampicillin agar plate from a

glycerol stock of His-hE1 (His-Aos1/Uba2) in BL21 cells.
overnight at 37 C.
overnight at 37 C.
5. Freeze pellet at 80 C until ready to proceed with
purification.
material.
8. During centrifugation prepare Ni-NTA agarose by washing
2 ml 50 % Ni-NTA agarose with E1 binding buffer in 50 ml
tube. Repeat twice.
9. Bind protein by mixing supernatant with Ni-NTA agarose,
rotate at 4 C for 1 h or longer.
10. Wash agarose by spinning down agarose, discard supernatant,
resuspend agarose in 1 ml of wash buffer. Transfer agarose to
column; wash with 12 ml of E1 wash buffer.
11. Elute with 3 ml of E1 elution buffer. Collect fractions and
analyze by SDS-PAGE. Pool fractions containing protein.
12. Dialyze overnight against 2 l enzyme dialysis buffer.
3.3 Purification
of SUMO E2 Enzyme
1. Streak out colonies onto an LB + ampicillin agar plate from a

glycerol stock of GST-Ubc9 in pGEX6p.1 in BL21 cells.
overnight at 37 C.
overnight at 37 C.
5. Freeze pellet at 80 C until ready to proceed with purification.
material.
460
Eric Cox et al.
8. During centrifugation prepare glutathione sepharose by washing 2 ml 50 % glutathione sepharose with 1 PBS in 50 ml
tube. Repeat twice.
9. Bind protein by mixing supernatant with glutathione sepharose, rotate at 4 C for 1 h or longer.
10. Wash sepharose by spinning down sepharose, discard supernatant, resuspend agarose in 1 ml of E2 wash buffer. Transfer
sepharose to column; wash with 12 ml of E2 wash buffer.
11. Cleave UBC9 from GST tag by transferring beads bound with
GST-precission protease and GST-UBC9 to the same column.
Parafilm column to prevent leakage. Incubate with shaking at
4 C overnight.
12. Allow sepharose to settle. Remove supernatant containing
purified untagged UBC9. Apply 4 ml of E2 wash buffer and
repeat.
13. Dialyze overnight against 2 l enzyme dialysis buffer.
3.4 SUMO Antibody
Labeling
1. Add 40 l 0.67 M borate buffer to 0.5 ml of 0.5 mg/ml

affinity-purified SUMO-1 antibody.
2. Add 0.5 ml of antibody in borate buffer to the vial of DyLight
reagent and vortex gently.
3. Briefly centrifuge to collect the sample in the bottom of the
tube.
4. Incubate the reaction mixture for 60 min at room temperature
protected from light.
5. Mix purification resin to ensure uniform suspension and add
400 l of the suspension into both spin columns. Centrifuge
for 45 s at 1,000 g to remove the storage solution. Discard
the used collection tubes and place the columns in new collection tubes.
6. Add 250270 l of the labeling reaction to each spin column
and mix the sample with the resin by vortexing.
7. Centrifuge columns for 45 s at 1,000 g to collect the purified
proteins. Combine the samples from both columns (0.5 ml
total).
8. Aliquot and store the labeled antibody at 20 C.
3.5 On-Chip
SUMOylation Assay
1. Remove arrays from 80 C freezer and immediately rinse by

quickly dunking in a beaker of 300 ml of TBST.
2. Using forceps, transfer each array to a well of a four-well plate
with 3 ml of SUMO blocking buffer per well.
461
3. Block protein microarray by gently shaking overnight at 4 C

(see Note 3).
4. Prepare the reaction mix and keep on ice. Add E1 and E2
enzymes and E3 ligase immediately before the end of the
blocking step. Set up two control experiments, one using antibody only, and one without the E3 ligase.
5. Remove one array from blocking buffer and carefully wick off
liquid by tapping the edge on a paper towel, give the bottom of
the array a quick wipe with a kimwipe and place array on the top
surface (protein side up) of a humidity chamber (see Note 4).
6. Add SUMO reaction mix to each slide carefully and place lifterslip on top, being careful to avoid bubbles (see Note 5).
7. Incubate at 37 C for 90 min (depending on enzyme activity).
8. Immediately after start of incubation, pre-warm appropriate volume of 1 % SDS to 55 C for later washing steps (see Note 6).
9. Remove coverslip by gently sliding off array (see Note 7).
10. Place arrays in a four-well plate and wash gently on orbital shaker
3 for 10 min at room temperature with 3 ml TBST per well.
11. Wash with 1 % SDS warmed to 55 C 3 for 5 min each.
12. Wash once with 3 ml TBST.
13. Dilute labeled SUMO1 antibody in blocking buffer at 1:1,000
dilution. Apply 200 l of the antibody mixture to each array
and add fresh coverslip. Incubate array with labeled SUMO1
antibody for 1 h at room temperature.
14. Remove coverslip by gently sliding off array and wash slides
three times for 10 min in 3 ml TBST in four-well plate.
15. Wash the slides once in 50100 ml milliQ water for 5 min to
remove residual salts from the surface of the microarray
(see Note 8).
16. Place each array horizontally into a micro slide box with kimwipes on the bottom. Centrifuge the box in a benchtop centrifuge for 3 min at 500 g.
17. Scan the microarray with a GenePix 4000B scanner with 5-m
resolution detection at 532 nm with the appropriate gain and
power settings (see Note 9). Save the scanned images as TIFF
files. All slides that will be compared should be scanned using
the same gain and power settings.
Notes
1. The humidity chamber is made by placing a folded paper towel
in the bottom of an empty pipette tip box (we usually use a
USA scientific stacked rack made for 200 l tips, e.g., #11111206), adding 1 in. of ddH2O, and replacing the tip holder on
462
Eric Cox et al.
the box. The arrays will sit on the tip holder and be covered
with the lid to maintain a humid environment.
2. SUMOylation of many proteins occurs in the absence of E3
ligases at relatively high E1 and E2 concentrations; therefore
E1 and E2 concentrations are usually used in the low end of
this range when using an E3 ligase. However, it is advisable to
establish the best SUMOylation conditions using a conventional tube-based assay, particularly E1 and E2 concentrations,
with the SUMO E3 ligase of interest and a known substrate.
The best result will allow clear visualization of enhanced
SUMOylation in the presence of the E3 ligase, before proceeding with the assay on-chip.
3. To minimize evaporation and potential photobleaching of any
fluorescently labeled protein beacons included on the protein
microarray, it is advised to wrap four-well plate with a layer of saran
wrap, then a layer of aluminum foil, before overnight incubation.
4. It is critical to strike the right balance of removing excess liquid
from the array while not over-drying at this step. The goal is to
remove enough liquid so as not to dilute the reaction mix,
while also not allowing the top (protein surface) to not become
completely dry. The best results can be obtained by being
ready to add reaction mix immediately, i.e., reaction mix tube
open, pipet set to correct volume and already loaded with a
pipet tip, after placing the chip on the surface of the tip holder.
5. Ensure that bubbles are pushed out of the reaction mix while
adding coverslip to the top surface of the array. One reliable
strategy is to first rest one end of the coverslip on the edge of
the array, then gradually lower the other end of the coverslip to
push out the bubbles.
6. 1 % SDS is used in order to stringently wash the array and remove
any non-covalently bound SUMO from the arrayed proteins.
7. Removing the coverslip is best achieved by gently sliding the
coverslip straight along the array with only the raised edges of
the coverslip contacting the outside edges of the chip, where
no proteins are printed. Allowing the raised edges of the coverslip to move across the array (contacting the region where
proteins are printed) can scratch the surface of the chip, causing high background signal in the scratched region, resulting
in lost information about proteins in the path of the scratch.
8. A wash with high purity water is essential to remove residual
salts that can result in high background when scanning.
9. To get the most meaningful information from your protein
microarray experiment, it is important to scan the chip with
the appropriate gain and power settings. A useful starting point
using GenePix software would be PMT Gain = 600,
Power = 100 %. Usually the gain value can then be raised or
463
lowered depending on the resulting image obtained at these

settings. Generally, it is desired to have the highest signal intensity values possible while avoiding signal saturation of important features.
References
1. Zhu H, Bilgin M, Bangham R et al (2001)
Global analysis of protein activities using proteome chips. Science 293:21012105
2. Zhu H, Hu S, Jona G et al (2006) Severe acute
respiratory syndrome diagnostics using a coronavirus protein microarray. Proc Natl Acad Sci
U S A 103:40114016
3. Chen CS, Korobkova E, Chen H et al (2008)
A proteome chip approach reveals new DNA
damage recognition activities in Escherichia
coli. Nat Methods 5:6974
4. Popescu SC, Popescu GV, Bachan S et al
(2007) Differential binding of calmodulinrelated proteins to their targets revealed
through high-density Arabidopsis protein
microarrays. Proc Natl Acad Sci U S A 104:
47304735
5. Lueking A, Possling A, Huber O et al (2003)
A nonredundant human protein chip for antibody screening and serum profiling. Mol Cell
6. Hu S, Li Y, Liu G et al (2007) A protein chip
approach for high-throughput antigen identification and characterization. Proteomics
7:21512161
7. Song Q, Liu G, Hu S et al (2010) Novel autoimmune hepatitis-specific autoantigens identified using protein microarray technology.
8. Hu S, Xie Z, Onishi A et al (2009) Profiling
the human protein-DNA interactome reveals
ERK2 as a transcriptional repressor of interferon signaling. Cell 139:610622
9. Hu S, Wan J, Su Y et al (2013) DNA methylation presents distinct binding sites for human
transcription factors. eLife 2:e00726
10. Jeong JS, Jiang L, Albino E et al (2012) Rapid
identification of monospecific monoclonal
antibodies using a human proteome microarray. Mol Cell Proteomics 11:O111.016253
11. Lee YI, Giovinazzo D, Kang HC et al (2014)
Protein microarray characterization of the
S-nitrosoproteome. Mol Cell Proteomics
13:6372
12. Tarrant MK, Rho HS, Xie Z et al (2012)

Regulation of CK2 by phosphorylation and
O-GlcNAcylation revealed by semisynthesis.
Nat Chem Biol 8:262269
13. Barry G, Briggs JA, Vanichkina DP et al (2014)
The long non-coding RNA Gomafu is acutely
regulated in response to neuronal activation
and involved in schizophrenia-associated alternative splicing. Mol Psychiatry 19:486494
14. Donnelly CJ, Zhang PW, Pham JT et al (2013)
RNA toxicity from the ALS/FTD C9ORF72
expansion is mitigated by antisense intervention. Neuron 80:415428
15. Hu CJ, Song G, Huang W et al (2012)
Identification of new autoantigens for primary
biliary cirrhosis using human proteome microarrays. Mol Cell Proteomics 11:669680
16. Geiss-Friedlander R, Melchior F (2007)
Concepts in sumoylation: a decade on. Nat
Rev Mol Cell Biol 8:947956
17. Oh Y, Chung KC (2013) UHRF2, a ubiquitin
E3 ligase, acts as a small ubiquitin-like modifier E3 ligase for zinc finger protein 131. J Biol
Chem 288:91029111
18. Liang Q, Deng H, Li X et al (2011) Tripartite
motif-containing protein 28 is a small
ubiquitin-related modifier E3 ligase and negative regulator of IFN regulatory factor 7.
J Immunol 187:47544763
19. Garcia-Gutierrez
P,
Juarez-Vicente
F,
Gallardo-Chamizo F et al (2011) The transcription factor Krox20 is an E3 ligase that
sumoylates its Nab coregulators. EMBO Rep
12:10181023
20. Dos Santos MT, Trindade DM, Goncalves KA
et al (2011) Human stanniocalcin-1 interacts
with nuclear and cytoplasmic proteins and acts
as a SUMO E3 ligase. Mol Biosyst 7:180193
21. Chu Y, Yang X (2011) SUMO E3 ligase activity of TRIM proteins. Oncogene 30:
11081116
22. Gill G (2004) SUMO and ubiquitin in the
nucleus: different functions, similar mechanisms? Genes Dev 18:20462059
Chapter 33
Amyloid-Binding Proteins: Affinity-Based Separation,
Proteomic Identification, and Optical Biosensor Validation
Alexei Medvedev, Olga Buneeva, Arthur Kopylov, Oksana Gnedenko,
Alexis Ivanov, Victor Zgoda, and Alexander A. Makarov
Abstract
The amyloid-beta peptide is considered as a key player in the development and progression of Alzheimers
disease (AD). Although good evidence exists that amyloid-beta accumulates inside cells, intracellular brain
amyloid-beta-binding proteins remain poorly characterized. Here we describe a protocol for affinity-based
profiling of amyloid-beta-binding proteins of rat brain, their proteomic identification and validation by a
surface plasmon resonance (SPR)-based analysis. It includes: (a) SPR-based selection of immobilization
conditions for beta-amyloid coupling and choice of appropriate resin for preparation of an affinity sorbent;
(b) immobilization of beta-amyloid on the selected resin; (c) preparation of biological samples (rat brain
homogenate) for affinity-based separation; (d) isolation of rat brain proteins by affinity chromatography
using amyloid-beta as the affinity ligand; (e) sample preparation for proteomic analysis; (f) proteomic
analysis and protein identification; (g) SPR-based validation of the interaction of identified proteins with
immobilized amyloid-beta.
Key words Amyloid-beta, Amyloid-beta-binding proteins, Affinity-based proteomic profiling, Rat brain
Introduction
The amyloid-beta peptide 142 formed during proteolytic processing of the amyloid precursor protein (APP) is considered as a key
player in the development or progression of Alzheimers disease
(AD) and other pathologies associated with formation of protein
aggregates in the central nervous system [16]. Although much
attention is paid to formation of extracellular amyloid plaques and
protein aggregates as well as to corresponding mechanisms of their
toxicity, good evidence exists that intracellular amyloid-beta can
accumulate intraneuronally and contribute to disease progression
[4, 713]. This suggests importance of amyloid-beta interaction
with particular intracellular targets. Indeed, interaction of amyloid-beta with intracellular catalase caused inactivation of this
enzyme and accumulation of hydrogen peroxide inside cells [14].
465
466
Alexei Medvedev et al.
This implies that oxidative stress induced in the cells exposed to

amyloid-beta may be (at least partially) associated with reduced
degradation of intracellular hydrogen peroxide [14].
In the cell amyloid-beta was found in different intracellular
compartments [4]. The latter suggests possibility of amyloid interaction with a broad range of proteins, which may be thus denominated as amyloid-beta-binding proteins. Although there are reports
on the interaction of amyloid-beta peptide with various intracellular proteins [14, 15] and development of protocols for affinity isolation of amyloid-beta-binding proteins [16], the affinity-based
proteomic profiling of brain proteins interacting with immobilized
amyloid-beta, their identification has not been performed yet.
Here we describe a protocol for affinity-based profiling of
amyloid-beta-binding proteins of rat brain, their proteomic identification and validation by a surface plasmon resonance (SPR)-based
analysis. It includes several interrelated blocks (Fig. 1):
(a) SPR-based selection of immobilization conditions for amyloidbeta coupling and choice of appropriate resin for preparation
of an affinity sorbent.
(b) Immobilization of amyloid-beta on the selected resin.
(c) Preparation of biological samples (rat brain homogenate) for
affinity-based separation.
Fig. 1 Scheme illustrating the experimental design
Proteomic Profiling of Amyloid-Binding Proteins
467
(d) Isolation of rat brain proteins by affinity chromatography using

amyloid-beta as the affinity ligand.
(e) Sample preparation for proteomic analysis.
(f) Proteomic analysis and protein identification.
(g) SPR-based validation of the interaction of identified proteins
with amyloid-beta.
Materials
2.1 SPR-Based
Biosensor
All SPR-based biosensor consumables are commercially available

from GE Healthcare as ready to use dry substances, which should
be dissolved using certain volumes of water.
1. Immobilization buffer A: Na-acetate buffer, pH 4.0. Prepare a
10 stock solution by adding 41 mL of 0.01 M glacial acetic
acid to a 100 mL graduated cylinder containing 9 mL of
0.01 M sodium acetate.
2. Immobilization buffer B: 0.01 M potassium phosphate buffer,
pH 4.75. Prepare a 10 stock solution. Transfer 88.7 mL of
0.1 M potassium phosphate monohydrate to a 100 mL graduated cylinder and add 11.3 mL of 0.1 M potassium phosphate
dehydrate and filter through a 0.45 m Whatman filter.
3. EDC:
1-ethyl-3-(3-dimethylaminopropyl)
hydrochloride (EDC).
carbodiimide
4. NHS: N-hydroxysuccinimide (NHS).

5. Activating mixture: 0.2 EDC, 0.05 NHS.
6. HBS-EP buffer: 0.01 M HEPES, pH 7.4, 0.15 M NaCl,
0.03 M EDTA, 0.005 % surfactant P20 (Polyoxyethylene
sorbitan).
7. Blocking solution: 1 M ethanolamine, pH 8.5.
8. Rabbit muscle glyceraldehyde-3-phosphate dehydrogenase
(GAPDH) (Sigma-Aldrich).
9. Running buffer: 0.05 M potassium phosphate buffer, pH 7.4.
Transfer 80.2 mL of 0.1 M potassium phosphate monohydrate
to a 100 mL graduated cylinder; add 19.8 mL of 0.1 M potassium phosphate dehydrate. Make up to 100 mL with water and
filter through a 0.45 m Whatman filter.
10. Washing buffer A: 1 M NaCl in running buffer.
11. Washing buffer B: 0.05 M glycineHCl, pH 9.5. Prepare 5
stock solution. Weigh 47 g glycine, add 400 mL of deionized
water, and adjust pH to 9.5 by 6 M HCl. After titration make
up the volume to 500 mL with water.
12. Biacore 3000 with Control software (GE Healthcare).
468
2.2 Components
for Affinity-Based
Separation
1. Biological material (rat or mouse brains).

2. Human amyloid-beta protein fragment (142) of the highest
purity available (Sigma-Aldrich).
3. Affi-Gel 10 (Bio-Rad).
4. Protease inhibitors: 0.5 % bacitracin water solution, 0.1 %
aprotinin water solution, and 0.1 M phenylmethylsulfonyl fluoride (PMSF) solution in 96 % ethanol.
5. Immobilization buffer: 0.1 M Na-acetate buffer, pH 4.0.
6. Washing buffer: 1 M Tris-acetate buffer, pH 7.0. Weigh 12.11
of Tris-base and add 70 mL deionized water. Adjust pH to 7.0
by glacial acetic acid. Make up the volume to 100 mL with
water.
7. Storage buffer: running buffer from 2.1 and add 0.02 % NaN3.
8. Solubilization solution: 0.05 M potassium phosphate buffer,
pH 7.4.
9. Triton X-100.
10. Elution buffer: 1 M glycine-HCl, pH 2.8, 0.15 M NaCl.
Dissolve 7.56 g glycine and 8.76 g sodium chloride (NaCl) in
70 mL water and adjust pH to 2.8 by 6 M HCl. Make up the
volume to 100 mL with water.
2.3 Proteomic
Analysis
1. Lysis buffer: 0.1 M TrisHCl, pH 8.5, 8 M urea, 0.015 M

dithiothreitol. Add 0.12 g Tris-base, 4.8 g urea, 0.023 g
dithiothreitol and water up to the total volume of 8 mL. Dissolve
the urea completely. Adjust pH to 8.5 by 6 M HCl and make
up the volume to 10 mL with water.
2. Alkylation solution: 0.05 M iodoacetamide. Dissolve 9.25 mg
iodoacetamide in 1 mL water.
3. 0.05 M ammonium bicarbonate solution. Dissolve 39.5 mg of
NH4HCO3 in 10 mL water.
4. Washing solution: 0.5 M NaCl. Dissolve 292 mg sodium chloride in 10 mL water.
5. Trypsin sequencing grade modified (specific activity 18,611 U/
mg) (Promega).
6. Mobile phase A: 0.08 % formic acid and 0.015 % TFA in water.
7. Mobile phase B: 0.08 % formic acid and 0.015 % TFA in 80 %
acetonitrile.
8. Mobile phase C: 0.08 % formic acid and 0.015 % TFA in 4 %
acetonitrile.
2.4
Instrumentation
1. Optical biosensor Biacore 3000 and the Control software (GE

Healthcare) for instrument operation.
2. Research grade sensor chips CM5 (GE HealthCare).
469
3. Homogenizer Ultra-Turrax T 10 (Cole-Parmer).

4. Compact liquid chromatography system AKTAprime plus (GE
Healthcare).
5. Polystyrene chromatographic columns Tricorn 5/20 (GE
Healthcare).
6. Spectrophotometer Cary 50 (Varien).
7. Microcentrifuge Eppendorf 5415 R (Eppendorf).
8. Peristaltic pump P-1 (GE Healthcare).
9. Centrifuge Heraeus megafuge 40 R (Thermo Scientific).
10. Centrifugal filters Amicon Ultra-15 (10 K) and Microcon-10
(10 K) (Merck Millipore Ltd).
11. Thermostat 5320 (Eppendorf).
2.5 LC-MS and Data
Evaluation
1. High resolution mass spectrometer Q Exactive (Thermo

Scientific) equipped with Easy-Spray ion source.
2. Column for reverse-phase chromatography PepMap RSLC
C18, 150 mm 50 m, particle size 2 m, pore size 100 A
(Thermo Scientific).
3. Chromatography RSLC-nano system Ultimate 3000 (Thermo
Scientific) composed of nano-flow and loading pumps, column
oven with integrated 10-port switching valve and microautosampler.
4. Data acquisition and analysis software Xcalibur version 2.2
SP1.48.
5. Data converter software Raw2MSM version 1.10.
6. Data analysis and evaluation software MASCOT server 16 CPU.
Methods
3.1 SPR-Based
Selection
of Immobilization
Conditions for BetaAmyloid Coupling
SPR-based selection of immobilization conditions and appropriate

resin are required for preparation of an affinity sorbent. Most frequently, affinity-based sorbents are prepared using cyanogen bromide (CNBr)-activated Sepharose [16]. In this case CNBr-activated
Sepharose coupling of ligands of interest occurs under alkaline
condition via their amino groups. However, in alkaline conditions
the beta-amyloid peptide (142) readily aggregates [17]. Taking
into consideration our SPR-based data on poor immobilization of
beta-amyloid peptide (142) on the optical biosensor chip surface
at pH 4.0 (Fig. 2), it appears that CNBr-activated Sepharose is
inapplicable for covalent immobilization of the monomeric form of
this peptide (see Note 1). Thus, we have chosen Affi-Gel 10 as an
alternative resin due to possibility of covalent immobilization
under a broad range of pH values.
Fig. 2 SPR-based selection of immobilization conditions for amyloid-beta coupling and validation of amyloidbeta interaction with glyceraldehyde-3-phosphate dehydrogenase. (a) Optimization of pH for buffer used for
amyloid-beta immobilization: 1amyloid-beta peptide (100 g/mL) in immobilization buffer B, 2HBS-EP

3.1.1 Optimization of pH
for Immobilization
471
1. Prepare human amyloid-beta 142 peptide 100 g/mL solution

in immobilization buffers A and B.
2. Inject the peptide solution, using a contact time of 2 min.
3. Set report points just before the start and end of the injection
to determine the level of electrostatically bound peptide.
3.2 Immobilization
of Amyloid-Beta
Peptide
1. Activate sensor chips CM5 surface (GE HealthCare) by the

activating mixture.
2. Inject amyloid-beta (100 g/mL in immobilization buffer A
or B) at a flow rate of 5 L/min for 30 min.
3. Wash 5 min with HBS-EP buffer.
4. Passivate active groups by injection of blocking solution for
1 min.
3.2.1 Binding
Measurements
3.3 Immobilization
of Amyloid Beta
Protein Fragment 142
on Affi-Gel 10
Proteinpeptide interactions are monitored by injecting various

concentrations of a model protein, glyceraldehyde-3-phosphate
dehydrogenase (GAPDH) (see Note 2), at a flow rate of 5 L/min
for 5 min; GAPDH was dissolved in running buffer at 144 g/
mL1.44 mg/mL. Regenerate the sensor surface after each measurement cycle by sequential injections of washing A and washing
buffer B with for 0.5 min at a flow rate of 50 L/min (Fig. 2).
1. Dissolve 1 mg amyloid beta protein fragment 142 in 250 L
distilled water.
2. Add 250 L immobilization buffer (see Subheading 2.2) and
mix gently.
3. Wash Affi-Gel 10 resin (0.5 mL) on a glass filter with 10 volumes of distilled water and 2 volumes of the immobilization
buffer A.
4. Transfer the resin to a 1.6 mL Eppendorf tube and eliminate
the buffer by aspiration after centrifuging of resin slurry at
1,485 g for 3 min.
5. Add amyloid beta protein fragment to the resin at a ratio of 1:1
(v/v) and incubate at 4 C overnight at a gentle stirring.
6. Remove the excess of the protein fragment by centrifuging
45 times in washing buffer, at 1,485 g for 3 min.
Fig. 2 (continued) buffer, pH 7.4, 3amyloid-beta peptide (100 g/mL) in immobilization buffer A. At pH 4.0
(immobilization buffer A) amyloid-beta demonstrates much better interaction with the surface of the optical
biosensor chip. (b) Immobilization of amyloid-beta peptide (100 g/mL) in 10 mM sodium acetate, pH 4.0
onto the surface of the optical biosensor chip. Arrows indicate injections: 1EDC/NHS; 2HBS-EP buffer;
3100 g/mL amyloid-beta in immobilization buffer A; 4blocking solution. (c) Interaction of GAPDH
(glyceraldehyde-3-phosphate dehydrogenase) with immobilized amyloid-beta peptide. Results are corrected
versus control channel (without amyloid-beta peptide)
472
7. Add the same buffer to the resin up to the suspension 1:1

(v/v) and incubate at 4 C for 4 h at a gentle stirring.
8. Wash 45 times with the same buffer and resuspend the resin
in storage buffer.
9. Store Affi-Gel 10 with the immobilized amyloid beta protein
fragment at 4 C. The immobilized protein fragment can be
kept under such conditions for more than 4 weeks.
10. Perform all the above-mentioned incubations (except the addition of amyloid beta protein fragment) with the control AffiGel 10.
3.4 Preparation
of Rat Brain
Homogenate
for Affinity-Based
Separation
1. Decapitate rats under light ether anesthesia.

2. Immediately dissect the brains and homogenize at a low speed
in solubilization buffer (1:3) (w/v), using an Ultra-Turrax T
10 homogenizer.
3. Measure protein concentration using Bradford protein assay
and spectrophotometer Cary 50 (Varian).
4. Add Triton X-100 (final concentration 3 %) to brain homogenates and incubate at 4 C for 1 h.
5. Dilute the brain homogenate lysates threefold with solubilization buffer and remove Triton-insoluble components by centrifugation at 23,755 g for 20 min at 4 C.
3.5 Isolation of Rat

Brain Proteins by
Affinity Chromatography Using
Amyloid-Beta
as the Affinity Ligand
1. Add the resultant supernatant (about 5 mg of protein/mL) to

the suspension of the affinity sorbent (Affi-Gel 10 with the
immobilized amyloid beta protein) (1:1, v/v).
2. Add protease inhibitors (bacitracin, aprotinin, and phenylmethylsulfonyl fluoride (PMSF)) up to the concentrations of
0.005 %, 0.01 %, and 0.001 M, respectively, and incubate the
suspension overnight at 4 C at a gentle stirring [18].
3. Perform the same incubations with the control Affi-Gel 10
without immobilized protein ligand to evaluate nonspecific
binding of proteins (see Note 3).
4. After the incubation wash the sorbent by centrifuging with
solubilization buffer up to the absence of the protein in the
washing (evaluated by baseline at OD280 on spectrophotometer Cary 50 (Varian)).
5. Pack the sorbent using the polystyrene chromatographic columns Tricorn 5/20 (1 0.75 mL) (GE Healthcare) and eluate
proteins by eluting buffer at a flow rate of 0.5 mL/min using
compact liquid chromatography system AKTAprime plus (GE
Healthcare) and peristaltic pump P-1 (GE Healthcare).
6. Concentrate the eluate up to 0.200 mL using Amicon Ultra15 centrifugal filter devices (Heraeus megafuge 40 R, bucketrotor, 2,121 g, 25 min, 4 C).
3.6 Sample
Preparation
for Proteomic Analysis
473
Extract, modify, and digest protein samples on filters using the

recently published FASP protocol [19].
1. Precipitate the proteins using chloroformmethanol method
[20]. Briefly, add methanol to the protein sample (1:2.5) vortex vigorously for 30 s and add chloroform (equal to the initial
volume of the protein sample), vortex again and centrifuge the
resultant mixture for 2 min at 23,755 g at 20 C for 2 min.
Discard the upper layer and add a threefold excess (versus initial sample) of methanol, vortex and centrifuge at 20 C for
3 min. Aspirate supernatant and dry the resultant sediment
under air for 2030 min.
2. Dissolve the sediment (of precipitated proteins) in 200 L of
lysis buffer for proteins denaturation and reducing of oxidized
disulfide bonds.
3. Pipette the samples onto the centrifugal filters Microcon-10
(Merck Millipore Ltd) and centrifuge at 1,485 g at 20 C for
40 min.
4. Add 100 L of alkylation solution in washing buffer and incubate at 20 C for 30 min in darkness.
5. Centrifuge under the same conditions for 40 min.
6. Wash twice with 200 L of washing solution (each time centrifuge under the same conditions for 40 min).
7. Add ammonium bicarbonate solution and sequencing grade
trypsin (see Notes 4 and 5) at a ratio of 1:100 = total mass of
enzyme:total mass of protein.
8. Incubate overnight at 37 C using Thermostat 5320
(Eppendorf).
9. Centrifuge under the same conditions for 40 min.
10. Wash with 100 L of washing solution.
3.7 Liquid
Chromatography:
Mass Spectrometry
Perform all mass spectrometry experiments using a high-resolution

mass spectrometer Q Exactive (Thermo Scientific) equipped with
Easy-Spray ion source and RSLC-nano Ultimate 3000 chromatography system (Thermo Scientific) composed of nano-flow and
loading pumps, column oven with integrated 10-port switching
valve and micro-autosampler, or any high resolution mass spectrometer available.
1. Select the positive ionization mode.
2. Use data-dependent acquisition mode at 70 K resolution in
MS and 35 K resolution in MS/MS.
3. Set the AGC targets at 1e6 ion for 20 ms and 5e4 ions for
75 ms at MS and MS/MS levels, respectively, with the underfill ratio of 10 %.
474
4. Set the instrument for top ten isolated MS peaks (window

1 m/z) to be used for the further tandem MS spectra acquisition for full duty cycle.
5. Apply high energy collision dissociation (HCD) for peptides
fragmentation with 25 % stepped 30 eV normalized collision
energy. Use 100 m/z as the first fixed mass for tandem spectra
acquisition.
6. Use active dynamic exclusion mode for 40 s and chromatography peak filtering (width of 15 s).
7. Trigger exclusion charge states of ions at z + 1 and z > 5+.
8. Use nano-LC Ultimate 3000+ system with mobile phase A and
mobile phase B for reverse-phase chromatographic separation
of digested peptides was performed on PepMap RSLC C18,
150 mm 50 m, particle size 2 m, pore size 100 A (Thermo
Scientific).
9. Load 1 L of a sample digest for 4 min at a flow rate 5 L/min
in the isocratic mode using mobile phase C onto Acclaim
PepMap C18 column (75 m 150 mm, 2 m particle size,
100 A pore size).
10. Separate peptides by means of linear elution gradient of mobile
phases A and B from 4 to 65 % of acetonitrile at a flow rate of
0.3 L/min following by rapid increase (within 3 min) of the
gradient to 98 % of mobile phase B and keep this mode remaining for 10 min. After the whole procedure perform column
post-analysis reconstitution under initial conditions for 10 min.
3.8 Database
Searching
and Identification
For peptide identification, MS/MS data collected from LC-MS

experiments are searched against Uniprot custom filtered database
using MASCOT (version 2.2) search algorithm. Spectra are
searched against Rat subset of SwissProt database.
1. Use the following search parameters: trypsin as the cutting
enzyme, set mass tolerance for the monoisotopic peptide window at 20 ppm (parts per million) and 0.05 Da for the MS/
MS tolerance window, one missed cleavage is allowed.
2. Perform parallel search of the data against decoy database, use
the results to estimate the q-value using the Percolator algorithm and filter at 1 % FDR (false discovery rate).
3. Specify carbamidomethylation of cysteine as fixed modification
and set methionine oxidation and cyclization of N-terminal
glutamine as variable modifications.
4. Set the following criteria of positive identification: minimum
score of 50, at least three positive identifications from three
different runs.
3.9 SPR-Based
Validation
of the Interaction
of Identified Proteins
with Amyloid- Beta
475
Further validation of the interaction between immobilized amyloidbeta and proteins identified by mass spectrometry can be obtained
using various methods. We recommend the SPR-based biosensor
analysis with purified proteins of interest (see Figs. 1 and 2).
Advantages of the SPR-based validation consist in the possibility of
repeated use of the optical biosensor cuvette with immobilized
amyloid-beta.
Notes
1. Recent results of isoelectric focusing studies indicate that low
molecular forms of monomeric amyloid-beta (142 and 140)
are characterized by a pI value of ca. 5, while at pI of 66.5
aggregates of high molecular weight are detected [17]. This
suggests that it is better to use more acidic solutions of amyloidbeta to prevent its aggregation.
2. Several studies have recognized glyceraldehyde-3-phosphate
dehydrogenase as the amyloid-binding protein (e.g., [18, 21]),
and therefore it can be used as a probe for interaction with
immobilized amyloid-beta.
3. Various intracellular highly abundant proteins (e.g., cytokeratins) may nonspecifically bind to a resin even without an immobilized affinity sorbent (see for example, ref. 22) and therefore
a control batch of the ligand-free sorbent has to be subjected
to all the same treatments but without amyloid-beta. The list
of identified amyloid-binding proteins will have to be corrected for possible presence of proteins nonspecifically bound
to the control sorbent.
4. Use sequencing grade trypsin only for mass spectrometry
experiments due to its high specificity and low contamination
of alpha-chymotrypsin. Also it is strongly recommended to use
modified trypsin (with acetylated lysins) to prolong activity of
enzyme and slow down its autolysis.
5. In the case of the use of non-sequencing grade trypsin its catalytic activity may be rather low; if so, it is recommended to add
up to 0.003 M of calcium chloride to increase ionic strength of
reaction mixture rather than sodium salts to minimize further
cationization during LC-MS analysis. However, it should be
noted that native trypsin may undergo autolysis, generating
pseudotrypsin, which exhibits a broadened substrate specificity
including chymotrypsin-like activity (see comments [23]). In
addition, trypsin may be contaminated with chymotrypsin,
and this fact has to be taken into consideration especially when
high enzyme/substrate ratios are used.
476
Acknowledgments
This work was supported by the Russian Science Foundation (grant
no. 14-24-00100) within the project The role of structural polymorphism of amyloid-beta in initiation of Alzheimers disease.
References
1. Schmechel DE, Saunders AM, Strittmatter WJ,
Crain BJ, Hulette CM, Joo SH, Pericak-Vance
MA, Goldgaber D, Roses AD (1993) Increased
amyloid beta-peptide deposition in cerebral
cortex as a consequence of apolipoprotein E
genotype in late-onset Alzheimer disease. Proc
Natl Acad Sci U S A 90:96499653
2. Saido TC, Iwatsubo T, Mann DM, Shimada H,
Ihara Y, Kawashima S (1995) Dominant and
differential deposition of distinct beta-amyloid
peptide species, A beta N3(pE), in senile
plaques. Neuron 14:457466
3. Hardy J, Selkoe DJ (2002) The amyloid
hypothesis of Alzheimers disease: progress and
problems on the road to therapeutics. Science
297:353356
4. LaFerla FM, Green KN, Oddo S (2007)
Intracellular amyloid- in Alzheimers disease.
Nat Rev Neurosci 8:499509
5. Masters CL, Selkoe DJ (2012) Biochemistry of
amyloid -protein and amyloid deposits in
Alzheimer disease. Cold Spring Harb Perspect
Med 2(6):a006262
6. Upadhaya A, Kosterin I, Kumar S, von Arnim
CA, Yamaguchi H, Fndrich M, Walter J, Thal
DR (2014) Biochemical stages of amyloid-
peptide aggregation and accumulation in the
human brain and their association with symptomatic
and
pathologically
preclinical
Alzheimers disease. Brain 137(pt 3):887903
7. Li QX, Maynard C, Cappai R, McLean CA,
Cherny RA, Lynch T, Culvenor JG, Trevaskis
J, Tanner JE, Bailey KA, Czech C, Bush AI,
Beyreuther K, Masters CL (1999) Intracellular
accumulation of detergent-soluble amyloidogenic A beta fragment of Alzheimers disease
precursor protein in the hippocampus of aged
transgenic mice. J Neurochem 72:24792487
8. Wirths O, Multhaup G, Czech C, Blanchard V,
Moussaoui S, Tremp G, Pradier L, Beyreuther K,
Bayer TA (2001) Intraneuronal A beta accumulation precedes plaque formation in betaamyloid precursor protein and presenilin-1
double-transgenic mice. Neurosci Lett 306:
116120
9. Butterfield DA, Sultana R, Poon HF (2006)
Redox proteomics: a new approach to investigate oxidative stress in Alzheimers disease. In:
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
Dalle-Donne I, Scaloni A, Butterfield DA (eds)

Redox proteomics: from protein modifications
to cellular dysfunction and diseases. Wiley,
Hoboken, pp 563603
Knobloch M, Konietzko U, Krebs DC, Nitsch
RM (2007) Intracellular A and cognitive deficits precede -amyloid deposition in transgenic
arcA mice. Neurobiol Aging 28:12971306
Reddy PH, Beal MF (2008) Amyloid beta,
mitochondrial dysfunction and synaptic damage: implications for cognitive decline in aging
and Alzheimers disease. Trends Mol Med 14:
4553
Spuch C, Ortolano S, Navarro C (2012) New
insights in the amyloid-beta interaction with
mitochondria. J Aging Res 2012:324968
Wu Z, Zhu Y, Cao X, Sun S, Zhao B (2014)
Mitochondrial toxic effects of A through mitofusins in the early pathogenesis of Alzheimers
disease. Mol Neurobiol 50(3):98696
Habib L, Lee MTC, Yang J (2010) Inhibitors
of catalase-amyloid interactions protect cells
from -amyloid-induced oxidative stress and
toxicity. J Biol Chem 285:3893338943
Hernandez-Zimbron LF, Luna-Munoz J,
Mena R, Vazquez-Ramirez R, Kubli-Garfias C,
Cribbs DH, Manoutcharian K, Gevorkian G
(2012) Amyloid- peptide binds to cytochrome
C oxidase subunit 1. PLoS One 7(8):e42344
Calero M, Rostagno A, Ghiso J (2012) Search
for amyloid-binding proteins by affinity chromatography. Methods Mol Biol 849:213223
Wiberg H, Ek P, Ekholm PF, Lannfelt L,
Emmer , Roeraade J (2010) Separation and
characterization of aggregated species of
amyloid-beta peptides. Anal Bioanal Chem
397:23572366
Schulze H, Schuler A, Stuber D, Dobeli H,
Langen H, Huber G (1993) Rat brain
glyceraldehyde-3-phosphate
dehydrogenase
interacts with the recombinant cytoplasmic
domain of Alzheimers beta-amyloid precursor
protein. J Neurochem 60:19151922
Wisniewski J, Zougman A, Nagaraj N, Mann M
(2009) Universal sample preparation method
for proteome analysis. Nat Methods 6:359362
Walker JM (ed) (2002) The protein protocol
handbook. Humana Press Inc., Totowa, NY

21. Verdier Y, Fldi I, Sergeant N, Flp L, Penke
Z, Janky T, Szcs M, Penke B (2008)
Characterization of the interaction between
Abeta 1-42 and glyceraldehyde phosphodehydrogenase. J Pept Sci 14:755762
22. Buneeva O, Gnedenko O, Zgoda V, Kopylov
A, Glover V, Ivanov A, Medvedev A, Archakov
477
A (2010) Isatin binding proteins of rat and

mouse brain: proteomic identification and
optical biosensor validation. Proteomics 10:
2337
23. Medvedev AE (2013) In macropore tryptic
digestion at acidic pH and its implication for
proteomics. Proteomics 13:31013102
Chapter 34
Proteomic Profiling by Nanomaterials-Based
Matrix-Assisted Laser Desorption/Ionization Mass
Spectrometry for High-Resolution Data and Novel Protein
Information Directly from Biological Samples
Abstract
Qualitative and quantitative analyses of global proteome samples derived from biocomplex mixtures are
very important to understand the cellular functions in cell biology. Matrix-assisted laser desorption/
ionization mass spectrometry (MALDI-MS)-based proteomics has recently become one of the most
informative and attractive core technologies in proteomics. Particularly, nanomaterials-based MALDI
mass spectrometric methods are quickly becoming a critical miniaturized bioanalytical tool for detecting
and discerning proteins from biocomplex samples. These MALDI-developed strategies allow highthroughput identification of proteins from highly complex mixtures including accurate mass measurement of peptides derived from total proteome digests and peptides/proteins separations from various
samples. The nanomaterials-integrated MALDI-MS technologies in protein arrays hold much promise
for interrogating the diverse and immense proteome in cell biology. As a result, nanomaterials-assisted
MALDI-MS-based proteomic workflow, including sample preparation, information on the local molecular composition, relative abundance and spatial distribution of peptides and proteins and their analysis,
should make the technology more easily available to a broad community and turn it into a powerful
methodology for bioanalysts.
Key words Nanomaterials, MALDI-MS, Proteomics, Sample preparation, Protein digestion
Introduction
A cornerstone of proteomics is the ability to monitor and to
characterize proteins in a cell or any biological sample. Mass spectrometry has become the bioanalytical technique for proteome
analyses, which has the ability to probe the covalent structure of
proteins. Among the mass spectrometric-based proteomics,
MALDI-MS-based proteomics is mature and an attractive technology for characterization and quantification of proteins in a biological sample or system [1]. Although MALDI-MS has been known
as a powerful bioanalytical technique for protein analysis, it still
479
480
faces significant challenges for complicated proteome research in

large scale due to large proportions of protein biomarkers exist in
extremely low abundances and thus require tedious sample pretreatment to trace the specific proteins [2, 3]. The successful detection of ultratrace biomolecules depends mostly on sample
pretreatment. In the last three decades, the biomolecular mass
spectrometric capabilities have been dramatically extended by the
development of nanomaterials-based MALDI-MS approaches for
biomolecules profiling in biocomplex samples [3]. This has led to
many MALDI-MS methodologies to be developed for proteome
research, with unique capabilities of specificity, sensitivity, speed,
and sampling with minimal volumes. Nanomaterials-based
MALDI-MS have greatly enhanced the analytical dynamic range
and limit of detection for biomolecules (proteins and peptides)
profiling in biocomplex samples (plasma, serum, other bodily fluids, or tissue) [13]. In MALDI-MS-based proteomics, nanomaterials exhibit as multifunctional roles such as matrices, and affinity/
concentrating probes for isolation and enrichment of ultratrace
target biomolecules prior to their identification [4]. As a result, the
workflow splits into three directions: (1) analysis of biomolecules
(peptides and large proteins) for reduced background noises, high
resolution and accuracy, (2) enrichment of tryptic digest products,
and (3) extraction and preconcentration proteins in biosamples.
1.1 Nanomaterials
as Matrices
The unique optical and physico-chemical properties of nanomaterials have been explored to be used as matrices for analysis of peptides and proteins with high mass accuracy and reduced background
signals because nanomaterials act as functional probes for generation of biomolecular mass spectra which produce better mass spectra than the conventional organic matrices [2]. The small nature
and unique physico-chemical properties of nanomaterials allow
them to enter/cross cellular membranes and their high surface area
to volume ratio provides a possibility to trap trace analytes with
high degree from complex samples. Furthermore, nanostructures
acted as good candidates to absorb incoming laser radiation and to
transfer the absorbed energy to the analytes, promoting their
desorption with high mass accuracy and with reduced background
noises, and eliminating the need for the organic matrix. Due to the
high dispersion ability of nanomaterials, nanomaterials-conjugated
analytes are efficiently and readily transferred between instruments,
which allow to detect ultratrace biomolecules with minimal sample
loss. A wide variety of nanomaterials including carbon-based nanostructures (carbon nanotubes, graphite, graphene), metallic (Au,
Ag, and Pt NPs), metal oxides (TiO2, ZnO, SiO2, Fe3O4, and
ZrO2), and semiconductor (ZnS, CdS, CdTe, and HgTe, etc.)
nanoparticles have been used as matrices for the analysis of biomolecules by MALDI-MS with good accuracy and high mass resolution [2, 57]. Due to their strong UV absorption cross-section of
Proteomic Profiling by Nanomaterials
481
nanomaterials, they acted as energy mediators for transferring

adsorbed energy to desorb and ionize surface ligands (analytes),
resulting in producing clean biomolecular mass spectra with high
mass sensitivity, resolution, and accuracy.
1.2 Nanomaterials
as Affinity Probes
In general, the bottom-up method has been widely accepted for

the routine identification of proteins in complex mixtures. In the
bottom up approach, proteins are quantified that can provide a
direct indication of the up-regulation or down-regulation of the
genes. Bottom-up-based MALDI-MS approaches involve the analysis of protein sequences from the protein digests (enzymatically or
chemically digesting proteins into small peptides). In comprehensive proteomics, separation and enrichment of low abundance peptides from very complex samples are challenging tasks for bioanalysts
[8]. Although not yet at the levels of data throughput and automation achieved in other genomic analyses such as DNA sequencing
or microarray gene expression analysis, global protein profiling
methods are rapidly evolving. However, it can be possible because
of recent improvements in MS instrumentation, protein and peptide separation techniques, computational data, and the availability
of complete sequence databases for many species. To meet shortgun proteomics, many efforts have been devoted on the combination of protein digestion, MS and MS/MS approaches for the
identification of protein sequence [8, 9]. Although these direct
MALDI-MS-based proteomics allow to analyze protein sequence
in protein mixtures, unfortunately, proteome analysis is not
straightforward due to the following reasons. First, the stoichiometry of target proteins is relatively difficult to know. Second, the
MS-based analytical techniques have a limited dynamic range for
the identification of minor sites in the target proteins. Third, repetitive sample transfers cause loss of peptides, yielding a low sequence
coverage for database searching with less confident protein identification. To solve these problems, various nanomaterials (metal
NPs, metal oxides, metal sulfides, and tellurides) have been used as
probes to speed up proteolysis and to capture the digested proteins
from biocomplex samples prior to their identification by
MALDI-MS with high degree of protein sequences [10, 11].
1.3 Nanomaterials
as Extracting
and Preconcentrating
Probes
Recent trends in MALDI-MS-based proteomics have been intensively focused on the improvement in quality of mass spectra of
target proteins, introduction of new technological developments
with MALDI-MS and especially in miniaturization, simplification,
and automation of the whole analytical procedure, to extract and
concentrate trace proteins in microvolume samples [12, 13]. Prior
to the identification of target proteins/peptides by MALDI-MS,
sample preparation step is essentially required to isolate them from
the sample matrix, and to enrich them to a concentration level
detectable by MALDI-MS. To achieve the objective, substantial
482
efforts have been made on the integration of nanomaterials in

miniaturization sample-pretreatment procedures (liquidliquid
microextraction (LLME) and single-drop microextraction
(SDME)) prior to their identification by MALDI-MS. Apart from
these, nanomaterial-based chips have been developed for isolation
and preconcentration of ultratrace biomolecules (peptides and
proteins) from various sample matrices prior to MALDI-MS analysis. These approaches improved the sample preparation performance, especially addresses the issues of sample and solvent
consumption, selectivity and recoveries, and speeding up the sample-treatment process, which is currently considered the bottleneck of proteome analysis.
In view of the above applications of nanomaterials in MALDIMS-based proteomics, this chapter is mainly focused on the analysis of proteins by nanomaterials-based MALDI-MS that has been
successful in our hands. It also discusses the potential applications
of nanomaterials-based MALDI-MS for deriving the protein composition of protein mixtures, to determine the members of proteins in biocomplex samples.
Materials and Instrumentation

1. Peptides
(methionine-enkephalin
(Met-enk),
leucineenkephalin (Leu-enk), HW6 (thiopeptide, HCKFWW), glutathione, gramicidin D, and valinomycin).
2. Proteins (trypsinogen, apo-transferrin (human), albumin from
human serum (HSA), albumin from bovine serum (BSA),
1-antitrypsin, immunoglobulin G (IgG), protein G, cytochrome c, trypsin, lysozyme, -casein, and -casein) were used
as standard analytes in MALDI-MS for proteomic profiling.
3. All proteins solutions (myoglobin, lysozyme, cytochrome c,
BSA, HSA, transferring, trypsin, -casein and -casein) were
prepared in (mg/mL) 50 mM of NH4HCO3 solution.
4. The standard solution of gramicidin D was prepared (mg/mL)
in methanol. 1.75 mM Leu-enk, 1.73 mM Met-enk, and
1.11 mM HW6 peptide solutions were prepared by dissolving
1 mg of above substances in deionized water.
5. The matrices solutions (0.133 M -cyano-4-hydroxycinnamic
acidCHCA, 1.33 mM 3,5-dimethoxy-4-hydroxycinnamic
acidSA, and 2,5-dihydroxybenzoic acid2,5-DHB, 20 mg/
mL) were prepared in 2:1 mixture of acetonitrilewater containing 0.1 % of trifluoroacetic acid (TFA).
6. All mass spectra were obtained in positive ion mode using a
Microflex MALDI-TOF-MS and 337 nm nitrogen laser was
483
used for irradiation of the sample. Ions produced by laser

desorption were stabilized during a delayed extraction period
of 200 ns before entering the mass analyzer and then accelerated through the TOF analyzer. The accelerating voltages
existed in the range from +20 to 20 kV. All experiments were
performed in a linear (m/z > 5,000) and reflectron (m/z < 5,000)
mode of TOF-MS.
7. The microwave oven has 2,450 MHz frequency and the maximum power is 700 W. A thermocouple probe was used to measure the temperatures of sample solutions prior to and after the
microwave experiments.
2.1 Functionalization
of Nanomaterials
1. Quantum dots (QDs) were synthesized by the well described

procedures in the literature [14]. Similarly, the unmodified
gold nanoparticles (Au NPs) were prepared by using NaBH4 as
a reducing agent. After preparation of bare QDs and Au NPs,
dopamine dithiocarbamate was attached on the surfaces of
QDs by the described procedures [14, 15]. Briefly, 0.549 mM
dopamine (104.1 mg), 0.549 mM carbon disulfide (40 L),
and triethylamine (5 L) were taken in a 2.0 mL glass vial and
sonicated for 5 min at room temperature. The formed dopamine dithiocarbamate was added into a 25 mL of nanomaterials (QDs and Au NPs) and then stirred for 15 min at room
temperature. Finally, the product was washed with ethanol to
remove reactants and unbound dopamine dithiocarbamate.
2. Vanillin, ethanol, acetic acid, and 4-aminothiophenol were
used for the synthesis of (4-mercaptophenyliminomethyl)-2methoxyphenol (Schiff base 3). Then, the bare Au NPs dispersed aqueous solution was added to the toluene solution
containing 1.66 mM of 4-mercaptophenyliminomethyl-2methoxyphenol and the solution was stirred vigorously for 5 h.
The Au NP was successfully transferred to the organic phase
(toluene), as indicated by the color change from light yellow to
dark brown [16].
3. 400 L of MPA was transferred into a 250 mL round bottom
flask containing 15 mL of deionized water. To this, 2 mL of
0.01 M cadmium nitrate was added dropwisely under N2 pressure with constant stirring for 1 h. The pH of the solution was
adjusted to 9 by adding ammonium hydroxide solution.
2.5 mL of 0.008 M sodium sulfide solution was quickly added
to the above solution at 96 C and solution was stirred for 2 h.
The obtained clean green-yellowish CdS QDs solution was
stored at 4 C until further use [17]. Similarly, HgTe nanostructures were prepared according to a previously described
procedure [7].
484
2.2 Preparation
of Hydrophobic
Nanomaterials
1. AgNO3, Se powders and octadecylamine (ODA), octadecanethiol (ODT) and 11-mercaptoundecanoic acid (MUA) were
used for the synthesis of functionalized silver selinide nanoparticles (Ag2Se NPs) [18]. The formed NPs were washed with
ethanol and then dispersed in toluene using an ultrasonicator
for further use.
2. Surface-modified BaTiO3 NPs were synthesized according to
the literature procedure [19]. Briefly, BaTiO3 NPs (0.5 g) surfaces were hydroxylated by refluxing with 200 mL of H2O2 for
4 h at 106 C. The hydroxylated BaTiO3 NPs were filtered and
then washed with deionized water for two times. The hydroxylated BaTiO3 NPs were directly added into 250 mL round
bottom flask that contained 50 mL of 50 % (v/v) aqueous
alcohol along with 0.5 (w/v) of 12-hydroxy octadecanoic acid
(HOA). The reaction mixture was heated at 90 C under vigorous stirring for 3 h. The reaction mixture was cooled to
room temperature and HOA-modified BaTiO3 NPs were
treated with 1 N HCl until the pH became 56 in order to
agglomerate the nanopowder and to facilitate NPs filtration.
The surface-modified BaTiO3 NPs were washed with deionized water and acetone in sequence, several times. The surfacemodified BaTiO3 NPs were dried in vacuum oven for 24 h and
then dispersed in toluene by ultrasonication for 10 min.
2.3
Bacteria Growth
Escherichia coli (12570) standard culture bacteria were obtained

from Bioresource Collection and Research Center, Taiwan.
Glassware and media were subjected to autoclave at 15 lbs of pressure for 15 min prior to bacteria culture. One colony of E. coli was
carefully taken up from a freshly prepared 24 h old streak plate culture using a sterile loop. The collected bacteria were cultured on
Luria Broth Agar (LBA) plates for 24 h at 37 C. All microbiological experiments were performed in a Biosafety Level 1 cabinet [19].
Methods
3.1 Functionalized
Nanomaterials
as Matrices
for Peptides
and Proteins
1. The stock solutions of peptides (Leu-enk, Met-enk, and HW6)

were diluted further for working concentrations. An aliquot
900 L of the peptide solutions were taken in a 1 mL polyethylene vial and 0.1 M HCl or NaOH was added to control pH
of the solution. 100 L of (0.56.0 M) Mn2+-doped ZnS
semiconductor nanoparticles solution was added and vortexed
for 30 min with speed of 86 g and then centrifuged the vials
at 6,797 g. The sample solutions (1 L) were pipetted onto
a stainless steel target plate and allowed to air-dry for 10 min
before the MALDI-MS analysis ([20], Note 1).
485
2. 750 L of standard solution of analytes (proteins or peptides)

were taken into a 1 mL polyethylene vial. 250 L of known
concentration of CdS QDs solution was added and pH of the
test solution was controlled by adding 0.1 M HCl or
NaOH. The sample vials were vortexed for 30 min at 86 g
and then 1 L of the above test solution was placed on the
target plate for MALDI-TOF-MS analysis [17].
3. The prepared HgTe nanostructures were used as concentration and desorption/ionization matrixes. Sample solutions
(1 mL) containing a protein, HgTe nanostructures (0.08),
and 300 mM ammonium citrate, pH 5.0 were equilibrated at
ambient temperature for 30 min. The mixtures were then centrifuged (3,823 g, 10 min), and all the supernatants were
removed. The pellets were resuspended in ammonium citrate
buffer (20 L), and the sample solutions (1 L) were pipetted
onto a stainless steel target plate and allowed to air-dry for
30 min prior to SALDI-MS measurement ([7], Note 2).
4. Furthermore, HgTe nanostructures are used as the matrix for
the investigation of two protein protein complexes:
1-antitrypsin trypsin and IgG protein G. Trypsin
(1.7 15 M), 1-antitrypsin (5 M), and Zn2+ ions (0.14 M)
were equilibrated in ammonium citrate solutions (20200 mM,
pH 8.0, 50 L) at 25 C for 1 h; protein G (220 M), IgG
(10 M), and Zn2+ ions (0.14 M) were equilibrated in
ammonium citrate solutions (20 mM, pH 5.0, 50 L) at 25 C
for 1 h. Equal volumes of the HgTe nanostructures (4, 10 L)
in the presence of 0.1 or 1 % Brij 76 and one of the protein
mixtures (10 L) were mixed by vortexing for 1 min. Aliquots
(1.0 L) of the mixtures were pipetted onto a stainless-steel
96-well MALDI target and dried in air at room temperature
for 30 min prior to SALDI-MS analysis [21].
3.2 MicrowaveAssisted Tryptic
Digestion
1. Eppendorf tubes are used to perform microwave-assisted tryptic

digestions of BSA, cytochrome c, -casein, and non-fat milk.
First, 50 M of the above proteins were dissolved in 50 mM of
NH4HCO3 buffer (pH 8.3) containing 8.0 M of urea. To this,
2.5 L of 1.8 M CaCl2 was added and then reduced by the addition of 5.0 L of 50 mM DTT for few minutes. The final concentration of each protein (18.0 M; 250 L) was prepared by
using 50 mM of NH4HCO3 and urea was maintained below
2 M. To these solutions, trypsin was added at an enzyme-toprotein ratio at 1:40 (w/w) for the microwave-assisted digestion. The above eppendorf tubes were placed in plastic rack and
then microwave irradiation was performed by applying
microwave power at 210 W for 50 s. After completion of microwave irradiation, sample solutions were subjected to cool for few
minutes prior to enrichment procedures with nanomaterials as
486
affinity probes. All the microwave tryptic-digested samples were

diluted to a certain concentration with 50 % (v/v) ACN that
contained 0.1 % of TFA.
2. 500 L of non-fat milk (50-folds of dilution) was denatured
by adding 50 mM of NH4HCO3 and 8 M of urea and the
sample vials were vortexed for a few minutes at room temperature. To this, 100 L of 50 mM DTT and 5.0 L of 1.8 M
CaCl2 were added and then 200 L of trypsin (mg/mL) was
added. The microwave irradiation was performed at 210 W
for 50 s. The sample vials were taken out and cooled for a few
minutes and digested proteins were enriched by the aforesaid
procedures.
3.3 Nanomaterials
as Affinity Probes
for the Enrichment
of Digested Proteins
A solution of the suspended nanomaterials (BaTiO3 NPs50 L,

12 mg/mL; QDs-100 L, 100 nM) was added to 150 L of a
microwave tryptic digest sample. The mixture was vortexed for
30 min at room temperature. Then, nanomaterials-conjugated
digested proteins were isolated by subjecting to centrifugation at
2,655 g for 10 min at room temperature. The obtained supernatant solution was removed using micropipette (see Note 3). The
nanomaterials-conjugated digested proteins (phosphoproteins and
non-phosphoproteins) were collected at the bottom of sample vial.
To avoid washing procedures, directly 1 L of 0.2 M 2,5-DHB
solution was mixed with 1.0 L of nanomaterials-conjugated
digested protein solution and then 0.5 L of nanomaterialsconjugated digested protein solution was placed on the MALDI
target for MALDI-MS analysis ([14], Note 4). The developed
workflow was successfully applied for the identification of phosphopeptides in milk ([21], Note 5).
3.4 Nanomaterials
as Extracting
and Concentrating
Probes for Enrichment
of Peptides
and Proteins
The nanomaterials-based SDME experiments were performed by

the following procedures. The target analytes with desired concentrations were spiked into a glass vial filled with 1 mL sample solution. A 10 L of microsyringe was used to draw 1 L of toluene
containing the modified nanomaterials. The microsyringe was
inserted into the sample solution through a PTFE-coated silicon
septum of screw cap of a glass vial. As soon as the sample was
extracted into the 1 L microdroplet of organic solvent (the modified nanomaterials prepared in toluene), the microdroplet was
drawn back into the microsyringe and then directly placed the 1 L
microdroplet onto the target plates for subsequent MALDI-MS
analysis ([16], Note 6).
Nanomaterials-based LLME: To this, 900 L of standard peptide
mixture (valinomycin0.40 M; gramicidin D2.0 M) was
taken into a 1.0 mL polyethylene vial and then pH of the sample
was adjusted by adding 1.0 M of HCl or NaOH. Then, 100 L of
487
dispersed functionalized Ag2Se NPs (2.0 mg/mL) was added and

then vortexed at 86 g for 30 min. The sample vials were allowed
to stand for 3 min to separate dispersed Ag2Se NPs in toluene
(organic) and aqueous layers. After that 2 L of dispersed functionalized Ag2Se NPs conjugated hydrophobic peptides (valinomycin and gramicidin D) was taken by using micropipette and
mixed with equal volume of 0.2 M 2,5-DHB. The above sample
was allowed to air dry prior to MALDI MS analysis [18].
For hydrophobic proteins in E. coli: 200 L of cultured E. coli
(1.5 108 cfu/mL) was taken into a 1.0 mL vial and then 200 L
of HOA-modified BaTiO3 NPs dispersed in toluene was added.
The vials were vortexed for 30 min at room temperature. The sample vials were kept on stands for 3 min to separate NPs contained
in organic layer and in aqueous layers. Then, 1.0 L of HOAmodified BaTiO3 NPs-conjugated hydrophobic proteins (upper
layer) was mixed with 1.0 L of 0.5 M SA. Finally, 1.0 L of above
solution was placed on the MALDI target plate, dried at room
temperature and then analyzed by MALDI-MS ([19], Note 7).
3.5 Peptides
and Proteins
Identification by
Nanomaterials-Based
MALDI-MS
Among different possible bioanalytical approaches, nanomaterialsbased MALDI mass spectrometry-based proteomics is increasingly
used to acquire the data important for understanding these processes. For example, Mn2+-doped ZnS-cysteine NPs-based
MALDI-MS was successfully used for the analysis of peptide mixture of Leu-enk (3.0 M), Met-enk (3.0 M), and HW6 (2.1 M)
(Fig. 1). It can be observed that Mn2+-doped ZnS-cysteine NPsbased MALDI-MS provided remarkable mass spectra for peptide
mixture, corresponding to the mass peaks of peptides at m/z 578.2,
595.3, 615.6, 906.6, and 929.5 for [Leu-enk + Na]+, [Metenk + Na]+, [Leu-enk + K]+, [HW6 + H]+, and [HW6 + Na]+, respectively. All ion signals were generated with sodium and potassium
adducts while only HW6 peptide was generated as protonated
ions. This reason is due to the alkali metals have greater affinity
toward the small molecules in MALDI-MS and also due to the
nanomaterials as the matrix lacking of the proton sources.
Moreover, large proteins were successfully detected in E. coli by
using HgTe nanostructures as matrices in MALDI-MS (Fig. 2).
The peaks at m/z 10,770, 21,539, 10,728, and 21,454 that correspond to the adducts [zSSAT1 + 2H]2+, [zSSAT1 + H]2+, [zSSAT1
R101A + 2H]2+, and [zSSAT1R101A + H]+, respectively, resulting
from the recombinant proteins of zSSAT1 or zSSAT1 R101A
under individual plasmids transformed into E. coli strain BL21.
Although zSSAT1 and zSSAT1 R101A share 99.5 % amino acid
sequence homology and only one amino acid variation, our
approach allowed differentiation of them, showing its potential for
the rapid detection of proteins. Interestingly, HgTe nanostructuresbased MALDI-MS was effectively detected protein G in both the
488
Fig. 1 MALDI-mass spectra of peptide mixture (Leu-enk, Met-enk, and HW6) using (a) CHCA and (b) Mn2+doped ZnS-cysteine nanoparticles as the matrices. Peak identities: m/z 578.2, 595.3, 615.6, 906.6 and 929.5
which are attributed to the [Leu-enk + Na]+, [Met-enk + Na]+, [Met-enk + K]+, [HW6 + H]+ and [HW6 + Na]+.
Reproduced from ref. 20 with permission from Royal Chemical Society
absence and presence of Brij 76, but the protein G IgG complex
was detectable only in the presence of 0.1 % Brij 76 [22]. The signals at m/z 26,023 and 13,019 represent the [protein G + H]+ and
[protein G + 2H]2+ adducts. The mass peaks at m/z 74,885, 49,984,
26,023, 13,019, 86,585, and 58,297 correspond to the
[IgG + 2H]2+, [IgG + 3H]3+, [protein G + H]+, [protein G + 2H]2+,
[IgG+ protein G + 2H]2+, and [IgG + protein G + 3H]3+ adducts
(Fig. 3). Using this approach, singly and multiply charged (1+, 2+,
3+, and 6+) adducts of IgG were observed and this approach successfully allowed to detect protein G (m/z 26,023) and IgG (m/z
149,931) at concentrations as low as 2 M (1 pmol) and 5 M
(2.5 pmol).
489
Fig. 2 Mass spectra of solutions containing HgTe nanostructures and the recombination proteins (0.01 g/L)
(a) zSSAT1 and (b) zSSAT1 R101A. SALDI-MS was performed in the linear mode. The peaks at m/z 10,770,
21,539, 10,728, and 21,454 represent the adducts [zSSAT1 + 2H]2+, [zSSAT1 + H]+, [zSSAT1R101A + 2H]2+, and
[zSSAT1 R101A + H]+, respectively. Reproduced from ref. 7 with permission from The American Chemical Society
The identification and characterization of post-translational

modifications (PTMs) of proteins can be easily studied by top-down
fragmentation of intact protein ions. However, traditional proteolysis requires 612 h and tedious separation procedures are needed
for their identification. These are limitations for rapid identification
of digested proteins with high selectivity and sensitivity. To overcome this problem, functionalized QDs are used as affinity probes
for the enrichment of microwave-tryptic digested BSA in
MALDI-MS (Fig. 4). Using this approach, the protein digestion
time was drastically reduced from 12 h to 50 s with good sequence
coverage. From the obtained MALDI mass spectrum, 11 proteolytic peptides (B1, B3, B6, B8, B9, B16, B21, B22, B24, B25, and B26)
were identified from microwave tryptic digest of BSA with low
sequence coverage of 26 %. In order to obtain more efficient microwave tryptic digestion of proteins, QDs-DDTC were used as concentrating probes for the enrichment of microwave-tryptic digest of
BSA without any pretreatment procedure. As a result, many digest
fragments of BSA (35 peptides from B1 to B35) were observed with
high signal-to-noise ratios by using QDs-DDTC as affinity probes.
Fig. 3 Mass spectra of IgG, protein G, and their complexes, recorded through SALDI-MS using HgTe nanostructures (a) in the absence and (b and c) in the presence of 0.1 % Brij 76. The samples were prepared in 20 mM
ammonium citrate (pH 5.0) containing 1 M Zn(II). The signals at m/z 149,931, 74,885, 49,984, 24,997,
26,023, 13,019, 86,585, and 58,297 represent the adducts [IgG + H]+, [IgG + 2H]2+, [IgG + 3H]3+, [IgG + 6H]6+,
[protein G + H]+, [protein G + 2H]2+, [IgG + protein G + 2H]2+, and [IgG + protein G + 3H]3+, respectively.
Reproduced from ref. 22 with permission from The American Chemical Society
Fig. 4 MALDI-TOF mass spectra of microwave tryptic-digested peptides originated from BSA (6.0 M) using
(a) 2,5-DHB as the conventional matrix and (b) QDs-DDTC as the affinity probes. Reproduced from ref. 14 with
permission from Elsevier
491
a
5000
7 2000
4000
1000
13
3000
Absolute intensity
1000
x10
1600
17
15
2000 1
0
4
14
1800
2000
2200
2400
22
10
15
4000
12 13
2000
11
1600
16
18
17
14
1800
19
2000
2200
22
10
1250
1500
2400
23
24
0
1000
21
20
1
1 2
21
20
1750
2000
m/z
2250
2500
2750
3000
3250
Fig. 5 (a) Direct MALDI mass spectrum of microwave tryptic digest of -casein (4.0 106 M). (b) MALDI mass
spectrum of microwave tryptic digest of -casein using BaTiO3 NPs (12 mg/mL) as enrichment probes.
Reproduced from ref. 21 with permission from Springer
These results indicate that the identified proteolytic peptides were

matched with the amino acid sequence coverage of 56 % for BSA.
Furthermore, 24 phosphopeptides (from 1 to 24) are effectively identified with improved signal intensities using BaTiO3 NPs
as enrichment probes from the microwave tryptic digest of -casein
(Fig. 5b). Figure 5 provides remarkable evidence that some of the
low abundance phosphopeptides (12, 16, 18, 19, and 20) are
effectively identified using BaTiO3 NPs. The BaTiO3 NPs were
used as enrichment probes to concentrate phosphopeptides from
non-fat milk. Notably, 21 phosphopeptides (from M1 to M21)
were effectively detected in MALDI-MS using BaTiO3 NPs as the
concentrating probes (Table 1). The phosphopeptide peaks of
-casein in non-fat milk are appeared at m/z 1,980.6, 2,350.5 and
2,475.1, which are due to dephosphorylated fragment ions (loss of
phosphorylated groups) from the ions at m/z 2,061.2, 2,431.3
and 2,556.2, respectively (data were not shown). The identified
phosphopeptides in non-fat milk are listed in Table 1.
Functionalized Au NPs were successfully integrated in SDME
technique for the extraction of milk proteins in milk. The mass
peaks at m/z 9,168.01, 11,498.10, 14,218.10, and 18,394.56 are
492
Table 1
List of identified phosphopeptides by MALDI-MS from the microwave tryptic digest of non-fat milk
(50-fold dilution) using BaTiO3 NPs as affinity probes
Sr. No
Peak number
Observed m/z
Phosphopeptide sequences
M1
924.4
DIGpSESTEDQAMEDIK (-S1/5873)
M2
976.2
YKVPQLEIVPNpSAEER (-S1/119134)
M3
1,003.3
NANEEEYSIGpSpSpSEEpSAEVATEEVK (-S2/6185)
M4
1,027.8
DIGpSEpSTEDQAMEDIKQ (-S1/4359)
M5
1,103.5
GNAEGpSpSDEEGKLVIDEPAK (-S1/180188)
M6
1,251.6
TKVIPYVRYL (-S2/213222)
M7
1,266.8
YLGYLEQLLR (-S1/91100)
M8
1,293.7
QMEAEpSIpSpSpSEEIVPNpSVEQ (-S1/7493)
M9
1,337.5
VNELpSKDIGpSEpSTEDQAMEDIK (-S1/5273)
10
M10
1,383.7
FFVAPFPEVFGK (-S1/3849)
11
M11
1,593.6
TVDMEpSTEVFTKK (-S2/153165)
12
M12
1,637.2
YLGYLEQLLRLKK (-S1/106118)
13
M13
1,641.3
FFVAPFPEVFGKEK (-S1/3851)
14
M14
1,759.8
HQGLPQEVLNENLLR (-S1/2337)
15
M15
1,769.3
LYQGPIVLNPWDQVK (-S2/114128)
16
M16
1,847.4
DIGpSETEDQAMEDIK (-S1/5873)
17
M17
1,927.6
DIGpSEpSTEDQAMEDIK (-S1/119134)
18
M18
2,105.0
TDAPSFSDIPNPIGSENSEK (-S1/189208)
19
M19
2,235.2
HPIKHQGLPQEVLNENLLR (-S1-(1937))
20
M20
2,618.7
NTMEHVpSpSpSEESIIpSQETYK (-S1/1736)
21
M21
2,678.0
VNELpSKDIGpSEpSTEDQAMEDIK (-S1/5273)
Reproduced from ref. 21 with permission from Springer. pS refers to phosphorylated serine unit
corresponded to proteoso pep. PP81 [1], 3-casein [2],

-lactoalbumin [3], and -lactoglobulin [4], respectively [16]. To
investigate the potential applications of functionalized nanomaterials in LLME for extraction and preconcentration of target proteins,
HOA-modified BaTiO3 NPs used as extracting and preconcentrating probes for LLME of hydrophobic proteins in E. coli prior to
their identification by MALDI-MS. Figure 6 displays the MALDI
mass spectrum of identified hydrophobic proteins in E. coli by using
HOA-modified BaTiO3 NPs-assisted LLME coupled with
MALDI-MS. Using this approach, 14 hydrophobic proteins were
successfully extracted and preconcentrated by using HOA-modified
493
Fig. 6 MALDI mass spectra of identified hydrophobic proteins in E. coli by using (a) HOA-modified BaTiO3 NPsassisted LLME along with SA (0.5 M) as the matrix and (b) SA (0.5 M) as the matrix without HOA-modified
BaTiO3 NPs. Reproduced from ref. 19 with permission from Elsevier
BaTiO3 NPs as hydrophobic affinity probes. Importantly, the signal

intensities of hydrophobic proteins (at m/z 5,090, 5,368, 7,009,
7,173, 7,861, and 8,405) were greatly enhanced (213 times) by
using HOA-modified BaTiO3 NPs as extracting and preconcentrating probes. The mass peaks at m/z 3,418, 7,185, 10,855, 5,878,
and 7,020 Da corresponded to the membrane proteins ecnB
(P56549), lpp (P69776), and osmE (P23933) and to hypothetical
membrane proteins yifL (P39166) and ygdI (P65292), respectively.
Similarly, the mass peak at m/z 8,888 corresponded to acetyl-acyl
carrier protein (ydhI; acetyl-ACP, P0A6A8; acetylation of the phosphopantetheine sulfur). Evidently, five lipoproteins (ecnB, lpp, osmE,
yifL, ygdI) and water-insoluble ATPase proteolipid (at m/z 8,282;
atpL, P68699) were effectively extracted, preconcentrated and then
identified by using HOA-modified BaTiO3 NPs-LLME coupled
with MALDI-MS. Unfortunately, only one peak (at m/z 8,405) is
generated with good signal intensity, the remaining peaks [27,
11, 12] were not well resolved and some of the peaks [1, 8, 10, 23]
were not generated by using SA as the matrix. These results indicated that HOA-modified BaTiO3 NPs have showed high affinity
toward hydrophobic proteins of E. coli, since HOA molecules on
the surfaces of BaTiO3 NPs have played a fundamental key role for
the efficient LLME of hydrophobic proteins in E. coli [19].
494
Notes
1. Nanomaterials exhibited high surface area, good photostability, which results to improve the solubility in aqueous media
and their conjugation with biomolecules.
2. The mass limit when SALDI-MS was used with HgTe nanostructures as matrixes reaches up to 150,000 Da (IgG), much
higher than those obtained using other kinds of nanomaterials.
The HgTe nanostructures-based MALDI-MS allowed to
detect weak drugprotein complexes (BSA/Y and hCAI-ACZ
complexes). Importantly, HgTe nanostructures allow the analyses of proteins and their complexes under mild conditions and
with greater tolerance toward salts.
3. BaTiO3 NPs are good candidatures for effective enrichment of
phosphopeptides from microwave tryptic digest of -casein.
Moreover, after centrifuging of BaTiO3 NPsphosphopeptides
conjugates from microwave tryptic digests, the trapped phosphopeptides are directly identified by MALDI-MS without
further washing procedure step. The entire procedure was
completed within 60 min.
4. The high degree of protein digestion was observed by using
following microwave conditions (microwave power: 210 W
and microwave irradiation time: 50 s) and trypsin-to-protein
ratio (1:30, w/w) was used for the digestion of cytochrome c,
lysozyme, and BSA proteins in the present study.
5. The BaTiO3 NPs acted as concentrating probes for phosphopeptides from microwave tryptic digest of -casein, since trace
levels of phosphopeptides (12, 16, 18, 19, and 20) are effectively enriched and identified without non-phosphorylated
peaks. It confirms that surfaces of BaTiO2 NPs have high capability to adsorb phosphopeptides from microwave tryptic
digest of -casein. It can be proved that BaTiO3 NPs have welldispersed in solution which can improve signal intensities, S/N
ratio and allowed unambiguous trapping of low abundance
phosphopeptides (Fig. 2).
6. The binary matrix approach applying Au NP-SDME microdroplets which can be homogeneously mixed with the organic
matrix (CHCA or SA) to form homogeneous crystals and thus
favorable for sensitive detection of peptides and proteins at low
concentration in MALDI-MS. In addition, the Au NP-SDME
can also serve as multifunctional nanoprobes for ionization,
enrichment, preconcentration, and desalting purposes for a
variety of peptides and proteins. Moreover, the sample can be
directly deposited onto the MALDI target plates and directly
sent for MALDI-MS analysis without the need for any further
washing steps or elution processes.
495
7. HOA-functionalized BaTiO3 NPs have attained hydrophobic

nature by the functionalization of BaTiO3 NPs surfaces with
HOA (17-alkyl chains), which facilitates to enhance the analytes extraction and preconcentration through hydrophobic
interactions. The use of HOA in MALDI-MS highly improves
its performance by acting as affinity probes and possibly as a
buffer in laser energy transfer to enhance the desorption/
ionization of hydrophobic proteins.
Acknowledgement
The authors greatly acknowledge Elsevier, Royal Chemical Society,
Springer, and American Chemical Society for giving copyright permission to reuse figures and some of the text for this chapter.
References
1. Godovac-Zimmermann J, Brown LR (2001)
Perspectives for mass spectrometry and functional proteomics. Mass Spectrom Rev 20:157
2. Chiang C-K, Chen W-T, Chang H-T (2011)
Nanoparticle-based mass spectrometry for the
analysis of biomolecules. Chem Soc Rev 40:
12691281
3. Zhu Z-J, Rotello VM, Vachet RW (2009)
Engineered nanoparticle surfaces for improved
mass spectrometric analyses. Analyst 134:
21832188
4. Kailasa SK, Cheng K-H, Wu H-F (2013)
Semiconductor nanomaterials-based fluorescence spectroscopic and matrix-assisted laser
desorption/ionization (MALDI) mass spectrometric approaches to proteome analysis.
Materials 6:57635795
5. Kailasa SK, Kiran K, Wu H-F (2008)
Comparison of ZnS semiconductor nanoparticles capped with various functional groups as
the matrix and affinity probes for rapid analysis
of cyclodextrins and proteins in surface-assisted
laser desorption/ionization time-of-flight mass
spectrometry. Anal Chem 80:96819688
6. Kailasa SK, Wu H-F (2011) Semiconductor
cadmium sulphide nanoparticles as matrices for
peptides and as co-matrices for the analysis of
large proteins in matrix-assisted laser desorption/ionization reflectron and linear time-offlight mass spectrometry. Rapid Commun Mass
Spectrom 25:271280
7. Chiang C-K, Yang Z, Lin Y-W, Chen W-T, Lin
H-J, Chang H-T (2010) Detection of proteins
and protein-ligand complexes using HgTe
8.
9.
10.
11.
12.
13.
14.
nanostructure matrixes in surface-assisted laser

desorption/ionization mass spectrometry. Anal
Chem 82:45434550
Kailasa SK, Wu H-F (2014) Recent developments in nanoparticle-based MALDI mass
spectrometric analysis of phosphoproteomes.
Microchim Acta 181:853864
Nesvizhskii AI, Aebersold R (2005)
Interpretation of shotgun proteomic data. Mol
Zhang Y, Li L, Yang P, Lu H (2012) On-plate
enrichment methods for MALDI-MS analysis
in proteomics. Anal Methods 4:26222631
Kailasa SK, Wu H-F (2014) Advances in
nanomaterial-based microwaves and infrared
wave-assisted tryptic digestion for ultrafast
proteolysis
and
rapid
detection
by
MALDI-MS. Comb Chem High Throughput
Screen 17:6879
Kailasa SK, Mehta VN, Wu H-F (2014) Recent
developments of liquid-phase microextraction
techniques directly combined with ESI- and
MALDI-mass spectrometric techniques for
organic and biomolecule assays. RSC Adv
4:1618816205
Nuhua AA, Basheera C, Saad B (2011) Liquidphase and dispersive liquidliquid microextraction techniques with derivatization: recent
applications in bioanalysis. J Chromatogr B
879:11801188
Kailasa SK, Wu H-F (2012) Functionalized
quantum dots with dopamine dithiocarbamate
as the matrix for the quantification of efavirenz
in human plasma and as affinity probes for
496
15.
16.
17.
18.

rapid identification of microwave tryptic
digested proteins in MALDI-TOF-MS.
Kailasa SK, Wu H-F (2012) One-pot synthesis of
dopamine dithiocarbamate functionalized gold
nanoparticles for quantitative analysis of small
molecules and phosphopeptides in SALDI- and
MALDI-MS. Analyst 137:16291638
Shastri L, Kailasa SK, Wu H-F (2010)
Nanoparticle-single drop microextraction as
multifunctional and sensitive nanoprobes: binary
matrix approach for gold nanoparticles modified
with
(4-mercaptophenyliminomethyl)-2methoxyphenol for peptide and protein analysis
in MALDI-TOF MS. Talanta 81:11761182
Ke Y, Kailasa SK, Wu H-F, Chen Z-Y (2010)
High resolution detection of high mass proteins up to 80,000 Da via multifunctional CdS
quantum dots in laser desorption/ionization
mass spectrometry. Talanta 83:178184
Kailasa SK, Wu H-F (2010) Surface modified
silver selinide nanoparticles as extracting probes
to improve peptide/protein detection via
nanoparticles-based liquid phase microextraction coupled with MALDI mass spectrometry.
Talanta 83:527534
19. Kailasa SK, Wu H-F (2013) Surface modified

BaTiO3 nanoparticles as the matrix for
phospholipids and as extracting probes for
LLME of hydrophobic proteins in Escherichia
coli by MALDIMS. Talanta 114:283290
20. Kailasa SK, Wu H-F (2010) Interference free
detection for small molecules: probing the
Mn2+-doped effect and cysteine capped effect
on the ZnS nanoparticles for coccidiostats and
peptide analysis in SALDI-TOF MS. Analyst
135:11151123
21. Kailasa SK, Wu H-F (2012) Rapid enrichment of phosphopeptides by BaTiO3 nanoparticles after microwave-assisted tryptic digest
of phosphoproteins, and their identification
by MALDI-MS. Microchim Acta 179:8390
22. Chen W-T, Chiang C-K, Lee C-H, Chang H-T
(2012) Using surface-assist laser desorption/
ionization mass spectrometry to detect proteins and proteinprotein complexes. Anal
Chem 84:19241930
23. Lpez-Ferrer D, Caas B, Vzquez J, Lodeiro
C, Rial-Otero R, Moura I, Capelo JL (2006)
Sample treatment for protein identification by
mass spectrometry-based techniques. TrAC
Trend Anal Chem 25:9961005
INDEX
A
Acetone precipitation .......................................................410
Affinity chromatography ................................. 160, 211222,
307309, 319, 467, 472
Affinity electrophoresis.....................................................327
Alkylation ..........128, 159, 170, 173, 196197, 206, 236239,
241242, 251253, 257, 261, 262, 267269, 272,
273, 278, 282, 312, 316, 371, 372, 375, 468, 473
Amyloid-binding proteins ........................................465475
Antibody........................................... 24, 26, 30, 33, 121, 131,
132, 135139, 141143, 147149, 196, 279, 285,
286, 291, 298, 301, 325, 327, 331, 338, 343345, 347,
383, 384, 386390, 394, 398, 402411, 430, 437,
438, 442445, 448450, 452, 457, 460, 462
APV Gaulin .......................................................................2
B
Bead-based array ......................................................441, 442
Bead beater .......................................................................2, 4
Bead impact ......................................................................27
Bead mill ..........................................................................57
Benzonase.......................... 337, 358, 360, 366, 444, 446, 457
Bicinchoninic acid (BCA) protein assay ........... 310, 313, 444
BioNeb cell disruption........................................................16
Bis-Tris SDS PAGE gels .......................... 328, 329, 344, 345
Blue native electrophoresis ...............................................419
Bradford protein assay .............................. 296, 396, 398, 472
Branson sonifier................................................................296
C
Carbamidomethylation ..................................... 288, 363, 474
Carbamylation ............................................ 30, 271, 318, 365
Carrier ampholytes ............................122, 159, 164, 396, 411
Cell culture
CHO-K1 cells ............................................................398
Hep 70.4 cells .............................................................451
LIM1863 cells ............................................ 180, 182, 192
primary skeletal muscle cells .........................................57
SW 480 cells ....................................... 168170, 173, 176
Cell disruption
bead impact methods ..................................................27
electromotive field ........................................................18
high pressure batch ................................... 2, 3, 1013, 19
high pressure flow ..................................... 2, 3, 1316, 19
low pressure ...........................................2, 3, 7, 16, 17, 19
mortar and pestle tissue grinders ....................................9

rotorstator homogenizer ...........................................79
ultrasonic processors ...............................................1619
Cell lysis
LC-MS/MS analysis .................................... 67, 259273
phosphoprotein analysis..............................................313
Cell wall ....................................................... 10, 73, 228, 263
Chaotrope.................................................................145, 156
Chloroplasts
Arabidopsis ......................................... 212214, 217, 221
pea ...................................................... 212, 214, 216, 221
Chromatography
affinity chromatography ............................ 160, 211222,
276, 306309, 319, 467, 472
hydrophilic interaction liquid chromatography
(HILIC) ................................276, 279281, 286, 288
hydrophobic interaction chromatography .......... 213, 219,
315, 369
ion exchange chromatography ............ 176, 213, 215, 219
reverse-phase liquid chromatography ................ 145, 154,
168, 174, 176, 369
size exclusion chromatography ......30, 212, 213, 218, 219
CID. See Collision induced dissociation
Co-immunoprecipitation..........................................443, 445
Collision induced dissociation (CID) ...................... 288, 300,
302, 357, 359, 363366
Coomassie Brilliant Blue G-250 ......................................333
Coomassie Brilliant Blue R-250.......................................251
Coomassie staining ...........................................................420
Crosslinking .....................................................................138
Culture medium .......................................... 49, 60, 169, 170,
174, 181, 182, 192, 194, 200203, 262, 278, 331, 332,
334, 393, 410, 411, 443, 445, 446
CyDye ........................121, 127, 156, 311, 315, 429, 431, 439
Cysteine labeling ......................................................157, 159
D
Data normalization
stain-free technology ..........................................381390
western blotting ..................................................381390
Detergents .................................... 28, 30, 105, 110, 119, 120,
124, 130, 133, 136, 146148, 168, 203, 250,
254, 257, 265, 270, 311, 334, 410, 411, 417, 421,
423, 424, 428, 451
Dimethylformamide (DMF) ........................... 122, 127, 131,
140, 144, 146, 311, 315, 429, 431
DOI 10.1007/978-1-4939-2550-6, Springer Science+Business Media New York 2015
497
PROTEOMIC PROFILING
498 Index
Disulfide bridges ..............................................................157
Dithioerythritol (DTE) ....................................................164
Dithiothreitol (DTT) .................................67, 105, 110, 120,
122, 123, 128, 130, 145, 157, 159, 164, 191, 195, 197,
214, 216, 226, 236, 237, 246, 251, 252, 261, 262, 266,
268, 269, 271273, 278, 296, 298, 311, 312, 358, 361,
371, 372, 375, 396, 397, 400, 409411, 424, 429, 430,
432, 433, 457, 468, 485, 486
DMEM. See Dulbeccos modified Eagle medium
DMF. See Dimethylformamide
Dounce homogenizer .......................................................384
Droplet low pressure nebulizer ...........................................17
Droplets impingement .......................................................16
DTT. See Dithiothreitol
Dulbeccos modified Eagle medium (DMEM) ........... 45, 48,
201, 260, 278, 280, 331, 443
Dynamic range compression.......................................99105
Dyno-Mill ............................................................................6
E
Ectosome ..................................................................167, 168
Electron microscopy ............ 94, 173, 175, 183, 191, 196, 206
Electrophoresis
affinity electrophoresis ................................................327
blue native PAGE .......................................................419
characterization of extracellular vesicles..............167168
2-D electrophoresis ................................... 158, 394, 396,
397, 399400, 410, 412, 413
differential in gel electrophoresis (DIGE) ............. 61, 63,
105, 156158, 430, 431, 433, 434, 436, 438, 439
free flow electrophoresis .....................................415424
isoelectric focusing ............................... 51, 119, 122, 154,
156, 158, 176, 231, 293302, 311, 319, 369, 419, 428,
429, 431, 432, 475
SDS-PAGE ................................................ 267, 312, 429
zone electrophoresis .................................... 416, 419, 421
Electrospray ionization (ESI) .......................... 111, 145, 154,
163, 362, 369
Escherichia coli (E. coli) ............. 7, 73, 452, 484, 487, 492, 493
ESI. See Electrospray ionization
Exosome
density-gradient separation ................................179206
electron microscopy ............................ 183, 191, 196, 206
immunoaffinity capture methods........................179206
ultracentrifugation ..............................................179206
F
Filter aided sample preparation (FASP) .............. 6670, 237,
240241, 473
mass spectrometry ..............................................237, 240
Formalin fixed paraffin embedded (FFPE) ...... 109114, 120
Free flow electrophoresis (FFE) ...............................415424
French press ..............................................................2, 1012
G
Glycoprotein.............................................................205, 357
enrichment .................................................................357
Glycosidases .....................................................................356
Glycosylation ....................................275291, 324, 355357,
366, 367, 456
Gradient centrifugation
exosomes .............................................................180, 181
lymphocytes ............................................................3341
mitochondria .................................................... 83, 93, 94
H
Heat stabilization of proteins .......................................2131
HeLa cells ..........147, 280, 289, 334, 338, 342, 371, 376, 377
HILIC. See Hydrophilic interaction liquid chromatography
Homogenizer................................................. 79, 13, 15, 78,
80, 82, 89, 9196, 131, 321, 384, 459, 472
Horseradish peroxidase (HRP)................................. 398, 402
western blotting ..................................................398, 402
Host cell protein
antibody coverage ....................................... 398, 403408
CHO cells ...................................394, 396, 410, 412, 413
2-D electrophoresis ..... 394, 396, 399400, 410, 412, 413
PDQuest ............................................................398, 403
ProteoMiner ...............................................................100
western blotting ..................................................393414
Human plasma ...................................................................56
Human serum................................................... 100, 101, 482
Hydrophilic interaction liquid chromatography
(HILIC) ................................276, 279281, 286, 288
I
ICPL. See Isotope-coded protein label
IEF. See Isoelectric focusing
Immobilized metal affinity chromatography
(IMAC) ........................................276, 279281, 284,
289, 291, 306, 308310, 314
Immobilized pH gradient (IPG) ..................... 100, 122, 127,
128, 154, 158, 231, 294, 296, 298, 300302, 311, 315,
319, 370, 371, 373, 375, 377, 396, 399401, 411, 412,
429, 430, 432, 436
Immunoaffinity ...........................33, 137, 139, 168, 179207
Immunoassay ...................................57, 60, 62, 100, 441452
multiplexing ........................................................441452
Immunoblotting .......................................118, 124, 204, 206,
227228, 327, 331, 340, 343
Immunoprecipitation
biotin ...........................................137, 141, 143, 147, 148
crosslinking .................................................................138
magnetic beads ....................135, 137, 138, 140, 142149
mass spectrometry ..............................................135149
protein AG ................................. 137140, 143, 147149
PROTEOMIC PROFILING: METHODS AND PROTOCOLS

499
Index
proteinprotein interaction .........................................135
streptavidin ................................. 137, 138, 141142, 147
In-gel digest .............................................118, 119, 129, 236,
238240, 250, 259261, 263, 266270, 325, 331
In-solution digest ............................................ 136, 138, 145,
147, 172174, 237, 241242, 260, 266270
IPG. See Immobilized pH gradient
Isoelectric focusing (IEF) .................................. 51, 119, 122,
154, 156, 158, 176, 231, 293302, 311, 319, 324, 369,
419, 428432, 475
Isoelectric point (pl) ..........................159, 294, 306, 369, 377
Isotope-coded protein label (ICPL) .........................162164
ITRAQ..................................................... 272, 290, 357, 429
L
Label-free proteomics....................................... 112114, 207
Liquid chromatography couples to mass spectrometry
(LC-MS) .................... 28, 71, 72, 140, 141, 143, 144,
48, 149, 160163, 236, 262, 369378, 469,
474, 475
Loading buffer........................................... 73, 148, 216, 220,
227, 278, 279, 282286, 290, 291
SDS-PAGE ........................................................227, 229
Lowry protein assay ..........................................................264
Lymphocytes
density gradient centrifugation ...............................3341
magnetic cell sorting ...............................................3341
sample preparation........................................................34
Lysis buffer .............................................45, 47, 67, 110, 111,
114, 140, 142, 144, 147, 148, 214, 216, 218, 220, 231,
237, 240, 246, 263, 265, 271, 272, 282, 290, 310, 313,
318, 319, 331, 332, 334, 370, 372, 417, 420, 444, 446,
447, 451, 457459, 468, 473
Lysis solution ....................................................................375
Lysosome ............................................................................76
M
Maldi. See Matrix-assisted laser desorption ionization
Mass spectrometry (MS)
electrospray ionization (ESI) ............................. 111, 112,
145, 154, 163, 240, 299, 301, 362, 369
Maldi .................................................. 163, 246, 479495
tandem .........................................111, 145, 155, 254, 474
Matrix-assisted laser desorption ionization
(Maldi) .................... 22, 163, 236, 245, 272, 479495
Matrix-assisted laser desorption ionization
time-of-flight mass spectrometry
(MALDI-TOF-MS) .............................. 82, 485, 490
Membrane proteins ...................159, 401, 402, 419, 428, 493
complexes ...........................................................415424
isolation with FFE ..................................... 416, 417
Microarray ........................................................ 455463, 481
human protein (HuProt) ......................................55463
Microfluidizer........................................................... 2, 15, 16
Microvesicles ..................... 167, 179, 186, 199, 201, 202, 206

isolation ......................................................................202
Mini Bead Beater .................................................................4
Mitochondria ................................................... 13, 58, 7596
isolation ..................................................................7596
Mortar and pestle tissue grinders .........................................9
Mouse brain ......................................277, 294, 295, 297, 468
isolation of synaptosomes ........................... 294, 295, 297
Multiplexing ............................................. 137, 154, 155, 159
Muscle human ..............................................................5563
N
Nanomaterials, MALDI MS....................................479495
N-terminal sequence analysis ...................................249, 254
O
Organelle isolation ........................75, 76, 211, 212, 221, 419
chloroplast, mitochondria ................75, 76, 211, 212, 221
P
Paraffin embedding ..........................................................117
Parr cell disruption bomb .......................................10, 1213
PBS. See Phosphate-buffered saline
Peptides
desalting.......................................................... 70, 71, 246
fractionation by isoelectric focusing ....................369, 373
glycopeptides .......................277283, 289, 291, 355367
phosphopeptides ........................................ 276283, 288,
289, 291, 294, 300, 302, 324, 326, 486, 491, 492, 494
reversed-phase (RP)-chromatography ............... 154, 160,
168, 176, 249, 369, 474
shotgun proteomics .......................70, 118, 207, 324, 369
Peroxisome .........................................................................76
Phenylmethylsulfonyl fluoride (PMSF) ................... 468, 472
protease inhibitor ................................................468, 472
Phosphatase inhibitor ........................................ 26, 147, 278,
296, 297, 310, 318, 319, 345, 444, 446
Phosphate-buffered saline (PBS) .................... 36, 3840, 45,
46, 49, 50, 57, 61, 63, 68, 101, 105, 121, 124, 125,
140, 142144, 147, 169, 170, 174, 188195, 202,
204, 260, 262, 263, 271, 358, 360, 370, 371, 396, 398,
444446, 448, 449, 457, 458, 460
Phosphopeptides
enrichment .......... 276, 278280, 282283, 294, 324, 494
identification............................................... 294, 300, 486
Phosphoproteins
affinity electrophoresis ................................................327
enrichment .................................................................326
fluorescent gel staining ...............................................318
Phosphorylation ........................................ 2227, 31, 56, 58,
114, 147, 275291, 294, 300302, 305, 306,
323351, 442, 443, 447, 448, 456
Phos-tag technology .................................................325326
Plasma proteins .................................................. 99101, 104
PROTEOMIC PROFILING
500 Index
Polytron ........................................................................2, 8, 9
Posttranslational modifications (PTM)
acetylation...........................................................276, 277
glycosylation ....................................... 276, 277, 324, 456
phosphorylation ........... 276, 277, 289, 294, 323, 324, 456
Preadipocytes ..........................................................4450, 52
Protease inhibitors ................ 50, 84, 110, 114, 214, 216, 260,
270, 278, 290, 310, 318, 358, 360, 444, 457, 468, 472
Protein assays
bicinchoninic acid (BCA) assay .................. 310, 313, 444
Bradford assay............................................ 111, 171, 172,
188189, 193, 203, 264, 296, 298, 396, 398, 472
detergent compatible (DC) assay....................... 120, 124,
130, 370, 372
Lowry assay ................................................................264
NanoDrop .......................................... 237, 238, 241246
Protein complexes..............168, 415417, 419421, 423, 424
Protein depletion ......................................................225232
Protein digestion
alkylation ..................... 170, 261, 262, 268, 269, 272, 273
filter aided sample preparation (FASP) ......... 6669, 237,
240241, 473
in-gel digest ........................................118, 119, 129, 236,
238240, 250, 259261, 263, 266270, 325, 331
in-solution digest ....................................... 136, 138, 145,
147, 172174, 237, 241242, 260, 266, 268270
Lys-C .................................................145, 278, 289, 296,
298, 302, 358, 361, 371, 372, 375
OASIS ................. 236, 238240, 278, 290, 358, 361, 362
reduction.....................................170, 173, 206, 236, 237,
239, 241242, 261, 262, 266, 268, 272, 278, 282, 375
STAGE-tip................................................. 197, 361, 362
trypsin.................................................111, 112, 114, 123,
137, 141, 145, 146, 163, 168, 170, 173, 176, 197, 236,
237, 239, 240, 246, 253, 261, 262, 267, 268, 278, 282,
290, 358, 361, 371, 372, 375
Protein extraction .................................... 110112, 117133,
164, 216, 220, 226, 231, 250, 266, 290, 431
Protein fractionation..........................139, 176, 245, 294, 427
Protein interaction ............................................ 442, 443, 452
Protein labeling
dimethyl labeling ........................................ 249257, 290
fluorescent dyes.................................... 79, 133, 154159,
309, 315, 318, 324, 428
stable isotopes ............................................. 155, 160164
(2,4,6-trimethoxyphenyl) phosphonium bromide
(TMPP) labeling ..........................................249257
Protein microarray ............................................ 455, 461, 462
Protein N-termini ....................................................249257
Protein precipitation
acetone ................................................................260, 265
Clean-up kit ................................111, 127, 311, 315, 410
methanol/chloroform......................... 168, 170, 172174,
176, 260, 265266, 473
trichloroacetic acid (TCA)...........218, 226, 260, 265, 266
Protein profiling ............................................... 293302, 481

ProteoMiner
dynamic range compression ..................................99105
high abundance proteins ......................... 60, 99, 100, 201
low abundance proteins ........................................99, 100
R
Reducing agents ...............................105, 119, 120, 124, 130,
318, 348, 483
Reversed-phase high performance liquid chromatography
(RP-HPLC) ......................................... 160, 198, 278
Ribulose 1,5 Bisphosphate Carboxylase/Oxygenase
(RuBisCO)
high abundant protein ................................................225
plant leaves .................................................................226
protamine sulfate ................................................225232
protein depletion ................................................225232
Rotor/stator homogenizers ...............................................79
S
Sample cleanup......................................... 118, 157, 164, 307
Sample preparation .....................................22, 24, 44, 57, 61,
6574, 105, 109, 110, 160, 202, 206, 235237,
257, 259, 278, 280281, 294, 307, 311, 315,
331332, 334, 369378, 384385, 396, 398,
416, 419421, 424, 429433, 443447, 467, 473,
481, 482
Secretome analysis ........................................................46, 59
Shotgun proteomics...............6574, 118, 207, 249, 324, 369
SILAC. See Stable isotope labeling with amino acids in cell
culture
Silver staining ............................................ 61, 154, 156, 219,
220, 232, 266, 272, 313, 317, 318
Solid phase extraction (SPE) ........................... 238, 243, 244,
358, 361, 371, 372, 374, 376
Sonicator
bath sonicator .......................................................68, 398
tip sonicator .................................................. 31, 264, 271
SPE. See Solid phase extraction
SPR. See Surface plasmon resonance
Stable isotope labeling with amino acids in cell culture
(SILAC) ................................135, 160162, 246, 357
Stain-free technology ...............................................381390
Surface plasmon resonance (SPR) ................... 326, 466, 467,
469471, 475
Swiss-Prot ................................................................280, 288
Sypro Ruby ........................ 133, 308, 320, 401, 403, 412, 413
T
TCEP. See Tris(2-carboxyethyl) phosphine
Tissue grinder ...................................................................2, 9
Tissue homogenization ................................ 118, 84, 85, 88
Tris(2-carboxyethyl) phosphine (TCEP)................. 140, 145,
157, 170, 237, 245, 273, 429, 431, 432, 439
PROTEOMIC PROFILING: METHODS AND PROTOCOLS

501
Index
Triton X-114 ....................................................................424
Trypsin .............................................26, 57, 67, 70, 111, 112,
114, 123, 126, 137, 141, 145147, 154, 163, 168, 170,
173, 176, 191, 197, 236, 237, 239243, 246, 253,
260262, 267, 268, 270, 271, 278, 282, 288, 290, 296,
298, 299, 301, 302, 340, 357, 358, 360, 361, 371, 372,
375, 468, 473475, 482, 485, 486, 494
2D fluorescence difference gel electrophoresis
(2D-DIGE) ......................................... 118120, 123,
126129, 428, 429, 431, 433438
2D-PAGE ................22, 23, 28, 118, 119, 129, 130, 428, 429
U
Ultracentrifugation ........................... 167172, 175, 179207
Ultrafiltration ...........................................................181, 182
Urine proteomics ................................................................66
V
Vacuum centrifuge ........................................... 123, 126, 192,
197, 240, 245, 252, 253, 267, 270, 283, 285, 286,
291, 359
W
Western blotting
data normalization ......................................................387
house keeping proteins .......................................381, 382
immunostaining .......................................... 384, 386388
loading control ....................................................381, 382
stain-free technology ..........................................381390
Wheaton Potter-Elvehjem tissue grinder .............................9
Z
Zone electrophoresis (ZE)........................................ 416, 419

Proteomic Profiling - Methods and Protocols

Загружено:

Сведения о документе

Авторское право

Доступные форматы

Поделиться этим документом

Поделиться или встроить документ

Параметры публикации

Этот документ был вам полезен?

Это неприемлемый материал?

Авторское право:

Доступные форматы

Proteomic Profiling - Methods and Protocols

Загружено:

Авторское право:

Доступные форматы

Methods in

Molecular Biology 1295

Anton Posch Editor

For further volumes:

11 Full-Length Protein Extraction Protocols and Gel-Based Downstream

26 In-Gel Peptide IEF Sample Preparation for LC/MS Analysis. . . . . . . . . . . . . .

FANG-KE HUANG Department of Biochemistry and Molecular Pharmacology, Kimmel

KATRIN MARCUS Medizinisches Proteom-Center, Ruhr-Universitt Bochum, Bochum, Germany

in Animal Production, University of Natural Resources and Applied Life Sciences

limited due to the generation of detrimental heat and/or shear

Bead Impactshaking vessel

Mixer Mill [1]

Bead Impactagitator shaft

French Press G-M [6]

Parr vessel [7]

APV Gaulin [8]

Mechanical/Physical Cell Disruption Methods

Bead impact shaking vessel

Bead impact agitator shaft

Rotor/stator shear by spinning shaft

High-pressure batch liquid expansion

High-pressure batch gas expansion

High-pressure flow high velocity shear

High-pressure flow opposed

Droplet low pressure droplet

Ultrasonic shear collapsing bubbles

Electro water separation

Type of biomaterial to be lysed

mechanical methods with at least one example of commercial

Bead Impact Methods: Shaking Vessel

All bead devices open the cells or homogenize tissues by throwing

Fig. 1 Bead impactshaken vesselMini Bead Beater-1 equipment

Fig. 2 Bead impactshaken vesselinside viewbeads moving in jar

materials than can be processed. The equipment is very low cost,

Denaturing of proteins due to high temperature or excessive

Mechanical/Physical Cell Disruption Methods

Detrimental contamination of the batch due to the wear of

Time savings can be achieved by ganging several samples into

A simple way to evaluate the suitability of bead shaking can be

Bead Impact Methods: Stirred Agitated Beads

Fig. 3 Bead impactagitated beadsDyno-Mill with 600 mL chamber

Denaturing of proteins due to high temperature or excessive

Hints for successful temperature control include: (1) Pre-chilling

Detrimental contamination of the batch due to the wear of

Mechanical/Physical Cell Disruption Methods

Fig. 4 Bead impactagitated stirred beadsinside viewspinning agitators

For small diameter cells (e.g. E. coli of 0.251.0 m) use beads

Rotor/Stator homogenizers consist of a rapidly spinning paddle

Fig. 5 RotorstatorShear by spinning shaftPolytron equipment

material is forced between the narrow gaps, there is a cutting

Choice of the rotor/stator geometry depending upon the

Mechanical/Physical Cell Disruption Methods

Fig. 6 Rotorstatorshear by spinning shaftinside view

would be best processed with a generator of 5 mm diameter.

Mortar and Pestle Tissue Grinders: Shear by Mechanical Pressure

High-Pressure Batch: Expanding Fluids

6.1.1 The French

The French Press G-M design consists of a stainless steel cylinder

6.1.2 The Parr Cell

Keeping the samples cold is achieved by pre-chilling of the

6.2.1 French Press G-M

Mechanical/Physical Cell Disruption Methods

Fig. 7 High-pressure batchLiquid expansion French Press G-M and pressure

Fig. 8 High-pressure batchgas expansionParr vessel equipment