![36 Chamrad, Körting, and Blüggel
peptides. Equally, analyzing the sample with differing MS
instrumentation (e.g., MALDI-MS and LC-ESI-MS/MS) will
give a complementary dataset.
Dedicated software is required to combine the outcome of
the database searches, as a combined search with, e.g., different
cleavage rules or mass spectrometric methods is not possible
using currently available sequence database search software. In
ProteinScape (Bruker Daltonik GmbH and Protagen AG), which
is a Proteomics Bioinformatics Platform (14, 15), an algorithm
for this task is integrated. Within ProteinScape, a new protein list
is built, combining all peptides from all searches. Additionally,
only the best matching sequence for each spectrum is annotated.
For complete protein characterization of therapeutic proteins, it
is necessary to show that the amino acid sequence, including
modifications such as glycosylation meets the expected patterns.
Second round searches with tools like ModiroTM
can help to
analyze existing modifications.
In case of LC-ESI data, the level of a specific modification can
be validated by the visualization of Extracted Ion Chromatograms
(EIC) of the modified and unmodified peptide. An EIC shows
the mass spectrometric signal intensity of a specific m/z value
over the retention time. With an overlay of two EICs, showing
the m/z of the modified and the unmodified peptide, the level of
modification can be detected (Figs. 3 and 4). If both signals are
visible, there should be a retention time shift between them.
2.8. Differential EIC
0
20000
40000
60000
80000
100000
120000
140000
39.0 40.0 41.0 42.0 43.0 44.0 45.0 46.0 47.0
intensity
retention time [min]
Fig. 3. Overlay of the extracted ion chromatograms of the unmodified and deamidated
peptide W.LNGKEY.K. The peak at 42.0 min is the unmodified peptide (m/z=723.3672),
the peak at 44.5 min the deamidated peptide (m/z=724.3512). By comparing the peak
intensities or areas a medium deamidation can be estimated. The lower signal at
42.0 min is the second isotope of the unmodified peptide which has nearly the same m/z
as the deamidated peptide.](https://image.slidesharecdn.com/1224231-250609082647-1e751f58/85/Data-Mining-In-Proteomics-From-Standards-To-Applications-1st-Edition-Michael-Hamacher-53-320.jpg)
Data Mining In Proteomics From Standards To Applications 1st Edition Michael Hamacher
- 1. Data Mining In Proteomics From Standards To Applications 1st Edition Michael Hamacher download https://ebookbell.com/product/data-mining-in-proteomics-from- standards-to-applications-1st-edition-michael-hamacher-2448462 Explore and download more ebooks at ebookbell.com
- 2. Here are some recommended products that we believe you will be interested in. You can click the link to download. Data Mining In Proteomics From Standards To Applications 1st Edition Michael Hamacher Martin Eisenacher Christian Stephan https://ebookbell.com/product/data-mining-in-proteomics-from- standards-to-applications-1st-edition-michael-hamacher-martin- eisenacher-christian-stephan-4648230 Fundamentals Of Data Mining In Genomics And Proteomics 1st Edition Daniel Berrar https://ebookbell.com/product/fundamentals-of-data-mining-in-genomics- and-proteomics-1st-edition-daniel-berrar-4287454 Data Mining In The Dark Darknet Intelligence Automation Brian Nafziger https://ebookbell.com/product/data-mining-in-the-dark-darknet- intelligence-automation-brian-nafziger-49473646 Data Mining In Grid Computing Environments 1st Edition Werner Dubitzky Editor https://ebookbell.com/product/data-mining-in-grid-computing- environments-1st-edition-werner-dubitzky-editor-2385300
- 3. Data Mining In Finance Advances In Relational And Hybrid Methods 1st Edition Boris Kovalerchuk https://ebookbell.com/product/data-mining-in-finance-advances-in- relational-and-hybrid-methods-1st-edition-boris-kovalerchuk-4199788 Data Mining In Crystallography 1st Edition Joannis Apostolakis Auth https://ebookbell.com/product/data-mining-in-crystallography-1st- edition-joannis-apostolakis-auth-4205450 Data Mining In Bioinformatics 1st Edition Jason T L Wang Mohammed J Zaki https://ebookbell.com/product/data-mining-in-bioinformatics-1st- edition-jason-t-l-wang-mohammed-j-zaki-4238968 Data Mining In Large Sets Of Complex Data 1st Edition Robson L F Cordeiro https://ebookbell.com/product/data-mining-in-large-sets-of-complex- data-1st-edition-robson-l-f-cordeiro-4241700 Data Mining In Structural Biology Signal Transduction And Beyond 1st Edition Ch Heldin Auth https://ebookbell.com/product/data-mining-in-structural-biology- signal-transduction-and-beyond-1st-edition-ch-heldin-auth-4284820
- 6. Me t h o d s i n Mo l e c u l a r Bi o l o g y ™ Series Editor John M. Walker School of Life Sciences University of Hertfordshire Hatfield, Hertfordshire, AL10 9AB, UK For other titles published in this series, go to www.springer.com/series/7651
- 8. Data Mining in Proteomics From Standards to Applications Edited by Michael Hamacher LeadDiscoveryCenterGmbH,Dortmund,Germany Martin Eisenacher and Christian Stephan MedizinischesProteom-Center,Ruhr-UniversitätBochum,Bochum,Germany
- 9. ISSN 1064-3745 e-ISSN 1940-6029 ISBN 978-1-60761-986-4 e-ISBN 978-1-60761-987-1 DOI 10.1007 /978-1-60761-987-1 Springer New York Dordrecht Heidelberg London © Springer Science+Business Media, LLC 2011 All rights reserved. This work may not be translated or copied in whole or in part without the written permission of the publisher (Humana Press, c/o Springer Science+Business Media, LLC, 233 Spring Street, New York, NY 10013, USA), except for brief excerpts in connection with reviews or scholarly analysis. Use in connection with any form of information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed is forbidden. The use in this publication of trade names, trademarks, service marks, and similar terms, even if they are not identified as such, is not to be taken as an expression of opinion as to whether or not they are subject to proprietary rights. While the advice and information in this book are believed to be true and accurate at the date of going to press, neither the authors nor the editors nor the publisher can accept any legal responsibility for any errors or omissions that may be made. The publisher makes no warranty, express or implied, with respect to the material contained herein. Printed on acid-free paper Humana Press is part of Springer Science+Business Media (www.springer.com) Editors Dr. Michael Hamacher Lead Discovery Center GmbH Dortmund Germany hamacher@lead-discovery.de Dr. Martin Eisenacher Medizinisches Proteom-Center Ruhr-Universität Bochum Bochum Germany martin.eisenacher@rub.de Dr. Christian Stephan Medizinisches Proteom-Center Ruhr-Universität Bochum Bochum Germany christian.stephan@rub.de
- 10. v Preface Inspired by the enormous impact of Genomics and the hopes that came along with it, biochemistry and its methods slowly evolved into what is now widely known as Proteomics. Scientists dedicated to mass spectrometry and gel-based technologies became aware of the powerful tools they hold in hand, dreaming of the quantitative analyses of proteins in cells, tissues, and diseases. Thus, Proteomics soon went from a shooting-star in the life science field to a must-have in each larger wet-lab group. Methods and technology developed rapidly, often much faster than the awareness of the special needs of the tools in use and even faster than standard protocols and standard formats could mature. Soon proteomics techniques created more and more data, while meaningful approaches for data handling, interpretation, and exchange sometimes were clearly behind, resulting in misinterpreted studies and frustrated colleagues from time to time. However, the know-how generated and experiences made especially in the last several years caused a rethinking of strategy design and data interpretation. Moreover, the elabo- ration of standards by such voluntarily driven groups as Proteomics Standards Initiative within the Human Proteome Organisation or the US institutions, Institute of Systems Biology (ISB), and National Institute of Standards and Technology (NIST), ushered in a new era of understanding and quality, proving how powerful Proteomics is when the tech- nology can be controlled through data generation, handling, and mining. This book reflects these new insights within the Proteomics community, taking the historical evolution as well as the most important international standardization projects into account so that the reader gets a feeling for the dynamism and openness in this field. Basic and sophisticated overviews are given in regard to proteomics technologies, stan- dard data formats, and databases – both local laboratory databases and public repositories. There are chapters dealing with detailed information concerning data interpretation strat- egies, including statistics, spectra interpretation, and analysis environments. Other chap- ters describe the HUPO initiatives or are about more specialized tasks, such as data annotation, peak picking, phosphoproteomics, spectrum libraries, LC/MS imaging, and splice isoforms. This volume also includes in-depth description of tools for data mining and visualization of Proteomics data, leading to modeling and Systems Biology approaches. To look beyond the Proteomics tasks and challenges, some chapters present insights into protein interaction network evolution, text mining, and random matrix approaches. All in all, we believe that this book is a well-balanced compendium for beginners and experts, offering a broad scope of data mining topics but always focusing on the current state-of-the-art and beyond. Enjoy! Dortmund, Germany Michael Hamacher Bochum, Germany Martin Eisenacher Bochum, Germany Christian Stephan
- 12. vii Contents Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix Part I Data Generation and Result Finding 1 Instruments and Methods in Proteomics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Caroline May, Frederic Brosseron, Piotr Chartowski, Cornelia Schumbrutzki, Bodo Schoenebeck, and Katrin Marcus 2 In-Depth Protein Characterization by Mass Spectrometry . . . . . . . . . . . . . . . . . . 27 Daniel Chamrad, Gerhard Körting, and Martin Blüggel 3 Analysis of Phosphoproteomics Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41 Christoph Schaab Part II Databases 4 The Origin and Early Reception of Sequence Databases . . . . . . . . . . . . . . . . . . . . 61 Joel B. Hagen 5 Laboratory Data and Sample Management for Proteomics . . . . . . . . . . . . . . . . . . 79 Jari Häkkinen and Fredrik Levander 6 PRIDE and “Database on Demand” as Valuable Tools for Computational Proteomics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93 Juan Antonio Vizcaíno, Florian Reisinger, Richard Côté, and Lennart Martens 7 Analysing Proteomics Identifications in the Context of Functional and Structural Protein Annotation: Integrating Annotation Using PICR, DAS, and BioMart . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107 Philip Jones 8 Tranche Distributed Repository and ProteomeCommons.org . . . . . . . . . . . . . . . 123 Bryan E. Smith, James A. Hill, Mark A. Gjukich, and Philip C. Andrews Part III Standards 9 Data Standardization by the HUPO-PSI: How has the Community Benefitted? . . 149 Sandra Orchard and Henning Hermjakob 10 mzIdentML: An Open Community-Built Standard Format for the Results of Proteomics Spectrum Identification Algorithms . . . . . . . . . . . . 161 Martin Eisenacher 11 Spectra, Chromatograms, Metadata: mzML-The Standard Data Format for Mass Spectrometer Output . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 179 Michael Turewicz and Eric W. Deutsch
- 13. viii Contents 12 imzML: Imaging Mass Spectrometry Markup Language: A Common Data Format for Mass Spectrometry Imaging . . . . . . . . . . . . . . . . . . . . . . . . . . . . 205 Andreas Römpp, Thorsten Schramm, Alfons Hester, Ivo Klinkert, Jean-Pierre Both, Ron M.A. Heeren, Markus Stöckli, and Bernhard Spengler 13 Tandem Mass Spectrometry Spectral Libraries and Library Searching . . . . . . . . . . 225 Eric W. Deutsch Part IV Processing and Interpretation of Data 14 Inter-Lab Proteomics: Data Mining in Collaborative Projects on the Basis of the HUPO Brain Proteome Project’s Pilot Studies . . . . . . . . . . . . 235 Michael Hamacher, Bernd Gröttrup, Martin Eisenacher, Katrin Marcus, Young Mok Park, Helmut E. Meyer, Kyung-Hoon Kwon, and Christian Stephan 15 Data Management and Data Integration in the HUPO Plasma Proteome Project . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 247 Gilbert S. Omenn 16 Statistics in Experimental Design, Preprocessing, and Analysis of Proteomics Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 259 Klaus Jung 17 The Evolution of Protein Interaction Networks . . . . . . . . . . . . . . . . . . . . . . . . . . 273 Andreas Schüler and Erich Bornberg-Bauer 18 Cytoscape: Software for Visualization and Analysis of Biological Networks . . . . . . 291 Michael Kohl, Sebastian Wiese, and Bettina Warscheid 19 Text Mining for Systems Modeling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 305 Axel Kowald and Sebastian Schmeier 20 Identification of Alternatively Spliced Transcripts Using a Proteomic Informatics Approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 319 Rajasree Menon and Gilbert S. Omenn 21 Distributions of Ion Series in ETD and CID Spectra: Making a Comparison . . . . 327 Sarah R. Hart, King Wai Lau, Simon J. Gaskell, and Simon J. Hubbard Part V Tools 22 Evaluation of Peak-Picking Algorithms for Protein Mass Spectrometry . . . . . . . . . 341 Chris Bauer, Rainer Cramer, and Johannes Schuchhardt 23 OpenMS and TOPP: Open Source Software for LC-MS Data Analysis . . . . . . . . . 353 Andreas Bertsch, Clemens Gröpl, Knut Reinert, and Oliver Kohlbacher 24 LC/MS Data Processing for Label-Free Quantitative Analysis . . . . . . . . . . . . . . . 369 Patricia M. Palagi, Markus Müller, Daniel Walther, and Frédérique Lisacek Part VI Modelling and Systems Biology 25 Spectral Properties of Correlation Matrices – Towards Enhanced Spectral Clustering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 381 Daniel Fulger and Enrico Scalas 26 Standards, Databases, and Modeling Tools in Systems Biology . . . . . . . . . . . . . . . 413 Michael Kohl 27 Modeling of Cellular Processes: Methods, Data, and Requirements . . . . . . . . . . . 429 Thomas Millat, Olaf Wolkenhauer, Ralf-Jörg Fischer, and Hubert Bahl Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 449
- 14. ix Contributors Philip C. Andrews • Departments of Biological Chemistry, Bioinformatics and Chemistry, University of Michigan, Ann Arbor, MI, USA Hubert Bahl • Division of Microbiology, Institute of Biological Sciences, University of Rostock, Rostock, Germany Chris Bauer • MicroDiscovery GmbH, Berlin, Germany Andreas Bertsch • Division for Simulation of Biological Systems, WSI/ZBIT, Eberhard-Karls-Universität Tübingen, Tübingen, Germany Martin Blüggel • Protagen AG, Dortmund, Germany Erich Bornberg-Bauer • Bioinformatics Division, Institute for Evolution and Biodiversity, School of Biological Sciences, University of Muenster, Münster, Germany Jean-Pierre Both • Commissariat à l’Énergie Atomique, Saclay, France Frederic Brosseron • Department of Functional Proteomics, Medizinisches Proteom- Center, Ruhr-Universität Bochum, Bochum, Germany Daniel Chamrad • Protagen AG, Dortmund, Germany Piotr Chartowski • Department of Functional Proteomics, Medizinisches Proteom- Center, Ruhr-Universität Bochum, Bochum, Germany Richard Côté • European Molecular Biology Laboratory, European Bioinformatics Institute, Cambridge, UK Rainer Cramer • The BioCentre and Department of Chemistry, The University of Reading, Whiteknights, Reading, UK Eric W. Deutsch • Institute for Systems Biology, Seattle, WA, USA Martin Eisenacher • Medizinisches Proteom-Center, Ruhr-Universität Bochum, Bochum, Germany Ralf-Jörg Fischer • Division of Microbiology, Institute of Biological Sciences, University of Rostock, Rostock, Germany Daniel Fulger • Department of Chemistry and WZMW, Computer Simulation Group, Philipps-University Marburg, Marburg, Germany Complex Systems Lagrange Lab, Institute for Scientific Interchange, Torino, Italy Simon J. Gaskell • Michael Barber Centre for Mass Spectrometry, School of Chemistry, Manchester Interdisciplinary Biocentre, University of Manchester, Manchester, UK Mark A. Gjukich • Departments of Biological Chemistry, Bioinformatics and Chemistry, University of Michigan, Ann Arbor, MI, USA Clemens Gröpl • Division for Simulation of Biological Systems, WSI/ZBIT, Eberhard-Karls-Universität Tübingen, Tübingen, Germany Bernd Gröttrup • Medizinisches Proteom-Center, Ruhr-Universität Bochum, Bochum, Germany Joel B. Hagen • Department of Biology, Radford University, Radford, VA, USA
- 15. x Contributors Jari Häkkinen • Department of Oncology, Clinical Sciences, Lund University, Lund, Sweden Michael Hamacher • Lead Discovery Center GmbH, Dortmund, Germany Sarah R. Hart • Michael Barber Centre for Mass Spectrometry, School of Chemistry, Manchester Interdisciplinary Biocentre, University of Manchester, Manchester, UK Institute for Science and Technology in Medicine/School of Medicine, Keele University, Staffordshire, UK Ron M. A. Heeren • FOM Institute for Atomic and Molecular Physics, Amsterdam, The Netherlands Henning Hermjakob • European Molecular Biology Laboratory, European Bioinformatics Institute, Cambridge, UK Alfons Hester • Justus Liebig University, Giessen, Germany James A. Hill • Departments of Biological Chemistry, Bioinformatics and Chemistry, University of Michigan, Ann Arbor, MI, USA Simon J. Hubbard • Faculty of Life Sciences, University of Manchester, Manchester, UK Philip Jones • European Molecular Biology Laboratory, European Bioinformatics Institute, Cambridge, UK Klaus Jung • Department of Medical Statistics, Georg-August-University Göttingen, Göttingen, Germany Ivo Klinkert • FOM Institute for Atomic and Molecular Physics, Amsterdam, The Netherlands Michael Kohl • Medizinisches Proteom-Center, Ruhr-Universität Bochum, Bochum, Germany Oliver Kohlbacher • Division for Simulation of Biological Systems, WSI/ZBIT, Eberhard-Karls-Universität Tübingen, Tübingen, Germany Gerhard Körting • Protagen AG, Dortmund, Germany Axel Kowald • Protagen AG, Dortmund, Germany Kyung-Hoon Kwon • Korea Basic Science Institute, Deajeon, Republic of Korea King Wai Lau • Faculty of Life Sciences, Michael Barber Centre for Mass Spectrometry, School of Chemistry, Manchester Interdisciplinary Biocentre, University of Manchester, Manchester, UK Fredrik Levander • Department of Immunotechnology and CREATE Health Strategic Centre for Translational Cancer Research, Lund University, Lund, Sweden Frédérique Lisacek • Proteome Informatics Group, Swiss Institute of Bioinformatics, Geneva, Switzerland Katrin Marcus • Department of Functional Proteomics, Medizinisches Proteom- Center, Ruhr-Universität Bochum, Bochum, Germany Lennart Martens • European Molecular Biology Laboratory, European Bioinformatics Institute, Cambridge, UK Caroline May • Department of Functional Proteomics, Medizinisches Proteom- Center, Ruhr-Universität Bochum, Bochum, Germany Rajasree Menon • Center for Computational Medicine and Biology and National Center for Integrative Biomedical Informatics, University of Michigan,
- 16. xi Contributors Ann Arbor, MI, USA Helmut E. Meyer • Medizinisches Proteom-Center, Ruhr-Universität Bochum, Bochum, Germany Thomas Millat • Systems Biology & Bioinformatics, Institute of Computer Science, University of Rostock, Rostock, Germany Markus Müller • Proteome Informatics Group, Swiss Institute of Bioinformatics, Geneva, Switzerland Gilbert S. Omenn • Departments of Medicine and Genetics, Center for Computational Medicine and Bioinformatics, Medical School and School of Public Health, University of Michigan, Ann Arbor, MI, USA Sandra Orchard • European Molecular Biology Laboratory, European Bioinformatics Institute, Cambridge, UK Patricia M. Palagi • Proteome Informatics Group, Swiss Institute of Bioinformatics, Geneva, Switzerland Young Mok Park • Korea Basic Science Institute, Daejeon, Republic of Korea Knut Reinert • Division for Simulation of Biological Systems, WSI/ZBIT, Eberhard- Karls-Universität Tübingen, Tübingen, Germany Florian Reisinger • European Molecular Biology Laboratory, European Bioinformatics Institute, Cambridge, UK Andreas Römpp • Justus Liebig University, Giessen, Germany Enrico Scalas • Department of Advanced Sciences and Technology, Laboratory on Complex Systems, University of East Piedmont Amedeo Avogadro, Alessandria, Italy Christoph Schaab • Kinaxo Biotechnologies GmbH, Martinsried, Germany Max Planck Institute of Biochemistry, Martinsried, Germany Sebastian Schmeier • South African National Bioinformatics Institute, University of the Western Cape, Bellville, South Africa Bodo Schoenebeck • Department of Functional Proteomics, Medizinisches Proteom-Center, Ruhr-Universität Bochum, Bochum, Germany Thorsten Schramm • Justus Liebig University, Giessen, Germany Johannes Schuchhardt • MicroDiscovery GmbH, Berlin, Germany Andreas Schüler • Bioinformatics Division, School of Biological Sciences, Institute for Evolution and Biodiversity, University of Muenster, Münster, Germany Cornelia Schumbrutzki • Department of Functional Proteomics, Medizinisches Proteom-Center, Ruhr-Universität Bochum, Bochum, Germany Bryan E. Smith • Departments of Biological Chemistry, Bioinformatics and Chemistry, University of Michigan, Ann Arbor, MI, USA Bernhard Spengler • Justus Liebig University, Giessen, Germany Christian Stephan • Medizinisches Proteom-Center, Ruhr-Universität Bochum, Bochum, Germany Markus Stöckli • Novartis Institutes for BioMedical Research, Basel, Switzerland Michael Turewicz • Medizinisches Proteom-Center, Ruhr-Universität Bochum, Bochum, Germany Juan Antonio Vizcaíno • European Molecular Biology Laboratory, European Bioinformatics Institute, Cambridge, UK Daniel Walther • Proteome Informatics Group, Swiss Institute of Bioinformatics,
- 17. xii Contributors Geneva, Switzerland Bettina Warscheid • Clinical & Cellular Proteomics, Medical Faculty and Center for Medical Biotechnology, Duisburg-Essen University, Essen, Germany Medizinisches Proteom-Center, Ruhr-Universität Bochum, Bochum, Germany Sebastian Wiese • Medizinisches Proteom-Center, Ruhr-Universität Bochum, Bochum, Germany Olaf Wolkenhauer • Systems Biology & Bioinformatics, Institute of Computer Science, University of Rostock, Rostock, Germany
- 18. Part I Data Generation and Result Finding
- 20. 3 Michael Hamacher et al. (eds.), Data Mining in Proteomics: From Standards to Applications, Methods in Molecular Biology, vol. 696, DOI 10.1007/978-1-60761-987-1_1, © Springer Science+Business Media, LLC 2011 Chapter 1 Instruments and Methods in Proteomics Caroline May, Frederic Brosseron, Piotr Chartowski, Cornelia Schumbrutzki, Bodo Schoenebeck, and Katrin Marcus Abstract In the past decade, major developments in instrumentation and methodology have been achieved in proteomics. For proteome investigations of complex biological samples derived from cell cultures, tis- sues, or whole organisms, several techniques are state of the art. Especially, many improvements have been undertaken to quantify differences in protein expression between samples from, e.g., treated vs. untreated cells and healthy vs. control patients. In this review, we give a brief insight into the main tech- niques, including gel-based protein separation techniques, and the growing field of mass spectrometry. The proteome describes the quantitative expression of genes within, e.g., a cell, a tissue, or body fluid at specific time points and under defined circumstances (1). In contrast to the genome, the proteome is highly dynamic and the protein expression pat- tern of cells in an organism varies depending on the physiological functions, differentiation status, and environmental factors. In addition, alternative splicing of mRNAs and a broad range of posttranslational modifications (e.g., phosphorylation, glycosyla- tion, and ubiquitination) increase proteome complexity (2, 3). Transcription analysis also does not allow insight into degradation and transport phenomena, alternative splicing, or posttransla- tional modifications. Furthermore, mRNA and protein levels often do not correlate (4, 5). All these influences are unconsid- ered in genome analysis and underline the importance of pro- teome analysis to obtain deeper insights into cellular functions. In general, proteome analysis provides a snap-shot of proteins expressed in a cell or tissue at a defined time point (1). Indeed, not only qualitative analysis resulting in a defined “protein inventory” 1. Introduction
- 21. 4 May et al. can be obtained, but differential proteome analysis also allows for the detection of distinct differences in protein expression. This is of implicit interest, e.g., in the fields of fundamental and clinical research in order to understand main cellular functions and physi- ological/pathophysiological processes. For proteome investiga- tion of complex biological samples derived from cell cultures, tissues, or whole organisms, several techniques were developed over the last decade, the most important of which are reviewed in the following paragraphs. Figure 1 gives a general overview of different workflows in proteomics. Fig. 1. General workflow for proteomics. Several different methods and technologies exist today which can be combined in order to achieve best results for a given scientific question. Most commonly used techniques and strategies are pre- sented in the following chapters. MS mass spectrometry; 1D-PAGE one-dimensional protein separation; 2D-PAGE two- dimensional protein separation; 2D-DIGE two-dimensional difference in gel electrophoresis.
- 22. 5 Instruments and Methods in Proteomics Gel-based approaches belong to the most frequently used assays in proteomics to separate proteins and to analyze them qualita- tively and quantitatively. For simple pre-separation of complex protein mixtures before mass spectrometric analysis, one- dimensional polyacrylamide gel electrophoresis (1D-PAGE) is often used. Additionally, two-dimensional approaches such as two-dimensional polyacrylamide gel electrophoresis (2D-PAGE) allow for the separation of up to 10,000 protein species (6), pro- viding the potential for global differential proteome analysis. Different gel-based methods especially differing in their respective resolution and application in proteomics are summarized in the following sections. One-dimensional polyacrylamide gel electrophoresis, according to Lämmli, with sodium dodecyl sulfate (SDS) as negative-charge detergent (7) is widely used for the separation of proteins accord- ing to their electrophoretic mobility. Due to SDS binding, the proteins are denaturated showing identical charge per unit pro- tein mass which after the application of an electric field results in fractionation by size (see Fig. 2). High mass proteins will be retained longer by the polyacrylamide network than smaller pro- teins. After visualization by one of several existing staining meth- ods, protein identification can easily be performed by mass spectrometry (MS) (see Subheading 3). The resolution of 1D-PAGE in contrast to that of 2D-PAGE is (see Subheading 2.2) rather low since the proteins are separated only according to their molecular mass. Nevertheless, 1D-PAGE is often used to achieve a pre-separation prior to MS or for the detection of proteins by subsequent Western blotting. Two-dimensional polyacrylamide gel electrophoresis was devel- oped in order to obtain higher resolved protein patterns than obtained using 1D-PAGE, offering a huge potential to give a comprehensive overview of the proteins present in the examined system. 2D-PAGE is a combination of two orthogonal separation techniques: in the first dimension, the proteins are separated according to their isoelectric point (Isoelectric Focusing: IEF), followed by a conventional SDS-PAGE in the second dimension. For IEF, two different techniques are described, namely, the car- rier-ampholyte (CA)-based (8, 9) and immobilized pH gradient (IPG) system (10, 11). The spot pattern can be visualized with several protein staining methods, which differ in sensitivity and dynamic range. For differential proteome analysis, spot patterns of related gels are compared with each other and protein species can be relatively quantified automatically using one of several 2. Gel-Based Protein Separation Techniques and Applications 2.1. One-Dimensional Protein Separation: 1D-PAGE 2.2. Two-Dimensional Protein Separation: 2D-PAGE
- 23. 6 May et al. available image analysis software tools (12). Differentially expressed proteins of interest are subsequently identified by MS (see Subheading 3). One drawback of 2D-PAGE is the fact that mainly hydrophilic proteins with a molecular weight of 5–150 kDa in a pH range of 3.5–10 can be analyzed. Especially hydrophobic/membrane proteins are underrepresented and must be analyzed with alternative gel-based methods such as 2D-benzyldimethyl-n-hexadecylammonium chloride (BAC)/SDS (13), 2D-cetyltrimethylammonium bromide (CTAB)/SDS (14, 15), SDS/SDS (16), and BlueNative-PAGE (17), or MS-based strategies (see below). Nevertheless, in combination with image analysis and MS, 2D-PAGE is still the method of choice to analyze complex protein samples. For more detailed description of 2D-PAGE, see Marcus et al. (18) and Rabilloud et al. (19). Fig. 2. 2D-IEF/SDS-PAGE of SH-SY5Ycells. The proteins of an SH-SY5Y cell lysate were separated according to their isoelectric point in the first dimension (isoelectric focusing) and to their electrophoretic mobility in the second dimension (SDS PAGE).After 2D-PAGE, protein spots were visualized with silver staining.
- 24. 7 Instruments and Methods in Proteomics The invention of two-dimensional difference in-gel electrophoresis (2D-DIGE) in 1997 drastically improved the technical reproduc- ibility of 2D-PAGE and the accurate quantification of different proteins in samples with high statistical significance (18, 20). Proteins of different samples are covalently labeled with spectrally resolvable fluorescent dyes (CyDyes™, GE Healthcare Europe GmbH) and afterwards separated simultaneously on the same gel. The application of an internal standard, optimally consisting of a mixture of all samples included in the study, allows accurate matching and normalization of protein spots in all gels, and with this highly accurate quantification (21). Two methodologies can be distinguished: CyDye™ minimal labeling and CyDye™ satura- tion labeling. For minimal labeling, dyes react with the e-amino group of lysine residues. Three to five percent of all proteins and only one lysine per protein on average are labeled. Three different dyes are available: Cy™2, Cy™3, and Cy™5. Saturation labeling allows for the analysis of scarce protein samples down to an amount of 3 mg per gel (15, 22). The label reacts with thiol groups of cysteine residues. All cysteine residues of all proteins are labeled. In this technique, two different dyes are available, Cy™3 and Cy™5. Protein patterns are digitalized using confocal fluorescent imagers, resulting in a gel image at a specific wavelength for each dye without any crosstalk. Appropriate analysis software allows for automated detection, background subtraction, quantification, normalization, and inter-gel matching. Similar to gel-based protein separation, MS is one of the most popular techniques in proteomics (23–25). In MS, the chemical compounds of a sample are ionized and the resulting charged molecules (ions) are analyzed according to their mass-to-charge (m/z) ratios. In proteomics, the molecules of interest are either proteins or peptides obtained from enzymatic digestion of pro- teins. MS can be used for the identification of either the peptides or the proteins, as well as for the quantification of the measured ion species. Up to date, several different MS setups and assays have been developed for use in proteome studies. Each of them has its own advantages and disadvantages, and is used for charac- teristic purposes, comprising identification of proteins from 2D-gel spots, description of peptides with chemical modifica- tions, and quantitative MS assays (18, 26–29). The following chapters illustrate the most important aspects of MS in proteomics and their characteristic applications. 2.2.1. 2D-DIGE: A Sophisticated Application 3. Mass Spectrometry- Based Techniques and Applications
- 25. 8 May et al. In general, a mass spectrometer consists of the following components: ion source, mass analyzer, and detector (30). The ion source is used to create protein or peptide ions usually by transferring posi- tive charged protons (H+ ) onto the molecules. The ionization is called “soft” because the chemical structure of the proteins or peptides remains unharmed. One or more mass analyzers are used to separate the ions by their m/z ratio or to fragment the ions for further sequence analysis. At last, the ions are passed to a detector connected to a PC with appropriate software for data analysis. Modern software tools include control programs for all parts of the mass spectrometer setup. Optional to this setup is the use of a chromatography system (widely HPLC) upstream of the ion source to reduce sample complexity (see Fig. 3). All hardware components are described in more detail in the following chapters. Different types of liquid chromatography (LC) are used in pro- teomics to complement gel-based separation techniques (29). The basic principle of LC is to separate solute analytes (e.g., pro- teins or peptides) in a fluid that flows over solid particles. The solution is referred to as the mobile phase, while the particles are termed the stationary phase. Depending on their differing chemi- cal and physical properties, different analyte species will interact in different ways with both phases. Usually, the stationary phase is packed into a column through which the mobile phase flow is led. This way, the analytes separate over time until they elute from the column. The time point in which a peptide elutes is called its retention time (RT). The amount of analytes eluting over the time is usually documented as a chromatogram by UV detectors. Different variants of LC systems each make use of special proper- ties of the analytes of interest, e.g., polarity or chemical functional groups. It is common to use LC for protein purification or 3.1. Setup of a Mass Spectrometer 3.1.1. Liquid Chromatography Techniques for Proteome Analysis Fig. 3. Setup of mass spectrometers. A typical mass spectrometer for proteomic purposes will be set up in the following way: high-performance liquid chromatography (HPLC) (optional), ion source, mass analyzer, detector, and personal com- puter. See the following chapters for details on hardware configuration.
- 26. 9 Instruments and Methods in Proteomics fractionation as one of the first steps in a proteome study. Nevertheless, peptides are more homogenous in size and polarity than proteins, and are thus better suited for chromatographic separation and analysis. Therefore, LC is a powerful tool to reduce the complexity of peptide samples, e.g., digested protein bands from 1D-gels or whole cell lysates (31). It is also used for the separation of less complex samples, such as 2D-gel spots. A major advantage of LC is the possibility to automate the separation progress. Modern automated systems can cover the whole separation progress, beginning with the loading of the sample onto the column up to the MS analysis of the eluted analytes (mostly peptides). This combination is referred to as LC–MS. Automation allows complex and elongated gradi- ents of mobile phase composition as well as the combination of several columns with different stationary phases in one analysis. An example for such sophisticated LC systems is the multi-dimen- sional protein identification technology (MuDPIT) (32). The peptide solution is separated first by strong cation exchange (SCX) with a pH gradient, followed directly by reversed phase (RP) chromatography using hydrophobic C18 material as the station- ary phase and a polar solution of water with increasing amount of organic compounds (33). MuDPIT runs can be prolonged to 12 or even more hours to increase their separation power. Another advantage of LC is the possibility of nano-size appli- cations with increased sensitivity. In nano-high pressure liquid chromatography (nano-HPLC), the mobile phase is pumped through capillary columns (34). The columns contain porous nonpolar particles serving as a hydrophobic solid phase with which the peptides can interact. The mobile phase is a polar fluid consisting mostly of a mixture of water, organic compounds such as acetonitrile, and low amounts of acids. For this reason, this type of HPLC is referred to as RP-HPLC. Usually, the amount of acetonitrile in the mixture is increased over the time of analysis following an automated gradient. As a result, hydrophilic pep- tides will elute first from the capillary column, followed by other peptides depending on their increasing hydrophobicity. Nano- HPLC is a very common proteomics method because even short runs (between 1 and 3 h) can be used to separate complex sam- ples. Additionally, it is possible to couple the chromatography system either directly (“online”) or indirectly (“offline”) with a mass spectrometer for subsequent MS analysis of the eluting pep- tides. In online LC–MS, the nano-HPLC system is connected directly with an electrospray ionization (ESI) ion source (see Subheading 3.1.2). This is possible because ESI requires liquid samples, which means the solution eluting from the nano-HPLC can be led directly into the ion source. Offline LC–MS establishes the connection between nano-HPLC and matrix-assisted laser desorption ionization (MALDI), which is another common
- 27. 10 May et al. ionization technique that requires samples in solid (crystallized) state (see Subheading 3.1.2). For this purpose, automated fraction- ators spot small amounts of liquid eluting from the nano-HPLC onto steel plates (“targets”) suitable for MALDI ion sources (31). One drawback of offline nano-LC–MALDI–MS in compari- son to online LC–ESI–MS is a longer analysis time. Indeed, spot- ted samples can be stored for some time, allowing for a re-investigation of the samples (for more details, see (29)). In principle, two main ionization methods are used in proteomics today, MALDI and ESI (23). In MALDI, the sample molecules are immobilized by co-crystallization in the presence of organic compounds such as alpha-cyano-4-hydroxycinnamic acid or 2,5-dihydroxybenzoic acid on a metal sample target (35). By administering laser energy to the samples, the matrix ions par- tially transfer their charge on the analyte molecules, producing mainly single-charged peptide ions. Since the pulsed laser operates rather in “shots” than continuously, MALDI is used primarily in combination with time of flight (TOF) analyzers (36). This com- bination is termed as MALDI-TOF, which is used in proteomics for analysis of proteins and peptides (37–39). ESI is another well-suited ionization method for biomole- cules such as peptides (23). Like MALDI, ESI is a “soft” method of ionization producing charged peptides in solution (40). ESI requires liquid samples which are delivered either by direct injec- tion with a syringe or “online” coupled with a (nano)-RP-HPLC system. The sample passes a capillary needle on which voltage is applied. As a result, charged droplets are generated at the capil- lary tip. The solvent partially evaporates, resulting in the reduc- tion of the droplets’ diameter and enhanced density of charges. The rising charge density leads to the so-called coulomb explo- sions which further reduce the diameter of the droplets. Hence, the analytes are dispersed as a fine spray (41, 42). Different mech- anisms have been discussed to describe the ESI process, which all end up with the fact that gas-phase ions are generated (43, 44). The ions are subsequently detected by the mass analyzer. One of the major advantages of ESI for proteomics is the possibility to separate highly complex peptide mixtures upstream by nano- HPLC, e.g., resulting from whole cell lysates. In general, both ionization techniques described above can be combined with different types of mass analyzers. Depending on the application desired, each combination is characterized by typical features such as enhanced mass accuracy, sensitivity, dynamic range, or resolving power. Therefore, for best perfor- mance, mass spectrometer setups favorable for, e.g., identifica- tion, quantification, high throughput analyses, or detection of modifications should differ from each other (for a comprehensive overview, see Domon and Aebersold (36)). 3.1.2. Ionization Methods
- 28. 11 Instruments and Methods in Proteomics Independent of the ionization technique, the molecular masses of free ions are measured in mass analyzers after passing them through a vacuum chamber. Different types of analyzers are often combined in a so-called hybrid mass spectrometer (24, 36). After the ions pass the analysis system, the detector measures the m/z ratios of all incoming ions and transfers this information to a computer. Most common in proteomics are TOF analyzers, dif- ferent types of ion traps, and high-resolution analyzers such as Fourier transform ion cyclotron resonance (FT-ICR) or the latest development, the orbitrap. In TOF analyzers, ions are accelerated by a potential between two electrodes (45). The analyzer itself is merely a vacuum tube. Ions with different masses pass the vacuum chamber with differ- ent velocities. By measuring the time the ions need until they reach the detector, the m/z ratio is calculated. TOF analyzers can reach resolutions of up to 15,000 full-width half-height maxi- mum (fwhm) with a mass accuracy of up to 2 ppm (36, 45, 46). In Q-Q-TOF instruments, two quadrupoles (Qs) are combined with a TOF analyzer. In the MS mode, the quadrupole serves as a guide for the ions toward the mass analyzer. In the MS/MS mode, where detailed peptide information is gained, the precursor ions are selected in the first quadrupole and subsequently frag- mented in the second quadrupole. This setup results in a high mass accuracy and high resolution of selected precursor ions (36). In a quadrupole (Q) analyzer, ions accelerated by strong elec- tric fields pass a set of stab electrodes arranged in cylindrical con- stellation (47, 48). Between the stab electrodes, an alternating electric field ensures that only ions of a defined mass can pass. In this way, the quadrupole acts as a mass filter. Furthermore, ions can be trapped in the electric fields for fragmentation. Quadrupoles are most common as parts of hybrid instruments, e.g., for focusing of the ion beam emitted from the ion source on the way to another mass analyzer with better resolution, like an orbitrap (49, 50). In addition, combinations of quadrupoles with TOF analyzers or as parts of FT-ICR mass spectrometers occur. Triple quadrupole (Q-Q-Q) instruments became more and more important in pro- teomics research. With the arrangement of three quadrupoles or two quadrupoles followed by a linear ion trap (LIT), new scan- ning methods such as product ion scanning, parent ion scanning (51, 52), neutral loss scanning (53, 54), and multiple reaction monitoring (55) (see Subheading 3.2) became feasible. All these scanning methods commonly use concomitant mass analyzers serving as a combination of mass filters and collision cells to enhance the sensitivity of a subset of ions one aims to analyze. In “ion trap” (IT) analyzers, ions are trapped and get accu- mulated over a given time in a physical device. Nonlinear ITs were first described by Paul et al. (56). The IT itself consists of two adversely arranged hyperbolic electrodes with a ring electrode 3.1.3. Types of Mass Analyzers and Hybrid Mass Spectrometers
- 29. 12 May et al. between them. This setup is used to establish dynamic electric fields in all three dimensions, which allows focusing of incoming ions in the center of the trap. From this point on, the ions can be selectively ejected and passed to the detector, or can be fragmented. This is usually done by collision-induced dissociation (CID) and/ or electron transfer dissociation (ETD) (see Subheading 3.2), combined with the activation of the ions induced by resonance to the changing electric fields (57). A detailed description of theory, instrumentation, and working modes can be found in ref. (58–62). Linear ion traps function as mass filters and simultaneously act as a storage device for specific ions. Ions that possess a defined m/z range can be trapped and stored before they are further passed through the detector. This is conducted by four electrode rods in a quadrupolar orientation describing a combination of alternating and co-current flows. Ions that reside within the adjusted m/z range oscillate through the drifting channel, whereas all other ions describe unstable flight paths and, there- fore, get stopped by collision with the electrodes. During the scanning of the mass field, both co-current (U) and alternating current (V) are simultaneously enhanced. With the change of this U/V ratio during the scan, the mass range of stable oscillation becomes shifted, resulting in a mass separation (49). LITs have the advantage of increased ion storage capacity compared to non- linear ion traps, leading to a higher sensitivity and dynamic range. In general, IT technology is characterized by MS/MS capabilities with unmatched sensitivity and fast data acquisition. Indeed, lim- ited resolution, low-ion trapping capacities, and space-charging effects result in low accuracy of the mass measurements. Fourier transform ion cyclotron resonance mass spectrometers are ITs with an additional homogeneous magnetic field (63, 64). The magnetic field forces ions into a circular path in which they cycle with high frequency, the so-called cyclotron circle frequency. By adding a changing electric field perpendicular to the magnetic field, a resonance between the ion mass and the cyclotron circle frequency is built up. In this process, energy is consumed from the changing electric field. This energy shift can be measured and transformed into m/z ratios by Fourier transformation. FT-ICR spectrometers reach high-resolution mass accuracy of up to 1.0 ppm (65). Nevertheless, FT-ICR spectrometers are less com- mon than other types because of their high operation expenses. The last important development in the field of mass analyzers was attained by the Orbitrap (66, 67). This type consists of a single, spindle-shaped electrode. In this setup, ions move on cir- cuits around the electrode and oscillate along the axis at the same time. The frequency of this oscillation is dependent on the masses and charges of the respective ions. On this basis, m/z can be calculated by Fourier transformation. Orbitrap analyzers reach
- 30. 13 Instruments and Methods in Proteomics resolutions and accuracies similar to those of FT-ICR analyzers combined with significantly lower operation expenses. For this reason, Orbitrap instruments become increasingly popular in pro- teome analysis (68). Mass spectrometry can be used for whole protein mass and pep- tide mass determination as well as peptide fragmentation analysis. Peptide fragmentation analysis became the most popular applica- tion over the years as it allows obtaining information not only about the mass and charge of a protein or peptide ion, but also on its chemical composition. Different main scanning methods suit- able for peptide mass and peptide fragmentation analysis can be distinguished, which are peptide mass fingerprinting (PMF) (69), post-source decay (PSD) (70), tandem-MS (also called MS/MS or MS²), product ion scanning (24, 36, 71), neutral loss (NL) scanning (53, 54), precursor ion scanning (PIS) (52, 72, 73), and multiple reaction monitoring (MRM) (36, 55, 74). Peptide mass fingerprinting or peptide mass mapping is based on the fact that digestion of a protein by enzymes will result in a specific mixture of peptides. When analyzed with a mass spectrom- eter, the peptide mixture will lead to a characteristic pattern of m/z values, the PMF. By comparing the PMF with databases, it is pos- sible to identify the corresponding protein (75). This makes PMF ideally suitable for the identification of proteins from low complex mixtures, e.g., 2D gel spots using MALDI-TOF MS (24). If the number of peptides for PMF analysis is not sufficient or the complete genome sequence of the analyzed species is unknown, fragmentation analysis can be performed for a more detailed and specific analysis. PSD, tandem-MS (MS/MS, MS2 ): The fragmentation of the peptide can be induced by metastable decay (PSD) (70), CID (76), or ETD (57). CID is an older, but still common technique that uses neutral gas molecules such as helium, nitrogen, or argon to transfer kinetic energy on the peptide ions, leading to fragmentation. In ETD, this is achieved by using fluoranthene radicals as electron donors that destabilize peptide ions by transferring the electron on them. ETD leads to different fragments than CID (see spectra inter- pretation). While CID is still the state of the art, especially for sequencing of peptide ions, ETD and combinations of both meth- ods have become important when analyzing posttranslational modi- fications such as phosphorylation or glycosylation (77–80) PSD analysis is restricted to MALDI-TOF/(TOF) instruments, whereas tandem-MS (MS/MS, MS2 ) analysis can be done on different types of instruments such as ITs, Q-Q-Qs, or Orbitraps. During MS frag- mentation analysis, peptide ions are automatically selected for fragmentation, resulting in predictable breakdown products. These fragment ions are recorded by the detector and give rise to the so- called PSD or tandem-MS (MS/MS, MS2 ) spectra. 3.2. Identification of Proteins by Mass Spectrometry: Scanning Methods and Fragmentation Types
- 31. 14 May et al. To date, the most common applications in proteomics use MS² spectra without further fragmentation for protein identifica- tion. This is due to the fact that generally samples in proteomics are analyzed after digestion of the proteins to peptides, and the resulting MS² spectra are sufficient for identification of the pep- tides. For detailed analyses of fragment ions, especially detection of posttranslational modifications, further fragmentations can be performed, resulting in MSn spectrometry (81, 82). Basically, the next described scanning modes are specialized MS/MS applica- tions for Q-Q-Q instruments which are used to enhance the selec- tivity and sensitivity for the measurement of a subset of ions. Product ion scanning is the most common method for sequenc- ing peptide ions generally on Q-Q-Q instruments (24, 36, 71). This scan determines, in a single experiment, all peptide (parent) m/z ratios that react to produce a selected product (daughter) ion m/z ratio. In Q-Q-Qs, one peptide of a specified m/z is selected in Q1 as a parent ion. In the next step, the parent ion is frag- mented in Q2. All resulting fragment ions are subsequently scanned in Q3. Usually, several parent ions of different m/z ratios are sequentially analyzed by stepwise alteration of the quadrupole field in Q1 in one MS run in this way. New developments in MS instrumentation today allows for product ion scanning with spe- cialized hybrid-TOF such as Q-TOF or TOF-TOF instruments. Converse to the product ion scan, the PIS is a scan that deter- mines, in a single experiment, all the product (daughter) ion m/z ratios that are produced by the reaction of a selected peptide (parent) ion m/z ratio. Parent ions of the whole mass range are transferred through Q1 and fragmented in Q2. Q3 is then fixed on a single fragment ion mass, filtering for pre-specified fragment ions selectively produced by the parent ions (73). This scanning method can be especially useful for the selective detection (and quantification) of posttranslational modifications such as glycosy- lation or phosphorylation (83, 84). Another selective scanning mode especially useful for the detection of protein/peptide phosphorylation or glycosylation is NL scanning verifying the loss of a neutral particle from a frag- mented parent ion (24, 85). Similar to PIS, in NL scanning, par- ent ions of the whole mass range are transferred through Q1 and fragmented in Q2. Q3 is not fixed on a special fragment mass but operates synchronously to Q1 scanning for a defined mass shift between precursor and fragment ion. In other words, only frag- ment ions that differ from their parent ion by a characteristic mass difference will reach the detector. Because the charge of the pep- tide ion does not change, this was designated as a neutral loss. NL scanning and PIS can be combined with product ion scanning for sequencing of the modified peptide ions. Multiplereactionmonitoringisonespecialapplicationinproteome analysis allowing for the targeted detection (and quantification) of
- 32. 15 Instruments and Methods in Proteomics pre-selected peptides in a complex peptide mixture. MRM analysis can be performed on Q-Q-Qs as well as on Q-hybrids such as Q-Q–LIT instruments (74, 86). In MRM (or single/selected reaction monitoring, SRM), Q1 serves as a mass filter for the selection of ions of a defined m/z ratio (Q1). Selected parent ions are fragmented in Q2 and pre-defined fragment ions are specifi- cally detected in Q3. The combination of pre-defined m/z ratios in Q1 and Q3, representing the precursor and a characteristic fragment ion, is called an MRM transition. Thus, MRM differs from the other scan types in the way that two pre-requisites have to be fulfilled in order to produce a signal in the detector: both ions, precursor and related fragment ion, need to be specifically measured in one scan. This makes the MRM scan highly specific even for low abundant peptide ions in complex mixtures. MRM can be used for all kinds of hypothesis-driven approaches where a specified protein/peptide of interest should be identified or even quantified (relatively or absolutely), e.g., in a complex protein mixture (87). All kinds of MS and MS/MS analyses result in the generation of the so-called raw data. These raw data containing information about the peptide masses and, in case of MS/MS data, also frag- ment ion masses and their intensities are transformed to a “peak list.” Identification of the peptide/protein is performed by using a search engine such as MASCOT (88) or Sequest (89) to search the peak list against a database of proteins “digested in silico,” meaning that the practically obtained MS and MS/MS data are directly compared with theoretically generated data from protein databases. Knowledge about sample preparation and separation conditions, type and mass accuracy of the mass spectrometer, and mode of peptide fragmentation (90) allows for a reliable peptide assignment (88, 89, 91). Typically, the algorithms give a probabil- ity value for the correctness of the identification. The peptides assigned should be unique for a protein species in order to annotate the analyzed spectrum clearly to only one protein. This kind of data analysis is possible only in cases where the genome of the investi- gated organism is sequenced and a database is available. Otherwise, de novo sequence analysis needs to be performed entailing manual interpretation and annotation of the MS/MS spectra in order to obtain sufficient information on the peptides’ sequence. Due to the described disadvantages of gel-based differential proteome analysis (see Subheading 2.2), over the last years worldwide efforts have led to the development of MS-based 3.3. MS-Data Interpretation 4. Quantitative Mass Spectrometry
- 33. 16 May et al. quantification methods. The fundamental idea with this was to shift the separation as well as quantification problem from protein to peptide level as peptides are much easier to handle than pro- teins due to their physic-chemical characteristics. Today, several MS-based quantification methods, including chemical, metabolic, enzymatic labeling, and label-free approaches ranging from the quantification of single peptides up to the quantification of pro- teins from whole cell lysates, exist that can be used as an alterna- tive or complementary setup to 2D-PAGE for analyzing complex protein and/or peptide mixtures. They include methods for rela- tive and absolute quantification such as label-free approaches (see Subheading 4.1.1); isotope labeling, e.g., isotope-coded affinity tags (ICAT) (92), isotope-coded protein labeling (ICPL) (93), isobaric tags for relative and absolute quantification (iTRAQ, TMT) (94), enzymatic labeling during protein hydrolysis in the presence of heavy (18 O-containing) water (95, 96), and stable iso- tope labeling with amino acids in cell culture (SILAC) (97); and absolute quantification of proteins (AQUA) (98, 99). For a gen- eral overview, see (28, 29, 100, 101). All the listed methods hold their advantages and disadvantages. Global internal standard (GIST) approaches where proteins are digested to peptides prior to labeling hold two major limitations: the high sample complexity results in the detection and quantification of only a limited num- ber of peptides (undersampling of the mass spectrometer), and by protein digestion prior to labeling, all information about the origi- nal belonging to the resulting peptide is lost. For protein-based chemical labeling, the main limitation is the incomplete labeling of the proteins resulting in falsified results. Today, the most accurate results are obtained with SILAC; this method is indeed mainly restricted to cells grown in culture and simple organisms. In the next two chapters, most frequently used methods for MS-based relative protein/peptide quantification are described shortly. Labeling of proteins or peptides with isotopes or other kinds of reagents distinguishable by MS is the most common strategy for gel-free protein quantification in proteomics. It is a universal approach as labeling is done after protein extraction. Over the years, several strategies have been developed which each suit dif- ferent needs. Usually, they are used for “shotgun” experiments starting directly on peptide level using LC–MS for separation, quantification, and sometimes even identification in one step. It is to be noted that these parameters depend much on the capabili- ties of the mass spectrometer used. Disadvantages of isotope labeling include cost expensiveness and the possibility of incom- plete labeling. Most of the state-of-the-art labeling chemistries are summarized by Julka and Regnier (100). 4.1. Relative Quantification 4.1.1. Isotope Labeling
- 34. 17 Instruments and Methods in Proteomics As the first method using isotopic labels for quantitative MS, the ICAT or cleavable ICAT (cICAT) was invented by Aebersold and co-workers in 1999 (92). The reagent with specificity toward side chains of cysteinyl residues consists of three elements: first, a reac- tive group toward thiol groups (cysteines); second, a linker con- taining either 12 C (light ICAT) or 13 C(heavy ICAT) atoms; and third, a biotin group that can be used for affinity purification before MS analysis. To quantify protein expression levels, e.g., of two different cell states, the protein mixture of the first cell state is labeled with light ICAT and the protein mixture of the second is labeled with the heavy ICAT. After pooling of both samples, they are enzymatically digested to peptides, separated with HPLC, and analyzed via MS. The light or heavy ICAT-modified peptides co-elute in HPLC and can be easily distinguished from each other by a 9-Da mass shift. The relative quantification is determined by the ratio of the peptide pairs (102). The main drawback is that ICAT cannot be used to quantify all proteins due to the fact that the number of proteins containing cysteines is restricted and only limited sequence coverage of the protein can be reached (28). As a result, information about protein isoforms, degradation prod- ucts, or posttranslational modifications, which are not located in the cysteine-containing peptide, are lost. The techniques isobaric tags for relative and absolute quantifi- cation (iTRAQ) and tandem mass tagging (TMT) were first introduced by Ross and Thompson, respectively (94, 103). Either protein or peptide labeling can be performed on lysine residues and/or the N-terminus. To date, eight different iTRAQ with eight different isobaric (same mass) mass tags, and six TMT reagents are available, allowing for multiplexing of samples. Isobaric peptides hold the advantage of identical migration prop- erties in the HPLC before MS analysis. Quantification is done after peptide fragmentation by the generation of label-specific low molecular weight reporter ion and signal integration. The different tags can be distinguished after peptide fragmentation as they result in different mass spectra. Therefore, this method allows the simultaneous determination of both identity and rela- tive abundance of the peptide species (104, 105). iTRAQ and TMT can also be used for absolute quantification. Indeed, both methods hold the described limitations of GIST approaches. Additionally, iTRAQ/TMT quantification cannot be obtained on all kinds of mass spectrometers as low molecular mass reporter ion region is not accessible in all instruments. Isotope-coded protein labeling is based on isotopic labeling of all free amino groups in proteins (93). Proteins from two differ- ent samples are extracted, alkylated, and labeled with either the isotope-free ICPL (light) or the isotope ICPL tag (heavy). After labeling, the protein mixtures are combined, optionally separated, e.g., by 1D-PAGE to reduce complexity, enzymatically digested, 4.1.1.1. Chemical Labeling
- 35. 18 May et al. and subsequently analyzed by MS (93). The heavy and light peptides differ in mass, and are visible as doublets in the mass spectra. Again, the peak intensities reflect relative quantitative information of the original proteins. The main advantage of this approach is the labeling already on protein level, circumventing all described limitations of the GIST approaches, although it holds the risk of incomplete protein labeling. Enzymatic labeling with heavy water (16 O/18 O method) uses the fact that during protein digestion with trypsin, Glu-C or Lys-C up to two O atoms are incorporated into the peptide. Thus, digestion in the presence of H2 18 O results in a peptide mass shift of 4 Da compared to that in peptides generated during diges- tion in the presence of normal H2 16 O. In a workflow using the 16 O/18 O method, the samples are independently digested in the presence of either H2 16 O or H2 18 O, and the samples are pooled and separated by HPLC, followed by peptide quantification and identification. This method is relatively simple; indeed, it holds the risk of back exchange of the O atoms and does not allow for multiplexing. Stable isotope labeling by amino acids in cell culture (SILAC) is a metabolic labeling based on the in vivo incorporation of specific amino acids into mammalian proteins (106). For example, mam- malian cells are grown up in a medium with normal essential amino acids (light label) and concomitantly in a medium with isotopic modified forms of essential amino acids (heavy label). After some proliferation cycles, the isotopic/normal amino acids incorporate completely into the cells. Protein extracts can be pooled, digested, and analyzed by MS. The heavy and light pep- tides elute as peak pairs separated by a defined mass difference. The ratios of the resulting relative peak intensities reflect the abundances of each measured peptide (107). Mainly, the isotopes 13 C, 15 N, 2 H, and 18 O are used for stable isotope labeling. The incorporation of the isotopes in proteins can be performed in cell culture and even in vivo in simple organisms such as Drosophila melanogaster, Caenorhabditis elegans, or mice (107, 108). For higher organisms, especially humans, this kind of metabolic label- ing is technically not feasible or completely impossible due to ethical reasons. To overcome the limitations of incomplete labeling, and also to spare costs and to reduce loss of proteins in the cause of sample prepa- ration, label-free MS approaches have been developed (101, 109). One disadvantage of label-free quantification indeed is that this technique does not allow multiplexing, and has a slight lack of sensitivity compared to labeling assays. Nevertheless, label-free approaches offer the opportunity to analyze samples with a 4.1.1.2. Metabolic Labeling 4.1.2. Label-Free Quantification
- 36. 19 Instruments and Methods in Proteomics protein amount that would be too low for labeling or 2D-DIGE strategies, since they omit many preparation steps. In spectral counting, the number of mass spectra repeatedly measured for one protein serves as a value for quantitation of this ion (109, 110). It could be shown that this number is propor- tional to the concentration of a peptide in a sample when ana- lyzed by nano-LC–MS (111). This is due to the fact that the higher the concentrations of a peptide, the longer it will take to elute from the HPLC system. Modern mass spectrometers can produce several MS² spectra in the time interval the peptides need to completely elute and be ionized by ESI. Disadvantages of spec- tral counting rise from the complexity of biological samples: Even with the best available LC system, co-eluting of peptides will still occur when analyzing complex mixtures such as cell lysates. Mass spectrometers will not be able to identify all co-eluting peptides at once. As a consequence, several replicated LC–MS runs will be needed to reach maximum identification results from one sample (111). This also leads to the second disadvantage of spectral counting that quantitative information can be obtained only from the peptides chosen as precursors, while information on less intensive or unselected peptides will be lost. Nevertheless, spec- tral counting is a cost-sparing alternative to labeling assays taking into account that this approach seems to be accurate, especially for high abundance proteins, but is highly sensitive to run-to-run variations (normalization is mandatory!). One of the latest quantitative MS methods that is still under development is comparative or differential LC–MS (112). This method utilizes the ability of mass spectrometers to record not only m/z and the intensity of the MS signal, but also RT. Softwares use these data to build contour plots in the form of heat maps, in which RT and m/z span up a plane, and MS intensity will be displayed in a color code (101). Quantitative information is obtained by integration of the volume of the m/z–RT intensity peaks. Software calculates the features which are the sum of all peaks generated by one peptide as quantitative factors. Special algorithms are used for normalization between the LC–MS runs. The advantage of this method is that it does not need any MS² spectra for quantitation, with the result that all signals recorded in one LC–MS run will be quantified. This could become the main advantage of comparative LC–MS, as the quantitative informa- tion should be more extensive than in spectral counting. Indeed, spectral counting still has advantages in sensitivity and reproduc- ibility (109). A major disadvantage of comparative LC–MS is that it allows no multiplexing and thus is more sensitive for run-to-run variations than labeling methods. Nevertheless, some studies report successful use of comparative LC–MS methods (example given by Johansson (113)).
- 37. 20 May et al. Intensive effort is spent currently to improve label-free quantification approaches, especially with respect to reproducibility, data analysis, and statistics. Over the last years, proteome research is more and more focused on the Absolute quantification of proteins (AQUA). AQUA per- mits the direct quantification of differences in proteins and post- translational modified protein expression levels (98). Therefore, chemically synthesized isotope peptides, which are unique for the proteins of interest, are used as internal standards by adding a known quantity to the analytical sample (114, 115). The ratio of synthetic to endogenous peptide is measured and the absolute level of the endogenous peptide can be precisely and quantita- tively calculated and consequently the absolute levels of proteins and posttranslational modified proteins are known (98). Although there are efforts to use MALDI, factors such as variable crystallization and laser ablation may lead to poor repro- ducibility, and thus generally ESI is the method of choice for AQUA (114). Before starting the AQUA approach, one has to adjust the peptide retention by RP chromatography, ionization efficiency, fragmentation via CID, and the amount of added stan- dards to fit with the dynamic detection range of the mass spec- trometer (see Gerber et al. for detailed information (98)). In a rather complex sample, the detection of the desired peptide likely competes with the detection of other isobaric peptides in the sample. This can be overcome by the combination of AQUA with MRM, allowing for a selective absolute quantification of the tar- get protein (115). This technique is of considerable benefit for, e.g., the absolute quantification of known biomarkers. Other available approaches for absolute quantification based on internal standards are QConCat (116) and protein standard for absolute quantification (PSAQ) (117). In the past decade, major developments in instrumentation and methodology have been achieved in proteomics. Powerful tech- niques have been established to identify and differentially quan- tify protein species of complex biological samples. Many proteomic laboratories are investigating new techniques to overcome consis- tent obstacles. Beyond alterations of the genome, the increasing advances in proteomics hold great promise for a comprehensive description of protein isoforms or even posttranslational modifi- cations. With the ongoing improvement of sample preparation techniques and mass spectrometer sensitivities, the resolution of quantifiable compounds will be further improved in proteomics 4.2. Absolute Quantification 5. Summary
- 38. 21 Instruments and Methods in Proteomics research allowing for the identification and especially reliable quantification of, e.g., physiologically relevant biomarkers indicating specific disease states. 1. For the electrophoretic separation of membrane proteins, conventional 2D-PAGE is not suitable. For this purpose, the application of specialized gel-based gel techniques such as CTAB- or BAC-SDS-PAGE, or MS-based methods is highly recommended (15, 118, 119). 2. Whenever a labeling approach is chosen for quantitative pro- teomics, labeling limitations have to be considered. For example, a saturation DIGE approach in 2D-DIGE will enhance the sensitivity but only cysteine residues will be labeled. Since cysteines are not found in all proteins, informa- tion about these proteins is lost. Moreover, peptide labeling might be more efficient than protein labeling. 3. In order to rule out labeling preferences, a dye swap should be included in 2D-DIGE experiments. This can be performed by switching the labeling dyes of samples A and B in two con- secutive experiments. 4. Protein differences between samples which have been found to be statistically valid in one technique need to be further validated by an independent method. 5. One has to consider that gel-based and MS-based techniques generally do not result in identical protein lists. Rather, both approaches complement each other. For a detailed and broad description of proteins within a sample, one may think about combining both approaches. Acknowledgments FB, PC, CS, BS, and KM are funded by the BMBF (grant 01 GS 08143). CM is supported by the Alma-Vogelsang Foundation. References 6. Notes 1. Wilkins MR, Sanchez JC, Gooley AA, Appel RD, Humphery-Smith I, Hochstrasser DF, Williams KL (1996) Progress with proteome projects: why all proteins expressed by a genome should be identified and how to do it. Biotechnol Genet Eng Rev 13:19–50 2. Patterson SD, Aebersold RH (2003) Proteomics: the first decade and beyond. Nat Genet 33(Suppl):311–323 3. Pandey A, Mann M (2000) Proteomics to study genes and genomes. Nature 405:837–846
- 39. 22 May et al. 4. Gygi SP, Rochon Y, Franza BR, Aebersold R (1999) Correlation between protein and mRNA abundance in yeast. Mol Cell Biol 19:1720–1730 5. Anderson NL, Anderson NG (1998) Proteome and proteomics: new technolo- gies, new concepts, and new words. Electrophoresis 19:1853–1861 6. Klose J, Kobalz U (1995) Two-dimensional electrophoresis of proteins: an updated pro- tocol and implications for a functional analy- sis of the genome. Electrophoresis 16:1034–1059 7. Laemmli UK (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227:680–685 8. Klose J (1975) Protein mapping by com- bined isoelectric focusing and electrophoresis of mouse tissues. A novel approach to testing for induced point mutations in mammals. Humangenetik 26:231–243 9. O’Farrell PH (1975) High resolution two- dimensional electrophoresis of proteins. J Biol Chem 250:4007–4021 10. Bjellqvist B, Ek K, Righetti PG, Gianazza E, Görg A, Westermeier R, Postel W (1982) Isoelectric focusing in immobilized pH gra- dients: principle, methodology and some applications. J Biochem Biophys Methods 6:317–339 11. Görg A, Postel W, Gunther S (1988) The current state of two-dimensional electropho- resis with immobilized pH gradients. Electrophoresis 9:531–546 12. Luhn S, Berth M, Hecker M, Bernhardt J (2003) Using standard positions and image fusion to create proteome maps from collec- tions of two-dimensional gel electrophoresis images. Proteomics 3:1117–1127 13. MacFarlane DE (1989) Two dimensional benzyldimethyl-n-hexadecylammonium chloride – sodium dodecyl sulfate preparative polyacrylamide gel electrophoresis: a high capacity high resolution technique for the purification of proteins from complex mix- tures. Anal Biochem 176:457–463 14. Eley MH, Burns PC, Kannapell CC, Campbell PS (1979) Cetyltrimethyl- ammonium bromide polyacrylamide gel electrophoresis: estimation of protein sub- unit molecular weights using cationic deter- gents. Anal Biochem 92:411–419 15. Helling S, Schmitt E, Joppich C, Schulenborg T, Mullner S, Felske-Muller S, Wiebringhaus T, Becker G, Linsenmann G, Sitek B, Lutter P, Meyer HE, Marcus K (2006) 2-D differ- ential membrane proteome analysis of scarce protein samples. Proteomics 6:4506–4513 16. Rais I, Karas M, Schägger H (2004) Two- dimensional electrophoresis for the isolation of integral membrane proteins and mass spectrometric identification. Proteomics 4:2567–2571 17. Schägger H, von Jagow G (1991) Blue native electrophoresis for isolation of membrane protein complexes in enzymatically active form. Anal Biochem 199:223–231 18. Marcus K, Joppich C, May C, Pfeiffer K, Sitek B, Meyer H, Stuehler K (2009) High- resolution 2DE. Methods Mol Biol 519:221–240 19. Rabilloud T, Vaezzadeh AR, Potier N, Lelong C, Leize-Wagner E, Chevallet M (2009) Power and limitations of electropho- retic separations in proteomics strategies. Mass Spectrom Rev 28:816–843 20. Unlu M, Morgan ME, Minden JS (1997) Difference gel electrophoresis: a single gel method for detecting changes in protein extracts. Electrophoresis 18:2071–2077 21. Alban A, David SO, Bjorkesten L, Andersson C, Sloge E, Lewis S, Currie I (2003) A novel experimental design for comparative two- dimensional gel analysis: two-dimensional difference gel electrophoresis incorporating a pooled internal standard. Proteomics 3:36–44 22. Sitek B, Luttges J, Marcus K, Kloppel G, Schmiegel W, Meyer HE, Hahn SA, Stuhler K (2005) Application of fluorescence differ- ence gel electrophoresis saturation labelling for the analysis of microdissected precursor lesions of pancreatic ductal adenocarcinoma. Proteomics 5:2665–2679 23. Nyman TA (2001) The role of mass spec- trometry in proteome studies. Biomol Eng 18:221–227 24. Aebersold R, Mann M (2003) Mass spec- trometry-based proteomics. Nature 422: 198–207 25. Steen H, Mann M (2004) The ABC’s (and XYZ’s) of peptide sequencing. Nat Rev Mol Cell Biol 5:699–711 26. Wuhrer M, Deelder AM, Hokke CH (2005) Protein glycosylation analysis by liquid chro- matography-massspectrometry.JChromatogr B Analyt Technol Biomed Life Sci 825: 124–133 27. Boersema PJ, Mohammed S, Heck AJ (2009) Phosphopeptide fragmentation and analysis by mass spectrometry, J. Mass Spectrom 44:861–878 28. Bantscheff M, Schirle M, Sweetman G, Rick J, Kuster B (2007) Quantitative mass spec- trometry in proteomics: a critical review. Anal Bioanal Chem 389:1017–1031
- 40. 23 Instruments and Methods in Proteomics 29. Urlaub H, Gronborg M, Richter F, Veenstra TD, Müller T, Tribl F, Meyer HE, Marcus K (2008) Common methods in proteomics. In: Nothwang HG, Pfeiffer SE (eds) Proteomics of the nervous system, 1st edn. Weinheim, Wiley-VCH 30. Glish GL, Vachet RW (2003) The basics of mass spectrometry in the twenty-first cen- tury. Nat Rev Drug Discov 2:140–150 31. Mitulovic G, Mechtler K (2006) HPLC techniques for proteomics analysis - a short overview of latest developments. Brief Funct Genomic Proteomic 5:249–260 32. Washburn MP, Wolters D, Yates JR III (2001) Large-scale analysis of the yeast proteome by multidimensional protein identification tech- nology. Nat Biotechnol 19:242–247 33. Nägele E, Vollmer M, Horth P, Vad C (2004) 2D-LC/MS techniques for the identification of proteins in highly complex mixtures. Expert Rev Proteomics 1:37–46 34. Chervet JP, Ursem M, Salzmann JP (1996) Instrumental requirements for nanoscale liqid chromatography. Anal Chem 68:1507–1512 35. Zaluzec EJ, Gage DA, Watson JT (1995) Matrix-assisted laser desorption ionization mass spectrometry: applications in peptide and protein characterization, Protein Expr Purif 6:109–123 36. Domon B, Aebersold R (2006) Mass spec- trometry and protein analysis. Science 312:212–217 37. Karas M, Hillenkamp F (1988) Laser desorp- tion ionization of proteins with molecular masses exceeding 10, 000 daltons. Anal Chem 60:2299–2301 38. Nordhoff E, Egelhofer V, Giavalisco P, Eickhoff H, Horn M, Przewieslik T, Theiss D, Schneider U, Lehrach H, Gobom J (2001) Large-gel two-dimensional electro- phoresis-matrix assisted laser desorption/ ionization-time of flight-mass spectrometry: an analytical challenge for studying complex protein mixtures. Electrophoresis 22: 2844–2855 39. Stuhler K, Meyer HE (2004) MALDI: more than peptide mass fingerprints. Curr Opin Mol Ther 6:239–248 40. Fenn JB, Mann M, Meng CK, Wong SF, Whitehouse CM (1989) Electrospray ioniza- tion for mass spectrometry of large biomole- cules. Science 246:64–71 41. Loo JA, Udseth HR, Smith RD (1989) Peptide and protein analysis by electrospray ionization-mass spectrometry and capillary electrophoresis-mass spectrometry. Anal Biochem 179:404–412 42. Cech NB, Enke CG (2001) Practical implica- tions of some recent studies in electrospray ionization fundamentals. Mass Spectrom Rev 20:362–387 43. Iribarne JV, Thomson BA (1976) On the evaporation of small ions from charged drop- lets. J Chem Phys 64:2287–2294 44. Dole M, Dole M, Mack LL, Mack LL, Hines RL, Hines RL, Mobley RC, Mobley RC, Ferguson LD, Ferguson LD, Alice MB, Alice MB (1968) Molecular Beams of Macroions. J Chem Phys 49:2240–2249 45. Wollnik H (1993) Time-of-flight mass ana- lyzers. Mass Spectrom Rev 12:89–114 46. Balogh MP (2004) Debating resolution and mass accuracy in mass spectrometry. Spectroscopy 19:34–40 47. Schwartz JC, Senko MW, Syka JE (2002) A two-dimensional quadrupole ion trap mass spectrometer. J Am Soc Mass Spectrom 13:659–669 48. March RE (2000) Quadrupole ion mass spectrometry: a view at the turn of the cen- tury. Int J Mass Spectrom 200:285–312 49. Douglas DJ, Frank AJ, Mao D (2005) Linear ion traps in mass spectrometry. Mass Spectrom Rev 24:1–29 50. Hager JW (2002) A new linear mass spec- tromter. Rapid Commun Mass Spectrom 16:512–526 51. Wilm M, Neubauer G, Mann M (1996) Parent ion scans of unseparated peptide mix- tures. Anal Chem 68:527–533 52. Steen H, Kuster B, Fernandez M, Pandey A, Mann M (2001) Detection of tyrosine phos- phorylated peptides by precursor ion scan- ning quadrupole TOF mass spectrometry in positive ion mode. Anal Chem 73: 1440–1448 53. HunterAP,GamesDE(1994)Chromatographic and mass spectrometric methods for the identi- fication of phosphorylation sites in phosphop- roteins. Rapid Commun Mass Spectrom 8:559–570 54. Schlosser A, Pipkorn R, Bossemeyer D, Lehmann WD (2001) Analysis of protein phosphorylation by a combination of elastase digestion and neutral loss tandem mass spec- trometry. Anal Chem 73:170–176 55. Yocum AK, Chinnaiyan AM (2009) Current affairs in quantitative targeted proteomics: multiple reaction monitoring-mass spec- trometry. Brief Funct Genomic Proteomic 8:145–157 56. Busch FV, Paul W (1961) Isotopentrennung mit dem elektrischen. Massenfilter Zeitschrift für Physik 164:581–587
- 41. 24 May et al. 57. Mikesh LM, Ueberheide B, Chi A, Coon JJ, Syka JE, Shabanowitz J, Hunt DF (2006) The utility of ETD mass spectrometry in proteomic analysis. Biochim Biophys Acta 1764:1811–1822 58. Wang Y, Franzen J (1992) The non-linear resonance QUISTOR Part1: Potential distri- bution in hyperboloidal QUISTORs. Int J Mass Spectrom Ion Processes 112:167–178 59. Wang Y, Franzen J, Wanczek KP (2009) The non-linear resonance ion trap. Part 2. A gen- eral theoretical analysis. Int J Mass Spectrom Ion Processes 124:125–144 60. Wang Y, Franzen J (1994) The non-linear ion trap. Part 3. Multipole components in three types of practical ion trap. Int J Mass Spectrom Ion Processes 132:155–172 61. Franzen J (1993) The non-linear ion trap: Part 4. Mass selcetive instability scan with multipole superposition. Int J Mass Spectrom Ion Processes 125:165–170 62. Franzen J (1994) The non-linear ion trap. Part 5. Nature of non-linear resonances and resonant ion ejection. Int J Mass Spectrom Ion Processes 130:15–40 63. Marshall AG, Hendrickson CL, Jackson GS (1998) Fourier transform ion cyclotron reso- nance mass spectrometry: a primer. Mass Spectrom Rev 17:1–35 64. Comisarow MB, Marshall AG (1974) Fourier transform ion cyclotron resonance spectros- copy. Chem Phys Lett 25:282–283 65. Goodlett DR, Bruce JE, Anderson GA, Rist B, Pasa-Tolic L, Fiehn O, Smith RD, Aebersold R (2000) Protein identification with a single accurate mass of a cysteine-con- taining peptide and constrained database searches. Anal Chem 72:1112–1118 66. Hu Q, Noll RJ, Li H, Makarov A, Hardman M, Graham CR (2005) The Orbitrap: a new mass spectrometer, J. Mass Spectrom 40:430–443 67. Perry RH, Cooks RG, Noll RJ (2008) Orbitrap mass spectrometry: instrumentation, ion motion and applications. Mass Spectrom Rev 27:661–699 68. Scigelova M, Makarov A (2006) Orbitrap mass analyzer–overview and applications in proteomics. Proteomics 6(Suppl 2):16–21 69. Aebersold R, Goodlett DR (2001) Mass spectrometry in proteomics. Chem Rev 101:269–295 70. Spengler B, Kirsch D, Kaufmann R, Jaeger E (1992) Peptide sequencing by matrix-assisted laser-desorption mass spectrometry. Rapid Commun Mass Spectrom 6:105–108 71. de Hoffmann E (1996) Tandem mass spec- trometry: a primer. J Mass Spectrom 31: 129–137 72. Steen H, Kuster B, Mann M (2001) Quadrupole time-of-flight versus triple- quadrupole mass spectrometry for the deter- mination of phosphopeptides by precursor ion scanning. J Mass Spectrom 36:782–790 73. Aldini G, Regazzoni L, Orioli M, Rimoldi I, Facino RM, Carini M (2008) A tandem MS precursor-ion scan approach to identify vari- able covalent modification of albumin Cys34: a new tool for studying vascular carbonyla- tion. J Mass Spectrom 43:1470–1481 74. Hopfgartner G, Varesio E, Tschappat V, Grivet C, Bourgogne E, Leuthold LA (2004) Triple quadrupole linear ion trap mass spec- trometer for the analysis of small molecules and macromolecules. J Mass Spectrom 39:845–855 75. Yates JR III, Speicher S, Griffin PR, Hunkapiller T (1993) Peptide mass maps: a highly informative approach to protein iden- tification. Anal Biochem 214:397–408 76. Johnson RS, Martin SA, Biemann K, Stults JT, Watson JT (1987) Novel fragmentation process of peptides by collision-induced decomposition in a tandem mass spectrome- ter: differentiation of leucine and isoleucine. Anal Chem 59:2621–2625 77. Chi A, Huttenhower C, Geer LY, Coon JJ, Syka JE, Bai DL, Shabanowitz J, Burke DJ, Troyanskaya OG, Hunt DF (2007) Analysis of phosphorylation sites on proteins from Saccharomyces cerevisiae by electron transfer dissociation (ETD) mass spectrometry. Proc Natl Acad Sci USA 104:2193–2198 78. Perdivara I, Petrovich R, Allinquant B, Deterding LJ, Tomer KB, Przybylski M (2009) Elucidation of O-glycosylation struc- tures of the beta-amyloid precursor protein by liquid chromatography-mass spectrome- try using electron transfer dissociation and collision induced dissociation. J Proteome Res 8:631–642 79. Alley WR Jr, Mechref Y, Novotny MV (2009) Characterization of glycopeptides by com- bining collision-induced dissociation and electron-transfer dissociation mass spectrom- etry data. Rapid Commun Mass Spectrom 23:161–170 80. Wiesner J, Premsler T, Sickmann A (2008) Application of electron transfer dissociation (ETD) for the analysis of posttranslational modifications. Proteomics 8:4466–4483 81. Carr SA, Huddleston MJ, Annan RS (1996) Selective detection and sequencing of phos- phopeptides at the femtomole level by mass spectrometry. Anal Biochem 239:180–192 82. Huddleston MJ, Bean MF, Carr SA (1993) Collisional Fragmentation of Glycopeptides by Electrospary Ionization LC/MS and LC/
- 42. 25 Instruments and Methods in Proteomics MS/MS: Methods for selective detection of glycopeptides in protein digests. Anal Chem 65:877–884 83. Annan RS, Carr SA (1997) The essential role of mass spectrometry in characterizing pro- tein structure: mapping posttranslational modifications. J Protein Chem 16:391–402 84. Williamson BL, Marchese J, Morrice NA (2006) Automated identification and quanti- fication of protein phosphorylation sites by LC/MS on a hybrid triple quadrupole linear ion trap mass spectrometer. Mol Cell Proteomics 5:337–346 85. Gadgil HS, Bondarenko PV, Treuheit MJ, Ren D (2007) Screening and sequencing of glycated proteins by neutral loss scan LC/ MS/MS method. Anal Chem 79: 5991–5999 86. Langenfeld E, Zanger UM, Jung K, Meyer HE, Marcus K (2009) Mass spectrometry- based absolute quantification of microsomal cytochrome P450 2D6 in human liver. Proteomics 9:2313–2323 87. Unwin RD, Griffiths JR, Leverentz MK, Grallert A, Hagan IM, Whetton AD (2005) Multiple reaction monitoring to identify sites of protein phosphorylation with high sensi- tivity. Mol Cell Proteomics 4:1134–1144 88. Perkins DN, Pappin DJ, Creasy DM, Cottrell JS (1999) Probability-based protein identifi- cation by searching sequence databases using mass spectrometry data. Electrophoresis 20:3551–3567 89. Eng JK, McCormack AL, Yates JR 3rd (1994) An approach to correlate tandem mass spectral data of peptides with amino acid sequences in a protein database. J Am Soc Mass Spectrom 5:976–989 90. BiemannK(1990)Appendix5.Nomenclature for peptide fragment ions (positive ions). Methods Enzymol 193:886–887 91. Zhang W, Chait BT (2000) ProFound: an expert system for protein identification using mass spectrometric peptide mapping infor- mation. Anal Chem 72:2482–2489 92. Gygi SP, Rist B, Gerber SA, Turecek F, Gelb MH, Aebersold R (1999) Quantitative anal- ysis of complex protein mixtures using iso- tope-coded affinity tags. Nat Biotechnol 17:994–999 93. Schmidt A, Kellermann J, Lottspeich F (2005) A novel strategy for quantitative pro- teomics using isotope-coded protein labels. Proteomics 5:4–15 94. Ross PL, Huang YN, Marchese JN, Williamson B, Parker K, Hattan S, Khainovski N, Pillai S, Dey S, Daniels S, Purkayastha S, Juhasz P, Martin S, Bartlet-Jones M, He F, Jacobson A, Pappin DJ (2004) Multiplexed protein quantitation in Saccharomyces cerevisiae using amine-reactive isobaric tag- ging reagents. Mol Cell Proteomics 3:1154–1169 95. Yao X, Freas A, Ramirez J, Demirev PA, Fenselau C (2001) Proteolytic 18 O labeling for comparative proteomics: model studies with two serotypes of adenovirus. Anal Chem 73:2836–2842 96. Staes A, Demol H, Van DJ, Martens L, Vandekerckhove J, Gevaert K (2004) Global differential non-gel proteomics by quantita- tive and stable labeling of tryptic peptides with oxygen-18. J Proteome Res 3:786–791 97. Ong SE, Blagoev B, Kratchmarova I, Kristensen DB, Steen H, Pandey A, Mann M (2002) Stable isotope labeling by amino acids in cell culture, SILAC, as a simple and accu- rate approach to expression proteomics. Mol Cell Proteomics 1:376–386 98. Gerber SA, Rush J, Stemman O, Kirschner MW, Gygi SP (2003) Absolute quantifica- tion of proteins and phosphoproteins from cell lysates by tandem MS. Proc Natl Acad Sci USA 100:6940–6945 99. Kito K, Ito T (2008) Mass spectrometry- based approaches toward absolute quantita- tive proteomics. Curr Genomics 9:263–274 100. Julka S, Regnier FE (2005) Recent advance- ments in differential proteomics based on stable isotope coding. Brief Funct Genomic Proteomic 4:158–177 101. Mueller LN, Brusniak MY, Mani DR, Aebersold R (2008) An assessment of soft- ware solutions for the analysis of mass spec- trometry based quantitative proteomics data. J Proteome Res 7:51–61 102. Shiio Y, Aebersold R (2006) Quantitative proteome analysis using isotope-coded affin- ity tags and mass spectrometry. Nat Protoc 1:139–145 103. Thompson A, Schafer J, Kuhn K, Kienle S, Schwarz J, Schmidt G, Neumann T, Johnstone R, Mohammed AK, Hamon C (2003) Tandem mass tags: a novel quantifi- cation strategy for comparative analysis of complex protein mixtures by MS/MS. Anal Chem 75:1895–1904 104. Aggarwal K, Choe LH, Lee KH (2006) Shotgun proteomics using the iTRAQ iso- baric tags. Brief Funct Genomic Proteomic 5:112–120 105. Bantscheff M, Boesche M, Eberhard D, Matthieson T, Sweetman G, Kuster B (2008) Robust and sensitive iTRAQ quantification on an LTQ Orbitrap mass spectrometer. Mol Cell Proteomics 7:1702–1713
- 43. 26 May et al. 106. Ong SE, Mann M (2005) Mass spectrome- try-based proteomics turns quantitative. Nat Chem Biol 1:252–262 107. Kruger M, Moser M, Ussar S, Thievessen I, Luber CA, Forner F, Schmidt S, Zanivan S, Fassler R, Mann M (2008) SILAC mouse for quantitative proteomics uncovers kindlin-3 as an essential factor for red blood cell func- tion. Cell 134:353–364 108. Krijgsveld J, Ketting RF, Mahmoudi T, Johansen J, Rtal-Sanz M, Verrijzer CP, Plasterk RH, Heck AJ (2003) Metabolic labeling of C. elegans and D. melanogaster for quantitative proteomics. Nat Biotechnol 21:927–931 109. Old WM, Meyer-Arendt K, Veline-Wolf L, Pierce KG, Mendoza A, Sevinsky JR, Resing KA, Ahn NG (2005) Comparison of label- free methods for quantifying human proteins by shotgun proteomics. Mol Cell Proteomics 4:1487–1502 110. Carvalho PC, Hewel J, Barbosa VC, Yates JR III (2008) Identifying differences in protein expression levels by spectral counting and feature selection. Genet Mol Res 7:342–356 111. Liu H, Sadygov RG, Yates JR III (2004) A model for random sampling and estimation of relative protein abundance in shotgun proteomics. Anal Chem 76:4193–4201 112. America AH, Cordewener JH (2008) Comparative LC-MS: a landscape of peaks and valleys. Proteomics 8:731–749 113. Johansson C, Samskog J, Sundstrom L, Wadensten H, Bjorkesten L, Flensburg J (2006) Differential expression analysis of Escherichia coli proteins using a novel soft- ware for relative quantitation of LC-MS/MS data. Proteomics 6:4475–4485 114. Lill J (2003) Proteomic tools for quantita- tion by mass spectrometry. Mass Spectrom Rev 22:182–194 115. Langenfeld E, Meyer HE, Marcus K (2008) Quantitative analysis of highly homologous proteins: the challenge of assaying the “CYP- ome” by mass spectrometry. Anal Bioanal Chem 392:1123–1134 116. Rivers J, Simpson DM, Robertson DH, Gaskell SJ, Beynon RJ (2007) Absolute multiplexed quantitative analysis of protein expression during muscle development using QconCAT. Mol Cell Proteomics 6:1416–1427 117. Brun V, Dupuis A, Adrait A, Marcellin M, Thomas D, Court M, Vandenesch F, Garin J (2007) Isotope-labeled protein standards: toward absolute quantitative proteomics. Mol Cell Proteomics 6:2139–2149 118. Basch JJ, Farrell HM Jr (1979) Charge separation of proteins complexed with sodium dodecyl sulfate by acid gel electro- phoresis in the presence of cetyltrimeth- ylammonium bromide. Biochim Biophys Acta 577:125–131 119. Akins RE, Tuan RS (1994) Separation of proteins using cetyltrimethylammonium bro- mide discontinuous gel electrophoresis. Mol Biotechnol 1:211–228
- 44. 27 Chapter 2 In-Depth Protein Characterization by Mass Spectrometry Daniel Chamrad, Gerhard Körting, and Martin Blüggel Abstract Within this chapter, various techniques and instructions for characterizing primary structure of proteins are presented, whereas the focus lies on obtaining as much complete sequence information of single proteins as possible. Especially, in the area of protein production, mass spectrometry-based detailed pro- tein characterization plays an increasing important role for quality control. In comparison to typical proteomics applications, wherein it is mostly sufficient to identify proteins by few peptides, several com- plementary techniques have to be applied to maximize primary structure information and analysis steps have to be specifically adopted. Starting from sample preparation down to mass spectrometry analysis and finally to data analysis, some of the techniques typically applied are outlined here in a summarizing and introductory manner. The field of Proteomics has been very successful in identifying the quantification of large sets of proteins (protein mixtures), for example, from whole organelles or cell lysates. Nowadays, hun- dreds of proteins within a complex sample can be easily identified by mass spectrometry, whereas only few peptides per protein are usually detected (1). This allows elucidating the name of the pro- tein via searching protein sequence databases. In addition to ana- lyzing complex protein mixtures, at least equally challenging is the art of in-depth characterization of individual proteins, or in other words, gaining as much primary structure information (including posttranslational modifications) as possible from a pro- tein of interest. In-depth protein characterization is of great importance, as it increases the chance to detect posttranslational modification (PTM), which modulates the activity of most eukaryote proteins. Also validating and distinguishing protein isoforms within a sample 1. Introduction Michael Hamacher et al. (eds.), Data Mining in Proteomics: From Standards to Applications, Methods in Molecular Biology, vol. 696, DOI 10.1007/978-1-60761-987-1_2, © Springer Science+Business Media, LLC 2011
- 45. 28 Chamrad, Körting, and Blüggel demands detailed elucidation of the protein sequence. Especially, therapeutic protein products require thorough characterization, for example, during protein engineering, protein production, and for first in men studies throughout routine testing. Mass spectrometry (MS) is an excellent tool for this purpose as it allows deducing the primary structure of proteins, including PTM by measuring mass per charge ratios (m/z) of peptide ions and corre- sponding peptide fragment ions in a high-throughput manner (2). Especially, the technology advances in recent years, including the increase in accuracy (today at ppm for peptides and peptide frag- ments), sensitivity (femtomol) and acquisition speed (more than 10,000 spectra/h) has turned MS into the most valuable analysis tool for detailed characterization of complex molecules like proteins. While high-throughput protein identification from peptide fragmentation (MS/MS) has become a standard in modern MS-based protein analytics, complete primary structure elucida- tion, including PTM is still a challenge due to various reasons: (a) Masses measured by MS are generally not unique, i.e., differ- ent amino acid sequences, including PTM may have identical or similar mass values, making them hard to distinguish. (b) Protein and peptide modifications can be induced by sample preparation and these must therefore be carefully distin- guished from original in vivo PTM. (c) Some protein sequence segments may be hard to monitor by MS, e.g., some peptides are hard to ionize or show poor fragmentation. (d) Protein modifications may not be homogenous, and due to numerous gene products caused by alternative splicing and combinations of modifications the protein mixture can be very complex. (e) Sample preparation methods have to be individually devel- oped as low protein concentration and interfering small mol- ecules like salt, detergent, and stabilizers in formulation are limiting or even preventing mass spectrometric analysis. In this chapter, we explore various current methods for comple- mentary primary structure elucidation via mass spectrometry. We also focus on sample preparation as this is an essential prerequisite to enable and improve primary structure discovery. Sample preparation methods for in-depth protein characteriza- tion by MS have to be developed to fulfill two aspects. On the one hand, sample preparation has to be performed to enable mass 2. Methods 2.1. Sample Preparation
- 46. 29 In-Depth Protein Characterization by Mass Spectrometry spectrometric analysis. On the other hand, it has to be designed in a way to minimize the risk of primary structure change due to the sample preparation. Adjuvants and contaminants, such as salt, detergent, or stabilizers, have the potential to prevent or reduce the results of mass spec- trometric analysis. In case of liquid chromatography coupled to electrospray ionization mass spectrometry, salts in millimolar con- centrations and even low detergent concentrations can be removed online within the HPLC setup (e.g., guard column or dedicated trapping column). For higher concentrations and for MALDI-MS applications, spinning columns (e.g., 3.5-kDa cutoff), dialysis (also available as microdialysis) or precipitation are the methods which are mostly applied. Additionally, separation techniques with high resolving power, such as reverse phase-HPLC or the combi- nation of SDS-PAGE (1D or 2D) with protein digestion, are also well suited to move to an MS compatible buffer, with salts like ammonia carbonate, solvents like water, acetonitril, methanol, and acids like formic or triflouracetic acid. Oxidationof,forexample,Methionine,deamidationof Asparagine, or truncation may occur under conditions of sample preparation. Additionally existing modifications (e.g., phosphorylation) may be removed (e.g., by contact to iron in not inert HPLC systems). Therefore, the sample preparation steps have to be limited to the minimum steps needed. Harsh conditions have to be avoided (e.g., 4 h, 37°C protein digestion method instead of 24 h, 37°C to avoid deamidation). There are no universal protocols as the methods have to be adopted and altered to meet several aspects: (a) Aim of analysis and intended MS technique. (b) Starting protein concentration and nature of buffer content. (c) Final protein amount and concentration needed. Additionally, protein specific aspects like hydrophobicity, tertiary structure, or modification often result in a need for protein- specific method development. Some general rules provide a guideline to method development: (a) Avoid any unnecessary step (e.g., multiple concentration, buf- fer changes). (b) Work at high protein concentrations so that only a minor frac- tion of the analyzed proteins is lost due to unspecific adsorp- tion and reduce unfavorable adjuvant to protein ratios. (c) Minimize harsh stress conditions like high temperature or RT for longer time, freeze/thaw cycle, extreme pH, lyophiliza- tion steps; oxidative stress. (d) Do not introduce any adjuvants where not needed. 2.1.1. Enabling Mass Spectrometric Analysis 2.1.2. Minimizing Risk of Primary Structure Change
- 47. 30 Chamrad, Körting, and Blüggel The primary structure of a biological molecule is the exact specification of its atomic composition and the chemical bonds connecting those atoms. For a high molecular weight protein like an antibody with approximately 20,000 atoms, the information of its primary structure is very complex. Fortunately, a good portion of this information can be reduced to the amino acid sequence. However, for proteins the primary structure is not only cov- ering the exact amino acid sequence, but also cross-links like dis- ulfide bridges and modifications. Microheterogeneity will add another level of complexity into sample characterization as it is present in many highly purified recombinant proteins as well. During the last 20 years, a huge number of mass spectromet- ric methods were developed to analyze the primary structure in detail. A full molecular weight determination by MS can provide a good insight for the verification of primary sequence and detec- tion of modification. MALDI-TOF-MS is robust in sample prep- aration and salt concentration and can give you accuracy with as low as a few Daltons for midsized proteins. With this accuracy, information on N-/C-terminal truncation or modifications like glycosylation or phosphorylation can be obtained. However, for modifications like deamidation, disulfid linkage, or even oxida- tion a higher accuracy may be needed. The ability of Electro Spray Ionization to measure the molecular weight of multiple highly charged ions in parallel results in a much better accuracy. For ESI- FT-MS measurement, these molecular weight determination can be in a sub-Dalton range. For a more detailed primary characterization, the protein has to be cleaved into subunits or peptides which are then measured by mass spectrometry. The “MALDI In Source Decay” method fragments a full intact protein within the mass spectrometer and enables here a direct sequencing of the N- and C-terminal sequence area. A sample preparation with a highly specific enzymatic diges- tion (e.g., Trypsin, Glu-C, Asp-N, etc.) will result into peptides which can be measured in a mixture (e.g., by MALDI-MS) or separated and analyzed by online LC-ESI-MS. With today’s instru- ments, these peptides can be measured with high sensitivity (fmol) and with highest mass accuracy (low to even sub-ppm level). In the same experiment, these peptides can be fragmented within the mass spectrometer and the resulting peptide fragment pattern will be recorded also with highest mass accuracy and sensitivity. With this ability and lab automation, it is possible to resolve also very complex primary structures and microheterogeneity of low abundant sequence variants. However, data analysis becomes increasingly important to unravel the full potential and latest improvements of mass spectrometry. 2.2. Primary Structure Elucidation by Mass Spectrometry
- 48. 31 In-Depth Protein Characterization by Mass Spectrometry Signal extraction and calibration are the most common first steps in the MS data interpretation process. Most software tools for MS-based protein analysis accept so-called peak lists, which are a collection of signals of a mass spectrum. Peak extraction is a com- plex task due to signal resolution, noise, signal overlapping, and the need for deisotoping. In case of ESI-MS, peptides and proteins are typically detected in various charge states (z), e.g., with z=1–4 for peptides, z=5–100 for proteins and complexes). In order to determine the exact molecular weight of a peptide or protein, the spectrum has to be deconvoluted (calculate M or MH+from m/z values). The information of the charge state can be derived directly from the given isotopic m/z signal pattern using software tools (3, 4). However, one should be aware that the applied software may fail to assign the correct charge state. In case of proteins, molecular mass is derived from m/z mass peaks of multiple charge states of the same protein. In case of time of flight (TOF) measurements calibration of the spectra is essential to obtain sufficient mass accuracy. Calibration can be done internally (e.g., using theoreti- cal m/z values of known peptides within the dataset, or by inject- ing substances in the MS instrument with each spectrum (“lock mass”)), or externally (using the calibration constants of an earlier run, which contains spectra of a known substance). After calibration, modern MS instruments can achieve a mass accuracy of few ppm. Fragmentation mass spectra of peptides can be correlated to protein sequences in a database in an automatic manner (5, 6). This can be done by dedicated protein sequence database search software (see Table 1). It is advantageous that this method does not require any a-priori knowledge about the analyzed proteins, and therefore it is often used as an initial step to identify all major protein components in a sample. 2.3. Signal Extraction 2.4. Peptide Fragmentation Fingerprinting Table 1 Overview on commonly used peptide fragmentation fingerprinting software Mascot http:/ /www.matrixscience.com/ MS-Seq http:/ /prospector.ucsf.edu/ Phenyx http:/ /www.genebio.com/products/phenyx/ Popitam http:/ /www.expasy.org/tools/popitam/ SEQUEST http:/ /fields.scripps.edu/sequest/ SpectrumMill http:/ /www.home.agilent.com/ X! Tandem http:/ /prowl.rockefeller.edu/prowl/
- 49. 32 Chamrad, Körting, and Blüggel Initially, the user has to define various input parameters carefully, such as the specificity of the applied proteolysis enzyme, maxi- mum allowed mass errors for peptide parent ion and fragment masses and the protein sequence database to be searched. Then, the software generates theoretical spectra by theoretical fragmen- tation of peptides obtained from in silico digestion of the searched database proteins. The obtained theoretical spectra are compared to the measured spectra and the result is a list of matching pep- tides and proteins. Commonly, the reported proteins and pep- tides are sorted by a specific search score that relates to the significance of the found database match. Protein and peptide modifications can be elucidated with this approach to some extent as typical database search engines that allow searching up to three different variable modifications (each amino acid in question is tested whether it is modified or not) and also fixed modifications (every amino acids is treated to be modi- fied). Also regarding enzyme nonspecificity, missed cleavage sites and even peak picking errors (e.g., failure to detect the correct monoisotopic peptide signal from overlapping isotopic distribu- tions) can be searched but generally applying these search strategies may lead to a drop in sensitivity. Therefore, it is advisable regarding only experimentally induced modifications (e.g., methionine-oxida- tion) and a maximum of one or two missed cleavages and no unspe- cific cleavage. In case of in-depth protein characterization, primary structure elucidation beyond this scope should be addressed by dedicated second round search engines (see below). Mass accuracy is crucial to obtain unambiguous results. The maximum allowed mass error parameters within the search should be set to at least two standard deviations (assuming a normal distri- bution, about 95% of the measurement errors fall in two times standard deviation). The standard deviation for mass measurements can be determined within routine MS-instrument calibration. Peptide masses determined by MS are generally not unique and each measured mass can randomly match a peptide from a sequence database. Therefore, a certain risk to obtain false posi- tive results remains. Assessing the correctness of a possible identi- fication is a challenging task. In fact, the probability that the match in question is correct cannot be calculated; however, most reported search scores relate to the probability that the observed peptide match is a pure random event (7, 8). In case of in-depth protein characterization, evaluation of sequence database search results is frequently not done automatically, but remains the task of an expert who manually inspects spectra matching to the pro- tein of interest. Usually, the primary structure detectable by a single database search is limited and must be extended by further experiments such as using a different cleavage enzyme, or using dedicated second round search engines.
- 50. 33 In-Depth Protein Characterization by Mass Spectrometry Standard database searches which can be seen as “first round” searches are limited in the elucidation of posttranslational modifi- cations, unspecific, and missed cleavages products, sequence errors, amino acid substitutions, and unsuspected mass shifts. For example, taking more than 200 described posttranslational modi- fications for all protein sequences of an organism into account would lead to an amount of peptides to be tested that impedes a brute force approach. Apart from the huge time exposure, simply the huge number of possible combinations leads to randomly matching sequences. To overcome this problem, second round searches have been developed, which work similar to peptide fragmentation fingerprinting described above but instead of searching a complete protein sequence database, only few selected protein sequences are regarded (9). Typically, protein identification is done in the first step using standard search algorithms. Second round searches are then used in the second step to elucidate previously unexplained spectra. In case of the software tools Mascot and Phenyx, the second round search feature is directly integrated, and can be triggered after the first round search. There is also a dedicated second round search tool named ModiroTM (http:/ /www.modiro.com) available. In case of ModiroTM , the user can enter own protein sequences, which is of, for example, special interest in case of therapeutic protein products from biotechnology. During the second round, search batches of unidentified spectra (e.g., whole LC-MS/MS runs) are screened in a sequential manner for various different posttranslational modifications, unknown mass shifts, unspecific cleavages, and sequence errors in one single step. A typical search result obtained by using ModiroTM is shown in Fig. 1. As genome sequencing capabilities have increased dramatically during the last decades, many organisms are sequenced today and sequences are available to the public community. However, genome sequence information is still lacking for many organisms at the same time while some of them are of interest in industrial or biochemical research. As MS/MS spectra of peptides are generated by fragmentation within the backbone of the peptide, the mass difference between two fragment ions directly provide information on the amino acid at a given peptide position. As a result, de novo sequencing is feasible for a peptide and partly also for proteins. However, each fragmenta- tion is highly sequence dependent, and the intensity of the different ions differs a lot for each fragment ion. Therefore, some positions maynotberesolved.Additionally,amassdifferencemaybeexplained by more than one amino acid combination leading to inconclusive sequences. As additional fragmentation (e.g., from internal frag- ments, side cleavage, doubly charged ions) may occure and overlay the ion series, the manual interpretation is quite laborious. 2.5. Second Round Searches 2.6. De Novo Sequencing
- 51. 34 Chamrad, Körting, and Blüggel Several software solutions were developed to perform an automated de novo sequencing (e.g., PEAKS (10), PepNovo (11), Lutefisk (12)). They provide the best guess of the sequence, at least a sequence tag. The accuracy of this prediction highly depends on the quality of the fragmentation spectra. Resulting peptide candidates can be easily searched for homology against sequence databases. MS-BLAST (13) is a dedicated alignment tool for this purpose. Fig. 1. Screenshots of the ModiroTM Software showing search parameter input and the obtained result page, including detected protein modifications in MS/MS datasets.
- 52. 35 In-Depth Protein Characterization by Mass Spectrometry Additionally, MS instrument providers deliver software packages where either a full de novo algorithm is incorporated or sequence tag generation is supported by interactive annotation of a resulting MS/ MS spectrum (e.g., BioTools, Bruker Daltonik GmbH). Although knowing that a given protein is derived from a non- sequenced organism, its MS/MS data should be analyzed in the first round by a search engine (see Subheading 2.4) with no or broad taxonomy restriction. For some peptides, the homology might be sufficient to pick up the homolog protein from another already sequenced organism, which reduces the workload for de novo sequencing. For isolated unknown proteins from an unsequenced organ- ism internal protein sequence parts are needed, in order to con- struct nucleotidic degenerative primers for PCR and subsequent DNA sequencing. For this purpose, high quality sequence infor- mation ideally form the C-terminal region and long (minimum 7, best 15 amino acids) stretches are best suited. In-depth characterization of protein requires the identification of the complete protein sequence. Usually, within a single MS analy- sis, some sequence areas are not identified or confirmed, as some peptides are outside the mass range detectable with a specific MS instrument, or have poor fragmentation. Therefore, it is advisable to make several MS runs, using different enzymes (or enzyme com- binations) for proteolysis, or to apply other sample preparation techniques. Ideally, missing sequence areas will be different for the different runs and applied techniques, yielding more com- plete sequence coverage after the combination of the found 2.7. Combination of Results (see Fig. 2) Fig. 2. Combining search results of MS/MS runs with several cleavage enzymes to get nearly complete sequence coverage.
- 53. 36 Chamrad, Körting, and Blüggel peptides. Equally, analyzing the sample with differing MS instrumentation (e.g., MALDI-MS and LC-ESI-MS/MS) will give a complementary dataset. Dedicated software is required to combine the outcome of the database searches, as a combined search with, e.g., different cleavage rules or mass spectrometric methods is not possible using currently available sequence database search software. In ProteinScape (Bruker Daltonik GmbH and Protagen AG), which is a Proteomics Bioinformatics Platform (14, 15), an algorithm for this task is integrated. Within ProteinScape, a new protein list is built, combining all peptides from all searches. Additionally, only the best matching sequence for each spectrum is annotated. For complete protein characterization of therapeutic proteins, it is necessary to show that the amino acid sequence, including modifications such as glycosylation meets the expected patterns. Second round searches with tools like ModiroTM can help to analyze existing modifications. In case of LC-ESI data, the level of a specific modification can be validated by the visualization of Extracted Ion Chromatograms (EIC) of the modified and unmodified peptide. An EIC shows the mass spectrometric signal intensity of a specific m/z value over the retention time. With an overlay of two EICs, showing the m/z of the modified and the unmodified peptide, the level of modification can be detected (Figs. 3 and 4). If both signals are visible, there should be a retention time shift between them. 2.8. Differential EIC 0 20000 40000 60000 80000 100000 120000 140000 39.0 40.0 41.0 42.0 43.0 44.0 45.0 46.0 47.0 intensity retention time [min] Fig. 3. Overlay of the extracted ion chromatograms of the unmodified and deamidated peptide W.LNGKEY.K. The peak at 42.0 min is the unmodified peptide (m/z=723.3672), the peak at 44.5 min the deamidated peptide (m/z=724.3512). By comparing the peak intensities or areas a medium deamidation can be estimated. The lower signal at 42.0 min is the second isotope of the unmodified peptide which has nearly the same m/z as the deamidated peptide.
- 54. 37 In-Depth Protein Characterization by Mass Spectrometry Another way of assuring that there are no major signals left unexplained can be done by coloring identified peptides in a base peak chromatogram (Fig. 5). Ideally, there should be no peaks left unexplained. If major signals are still unexplained, the corre- sponding MS and MS/MS spectra must be analyzed further. In-depth protein characterization by MS is significantly different from the task to identify proteins from simple or complex mixtures. The whole analysis process from sample preparation to MS acquisition 3. Conclusions Fig. 4. MS spectra of the deamidation of Fig. 3. The first spectrum is the unmodified peptide at 42.0 min, the second spectrum the deamidated peptide at 44.5 min. 0 50000 100000 150000 200000 250000 300000 350000 400000 0 20 40 60 80 100 120 140 Fig. 5. Base peak chromatogram with identified peaks colored. Most of the MS run is explained. The remaining peak at 42 min was assigned to a peptide containing glycan, but the MS/MS fragmentation was not sufficient for identification.
- 55. Other documents randomly have different content
- 56. Braithwaite, and of the brothers and sisters home on leave, and of the garden earnestly dedicated to potatoes, or the small family of Ceylon nephews and nieces deposited continually in her charge by their parents? Poor little Pamela! She had had a burdened life; the assiduities of maternity and none of its initial romance. With her large, clear eyes, very far apart, she had always a wistful look; but it was that of a child watching a game and waiting for its turn to come in, and no creature could have given less the impression of weariness or routine. For she had remained, even at thirty-three, the merely bigger sister; an atmosphere of schoolroom tea and the nurture of rabbits and guinea-pigs still hanging about her; her resource and cheerfulness seeming concerned always with the organizing of games, the care of pets, and the soothing of unimportant distresses. Tall, in her scant tweed skirts, her much-repaired white blouse, her slender feet laced into heavy boots, gardening gloves on her hands, so Rosamund had last seen her, a year ago, just before Charlie had been killed, when she had straightened herself from moulding potatoes in the lawn borders and had come forward with her pretty smile to greet her visitor and take her in to tea. Frank had been killed since then, as well as Charlie, but at that time, for both households, the war was splendid adventure rather than sorrow. Mr. Braithwaite, in the sunny, shabby drawing-room, had stumbled up among his wrappings, to point out to her his accurate flags, advancing or retreating on the many maps that were pinned upon the walls. Frank’s last letter had been read to her, and Dick’s and Eustace’s; and Pamela had come in and out, helping the maid with the tea (the Braithwaite maids were always as cheerful and desultory as the family, and Rosamund never remembered seeing one of them who had not her cap askew or her cuffs untied), standing to butter the bread herself, the side of the loaf before cutting the slice, after her old schoolroom fashion; her discreet yet generous use of the butter—the crust covered to a nicety and no lumps on the crumb —seeming to express her, as did the pouring out of the excellent tea, drawn to a point and never over, and the pleasant, capacious cups with their gilt rims and the immersed rose which, as one drank, discovered itself at the bottom. A sweet, old-fashioned, homely creature; like the evening primroses; like them, obliterated, unnoticed in daylight; and like them now, becoming visible, becoming personal, even becoming tragic at this nocturnal hour; for was this really Pamela, sweet, prosaic Pamela, sobbing so broken-heartedly
- 57. beside her? How meagre, intellectual, and unsubstantial her own grief seemed to Rosamund as she listened, almost aghast, her arm about Pamela’s shoulders; and her instinct told her: “It is a man. It is some one she loves— not Frank, but some one she loves far more—who is dead. It is something final and fatal that has broken her down like this.” And aloud she repeated: “Can you tell me, Pamela dear? Please try to tell me. It may help you to tell.” Her own heart was shaken and tears were in her own eyes. Between her sobs Pamela answered, “I love him—I love him so much. He is dead. And sometimes I can’t bear it.” Rosamund had never heard of a love-affair. But these years of war had done many things, had found out even the hidden Pamelas. “I didn’t know.—My poor child!—I never heard. Were you engaged?” She had Pamela’s ringless hand in hers. “No! No! It wasn’t that. No—I’ve never had any one like that. No one ever knew. He never knew.” Pamela lifted her head. Her face seemed now only a message emerging from the darkness; shadowed light upon the shadow, it was expression rather than form. “May I tell you?” she said. “Can you forgive my telling you—here and now,—and to-night, when you’ve come to be with him? It was Mr. Hayward I loved. I’ve always loved him. He has been all my life. Ever since you first came here to live.” Rosamund gazed at her, and through all her astonishment there ran an undertone of accomplished presage. Yes, that was it, of course. Had she not been feeling it, seeking it all the evening?—or had it not been seeking her? Here it was, then, the lacking emptiness. Desolate voids seemed to open upon her in Pamela’s shadowy eyes. She tightly held the ringless hand and felt, presently, that she pressed it against her heart where something pierced her. Was it pity for Pamela? or for Charlie? This was his; had always been his. And Pamela, who had had nothing, had lost everything. “My dear!” she murmured. “Oh, how kind you are!” said Pamela. She sat quiet, looking down at their two hands held against Rosamund’s heart. And with all the austerity of her grief she had never been more childlike in Rosamund’s eyes. Like a child, once the barriers of shyness were down and trust established, she would confide everything. Rosamund knew how it must help her to confide. “Tell me if you will,” she said. “I am glad you loved him, if it has not hurt you too much. You
- 58. understand, don’t you, that I must be glad—for him?” “Yes, oh, yes; I understand. How beautiful of you to see it all!—Even though it’s so little, it is his; something he did; and so you must care. But I don’t think there’s much to tell; nothing about him that you don’t know.” “About you, then. About what he was to you.” “That would simply be my whole life,” said Pamela. “It’s so wonderful of you to understand and not to blame me. So many people would have thought it wrong; but it came before I knew what it was going to be, and I never can feel that it was wrong. He never knew. And even if he had, it couldn’t have made any difference. It must be because of that that I can tell you. If you hadn’t been so happy, if it hadn’t been so perfect—for you and him—I don’t think that I could have told. I should just have rushed away when you came in and hidden from you.” “Why?” asked Rosamund after a moment. She heard something in her own voice that Pamela would not hear. “I don’t quite know why,” said Pamela; “but don’t you feel it too? Perhaps if it hadn’t been so perfect, even my little outside love might have hurt you—or troubled you—to hear about. But I see now that you are the only person in the world who could care to hear. It is a comfort to tell you. I am so glad you came.” Pamela turned her eyes upon her and it was almost with her smile. “When I see you like this I can believe that he is here, listening with you, and sorry for me, too.” How like an evening primrose she was! Rosamund could see her clearly now: the candid oval of the face, the eyes, the innocent, child forehead with thick, fair hair falling across it. “Yes. Go on,” she said, smiling back. She was not worthy of Pamela, and poor Charlie was not worthy of her; but no human being is worthy of a flower. And though so innocent, she was not stupid; subtlety like a fragrance was about her as she said, “You can comfort me because you have so much to comfort with.” “So much grief, or so much remembered happiness?” “They go together, don’t they?” said Pamela. “Every sort of fulness. But I needn’t try to get it clear. You understand. I always thought that perhaps people who had fulness couldn’t; now I see that I was mistaken.” “Have you been very unhappy, dear child?”
- 59. "Until now? While he was here? Oh, no, I have been lonely. Even before he came, even though my life was so crowded, it was rather lonely. I never had any one of my own, for myself. But afterwards, even if I felt lonely, I was happy. At least, after just at first. Because, just at first, it was miserable, for I couldn’t help longing to see him more and to have him like me more, and that made me understand that I was in love with him, and I was frightened. I can’t explain clearly about it, even to myself. But I was very, very unhappy. Perhaps you remember the time when I was twenty, and got so run down, and they sent me to Germany to my old governess—the only time I ever went away from home, out of England. It was a miserable time. I tried not to think of him and not to care. But I had to come back, and he was there, and I knew I couldn’t stop caring, and that all I could do about it was to try to be better because of him,—you know,—and make people happier, and not think of myself, but of him and them. And everything changed after that. I was never frightened any more, and though perhaps it wasn’t exactly happiness, it was, sometimes, I believe, almost better. I can’t explain it, but what I mean is in some poetry. I never cared much about poetry till he came. Then I seemed to understand things I’d never understood before, and to feel everything that was beautiful. “You remember how dear he was to us all—to the boys and me. I always shared in everything they did. Every bit of this country is full of him; I could never bear to go away and leave it. I want always to stay here till I die.—Flowers and birds—wasn’t he wonderful about them? And our walks in the woods! He saw everything, and made us see it. I never woke in the morning without thinking, Will he come to-day? What will he say and do? I was never tired of watching him and listening to him. All his little ways—you know. When I pleased him,—sometimes I saw the bird we were watching for first, or caught my trout well,—it was a red-letter day. And in big things—to feel I should have pleased him if he’d known. It was he who helped me in every way, without knowing it. And I took more and more joy in you. At first I had felt dreadfully shy with you—and afraid of you. You were so clever, with all your books and music and friends, and you didn’t seem to need anything. But afterwards you were so kind, that, though I was always shy, I was not frightened any longer. I used to think about you so much, and imagine what he felt about you—and you about him.—You won’t mind my saying it, I know. Perhaps you remember the way I used so often, in the evenings, to walk past with the children, and say good-night over the wall.
- 60. That was to see you and him walking together. You were so beautiful! You are far and far away the most beautiful person I’ve ever known. I always noticed everything you wore, and how your hair was done. I was glad when you took it down from the knot and had it all at the back, as you do now. And the lovely pale blue dress, with the little flounces—do you remember? —a summer dress of lawn. I did love that. And the white linen coats and skirts, and the big white hat with the lemon-coloured bow. Your very shoes —those grey ones you always had, with the low heels and little silver buckles. No one had such lovely clothes. And the way you poured out tea and looked across the table at one. Always like a beautiful muse—you don’t mind my saying it?—a little above everything, and apart, and quietly looking on.—How I understood what he felt for you! I felt it, too, I think, with him.” Yes, dear flower and child, she had: offering to Charlie that last tribute of a woman’s worship, the imaginative love of the woman he loves; cherishing the cruelly sweet closeness of that piercing community. How she had idealized them both. How she had idealized Charlie’s love. Charlie had never seen her like this. Charlie had never dreamed of her as a muse, above, apart, and quietly watching. Why, with Pamela’s Charlie she herself could almost have been in love! “What did you talk about, you and he,” she asked, “when you were together?” Their sylvan life, Pamela’s and Charlie’s, was almost as unknown to her as that of the birds they watched. She had almost a soft small hope that perhaps Pamela could show her something she had missed. “Did you ever talk about poetry, for instance?” “No; never about things like that,” Pamela answered. “He talked more to the boys than to me; he talked to us all together—about what we were doing. But I used to love listening to him when he came and talked to father. Politics, you know; and the way things ought to be done. He was a great deal discouraged, you remember, by the way they were being done. All those unjust taxes, you know. He wanted, he used always to say, to give to the poor himself; he loved taking care of them. But he hated that his money should be taken from him like that, against his will. And he always, always foresaw the war; always knew that Germany was plotting, and how England swarmed with spies. He thought we ought to have declared war upon her long ago and struck first.—I’m rather glad we didn’t, aren’t you? because then, in a way, we should have been in the wrong rather than they;
- 61. but of course he felt it as a statesman, not like an ignorant woman.—You think Germany plotted, too?” “Yes, oh, yes.” How glad Rosamund was to be able to think it, to be able, here, with a clear conscience, to remember that, on the theme of Germany’s craft and crime, she and Charlie had thought quite sufficiently alike. “But I am with you about not striking first.” “Are you really?” There was surprise in Pamela’s voice. She did not dwell on the slight perplexity. “Of course, he always worsted father if he disagreed. It was rather wicked of me, but I couldn’t help enjoying seeing father worsted. He’d never thought things out, as Mr. Hayward had. But that’s what he talked about—things like that—and you.” “Me?” Rosamund’s voice was gentle, meditative—her old voice of the encounters with Charlie. How she could hear him through all Pamela’s candid recitative! "He was always thinking about you. ‘My wife says so and so. My wife agrees with me about it. I brought my wife last night to see it as I do.’ Oh, you were with him in everything! It was so beautiful to see and hear! I used to imagine that the Brownings were like that—after I read their lives. He was a sort of poet, wasn’t he? Any one so loving and so happy is a sort of poet—even if they don’t write poetry. Down in the meadows one day, when we were watching lapwings, he and I and the boys,—he wanted to show us a nest; you know how difficult they are to find,—you passed up on the hillside, with Philip and Giles. We could see you against the larchwood, they in their holland smocks and you in white, with the white-and-yellow hat. I shall never forget the way he stood up and smiled, his eyes following you. 'There’s Rosamund and the progeny,' he said.—You know the dear, funny way he had of saying things." Yes—she knew it. Yet tears had risen to Rosamund’s eyes. Dear old Charlie; dear, old, tiresome Charlie! The tears had come as she saw him standing to look after her and his boys; but there was nothing more, nothing that she could give to Pamela, not one crumb of enrichment from what Pamela believed to be her great store. Pamela had seen all—and more than all—that there was to see. In her own silence now she was aware of a growing oppression. She was too silent, even for one mute from the depth and sacredness of memory. Might not such silence seem to reprove Pamela’s flooding confidence? She
- 62. struggled with her thoughts. “The lapwings?” she heard herself murmuring. “I remember his showing me a nest. How he loved birds and how much he knew about them! Weren’t you with us on the day we put up all the nesting- boxes here? Do you remember how he planned for the placing of each one, each bird to have its own appropriate domain? It was a lovely day, in very early spring.” “Oh—do you remember that?” How Pamela craved the crumb was shown by her lightened face; it was almost happy, as it turned to Rosamund, with its sense of recovered treasures. "Very early spring—March. Snowdrops were up over there,—and there,—and there were daffodils at the foot of the wall. You were in blue: a frieze coat and skirt of Japanese blue, with a grey silk scarf and a little soft grey hat with a blue wing in it; and you said,—you were standing just over there, near the pond,—‘We can always count on tits.’—But you did get robins, too, and thrushes in the big boxes; and then the splendid year when the nut-hatches came to the box down in the orchard. And you were tying up one box, but it was too high and he came and did it for you. I can see you both so plainly, your hands stretching up against the sky. Tall as you are he was taller; his head seemed to tower up into the branches. Such a blue sky it was! And afterwards we had tea in the drawing-room, and the tea wasn’t strong enough for him, and you liked China and he Indian tea. And you teased him and said that you had always to make him the little brown pot all for himself. He said, ‘Tea never tastes so right as out of a brown pot.’ There were white tulips growing in a bowl on the tea-table. And then you played to us. And you sang—‘I need no star in heaven to guide me.’—He was so fond of that. Oh, do you remember it all, too?" All—all. Rosamund, though her tears fell, felt her cheek flushing in the darkness. How often he had asked for "I need no star in heaven to guide me"! How often she had sung it to him, rejoicing so soon, while she threw the proper tumultuous fervour that Charlie loved into the foolish air, in the atoning thought that already Philip’s favourite was “Der Nussbaum” and that even little Giles asked for “the sheep song,” the bleak, beautiful old Scottish strain: “Ca' the yowes to the knowes,” with its sweetest drop to “my bonnie dearie.” “Oh—give us something cheerful!” Charlie would exclaim after it. “I remember it all, dear,” she answered; and there was silence for a while.
- 63. “How do you bear it?” Pamela whispered suddenly. The hour, the stillness, the hands that held her, drew her past the last barrier. Her broken heart yearned for the comfort that the greater loss alone could give. What was the strength that enabled his wife to sit there so quietly, so gently, so full of peace and pity? Rosamund felt herself faltering, stumbling, as she heard the inevitable question, and knew, as it came, that even Pamela’s heavenly blindness might not protect her, unless she could be very careful, from horrid loss or suspicion. To touch with a breath of her daylight reality that silver world of recollection would be to desecrate. Could she hold her breath and tread softly while she answered? Yes, surely. Surely she, who had hidden through all the years from Charlie, could hide from Pamela, although Pamela already was nearer than Charlie and knew her better than he had ever done. All the old strength and resource welled up in her, protecting this lovely thing, as, after the long moment, not looking at Pamela, but into Charlie’s garden, she found the right answer. “You see, dear, it is so different with me. You have only your memories. I have the boys—his boys—to live for.” It was right. It was the only answer. She heard Pamela’s long, soft breaths, full of a gentle awe, and felt her hand more tightly clasped. Once the right step was taken, it was easier to go on: "I want to tell you why I am so glad to have found you here, Pamela dear. You’ll understand, I think, when I say that motherhood lives in the present and future, and is almost cruel, cruel to everything not itself, for it forgets the past in the present. Do you see,"—she found the beautiful untruth,—“he is so much in them for me, that I might almost forget him in them—forget to mourn him, as one would if they were not there. So do you see why it comforts me to know that, while I must go on into the future with them, you will be keeping him here and remembering?” She could look at Pamela now, in safety, and she turned to her, finding rapt eyes upon her. “Come here often, won’t you, when I’m away as well as when I’m here. We must make it all look again as it did when he was with us—flowers and trees and bird-boxes. You will help me in it all and you will think of him here and love him. I know what happiness you meant to him—more than he was aware of. You were a beautiful part of his life. You say you were
- 64. always, for him, only together, with the boys. That is only partly true. He used often to speak of you to me, the little passing things people say of any one they are very fond of and take for granted. He appreciated you and counted upon you. I came here so sad, Pamela, so burdened. I’ve never been sadder in my life than I was to-night as I walked here. And you have lifted it all. It makes all the difference to know that you are here, in his garden, remembering him. More difference than I can say.” It was an unutterable gratitude that, with her tears, with love and pity and reverence, welled up in her, seeing what Pamela had done. The garden was no longer empty, and Charlie not forgotten. In the night of his death and disappearance this flower had become visible. Always, when she thought of him, she would think of evening primroses and of Pamela, so that it would be with tenderness, with the understanding, homely, unexacting, consecrating, that Pamela gave; Pamela herself becoming a gift from Charlie; emerging from the darkness, evident and beautiful,—almost another child whose future she must carry in her heart; though the only gift she could give her now, in return for all that she had given, was the full and free possession of the past, where, outside the garden wall, she had been a wistful onlooker. She felt that she opened the gate, drew Pamela in, and put into her keeping all the keys that had weighed so heavily in her unfitted hands.
- 65. AUTUMN CROCUSES I HAT you need is a complete change, and quiet,” said his cousin Dorothy. Guy, indeed, in spite of his efforts to keep up appearances, was a dismal figure. He had been passing the teacups and the bread and butter, enduring all the jests about sugar-rations and margarine, and enduring, which was so much worse, the complacencies over the approaching end of the war. His haggard face, narrow-jawed and high-foreheaded, expressed this endurance rather than any social amenity, and he was aware that Aunt Emily could hardly feel that the presence of her poet and soldier nephew added much to her tea- party. Indeed, the chattering, cheerful women affected his nerves almost as painfully as did the sound of the motor-buses when—every day it happened —he stopped on the curb, after leaving his office in Whitehall, and wondered how long it would take him to summon courage to cross the street. He felt, then, like breaking down and crying; and he felt like it now when they said, “Isn’t it all too splendid!” Cousin Dorothy was as chattering and as cheerful as the rest of them, and she had every reason to be, he remembered, with Tom, her fiancé, ensconced in Paris, safe after all his perils. Dorothy, though like everybody else she had worked hard during the war, had seen nothing and lost nothing. And she had never had any imagination. All the same, he was thankful when she rescued him from the woman who would talk to him idiotically about his poetry (she evidently hadn’t understood a word of it), and took him into a quiet nook near the piano. It might, then, have been mere consanguinity, for he had never before found intimacy possible where Dorothy was concerned; or it might have been a symptom of his state (his being at Aunt Emily’s tea-party at all was
- 66. that!); but, at all events after admitting that Mrs. Dickson had been boring him, he found himself presently confessing his terrors about the motor- buses, his terror of the dark, his sleeplessness and general disintegration. His nervous laugh was a concession to Dorothy’s possible misunderstanding; but as he went on, he felt himself almost loving her for the matter-of-factness she infused into her sympathy. After all, even good old Dorothy wasn’t stupid enough to suspect him of cowardice; and although, from a military point of view, he had made such a mess of it (invalided home again and again on account of digestive complaints, and finally, last spring, transferred to his small official post in London), to any one, really, who had at all followed his career, it would be apparent that no one could have stuck harder to the loathly job. He had felt it that, and only that, even while, prompted by pride, he had made his effort to enlist, in the first months of the war. It had been with a deep relief that he had found himself at once rejected and free to stay behind, free to serve humanity with his gift rather than with his inefficiency; for he took his poetic vocation with a youthful seriousness. And when, later on, through one of the blunders of medical examinations, he was drawn into the net of conscription, no one could have denied that he marched off to the shambles with unflinching readiness. Dorothy, he saw, took courage all along for granted: “It’s simply a case of shell-shock,” she said, as if it were her daily fare; “you’re queer and jumpy, and you can’t stand noise. It’s quite like Tommy.” He couldn’t associate Tommy, short-nosed, round-headed, red-eared Tommy, with anything of the sort, and said so in some resentment. But Dorothy assured him that for some months—just a year ago—Tommy had been at home on sick leave, and really bad enough for anything. “He suffered in every way just as you do.” Guy was quite sure he hadn’t, but he did not want to argue about it. For nothing in the world would he have defined to Dorothy what he really suffered. “It’s country air you need; country food and country quiet,” Dorothy went on. “You can get away?” “Oh, yes; I can get away all right. Old Forsyth is most decent about it. He was telling me this morning that I ought to take a month.”
- 67. “I wonder if Mrs. Baldwin could have you at Thatches,” Dorothy mused. “Tommy got well directly.” “Mrs. Baldwin?” His voice, he knew, expressed an unflattering scepticism, but he couldn’t help it. “Is she at home—an institution?” He saw Mrs. Baldwin, hatefully tactful, in a Red Cross uniform. “No, thank you, my dear.” “Of course not. What do you take me for?” Dorothy kept her competent eyes upon him. “It’s not even a P.G. place—at all events, not a regular one, though of course you do pay for your keep. She has very narrow means and takes friends sometimes, and, since the war, it’s just happened—by people telling each other, as I’m telling you—to be shell-shock cases rather particularly. It’s a lovely country, and a dear, quaint little cottage, and she does you most awfully well, Tommy said.” “I don’t like the idea of settling down like that on a stranger.” "But she wouldn’t be a stranger. You’d go through me, and I feel as if I knew her already through Tommy. He said he was at home at once. ‘Cosy,’ was how he expressed it. And you get honey on your bread at tea and cream in your coffee at breakfast, and all sorts of delightful things en casserole, that she cooks with her own hands, quite equal, Tommy said, to the French. And, Tommy knows, now, you see." “It’s Mrs. Baldwin herself who frightens me. She frightens me more than the motor-buses in Whitehall.” “That’s just what she won’t do. She’s perfectly sweet. Cosy. Middle-aged. A widow. Her nice old father lives with her, and Tommy liked him so much, too. You help her to garden, and with the bees, you know. And the old father plays chess with you in the evenings. There’s a stream near by where you can fish if you want to. It’s late for that, of course; but Tommy got some quite good sport; he was there at just this time of year. And he said that it was most awfully jolly country, and that the meadows all about were full of autumn crocuses.” “Autumn crocuses? In the fields? I’ve never seen them wild.” “They do grow wild, though, in some parts of England. They are wild there. Tommy particularly wrote about them. He said one walked down to the stream among the autumn crocuses.” Dorothy was baiting her hook very prettily, and he gloomily smiled his recognition of it. “They do sound attractive,” he owned. He hadn’t imagined Tom a man to notice crocuses, and he was the more inclined to trust his
- 68. good impressions further. After all, apart from Mrs. Baldwin and her father, the country, with honey, cream, and autumn crocuses, was a happy combination, if he had been in condition for feeling anything happy. What would Dorothy have thought of him, could she have known that, while they talked, her rosy, bonnie face kept constantly, before his haunted eyes, dissolving into a skull? Faces had a way of doing this with him since his last encounter with the war in the spring. And all the people talking in the room squeaked and gibbered. How could they go on talking? How could they go on living—after what had happened? How could he? The familiar nausea rose in him even as he forced himself to smile and say, “Well, could she have me—Mrs. Baldwin?” He could not have made an effort to find a place for himself. Such efforts, he felt sure, would have landed him at some God-forsaken farmhouse miles from the station, where the beds were damp and the meat tough; or, even worse, at a Bournemouth hotel, amid orchestras and people who made a point of dressing for dinner. But, if some one found it for him, he would let himself be pushed off. “I’m sure she could,” said Dorothy with conviction. “I have her address and I’ll write to-night and tell her all about you: that you’re a rising poet, and that your friends and relations will be so grateful if she’ll do for you what she did for Tommy.” He had an ironic glance for her “rising.” His relations—and Aunt Emily and her brood were the nearest left to him—had never in the least taken in his standing or realized that he was, among people who knew, looked upon as completely risen. At the same time, sunken was what he felt himself; drowned deep; too deep, he sometimes thought, for recovery. His last little volume had been like a final fight for breath. He had written most of it over there, after Ronnie’s death and before his own decisive breakdown, and he knew it a result as much of his malady as of his war experience. He wondered now, anew, whether these people had really read the poems. If they had, it only showed how impervious to reality they must remain. And there had actually been one, written after one of his leaves, called “Eating Bread-and-Butter,” that should indeed have embarrassed them, had they remembered it, inviting them to eat it with him in a trench with unburied comrades lying in No-Man’s Land before them. His head, as he thought of that,—from unburied comrades passing to unburied friends,—
- 69. gave a nervous, backward jerk, for he had told himself before that he must stop thinking in certain directions; and indeed the poems had helped to exorcise the obsession at the time when they had been written. All the same, it was very strange—such a poet at such a tea-party. He had plunged into Aunt Emily’s tea-party as he plunged nowadays into anything that presented itself as offering distraction. And now, as he said, “Well, if you’ll put it through, I’ll go, and be very grateful to you,” he felt that he was making another plunge into Mrs. Baldwin’s cottage. II IT was a pretty cottage he found, as, on the September evening, his station fly drew up at the wicket-gate. They had come a long way from the station, and, after leaving a small village, the winding lane, too, had seemed long. He saw, nevertheless, as he alighted, that the rustic building, old stones below and modern thatch above, could not be far from the central group of which it formed an adjunct; for it had been contrived, by devices dear to the heart of the week-ender, from two or three labourers' cottages thrown into one and covered all over with the capacious and brooding thatch. “Quaint,” Dorothy’s really inevitable word, altogether expressed it, from the box hedges that ran on either side of the flagged path, to the pale yellow hollyhocks beside the door. A round-cheeked country girl, neatly capped and aproned, opened the door on a square, rush-matted hall; and beyond that he saw a room full of the sunset, where a table was being laid and from which Mrs. Baldwin came out to greet him. She was not tall, and had thick, closely bound braids. He had dreaded finding himself at once dealt with as a case; but Mrs. Baldwin’s manner was not even that of one accustomed to paying guests. Her murmur of welcome, her questions about his journey, her mild directions as she led him up to his room, "Be careful at this landing, the level of the floor goes up and the beam comes down so low,"—were rather those of a shy and entirely unprofessional hostess. He thought, as soon as he took in his room, with its voile-de-Gènes hangings and dear old furniture, that he pleased her by saying, “What a delicious room!” and even more when, on going to the wide, low, mullioned
- 70. window, its panes open to the west, he added, “And what a delicious view!” There were meadows and tall hedgerow elms, and, running in a tranquil band of brightness, the stream that reflected the sky. She did not say that she was glad he liked it, but her very gentle smile at the welcome it all made for him was part of the welcome. What she did say was, with the little air of shy preoccupation, while she wrung her finger-tips together, those of one hand in those of the other, “I think the water’s very hot. I have a rather young little maid. You’ll tell me if you want anything. Are three blankets and the down quilt enough? The nights are rather cold already.” He said that three would be perfect, secure, from his glance at the deep, comely bed, that they would be beautifully thick and fleecy. “Then you’ll come down to us when you are ready.” She stood in the door to look round again. “Matches here, you see; biscuits in the little earthenware box; and the spirit-lamp is in case you should wake in the night —you could make yourself a cup of cocoa? Everything is there—cocoa, milk, and sugar. It usually sends one off again directly.” It was all the slightly shy hostess rather than the businesslike soother and sustainer; and, no, it wasn’t a bit cosy. He repudiated that word indignantly, while he washed—the water was very hot, admirably hot; there was a complacency about cosy, and Mrs. Baldwin had no complacency, though she was, for all her shyness and the unconscious gestures of physical nervousness, composed. Her hands, he remembered, recalling their little trick,—he had noticed it in the hall,—were like a child’s; not the hands of a practical housewife. Yet, from the look of that bed (yes, thank heaven, a box-spring mattress!), from the heat of the water, and, above all, the deft and accessible grouping of the spirit-lamp and its adjuncts, she proved that she knew how to make one comfortable. There were the meadows and—going again to the window, he wondered leaning out,—could he see the autumn crocuses? Yes, surely; even at this evening hour his eyes distinguished the pale yet delicately purpling tint that streaked the pastoral verdure. What a delicious place, indeed! He stood, absorbed in looking out, until the maid came to say that supper would be ready in five minutes. The long room, the living-room,—for it combined, he saw, all social functions,—also faced the meadows at the back of the house, and the
- 71. primrose coloured sunset still filled it as he entered. Mrs. Baldwin was busying herself with the table, and an old gentleman with a very long white beard rose, with much dignity, from the grandfather’s chair near a window- seat. Mr. Haseltine, so his daughter named him, had more the air of seeing the visitor as a P.G., perhaps even as a shell-shock patient; but he was a nice old man, Guy felt, although his beard was too long. He wore a brown velveteen jacket, and Guy surmised that he might have been a writer or scholar of some not very significant sort. “Yes, we think ours a very favored nook indeed,” he said, as Guy again praised the prospect. “Yes; three cottages. Very happily contrived, is it not? There is a clever builder in the next town. He kept the old fireplace, you see; that end was a kitchen and the beams are all the old ones. Three gardens, too, thrown into one; but that is entirely my daughter’s creation. Pig-styes used to be in that corner.” Guy looked out at the squares of colour, the low beds of mignonette, the phloxes, larkspurs, and the late sweet-peas a screen of stained-glass tints against the sky. Where the pig-styes had been was a little thatched summer- house with rustic seat and table. The bee-hives were just outside the hedge, at an angle of the meadow. Mr. Haseltine continued to talk while Mrs. Baldwin and the maid came in and out, carrying tea and eggs and covered dishes. “I hope you don’t mind high tea,” she said. “It seems to go with our life here.” He felt that high tea was his favourite meal. There was a big white earthenware bowl on the table, filled with sweet-peas. “Where do you get the old-fashioned colours?” he asked her. “I thought the growers had extirpated them; one sees only the long-stemmed ones nowadays, with the tiresome artistic shades.” He pleased her again, he felt sure, and she told him that she always saved the seed, liking the old bright colours better, too. He was glad that he had come, although Mr. Haseltine’s beard was too long and he feared that he would prove talkative in the worst way, the deliberate and retaining way. He liked the smell of everything,—a mingling of sweet-peas, rush-matting, and China tea,—and the look of everything; good, unpretentious old oak furniture, fresh, if faded, chintzes, and book- lined walls; and he presently liked the taste of everything too.
- 72. “I feel already as if I should sleep to-night,” he said to Mrs. Baldwin. She sat behind the tea-urn a little distracted, if anything so mild could be called distraction, by the plunging movements of the little maid as she moved about the table. “That will do nicely, Cathy,” she said. “We can manage now. You can bring in some more hot water if I ring.—Oh, I do hope you’ll sleep. People usually sleep here.” She was hardly middle-aged, though, after Dorothy’s bright browns and pinks, Tommy might well have thought her so. Many years older than Dorothy, of course, yet how many he could not in the least compute. There was an agelessness, with something tough and solid, about her; she was as little slender as she was stout; she might, with her neutral tints,—hair, skin, dress,—have looked almost the same at sixty as she did now. She wasn’t pale, or sallow, or sunburned; yet her complexion seemed so to go with her hair that the whole head might have been carved in some pleasantly tinted stone. Only her eyes gave any depth of difference; gentle eyes, like a grey- blue breadth of evening. She had a broad, short face and broad, beautifully drawn lips, and looked almost mysteriously innocent. Guy took her in to this extent, swift as he was at taking people in, and sensitive as he was to what he found. He felt sure—and the depth of comfort it gave him made him aware of all the reluctances Dorothy’s decision had overborne—that she hadn’t the ghost of a method or of a theory. Shell-shock people had merely happened to come and had happened to get well quickly. He even gathered, as the peaceful evening wore on,— Cathy clearing, placid lamps lighted, the windows still left open to the twilight—that she didn’t really think very much about her cases, in so far as they were cases and not guests. Having done her best in the way of blankets, hot water, and spirit-kettles, and seen them settled down into the life she had made for herself,—and not at all for them,—she went her own way, irresponsible and unpreoccupied. To-night she didn’t attempt to entertain him. It was Mr. Haseltine, at supper, who kept up the conversation, and with the air of always keeping it up, with even the air, Guy imagined once or twice, of feeling it specially his part to make amends, in that sort of resource, for his dear daughter’s deficiency. She was, Guy saw, very much his dear daughter; but he felt sure that it had never entered the old gentleman’s head that any one would find her interesting when he himself was there.
- 73. After supper she was occupied for a little while at her desk, adding up figures, it appeared, in house-books; for she came to her father and asked him if he would do a column for her. “It has come out differently three times with me,” she confessed, but without ruefulness. “I’m so dull at my accounts!” Guy, as Mr. Haseltine fumbled for his large tortoise-shell eyeglasses, offered to help her, and then came over and sat beside the desk and did the rest of the sums for her. She was tidying up for the month, she told him, and always found it rather confusing. “It’s having to put the pennies, which are twelves, into pounds, which are twenties, isn’t it?” she said, and thanked him so much. But this could hardly be called entertaining him, nor could it, when he accompanied her across the lane in the now deepening dusk, to shut up her fowls. After that, there was the game of chess, during which Mrs. Baldwin absented herself a good deal, helping Cathy, Guy imagined, with the beds and hot-water bottles; and at nine-thirty they all lighted their candles and went upstairs. Bedtime had been, for many months, his most dreaded moment. The door shut him in and shut away the last chance of alleviation. There was nothing for it but to stretch himself haggardly on his couch and cling to every detail in the day’s events, or in the morrow’s prospects, that might preserve him from the past. To fight not to remember was a losing game, and filled one’s brain with the white flame of insomnia. He had found that it was when, exhausted by the fruitless effort, he suffered the waiting vultures to settle upon him, abandoned himself to the beaks and talons, that, through the sheer passivity of anguish, oblivion most often came. To-night, from the habit of it, his mind braced itself as he came into the room, and he was aware, as he had been for nearly a year now, that Ronnie’s face was waiting, as it were, on the outskirts of consciousness, to seize upon him. But, after he had lighted the candles on his dressing-table and the candles on the mantelpiece, taken off his coat, and started undressing, he found that his thoughts, quite effortlessly, were engaged with his new surroundings, old Mr. Haseltine’s beard and eyeglasses occupying them, and the clucking noise he made in drinking the glass of hot ginger and water that had been brought to them on a tray while they played; Mrs. Baldwin’s accounts, her fowls, and the colour of her eyes. He decided that
- 74. the colour was Wedgwood, or perhaps periwinkle blue—some very dense, quiet colour. As he moved about the room, this protective interest came to him from the little objects he made acquaintance with: the round Venetian box, dim gilt and blue and red, on the chest of drawers in which he found a handful of tiny shells—shells, no doubt, that Mrs. Baldwin had picked up during a seaside outing; the faded old blue leather blotter on the writing-table, marked E. H., which had probably been hers since maiden days (and did E stand for Ethel or Edith or Ellen?); the pretty lettering in fine black script of the writing-paper so pleasantly stacked; the dear old Dutch coffee-pot and jug on the mantelpiece, and the bowl of mignonette that she, of course, had arranged. He sank his face into its fragrance, and peace seemed breathed upon him from the flowers. He was wondering, as he got into bed, with a glance, before he blew out the candle, at the birds and branches, the whites and blacks and roses of the voile-de-Gènes, whether he would find the autumn crocuses open in the meadows next morning; it had looked like the evening of another fine day. Then, the candle out, his thoughts, for a little while, were tangled in the magical dreamland of the voile-de-Gènes, and the breath of the mignonette seemed to lie upon his eyelids with a soft compulsion to peace, until, all thought sliding suddenly away, he dropped into delicious slumber. III HT found the crocuses open, before breakfast. Only Cathy was in the living- room, sweeping, when he crossed it, though he thought he heard Mrs. Baldwin in the kitchen. A robin was singing on a spray over the summer- house. The sky arched pale and high; and though there was no mist in the air, its softness made him think of milk. From the garden he passed into the meadows, and, almost at once, saw, everywhere, the fragile, purple flowers about him, if purple were not too rich a word for their clear, cold tint. Lower down, near the stream, they made him think of the silver bobbins set playing by great rain drops when they fall heavily upon wide, shallow pools of water; and they seemed to grow even more thickly in the farther meadow beyond the wooden bridge. A sense of bliss was upon him as he walked among the flowers. He had
- 75. never seen anything more lovely, and all but the darker buds were open, showing pale golden hearts to the sun. Yet, by the time that he had crossed the bridge, leaning on the high rail to look down into the limpid, sliding water, he knew that it could never stay at that or mean that for him. He had seen fields of flowers in France, and, while the horrors there had been enacted, these fields of crocuses, year after year, had bloomed. What they meant for his mind was the unbridged chasm between nature and the sufferings of man. Only when one ceased to be a man, ceased to remember and to think, could such a day, such sights, bring the unreasoning joy. Walking back, he saw, as he approached the house, that Mrs. Baldwin was standing at the garden-gate, and, bare-headed, in the linen dress of pale lavender, she made him at once think of the crocuses, or they of her. Their gentleness was like her, their simplicity, and something, too,—for he felt this in her,—of unearthliness. More perhaps, than any other flower they seemed to belong to the air rather than to the ground, and, with their faint, pale stalks, their fragile petals unconfined by leaf or calyx, to be rising like emanations from the sod and ready to dissolve in mist into the sunlight. “You’ve had a little walk?” Mrs. Baldwin asked him as they met. He said he had been looking at the crocuses. “Are they really crocuses?” he questioned. “I’ve never seen them wild before.” “They’re not real crocuses,” she said, “though those grow wild, too, in a few places in England. These flowers are always called autumn crocuses hereabouts; but they are really, botanically, meadow saffron; and they grow wild in a great many places. You see they are not so dark a purple as the wild crocus, and they are much taller, and the petals are more pointed. Much more beautiful flowers, I think.” “Meadow saffron. That’s a pretty name, too. But I think I’ll go on calling them autumn crocuses. They were one of the reasons that made me want to come here,” he told her. They were leaning on the little garden-gate looking over the meadows. “Really? Did you hear about them?” He told her what Dorothy had said, passed on from the appreciative Tommy, and she said again, “Really!” and with surprise, so that, laughing a little, he said that he believed she would never have thought of Mr. Barnet as an appreciator of crocuses. She laughed a little, too, confessing to a
- 76. community of perception where Tommy was concerned, and remarked that it was very nice of him to have cared. “What he talked about,” she said, “was the food. He was never done praising my coffee. It’s time for coffee now,” she added. Guy, as they went in, said that, after all, if that was what Tommy talked about, he wondered that his caring for the crocuses should have surprised her, for he was sure that the one was almost as poetical as the others. It was poetical, indeed, as she made it, in a delightful and complicated apparatus, glass and brass and premonitory scented steam; and the milk was as hot as the water had been, and there was cream. “How do you manage it, in these days?” he asked. But she said that it wasn’t wickedness and bribery, really: she and Cathy skimmed it from the milk that was brought from the nearest farm. He realized that he was himself talking about the food just as Tommy had done; just as the chattering women at Aunt Emily’s tea-party had done; just as everybody, of course, had been doing in England ever since food became such an important matter. But it was Mrs. Baldwin who made him do it; for though unearthly, she was deliciously prosaic. He felt that anew when he heard her going about the house in her low-heeled little shoes, with Cathy. They did, evidently, all the work, and how fresh, composed, and shining everything was. The living-room, with its happy southern windows, its tempting writing-tables, its flowers and books, was an embodiment of the poetry that only such prose can secure. Guy, while Mr. Haseltine sat behind his rustling Times, strolled before the shelves, surprised, presently, at their range of subject. Surely not Mrs. Baldwin’s, such reading; hardly, he thought, Mr. Haseltine’s. He took down a volume of Plotinus and found, on the fly-leaf, “Oliver Baldwin,” written in a small, scholarly hand. That explained it, then. Her husband’s. The Charles d’Orleans, too, the Fustel de Coulanges, the Croce, and the Dante, with marginal notes. He had been a man of letters, perhaps. Of the dozen books he took down to examine, only one was initialled “E. H.,” and that, suitably, was Dominique. But it had been given her by “O. B.” As in the garden, presently, he and the old gentleman walked up and down, smoking, Guy asked him, with the diffidence natural to the question, whether his son-in-law, Mrs. Baldwin’s husband, had been killed in the war; though he couldn’t imagine her a war-widow. One didn’t indeed think of
- 77. her in connection with marrying and giving in marriage—that was part of the unearthliness; yet widowhood, permanent widowhood, seemed a suitable state. She was not girlish, nor was she wifely. She was widowed, and it had happened, he felt sure, in spite of his question, long ago. As he had expected, his companion replied, “Ah, no; he died eight, nine years since.” And Mr. Haseltine then went on to tell, taking the war as the obvious interest, and not without the satisfaction that Guy had so often met and so often loathed, that he had lost dear ones. “Children of my eldest son. Fine lads. Brave boys. One in the first month—at the Marne; the other only last year, flying. Yes; I’ve done my bit,” said Mr. Haseltine, with the fatuity that he was so plentifully companioned in displaying. “Bit.” Odious word. His “bit.” Why his? Had any one written a poem on the formula coming from the lips of those for whom others had died? A scattered, flagellating line or two floated through Guy’s mind. Something about barbed wire came in. He wondered how old Mr. Haseltine would have felt about his “bit,” hung up on that and unable to die. He wondered where the fine lads now lay. No more coffee for them, with cream in it; no more robins singing; no more strolling smokes among mignonette in the sunlight. How they were forgotten, already, except for trophies, for self- glorification to display! How pleased, how smug this rescued, comfortable world! Something of his distaste attached itself even to Mrs. Baldwin when she next appeared. Something irritating him in her peacefulness. She, too, had seen nothing and lost nothing. But, at all events, she wouldn’t, he knew that, take any stand on the two nephews to claim her “bit.” There was nothing fatuous about Mrs. Baldwin. The slight distaste still lingered, however, and he found himself wondering once or twice, during the day that passed, in spite of it, so pleasantly, whether she wasn’t, for all his idealizing similes, a stupid as well as a sweet woman. It was not because of filial self-effacement that she let her father do all the talking at meals: it was simply because she had nothing to say, and the good old boy was quite right in taking his responsibility for granted. The person who could talk was the responsible person. Her mind, though so occupied, was quite singularly inactive and, he was sure, completely uncritical. She didn’t find her father in the least a bore, or suspect that anybody else might find him so. She did find, Guy felt sure, satisfaction in all her occupations. He heard her laughing—a quiet little laugh—with Cathy in the kitchen; and in the afternoon, when he helped her to prick out seedlings, her attentive profile—
- 78. as, after he had dug each hole, she dropped in the little plant, pressed the earth about its roots, and fixed it in its place—made him think of the profile of a child putting its dolls to bed. They planted three beautiful long rows, and Guy was quite tired by tea-time, for though they had high tea at half- past six, they were not deprived of the precious afternoon pause, taking place as it did at the unaccustomed but pleasing hour of four. After tea she went to see some people in the village, Mr. Haseltine dozed in his chair, and Guy took a long walk. So the days went on, and at the end of a week he was able to write to Dorothy and tell her that he was sleeping wonderfully and that Mrs. Baldwin’s cottage was all that she had pictured it. By the end of the week he had even grown rather attached to Mr. Haseltine, and he enjoyed playing chess with him every evening; and sometimes they had a game in the afternoon when tea was over. The undercurrent of irritation still flowed, but he had learned to put up with the old gentleman and to circumvent his communicativeness, and in the case of Mrs. Baldwin he more and more felt that she was the sort of person to whom one would, probably, forgive anything. It had become evident to him that what might be dulness might also be unawareness. That was a certain kind of dulness, it was true, but it didn’t preclude capacity for response if the proper stimulus were applied. It amused him to note that if none of the nearly inevitable jars of shared life seemed ever to occur between her and her father, it was simply because, when a difference arose, she remained unconscious of it unless it were put before her. Nothing could have been less in the line of selfishness; it was she who thought of him, of his comfort and happiness, and who ordered her life to further them; he, in this respect, was passive; but Guy felt that the poor old boy often brooded in some disconsolateness over small trials and perplexities that a companion more alert to symptoms would have discerned and dispelled at once. Mr. Haseltine even, sometimes, confided such grievances to the P.G. “I don’t want to bother Effie about it,” he said;—E. had stood for Effie- -“she’s a dreamy creature and very forgetful. But it’s quite evident to me that the rector and his wife have been expecting to be asked to tea to meet you. I’ve just been talking to them in the lane, and I saw it plainly. They had asked us to bring you before you arrived, hearing we were to have another guest,—they’ve always been most kind and neighbourly in helping us to entertain our new friends,—and I really don’t know why Effie should have
- 79. got out of it. I usually have to remind her, it’s true. But I sometimes get tired of always having to. She doesn’t care for them herself; but that’s no reason why you might not. We have few enough interests to offer visitors.” Guy was glad to have escaped the rectory tea, though he did not say this in assuring Mr. Haseltine that the entertainment offered at Thatches was absolutely to his taste. He was completely out of place at any rectory; he could imagine no rector who would not find his poems pernicious; but he felt that there was justice in Mr. Haseltine’s contention. He might have cared for them. As it was, Mr. Haseltine was brought once again to reminding her. It was evident then that she was ready to please anybody or everybody. “Ask them? Ought I to ask them?” “My dear, it’s ten days since they sent their invitation. They spoke again —and it’s the second time—of having been so sorry not to see us, when I met them yesterday, in the lane. I don’t know why you did not go.” “I thought it would bore Mr. Norris, father. He came here for quiet, you know. But would it bore you?” she asked Guy. “They are very nice. I don’t mean that.” “It’s certainly very pleasant being quiet,” said Guy; “but if Mr. Haseltine likes having them, I assure you that people don’t frighten me in the least.” “Oh, not on my account,” Mr. Haseltine protested. “I see our good friends continually. It is of them I am thinking, as well as of Mr. Norris. He might find them more interesting than you do, Effie, and they will, I fear, be hurt.” Now that it was put before her, Mrs. Baldwin did it every justice, rising from the breakfast-table, where she had just finished, to go to her desk, and murmuring as she went, “I hadn’t thought of that. They might be hurt. So, if it won’t bore you, Mr. Norris.” And the Laycocks were asked, and did indeed bore Guy sadly. It was on the night after their visit—Mr. Laycock had questioned him earnestly about his personal impressions of the war and to evade him had been wearying—that Guy, for the first time, really, since he had come, found sleep difficult and even menaced. It was because of that, he felt sure, looking back on it, that the curious occurrence of the next day took place— curious, and, had it taken place in the presence of any one else,
- 80. embarrassing. But what made it most curious was just that; he had not felt it embarrassing to break down and sob before Mrs. Baldwin. The morning had begun badly. The breakfast-table papers had been full of the approaching victory. Mr. Haseltine read out passages from the Times as he broke his toast and drank his coffee. He had reiterated the triumph of his long conviction, and Mrs. Baldwin had murmured assent. “All’s well with the world,” was the suffocating assurance that seemed to breathe from them both. “All’s blue.” Was hell forgotten like that? What if the war were won? Of course, it had to be won—that was an unquestioned premise that had underlain his rebellions as well as Mr. Haseltine’s complacencies since the beginning. But what of it? No victory could redeem what had been done. He went out into the garden, to be away from Mr. Haseltine, as soon as he could, and took a book into the summer-house; and it was here, a little later, that Mrs. Baldwin, seeing him as she passed, her garden-basket on her arm, paused to ask him, with her smile of the shy hostess, if he were all right. She didn’t often ask him that, and he saw at once that his recent recalcitrancy to rejoicing had pierced even her vagueness. He knew that he still looked recalcitrant, and he was determined not to soften the overt opposition rising in him; so he raised his eyes to her over his book and said that he was not, perhaps, feeling very fit that morning. Mrs. Baldwin hesitated at the entrance to the summer-house. She looked behind her at the garden and up at the roses clustering over the lintel under the thatch; she even took out her scissors, in the uncertainty that, evidently, beset her, and snipped off a dead rose, and she said presently, “It was all that talk about the war, wasn’t it—when what you must ask is to forget it.” “Oh, I don’t ask that at all,” said Guy. “I should scorn myself for forgetting it.” She glanced in again at him, mildly. “I want to forget what’s irrelevant, like victory,” he said; “but not what is relevant, like irremediable wrong.” Her awareness had not, of course, gone nearly as far as this. She kept her eyes on him, and he was glad to feel that he could probably shock her. “You see,” he found himself saying, “I saw the wrong. I saw the war—at the closest quarters.” “Yes—oh, yes,” Mrs. Baldwin murmured. “For me, tragedy doesn’t cease to exist when it’s shovelled underground. If one goes down into hell,
- 81. one doesn’t want to forget the fact—though one may hope to forget the torments and horrors; one wants, rather, to remember that hell exists—and to try and square life with that actuality.” There was silence after this for a moment, and he imagined that she was very much at a loss. Her next words seemed indeed to express nothing so much as her failure to follow—that and a silliness really rather adorable, had he been in a mood to find it anything but exasperating. “But, still—hell doesn’t exist, does it?” she offered him for his appeasement. Guy laughed. “Doesn’t it? When things like this war can happen? How could it ever have existed but in men’s hearts? It’s there that it smoulders and, when its moment comes, leaps out to blast the world.” He could talk to her like this because she was too simple to suspect in him a poetical attitudinizing; any one else would of course suspect it. Guy was even aware that to any one else that was what it would have been. She looked kind and troubled and as much as ever at a loss. She didn’t know at all how to deal with the patient, and she was evidently uncertain what to do, since it might seem heartless to go away and leave him to his black thoughts, yet intrudingly intimate to come and sit down beside him. Nothing could be less intimate than Mrs. Baldwin. It was he, of course, who was tasteless in talking to her in a vein appropriate only to intimacy. “Don’t bother over me,” he said, offering her the patent artifice of a smile. “I’m simply a bad case. You mustn’t let me trouble you. You must just turn your back on me when I’m like this.” It was not poetic attitudinizing now; there was in his voice a quaver of grief and she responded to it at once. “Oh, but I don’t like to do that. I do wish I could be of some help. I see you haven’t slept, for you look so tired, as you did when you first came. And Mr. Laycock did bore you. It’s wrong of people to talk to you about the war.” For the first time he saw in the eyes fixed upon him, pity, evident pity and solicitude. And before it he felt himself crumble suddenly. He saw all the reasons she had for pitying him, did she but know. He saw Ronnie’s face again; he saw his own haunted night and his own grief. He wanted her to see it. “Oh—one can’t be guarded like that,” he murmured; “I must try to get used to it. But—I didn’t sleep; that’s true. I’m so horribly afraid of not sleeping. You can’t imagine what it is. I’ve the most awful visions.” And
- 82. leaning his elbows on the table, he put his hands before his face and began to cry. She stood there; he did not hear her move at first; and then she entered and sat down on the seat beside him. But she said nothing and did not touch him. He had had in all the tumult of his disintegration, a swift passage of surmise; would she not draw his head upon her shoulder, like a mother, and comfort him? But that would have broken him down heaven knew how much further. He cried frankly, articulating presently, “It’s my nerves, you know; they have all gone to pieces. I lost my friend; my dearest friend. For months I didn’t sleep.” Mrs. Baldwin’s silence was not oppressive, or repressive either. He heard her hands move slightly on the basket she held on her knees and the soft chafing in the folds of her linen bodice that her breathing made. It was an accepting stillness and it presently quieted him; more than that, it enabled him at last to lift his head and look at her without feeling ashamed of himself. Oddly enough, he knew that he, perhaps, ought to be. He could have helped himself. There had been an element of wilfulness in his breakdown; he had wanted her to see; but, even had she known this about him, he would not have felt ashamed. She was so curiously a person with whom one could not associate blames and judgments. She was an accepting person. She wasn’t looking at him, but out at the sweet, bright, autumnal little garden; and as her eyes came to him, he felt them full of thought; felt, for the first time, sure that, whatever she might be, she was not dull. He could not remember, looking back at the little scene, that she had said a single further word. He did not think that he had said anything further. He was helping her, a little while after, to prune the Aimée Vibert rose that had grown with great unruliness over the little tool-house near the kitchen door. “It will really pull it down unless we cut out some of these great branches,” she had said, as, equipped with stout gloves, they had worked away together, unfastening the tangled trails and stretching them out on the ground. So displayed, the Aimée Vibert was drastically dealt with, and it was midday before they finished fastening the thinned and shortened shoots into place.
- 83. She had said nothing further; but he believed that, for the first time, her thought really included him. He had been put before her. She was different afterwards. He had become an individual to her, and had ceased to be merely the paying guest. IV THE third week came. There was rain, rather sad September rain, for a day or two. They sat in the evenings before the wide fireplace where logs blazed. Mrs. Baldwin, at his suggestion, read aloud to them Fabre’s Souvenirs Entomologiques. She read French prettily, better than he did himself, and he was a little chagrined once or twice to find that she knew it better, priding himself on his French as he did. He had lived for a year in Paris, with Ronnie, before the war. The horrors of the grim, complicated underworld revealed by the French seer distressed him. Mrs. Baldwin did not feel them as he did, feeling the marvels rather than the horrors, perhaps. She laughed a little, rather callously, at the ladies who devoured their husbands, and seemed pleased by the odious forethought of the egg-laying mothers. She shared Fabre’s humorous dispassionateness, if not the fond partiality which, while it made him the more charming, didn’t, Guy insisted, make his horrid wasps and beetles a bit more so. As usual, she vexed him a little, even while, more and more, he felt her intelligent; perhaps she vexed him all the more for that. “She’s so devilishly contented with the world,” he said to himself sometimes, even while he smiled, remembering her laughter. Old Mr. Haseltine fell asleep one night while she read, and to be together there before the fire, the old man sleeping beside them, made them nearer than they had ever been before. Guy was aware of this nearness while he listened and while he watched her hand, short, like a child’s (and her face was so short) support the book, and her eyelashes dropping down the page or raised to a fresh one. When he went to his room that night, he stood still for a long time, his candle in his hand, listening to the soft beat of the rain against the window. He was hardly ever now afraid of being alone, or of the dark, and he stood there musing and listening, while he still seemed to see Mrs. Baldwin’s
- 84. hand as it held the book, and her reading profile. Her life seemed to breathe upon him and he rested in it. He slept deliciously. “Did you know that I write?” he asked her next day. He had wondered about this once or twice before. “Oh, yes; your cousin, in her letter, you know, told me that you wrote,” said Mrs. Baldwin. They were in the living-room after midday dinner, and alone. She looked up at him very kindly from the papers and letters she was sorting at her desk. “You’ve never heard of my effusions otherwise, though?” He put on a rueful air. “Such is fame!” “Are you famous?” Her smile was a little troubled. “I don’t follow things, you know, living here as I do.” “You read the papers. I have had reviews: good ones.” “I don’t read them very regularly,” she admitted. “And I so often don’t remember the names of people in reviews, even when I’ve liked what is said of them. Have you any of your poems here? Perhaps you’ll let me read them.” He felt, with the familiar chagrin, that she would never, of herself, have thought of asking him. “Yes, my last volume. It’s just out.” He was going for a walk in the rain with Mr. Haseltine that afternoon. There was an old church in the neighbouring village that his friend wanted him to see. Mrs. Baldwin had letters to write. “Will you have time to look at it while we are out?” he asked. Although she had shown so little interest in him, he was eager, pathetically so, he felt, that she should read and care about his poems. She said that it was just the time: her letters would not take long. And so he ran up to his room and got the little book for her: Burnt Offerings. All the time that he was walking with Mr. Haseltine and seeing the church, and the old manor house that took them a half mile further, he wondered what she was thinking about his poems. By the time they had returned the rain had ceased. A warm September sunlight diffused itself. Veils lifted from the stream and trailed upon the lower meadows. The sky grew clear and the leaves all sparkled. They found
- 85. that Mrs. Baldwin had had her cup of tea, for it was past four; but all had been left in readiness for them, the kettle boiling; and after Guy had swallowed his, he went out and saw her walking down among the crocuses. “Oh, you are back?” she said when he joined her. “I wanted to be there to give you your tea. Was it all right?” “Perfectly,” he said. “We put in just your number of spoonfuls.” Mrs. Baldwin wore her little knitted jacket and had put on her white, rubber-soled canvas shoes against the wet; but her head, with its thick, close braids, was bare to the sunlight. “I had to come out as soon as it stopped raining,” she said; “and I’m afraid I simply forgot to look out for you and father.” Her gentleness had always seemed contentment; this afternoon it seemed happiness, and he had never seen her look so young. He wondered if she were going to take him so dreadfully aback as not even to mention his poems; if she had simply forgotten them, too. Already her demeanour, unclouded, almost radiant, inflicted a wound; she had either forgotten, or she had cared little indeed, since she could look like that. But, after he had commented, consentingly, on the lovely hour, she went on with a change of tone, a voice a little shy, “I’ve read the poems. Thank you so much for letting me see them.” “You read all of them?” “Yes. I didn’t write my letters.” “I hope you read them, then, because you cared for them.” She didn’t answer for a moment, walking along and placing the small white feet carefully among the crocuses. “They are very sad,” she then said. He was aware, after an instant of adjustment to the blow, that she made him very angry. Terrible, his poems, searing, scorching; wicked, if one would; but not sad. “Oh!” he murmured; and he wondered if the divided feeling she had from the first roused in him had been this hatred, not perhaps of her, but of her unvarying acquiescence, her untroubled inadequacy. “They interested me very much,” she said, feeling, no doubt, that, whatever he was, he was not pleased. “They made me see, I mean, all the things you have been through.”
- 86. “Sad things, you call them. You know, I rather feel as if I’d heard you call hell sad.” She looked up at him quickly, and it was now she who was taken aback and, as she had been the other day, at a loss. And, as on the other day, she found the same answer, though she offered it deprecatingly, feeling his displeasure. “But hell doesn’t exist.” “Don’t you think anything horrible exists?” They turned at the end of the meadow. It seemed to him, although he felt as if he hated her, that they were suddenly intimate in their antagonism. He would force that antagonism, and its intimacy, upon her—to its last implication. “Horrible? Oh, yes, yes!” she said, startled, and that was, he reflected grimly, to the good. “But it would have to be irretrievable, wouldn’t it, to be hell?” she urged. “Do you suggest that it’s not irretrievable? You own it’s horrible. Irretrievably horrible, I call it. And that’s what I call hell. Yet all that you can find to say of my poems is that they are sad.” She hesitated, feeling her way, hearing in the recurrent word how it had rankled. “I meant sad, I think, because of you; because you had suffered so much.” “You seem always to imply that one might not have suffered!” And thrusting aside her quickly murmured, “Oh, no, no!” he went on: “I can’t understand your attitude of mind. Do you realize at all, I sometimes wonder, what it has all meant, this nightmare we are living in—we, that is, to whom it came? Can you imagine what it was to me to see boys, dead boys, buried stealthily, at night, under fire? Boys so mangled, so disfigured—you read that poem, 'Half a Corpse'?—that their mothers wouldn’t have known them; featureless, dismembered boys, heaped one upon the other in the mud. Has your mind ever dwelt upon the community of corruption in which they lie, as their mothers' minds must dwell? I do not understand you. I do not understand how you can dare to call such things sad.” His own wrath shook and yet sustained him, though he knew a fear lest he had gone too far; but in her silence—they had reached the other end of the meadow and turned again in their walk—he felt that there was no resentment. It was as if she realized that those who have returned from hell cannot be asked to stop and pick their words with courtesy, and accepted his
- 87. vehemence, if not his blame; and again, when she spoke at last, he felt that her bewilderment had settled into thought. “Yes, I can imagine,” she said. “But no, I don’t think that my mind has dwelt on those things. If I were their mothers, I don’t think that my mind would dwell, as you say. Something would burn through. There are other kinds of suffering—better kinds; they help, I believe. And, for that kind, it is worse, but is it so much worse than in ordinary life? That is what happens all the time when there is no war; dreadful changes in the dead; and burials. They are not quite so near each other in a churchyard, and their graves are named; but do you think that makes it easier to bear?” He felt now as if it were insult she was offering him. "You deny all tragedy to war, then? It’s all to you on a level with an Elegy in a Country Churchyard, with curfew and rector and primrose- wreaths? You read 'His Eyes,'"—Guy’s voice had a hoarser note, but, mingled with the sincerity of what, at last, he knew he was to tell her, the very centre of his sick heart, went a surface appreciation of what he had just said and of how curfew and rector and primrose-wreaths would go into a bitter poem one day,—"you read that poem of mine at the end of the book. ‘His Eyes’ is about myself and my friend Ronnie Barlow, the artist; you never heard of him, I know. He hung, with shattered legs, dying, just in front of us, on the barbed wire, for three days and nights. When he could speak, it was to beg to be shot. We tried to get to him, four, five times; it was no good. There was barbed wire between, and the Germans spotted us every time. He died during the third night, and next morning I found him looking at me—as he had looked during these three days—his torment and his reproach. And so he went on looking until the rats came and he had no more eyes to look with. Will you tell me that that is no worse than the deaths died in the parishes of England? Will you tell me that it’s the sort of death died by the cheery, mature gentlemen who ate their dinners and slept warm and dropped a tear—while they did their ‘bit’ in their Government offices—over the brave lads saving England?" He had taken refuge from Ronnie in hatred of those whom, in the poem, he called his murderers, and his voice was weighted with its fierce indictment. In the pause that followed he had time to wonder if she found him, at last, intolerable. She walked beside him, still looking down, and it might well have been in a chill withdrawal. He almost expected to hear her,
- 88. Welcome to our website – the perfect destination for book lovers and knowledge seekers. We believe that every book holds a new world, offering opportunities for learning, discovery, and personal growth. That’s why we are dedicated to bringing you a diverse collection of books, ranging from classic literature and specialized publications to self-development guides and children's books. More than just a book-buying platform, we strive to be a bridge connecting you with timeless cultural and intellectual values. With an elegant, user-friendly interface and a smart search system, you can quickly find the books that best suit your interests. Additionally, our special promotions and home delivery services help you save time and fully enjoy the joy of reading. Join us on a journey of knowledge exploration, passion nurturing, and personal growth every day! ebookbell.com