Methods in Enzymology 478 - Glycomics

METHODS IN ENZYMOLOGY Editors-in-Chief JOHN N. ABELSON AND MELVIN I. SIMON Division of Biology California Institute of ...

Author: Minoru Fukuda

21 downloads 1652 Views 24MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form

DOWNLOAD PDF

METHODS IN ENZYMOLOGY Editors-in-Chief

JOHN N. ABELSON AND MELVIN I. SIMON Division of Biology California Institute of Technology Pasadena, California Founding Editors

SIDNEY P. COLOWICK AND NATHAN O. KAPLAN

Academic Press is an imprint of Elsevier 525 B Street, Suite 1900, San Diego, CA 92101-4495, USA 30 Corporate Drive, Suite 400, Burlington, MA 01803, USA 32 Jamestown Road, London NW1 7BY, UK First edition 2010 Copyright # 2010, Elsevier Inc. All Rights Reserved. No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means electronic, mechanical, photocopying, recording or otherwise without the prior written permission of the publisher Permissions may be sought directly from Elsevier’s Science & Technology Rights Department in Oxford, UK: phone (+44) (0) 1865 843830; fax (+44) (0) 1865 853333; email: permissions@ elsevier.com. Alternatively you can submit your request online by visiting the Elsevier web site at http://elsevier.com/locate/permissions, and selecting Obtaining permission to use Elsevier material Notice No responsibility is assumed by the publisher for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions or ideas contained in the material herein. Because of rapid advances in the medical sciences, in particular, independent verification of diagnoses and drug dosages should be made For information on all Academic Press publications visit our website at elsevierdirect.com ISBN: 978-0-12-381001-4 ISSN: 0076-6879 Printed and bound in United States of America 10 11 12 10 9 8 7 6 5 4 3 2 1

CONTRIBUTORS

Sara Al-Chalabi Division of Molecular Biosciences, Faculty of Natural Sciences, Imperial College London, London, United Kingdom Maho Amano Laboratory of Advanced Chemical Biology, Graduate School of Advanced Life Science, Frontier Research Center for Post-Genome Science and Technology, Hokkaido University, Sapporo, Japan I. Jonathan Amster Department of Chemistry, University of Georgia, Athens, Georgia Hiromune Ando Department of Applied Bioorganic Chemistry, Faculty of Applied Biological Sciences, Gifu University, Yanagido, Gifu, and Institute for Integrated Cell-Material Sciences (iCeMS), Kyoto University, Yoshida Ushinomiya-cho, Sakyo-ku, Kyoto, Japan Aristotelis Antonopoulos Division of Molecular Biosciences, Faculty of Natural Sciences, Imperial College London, London, United Kingdom Timor Baasov The Edith and Joseph Fischer Enzyme Inhibitors Laboratory, Schulich Faculty of Chemistry, Technion—Israel Institute of Technology, Haifa, Israel George Barany Department of Chemistry, University of Minnesota, Minneapolis, Minnesota, USA Adam W. Barb Complex Carbohydrate Research Center, University of Georgia, Athens, Georgia, USA Michelle R. Bond Division of Translational Research, Department of Internal Medicine, University of Texas Southwestern Medical Center, Dallas, and Department of Chemistry, Stanford University, Stanford, California, USA

xv

xvi

Contributors

Andrew J. Borgert Center for Magnetic Resonance Research, University of Minnesota, Minneapolis, Minnesota, USA Terry D. Butters Department of Biochemistry, Oxford Glycobiology Institute, University of Oxford, Oxford, United Kingdom Ke´vin Canis Division of Molecular Biosciences, Faculty of Natural Sciences, Imperial College London, London, United Kingdom Wengang Chai Glycosciences Laboratory, Faculty of Medicine, Imperial College London, Northwick Park Hospital Campus, Harrow, Middlesex, United Kingdom Yasunori Chiba Research Center for Medical Glycoscience, National Institute of Advanced Industrial Science and Technology (AIST), Open Space Laboratory C-2, Umezono, Tsukuba, Ibaraki, Japan Robert Childs Glycosciences Laboratory, Faculty of Medicine, Imperial College London, Northwick Park Hospital Campus, Harrow, Middlesex, United Kingdom Richard D. Cummings Department of Biochemistry, Emory University School of Medicine, Atlanta, Georgia, USA Anne Dell Division of Molecular Biosciences, Faculty of Natural Sciences, Imperial College London, London, United Kingdom Steffen Eller Department of Biomolecular Systems, Max Planck Institute of Colloids and Interfaces, Research Campus Potsdam-Golm, Potsdam, and Department of Biology, Chemistry and Pharmacy, Institute of Chemistry and Biochemistry, Free University Berlin, Berlin, Germany A. Tony Etienne Division of Molecular Biosciences, Faculty of Natural Sciences, Imperial College London, London, United Kingdom Ten Feizi Glycosciences Laboratory, Faculty of Medicine, Imperial College London, Northwick Park Hospital Campus, Harrow, Middlesex, United Kingdom

Contributors

xvii

Yukari Fujimoto Department of Chemistry, Graduate School of Science, Osaka University, Machikaneyama, Toyonaka, Osaka, Japan Koichi Fukase Department of Chemistry, Graduate School of Science, Osaka University, Machikaneyama, Toyonaka, Osaka, Japan Michiko N. Fukuda Glycobiology Unit, Tumor Microenvironment Program, Cancer Center, SanfordBurnham Medical Research Institute, La Jolla, California, USA Paola Grassi Division of Molecular Biosciences, Faculty of Natural Sciences, Imperial College London, London, United Kingdom Naoko Goto-Inoue Department of Molecular Anatomy, Hamamatsu University School of Medicine, Handayama, Higashi-ku, Hamamatsu, Shizuoka, Japan Rebecca Harrison Division of Molecular Biosciences, Faculty of Natural Sciences, Imperial College London, London, United Kingdom Stuart M. Haslam Division of Molecular Biosciences, Faculty of Natural Sciences, Imperial College London, London, United Kingdom Takahiro Hayasaka Department of Molecular Anatomy, Hamamatsu University School of Medicine, Handayama, Higashi-ku, Hamamatsu, Shizuoka, Japan Jun Hirabayashi Research Center for Medical Glycoscience, National Institute of Advanced Industrial Science and Technology (AIST), Central 2, Umezono, Ibaraki, Japan Yuzuru Ikehara Research Center for Medical Glycoscience, National Institute of Advanced Industrial Science and Technology (AIST), Central 2, Umezono, Ibaraki, Japan Hideharu Ishida Department of Applied Bioorganic Chemistry, Faculty of Applied Biological Sciences, Gifu University, Yanagido, Gifu, Japan Mohd Nazri Ismail Division of Molecular Biosciences, Faculty of Natural Sciences, Imperial College London, London, United Kingdom

xviii

Contributors

Yoko Itakura Research Center for Medical Glycoscience, National Institute of Advanced Industrial Science and Technology (AIST), Central 2, Umezono, Ibaraki, Japan Hiromi Ito Research Center for Medical Glycoscience, National Institute of Advanced Industrial Science and Technology (AIST), Open Space Laboratory C-2, Umezono, Tsukuba, Ibaraki, Japan Masayuki Izumi Department of Chemistry, Graduate School of Science, Osaka University, Machikaneyama, Toyonaka, Osaka, Japan Jihye Jang-Lee Division of Molecular Biosciences, Faculty of Natural Sciences, Imperial College London, London, United Kingdom Yasuhiro Kajihara Department of Chemistry, Graduate School of Science, Osaka University, Machikaneyama, Toyonaka, Osaka, Japan Akihiko Kameyama Research Center for Medical Glycoscience, National Institute of Advanced Industrial Science and Technology (AIST), Open Space Laboratory C-2, Umezono, Tsukuba, Ibaraki, Japan Jeyakumar Kandasamy The Edith and Joseph Fischer Enzyme Inhibitors Laboratory, Schulich Faculty of Chemistry, Technion—Israel Institute of Technology, Haifa, Israel Koichi Kato Institute for Molecular Science and Okazaki Institute for Integrative Bioscience, National Institutes of Natural Sciences, Higashiyama, Myodaiji, Okazaki, Aichi, and Department of Structural Biology and Biomolecular Engineering, Graduate School of Pharmaceutical Sciences, Nagoya City University, Tanabe-dori, Mizuho-ku, Nagoya; The Glycoscience Institute, Ochanomizu University, Ohtsuka, Bunkyo-ku, Tokyo, Japan Norihito Kawasaki Department of Integrated Biosciences, Graduate School of Frontier Sciences, The University of Tokyo, Chiba, Japan Kay-Hooi Khoo Institute of Biological Chemistry, Academia Sinica, Nankang, Taipei, Taiwan Makoto Kiso Department of Applied Bioorganic Chemistry, Faculty of Applied Biological Sciences, Gifu University, Yanagido, Gifu, and Institute for Integrated Cell-Material

Contributors

xix

Sciences (iCeMS), Kyoto University, Yoshida Ushinomiya-cho, Sakyo-ku, Kyoto, Japan Ken Kitajima Bioscience and Biotechnology Center, Graduate School of Bioagricultural Science, Nagoya University, Nagoya, Japan Jennifer J. Kohler Division of Translational Research, Department of Internal Medicine, University of Texas Southwestern Medical Center, Dallas, USA Atsushi Kuno Research Center for Medical Glycoscience, National Institute of Advanced Industrial Science and Technology (AIST), Central 2, Umezono, Ibaraki, Japan Tatiana N. Laremore Department of Chemistry and Chemical Biology, Rensselaer Polytechnic Institute, Troy, New York, USA Franklin E. Leach III Department of Chemistry, University of Georgia, Athens, Georgia ¨nen Anne Leppa Department of Biosciences, Division of Biochemistry, University of Helsinki, Viikinkaari, Helsinki, Finland Robert J. Linhardt Department of Chemistry and Chemical Biology, and Departments of Chemical and Biological Engineering and Biology, Center for Biotechnology and Interdisciplinary Studies, Rensselaer Polytechnic Institute, Troy, New York, USA Mian Liu Complex Carbohydrate Research Center, University of Georgia, Athens, Georgia, USA Yan Liu Glycosciences Laboratory, Faculty of Medicine, Imperial College London, Northwick Park Hospital Campus, Harrow, Middlesex, United Kingdom David Live Complex Carbohydrate Research Center, University of Georgia, Athens, Georgia, USA Todd L. Lowary The Alberta Ingenuity Centre for Carbohydrate Science, Department of Chemistry, University of Alberta, Edmonton, Alberta, Canada Shino Manabe RIKEN Advanced Science Institute, Saitama, Japan

xx

Contributors

Atsushi Matsuda Research Center for Medical Glycoscience, National Institute of Advanced Industrial Science and Technology (AIST), Central 2, Umezono, Ibaraki, Japan Hitoshi Matsumoto Department of Molecular Biochemistry and Clinical Investigation, Osaka University Graduate School of Medicine, Yamada-oka, Suita, Japan Takahiko Matsushita Graduate School of Advanced Life Science and Frontier Research Center for Post-Genome Science and Technology, Hokkaido University, Sapporo, Japan Eiji Miyoshi Department of Molecular Biochemistry and Clinical Investigation, Osaka University Graduate School of Medicine, Yamada-oka, Suita, Japan Kenta Moriwaki Department of Molecular Biochemistry and Clinical Investigation, Osaka University Graduate School of Medicine, Yamada-oka, Suita, Japan Claudia Muhle-Goll European Molecular Biology Laboratory, Heidelberg, Germany Hisashi Narimatsu Research Center for Medical Glycoscience, National Institute of Advanced Industrial Science and Technology (AIST), Open Space Laboratory C-2, Umezono, Tsukuba, Ibaraki, Japan Shin-Ichiro Nishimura Laboratory of Advanced Chemical Biology, Graduate School of Advanced Life Science, Frontier Research Center for Post-Genome Science and Technology, Hokkaido University, Sapporo, Japan Simon J. North Division of Molecular Biosciences, Faculty of Natural Sciences, Imperial College London, London, United Kingdom Igor Nudelman The Edith and Joseph Fischer Enzyme Inhibitors Laboratory, Schulich Faculty of Chemistry, Technion—Israel Institute of Technology, Haifa, Israel Mary K. O’Reilly Departments of Chemical Physiology and Molecular Biology, The Scripps Research Institute, La Jolla, California, USA Ryo Okamoto Department of Chemistry, Graduate School of Science, Osaka University, Machikaneyama, Toyonaka, Osaka, Japan

Contributors

xxi

Angelina S. Palma Glycosciences Laboratory, Faculty of Medicine, Imperial College London, Northwick Park Hospital Campus, Harrow, Middlesex, United Kingdom Poh-Choo Pang Division of Molecular Biosciences, Faculty of Natural Sciences, Imperial College London, London, United Kingdom James C. Paulson Departments of Chemical Physiology and Molecular Biology, The Scripps Research Institute, La Jolla, California, USA Varvara Pokrovskaya The Edith and Joseph Fischer Enzyme Inhibitors Laboratory, Schulich Faculty of Chemistry, Technion—Israel Institute of Technology, Haifa, Israel Myles B. Poulin The Alberta Ingenuity Centre for Carbohydrate Science, Department of Chemistry, University of Alberta, Edmonton, Alberta, Canada Chihiro Sato Bioscience and Biotechnology Center, Graduate School of Bioagricultural Sciences, Nagoya University, Nagoya, Japan Takashi Sato Research Center for Medical Glycoscience, National Institute of Advanced Industrial Science and Technology (AIST), Open Space Laboratory C-2, Umezono, Tsukuba, Ibaraki, Japan Peter H. Seeberger Department of Biomolecular Systems, Max Planck Institute of Colloids and Interfaces, Research Campus Potsdam-Golm, Potsdam, and Department of Biology, Chemistry and Pharmacy, Institute of Chemistry and Biochemistry, Free University Berlin, Arnimallee, Berlin, Germany Mitsutoshi Setou Department of Molecular Anatomy, Hamamatsu University School of Medicine, Handayama, Higashi-ku, Hamamatsu, Shizuoka, Japan Atsushi Shimoyama Department of Chemistry, Graduate School of Science, Osaka University, Machikaneyama, Toyonaka, Osaka, Japan Shinichiro Shinzaki Department of Molecular Biochemistry and Clinical Investigation, Osaka University Graduate School of Medicine, Yamada-oka, Suita, Japan

xxii

Contributors

Kemal Solakyildirim Department of Chemistry and Chemical Biology, Rensselaer Polytechnic Institute, Troy, New York, USA Katsunori Tanaka Department of Chemistry, Graduate School of Science, Osaka University, Machikaneyama, Toyonaka, Osaka, Japan Hiroaki Tateno Research Center for Medical Glycoscience, National Institute of Advanced Industrial Science and Technology (AIST), Central 2, Umezono, Ibaraki, Japan Alana Trollope Division of Molecular Biosciences, Faculty of Natural Sciences, Imperial College London, London, United Kingdom Markus Weishaupt Department of Biomolecular Systems, Max Planck Institute of Colloids and Interfaces, Research Campus Potsdam-Golm, Potsdam, and Department of Biology, Chemistry and Pharmacy, Institute of Chemistry and Biochemistry, Free University Berlin, Berlin, Germany Chad M. Whitman Division of Translational Research, Department of Internal Medicine, University of Texas Southwestern Medical Center, Dallas, and Department of Chemistry, Stanford University, Stanford, California, USA Yoshiki Yamaguchi Structural Glycobiology Team, Systems Glycobiology Research Group, Chemical Biology Department, RIKEN, Advanced Science Institute, Hirosawa, Wako, Japan Nao Yamakawa Bioscience and Biotechnology Center, Graduate School of Bioagricultural Sciences, Nagoya University, Nagoya, Japan Kazuo Yamamoto Department of Integrated Biosciences, Graduate School of Frontier Sciences, The University of Tokyo, Chiba, Japan Naoki Yamamoto Department of Chemistry, Graduate School of Science, Osaka University, Machikaneyama, Toyonaka, Osaka, Japan Jian Yin Department of Biomolecular Systems, Max Planck Institute of Colloids and Interfaces, Research Campus Potsdam-Golm, Potsdam, and Department of Biology, Chemistry and Pharmacy, Institute of Chemistry and Biochemistry, Free University Berlin, Arnimallee, Berlin, Germany

Contributors

xxiii

Tohru Yoneyama Department of Urology, Hirosaki University of Medicine, Hirosaki, Japan Seok-Ho Yu Division of Translational Research, Department of Internal Medicine, University of Texas Southwestern Medical Center, Dallas, USA Shin-Yi Yu Institute of Biological Chemistry, Academia Sinica, Nankang, Taipei, Taiwan Yibing Zhang Glycosciences Laboratory, Faculty of Medicine, Imperial College London, Northwick Park Hospital Campus, Harrow, Middlesex, United Kingdom

PREFACE

In 2006, we published three volumes in the Methods in Enzymology dedicated to Glycobiology field as follows: Glycobiology (Volume 415), Glycomics (Volume 416), and Functional Glycomics (Volume 417). We have seen a tremendous progress in the Glycobiology field since then. In particular, an explosive progress has been made in immunology, neuroglycbiology, glycomics, signal transduction, and many other disciplines, examining each unique system and employing new technology. The Academic Press kindly gave another opportunity to update the introduction of new methods to a large variety of readers who like to contribute to the advancement of Glycosciences. In the current series of Methods in Enzymology, Glycomics (Volume 478), Functional Glycomics (Volume 479), and Glycobiology (Volume 480) have been dedicated to disseminate information on the methods of determining the biological roles of carbohydrates, thanks to Academic Press Manager, particularly to Ms. Zoe Kruze and Ms. Delsy Retchagar. In the current book A Glycomics (Volume 478), wide topics of glycomics are covered, and glycomics revealed by mass spectrometric analysis, by carbohydrate-binding proteins, and chemical glycobiology are described. The latter include protein–carbohydrate interaction, synthetic carbohydrate chemistry, and identification of carbohydrate-binding protein by carbohydrate mimicry peptides. I have tried to present as new development as possible of these expanding fields in this book. The second volume (Volume 479) covers new development in glycosciences, including functional studies of glycosylation in stem cells, functions revealed by gene knockout mouse, glycan defects in muscular dystrophy and carcinoma. The third volume (Volume 480) covers proteoglycan function, infection, immunity, and carbohydrate-binding proteins including galectin, and new development including O-glycosylation in Notch and other signaling. I believe that we have a collection of outstanding contributors who represent respective expertise and field. I believe that this book will be useful to a wide variety of readers from graduate students, researchers in academic, and industry, to those who would like to teach Glycobiology and Glycosciences at various levels. We hope that this book will contribute to further explosive progress in Glycosciences and Glycobiology. MINORU FUKUDA xxv

METHODS IN ENZYMOLOGY

VOLUME I. Preparation and Assay of Enzymes Edited by SIDNEY P. COLOWICK AND NATHAN O. KAPLAN VOLUME II. Preparation and Assay of Enzymes Edited by SIDNEY P. COLOWICK AND NATHAN O. KAPLAN VOLUME III. Preparation and Assay of Substrates Edited by SIDNEY P. COLOWICK AND NATHAN O. KAPLAN VOLUME IV. Special Techniques for the Enzymologist Edited by SIDNEY P. COLOWICK AND NATHAN O. KAPLAN VOLUME V. Preparation and Assay of Enzymes Edited by SIDNEY P. COLOWICK AND NATHAN O. KAPLAN VOLUME VI. Preparation and Assay of Enzymes (Continued) Preparation and Assay of Substrates Special Techniques Edited by SIDNEY P. COLOWICK AND NATHAN O. KAPLAN VOLUME VII. Cumulative Subject Index Edited by SIDNEY P. COLOWICK AND NATHAN O. KAPLAN VOLUME VIII. Complex Carbohydrates Edited by ELIZABETH F. NEUFELD AND VICTOR GINSBURG VOLUME IX. Carbohydrate Metabolism Edited by WILLIS A. WOOD VOLUME X. Oxidation and Phosphorylation Edited by RONALD W. ESTABROOK AND MAYNARD E. PULLMAN VOLUME XI. Enzyme Structure Edited by C. H. W. HIRS VOLUME XII. Nucleic Acids (Parts A and B) Edited by LAWRENCE GROSSMAN AND KIVIE MOLDAVE VOLUME XIII. Citric Acid Cycle Edited by J. M. LOWENSTEIN VOLUME XIV. Lipids Edited by J. M. LOWENSTEIN VOLUME XV. Steroids and Terpenoids Edited by RAYMOND B. CLAYTON xxvii

xxviii

Methods in Enzymology

VOLUME XVI. Fast Reactions Edited by KENNETH KUSTIN VOLUME XVII. Metabolism of Amino Acids and Amines (Parts A and B) Edited by HERBERT TABOR AND CELIA WHITE TABOR VOLUME XVIII. Vitamins and Coenzymes (Parts A, B, and C) Edited by DONALD B. MCCORMICK AND LEMUEL D. WRIGHT VOLUME XIX. Proteolytic Enzymes Edited by GERTRUDE E. PERLMANN AND LASZLO LORAND VOLUME XX. Nucleic Acids and Protein Synthesis (Part C) Edited by KIVIE MOLDAVE AND LAWRENCE GROSSMAN VOLUME XXI. Nucleic Acids (Part D) Edited by LAWRENCE GROSSMAN AND KIVIE MOLDAVE VOLUME XXII. Enzyme Purification and Related Techniques Edited by WILLIAM B. JAKOBY VOLUME XXIII. Photosynthesis (Part A) Edited by ANTHONY SAN PIETRO VOLUME XXIV. Photosynthesis and Nitrogen Fixation (Part B) Edited by ANTHONY SAN PIETRO VOLUME XXV. Enzyme Structure (Part B) Edited by C. H. W. HIRS AND SERGE N. TIMASHEFF VOLUME XXVI. Enzyme Structure (Part C) Edited by C. H. W. HIRS AND SERGE N. TIMASHEFF VOLUME XXVII. Enzyme Structure (Part D) Edited by C. H. W. HIRS AND SERGE N. TIMASHEFF VOLUME XXVIII. Complex Carbohydrates (Part B) Edited by VICTOR GINSBURG VOLUME XXIX. Nucleic Acids and Protein Synthesis (Part E) Edited by LAWRENCE GROSSMAN AND KIVIE MOLDAVE VOLUME XXX. Nucleic Acids and Protein Synthesis (Part F) Edited by KIVIE MOLDAVE AND LAWRENCE GROSSMAN VOLUME XXXI. Biomembranes (Part A) Edited by SIDNEY FLEISCHER AND LESTER PACKER VOLUME XXXII. Biomembranes (Part B) Edited by SIDNEY FLEISCHER AND LESTER PACKER VOLUME XXXIII. Cumulative Subject Index Volumes I-XXX Edited by MARTHA G. DENNIS AND EDWARD A. DENNIS VOLUME XXXIV. Affinity Techniques (Enzyme Purification: Part B) Edited by WILLIAM B. JAKOBY AND MEIR WILCHEK


VOLUME XXXV. Lipids (Part B) Edited by JOHN M. LOWENSTEIN VOLUME XXXVI. Hormone Action (Part A: Steroid Hormones) Edited by BERT W. O’MALLEY AND JOEL G. HARDMAN VOLUME XXXVII. Hormone Action (Part B: Peptide Hormones) Edited by BERT W. O’MALLEY AND JOEL G. HARDMAN VOLUME XXXVIII. Hormone Action (Part C: Cyclic Nucleotides) Edited by JOEL G. HARDMAN AND BERT W. O’MALLEY VOLUME XXXIX. Hormone Action (Part D: Isolated Cells, Tissues, and Organ Systems) Edited by JOEL G. HARDMAN AND BERT W. O’MALLEY VOLUME XL. Hormone Action (Part E: Nuclear Structure and Function) Edited by BERT W. O’MALLEY AND JOEL G. HARDMAN VOLUME XLI. Carbohydrate Metabolism (Part B) Edited by W. A. WOOD VOLUME XLII. Carbohydrate Metabolism (Part C) Edited by W. A. WOOD VOLUME XLIII. Antibiotics Edited by JOHN H. HASH VOLUME XLIV. Immobilized Enzymes Edited by KLAUS MOSBACH VOLUME XLV. Proteolytic Enzymes (Part B) Edited by LASZLO LORAND VOLUME XLVI. Affinity Labeling Edited by WILLIAM B. JAKOBY AND MEIR WILCHEK VOLUME XLVII. Enzyme Structure (Part E) Edited by C. H. W. HIRS AND SERGE N. TIMASHEFF VOLUME XLVIII. Enzyme Structure (Part F) Edited by C. H. W. HIRS AND SERGE N. TIMASHEFF VOLUME XLIX. Enzyme Structure (Part G) Edited by C. H. W. HIRS AND SERGE N. TIMASHEFF VOLUME L. Complex Carbohydrates (Part C) Edited by VICTOR GINSBURG VOLUME LI. Purine and Pyrimidine Nucleotide Metabolism Edited by PATRICIA A. HOFFEE AND MARY ELLEN JONES VOLUME LII. Biomembranes (Part C: Biological Oxidations) Edited by SIDNEY FLEISCHER AND LESTER PACKER

xxix

xxx


VOLUME LIII. Biomembranes (Part D: Biological Oxidations) Edited by SIDNEY FLEISCHER AND LESTER PACKER VOLUME LIV. Biomembranes (Part E: Biological Oxidations) Edited by SIDNEY FLEISCHER AND LESTER PACKER VOLUME LV. Biomembranes (Part F: Bioenergetics) Edited by SIDNEY FLEISCHER AND LESTER PACKER VOLUME LVI. Biomembranes (Part G: Bioenergetics) Edited by SIDNEY FLEISCHER AND LESTER PACKER VOLUME LVII. Bioluminescence and Chemiluminescence Edited by MARLENE A. DELUCA VOLUME LVIII. Cell Culture Edited by WILLIAM B. JAKOBY AND IRA PASTAN VOLUME LIX. Nucleic Acids and Protein Synthesis (Part G) Edited by KIVIE MOLDAVE AND LAWRENCE GROSSMAN VOLUME LX. Nucleic Acids and Protein Synthesis (Part H) Edited by KIVIE MOLDAVE AND LAWRENCE GROSSMAN VOLUME 61. Enzyme Structure (Part H) Edited by C. H. W. HIRS AND SERGE N. TIMASHEFF VOLUME 62. Vitamins and Coenzymes (Part D) Edited by DONALD B. MCCORMICK AND LEMUEL D. WRIGHT VOLUME 63. Enzyme Kinetics and Mechanism (Part A: Initial Rate and Inhibitor Methods) Edited by DANIEL L. PURICH VOLUME 64. Enzyme Kinetics and Mechanism (Part B: Isotopic Probes and Complex Enzyme Systems) Edited by DANIEL L. PURICH VOLUME 65. Nucleic Acids (Part I) Edited by LAWRENCE GROSSMAN AND KIVIE MOLDAVE VOLUME 66. Vitamins and Coenzymes (Part E) Edited by DONALD B. MCCORMICK AND LEMUEL D. WRIGHT VOLUME 67. Vitamins and Coenzymes (Part F) Edited by DONALD B. MCCORMICK AND LEMUEL D. WRIGHT VOLUME 68. Recombinant DNA Edited by RAY WU VOLUME 69. Photosynthesis and Nitrogen Fixation (Part C) Edited by ANTHONY SAN PIETRO VOLUME 70. Immunochemical Techniques (Part A) Edited by HELEN VAN VUNAKIS AND JOHN J. LANGONE


xxxi

VOLUME 71. Lipids (Part C) Edited by JOHN M. LOWENSTEIN VOLUME 72. Lipids (Part D) Edited by JOHN M. LOWENSTEIN VOLUME 73. Immunochemical Techniques (Part B) Edited by JOHN J. LANGONE AND HELEN VAN VUNAKIS VOLUME 74. Immunochemical Techniques (Part C) Edited by JOHN J. LANGONE AND HELEN VAN VUNAKIS VOLUME 75. Cumulative Subject Index Volumes XXXI, XXXII, XXXIV–LX Edited by EDWARD A. DENNIS AND MARTHA G. DENNIS VOLUME 76. Hemoglobins Edited by ERALDO ANTONINI, LUIGI ROSSI-BERNARDI, AND EMILIA CHIANCONE VOLUME 77. Detoxication and Drug Metabolism Edited by WILLIAM B. JAKOBY VOLUME 78. Interferons (Part A) Edited by SIDNEY PESTKA VOLUME 79. Interferons (Part B) Edited by SIDNEY PESTKA VOLUME 80. Proteolytic Enzymes (Part C) Edited by LASZLO LORAND VOLUME 81. Biomembranes (Part H: Visual Pigments and Purple Membranes, I) Edited by LESTER PACKER VOLUME 82. Structural and Contractile Proteins (Part A: Extracellular Matrix) Edited by LEON W. CUNNINGHAM AND DIXIE W. FREDERIKSEN VOLUME 83. Complex Carbohydrates (Part D) Edited by VICTOR GINSBURG VOLUME 84. Immunochemical Techniques (Part D: Selected Immunoassays) Edited by JOHN J. LANGONE AND HELEN VAN VUNAKIS VOLUME 85. Structural and Contractile Proteins (Part B: The Contractile Apparatus and the Cytoskeleton) Edited by DIXIE W. FREDERIKSEN AND LEON W. CUNNINGHAM VOLUME 86. Prostaglandins and Arachidonate Metabolites Edited by WILLIAM E. M. LANDS AND WILLIAM L. SMITH VOLUME 87. Enzyme Kinetics and Mechanism (Part C: Intermediates, Stereo-chemistry, and Rate Studies) Edited by DANIEL L. PURICH VOLUME 88. Biomembranes (Part I: Visual Pigments and Purple Membranes, II) Edited by LESTER PACKER

xxxii


VOLUME 89. Carbohydrate Metabolism (Part D) Edited by WILLIS A. WOOD VOLUME 90. Carbohydrate Metabolism (Part E) Edited by WILLIS A. WOOD VOLUME 91. Enzyme Structure (Part I) Edited by C. H. W. HIRS AND SERGE N. TIMASHEFF VOLUME 92. Immunochemical Techniques (Part E: Monoclonal Antibodies and General Immunoassay Methods) Edited by JOHN J. LANGONE AND HELEN VAN VUNAKIS VOLUME 93. Immunochemical Techniques (Part F: Conventional Antibodies, Fc Receptors, and Cytotoxicity) Edited by JOHN J. LANGONE AND HELEN VAN VUNAKIS VOLUME 94. Polyamines Edited by HERBERT TABOR AND CELIA WHITE TABOR VOLUME 95. Cumulative Subject Index Volumes 61–74, 76–80 Edited by EDWARD A. DENNIS AND MARTHA G. DENNIS VOLUME 96. Biomembranes [Part J: Membrane Biogenesis: Assembly and Targeting (General Methods; Eukaryotes)] Edited by SIDNEY FLEISCHER AND BECCA FLEISCHER VOLUME 97. Biomembranes [Part K: Membrane Biogenesis: Assembly and Targeting (Prokaryotes, Mitochondria, and Chloroplasts)] Edited by SIDNEY FLEISCHER AND BECCA FLEISCHER VOLUME 98. Biomembranes (Part L: Membrane Biogenesis: Processing and Recycling) Edited by SIDNEY FLEISCHER AND BECCA FLEISCHER VOLUME 99. Hormone Action (Part F: Protein Kinases) Edited by JACKIE D. CORBIN AND JOEL G. HARDMAN VOLUME 100. Recombinant DNA (Part B) Edited by RAY WU, LAWRENCE GROSSMAN, AND KIVIE MOLDAVE VOLUME 101. Recombinant DNA (Part C) Edited by RAY WU, LAWRENCE GROSSMAN, AND KIVIE MOLDAVE VOLUME 102. Hormone Action (Part G: Calmodulin and Calcium-Binding Proteins) Edited by ANTHONY R. MEANS AND BERT W. O’MALLEY VOLUME 103. Hormone Action (Part H: Neuroendocrine Peptides) Edited by P. MICHAEL CONN VOLUME 104. Enzyme Purification and Related Techniques (Part C) Edited by WILLIAM B. JAKOBY


xxxiii

VOLUME 105. Oxygen Radicals in Biological Systems Edited by LESTER PACKER VOLUME 106. Posttranslational Modifications (Part A) Edited by FINN WOLD AND KIVIE MOLDAVE VOLUME 107. Posttranslational Modifications (Part B) Edited by FINN WOLD AND KIVIE MOLDAVE VOLUME 108. Immunochemical Techniques (Part G: Separation and Characterization of Lymphoid Cells) Edited by GIOVANNI DI SABATO, JOHN J. LANGONE, AND HELEN VAN VUNAKIS VOLUME 109. Hormone Action (Part I: Peptide Hormones) Edited by LUTZ BIRNBAUMER AND BERT W. O’MALLEY VOLUME 110. Steroids and Isoprenoids (Part A) Edited by JOHN H. LAW AND HANS C. RILLING VOLUME 111. Steroids and Isoprenoids (Part B) Edited by JOHN H. LAW AND HANS C. RILLING VOLUME 112. Drug and Enzyme Targeting (Part A) Edited by KENNETH J. WIDDER AND RALPH GREEN VOLUME 113. Glutamate, Glutamine, Glutathione, and Related Compounds Edited by ALTON MEISTER VOLUME 114. Diffraction Methods for Biological Macromolecules (Part A) Edited by HAROLD W. WYCKOFF, C. H. W. HIRS, AND SERGE N. TIMASHEFF VOLUME 115. Diffraction Methods for Biological Macromolecules (Part B) Edited by HAROLD W. WYCKOFF, C. H. W. HIRS, AND SERGE N. TIMASHEFF VOLUME 116. Immunochemical Techniques (Part H: Effectors and Mediators of Lymphoid Cell Functions) Edited by GIOVANNI DI SABATO, JOHN J. LANGONE, AND HELEN VAN VUNAKIS VOLUME 117. Enzyme Structure (Part J) Edited by C. H. W. HIRS AND SERGE N. TIMASHEFF VOLUME 118. Plant Molecular Biology Edited by ARTHUR WEISSBACH AND HERBERT WEISSBACH VOLUME 119. Interferons (Part C) Edited by SIDNEY PESTKA VOLUME 120. Cumulative Subject Index Volumes 81–94, 96–101 VOLUME 121. Immunochemical Techniques (Part I: Hybridoma Technology and Monoclonal Antibodies) Edited by JOHN J. LANGONE AND HELEN VAN VUNAKIS VOLUME 122. Vitamins and Coenzymes (Part G) Edited by FRANK CHYTIL AND DONALD B. MCCORMICK

xxxiv


VOLUME 123. Vitamins and Coenzymes (Part H) Edited by FRANK CHYTIL AND DONALD B. MCCORMICK VOLUME 124. Hormone Action (Part J: Neuroendocrine Peptides) Edited by P. MICHAEL CONN VOLUME 125. Biomembranes (Part M: Transport in Bacteria, Mitochondria, and Chloroplasts: General Approaches and Transport Systems) Edited by SIDNEY FLEISCHER AND BECCA FLEISCHER VOLUME 126. Biomembranes (Part N: Transport in Bacteria, Mitochondria, and Chloroplasts: Protonmotive Force) Edited by SIDNEY FLEISCHER AND BECCA FLEISCHER VOLUME 127. Biomembranes (Part O: Protons and Water: Structure and Translocation) Edited by LESTER PACKER VOLUME 128. Plasma Lipoproteins (Part A: Preparation, Structure, and Molecular Biology) Edited by JERE P. SEGREST AND JOHN J. ALBERS VOLUME 129. Plasma Lipoproteins (Part B: Characterization, Cell Biology, and Metabolism) Edited by JOHN J. ALBERS AND JERE P. SEGREST VOLUME 130. Enzyme Structure (Part K) Edited by C. H. W. HIRS AND SERGE N. TIMASHEFF VOLUME 131. Enzyme Structure (Part L) Edited by C. H. W. HIRS AND SERGE N. TIMASHEFF VOLUME 132. Immunochemical Techniques (Part J: Phagocytosis and Cell-Mediated Cytotoxicity) Edited by GIOVANNI DI SABATO AND JOHANNES EVERSE VOLUME 133. Bioluminescence and Chemiluminescence (Part B) Edited by MARLENE DELUCA AND WILLIAM D. MCELROY VOLUME 134. Structural and Contractile Proteins (Part C: The Contractile Apparatus and the Cytoskeleton) Edited by RICHARD B. VALLEE VOLUME 135. Immobilized Enzymes and Cells (Part B) Edited by KLAUS MOSBACH VOLUME 136. Immobilized Enzymes and Cells (Part C) Edited by KLAUS MOSBACH VOLUME 137. Immobilized Enzymes and Cells (Part D) Edited by KLAUS MOSBACH VOLUME 138. Complex Carbohydrates (Part E) Edited by VICTOR GINSBURG


xxxv

VOLUME 139. Cellular Regulators (Part A: Calcium- and Calmodulin-Binding Proteins) Edited by ANTHONY R. MEANS AND P. MICHAEL CONN VOLUME 140. Cumulative Subject Index Volumes 102–119, 121–134 VOLUME 141. Cellular Regulators (Part B: Calcium and Lipids) Edited by P. MICHAEL CONN AND ANTHONY R. MEANS VOLUME 142. Metabolism of Aromatic Amino Acids and Amines Edited by SEYMOUR KAUFMAN VOLUME 143. Sulfur and Sulfur Amino Acids Edited by WILLIAM B. JAKOBY AND OWEN GRIFFITH VOLUME 144. Structural and Contractile Proteins (Part D: Extracellular Matrix) Edited by LEON W. CUNNINGHAM VOLUME 145. Structural and Contractile Proteins (Part E: Extracellular Matrix) Edited by LEON W. CUNNINGHAM VOLUME 146. Peptide Growth Factors (Part A) Edited by DAVID BARNES AND DAVID A. SIRBASKU VOLUME 147. Peptide Growth Factors (Part B) Edited by DAVID BARNES AND DAVID A. SIRBASKU VOLUME 148. Plant Cell Membranes Edited by LESTER PACKER AND ROLAND DOUCE VOLUME 149. Drug and Enzyme Targeting (Part B) Edited by RALPH GREEN AND KENNETH J. WIDDER VOLUME 150. Immunochemical Techniques (Part K: In Vitro Models of B and T Cell Functions and Lymphoid Cell Receptors) Edited by GIOVANNI DI SABATO VOLUME 151. Molecular Genetics of Mammalian Cells Edited by MICHAEL M. GOTTESMAN VOLUME 152. Guide to Molecular Cloning Techniques Edited by SHELBY L. BERGER AND ALAN R. KIMMEL VOLUME 153. Recombinant DNA (Part D) Edited by RAY WU AND LAWRENCE GROSSMAN VOLUME 154. Recombinant DNA (Part E) Edited by RAY WU AND LAWRENCE GROSSMAN VOLUME 155. Recombinant DNA (Part F) Edited by RAY WU VOLUME 156. Biomembranes (Part P: ATP-Driven Pumps and Related Transport: The Na, K-Pump) Edited by SIDNEY FLEISCHER AND BECCA FLEISCHER

xxxvi


VOLUME 157. Biomembranes (Part Q: ATP-Driven Pumps and Related Transport: Calcium, Proton, and Potassium Pumps) Edited by SIDNEY FLEISCHER AND BECCA FLEISCHER VOLUME 158. Metalloproteins (Part A) Edited by JAMES F. RIORDAN AND BERT L. VALLEE VOLUME 159. Initiation and Termination of Cyclic Nucleotide Action Edited by JACKIE D. CORBIN AND ROGER A. JOHNSON VOLUME 160. Biomass (Part A: Cellulose and Hemicellulose) Edited by WILLIS A. WOOD AND SCOTT T. KELLOGG VOLUME 161. Biomass (Part B: Lignin, Pectin, and Chitin) Edited by WILLIS A. WOOD AND SCOTT T. KELLOGG VOLUME 162. Immunochemical Techniques (Part L: Chemotaxis and Inflammation) Edited by GIOVANNI DI SABATO VOLUME 163. Immunochemical Techniques (Part M: Chemotaxis and Inflammation) Edited by GIOVANNI DI SABATO VOLUME 164. Ribosomes Edited by HARRY F. NOLLER, JR., AND KIVIE MOLDAVE VOLUME 165. Microbial Toxins: Tools for Enzymology Edited by SIDNEY HARSHMAN VOLUME 166. Branched-Chain Amino Acids Edited by ROBERT HARRIS AND JOHN R. SOKATCH VOLUME 167. Cyanobacteria Edited by LESTER PACKER AND ALEXANDER N. GLAZER VOLUME 168. Hormone Action (Part K: Neuroendocrine Peptides) Edited by P. MICHAEL CONN VOLUME 169. Platelets: Receptors, Adhesion, Secretion (Part A) Edited by JACEK HAWIGER VOLUME 170. Nucleosomes Edited by PAUL M. WASSARMAN AND ROGER D. KORNBERG VOLUME 171. Biomembranes (Part R: Transport Theory: Cells and Model Membranes) Edited by SIDNEY FLEISCHER AND BECCA FLEISCHER VOLUME 172. Biomembranes (Part S: Transport: Membrane Isolation and Characterization) Edited by SIDNEY FLEISCHER AND BECCA FLEISCHER


xxxvii

VOLUME 173. Biomembranes [Part T: Cellular and Subcellular Transport: Eukaryotic (Nonepithelial) Cells] Edited by SIDNEY FLEISCHER AND BECCA FLEISCHER VOLUME 174. Biomembranes [Part U: Cellular and Subcellular Transport: Eukaryotic (Nonepithelial) Cells] Edited by SIDNEY FLEISCHER AND BECCA FLEISCHER VOLUME 175. Cumulative Subject Index Volumes 135–139, 141–167 VOLUME 176. Nuclear Magnetic Resonance (Part A: Spectral Techniques and Dynamics) Edited by NORMAN J. OPPENHEIMER AND THOMAS L. JAMES VOLUME 177. Nuclear Magnetic Resonance (Part B: Structure and Mechanism) Edited by NORMAN J. OPPENHEIMER AND THOMAS L. JAMES VOLUME 178. Antibodies, Antigens, and Molecular Mimicry Edited by JOHN J. LANGONE VOLUME 179. Complex Carbohydrates (Part F) Edited by VICTOR GINSBURG VOLUME 180. RNA Processing (Part A: General Methods) Edited by JAMES E. DAHLBERG AND JOHN N. ABELSON VOLUME 181. RNA Processing (Part B: Specific Methods) Edited by JAMES E. DAHLBERG AND JOHN N. ABELSON VOLUME 182. Guide to Protein Purification Edited by MURRAY P. DEUTSCHER VOLUME 183. Molecular Evolution: Computer Analysis of Protein and Nucleic Acid Sequences Edited by RUSSELL F. DOOLITTLE VOLUME 184. Avidin-Biotin Technology Edited by MEIR WILCHEK AND EDWARD A. BAYER VOLUME 185. Gene Expression Technology Edited by DAVID V. GOEDDEL VOLUME 186. Oxygen Radicals in Biological Systems (Part B: Oxygen Radicals and Antioxidants) Edited by LESTER PACKER AND ALEXANDER N. GLAZER VOLUME 187. Arachidonate Related Lipid Mediators Edited by ROBERT C. MURPHY AND FRANK A. FITZPATRICK VOLUME 188. Hydrocarbons and Methylotrophy Edited by MARY E. LIDSTROM VOLUME 189. Retinoids (Part A: Molecular and Metabolic Aspects) Edited by LESTER PACKER

xxxviii


VOLUME 190. Retinoids (Part B: Cell Differentiation and Clinical Applications) Edited by LESTER PACKER VOLUME 191. Biomembranes (Part V: Cellular and Subcellular Transport: Epithelial Cells) Edited by SIDNEY FLEISCHER AND BECCA FLEISCHER VOLUME 192. Biomembranes (Part W: Cellular and Subcellular Transport: Epithelial Cells) Edited by SIDNEY FLEISCHER AND BECCA FLEISCHER VOLUME 193. Mass Spectrometry Edited by JAMES A. MCCLOSKEY VOLUME 194. Guide to Yeast Genetics and Molecular Biology Edited by CHRISTINE GUTHRIE AND GERALD R. FINK VOLUME 195. Adenylyl Cyclase, G Proteins, and Guanylyl Cyclase Edited by ROGER A. JOHNSON AND JACKIE D. CORBIN VOLUME 196. Molecular Motors and the Cytoskeleton Edited by RICHARD B. VALLEE VOLUME 197. Phospholipases Edited by EDWARD A. DENNIS VOLUME 198. Peptide Growth Factors (Part C) Edited by DAVID BARNES, J. P. MATHER, AND GORDON H. SATO VOLUME 199. Cumulative Subject Index Volumes 168–174, 176–194 VOLUME 200. Protein Phosphorylation (Part A: Protein Kinases: Assays, Purification, Antibodies, Functional Analysis, Cloning, and Expression) Edited by TONY HUNTER AND BARTHOLOMEW M. SEFTON VOLUME 201. Protein Phosphorylation (Part B: Analysis of Protein Phosphorylation, Protein Kinase Inhibitors, and Protein Phosphatases) Edited by TONY HUNTER AND BARTHOLOMEW M. SEFTON VOLUME 202. Molecular Design and Modeling: Concepts and Applications (Part A: Proteins, Peptides, and Enzymes) Edited by JOHN J. LANGONE VOLUME 203. Molecular Design and Modeling: Concepts and Applications (Part B: Antibodies and Antigens, Nucleic Acids, Polysaccharides, and Drugs) Edited by JOHN J. LANGONE VOLUME 204. Bacterial Genetic Systems Edited by JEFFREY H. MILLER VOLUME 205. Metallobiochemistry (Part B: Metallothionein and Related Molecules) Edited by JAMES F. RIORDAN AND BERT L. VALLEE


xxxix

VOLUME 206. Cytochrome P450 Edited by MICHAEL R. WATERMAN AND ERIC F. JOHNSON VOLUME 207. Ion Channels Edited by BERNARDO RUDY AND LINDA E. IVERSON VOLUME 208. Protein–DNA Interactions Edited by ROBERT T. SAUER VOLUME 209. Phospholipid Biosynthesis Edited by EDWARD A. DENNIS AND DENNIS E. VANCE VOLUME 210. Numerical Computer Methods Edited by LUDWIG BRAND AND MICHAEL L. JOHNSON VOLUME 211. DNA Structures (Part A: Synthesis and Physical Analysis of DNA) Edited by DAVID M. J. LILLEY AND JAMES E. DAHLBERG VOLUME 212. DNA Structures (Part B: Chemical and Electrophoretic Analysis of DNA) Edited by DAVID M. J. LILLEY AND JAMES E. DAHLBERG VOLUME 213. Carotenoids (Part A: Chemistry, Separation, Quantitation, and Antioxidation) Edited by LESTER PACKER VOLUME 214. Carotenoids (Part B: Metabolism, Genetics, and Biosynthesis) Edited by LESTER PACKER VOLUME 215. Platelets: Receptors, Adhesion, Secretion (Part B) Edited by JACEK J. HAWIGER VOLUME 216. Recombinant DNA (Part G) Edited by RAY WU VOLUME 217. Recombinant DNA (Part H) Edited by RAY WU VOLUME 218. Recombinant DNA (Part I) Edited by RAY WU VOLUME 219. Reconstitution of Intracellular Transport Edited by JAMES E. ROTHMAN VOLUME 220. Membrane Fusion Techniques (Part A) Edited by NEJAT DU¨ZGU¨NES VOLUME 221. Membrane Fusion Techniques (Part B) Edited by NEJAT DU¨ZGU¨NES VOLUME 222. Proteolytic Enzymes in Coagulation, Fibrinolysis, and Complement Activation (Part A: Mammalian Blood Coagulation Factors and Inhibitors) Edited by LASZLO LORAND AND KENNETH G. MANN

xl


VOLUME 223. Proteolytic Enzymes in Coagulation, Fibrinolysis, and Complement Activation (Part B: Complement Activation, Fibrinolysis, and Nonmammalian Blood Coagulation Factors) Edited by LASZLO LORAND AND KENNETH G. MANN VOLUME 224. Molecular Evolution: Producing the Biochemical Data Edited by ELIZABETH ANNE ZIMMER, THOMAS J. WHITE, REBECCA L. CANN, AND ALLAN C. WILSON VOLUME 225. Guide to Techniques in Mouse Development Edited by PAUL M. WASSARMAN AND MELVIN L. DEPAMPHILIS VOLUME 226. Metallobiochemistry (Part C: Spectroscopic and Physical Methods for Probing Metal Ion Environments in Metalloenzymes and Metalloproteins) Edited by JAMES F. RIORDAN AND BERT L. VALLEE VOLUME 227. Metallobiochemistry (Part D: Physical and Spectroscopic Methods for Probing Metal Ion Environments in Metalloproteins) Edited by JAMES F. RIORDAN AND BERT L. VALLEE VOLUME 228. Aqueous Two-Phase Systems Edited by HARRY WALTER AND GO¨TE JOHANSSON VOLUME 229. Cumulative Subject Index Volumes 195–198, 200–227 VOLUME 230. Guide to Techniques in Glycobiology Edited by WILLIAM J. LENNARZ AND GERALD W. HART VOLUME 231. Hemoglobins (Part B: Biochemical and Analytical Methods) Edited by JOHANNES EVERSE, KIM D. VANDEGRIFF, AND ROBERT M. WINSLOW VOLUME 232. Hemoglobins (Part C: Biophysical Methods) Edited by JOHANNES EVERSE, KIM D. VANDEGRIFF, AND ROBERT M. WINSLOW VOLUME 233. Oxygen Radicals in Biological Systems (Part C) Edited by LESTER PACKER VOLUME 234. Oxygen Radicals in Biological Systems (Part D) Edited by LESTER PACKER VOLUME 235. Bacterial Pathogenesis (Part A: Identification and Regulation of Virulence Factors) Edited by VIRGINIA L. CLARK AND PATRIK M. BAVOIL VOLUME 236. Bacterial Pathogenesis (Part B: Integration of Pathogenic Bacteria with Host Cells) Edited by VIRGINIA L. CLARK AND PATRIK M. BAVOIL VOLUME 237. Heterotrimeric G Proteins Edited by RAVI IYENGAR VOLUME 238. Heterotrimeric G-Protein Effectors Edited by RAVI IYENGAR


VOLUME 239. Nuclear Magnetic Resonance (Part C) Edited by THOMAS L. JAMES AND NORMAN J. OPPENHEIMER VOLUME 240. Numerical Computer Methods (Part B) Edited by MICHAEL L. JOHNSON AND LUDWIG BRAND VOLUME 241. Retroviral Proteases Edited by LAWRENCE C. KUO AND JULES A. SHAFER VOLUME 242. Neoglycoconjugates (Part A) Edited by Y. C. LEE AND REIKO T. LEE VOLUME 243. Inorganic Microbial Sulfur Metabolism Edited by HARRY D. PECK, JR., AND JEAN LEGALL VOLUME 244. Proteolytic Enzymes: Serine and Cysteine Peptidases Edited by ALAN J. BARRETT VOLUME 245. Extracellular Matrix Components Edited by E. RUOSLAHTI AND E. ENGVALL VOLUME 246. Biochemical Spectroscopy Edited by KENNETH SAUER VOLUME 247. Neoglycoconjugates (Part B: Biomedical Applications) Edited by Y. C. LEE AND REIKO T. LEE VOLUME 248. Proteolytic Enzymes: Aspartic and Metallo Peptidases Edited by ALAN J. BARRETT VOLUME 249. Enzyme Kinetics and Mechanism (Part D: Developments in Enzyme Dynamics) Edited by DANIEL L. PURICH VOLUME 250. Lipid Modifications of Proteins Edited by PATRICK J. CASEY AND JANICE E. BUSS VOLUME 251. Biothiols (Part A: Monothiols and Dithiols, Protein Thiols, and Thiyl Radicals) Edited by LESTER PACKER VOLUME 252. Biothiols (Part B: Glutathione and Thioredoxin; Thiols in Signal Transduction and Gene Regulation) Edited by LESTER PACKER VOLUME 253. Adhesion of Microbial Pathogens Edited by RON J. DOYLE AND ITZHAK OFEK VOLUME 254. Oncogene Techniques Edited by PETER K. VOGT AND INDER M. VERMA VOLUME 255. Small GTPases and Their Regulators (Part A: Ras Family) Edited by W. E. BALCH, CHANNING J. DER, AND ALAN HALL

xli

xlii


VOLUME 256. Small GTPases and Their Regulators (Part B: Rho Family) Edited by W. E. BALCH, CHANNING J. DER, AND ALAN HALL VOLUME 257. Small GTPases and Their Regulators (Part C: Proteins Involved in Transport) Edited by W. E. BALCH, CHANNING J. DER, AND ALAN HALL VOLUME 258. Redox-Active Amino Acids in Biology Edited by JUDITH P. KLINMAN VOLUME 259. Energetics of Biological Macromolecules Edited by MICHAEL L. JOHNSON AND GARY K. ACKERS VOLUME 260. Mitochondrial Biogenesis and Genetics (Part A) Edited by GIUSEPPE M. ATTARDI AND ANNE CHOMYN VOLUME 261. Nuclear Magnetic Resonance and Nucleic Acids Edited by THOMAS L. JAMES VOLUME 262. DNA Replication Edited by JUDITH L. CAMPBELL VOLUME 263. Plasma Lipoproteins (Part C: Quantitation) Edited by WILLIAM A. BRADLEY, SANDRA H. GIANTURCO, AND JERE P. SEGREST VOLUME 264. Mitochondrial Biogenesis and Genetics (Part B) Edited by GIUSEPPE M. ATTARDI AND ANNE CHOMYN VOLUME 265. Cumulative Subject Index Volumes 228, 230–262 VOLUME 266. Computer Methods for Macromolecular Sequence Analysis Edited by RUSSELL F. DOOLITTLE VOLUME 267. Combinatorial Chemistry Edited by JOHN N. ABELSON VOLUME 268. Nitric Oxide (Part A: Sources and Detection of NO; NO Synthase) Edited by LESTER PACKER VOLUME 269. Nitric Oxide (Part B: Physiological and Pathological Processes) Edited by LESTER PACKER VOLUME 270. High Resolution Separation and Analysis of Biological Macromolecules (Part A: Fundamentals) Edited by BARRY L. KARGER AND WILLIAM S. HANCOCK VOLUME 271. High Resolution Separation and Analysis of Biological Macromolecules (Part B: Applications) Edited by BARRY L. KARGER AND WILLIAM S. HANCOCK VOLUME 272. Cytochrome P450 (Part B) Edited by ERIC F. JOHNSON AND MICHAEL R. WATERMAN VOLUME 273. RNA Polymerase and Associated Factors (Part A) Edited by SANKAR ADHYA


xliii

VOLUME 274. RNA Polymerase and Associated Factors (Part B) Edited by SANKAR ADHYA VOLUME 275. Viral Polymerases and Related Proteins Edited by LAWRENCE C. KUO, DAVID B. OLSEN, AND STEVEN S. CARROLL VOLUME 276. Macromolecular Crystallography (Part A) Edited by CHARLES W. CARTER, JR., AND ROBERT M. SWEET VOLUME 277. Macromolecular Crystallography (Part B) Edited by CHARLES W. CARTER, JR., AND ROBERT M. SWEET VOLUME 278. Fluorescence Spectroscopy Edited by LUDWIG BRAND AND MICHAEL L. JOHNSON VOLUME 279. Vitamins and Coenzymes (Part I) Edited by DONALD B. MCCORMICK, JOHN W. SUTTIE, AND CONRAD WAGNER VOLUME 280. Vitamins and Coenzymes (Part J) Edited by DONALD B. MCCORMICK, JOHN W. SUTTIE, AND CONRAD WAGNER VOLUME 281. Vitamins and Coenzymes (Part K) Edited by DONALD B. MCCORMICK, JOHN W. SUTTIE, AND CONRAD WAGNER VOLUME 282. Vitamins and Coenzymes (Part L) Edited by DONALD B. MCCORMICK, JOHN W. SUTTIE, AND CONRAD WAGNER VOLUME 283. Cell Cycle Control Edited by WILLIAM G. DUNPHY VOLUME 284. Lipases (Part A: Biotechnology) Edited by BYRON RUBIN AND EDWARD A. DENNIS VOLUME 285. Cumulative Subject Index Volumes 263, 264, 266–284, 286–289 VOLUME 286. Lipases (Part B: Enzyme Characterization and Utilization) Edited by BYRON RUBIN AND EDWARD A. DENNIS VOLUME 287. Chemokines Edited by RICHARD HORUK VOLUME 288. Chemokine Receptors Edited by RICHARD HORUK VOLUME 289. Solid Phase Peptide Synthesis Edited by GREGG B. FIELDS VOLUME 290. Molecular Chaperones Edited by GEORGE H. LORIMER AND THOMAS BALDWIN VOLUME 291. Caged Compounds Edited by GERARD MARRIOTT VOLUME 292. ABC Transporters: Biochemical, Cellular, and Molecular Aspects Edited by SURESH V. AMBUDKAR AND MICHAEL M. GOTTESMAN

xliv


VOLUME 293. Ion Channels (Part B) Edited by P. MICHAEL CONN VOLUME 294. Ion Channels (Part C) Edited by P. MICHAEL CONN VOLUME 295. Energetics of Biological Macromolecules (Part B) Edited by GARY K. ACKERS AND MICHAEL L. JOHNSON VOLUME 296. Neurotransmitter Transporters Edited by SUSAN G. AMARA VOLUME 297. Photosynthesis: Molecular Biology of Energy Capture Edited by LEE MCINTOSH VOLUME 298. Molecular Motors and the Cytoskeleton (Part B) Edited by RICHARD B. VALLEE VOLUME 299. Oxidants and Antioxidants (Part A) Edited by LESTER PACKER VOLUME 300. Oxidants and Antioxidants (Part B) Edited by LESTER PACKER VOLUME 301. Nitric Oxide: Biological and Antioxidant Activities (Part C) Edited by LESTER PACKER VOLUME 302. Green Fluorescent Protein Edited by P. MICHAEL CONN VOLUME 303. cDNA Preparation and Display Edited by SHERMAN M. WEISSMAN VOLUME 304. Chromatin Edited by PAUL M. WASSARMAN AND ALAN P. WOLFFE VOLUME 305. Bioluminescence and Chemiluminescence (Part C) Edited by THOMAS O. BALDWIN AND MIRIAM M. ZIEGLER VOLUME 306. Expression of Recombinant Genes in Eukaryotic Systems Edited by JOSEPH C. GLORIOSO AND MARTIN C. SCHMIDT VOLUME 307. Confocal Microscopy Edited by P. MICHAEL CONN VOLUME 308. Enzyme Kinetics and Mechanism (Part E: Energetics of Enzyme Catalysis) Edited by DANIEL L. PURICH AND VERN L. SCHRAMM VOLUME 309. Amyloid, Prions, and Other Protein Aggregates Edited by RONALD WETZEL VOLUME 310. Biofilms Edited by RON J. DOYLE


VOLUME 311. Sphingolipid Metabolism and Cell Signaling (Part A) Edited by ALFRED H. MERRILL, JR., AND YUSUF A. HANNUN VOLUME 312. Sphingolipid Metabolism and Cell Signaling (Part B) Edited by ALFRED H. MERRILL, JR., AND YUSUF A. HANNUN VOLUME 313. Antisense Technology (Part A: General Methods, Methods of Delivery, and RNA Studies) Edited by M. IAN PHILLIPS VOLUME 314. Antisense Technology (Part B: Applications) Edited by M. IAN PHILLIPS VOLUME 315. Vertebrate Phototransduction and the Visual Cycle (Part A) Edited by KRZYSZTOF PALCZEWSKI VOLUME 316. Vertebrate Phototransduction and the Visual Cycle (Part B) Edited by KRZYSZTOF PALCZEWSKI VOLUME 317. RNA–Ligand Interactions (Part A: Structural Biology Methods) Edited by DANIEL W. CELANDER AND JOHN N. ABELSON VOLUME 318. RNA–Ligand Interactions (Part B: Molecular Biology Methods) Edited by DANIEL W. CELANDER AND JOHN N. ABELSON VOLUME 319. Singlet Oxygen, UV-A, and Ozone Edited by LESTER PACKER AND HELMUT SIES VOLUME 320. Cumulative Subject Index Volumes 290–319 VOLUME 321. Numerical Computer Methods (Part C) Edited by MICHAEL L. JOHNSON AND LUDWIG BRAND VOLUME 322. Apoptosis Edited by JOHN C. REED VOLUME 323. Energetics of Biological Macromolecules (Part C) Edited by MICHAEL L. JOHNSON AND GARY K. ACKERS VOLUME 324. Branched-Chain Amino Acids (Part B) Edited by ROBERT A. HARRIS AND JOHN R. SOKATCH VOLUME 325. Regulators and Effectors of Small GTPases (Part D: Rho Family) Edited by W. E. BALCH, CHANNING J. DER, AND ALAN HALL VOLUME 326. Applications of Chimeric Genes and Hybrid Proteins (Part A: Gene Expression and Protein Purification) Edited by JEREMY THORNER, SCOTT D. EMR, AND JOHN N. ABELSON VOLUME 327. Applications of Chimeric Genes and Hybrid Proteins (Part B: Cell Biology and Physiology) Edited by JEREMY THORNER, SCOTT D. EMR, AND JOHN N. ABELSON

xlv

xlvi


VOLUME 328. Applications of Chimeric Genes and Hybrid Proteins (Part C: Protein–Protein Interactions and Genomics) Edited by JEREMY THORNER, SCOTT D. EMR, AND JOHN N. ABELSON VOLUME 329. Regulators and Effectors of Small GTPases (Part E: GTPases Involved in Vesicular Traffic) Edited by W. E. BALCH, CHANNING J. DER, AND ALAN HALL VOLUME 330. Hyperthermophilic Enzymes (Part A) Edited by MICHAEL W. W. ADAMS AND ROBERT M. KELLY VOLUME 331. Hyperthermophilic Enzymes (Part B) Edited by MICHAEL W. W. ADAMS AND ROBERT M. KELLY VOLUME 332. Regulators and Effectors of Small GTPases (Part F: Ras Family I) Edited by W. E. BALCH, CHANNING J. DER, AND ALAN HALL VOLUME 333. Regulators and Effectors of Small GTPases (Part G: Ras Family II) Edited by W. E. BALCH, CHANNING J. DER, AND ALAN HALL VOLUME 334. Hyperthermophilic Enzymes (Part C) Edited by MICHAEL W. W. ADAMS AND ROBERT M. KELLY VOLUME 335. Flavonoids and Other Polyphenols Edited by LESTER PACKER VOLUME 336. Microbial Growth in Biofilms (Part A: Developmental and Molecular Biological Aspects) Edited by RON J. DOYLE VOLUME 337. Microbial Growth in Biofilms (Part B: Special Environments and Physicochemical Aspects) Edited by RON J. DOYLE VOLUME 338. Nuclear Magnetic Resonance of Biological Macromolecules (Part A) Edited by THOMAS L. JAMES, VOLKER DO¨TSCH, AND ULI SCHMITZ VOLUME 339. Nuclear Magnetic Resonance of Biological Macromolecules (Part B) Edited by THOMAS L. JAMES, VOLKER DO¨TSCH, AND ULI SCHMITZ VOLUME 340. Drug–Nucleic Acid Interactions Edited by JONATHAN B. CHAIRES AND MICHAEL J. WARING VOLUME 341. Ribonucleases (Part A) Edited by ALLEN W. NICHOLSON VOLUME 342. Ribonucleases (Part B) Edited by ALLEN W. NICHOLSON VOLUME 343. G Protein Pathways (Part A: Receptors) Edited by RAVI IYENGAR AND JOHN D. HILDEBRANDT VOLUME 344. G Protein Pathways (Part B: G Proteins and Their Regulators) Edited by RAVI IYENGAR AND JOHN D. HILDEBRANDT


xlvii

VOLUME 345. G Protein Pathways (Part C: Effector Mechanisms) Edited by RAVI IYENGAR AND JOHN D. HILDEBRANDT VOLUME 346. Gene Therapy Methods Edited by M. IAN PHILLIPS VOLUME 347. Protein Sensors and Reactive Oxygen Species (Part A: Selenoproteins and Thioredoxin) Edited by HELMUT SIES AND LESTER PACKER VOLUME 348. Protein Sensors and Reactive Oxygen Species (Part B: Thiol Enzymes and Proteins) Edited by HELMUT SIES AND LESTER PACKER VOLUME 349. Superoxide Dismutase Edited by LESTER PACKER VOLUME 350. Guide to Yeast Genetics and Molecular and Cell Biology (Part B) Edited by CHRISTINE GUTHRIE AND GERALD R. FINK VOLUME 351. Guide to Yeast Genetics and Molecular and Cell Biology (Part C) Edited by CHRISTINE GUTHRIE AND GERALD R. FINK VOLUME 352. Redox Cell Biology and Genetics (Part A) Edited by CHANDAN K. SEN AND LESTER PACKER VOLUME 353. Redox Cell Biology and Genetics (Part B) Edited by CHANDAN K. SEN AND LESTER PACKER VOLUME 354. Enzyme Kinetics and Mechanisms (Part F: Detection and Characterization of Enzyme Reaction Intermediates) Edited by DANIEL L. PURICH VOLUME 355. Cumulative Subject Index Volumes 321–354 VOLUME 356. Laser Capture Microscopy and Microdissection Edited by P. MICHAEL CONN VOLUME 357. Cytochrome P450, Part C Edited by ERIC F. JOHNSON AND MICHAEL R. WATERMAN VOLUME 358. Bacterial Pathogenesis (Part C: Identification, Regulation, and Function of Virulence Factors) Edited by VIRGINIA L. CLARK AND PATRIK M. BAVOIL VOLUME 359. Nitric Oxide (Part D) Edited by ENRIQUE CADENAS AND LESTER PACKER VOLUME 360. Biophotonics (Part A) Edited by GERARD MARRIOTT AND IAN PARKER VOLUME 361. Biophotonics (Part B) Edited by GERARD MARRIOTT AND IAN PARKER

xlviii


VOLUME 362. Recognition of Carbohydrates in Biological Systems (Part A) Edited by YUAN C. LEE AND REIKO T. LEE VOLUME 363. Recognition of Carbohydrates in Biological Systems (Part B) Edited by YUAN C. LEE AND REIKO T. LEE VOLUME 364. Nuclear Receptors Edited by DAVID W. RUSSELL AND DAVID J. MANGELSDORF VOLUME 365. Differentiation of Embryonic Stem Cells Edited by PAUL M. WASSAUMAN AND GORDON M. KELLER VOLUME 366. Protein Phosphatases Edited by SUSANNE KLUMPP AND JOSEF KRIEGLSTEIN VOLUME 367. Liposomes (Part A) Edited by NEJAT DU¨ZGU¨NES VOLUME 368. Macromolecular Crystallography (Part C) Edited by CHARLES W. CARTER, JR., AND ROBERT M. SWEET VOLUME 369. Combinational Chemistry (Part B) Edited by GUILLERMO A. MORALES AND BARRY A. BUNIN VOLUME 370. RNA Polymerases and Associated Factors (Part C) Edited by SANKAR L. ADHYA AND SUSAN GARGES VOLUME 371. RNA Polymerases and Associated Factors (Part D) Edited by SANKAR L. ADHYA AND SUSAN GARGES VOLUME 372. Liposomes (Part B) Edited by NEJAT DU¨ZGU¨NES VOLUME 373. Liposomes (Part C) Edited by NEJAT DU¨ZGU¨NES VOLUME 374. Macromolecular Crystallography (Part D) Edited by CHARLES W. CARTER, JR., AND ROBERT W. SWEET VOLUME 375. Chromatin and Chromatin Remodeling Enzymes (Part A) Edited by C. DAVID ALLIS AND CARL WU VOLUME 376. Chromatin and Chromatin Remodeling Enzymes (Part B) Edited by C. DAVID ALLIS AND CARL WU VOLUME 377. Chromatin and Chromatin Remodeling Enzymes (Part C) Edited by C. DAVID ALLIS AND CARL WU VOLUME 378. Quinones and Quinone Enzymes (Part A) Edited by HELMUT SIES AND LESTER PACKER VOLUME 379. Energetics of Biological Macromolecules (Part D) Edited by JO M. HOLT, MICHAEL L. JOHNSON, AND GARY K. ACKERS VOLUME 380. Energetics of Biological Macromolecules (Part E) Edited by JO M. HOLT, MICHAEL L. JOHNSON, AND GARY K. ACKERS


xlix

VOLUME 381. Oxygen Sensing Edited by CHANDAN K. SEN AND GREGG L. SEMENZA VOLUME 382. Quinones and Quinone Enzymes (Part B) Edited by HELMUT SIES AND LESTER PACKER VOLUME 383. Numerical Computer Methods (Part D) Edited by LUDWIG BRAND AND MICHAEL L. JOHNSON VOLUME 384. Numerical Computer Methods (Part E) Edited by LUDWIG BRAND AND MICHAEL L. JOHNSON VOLUME 385. Imaging in Biological Research (Part A) Edited by P. MICHAEL CONN VOLUME 386. Imaging in Biological Research (Part B) Edited by P. MICHAEL CONN VOLUME 387. Liposomes (Part D) Edited by NEJAT DU¨ZGU¨NES VOLUME 388. Protein Engineering Edited by DAN E. ROBERTSON AND JOSEPH P. NOEL VOLUME 389. Regulators of G-Protein Signaling (Part A) Edited by DAVID P. SIDEROVSKI VOLUME 390. Regulators of G-Protein Signaling (Part B) Edited by DAVID P. SIDEROVSKI VOLUME 391. Liposomes (Part E) Edited by NEJAT DU¨ZGU¨NES VOLUME 392. RNA Interference Edited by ENGELKE ROSSI VOLUME 393. Circadian Rhythms Edited by MICHAEL W. YOUNG VOLUME 394. Nuclear Magnetic Resonance of Biological Macromolecules (Part C) Edited by THOMAS L. JAMES VOLUME 395. Producing the Biochemical Data (Part B) Edited by ELIZABETH A. ZIMMER AND ERIC H. ROALSON VOLUME 396. Nitric Oxide (Part E) Edited by LESTER PACKER AND ENRIQUE CADENAS VOLUME 397. Environmental Microbiology Edited by JARED R. LEADBETTER VOLUME 398. Ubiquitin and Protein Degradation (Part A) Edited by RAYMOND J. DESHAIES VOLUME 399. Ubiquitin and Protein Degradation (Part B) Edited by RAYMOND J. DESHAIES

l


VOLUME 400. Phase II Conjugation Enzymes and Transport Systems Edited by HELMUT SIES AND LESTER PACKER VOLUME 401. Glutathione Transferases and Gamma Glutamyl Transpeptidases Edited by HELMUT SIES AND LESTER PACKER VOLUME 402. Biological Mass Spectrometry Edited by A. L. BURLINGAME VOLUME 403. GTPases Regulating Membrane Targeting and Fusion Edited by WILLIAM E. BALCH, CHANNING J. DER, AND ALAN HALL VOLUME 404. GTPases Regulating Membrane Dynamics Edited by WILLIAM E. BALCH, CHANNING J. DER, AND ALAN HALL VOLUME 405. Mass Spectrometry: Modified Proteins and Glycoconjugates Edited by A. L. BURLINGAME VOLUME 406. Regulators and Effectors of Small GTPases: Rho Family Edited by WILLIAM E. BALCH, CHANNING J. DER, AND ALAN HALL VOLUME 407. Regulators and Effectors of Small GTPases: Ras Family Edited by WILLIAM E. BALCH, CHANNING J. DER, AND ALAN HALL VOLUME 408. DNA Repair (Part A) Edited by JUDITH L. CAMPBELL AND PAUL MODRICH VOLUME 409. DNA Repair (Part B) Edited by JUDITH L. CAMPBELL AND PAUL MODRICH VOLUME 410. DNA Microarrays (Part A: Array Platforms and Web-Bench Protocols) Edited by ALAN KIMMEL AND BRIAN OLIVER VOLUME 411. DNA Microarrays (Part B: Databases and Statistics) Edited by ALAN KIMMEL AND BRIAN OLIVER VOLUME 412. Amyloid, Prions, and Other Protein Aggregates (Part B) Edited by INDU KHETERPAL AND RONALD WETZEL VOLUME 413. Amyloid, Prions, and Other Protein Aggregates (Part C) Edited by INDU KHETERPAL AND RONALD WETZEL VOLUME 414. Measuring Biological Responses with Automated Microscopy Edited by JAMES INGLESE VOLUME 415. Glycobiology Edited by MINORU FUKUDA VOLUME 416. Glycomics Edited by MINORU FUKUDA VOLUME 417. Functional Glycomics Edited by MINORU FUKUDA


li

VOLUME 418. Embryonic Stem Cells Edited by IRINA KLIMANSKAYA AND ROBERT LANZA VOLUME 419. Adult Stem Cells Edited by IRINA KLIMANSKAYA AND ROBERT LANZA VOLUME 420. Stem Cell Tools and Other Experimental Protocols Edited by IRINA KLIMANSKAYA AND ROBERT LANZA VOLUME 421. Advanced Bacterial Genetics: Use of Transposons and Phage for Genomic Engineering Edited by KELLY T. HUGHES VOLUME 422. Two-Component Signaling Systems, Part A Edited by MELVIN I. SIMON, BRIAN R. CRANE, AND ALEXANDRINE CRANE VOLUME 423. Two-Component Signaling Systems, Part B Edited by MELVIN I. SIMON, BRIAN R. CRANE, AND ALEXANDRINE CRANE VOLUME 424. RNA Editing Edited by JONATHA M. GOTT VOLUME 425. RNA Modification Edited by JONATHA M. GOTT VOLUME 426. Integrins Edited by DAVID CHERESH VOLUME 427. MicroRNA Methods Edited by JOHN J. ROSSI VOLUME 428. Osmosensing and Osmosignaling Edited by HELMUT SIES AND DIETER HAUSSINGER VOLUME 429. Translation Initiation: Extract Systems and Molecular Genetics Edited by JON LORSCH VOLUME 430. Translation Initiation: Reconstituted Systems and Biophysical Methods Edited by JON LORSCH VOLUME 431. Translation Initiation: Cell Biology, High-Throughput and Chemical-Based Approaches Edited by JON LORSCH VOLUME 432. Lipidomics and Bioactive Lipids: Mass-Spectrometry–Based Lipid Analysis Edited by H. ALEX BROWN VOLUME 433. Lipidomics and Bioactive Lipids: Specialized Analytical Methods and Lipids in Disease Edited by H. ALEX BROWN

lii


VOLUME 434. Lipidomics and Bioactive Lipids: Lipids and Cell Signaling Edited by H. ALEX BROWN VOLUME 435. Oxygen Biology and Hypoxia Edited by HELMUT SIES AND BERNHARD BRU¨NE VOLUME 436. Globins and Other Nitric Oxide-Reactive Protiens (Part A) Edited by ROBERT K. POOLE VOLUME 437. Globins and Other Nitric Oxide-Reactive Protiens (Part B) Edited by ROBERT K. POOLE VOLUME 438. Small GTPases in Disease (Part A) Edited by WILLIAM E. BALCH, CHANNING J. DER, AND ALAN HALL VOLUME 439. Small GTPases in Disease (Part B) Edited by WILLIAM E. BALCH, CHANNING J. DER, AND ALAN HALL VOLUME 440. Nitric Oxide, Part F Oxidative and Nitrosative Stress in Redox Regulation of Cell Signaling Edited by ENRIQUE CADENAS AND LESTER PACKER VOLUME 441. Nitric Oxide, Part G Oxidative and Nitrosative Stress in Redox Regulation of Cell Signaling Edited by ENRIQUE CADENAS AND LESTER PACKER VOLUME 442. Programmed Cell Death, General Principles for Studying Cell Death (Part A) Edited by ROYA KHOSRAVI-FAR, ZAHRA ZAKERI, RICHARD A. LOCKSHIN, AND MAURO PIACENTINI VOLUME 443. Angiogenesis: In Vitro Systems Edited by DAVID A. CHERESH VOLUME 444. Angiogenesis: In Vivo Systems (Part A) Edited by DAVID A. CHERESH VOLUME 445. Angiogenesis: In Vivo Systems (Part B) Edited by DAVID A. CHERESH VOLUME 446. Programmed Cell Death, The Biology and Therapeutic Implications of Cell Death (Part B) Edited by ROYA KHOSRAVI-FAR, ZAHRA ZAKERI, RICHARD A. LOCKSHIN, AND MAURO PIACENTINI VOLUME 447. RNA Turnover in Bacteria, Archaea and Organelles Edited by LYNNE E. MAQUAT AND CECILIA M. ARRAIANO VOLUME 448. RNA Turnover in Eukaryotes: Nucleases, Pathways and Analysis of mRNA Decay Edited by LYNNE E. MAQUAT AND MEGERDITCH KILEDJIAN


liii

VOLUME 449. RNA Turnover in Eukaryotes: Analysis of Specialized and Quality Control RNA Decay Pathways Edited by LYNNE E. MAQUAT AND MEGERDITCH KILEDJIAN VOLUME 450. Fluorescence Spectroscopy Edited by LUDWIG BRAND AND MICHAEL L. JOHNSON VOLUME 451. Autophagy: Lower Eukaryotes and Non-Mammalian Systems (Part A) Edited by DANIEL J. KLIONSKY VOLUME 452. Autophagy in Mammalian Systems (Part B) Edited by DANIEL J. KLIONSKY VOLUME 453. Autophagy in Disease and Clinical Applications (Part C) Edited by DANIEL J. KLIONSKY VOLUME 454. Computer Methods (Part A) Edited by MICHAEL L. JOHNSON AND LUDWIG BRAND VOLUME 455. Biothermodynamics (Part A) Edited by MICHAEL L. JOHNSON, JO M. HOLT, AND GARY K. ACKERS (RETIRED) VOLUME 456. Mitochondrial Function, Part A: Mitochondrial Electron Transport Complexes and Reactive Oxygen Species Edited by WILLIAM S. ALLISON AND IMMO E. SCHEFFLER VOLUME 457. Mitochondrial Function, Part B: Mitochondrial Protein Kinases, Protein Phosphatases and Mitochondrial Diseases Edited by WILLIAM S. ALLISON AND ANNE N. MURPHY VOLUME 458. Complex Enzymes in Microbial Natural Product Biosynthesis, Part A: Overview Articles and Peptides Edited by DAVID A. HOPWOOD VOLUME 459. Complex Enzymes in Microbial Natural Product Biosynthesis, Part B: Polyketides, Aminocoumarins and Carbohydrates Edited by DAVID A. HOPWOOD VOLUME 460. Chemokines, Part A Edited by TRACY M. HANDEL AND DAMON J. HAMEL VOLUME 461. Chemokines, Part B Edited by TRACY M. HANDEL AND DAMON J. HAMEL VOLUME 462. Non-Natural Amino Acids Edited by TOM W. MUIR AND JOHN N. ABELSON VOLUME 463. Guide to Protein Purification, 2nd Edition Edited by RICHARD R. BURGESS AND MURRAY P. DEUTSCHER VOLUME 464. Liposomes, Part F Edited by NEJAT DU¨ZGU¨NES

liv


VOLUME 465. Liposomes, Part G Edited by NEJAT DU¨ZGU¨NES VOLUME 466. Biothermodynamics, Part B Edited by MICHAEL L. JOHNSON, GARY K. ACKERS, AND JO M. HOLT VOLUME 467. Computer Methods Part B Edited by MICHAEL L. JOHNSON AND LUDWIG BRAND VOLUME 468. Biophysical, Chemical, and Functional Probes of RNA Structure, Interactions and Folding: Part A Edited by DANIEL HERSCHLAG VOLUME 469. Biophysical, Chemical, and Functional Probes of RNA Structure, Interactions and Folding: Part B Edited by DANIEL HERSCHLAG VOLUME 470. Guide to Yeast Genetics: Functional Genomics, Proteomics, and Other Systems Analysis, 2nd Edition Edited by GERALD FINK, JONATHAN WEISSMAN, AND CHRISTINE GUTHRIE VOLUME 471. Two-Component Signaling Systems, Part C Edited by MELVIN I. SIMON, BRIAN R. CRANE, AND ALEXANDRINE CRANE VOLUME 472. Single Molecule Tools, Part A: Fluorescence Based Approaches Edited by NILS WALTER VOLUME 473. Thiol Redox Transitions in Cell Signaling, Part A Chemistry and Biochemistry of Low Molecular Weight and Protein Thiols Edited by ENRIQUE CADENAS AND LESTER PACKER VOLUME 474. Thiol Redox Transitions in Cell Signaling, Part B Cellular Localization and Signaling Edited by ENRIQUE CADENAS AND LESTER PACKER VOLUME 475. Single Molecule Tools, Part B: Super-Resolution, Particle Tracking, Multiparameter, and Force Based Methods Edited by NILS G. WALTER VOLUME 476. Guide to Techniques in Mouse Development, Part A: Mice, Embryos, and Cells, 2nd Edition Edited by PAUL M. WASSARMAN AND PHILIPPE M. SORIANO VOLUME 477. Guide to Techniques in Mouse Development, Part B: Mouse Molecular Genetics, 2nd Edition Edited by PAUL M. WASSARMAN AND PHILIPPE M. SORIANO VOLUME 478. Glycomics Edited by MINORU FUKUDA

C H A P T E R

O N E

Mass Spectrometric Analysis of Sulfated N- and O-Glycans Kay-Hooi Khoo and Shin-Yi Yu Contents 4 7 7 9 10 13 14

1. Introduction and Overview 2. Sample Preparation 2.1. From biological sources to glycoprotein extracts 2.2. From glycoproteins to released N- and O-glycans 2.3. Permethylation and microscale fractionation 3. MS Analyses and Data Interpretation 3.1. MALDI-based MS and MS/MS analysis 3.2. Interpretation of MALDI-MS profile of permethylated sulfated glycans 3.3. CID MS/MS of permethylated sulfated glycans 4. Future Perspectives Acknowledgments References

17 20 23 24 24

Abstract Sulfated N- and O-glycans carried on a myriad of cell-surface adhesion molecules and receptors are often not detected by current approaches in mass spectrometry (MS)-based glycomic mapping of cells and tissues. This is in part due to a natural lower abundance, compounded further by their negatively charged nature, which adversely disfavors their ionization and detection amid a sea of often much more abundant, nonsulfated, sialylated glycans. However, this particular limitation can actually be taken advantage of to effect highly selective enrichment and sensitive MS screening in negative ion mode, provided the ubiquitous sialic acids can first be neutralized. It has been demonstrated that permethylation not only fulfills this role adequately but further confers better MS/MS fragmentation characteristics for more efficient structural mapping and sequencing. Protocols and general practical considerations are described here which would enable one to readily prepare permethylated

Institute of Biological Chemistry, Academia Sinica, Nankang, Taipei, Taiwan Methods in Enzymology, Volume 478 ISSN 0076-6879, DOI: 10.1016/S0076-6879(10)78001-0

#

2010 Elsevier Inc. All rights reserved.

3

4

Kay-Hooi Khoo and Shin-Yi Yu

sulfated glycans, fractionate them away from the more abundant nonsulfated ones in simple steps for high-sensitivity MS analysis, and sensibly interpret the initial sulfoglycomic screening data thus obtained.

1. Introduction and Overview Sulfation, occurring on specific locations of terminal oligosaccharyl epitopes (Fig. 1.1), modifies the physicochemical properties of a glycotope and thereby alters its cognate recognition by specific endogenous lectins that perceive and translate the encoded immunobiological functions. It is clear that such specific recognition codes imparted by the unique sulfation pattern on an assorted surface markers of leukocytes and endothelial venules (Kawashima, 2006; Rosen, 2004) cannot be critically delineated without them being first structurally defined within the full context of their underlying carriers. Mass spectrometry (MS) and various modes of separation techniques, coupled with chemical modifications and glycosidase digestions, are the corner stones of high-sensitivity approach in structural determination of complex glycans derived from biological sources (Geyer and Geyer, 2006; North et al., 2009). More recently, these glycosylation analyses have evolved into what is referred to as glycomics, which involves mapping the lacdiNAc S 4 4 GalNAc-4-O- S HNK-1 4 3

3 S

3⬘ sulfo sialyl LeX 6⬘ sulfo sialyl LeX 6 sulfo sialyl LeX 6 sulfo sialyl S 6′ S 6 LacNAc 4 3 4 4 3 3 S 6 3 S ± 3 3 6 ± ± ± ± ± 4 Gal-3-O-

Gal-6-O-

S

S

GlcNAc-6-O-

S

S

4 SdA

3

4 3 SLeX

± S

GlcA-3-O- S 4 3

Gal

GlcA

GlcNAc GalNAc

Neu5Ac Fuc

3 ±

4 3

3 ±

±

6 4 3

6

6

±

3 3 MECA-79 epitope (sulfated LacNAc on extended core 1)

Figure 1.1 Various commonly occurring sulfated glycotopes. Additional sulfation on the GlcNAc or Gal would preclude formation of sialyl Lewis X (SLeX), yielding instead sulfo sialyl LeX. Since sulfation on GlcNAc precedes its subsequent b4-galactosylation, a2-3 sialylation, and a3 fucosylation in that order, various forms of incompletely sialylated and/or fucosylated sulfo LacNAc are also commonly found, along with nonextended sulfated GlcNAc termini. In B cells, occurrence of a2-6 sialylated 6-sulfo LacNAc has been reported (Kimura et al., 2007). The critical peripheral lymph node addressins that mediate lymphocyte homing are thought to correspond to sialomucins with Core 2 O-glycans carrying 6-sulfo sialyl LeX on both arms. However, various incompletely elaborated sulfated glycotopes are also present. The minimal epitope recognized by mAb MECA-79, which stains HEV, corresponds to a sulfated LacNAc extending out from Core 1. Addition of terminal GalNAc on sialyl LacNAc leads to formation of the SdA antigen, which is currently not known if it may also be 6-O-sulfated on the internal LacNAc unit.

MS Analysis of Sulfated Glycans

5

complexity and characteristics of the glycome of a cell, tissue, or organism, at a particular pathophysiological or genetically altered state (Haslam et al., 2006; Zaia, 2008). While a single dimensional MS mapping proves to be relatively straightforward and informative in cases of providing a ‘‘first rough impression’’ and identifying drastic glycomic changes, it is clear from the onset that more subtle glycomic changes are often apparent only by applications of multistage fractionation and MS analyses at increasing levels of sophistication. Added to the need in resolving complex isomeric mixtures is a practical demand for an ever increasing sensitivity to detect glycans of low abundance, and/or those multiply substituted with negatively charged constituents. The sulfated glycome, in particular, is often refractory to MS-based glycomic mapping, and yet arguably constitutes one of the most glycobiologically relevant structural entities. At present, analysis of the more abundant sulfated O-glycans derived from epithelial mucins is less of a problem. Underivatized native O-glycans are commonly separated into neutral and negatively charged fractions by microscale fractionation on anion exchange media, followed by desalting on Carbograph or porous graphitized carbon (PGC) media, before subjecting to micro- or nanoLC–MS/MS on a PGC-based capillary column, in negative ion mode (Karlsson and Thomsson, 2009; Robbe-Masselot et al., 2009). A tremendous amount of structural information can be obtained from the reconstructed MS profiles and MS/MS data on those peaks that were selected for analysis. Even so, a closer scrutiny of the reported data often leads to many uncertainties with respect to the extent of isomeric variations, linkages, and location of sulfates. Such is the intrinsic nature of an automated LC–MS/MS runs, with single-stage fragmentation (MS2) on native glycans. Another good recent work is illustrated by one working on the sulfated glycans derived from glycolipids of colonic cancer cells (Shida et al., 2009). The reducing end of the native glycans was tagged with a fluorophore, for example pyridylamine, to allow two-dimensional HPLC separation. Structural identification of each of the detected peaks was mostly inferred from molecular weight information by MS analysis, the HPLC coordinates as compared to standard references, and their shifts after specific exoglycosidase digestions. On the other hand, for membrane-bound sialomucins such as those isolated from lymphoid cells or tissues, practical cases of applications using the aforementioned approaches have yet to be reported, presumably due to limited sample amount and the complexity of tissue extraction. A more recent work on isolated CD34 from tonsils (Hernandez Mir et al., 2009) illustrates the current limitations. The expected 6-sulfo sialyl LeX epitope was only detectable in a minor population of CD34, after three sequential steps of affinity capture, using first a general lectin against sialylated mucins, then mAb against CD34, and finally L-selectin to enrich for the binding population. In this and many other examples, the final yield of glycans is

6


often only sufficient for a rough MALDI-MS mapping in negative ion mode without MS/MS, while limited structural information was obtained by observing mass shifts after selective exoglycosidase digestion. In comparison with MS analysis of native glycans with or without reducing end tagging, MS analysis of permethylated glycans may not rank as the most sensitive MS technique to just obtain molecular weight information, but clearly endowed with several distinct advantages, particularly in deriving detailed structural information on novel epitopes through a series of reliable linkage-specific fragment ions. Without the O-Me tag, multiple cleavages are often indistinguishable from single cleavage and thus not conducive to definitive assignment of the branching pattern. This is highly relevant as the specifically located sialic acids and fucoses (Fig. 1.1), which collectively constitute the various important glycotopes, are also those that are most readily lost during collision-induced dissociation (CID) MS/MS in preference to cleavages at the backbone LacNAc unit, rendering assignment often ambiguous. Likewise, MS in-source neutral loss of these labile residues during ionization process cannot be distinguished from naturally existing range of glycoform heterogeneity with various degrees of incomplete sialylation and fucosylation. Another major issue is sialylation, which also contributes to negative charges unless first neutralized (often inefficient and incomplete (Toyoda et al., 2008)) or removed (not desirable). In contrast, MALDI-MS mapping of permethylated glycans coupled with a complementary low- and high-energy CID on Q/TOF and TOF/ TOF, respectively, is unrivaled in its simplicity, robustness, and definitive sequencing ability (Yu et al., 2006). Furthermore, inherent within the sample preparation procedures are effective means to remove hydrophilic salts from biological matrices, especially when dealing with glycomics of whole cell lysates. We have thus taken the lead over the last few years to develop the enabling sample preparation techniques for sulfoglycomics, based on MALDI-MS and CID MS/MS analyses of permethylated glycans. Both our own data (Mitoma et al., 2007; Yu et al., 2009) and that similarly undertaken by others (Lei et al., 2009) have since shown that the commonly used NaOH/DMSO slurry permethylation method first introduced by Ciucanu and Kerek (Ciucanu and Kerek, 1984) can fully retain the sulfate while neutralizing all negative charges carried on the sialic acids. This derivatization effectively leaves the sulfated glycans as the only negatively charged species and thus can be preferentially detected in negative ion mode screening at high sensitivity, and/or efficiently separated from the overwhelmingly high abundant, nonsulfated species, for them to be additionally detected by MALDI-MS in positive ion modes and subjected to linkage and sequence informative high-energy CID MS/MS, albeit at a lower sensitivity. We describe below a routinely applicable protocol as practiced in our laboratories for such a first screen MALDI-MS analysis while the more


7

advanced approaches, some of which still under development and must be tailored for each specific applications, are only briefly discussed in general terms under future perspectives. Please read chapter 4 of Volume 480 of this series for ‘‘Glycomics Profiling of Heparan Sulfate Structure Activity’’.

2. Sample Preparation Over the last 5 years or so, a generic glycomic analysis workflow and associated methodologies for handling glycoprotein extracts from biological fluids, cells, and tissue samples are already in routine operation in many leading glycoanalytical laboratories, and well described in published work, including chapters in this series. The overall workflow we routinely used is essentially similar to the one adopted and publicized by the analytical core of the Consortium for Functional Glycomics ( Jang-Lee et al., 2006). In brief, following detergent or guanidine chloride extraction from cells, reduction alkylation and tryptic digestion of the proteins, N-glycans are released by PNGase F whereas the O-glycans are additionally released from the de-Nglycosylated peptides by reductive elimination. These glycan preparation steps are not further described in detail here other than a general descriptive account, with notes given in cases where alternatives are recommended and can be favorably considered. We have since extended this platform to analysis of sulfated glycans by further optimizing the permethylation protocols and to introduce screening by MALDI-MS in negative ion mode using a better matrix to enhance detection sensitivity (Yu et al., 2009). The key issue as compared with previous permethylation protocols (Dell et al., 1994; Jang-Lee et al., 2006) is to avoid aqueous-phase partition particularly for smaller O-glycans, and/ or the multisulfated ones. Subsequent microscale fractionation based on amine-beads packed in microtips allows enrichment of the sulfated glycans away from the predominating nonsulfated ones, so as to enable better chances of detection and sequencing in positive ion mode, taking advantages of the fragmentation characteristics already established. This extended workflow (Fig. 1.2) represents the first screen sulfoglycomic strategy that would be applied to any not previously analyzed sample to confirm the presence of sulfated N- and/or O-glycans, and to gauge their complexity, abundance, and most salient features.

2.1. From biological sources to glycoprotein extracts Apart from secreted glycoproteins found in biological fluids or culture media, most glycoproteins of interest are membrane glycoproteins, which are typically extracted from cell lysates using either chaotropic agents (e.g.,

8


Glycoproteins extracts Glycome cells and tissue

Glycans

NaOH/DMSO perMe glycans

C18 SPE

± Glycosidase ± Fractionation

Sulfated + nonsulfated glycans

NH2 SPE

Analysis of nonsulfated glycans by +ve MALDI-MS and MS/MS Screening for presence of sulfated glycans by –ve MALDI-MS

Nonsulfated glycans Sulfated glycans

–ve & +ve MS + MS/MS

Figure 1.2 Schematic workflow for a sulfoglycomic analysis. Extending from released glycans, the key steps are C18 Sep-Pak clean up of the permethylated glycans and the subsequent amine-beads fractionation of the sulfated permethylated glycans from the nonsulfated ones if preliminary screen by MS in negative ion mode reveals their presence. SPE, solid-phase extraction.

urea, guanidine hydrochloride) or mild detergents. With the former, cell lysates are often first delipidated by organic solvent, usually a combination of chloroform/methanol/water, and this fraction can be used for analysis of glycolipids if needed (Guerardel et al., 2006; Parry et al., 2007). However, optimizing for a good recovery of glycolipids may require repeated cycles of extractions with organic solvents of increasing polarity, which may compromise the yield of glycoproteins from remaining cell pellets. This should be avoided if only the N- and O-glycans are to be profiled. In our laboratories, we typically use 6 M guanidine hydrochloride (in 50 mM Tris–HCl, pH 8.4) for extraction after a simple delipidation of lyophilized cell lysates (Yu et al., 2009). Solubilized (glyco)proteins are then reduced by 20 mM dithiothreitol (Sigma) in 6 M guanidine–HCl, followed by alkylation with 50 mM iodoacetamide (Sigma) in 6 M guanidine–HCl, and subsequently dialyzed against ddH2O. The alternative use of detergents (e.g., 1% CHAPS, 1% Triton X-100) to solubilize membrane proteins from total cell lysates or microsomal fraction without the first delipidation step often produces cleaner extracts and sometimes is needed to preserve the conformation of glycoproteins for subsequent affinity capture or activity assays used in purification. It is popular for glycoproteins to be further subjected to rounds of lectin/ antibody captures, with possibilities for buffer exchange, or for glycans eventually to be recovered from protein SDS-PAGE gel bands or PVDF blots (Hernandez Mir et al., 2009). Otherwise, a common problem for MSbased glycomic analysis is the subsequent removal of detergents which is not trivial. The selective loss introduced by methods such as trichloroacetic acid, acetone, or ethanol precipitation, or passing through specialized detergentremoving columns, can be an important factor needs considerations. In this context, CHAPS, which forms low molecular weight micelles is more readily removed by dialysis or gel filtration than Triton X-100, which has a low critical micelle concentration (CMC) and forms high molecular weight micelles. The use of 1% CHAPS has recently been successfully


9

demonstrated for direct glycomic analysis (Babu et al., 2009) while we have been able to use 1% Triton X-100 followed by TCA precipitation, with no significant difference in the resulting glycomic profiles as compared with those obtained through guanidine–HCl. However, neither has been critically evaluated for applications to sulfoglycomics.

2.2. From glycoproteins to released N- and O-glycans (Glyco)proteins are commonly digested with a combination of trypsin (Sigma) and chymotrypsin (Sigma), each for 4 h, in 50 mM ammonium bicarbonate buffer, pH 8.4, at 37 C, followed by treatment with PNGase F (Roche) (Note 1). Released N-glycans are isolated from the de-N-glycosylated peptides by passing through C18 Sep-Pak cartridge (Waters, Part No. WAT051910) in 5% acetic acid (Note 2). Retained peptides are then eluted with 20–60% propanol/5% acetic acid and used for subsequent release of O-glycans by reductive elimination (1 M NaBH4/ 0.05 N NaOH, 37 C, 3 days) (Note 3). After terminating the chemical reaction by dropwise addition of glacial acetic acid on ice, the neutralized sample is taken through Dowex 50 8 column (50–100 mesh, Hþ form, Bio-Rad) in 5% acetic acid, dried, and coevaporated with 10% acetic acid in methanol to remove borates. Notes 1. Sequencing grade trypsin is recommended if a portion of the tryptic peptides is to be additionally subjected to proteomic analysis for peptide mapping and protein identification. Otherwise, any good commercial source of trypsin supplemented by less specific chymotrypsin will be preferable to effect more extensive proteolytic cleavages at lower cost. Optional reverse-phase isolation of the resulting glycopeptides/peptides by, for example, C18 Sep-Pak can be performed prior to N-glycan release by PNGase F, so as to obtain a cleaner sample free from contaminating free glycans, as previously recommended ( Jang-Lee et al., 2006). However, it may run the risk of losing glycopeptides that are too hydrophilic and thus not favorably retained by C18. For this reason, we have omitted this additional step and relying instead on other subsequent clean up steps at the glycan level. 2. For the same reason as above and to minimize sample handling so as to optimize yield, the PNGase F digested sample can be directly subjected to reductive elimination without first isolating away the released N-glycans. This will lead to reduced N-glycans and O-glycans to be recovered within a single fraction after subsequent desalting steps. The disadvantage, however, is that it may unnecessarily complicate the resulting glycomic map, especially if the usually smaller O-glycans extend to larger ones and overlap with the mass range populated by the N-glycans.

10


3. Lower temperature and longer reaction time are used instead of the usual reductive elimination conditions (45 C, 16 h) to avoid too harsh a treatment of the sulfated O-glycopeptides. For the O-glycans, the released oligoglycosyl alditols can be further subjected to mild periodate cleavages to derive the C2 and C4 halves for mapping of 6-arm and 3-arm extensions, respectively (Wu et al., 2007). The additional use of endoglycosidases such as endo-b-galactosidase and a range of exoglycosidases are common practice designed to help assignment of stereoisomers and terminal epitopes. The glycans derived thereof, with and without additional chemical or enzymatic cleavages, are then subjected to permethylation and first screened by MALDI-MS, with several most intense peaks and/or those of ambiguous molecular composition assignment be selected for a complementary low- and high-energy CID MALDI MS/MS sequencing (Yu et al., 2006).

2.3. Permethylation and microscale fractionation Strategic considerations—In general, if the cells to be analyzed contain any significant amount of sulfated glycans, such as the leukocytes, or cell lines transfected with sulfotransferases, it is a good practice to start with permethylating the glycan sample derived from only an equivalent of 107 cells or less as a first attempt. This means that one can start with 1/10th of a sample prepared from 108 cells or 1/5th of that from 5 107 cells, for example, and leave enough native sample for additional clean-up, fractionation, desialylation, and/or glycosidase digestions, when needed. On the other hand, if the yield of the sulfated glycans is significantly less than anticipated as revealed by first MS screen, all or the majority of the remaining samples can be permethylated to attempt securing at least a decent mapping at the MS level. As described above, the actual permethylation step including the preparation of the NaOH/DMSO slurry is similar to the conventional methods already well described in the literature. Any laboratory already well versed in NaOH permethylation can readily adapt their own protocols by incorporating a suitable clean up step after the reaction. In this context, a direct fractionation of the permethylated glycans by reverse-phase C18 Sep-Pak cartridge often leads to overlapping separations and can be avoided by pooling together the eluates, if so desired. Subsequent anion exchange step can be more efficient in isolating the negatively charged sulfated glycans away from other nonbinding, nonsulfated, neutral glycans. The main purpose of the initial C18 Sep-Pak step then is to simply function as a cleanup procedure for the reaction mixtures containing sodium salts and DMSO, substituting for the more commonly used chloroform/water partition. The latter will not efficiently retain the sulfated glycans in the organic phase except for the monosulfated, larger N-glycans. Under these general considerations, many


11

variations in terms of judicious choice of the microscale fractionation scheme are possible and largely sample dependent. The protocols described below include a description of the anticipated fractionation behaviors of each different classes of N- and O-glycans, based on which one can device a permutation of steps most sensible for the sample to be analyzed. 2.3.1. Methods (1) An aliquot of the glycan sample to be permethylated is dried down in a screw-capped glass tube or reaction vial. (2) Add approximately 200 ml of a slurry of finely ground NaOH pellets (Merck, pellets for analysis, ISO grade) in dimethyl sulfoxide (DMSO, Merck, purity 99.9%, max. 0.025% H2O) to the sample, followed by addition of 50–100 ml of methyl iodide (Merck, purity 99%, stabilized with silver for synthesis). (3) Gently vortex the reaction mixture for 3 h at 4 C and then quench the reaction on ice with 200 ml of cold water, followed by careful neutralization with 5% aqueous acetic acid until reaching a final pH of about 5–6 (checked with pH indicator paper). Note: Nonsialylated, sulfated glycans are often more readily fully permethylated, with a reaction time ranging from less than 20 min to longer than 3 h, at room temperature or 4 C, without any adverse effect. In contrast, best yield of fully permethylated, sialylated, sulfated glycans appears to be favored by lowering the reaction temperature concomitant with a longer reaction time. (4) Pass the neutralized reaction mixture directly through a primed C18 Sep-Pak cartridge (Waters, Part No. WAT051910) to obtain the desalted, permethylated glycans. a. Condition the Sep-Pak cartridge by sequential washing with acetonitrile (3 ml), methanol (3 ml), and water (3–5 ml) before sample loading. b. Hydrophilic salts and contaminants are step-wise washed off with 5 ml each of water, 2.5% and 10% acetonitrile. c. Permethylated glycans carrying one or two negative charges conferred by sulfates are next eluted with 5 ml of 25% acetonitrile. Note that in addition to disulfated N-and O-glycans, part of monosulfated O-glycans and smaller N-glycans, as well as some small nonsulfated O-glycans will be collected in this fraction. d. Permethylated glycans with a single charge can then be eluted with the next 5 ml of 50% acetonitrile. Note that a substantial amount of monosulfated N-glycans and larger monosulfated O-glycans, along with most of the nonsulfated N-glycans and O-glycans will be found in this fraction. A final elution step of 75–100% acetonitrile can be included to recover any larger, nonsulfated glycans.

12

(5)

(6) (7)

(8)


e. It may be advantageous to pool the 25–50% acetonitrile fractions since some overlapping of monosulfated glycans in both these two fractions is inevitable. On the other hand, it may be desirable to keep them separate as the earlier eluting 25% acetonitrile fraction is enriched with multiply sulfated glycans that are not overwhelmed by the neutral and monosulfated ones. For applications to MS analyses, the permethylated samples thus obtained can be further cleaned up and/or concentrated by using a ZipTipC18 (Millipore, Cat. No. ZTC18S096) or any other equivalent microtip devices. Occasionally, samples eluted off from the initial C18 Sep-Pak may not be sufficiently clean or free of suppressing contaminants to register any MS signal but will do so after such additional microscale clean up. a. Redissolve the permethylated glycans in 20 ml of 10% acetonitrile. b. Condition the ZipTipC18 with 10 ml of 50% acetonitrile followed by 10 ml of 0.1% trifluoroacetic acid. c. Pipette the sample in ZipTipC18 several times (10–20 times) to allow binding. d. Wash away salts and other hydrophilic contaminants with 10 ml of 0.1% trifluoroacetic acid (repeated pipetting for three to five times). e. Elute the permethylated sulfated glycans with 10 ml of 50% acetonitrile/0.1% trifluoroacetic acid, which can be collected into microtubes or directly spotted onto the MALDI target plate. Note that direct spotting from ZipTip is a most efficient way to apply as much of the sample for MALDI-MS analyses. MALDI-MS screening in positive and negative ion modes (see later). Although permethylation can be effected on crude biological extracts, excessive high salt content and other contaminants may prevent efficient full methylation. This is not normally a severe problem for analysis of nonsulfated glycans in positive ion mode but glycans with undermethylation on the sialic acids have been observed to present the major signals in negative ion mode instead of the less abundant sulfated glycans. Under such circumstances, the permethylated sample can be subjected to a second round of permethylation, which often improves overall signal quality and reduces or abolishes negative ion signals contributed by under-methylated species. Alternatively, a further clean up of the remaining native N- and Oglycan samples by graphitized carbon column prior to permethylation is recommended. a. Condition the carbon column (graphitized carbon column, Supleco, Cat. No. 57088) by sequential washing with acetonitrile (3 ml), 75% acetonitrile/0.1% trifluoroacetic acid (3 ml), 50% acetonitrile/0.1% trifluoroacetic acid (3 ml), 25% acetonitrile/0.1% trifluoroacetic acid (3 ml), and water (3–5 ml).


13

Note: Other micro-column or microtip packed with porous or non-PGC beads can be used depending on sample amount. b. Dissolve the native glycans in 200 ml of water for loading onto the graphitized carbon column. Salts are washed away with 2 5 ml water. c. Step-wise elute the glycans with 2 ml of 25% acetonitrile/0.1% trifluoroacetic acid, 50% acetonitrile/0.1% trifluoroacetic acid, and 75% acetonitrile/0.1% trifluoroacetic acid, and dry down. d. Note that simple fractionation can also be accomplished with the graphitized carbon column by first eluting the neutral glycans in 25–75% acetonitrile, and then the acidic glycans in 25–75% acetonitrile/0.1% trifluoroacetic acid. (9) To facilitate MS detection and analyses in positive ion mode, the permethylated sulfated glycans can be conveniently separated from coeluting nonsulfated ones by using a microcolumn device self-packed with amine-beads. ˚ pore a. Take up the amine beads (Nucleosil, 5 mm particle size, 100 A size) in methanol and pack into a pipette tip with its tapered end plugged by filter paper. Depending on the sample quantity to be handled, the volume of the packed beads can range from as little as 0.5 ml similar to a ZipTip size, to about 5 ml or more, using microtips of different sizes in conjunction with different wash/elution volumes (up to 5 or more of the bead volume). b. Condition the packed amine microtip (total bead volume 2 ml) by sequential washing with 10 ml each of 95% acetonitrile/0.1% formic acid, 50% acetonitrile/ 0.1% formic acid, and 95% acetonitrile/0.1% formic acid. c. Dissolve the permethylated glycans in acetonitrile for loading, with two to three times reloading of the flow through. Nonsulfated and thus neutral permethylated glycans and other hydrophobic contaminants are collected in the final unbound fraction, to be pooled with additional 50 ml wash of 95% acetonitrile. d. Elute the monosulfated permethylated glycans with 30 ml of 2.5 mM ammonium acetate in 50% acetonitrile, and di- or multisulfated ones with 30 ml of 10 mM ammonium acetate in 50% acetonitrile, respectively.

3. MS Analyses and Data Interpretation In general, MS analyses of the permethylated glycan samples can be considered and undertaken at three different stages within the overall workflow. The first stage serves primarily as a quick screen to detect the presence

14


or otherwise of sulfated glycans by MALDI-MS in negative ion mode while an accompanying positive ion mapping would inform the overall glycomic complexity. Any excessive contaminants, high degree of incomplete methylation, or unexpected low yield, will be revealed at this stage. As described above, a repermethylation of the permethylated sample, or to attempt a second permethylation after additional sample clean up, may help to improve data quality. More importantly, screening at this stage will inform if there is sufficient amount of sulfated glycans detected for subsequent stages of analyses, aiming at gaining more structural information. Depending on the signal intensity registered, it can be decided if there is merit in further isolating the sulfated glycans by additional amine-beads fractionation, so as to enable MS/ MS analysis in positive ion mode. Alternatively, more of the remaining native sample (or from a new preparation) should be permethylated, with or without precleaning, fractionation, desialylation, and other treatments. Either during the initial screen or subsequent ones after additional sample preparation, one to several major peaks of interest can be selected for MS/MS analysis. These may not always be the most abundant molecular ions detected but ideally correspond to the ones that are most informative with respect to producing data that will confirm the presence of a particular sulfated glycotope. In both these first and second stages, MALDI-based MS and MS/MS analyses is preferable since it is more straightforward, rapid, easier to operate, and thus less demanding in terms of technical skill and instrument time. However, for laboratory without MALDI-based instrument, offline nanospray (nanoESI)based analyses is equally applicable. Based on the data collated and knowledgebased interpretation, a final third stage may be carried out, aiming at either a targeted approach to identify specific glycan/glycotope of interest not revealed by initial mapping, or a more comprehensive glycomic mapping at MS2 level, usually by subjecting to automated LC–MS/MS. While general protocols for sample preparation can be formulated and readily followed by most laboratories equipped to carry out general biochemical analysis, the actual mileage gained with respect to the MS data quality and final useful information that can be gleaned will vary greatly among different laboratories supported by respective common MS facilities. The MS data acquisition parameter itself is highly instrument dependent. So are the performance (sensitivity, accuracy, and resolution) and afforded MS/ MS fragmentation characteristics. Only some general guidelines can be given here, which are based primarily on our own experience with our own MALDI-MS instrument (AB 4700 Proteomics Analyzer).

3.1. MALDI-based MS and MS/MS analysis As mentioned, the best conditions for spotting permethylated glycan samples onto MALDI target plate is instrument (the material and format of the target plate) and environment (humidity, ambient temperature) dependent,


15

and need to be carefully optimized by individual laboratories for best results. In these respects, the commercially available trisulfated neocarrahexose standard (Dextra Laboratory, Cat. No. C1010), which can be permethylated directly, and the bovine thyroid stimulating hormone (bTSH, Sigma-Aldrich, Product No. T8931), which requires a full procedure of tryptic digest, PNGase F treatment, and RP C18 Sep-Pak before permethylation, and thus can be additionally used to check against the glycan release protocols, are very useful performance benchmark standards. For highest sensitivity, the ‘‘best’’ MALDI matrices to be used for detecting permethylated glycan samples in positive and negative ion modes are found to be different. In general, the 2,5-dihydroxybenzoic acid (DHB) matrix is commonly considered as well suited for detecting glycans in the positive ion mode. However, it was found to be less sensitive in supporting detection of negatively charged glycans in negative ion mode, especially when the sample contains significant amount of salts. Alternative matrices such as arabinosazone (Chen et al., 1997), 2,4,6-trihydroxyacetophenone (THAP) (Papac et al., 1996), and 3,4-diaminobenzophenone (DABP), have each found preferences and registered better results in different labs. In our hands, DABP, which was originally used to detect oligonucleotides by MALDI-MS (Fu et al., 2006; Xu et al., 2006), was found to give best sensitivity for detecting permethylated sulfated glycans in the negative ion mode, at a laser energy setting similar to that used for DHB. It does, however, induce a higher degree of in source neutral loss of sodium sulfite (102) from di- and multiply sulfated glycans in negative ion mode. 3.1.1. Methods (1) Premix an aliquot (typically 0.5 ml) of suitably concentrated permethylated sulfated glycan sample in acetonitrile 1:1 (v:v) with DHB matrix (10 mg/ml in 50% acetonitrile), or 1:1 (v:v) with DABP matrix (10 mg/ml in 75% acetonitrile/0.05% trifluoroacetic acid) (Acros Organics, NJ, USA) in a microtube, and then carefully spot 0.5–1 ml of the sample/matrix mixture onto a heated MALDI target plate (placed above a 50 C hot plate). (2) In cases where the analyte fails to crystallize well in a small focused spot, a respotting of new aliquot tends to produce better results than recrystallization. A substantial amount of practice and experience is needed to be able to comfortably handle the spotting process, which can be an important factor to successful analysis. It is a good practice to keep the matrix solution relatively fresh and observe that it remains mostly colorless for DHB, while the yellow DABP solution does not turn brownish red. (3) Optimal instrument parameter and laser energy setting can be predetermined against the aforementioned standards.

16


a. A fully permethylated, trisulfated neocarrahexose is normally eluted in the 25% acetonitrile fraction from C18 Sep-Pak. Using DHB as matrix, MALDI-MS analysis should afford a major [Mþ4Na3H]þ molecular ion signal at m/z 1419.2922 (accurate monoisotopic mass) in positive ion mode, and an [Mþ2Na3H] signal at m/z 1373.3126 (accurate monoisotopic mass) in negative ion mode. Up to 3 and 2 degrees of successive losses of sodium sulfite from the parent ion may be observed in positive and negative ion modes, respectively. The effect of substituting the DHB matrix with DABP for analysis in negative ion mode can be evaluated, particularly in terms of sensitivity gain versus extent of sodium sulfite loss. Importantly, fully methylated disulfated molecular ions should not be detected at any significant level to give assurance that the permethylation process does not induce unwarranted desulfation. As a gauge of sensitivity, 100 pmol of starting material should afford strong signals when 1/20th equivalent of the permethylated sample is spotted. b. The bTSH N-glycans include both mono- and disulfated glycans. Using DHB as matrix, MALDI-MS analysis in negative ion mode should give a major [MþNa2H] molecular ion signal at m/z 2456.0548, corresponding to a core fucosylated biantennary complex type N-glycan with two sulfated LacdiNAc termini. As above, loss of sodium sulfite (102 U) may be observed along with an [MH] signal at 88 U lower, corresponding to monosulfated species. The relative amount of these signals varies according to instrument, matrix used, and sample preparation but the disulfated species should always be significantly more abundant than the monosulfated one, which may even not be detected at all. Other major signals that can be detected include the [MH] of hybrid type N-glycans containing one sulfated LacdiNAc, with and without fucose, at m/z 2081.9729, and 1907.8837, respectively. Switching to positive ion mode allows one to further evaluate how readily these negatively charged mono- and disulfated glycans can be observed, with the disulfated biantennary glycan now detected as [Mþ3Na2H]þ at m/z 2502.03436 or 102 U thereof due to loss of one or more sodium sulfite. In this regard, bTSH is a good standard as it does not contain significant level of nonsulfated glycans and thus allow the permethylated sulfated glycans to be directly observed in the positive ion mode, without further fractionation. (4) For MALDI-MS instrument equipped with capability of CID MS/MS such as the MALDI Q/TOF or MALDI TOF/TOF, the observed signals can be selected for MS/MS analyses in positive and/or negative ion modes. Similarly, the sulfated glycans from bTSH can first be evaluated for their fragmentation characteristics and optimization of laser and CID


17

settings. Prior to this, it is expected that the MS instruments have been well set up to perform MS/MS analysis on nonsulfated permethylated glycans and one is familiar with the fragmentation pattern afforded.

3.2. Interpretation of MALDI-MS profile of permethylated sulfated glycans MALDI-MS normally affords only singly charged molecular ions, typically as sodiated forms for permethylated glycans in positive ion mode. In general, each sulfate group, when present, will be additionally balanced by an extra sodium cation. Glycans carrying n sulfate groups will therefore be detected as molecular ion species conforming to the general formula of [Mþ (nþ1)NanH]þ. When switching to negative ion mode, permethylated glycans, with all OH groups being derivatized to O-Me and the carboxylic group of sialic acids methyl esterified, do not deprotonate or acquire a counter anion readily and thus are not normally observed unless sulfated. This feature makes detection in negative ion mode rather selective and extremely informative for a first screen for the presence of sulfated glycans before attempting any further fractionation. Monosulfated species typically ionize as [MH] whereas additional sulfates are likewise counter balanced by sodium, giving rise to singly charged [Mþ(n1)NanH] molecular ions in the negative ion mode. It is useful to note that the same sulfated glycan signals detected in negative ion mode are thus expected to shift to 46 U higher in positive ion mode by carrying two extra sodium. It is also consistent with the observation that each neutral loss of sulfate occurs in the form of losing both the sulfite (80 U) and its sodium counter cation (22 U), and registered as 102 U. This essentially produces an ‘‘under-methylated’’ species, with the original O-SO3Na being converted to free OH. A signal detected at 14 U lower than a monosulfated species can thus be interpreted as suggestive evidence for the presence of disulfated species having lost a sodium sulfite (Fig. 1.3A). In our experience, intact multiply sulfated glycans from cells and tissues, as opposed to more abundant standard sulfated glycoprtoeins, are not readily observed in a MALDI-MS-based sulfoglycomic analysis, even in negative ion mode. Instead, these can be more favorably detected as ‘‘under-methylated’’ monosulfated species at multiples of 14 U lower, after losing their extra sodium sulfite moieties. The use of DABP to increase overall detection sensitivity in negative ion mode unfortunately also promotes loss of sodium sulfite to the extent that the parents that retain the extra sulfates may be completely absent. Since these in-source generated fragment ions are indistinguishable from molecular ion signals corresponding to genuine undermethylated species, additional evidence is highly desirable.

18


A

14 u

100 % Intensity

1821.7

14 u

1835.7 50

2182.9

1747.7

2196.9

0 B

S

100

% Intensity

S

50

1747.8

1835.9 2197.1

0 1680

% Intensity

C

100

−SO Na 3

−SO

3Na

1821.9

S2

Na

2183.1

S2 Na

50

0 1680

2010 m/z

2340

Figure 1.3 Distinguishing disulfated glycans from under-methylation by fractionation. Without separation (A), signals at 14 U lower than a monosulfated species may indicate the presence of a genuinely under-methylated species, or a disulfated species of the same glycosyl composition having lost its extra sulfate group by in source fragmentation. By further subjecting this permethylated sample to fractionation by amine-beads, monosulfated glycans would be eluted in the earlier 2.5 mM NH4OAc/50% acetonitrile fraction (B) whereas the disulfated glycans in the 10 mM NH4OAc/50% acetonitrile fraction (C).

In general, the fractionation scheme as described in previous sections will help. By C18 Sep-Pak, di- and multiply sulfated species are normally collected only in the 25% acetonitrile fraction and thus the 14 U or ‘‘under-methylation’’ signals should mostly be detected there and not in the 50% acetonitrile fraction. Alternatively, if fractionation is not attempted at the C18 Sep-Pak stage, the mono- and disulfated species that constitute the 14 U signal pairs detected initially can be fractionated subsequently by


19

differential elution off the amine-beads (Fig. 1.3B and C). A more direct evidence can then be obtained by subjecting the fraction enriched with multiply sulfated glycans to an offline nanoESI-MS analysis, which is more likely to retain the extra sulfate moieties. With experience and performed carefully, any genuine under-methylation should be minimum and generally does not pose a significant problem. However, for sulfoglycomic analysis in negative ion mode, under-methylation occurring at the carboxylic group of sialic acid can contribute to preferential detection of nonsulfated, sialylated glycans. Care should therefore be taken to not mistakenly attribute these negative ion signals to sulfated glycans, which may in turn be obscured. This issue cannot be addressed by fractionation but can be alleviated by repermethylation. Removal of sialic acids by neuraminidase prior to permethylation will overcome this problem and confirm the presence of sulfated glycans, but at the expense of losing any information on sialylated, sulfated epitope. With the above considerations in mind, one can set out to assign the glycosyl composition of the detected peaks with ease, using the general formula of X m=z ¼ ðCH3 þ OCH3 Þ þ glycosyl residual masses þ ðOSO 3 OCH3 Þ for singly charged, permethylated, monosulfated [MH] molecular ion in negative ion mode. (CH3 þ OCH3) accounts for the reducing and nonreducing end masses, which is 46 U in nominal mass and can be modified accordingly if the reducing end is reduced, or further tagged. For example, another 16 U (2H þ CH2) is added for the O-glycans released by reductive elimination. The first sulfated group added will account for a mass increment of 65 U and contributing to the negative charge. Thereafter, each additional sulfate group contributes to 88 U mass increment, assuming an O-SO3Na substituting for O-Me, as discussed above. For simplicity, one canPderive the m/z for [MH] of permethylated, sulfated molecular ions as glycosyl residual masses þ 111 U þ 88(n 1), where n ¼ number of sulfate groups contained. A full list of the glycosyl residual masses has been tabulated (Dell et al., 1994), the most common ones being 204, 245, 174, 361, and 391 U, for the nominal masses of Hex, HexNAc, dHex, Neu5Ac, and Neu5Gc, respectively. The calculation for positive molecular ions is similar, but adding another 46 U in general to account for the extra two sodium, as discussed above. With some glycobiology background knowledge particularly a familiarity with the range of core and terminal structures permitted by the host biosynthesis machinery, manual assignment is relatively simple. It can also be easily automated using a variety of publicly available software program, or self-written scripts. One should, however, bear in mind that such

20


glycosyl composition assignment is often equivocal because of similarity in mass of several glycosyl permutation. This includes the 88 U increment, which can be assigned as carrying an extra O-SO3Na in place of O-Me as described, or an extra LacNAc unit (449 U) in place of Neu5Ac (361 U). 2 Fuc (174 2 ¼ 348 U) differs from a Neu5Ac by 13 U, which can be confused with aforementioned under-methylation (14 U), especially at high masses when the overlapping isotopic clusters cannot be adequately delineated. In nonhuman samples, the additional presence of Neu5Gc further complicate assignment as the difference between Neu5Ac and Neu5Gc at 30 U, also coincides with the difference between Hex and dHex, and that nonhuman samples often also carries extra a3-Gal caps. All these factors contribute to uncertainty associated with glycosyl composition based on MS data alone, and is particularly problematic with low resolution and/or accuracy data. Provided sample amount is not limiting, further MS/MS analysis on selected or all major peaks is therefore highly recommended.

3.3. CID MS/MS of permethylated sulfated glycans As noted, direct MALDI CID MS/MS manually acquired on selected peaks after initial MS mapping is most straightforward and elegantly simple for novice and expert alike. In our laboratories, routine positive ion mode MS/ MS data are manually acquired on both MALDI Q/TOF (Q/TOF Ultima, Micromass) and MALDI TOF/TOF (AB 4700 Proteomic Analyzer) for low- and high-energy CID, respectively. We also have limited experience on the AB 4800 MALDI TOF/TOF instrument and verified that similar high-energy CID fragmentation pattern as afforded by 4700 instrument can be obtained, with a caveat that the instrument must be specifically tuned up. The advantage of the TOF/TOF instruments is that, depending on analytical need, glycan fragmentation pattern akin to low-energy CID as afforded by ion traps, or Q/TOFs can also be obtained, in addition to what it is supposed to afford, namely a true high-energy CID. It should be further noted that, multiple collision events producing multiple cleavages occur more readily in the collision cell of a Q/TOF whereas most cleavages including those unique concerted double cleavages on the same glycosyl residue as observed on a TOF/TOF (Spina et al., 2004; Yu et al., 2006) are likely resulting from single collision. This translates into a fundamental difference between the two modes of CID MS/MS in that successive neutral losses of terminal structural motifs delimited by HexNAc, or the Neu5Ac, will not normally be observed on TOF/TOF. Conversely, with higher collision energy at kilovolt range, only the TOF/TOF readily provides linkage-specific cross ring and other concerted cleavages, which are more demanding in assignment. These fragmentation characteristics (Fig. 1.4) have been extensively discussed (Yu et al., 2006, 2008) and not repeated here, except for a few key points to help beginners to enter the field.

21


A

B 1,5

X

Y

MeO

H

C′′/Y O

O 1,5

E

B

MeNAc

MeO

G O R

O,4

D

A

X

C′′ [ C-2 H ]

O

A MeO

R-O

Y S

Y

O

3,5

B

O

O

MeO

N-Glycan O

O

O

O

S B

Core 1

S B

B

3 O

O Y

Core 2 O 6

Z

3 O

Z

O-Glycans

Figure 1.4 Common fragmentation pattern observed under MALDI CID MS/MS. The Domon and Costello nomenclature (Domon and Costello, 1988) for the glycosidic cleavage Y, Z and B, C ions, as well as the cross ring cleavage X and A ions, are commonly adopted. For high-energy CID MS/MS on MALDI-TOF/TOF, Spina et al. (2004) extended the nomenclature system to additional concerted cleavage ions around the ring, namely the E, F, G, ions. Further incorporating the D ion proposed by Harvey (2005), we integrated the system and additionally introduced the H ion, along with C00 , and C00 /Y ions (Yu et al., 2006). Only the more readily produced and/or useful ions are illustrated in (A). For low-energy CID MS/MS, the most abundant ions are the B and Y ions (B). Multiple cleavages often produce fragment ions corresponding to the core having lost all antennary extension. Z ions are mostly restricted to elimination of the 3linked substituents and most commonly observed for eliminating the 3-arm of O-glycan cores, and the Fuc of LeX.

(1) In reference to the commonly adopted Domon and Costello nomenclature (Domon and Costello, 1988), the most universal pair of fragment ions afforded by all different MS instruments is the B and Y ions. In the case of permethylated glycans, the Y ions correspond mostly to loss of terminal Neu5Ac and Neu5Ac-R-HexNAc, where R can be any combination of Hex and Fuc. These are most useful in identifying nonreducing terminal epitopes and would be further corroborated by the abundant B ions observed at the low mass range (Fig. 1.4B). While the masses of Y ions are dependent on the parent, the neutral loss itself and the corresponding B ions are characteristics of each specific epitope. It is a distinct limitation of ion traps that detection of these low mass B ions are disfavored by the 1/3rd cut-off rule although somewhat compensated by ability to perform MSn. (2) In Q/TOFs, successive neutral losses of the antennary units are common and by enumerating the free OH groups carried on the remaining core structures, the number of branching can be readily inferred. In TOF/TOF, all B and Y ions are normally resulting from single cleavage and are likewise mostly restricted to HexNAc and Neu5Ac. At these and other sites, high-energy CID is more prone to provide a complete series of cross ring cleavage 1,5X ions, which is very informative for a

22


full sequencing of glycosyl residues delimited by HexNAc. This is particularly so for some unusual terminal epitopes found in lower animals with additional galactosylation. For mammalian structures, the 1,5X ions derived from cleavages at the mannoses on the 3- and 6-arms of the trimmanosyl core is advantageous in defining the respective antennary substituents. (3) True linkage-specific ions are normally represented by the A ions, for example, 3,5A and 2,4A ions, which unfortunately are not abundantly produced even with high-energy CID, especially when sample amount is limiting. We found that other concerted cleavages such as the D and G ions are more readily observed instead and can provide linkage information (Fig. 1.4A). (4) Importantly, in positive ion mode, the additional presence of 6-OSO3Na on HexNAc or Hex does not significantly alter the established fragmentation pattern (Yu et al., 2009). In contrast to its ready loss during ionization in source, as discussed earlier, sulfated species selected for CID MS/MS does not exhibit further neutral loss of sodium sulfite. This allows direct applications of all previously established fragmentation patterns and rules to sulfated glycans but taking into account the mass increment conferred by the sodium sulfite moiety. (5) In negative ion mode, all fragment ions should retain a negative charge for them to be detected and therefore in the case of permethylated sulfated glycans, only ions retaining the sulfate moiety will generally be observed. An abundant ion at m/z 97 is ubiquitously present, corresponding to HSO4 and serves to confirm that the selected parent ion is indeed sulfated, but does not provide any further structural information. More useful are the B ions at m/z 324, 528, 702, 889/ 919, and 1063/1093, corresponding to sulfated HexNAc, LacNAc, LeX, Neu5Ac/Gc-LacNAc, and Neu5Ac/Gc-LeX, respectively, which will inform the kind of terminal sulfated epitopes presented (Mitoma et al., 2007). In practice, real cases of MALDI MS/MS applications to sulfoglycomics of biological samples are limited not in technical principles but more so by sensitivity. We have demonstrated that all the useful fragmentation described previously (Yu et al., 2006) can be similarly reproduced on sulfated glycans by a complementary of low- and high-energy CID in positive ion mode, including the O-glycans derived from Peyers patches of a single mouse (Yu et al., 2009). However, in further push for sensitivity to get below the 1–10 million cell threshold, especially for cells that do carry sulfated glycans but only at very low abundance or to specifically identify a particular unique sulfated glycotope, a further leap of MS/MS sensitivity is needed. Otherwise, sulfoglycomic mapping will only be realized at MS level.


23

4. Future Perspectives This chapter serves primarily to document a readily adapted set of protocols for preparing samples suitable for sulfoglycomic analysis based on MALDI-MS analysis of permethylated glycans. Taking into consideration the technical skills and time required, MALDI-MS screening in negative ion mode for permethylated glycans, with and without additional fractionation away of the much more abundant nonsulfated glycans, represents the most sensitive way of detecting the presence of monosulfated N- and O-glycans. Subsequent to this first screen, two immediate technical hurdles need to be crossed for advancing further sulfoglycomics. The first is to identify multiply sulfated species. It can almost be assumed that if monosulfated glycans are detected, di and multiply sulfated ones should also be present. The only issue is how abundant and whether the current experimental approach is sufficiently sensitive to detect them. The second is to effect comprehensive MS/MS, ideally on each of the detected sulfated glycan peaks, which again boils down to the issue of sensitivity versus how much starting materials one is willing to prepare. At a more structural details level, a critical issue is the ability to distinguish the positions of sulfate and thereby define the particular sulfo-glycotopes, for example, 6- or 60 -sulfo LeX. Our first attempt at this direction is to prime the sample preparation procedures toward MS/MS analysis in positive ion mode. With MALDI TOF/TOF, all issues related to structural details can, in principle, be addressed provided sample amount is not limiting in affording decent positive ion signals in the first place. However, to progress further, we have increasingly been relying on nanoESI-MS and MS/MS in more advanced MS instruments, such as the Thermo LTQ-Orbitrap Velos system. An offline analysis often allows detection at high resolution and accuracy of multiply sulfated glycans without losing the extra sulfates during MS ionization, at a sensitivity comparable to or better than MALDI-MS in negative ion mode. Signals observed can then be directly selected for MS/ MS, with a choice of CID MSn using the ion trap or a more Q/TOF like MS2 analysis using the HCD cell coupled with Orbitrap detection. Alternatively, a total ion mapping (Aoki et al., 2007) can be performed by fragmenting across the entire useful mass range using the HCD to avoid low mass cut-off and to use Orbitrap measurement for higher resolution. The initial results are very promising. With stable nanospray, the entire process can be accomplished within 10–15 min. Finally, an LC–MS/MS with several permutation of scan functions can be programmed for an even more in depth mining of the sulfoglycome.

24


We are thus cautiously optimistic about the future with more anticipated advances in MS instruments. In the meantime, it is hoped that all laboratories engaging in glycobiology research can adequately prepare the permethylated sample to the stage described in details in this chapter, including a first screen by MALDI-MS in negative ion mode, while leaving the more advanced MS analysis to experts with access to state-of-the-art MS instruments.

ACKNOWLEDGMENTS The authors wish to acknowledge financial support provided by Academia Sinica and Taiwan NRPGM to the common Mass Spectrometry Facilities for Proteomics and Glycomics, established at the Institute of Biological Chemistry, Academia Sinica, over the last 5 years or so to make this body of work in MS methodologies development for sulfoglycomics possible.

REFERENCES Aoki, K., Perlman, M., Lim, J. M., Cantu, R., Wells, L., and Tiemeyer, M. (2007). Dynamic developmental elaboration of N-linked glycan complexity in the Drosophila melanogaster embryo. J. Biol. Chem. 282, 9127–9142. Babu, P., North, S. J., Jang-Lee, J., Chalabi, S., Mackerness, K., Stowell, S. R., Cummings, R. D., Rankin, S., Dell, A., and Haslam, S. M. (2009). Structural characterisation of neutrophil glycans by ultra sensitive mass spectrometric glycomics methodology. Glycoconj. J. 26, 975–986. Chen, P., Baker, A. G., and Novotny, M. V. (1997). The use of osazones as matrices for the matrix-assisted laser desorption/ionization mass spectrometry of carbohydrates. Anal. Biochem. 244, 144–151. Ciucanu, I., and Kerek, F. (1984). A simple and rapid method for the permethylation of carbohydrates. Carbohydr. Res. 131, 209–217. Dell, A., Reason, A. J., Khoo, K. H., Panico, M., McDowell, R. A., and Morris, H. R. (1994). Mass spectrometry of carbohydrate-containing biopolymers. Methods Enzymol. 230, 108–132. Domon, B., and Costello, C. E. (1988). A systematic nomenclature for carbohydrate fragmentations in FAB-MS/MS spectra of glycoconjugates. Glycoconj. J. 5, 397–409. Fu, Y., Xu, S., Pan, C., Ye, M., Zou, H., and Guo, B. (2006). A matrix of 3,4-diaminobenzophenone for the analysis of oligonucleotides by matrix-assisted laser desorption/ ionization time-of-flight mass spectrometry. Nucleic Acids Res. 34, e94. Geyer, H., and Geyer, R. (2006). Strategies for analysis of glycoprotein glycosylation. Biochim. Biophys. Acta 1764, 1853–1869. Guerardel, Y., Chang, L. Y., Maes, E., Huang, C. J., and Khoo, K. H. (2006). Glycomic survey mapping of zebrafish identifies unique sialylation pattern. Glycobiology 16, 244–257. Harvey, D. J. (2005). Structural determination of N-linked glycans by matrix-assisted laser desorption/ionization and electrospray ionization mass spectrometry. Proteomics 5, 1774–1786. Haslam, S. M., North, S. J., and Dell, A. (2006). Mass spectrometric analysis of N- and O-glycosylation of tissues and cells. Curr. Opin. Struct. Biol. 16, 584–591.


25

Hernandez Mir, G., Helin, J., Skarp, K. P., Cummings, R. D., Makitie, A., Renkonen, R., and Leppanen, A. (2009). Glycoforms of human endothelial CD34 that bind L-selectin carry sulfated sialyl Lewis x capped O- and N-glycans. Blood 114, 733–741. Jang-Lee, J., North, S. J., Sutton-Smith, M., Goldberg, D., Panico, M., Morris, H., Haslam, S., and Dell, A. (2006). Glycomic profiling of cells and tissues by mass spectrometry: Fingerprinting and sequencing methodologies. Methods Enzymol. 415, 59–86. Karlsson, N. G., and Thomsson, K. A. (2009). Salivary MUC7 is a major carrier of blood group I type O-linked oligosaccharides serving as the scaffold for sialyl Lewis x. Glycobiology 19, 288–300. Kawashima, H. (2006). Roles of sulfated glycans in lymphocyte homing. Biol. Pharm. Bull. 29, 2343–2349. Kimura, N., Ohmori, K., Miyazaki, K., Izawa, M., Matsuzaki, Y., Yasuda, Y., Takematsu, H., Kozutsumi, Y., Moriyama, A., and Kannagi, R. (2007). Human Blymphocytes express alpha2-6-sialylated 6-sulfo-N-acetyllactosamine serving as a preferred ligand for CD22/Siglec-2. J. Biol. Chem. 282, 32200–32207. Lei, M., Mechref, Y., and Novotny, M. V. (2009). Structural analysis of sulfated glycans by sequential double-permethylation using methyl iodide and deuteromethyl iodide. J. Am. Soc. Mass Spectrom. 20, 1660–1671. Mitoma, J., Bao, X., Petryanik, B., Schaerli, P., Gauguet, J. M., Yu, S. Y., Kawashima, H., Saito, H., Ohtsubo, K., Marth, J. D., Khoo, K. H., von Andrian, U. H., et al. (2007). Critical functions of N-glycans in L-selectin-mediated lymphocyte homing and recruitment. Nat. Immunol. 8, 409–418. North, S. J., Hitchen, P. G., Haslam, S. M., and Dell, A. (2009). Mass spectrometry in the analysis of N-linked and O-linked glycans. Curr. Opin. Struct. Biol. 19, 498–506. Papac, D. I., Wong, A., and Jones, A. J. (1996). Analysis of acidic oligosaccharides and glycopeptides by matrix-assisted laser desorption/ionization time-of-flight mass spectrometry. Anal. Chem. 68, 3215–3223. Parry, S., Ledger, V., Tissot, B., Haslam, S. M., Scott, J., Morris, H. R., and Dell, A. (2007). Integrated mass spectrometric strategy for characterizing the glycans from glycosphingolipids and glycoproteins: Direct identification of sialyl Le(x) in mice. Glycobiology 17, 646–654. Robbe-Masselot, C., Herrmann, A., Maes, E., Carlstedt, I., Michalski, J. C., and Capon, C. (2009). Expression of a core 3 disialyl-Le(x) hexasaccharide in human colorectal cancers: A potential marker of malignant transformation in colon. J. Proteome Res. 8, 702–711. Rosen, S. D. (2004). Ligands for L-selectin: Homing, inflammation, and beyond. Annu. Rev. Immunol. 22, 129–156. Shida, K., Misonou, Y., Korekane, H., Seki, Y., Noura, S., Ohue, M., Honke, K., and Miyamoto, Y. (2009). Unusual accumulation of sulfated glycosphingolipids in colon cancer cells. Glycobiology 19, 1018–1033. Spina, E., Sturiale, L., Romeo, D., Impallomeni, G., Garozzo, D., Waidelich, D., and Glueckmann, M. (2004). New fragmentation mechanisms in matrix-assisted laser desorption/ionization time-of-flight/time-of-flight tandem mass spectrometry of carbohydrates. Rapid Commun. Mass Spectrom. 18, 392–398. Toyoda, M., Ito, H., Matsuno, Y. K., Narimatsu, H., and Kameyama, A. (2008). Quantitative derivatization of sialic acids for the detection of sialoglycans by MALDI MS. Anal. Chem. 80, 5211–5218. Wu, A. M., Khoo, K. H., Yu, S. Y., Yang, Z. G., Kannagi, R., and Watkins, W. M. (2007). Glycomic mapping of pseudomucinous human ovarian cyst glycoproteins: Identification of Lewis and sialyl Lewis glycotopes. Proteomics 7, 3699–3717. Xu, S., Ye, M., Xu, D., Li, X., Pan, C., and Zou, H. (2006). Matrix with high salt tolerance for the analysis of peptide and protein samples by desorption/ionization time-of-flight mass spectrometry. Anal. Chem. 78, 2593–2599.

26


Yu, S. Y., Wu, S. W., and Khoo, K. H. (2006). Distinctive characteristics of MALDI-Q/ TOF and TOF/TOF tandem mass spectrometry for sequencing of permethylated complex type N-glycans. Glycoconj. J. 23, 355–369. Yu, S. Y., Khoo, K. H., Yang, Z., Herp, A., and Wu, A. M. (2008). Glycomic mapping of O- and N-linked glycans from major rat sublingual mucin. Glycoconj. J. 25, 199–212. Yu, S. Y., Wu, S. W., Hsiao, H. H., and Khoo, K. H. (2009). Enabling techniques and strategic workflow for sulfoglycomics based on mass spectrometry mapping and sequencing of permethylated sulfated glycans. Glycobiology 19, 1136–1149. Zaia, J. (2008). Mass spectrometry and the emerging field of glycomics. Chem. Biol. 15, 881–892.

C H A P T E R

T W O

Mass Spectrometric Analysis of Mutant Mice Simon J. North, Jihye Jang-Lee, Rebecca Harrison, Ke´vin Canis, Mohd Nazri Ismail, Alana Trollope, Aristotelis Antonopoulos, Poh-Choo Pang, Paola Grassi, Sara Al-Chalabi, A. Tony Etienne, Anne Dell, and Stuart M. Haslam Contents 1. Overview 2. Methods 2.1. General advice 2.2. Protocol 1: Preparation of samples for glycan release 2.3. Protocol 2: Preparation of glycans for analysis 2.4. Protocol 3: Optional sample preparation steps 2.5. Protocol 4: Derivatization and analysis of released samples 2.6. Protocol 5: Glycobioinformatics 3. Interpretation of Glycomic Data 4. Example Project: Characterization of Pancreatic Tissue from WildType and Mgat4a Knockout Mice 4.1. MALDI-TOF MS mass fingerprinting 4.2. MALDI-TOF/TOF MS sequencing 4.3. Enzymatic digestion—a-galactosidase treatment 4.4. Linkage analysis by GC–MS 4.5. Summary 5. Summary of Glycan Structural Observations in Murine Tissues, Cells, and Knockouts Acknowledgments References

28 31 31 31 34 40 49 57 59 62 62 62 65 67 69 69 73 74

Abstract Mass spectrometry (MS) has proven to be the preeminent tool for the rapid, high-sensitivity analysis of the primary structure of glycans derived from diverse biological sources including cells, fluids, secretions, tissues, and Division of Molecular Biosciences, Faculty of Natural Sciences, Imperial College London, London, United Kingdom Methods in Enzymology, Volume 478 ISSN 0076-6879, DOI: 10.1016/S0076-6879(10)78002-2

#


27

28

Simon J. North et al.

organs. These analyses are anchored by matrix-assisted laser desorption ionization time of flight (MALDI-TOF) analysis of permethylated derivatives of glycan pools released from the samples, to produce glycomic mass fingerprints. The application of complimentary techniques, such as chemical and enzymatic digestions, GC–MS linkage analysis, and tandem mass spectrometry (MS/MS) utilizing both electrospray (ES) and MALDI-TOF/TOF, together with bioinformatic tools allows the elucidation of incrementally more detailed structural information from the sample(s) of interest. The mouse as a model organism offers many advantages in the study of human biology, health, and disease; it is a mammal, shares 99% genetic homology with humans and its genome supports targeted mutagenesis in specific genes to produce knockouts efficiently and precisely. Glycomic analyses of tissues and organs from mice genetically deficient in one or more glycosylation gene and comparison with data collected from wild-type samples enables the facile identification of changes and perturbations within the glycome. The Consortium for Functional Glycomics (CFG) has been applying such MS-based glycomic analyses to a range of murine tissues from both wild-type and glycosylation-knockout mice in order to provide a repository of structural data for the glycobiology community. In this chapter, we describe in detail the methodologies used to prepare, derivatize, purify, and analyze glycan pools from mouse organs and tissues by MS. We also present a summary of data produced from the CFG systematic structural analysis of wild-type and knockout mouse tissues, together with a detailed example of a glycomic analysis of the Mgat4a knockout mouse.

1. Overview The mouse model system has long been providing the field of glycobiology with a steady stream of important information concerning the function of glycans and glycan-associated molecules in mammalian systems. By using targeted mutagenesis to knockout specific genes, the means are provided to investigate the biological importance of individual glycosyltransferases or specific glycan epitopes (Ioffe and Stanley, 1994; Metzler et al., 1994). Such methods have been used to reveal important aspects of glycan functions in the mammalian system (Bhaumik et al., 1998; Chui et al., 1997; Ellies et al., 1998), but have been somewhat limited by the lack of structural information. Complimentary knowledge of the precise nature of the glycan structural alterations perpetrated by these genetic abnormalities is an essential component in the investigation of these mice and their phenotypes, capable not only of providing concrete confirmation of structural features thereby proving or disproving hypotheses (Moody et al., 2001, 2003; North et al., 2010b), but also of indirectly revealing previously unknown biosynthetic information (Akama et al., 2006). However, though the glycosylation of various murine glycoproteins and some specific tissues has been established

Mass Spectrometric Analysis of Mutant Mice

29

structurally (Parry et al., 2006, 2007; Sutton-Smith et al., 2000, 2002), the glycan repertoires—or glycomes—of many murine tissues, organs, and cell types are poorly defined. Recognizing this issue, the Consortium for Functional Glycomics (CFG) has dedicated considerable resources toward providing a repository of information concerning the whole system analysis of the N-, O-, and glycolipid-linked structural glycomes of murine tissues. This data represents a foundation of structural knowledge and is available free of charge as a resource for the scientific community (CFG). Systematic structural characterizations of tissues and organs from pairs of age and sex matched C57BL/6 littermates as well as cell populations, for both wild-type and genetic knockouts, have been carried out utilizing a modern glycomics methodology based around mass spectrometry (MS). Structural glycobiological advancements made in the last 30 years have been driven in large part by advancements in mass spectrometric technology and techniques. MS today plays an integral role in the structural characterization of N- and O-linked glycans, glycolipids, and glycoconjugates. Within this chapter, we describe a refined, heavily tested glycomic methodology (Dell et al., 1994; Jang-Lee et al., 2006; Sutton-Smith et al., 2000) specifically tailored to structurally define N-linked, O-linked, and glycolipid-derived glycans from murine tissues, organs, and cell populations. The underlying strategy is based upon methods developed in the 1980s (Dell and Ballou, 1983; Fukuda et al., 1984a,b, 1985) and has been periodically updated to suit modern instrumentation and include other methodological developments (Dell, 1990; Dell et al., 1994; Jang-Lee et al., 2006; Sutton-Smith et al., 2000), but is still based upon the principle that permethylated glycans yield molecular ions at very high sensitivity, irrespective of the type of MS employed (Wada et al., 2007, 2010). In brief, pools of glycans are released sequentially from proteolytically digested biological samples. These glycan pools—N-linked, O-linked, and polar and nonpolar glycolipid-derived glycans—are then derivatized by permethylation prior to analysis by a range of mass spectrometric techniques. At present, the most commonly used instrumentation for profiling work is matrix-assisted laser-desorption ionization time-of-flight mass spectrometry (MALDI-TOF MS) or the tandem time-of-flight version MALDI-TOF/ TOF MS. Sequence information is obtained by collisionally activated decomposition (CAD) carried out by tandem mass spectrometry equipment, with electrospray (ES) and MALDI-TOF/TOF methods described in detail. Methods for the optional chemical or enzymatic degradation of the samples and linkage analysis by gas chromatography–mass spectrometry (GC–MS), as well as guidelines for the utilization of some glycoinformatic tools are also included. The complete methodology is shown in Fig. 2.1. The information generated from these methods are commonly further enhanced by the parallel analysis by glycan gene microarray screening (Bax et al., 2007; Comelli et al., 2006; Montpetit et al., 2009).

30


Protocol 1: Preparation of mice

Protein pellet

2.1: Cleavage of S-S bridges and protection of Cys

Supernatant 3.2b: Polar recovery 3.2c: Nonpolar recovery Upper phase

2.2: Proteolytic digest

Lower phase 2.3: Glycopeptide purification

Polar glycolipids

Nonpolar glycolipids

2.4: N-glycan release 2.5: N-glycan purification Propanol fraction

3.2d: Glycolipid oligosaccharide release 3.2e: Glycolipid oligosaccharide purification

2.6: O-glycan release

Aq. fraction

Polar glycolipid derived glycans

Nonpolar glycolipid derived glycans

2.7: O-glycan purification

N-linked glycans

O-linked glycans

Protocol 2: Preparation of glycans for analysis

Protocol 3: Optional sample preparation steps

3.2a: Glycolipid partitioning

1.2: Homogenization 1.3: Cell lysis

1.1: Excision of tissues/organs or isolation of cells

3.1: Homogenization/lysis with glycolipid extraction

3.3: Chemical digestion of released glycans 3.4: Enzymatic digestion of released glycans

Protocol 4: Derivatisation and analysis of released samples

4.1: Permethylation of released glycan pools 4.2: Permethylated glycan purification Permethylated glycan pools

4.3: MALDI-TOF Mass spectrometry Mass-fingerprinting 4.5: Electrospray Mass spectrometry Sequencing

Protocol 5: Bioinformatics

4.6: Partially methylated alditol acetates

4.4: MALDI-TOF-TOF Mass spectrometry Sequencing

5a: Cartoonist 5b: Glycoworkbench

4.7: Gas-chromatography Mass spectrometry Linkage analysis

Glycan structural data

CFG database

Figure 2.1 Overall strategy for the preparation and analysis of glycans from biological samples.


31

By employing these methods, the analytical glycotechnology core (Core C) of the CFG has produced detailed structural data on hundreds of murine tissue, organ, and cell samples. All of this data is deposited in the open access CFG databases (http://www.functionalglycomics.org) and is too vast a resource to completely distil into a single chapter (North et al., 2010a). However, within the second half of this chapter, we have compiled tables summarizing the major structural features observed in the wild-type and transgenic mouse samples analyzed by Core C. Sample MALDI mass spectra of selected mouse tissue and cell samples are also included, as well as an exemplar application of the methodology in the comparative analysis of a wild-type and mgat4a knockout mouse pancreas sample.

2. Methods 2.1. General advice 1. For all aqueous solutions, water additions, and cleaning of equipment, use ‘‘ultrapure’’ 18.2 MOcm3 distilled/deionized water. 2. Wherever possible, use glassware, Pyrex disposable culture tubes, and glass disposable pipettes rather than plastic. It should be noted that acid hydrolysis reactions involving hydrofluoric acid (HF) is the major exception to this rule (see Protocol 3.3a). 3. Ensure all glassware is thoroughly cleaned and dried before use. 4. Avoid the use of detergents wherever possible. Do not wash equipment with detergents, for example. 5. Avoid the use of anything that could contaminate your sample with carbohydrates (cellulose from culture-tube screw-caps or tissue paper while cleaning equipment, for example).

2.2. Protocol 1: Preparation of samples for glycan release 1.1: Excision of tissues/organs or isolation of cells 1. Mouse tissues and organs analyzed by the CFG were harvested from 6to 8-week old sex-matched C57BL/6 mice obtained from the Scripps Research Institute (La Jolla, CA) custom breeding core and sacrificed by cervical dislocation. The excised samples were snap-frozen immediately after harvesting and stored at 80 C ready for homogenization. Various commercial sources for transgenic mice are available, including The Jackson Laboratory ( JAX), The Mutant Mouse Regional Resource Centre (MMRRC), or the Knock Out Mouse Project (KOMP). 2. Cells, once isolated according to the relevant protocol, should be pelleted having been washed three times with phosphate buffered saline (PBS),

32


especially if they might contain any trace of culture media (fetal calf serum (FCS) especially is a very rich source of contaminating glycoproteins). The cells should be resuspended in 100 mM ammonium bicarbonate in water, boiled for 10 min, and then lyophilized and stored at 80 C ready for cell lysis. 1.2: Homogenization For screening the glycans in mammalian tissues, 100–400 mg of tissue is sufficient for a series of MS analyses including sequential exoglycosidase digestions, MS/MS studies, and linkage analyses. As a rough guide to physical amounts, high-quality mapping data can be obtained from 10% of a single mouse kidney. The use of a dedicated electric homogenizer removes the need for dicing of tissues or organs in most cases, but for larger or more awkward samples, mince the excised tissue into small (1–2 mm3) cubes using a clean scalpel and glass Petri dish, prior to homogenization. Obtain an approximate wet-weight of the sample prior to homogenization. Note: if you intend to extract glycolipids (optional) in addition to N- and O-linked glycans, see Protocol 3.1a. Materials 1. Homogenization buffer (50 ml): 1% CHAPS (v/v) in 25 mM Tris, 150 mM sodium chloride (NaCl), 5 mM EDTA in water, pH 7.4 (adjusted with dilute acetic acid). Place on ice or refrigerate. 2. Dialysis buffer (4.5 l): 50 mM ammonium hydrogen carbonate, pH 7.4 (adjusted with dilute acetic acid). Store at 4 C. 3. Cleaning solution A (50 ml): 80% (v/v) methanol in H2O. 4. Cleaning solution B (50 ml): 33%:33%:33% (v:v:v) methanol, formic acid, H2O 5. Cleaning solution C (50 ml): 50% (v/v) chloroform in methanol 6. Methanol 7. Snakeskin dialysis tubing, cut off point of 7 kDa (Pierce, Prod # 68700). Soak tubing in water prior to usage. 8. Ingenieurbu¨ro CAT X-120 homogenizer with a T6 or 6.1 dispersion shaft (a T17 shaft is also useful for large tissues) 9. Glass culture tubes (13 100 mm, Corning), culture tube caps (Fisher), Teflon inserts (Owens Polyscience), dialysis clips, magnetic stirrer, 15 and 50 ml FalconÒ (Blue MaxTM) polypropylene tubes (BDH). Method 1. Cleaning the homogenizer: a. Immerse the tip of the shaft into cleaning solution A in a 50 ml Falcon tube and operate at low to intermediate settings for 60 s.


33

b. Examine the tip and carefully remove any debris with a fine needle taking care not to scratch the shaft. Repeat this process if required. c. Remove the dispersion shaft assembly, place into Cleaning Solution B in a 50 ml Falcon tube and sonicate for 10 min. d. Reassemble the homogenizer, and repeat step (1) with the following solutions: (a) methanol, (b) Cleaning Solution C, (c) methanol, (d) H2O (twice) e. Operate at a low to intermediate setting in an empty Falcon tube for 1 min. The homogenizer is now ready for use. 2. Place the sample (depending on size) into a fresh 15 or 50 ml Falcon tube, add 5 or 10 ml (respectively) of ice cold homogenization buffer and place the tube on ice. Place the tip of the homogenizer at the bottom of the tube and operate in 10 s bursts (pausing for 10 s in-between) at intermediate to high settings for 2 min, while keeping the sample on ice. Ensure that the tissue is properly homogenized. 3. Transfer the homogenized sample into the presoaked dialysis tubing and dialyse against the dialysis buffer at 4 C for 48 h. Use constant stirring and replace the dialysis buffer at regular intervals (approximately every 12 h). 4. Once dialysis is complete, transfer the sample into labeled glass culture tubes, cover with perforated Parafilm and lyophilize. 1.3: Cell lysis For the screening of cell isolates, 1 106 cells is considered the recommended lower limit to produce MALDI-MS profiling data from N- and O-linked glycoprotein-derived glycans. Larger cell numbers are usually required if further analytical techniques such as MALDI and/or ES-MS/ MS, chemical and enzymatic digestions, and assignment of glycolipid-derived glycan residues is desired. Using smaller sample quantities is possible, but will likely limit the types of analysis possible, leading to lower levels of structural information. The use of multiple independent biological replicates is also advised. Note: if you intend to extract glycolipids (optional) in addition to N- and O-linked glycans, see Protocol 3.1b. Materials 1. Cleaning solution A (50 ml): 80% (v/v) methanol in H2O 2. Cleaning solution B (50 ml): 33%:33%:33% (v:v:v) methanol, formic acid, H2O 3. Cleaning solution C (50 ml): 50% (v/v) chloroform in methanol 4. Methanol 5. Extraction buffer (50 ml): 1% CHAPS (v/v) in 25 mM Tris, 150 mM sodium chloride (NaCl), 5 mM EDTA in water, pH 7.4 (adjusted with dilute acetic acid). Place on ice or refrigerate.

34


6. Snakeskin dialysis tubing, cut off point of 7 kDa (Pierce, Prod # 68700). Soak tubing in water prior to usage. 7. Sonicator (Vibra-Cell-Sonics, model CV188), glass culture tubes (13 100 mm, Corning), culture tube caps (Fisher), Teflon inserts (Owens Polyscience), dialysis clips, magnetic stirrer, 15 ml FalconÒ (Blue MaxTM) polypropylene tubes (BDH). Method 1. Cleaning the sonicator: a. Immerse the head of the sonicator into water in a 50 ml Falcon tube and operate on continuous mode at 20 amps for 10 s. b. Repeat step 1 with Cleaning Solution B c. Immerse the sonicator in water in a 50 ml Falcon tube and sonicate using a sonication bath for 10 min. d. Repeat step (1) with the following solutions: (a) water, (b) methanol, (c) Cleaning Solution C, (d) methanol, (e) H2O (twice) e. Operate continuous mode at 20 amps in an empty Falcon tube for 1 min. The sonicator is now ready for use. 2. Resuspend the lyophilized cell pellet by addition of 2 ml of ice cold extraction buffer in a 15 ml Falcon tube. Place on ice. 3. Immerse the tip of the sonicator into the sample and sonicate for 10 s on continuous mode at 40 amps. Pause for 15 s. Repeat five times. 4. Transfer the lysed sample into the presoaked dialysis tubing and dialyse against the dialysis buffer at 4 C for 48 h. Use constant stirring and replace the dialysis buffer at regular intervals (approximately every 12 h). 5. Once dialysis is complete, transfer the sample into 15 ml Falcon tubes, cover with perforated Parafilm and lyophilize.

2.3. Protocol 2: Preparation of glycans for analysis 2.1: Cleavage of S-S bridges and protection of Cys In order to efficiently release glycans using enzymes, the glycoprotein must first be denatured to allow access to potentially inaccessible glycosylation sites. This is routinely achieved by means of reduction/alkylation of the S-S bridges, followed by proteolytic digestion. This slightly modified protocol adds GuHCl to assist solubilization of the more difficult samples. 2.1: Materials 1. Tris buffer (50 ml): 0.6 M Tris, pH 8.5 (adjusted with dilute acetic acid). Degas by gently bubbling nitrogen gas through the solution for 30 min. 2. GuHCl stock solution (10 ml): 8 M guanidine hydrochloride (GuHCl) in water.


35

3. Tris–GuHCl buffer (5 ml): 0.6 M Tris in 4 M guanidine hydrochloride (GuHCl), pH 8.5 (adjusted with HCl). Add 0.363 g Tris to 2.5 ml of GuHCl Stock Solution. Adjust pH and top up to 5 ml with water. Degas by gently bubbling nitrogen gas through the solution for 30 min. 4. DTT solution (5 ml): 2 mg/ml dithiothreitol (DTT) in Tris–GuHCl Buffer. 5. IAA solution (5 ml): 12 mg/ml iodoacetic acid (IAA) in Tris buffer. 6. Dialysis buffer (4.5 l): 50 mM ammonium hydrogen carbonate, pH 7.4 (adjusted with dilute acetic acid). Store at 4 C. 7. Snakeskin dialysis tubing, cut off point of 7 kDa (Pierce, Prod # 68700). Soak tubing in water prior to usage. 8. Glass culture tubes (13 100 mm, Corning), culture tube caps (Fisher), Teflon inserts (Owens Polyscience), dialysis clips, magnetic stirrer, heating block at 37 C. Method 1. Add 1–2 ml of the DTT Solution to the sample, use the minimum volume depending on solubility (for more difficult or larger samples, split into aliquots). If dealing with a protein pellet, the exact amount will depend upon pellet size—again, for larger samples, split into aliquots. Mix thoroughly, breaking up lumps as necessary, until the vast majority of the sample is solubilized. Cap the tube and incubate at 50 C for 2 h. Briefly centrifuge. 2. Add 1 ml of the IAA solution to the sample tube, cap the tube, and incubate for a further 90 min at room temperature in the dark. 3. The reaction is terminated by dialysis. Transfer the sample into the presoaked dialysis tubing and dialyse against the dialysis buffer at 4 C for 48 h. Use constant stirring and replace the dialysis buffer at regular intervals (approximately every 12 h). 4. Once dialysis is complete, transfer the sample into labeled glass culture tubes (or 15 ml Falcon tubes if volume is an issue), cover with perforated Parafilm and lyophilize. The sample is now ready for proteolytic digestion. 2.2: Proteolytic digestion For most glycomic experiments, trypsin digestion is ideal. However, alternatives such as endoproteinase Glu-C, chymotrypsin, or cyanogen bromide can be substituted. The method below is sufficient for samples of up to around 1 g of intact starting material. For larger samples, more trypsin will be required—typically a 1:50/1:100 ratio of enzyme to protein (w/w) is required to digest the sample. Be aware that large quantities of trypsin do carry a risk of contamination from exogenous glycans.

36


Materials 1. Digestion buffer (50 ml): 50 mM ammonium hydrogen carbonate, pH 8.4 (adjusted with aqueous ammonia). 2. Dilute acetic acid (50 ml): 5% (v/v) acetic acid in H2O 3. Trypsin solution: 1 mg/ml porcine pancreas trypsin in digestion buffer. Weigh out 0.5–1.5 mg of TPCK-treated trypsin and add an appropriate amount of digestion buffer (i.e., 0.5–1.5 ml) to make a 1 mg/ml solution. 4. Heating block at 37 C. Method 1. Add 1 ml of the trypsin solution to the reduced/carboxymethylated sample and incubate at 37 C for 16 h. Ensure the sample is completely suspended in the solution, topping up with digestion buffer if necessary. 2. Terminate the reaction by heating to 100 C for 2 min. 3. Add 1 drop of dilute acetic acid (more may cause precipitation) to neutralize the sample. 4. Proceed directly to glycopeptide purification (i.e., do not lyophilize) 2.3: Glycopeptide purification This purification step is necessary to remove any remaining hydrophilic contaminants from the peptides and putative glycopeptides prior to glycan release. Materials 1. Purification column: OasisÒ HLB Plus cartridge (Waters). 2. Dilute acetic acid (200 ml): 5% (v/v) acetic acid in H2O 3. Elution buffers (4 50 ml): 20%, 40%, and 100% (v:v) propan-1-ol in dilute acetic acid. 4. Methanol 5. Glass culture tubes (13 100 mm, Corning), culture tube caps (Fisher), Teflon inserts (Owens Polyscience), retort stand and clamp, 20 ml glass syringe. Method 1. Assemble the glass syringe with the Oasis purification column attached. Clamp the body of the syringe in a retort stand and position a waste beaker underneath the apparatus. Ensure all solutions are prepared in advance and you have labeled three clean screw-capped culture glass tubes. 2. Condition the Oasis column by eluting sequentially with 5 ml methanol, 5 ml dilute acetic acid, 5 ml propan-1-ol, and 15 ml dilute acetic acid.


37

3. Carefully load the sample dropwise directly onto the column. For most samples from whole tissues, splitting the sample between multiple columns will be necessary. 4. Wash the sample with 20 ml dilute acetic acid. This fraction will only contain hydrophilic contaminants and may be discarded. 5. Elute stepwise with 4 ml of each of the three Elution buffers, collecting into clean, labeled glass culture tubes. 6. Reduce the volume of the fractions in a vacuum centrifuge (SpeedVacÒ) until you can combine the fractions into a single tube. 7. Cover with perforated Parafilm and lyophilize. Keep at 80 C ready for glycan release. 2.4: N-Glycan release Release of N-linked glycans from the peptide backbone is commonly achieved enzymatically by way of PNGase F (Tarentino and Plummer, 1994; Tarentino et al., 1985) or PNGase A (Tretter et al., 1991). PNGase A is used in situations where samples contain a1-3 fucose linked to the reducing end N-acetylglucosamine (GlcNAc). This is a common modification in plants and invertebrates, but is not known to naturally occur in mammalian samples, so barring special circumstances PNGase F digestion is sufficient. Materials 1. Peptide: N-glycosidase F (PNGase F) in glycerol (Roche EC 3.5.1.52). This can be kept for long periods of time in the freezer at 20 C. If lyophilized PNGase F is used, then once it has been dissolved use immediately as the activity of the enzyme may decline. 2. Enzyme buffer (50 ml): 50 mM ammonium hydrogen carbonate, pH 8.4 (adjusted with ammonia). 3. Heating block at 37 oC. Method 1. Dissolve the lyophilized propan-1-ol fraction(s) in 200 ml of enzyme buffer and combine if necessary. 2. Add 3–4 U of PNGase F and incubate for 24–48 h at 37 C, adding a fresh 3–4 U aliquot of PNGase F after 24 h. 3. Cover with perforated Parafilm and lyophilize. Keep at 80 C ready for purification. 2.5: N-Glycan purification This purification step is necessary to separate the released, hydrophilic N-glycans from the remaining peptides and O-glycopeptides.

38


Materials 1. Purification column: C18 Sep-PakÒ Short Body Classic Cartridges (Waters) 2. Dilute acetic acid (200 ml): 5% (v/v) acetic acid in H2O 3. Elution buffers (4 50 ml): 20%, 40%, and 100% (v:v) propan-1-ol in dilute acetic acid. 4. Methanol 5. Glass culture tubes (13 100 mm, Corning), culture tube caps (Fisher), Teflon inserts (Owens Polyscience), retort stand and clamp, 5 ml glass syringe. Method 1. Assemble the glass syringe with the Sep-Pak purification column attached. Clamp the body of the syringe in a retort stand and position a waste beaker underneath the apparatus. Ensure all solutions are prepared in advance and you have labeled four clean screw-capped culture glass tubes. 2. Condition the Sep-Pak column by eluting sequentially with 5 ml methanol, 5 ml dilute acetic acid, 5 ml propan-1-ol, and 15 ml dilute acetic acid. 3. If necessary, resuspend the sample in a small volume of dilute acetic acid, before carefully loading the sample dropwise directly onto the column. 4. Elute stepwise with 5 ml of dilute acetic acid (do not discard, see note below) followed by 4 ml of each of the three elution buffers, collecting into clean, labeled glass culture tubes. 5. Reduce the volume of the fractions in a vacuum centrifuge (SpeedVacÒ) to approximately 1–2 ml. Combine the propan-1-ol fractions into a single tube. 6. Cover each sample with perforated Parafilm and lyophilize. Keep at 80 C in preparation for derivatization (Protocol 4.1) or further chemical/enzymatic degradations (optional, Protocols 3.3 and 3.4). Note: The acetic acid fraction contains the total released N-glycan pool. The combined propan-1-ol fraction contains the remaining O-glycopeptides. 2.6: O-Glycan release An enzyme, O-glycosidase, is available and capable of releasing some O-glycans. However, it has a very restricted specificity so is thus only suitable for specialist use in conjunction with other enzymes. O-Glycan release is therefore typically achieved chemically by way of an alkaline b-elimination. The reaction is carried out under reducing conditions to prevent glycan degradation by the ‘‘peeling’’ reaction by reducing terminal N-acetylgalactosamine (GalNAc) residues to their alditol forms and hence is commonly referred to as reductive elimination.


39

Materials 1. 0.1 M potassium hydroxide (KOH) stock solution (100 ml) 2. 1 M potassium borohydride (KBH4) solution in 0.1 M KOH (0.5–1.5 ml): weigh out 25–50 mg of KBH4 and dissolve in the 0.1 M KOH stock solution to give a concentration of 54–55 mg/ml. 3. Glacial acetic acid 4. Heating block at 45 C. Method 1. Add 400 ml of the KBH4 solution to the combined propan-1-ol fractions, in a glass culture tube. 2. Cap the tube with a Teflon-lined screw top and incubate at 45 C for 20–24 h. Briefly centrifuge. 3. Terminate the reaction by addition of a few (3–5) drops of glacial acetic acid, adding dropwise until the effervescence is observed to cease. 4. The sample is now ready for purification by cation exchange chromatography. 2.7: O-Glycan purification The freshly released O-glycans first need to be desalted before any further experiments are carried out. This is achieved by use of a Dowex cation exchange column. Materials 4 M hydrochloric acid (HCl) (500 ml) Dilute acetic acid (200 ml): 5% (v/v) acetic acid in H2O Methanolic acetic acid: 10% (v/v) acetic acid in methanol (50 ml) Washed Dowex beads (50 W-X8 (H) 50–100 mesh) (100 g): Place 100 g of dry beads in a 250 ml screw cap glass bottle. Add 100 ml of 4 M HCl and decant. Repeat this two more times, then wash the beads by adding, agitating, and decanting 25–30 times with water to remove residual HCl. Wash the beads three times with dilute acetic acid before finally storing them immersed in dilute acetic acid ( 200 ml). The treated beads can be kept equilibrated in this state for many months. 5. Glass culture tubes (13 100 mm, Corning), culture tube caps (Fisher), Teflon inserts (Owens Polyscience), glass Pasteur pipettes, glass wool, silicone tubing (length 10 cm, internal diameter of 1–2 mm), adjustable (screw) clips, retort stand, and clamp. 1. 2. 3. 4.

40


Method 1. Clamp the desalting column (a Pasteur pipette plugged at the tapered end with a small amount of glass wool) in a retort stand with a waste beaker beneath. Insert the tapered end of the Pasteur pipette into a piece of silicone tubing (10 cm length with internal bore diameter of 1–2 mm) and use a screw-adjustable clip to close the tip of the tubing. 2. Fill the column with dilute acetic acid. Open the clip to allow the acid to slowly run out. As the acid level drops, fill the column with washed Dowex beads. Fill the column to the crimp in the Pasteur. 3. Wash the column with 20 ml of dilute acetic acid. Do this slowly, do not allow the level of liquid to drop below the Dowex level. Use the clip to control flow. 4. Place a labeled glass culture tube under the column and load the reductively eliminated sample, allowing it to slowly flow onto the column, closing the clip once the entire sample has moved into the beads. 5. Top up the column with dilute acetic acid before opening the clip and allowing the liquid to slowly pass through the column and into the collection tube. Keep topping up the column with dilute acetic acid until 2.5 ml of eluent has been collected. 6. Replace the collection tube with a fresh labeled tube and repeat step 5. 7. Reduce the volume of the two samples in a vacuum centrifuge (SpeedVacÒ) to approximately 1–2 ml. Cover the samples with perforated Parafilm and lyophilize. 8. Borates introduced during the reductive elimination step now need to be removed by coevaporation. Add 0.5 ml of methanolic acetic acid to the lyophilized sample, before evaporating under a gentle stream of nitrogen gas at room temperature. Repeat this process four times. 9. The first 2.5 ml sample now contains the total released and reduced O-glycan pool. Keep at 80 C in preparation for derivatization (Protocol 4.1) or further chemical/enzymatic degradations (optional, Protocols 3.3 and 3.4). The second 2.5 ml aliquot should not contain much beyond contaminating material, but should be stored at 80 C in case it is needed.

2.4. Protocol 3: Optional sample preparation steps 3.1a. Homogenization with glycolipid extraction See Protocol 1.2 for general notes on tissue preparation. Before proceeding, weigh the tissue and calculate the volume of water represented, using the assumptions that 80% of tissue mass is water and 1 g H2O ¼ 1 ml H2O.


41

Materials Cleaning solutions (see Protocol 1.2) Methanol Chloroform Tris buffer (50 ml): 0.6 M Tris, pH 8.5 (adjusted with dilute acetic acid). Degas by gently bubbling nitrogen gas through the solution for 30 min. 5. Snakeskin dialysis tubing, cut off point of 7 kDa (Pierce, Prod # 68700). Soak tubing in water prior to usage. 6. Ingenieurbu¨ro CAT X-120 homogenizer with a T6 or 6.1 dispersion shaft (a T17 shaft is also useful for large tissues) 7. Glass culture tubes (13 100 mm, Corning), culture tube caps (Fisher), Teflon inserts (Owens Polyscience), dialysis clips, magnetic stirrer, benchtop centrifuge, 15 and 50 ml FalconÒ (Blue MaxTM) polypropylene tubes (BDH).

1. 2. 3. 4.

Method 1. Clean the homogenizer as described in Protocol 1.2 2. Place the sample (depending on size) into a fresh 15 or 50 ml Falcon tube, add 4 volumes of ice-cold water and place the tube on ice. Place the tip of the homogenizer at the bottom of the tube and operate in 10 s bursts (pausing for 10 s in-between) at intermediate to high settings for 2 min, while keeping the sample on ice. Ensure that the tissue is properly homogenized. 3. Calculate the aqueous volume in ml by adding the total volume of water used in step 2 in ml to 80% of the tissue mass in g. Add 2.67 times the aqueous volume of methanol to the sample and mix vigorously at room temperature. 4. Add 1.33 times the aqueous volume of chloroform and mix vigorously at room temperature. 5. Centrifuge at 3000 rpm for 10 min. At this point, the glycolipids will be contained within the supernatant, with the pellet containing the remaining protein extract. 6. Carefully remove the supernatant from the pellet. This fraction is now ready for glycolipid partitioning (Protocol 3.2). 7. Remove excess methanol/chloroform from the pellet by placing under a gentle stream of nitrogen gas for 1–2 min. Do not allow the sample to dry completely. 8. Add 50 ml of Tris buffer to the pellet and replace under the nitrogen stream. Again, do not allow the sample to dry. The protein pellet is now ready for cleavage of S-S bridges and protection of Cys (Protocol 2.1). 3.1b: Cell lysis with glycolipid extraction See Protocol 1.3 for general notes on cell isolate preparation.

42


Materials Cleaning solutions (see Protocol 1.3) Methanol Chloroform Snakeskin dialysis tubing, cut off point of 7 kDa (Pierce, Prod # 68700). Soak tubing in water prior to usage. 5. Sonicator (Vibra-Cell-Sonics, model CV188), glass culture tubes (13 100 mm, Corning), culture tube caps (Fisher), Teflon inserts (Owens Polyscience), dialysis clips, magnetic stirrer, centrifuge, 15 ml FalconÒ (Blue MaxTM) polypropylene tubes (BDH).

1. 2. 3. 4.

Method 1. Cleaning the sonicator: a. Immerse the head of the sonicator into water in a 50 ml Falcon tube and operate on continuous mode at 20 amps for 10 s. b. Repeat step 1 with cleaning solution B c. Immerse the sonicator in water in a 50 ml Falcon tube and sonicate using a sonication bath for 10 min. d. Repeat step (1) with the following solutions: (a) water, (b) methanol, (c) Cleaning Solution C, (d) methanol, (e) H2O (twice) e. Operate continuous mode at 20 amps in an empty Falcon tube for 1 min. The sonicator is now ready for use. 2. Resuspend the lyophilized cell pellet by addition of 2 ml of ice cold water in a 15 ml Falcon tube. Place on ice. 3. Immerse the tip of the sonicator into the sample and sonicate on ice for 10 s on continuous mode at 40 amps. Pause for 15 s. Repeat five times. 4. Add 2.67 sample volumes of methanol, cap, and mix vigorously 5. Add 1.33 sample volumes of chloroform, cap, and mix vigorously 6. Centrifuge at 3000 rpm for 10 min At this point, the glycolipids will be contained within the supernatant, with the pellet containing the remaining protein extract. 7. Carefully remove the supernatant from the pellet. This fraction is now ready for glycolipid partitioning (Protocol 3.2). 8. Remove excess methanol/chloroform from the pellet by placing under a gentle stream of nitrogen gas for 1–2 min. Do not allow the sample to dry completely. 9. Add 50 ml of Tris buffer to the pellet and replace under the nitrogen stream. Again, do not allow the sample to dry. 10. The protein pellet is now ready for cleavage of S-S bridges and protection of Cys (Protocol 2.1).


43

3.2: Glycolipid preparation The supernatant collected from the homogenization with glycolipid extraction procedure (Protocol 3.1) contains the extracted glycolipid pool from the sample, which can now be further partitioned into polar and nonpolar glycolipids. These pools are then purified, the oligosaccharide portions are released, and the glycolipid-derived glycans purified. Materials 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15.

Methanol Chloroform Acetonitrile (MeCN) Butan-1-ol Acetonitrile/TFA solution 1(50 ml): 80% acetonitrile in 0.1% (v/v) trifluoroacetic acid (TFA) in H2O Acetonitrile/TFA solution 2 (50 ml): 25% acetonitrile in 0.05% (v/v) TFA in H2O Methanol/water solution (50 ml): 50% (v/v) methanol in H2O Methanol/chloroform solution (20 ml): 50% (v/v) methanol in chloroform Dilute acetic acid (50 ml): 5% (v/v) acetic acid in H2O Purification column 1: tC18 (Plus) Sep-PakÒ Cartridge (Waters) Purification column 2: C18 Sep-PakÒ Short Body Classic Cartridges (Waters) Purification column 3: Hypercarb PGC column (Hypersep SPE column, 6 ml column volume, 1 g bed weight, Thermo) Digestion buffer (50 ml): Prepare 50 ml of 50 mM sodium acetate in water, pH 5.5 (adjusted with acetic acid). Add 100 mg sodium cholate to produce a 0.2% w/v solution. Ceramide glycanase, rEGCase II (Takara) Glass culture tubes (13 100 mm, Corning), culture tube caps (Fisher), Teflon inserts (Owens Polyscience), 10 ml glass syringe, benchtop centrifuge, retort stand, and clamp.

3.2a: Glycolipid partitioning 1. Measure the volume of the supernatant collected from the homogenization with glycolipid extraction procedure (Protocol 3.1). 2. Add 0.173 volumes of water and mix thoroughly. 3. Centrifuge for 15 min at 3000 rpm. 4. Separate the upper layer (polar glycolipids) and lower layer (nonpolar glycolipids) and place into a clean, dry glass culture tubes.

44


3.2b: Polar glycolipid recovery 1. Assemble the glass syringe with the tC18 (Plus) Sep-Pak purification column attached. Clamp the body of the syringe in a retort stand and position a waste beaker underneath the apparatus. Ensure all solutions are prepared in advance and you have labeled two clean screw-capped culture glass tubes. 2. Condition the Sep-Pak column by eluting sequentially with 5 ml methanol, 5 ml methanol/water solution, 5 ml methanol/chloroform solution, and 15 ml methanol/water solution. 3. Carefully load the sample onto the column, using the glass syringe as a reservoir. 4. Wash the column with 15 ml of methanol/water solution. 5. Elute stepwise with 5 ml of methanol followed by 5 ml of methanol/ chloroform solution, collecting into clean, labeled glass culture tubes. 6. Reduce the volumes of both fractions under a gentle stream of nitrogen gas, then combine. Replace under the nitrogen flow and evaporate to dryness. The polar glycolipids are now ready for oligosaccharide release. 3.2c: Nonpolar glycolipid recovery 1. 2. 3. 4.

Add 15 ml chloroform to the lower layer (nonpolar glycolipid) sample. Add 15 ml water and mix vigorously. Centrifuge for 15 min at 3000 rpm. Discard upper aqueous layer, transfer the lower chloroform layer to a fresh tube and repeat. Reduce the volume of the remaining chloroform layer under a gentle stream of nitrogen gas and before the sample dries completely, transfer to a clean glass culture tube before evaporating any remaining solvent. The nonpolar glycolipids are now ready for oligosaccharide release.

3.2d: Glycolipid oligosaccharide release 1. Resuspend the sample (either the polar glycolipids or the nonpolar glycolipids) in a minimum volume (0.2–1 ml) of digestion buffer. 2. Add 25 mU of ceramide glycanase and incubate at 37 C for 24 h. 3. Add a further 25 mU of enzyme and incubate at 37 C for a further 24 h. 3.2e: Glycolipid oligosaccharide purification 1. Add water to the ceramide digest sample to bring the volume up to 2 ml. 2. Add 2 ml of butan-1-ol and mix vigorously. Discard the upper (butan-1-ol) layer. 3. Repeat steps 1–2 twice more, before removing any lingering butan-1-ol by evaporation under a gentle flow of nitrogen gas for 20 min.


45

4. Assemble the glass syringe with the C18 Sep-PakÒ purification column attached. Clamp the body of the syringe in a retort stand and position a waste beaker underneath the apparatus. Ensure all solutions are prepared in advance and you have labeled three clean screw-capped culture glass tube. 5. Condition the Sep-Pak column by eluting sequentially with 5 ml methanol and 10 ml dilute acetic acid. 6. Carefully load the sample dropwise directly onto the column. 7. Elute with 5 ml of dilute acetic acid, collecting into a clean, labeled glass culture tube. 8. Further purify eluted glycans by Hypercarb column chromatography. Use a 5 ml syringe fitted to push the solvent through the column. 9. Condition the Hypercarb column by eluting sequentially with 3 columns of Acetonitrile/TFA solution 1, and 3 columns of water. 10. Carefully load the sample (from step 7) 11. Wash with 3 columns of water. 12. Elute with 2 columns of Acetonitrile/TFA solution 2. Collect in clean glass culture tube. 13. Reduce the volume of the fraction in a vacuum centrifuge (SpeedVacÒ) to approximately 1–2 ml. 14. Cover each sample with perforated Parafilm and lyophilize. Keep at 80 C in preparation for derivatization (Protocol 4.1) or further chemical/enzymatic degradations (optional, Protocols 3.3 and 3.4). 3.3: Chemical digestion of released glycans The treatment of released, purified glycan pools with chemical or enzymatic (Protocol 3.4) degradation processes is a powerful tool and can assist MS screening and MS/MS sequencing of glycans by providing information on structural features, linkages, monosaccharide identification, and stereochemistry. These experiments are most usually carried out following an initial analysis of the sample by MS or MS/MS, in order that the degradation (s) employed may be tailored to answer specific structural ambiguities. 3.3a: Acid hydrolysis Acid hydrolysis using HF is a popular chemical treatment employed in the analysis of carbohydrates. The process has been shown to release phosphorylcholine (PC) from N-glycans (Haslam et al., 1997) and is capable of hydrolyzing phosphodiester bonds in glycosylphosphatidilinositol(GPI)anchored proteins (Schneider and Ferguson, 1995). Most commonly, however, it is utilized to selectively cleave fucose residues. HF treatment removes a1-3-linked fucose residues rapidly, while a1-2,4 and 6-linked fucoses are released at a much slower rate.

46


Materials 1. 48% HF solution 2. Dilute acetic acid (50 ml): 5% (v/v) acetic acid in H2O 3. Glass culture tubes (13 100 mm, Corning), culture tube caps (Fisher), Teflon inserts (Owens Polyscience), 1.5 ml Eppendorfs. Carry out all reactions in plastic, since HF will dissolve glass. Handle with care. Method 1. Add 50 ml of HF solution to the purified, released sample in a plastic Eppendorf and incubate at 4 C for 20 h. 2. Terminate the reaction by drying under a gentle stream of nitrogen gas. 3. Redissolve the sample in a minimum volume of dilute acetic acid and transfer to a fresh glass culture tube. Cover the sample with perforated Parafilm and lyophilize. 4. The sample is now ready for further degradations via enzymatic treatment or derivatization by permethylation (see Protocol 3.4 (optional) or Protocol 4.1) 3.3b: Mild periodate oxidation Periodate oxidation has long been a widely used tool in carbohydrate chemistry (Angel and Nilsson, 1990; Bobbitt, 1956). It is particularly useful for O-glycan structural determination. Under mild conditions, cleavage occurs specifically between the C4 and C5 carbons of the core N-acetylgalactosaminitol. This allows the O-glycan core type to be determined (Stoll et al., 1990). Materials 1. Oxidation buffer (50 ml): 100 mM ammonium acetate (pH 6.5, adjusted with acetic acid) 2. Oxidation solution (1 ml): 20 mM sodium m-periodate in oxidation buffer 3. Ethylene glycol 4. Reduction buffer (50 ml): 2 M ammonium hydroxide 5. Reduction solution (5 ml): 10 mg/ml sodium borohydride (NaBH4) in reduction buffer 6. Acetic acid Method 1. Add 100 ml of oxidation solution to the released O-glycan sample and incubate at 0 C for 20 h in the dark 2. Terminate the reaction with the addition of 2–3 ml of ethylene glycol


47

3. Stand at room temperature for 1 h, then cover the sample with perforated Parafilm and lyophilize. 4. Add 400 ml of reduction solution and incubate at room temperature for 2 h 5. Terminate the reaction by addition of a few (3–5) drops of glacial acetic acid, adding dropwise until the effervescence is observed to cease. 6. The sample is now ready for purification by cation exchange chromatography (see Protocol 2.7: O-glycan purification) 3.4: Enzymatic digestion of released glycans Enzymatic digestions of glycan pools are relatively simple experiments that are capable of yielding large amounts of structural information. Most commonly, the enzymes utilized fall into one of three categories—glycosyltransferases (which add sugar residues to nonreduced positions), exoglycosidases (which cleave nonreducing end terminal sugar residues), and endoglycosidases (which cleave internal glycosidic bonds). These enzymes typically have extremely specific substrates, only cleaving or extending glycans with the correct residue, linkage, and anomeric stereochemistry. Materials 1. a-Mannosidase Cleaves all a(1-2,3,6)-linked mannose residues (Lundblad et al., 1976) Digestion buffer (50 ml): 50 mM ammonium acetate, pH 4.5 (adjusted with dilute acetic acid) Enzyme solution: 0.5 U of a-mannosidase ( Jack Bean, EC 3.2.1.24 (Sigma)) in 100 ml of digestion buffer. 2. a-Galactosidase Cleaves all a-linked, nonreducing terminal galactose residues (Kobata, 1979). Digestion buffer (50 ml): 50 mM ammonium formate, pH 6.0 (adjusted with HCl) Enzyme solution: 0.5 U of a-galactosidase (green coffee beans, EC 3.2.1.22 (Sigma)) in 100 ml of digestion buffer. 3. b-Galactosidase Cleaves all b-linked, nonreducing terminal galactose residues. Fucose linked to the penultimate N-acetylglucosamine will block cleavage of the galactose (Miyatake and Suzuki, 1975). Digestion buffer (50 ml): 50 mM ammonium acetate, pH 4.6 (adjusted with dilute acetic acid) Enzyme solution: 10 mU of b-galactosidase (bovine testes, EC 3.2.1.23 (Sigma)) in 100 ml of digestion buffer.

48


4. b-N-Acetylhexosaminidase Cleaves terminal b-N-acetylglucosamine and b-N-acetylgalactosamine residues (Kobata, 1979) Digestion buffer (50 ml): 50 mM ammonium formate, pH 5.0 (adjusted with HCl) Enzyme solution: b-N-acetylhexosaminidase (bovine kidney, EC 3.2.1.52 (Sigma)) in 100 ml of digestion buffer. 5. Endo-b-galactosidase Cleaves internal b(1-4) galactose linkages in unbranched, repeating poly-N-acetyllactosamine [GlcNAcb(1-3)Galb(1-4)]n structures (Scudder et al., 1983) Digestion buffer (50 ml): 50 mM ammonium acetate, pH 5.5 (adjusted with dilute acetic acid) Enzyme solution: 5 mU of endo-b-galactosidase (Bacteroides fragilis, EC 3.2.1.103 (Sigma)) in 100 ml of digestion buffer. 6. a(2-3,6,8) Neuraminidase Cleaves all nonreducing terminal sialic acid residues (Corfield et al., 1983). Digestion buffer (50 ml): 50 mM ammonium acetate, pH 5.5 (adjusted with dilute acetic acid) Enzyme solution: 20 U of a(2-3,6,8) neuraminidase (Vibrio cholerae, EC 3.2.1.18 (Sigma)) in 100 ml of digestion buffer. 7. a(2-3) Neuraminidase Cleaves exclusively the nonreducing terminal a(2-3) unbranched sialic acid residues (Corfield et al., 1983) Digestion buffer (50 ml): 50 mM ammonium acetate, pH 5.5 (adjusted with dilute acetic acid) Enzyme solution: 0.025 U of a2-3-neuraminidase (Streptococcus pneumoniae, EC 3.2.1.18 (Sigma)) in 100 ml of digestion buffer. 8. Sialidase A Cleaves all nonreducing terminal branched and unbranched sialic acids, including the Sda epitope (GaINAcbl-4[NeuAca2-3]Galbl4GlcNAc-R) (Uchida et al., 1977) Digestion buffer: 50 mM sodium acetate, pH 5.5 (adjusted with dilute acetic acid) Enzyme solution: 170 mU of sialidase A (recombinant from Arthrobacter ureafacien, expressed in E. Coli, EC 3.2.1.18 (Glyko)) in 100 ml of digestion buffer. Method 1. Add 100 ml of the chosen enzyme solution (see above) to the released, purified glycan sample and incubate at 37 C for 48 h. 2. Add additional 100 ml aliquots of freshly prepared enzyme solution every 12 h.


49

3. Cover the sample with perforated Parafilm and lyophilize. Keep at 80 C in preparation for derivatization (Protocol 4.1) or further chemical/ enzymatic degradations.

2.5. Protocol 4: Derivatization and analysis of released samples 4.1: Permethylation of released glycan pools Analysis of native glycans is possible, but since they do not ionize as well as other biomolecules (such as peptides (Harvey, 1999)), it is desirable to derivatize prior to analysis. This bestows a compensatory increase in sensitivity, as well as a hydrophobicity increase which is advantageous when trying to remove salt contaminants. Derivatization methods can be broadly categorized into those that employ reducing end tagging and those that serve to protect most or all of the functional groups. Tagging facilitates chromatographic separation and enhances reducing-end fragmentation, with current common tagging reagents including aminopyridine (Kuraya and Hase, 1996), 2-aminobenzamide (Chen and Flynn, 2007), and 1-phenyl-3-methyl pyrazolone (Mason et al., 2006). Protection of the functional groups by permethylation is generally preferable, however. In addition to greatly enhancing the sensitivity of detection with the smallest increase in molecular weight, permethyl derivatives fragment very selectively, producing a limited number of easily interpretable structurally diagnostic ions and allowing easy discrimination between single and multiple cleavage events (Dell, 1990; Dell et al., 1994; Jang-Lee et al., 2006). Materials 1. Methyl iodide 2. Anhydrous dimethyl sulfoxide (10 ml): careful use of a septum-sealed bottle or ampoules of anhydrous dimethyl sulfoxide (DMSO) such as HiDry Anhydrous DMSO solvent (Romil) removes the need for calcium hydride treatment (Dell, 1990) 3. Sodium hydroxide pellets 4. Chloroform 5. Glass pestle and mortar, glass Pasteur pipettes. Ensure all equipment is completely dry before use. Method 1. Place approximately 3–5 sodium hydroxide pellets into the mortar and add 3 ml of DMSO

50


2. Crush the pellets into a slurry, adding more DMSO if necessary. This step needs to be carried out quickly in order to avoid excessive moisture absorption. 3. Using a glass Pasteur pipette, add a small amount (0.5–1 ml) of the slurry directly to the purified glycan sample 4. Add 0.5 ml of methyl iodide, cap the tube, and mix vigorously (use an automatic shaker if available) for 10 min at room temperature. 5. Quench the reaction by slow, dropwise addition of water with constant agitation, until the effervescence is observed to cease. 6. Add 2 ml of chloroform 7. Top up with water to give a total volume of 5 ml and mix vigorously. 8. Allow the mixture to separate into two layers (briefly centrifuge if necessary) 9. Discard upper aqueous layer, and repeat steps 7 and 8 four more times 10. Dry the remaining chloroform layer under a gentle stream of nitrogen gas. The permethylated glycans are now ready for permethylated glycan purification. 4.2: Permethylated glycan purification The newly hydrophobic permethyl derivatives can now be separated from any aqueous contaminants (such as salts, a common culprit of poor MS) by way of a chromatographic separation. Materials 1. Purification column: C18 Sep-PakÒ Short Body Classic Cartridges (Waters) 2. Elution buffers (4 50 ml): 15%, 35%, 50%, and 75% (v:v) acetonitrile in water. 3. Acetonitrile 4. Methanol 5. Methanol/water solution (10 ml): 50% (v/v) methanol in H2O 6. Glass culture tubes (13 100 mm, Corning), culture tube caps (Fisher), Teflon inserts (Owens Polyscience), retort stand and clamp, 5 ml glass syringe. Method 1. Assemble the glass syringe with the C18 Sep-Pak purification column attached. Clamp the body of the syringe in a retort stand and position a waste beaker underneath the apparatus. Ensure all solutions are prepared in advance and you have labeled four clean screw-capped culture glass tubes. 2. Condition the Sep-Pak column by eluting sequentially with 5 ml methanol, 5 ml water, 5 ml acetonitrile, and 15 ml water.


51

3. Redissolve the sample in 200 ml of methanol/water and carefully load the sample dropwise directly onto the column. 4. Elute stepwise with 5 ml of water (discard) followed by 3 ml of each of the four elution buffers (15%, 35%, 50%, and 75%), collecting into clean, labeled glass culture tubes. 5. Reduce the volume of the fractions in a vacuum centrifuge (SpeedVacÒ) to approximately 0.5 ml. 6. Cover each sample with perforated Parafilm and lyophilize. The samples are now ready for MS analysis. Note: The fraction that contains the largest proportion of permethylated glycans depends upon the glycan pool. Smaller, less hydrophobic O-glycans tend to elute mostly in the 35% fraction, while the larger N-glycans tend to be observed eluting in the 50% fraction of their respective separations. Keep all the fractions, storing in screw-capped tubes at room temperature. Mass spectrometric analysis of derivatized glycan pools Early glycomic investigations utilized fast atom bombardment (FAB) mass spectrometry (Dell and Ballou, 1983; Fukuda et al., 1984b) for the analysis of glycan mixtures. This has since been superseded by matrix-assisted laser desorption-ionization time-of-flight mass spectrometry (MALDI-TOF MS) on both single and double TOF instruments for mass-mapping of samples, offering increases in throughput and sensitivity in the analysis of derivatized isolated pools of glycans (Yu et al., 2006). Interpretation of MALDI spectra produced from the methods described here are based upon prior knowledge of glycan biosynthesis, with specific masses usually coinciding with unique glycan compositions. These MALDI ‘‘fingerprints’’ or mass maps are used to inform prior to further experimentation designed to clarify antennal sequences or branching patterns. This may take the form of enzymatic or chemical treatments (Protocols 3.3 and 3.4) or by application of tandem mass spectrometry (MS/MS) to the fragmentation of individual molecular ions in the sample. A diverse array of such instrumentation is available from all of the major MS manufacturers. Though a detailed discussion of their various merits and limitations is outside the scope of this chapter (Dell et al., 2007, 2008; North et al., 2009; Zaia, 2008), alternatives for the tandem and multistage (MSn) mass spectrometric analysis of glycans include the older triple-quadrupole ES instrumentation (Dell et al., 1994; Fenn et al., 1989; Yost and Enke, 1978), ion traps (especially the more modern linear ion traps (Wang et al., 2006)), and electron capture dissociation (ECD) and electron transfer dissociation (ETD), two methods which do not solely use the more common collisionally activated dissociation (CAD) method to induce fragmentation, instead using additional, alternate mechanisms (Perdivara et al., 2009; Sihlbom et al., 2009). The techniques employed at present in our laboratory are matrix-assisted laser

52


desorption-ionization, tandem time-of-flight mass spectrometry (MALDITOF-TOF MS) and ES tandem mass spectrometry (ES-MS/MS), both of which are common instrumentation in biopolymer MS facilities. The fragmentation pathways established for permethyl derivatives of carbohydrates using FAB-MS (Dell, 1987) are preserved in all CAD-based tandem mass spectrometric techniques. The major ions produced arise from cleavage of the positively charged parent ions on the nonreducing side of the glycosidic bond to form an oxonium ion (A-type cleavages, producing B ions), glycosidic bond cleavages on either side of the glycosidic oxygen (bcleavages, producing Y ions), and cross-ring cleavages producing A-ions, where the charge may reside on either the reducing or nonreducing end of the molecule, depending upon the nature of the sample. The mechanisms for the generation of these daughter ions (as well as other products) are well documented elsewhere (Dell et al., 1994, 2007; Mechref and Novotny, 2002), and standardized nomenclature for the assignment of glycan fragments in MS follows the system suggested in 1988 (Domon and Costello, 1988). 4.3: MALDI-TOF MS Prior to analysis by MALDI-TOF MS, the sample is cocrystallized with a large excess of a matrix material, a low molecular weight material which absorbs strongly at the wavelength of the incident laser, and the mixture is deposited on a metal plate. Irradiation of this mixture by the laser induces the accretion of a large amount of energy in the condensed phase through electronic excitation of the matrix molecules. This causes desorption of analyte and matrix ions from the surface of the crystal. MALDI is regarded as a ‘‘soft’’ ionization method, producing mainly singly charged ions of the form [MþH]þ or related salt adducts such as [MþNa]þ or [MþK]þ, with little fragmentation. This makes it an ideal technique for the initial ‘‘fingerprinting’’ or mass-profiling of glycan mixtures. Materials 1. Matrix for permethylated glycans (100–500 ml): 20 mg/ml 2,5-dihydroxybenzoic acid (DHB) in 5:5 (v/v) methanol/water. Make fresh and keep in the dark. For glycolipid-derived samples, use a 10 mg/ml solution. 2. Matrix for peptide calibrants (100–500 ml): 10 mg/ml a-cyano-4-hydroxycinnamic acid in 50% acetonitrile in 0.1% (v/v) TFA in water. Make fresh and keep in the dark. 3. External peptide calibrant solution: 100 mg/ml of peptide/protein mix (leucine enkephalin, bradykinin (fragment-8), angiotensin I, adrenocorticotropic hormone fragment (ACTH) 1–17, ACTH fragment 18–39,


53

ACTH fragment 7–38, ACTH fragment 1–39 and insulin (bovine pancreas)) in 0.1% (v/v) TFA in water. 4. Methanol 5. ABI Voyager-DE STR MALDI-TOF (or similar), 100 spot MALDI plate, 0.5 ml Eppendorf tubes. Method 1. Thoroughly clean the MALDI plate with methanol before a fresh use and ensure it is completely dry. 2. Dissolve the dry, permethylated glycan sample in 10 ml of methanol 3. Take a 1 ml aliquot of the sample and mix with 1 ml of matrix solution in a fresh tube 4. Take a 2 ml aliquot of the external calibrant solution and mix with 2 ml of matrix solution in a fresh tube 5. Spot a 1 ml aliquot of each sample to be analyzed (plus calibrants) onto individual spots on the MALDI plate, and dry under vacuum for 20 min. Once completely dry, the plate is ready to be loaded into the MALDI mass spectrometer. 6. Before sample analysis, perform a calibration. Use a laser intensity of 2200 and record 50 shots/spectrum. Accumulate 3–4 spectra and calibrate. 7. Typical MALDI parameters for permethylated glycan analysis Mode: Positive reflectron, with delayed extraction Mass range: m/z 500–5000 Low mass gate: m/z 500 Accelerating volts: 20 kV Grid: 60–78% Delay time: 220–280 ns 4.4: MALDI-TOF-TOF tandem mass spectrometry MALDI-TOF-TOF MS is functionally identical to MALDI-TOF MS, as described in Protocol 4.3 for the ABI Voyager, with an additional TOF analyser arranged in tandem for the analysis of CAD fragments. However, this state of the art instrumentation offers increased performance in terms of upper mass range, sensitivity, and signal-to-noise ratios in addition to the ability to generate and analyze glycan fragments (Babu et al., 2009; North et al., 2010b; Pang et al., 2009; Stone et al., 2009). These instruments are capable of producing CAD fragments at both low and high energies, while preserving the resolution and sensitivity of the single-TOF instrumentation (Vestal and Campbell, 2005).

54


Materials 1. Matrix for permethylated glycans (100–500 ml): 20 mg/ml 2,5-DHB acid in 5:5 (v/v) methanol/water. Make fresh and keep in the dark. For glycolipid-derived samples, use a 10 mg/ml solution. 2. Matrix for peptide calibrants (100–500 ml): 10 mg/ml a-cyano-4-hydroxycinnamic acid in 50% acetonitrile in 0.1% (v/v) TFA in water. Make fresh and keep in the dark. 3. External peptide calibrant solution (MS mode): 100 mg/ml of peptide/ protein mix (Calmix—leucine enkephalin, bradykinin (fragment-8), angiotensin I, ACTH fragment 1–17, ACTH fragment 18–39, ACTH fragment 7–38, ACTH fragment 1–39 and insulin (bovine pancreas)) in 0.1% (v/v) TFA in water. 4. External peptide calibrant solution (MS/MS mode) (1 ml): 1 pmol/ml solution of [Glu1]-fibrinopeptide B in 1:3 (v/v) acetonitrile/5% acetic acid (v/v) in water 5. Methanol 6. ABI 4800 MALDI TOF/TOFTM (or similar), MALDI plate, 0.5 ml Eppendorf tubes. Method 1. Preparation of the MALDI plate, samples, and spotting are as described in Protocol 4.3: MALDI-TOF MS, remembering to additionally spot 1 ml of MS/MS mode calibration solution mixed with an equal volume of peptide matrix. 2. Before MS sample analysis, perform a calibration upon the MS peptide calibrant spot. Use a laser intensity of 3500 and record 50 shots/spectrum. Accumulate 10 spectra and calibrate. 3. Before MS/MS sample analysis, perform an MS/MS mode calibration upon the MS/MS peptide calibrant spot, using the above settings. 4. Typical MALDI TOF/TOF parameters for permethylated glycan sequencing Mode: Positive reflectron, with metastable ion suppression Laser intensity: 5000–6000 Shots: 1000 shots/spectrum, accumulate 10 spectra Mass range: m/z 50–5000 Accelerating volts: 20 kV Source-collision cell potential difference: 1 kV CID: On CID gas type: Inlet 1—Medium weight Collision gas: Argon Collision gas pressure: 10 4 mbar


55

4.5: Electrospray tandem mass spectrometry (ES-MS/MS) The ES is produced by applying a large electrical field, under atmospheric pressure, to a liquid passing through a conductively coated capillary tube, producing an ‘‘electrospray’’ of charged droplets. This aerosol eventually gives rise to gaseous ions, whose charge distribution is proportional to the number of ionizable groups in the molecule. ES-MS/MS instrumentation is very helpful to have access to, since multiple different types of mass spectrometer generate useful complimentary data. The widely available Q-TOF style ES-MS instrumentation (Morris et al., 1996) is ideal, employing relatively low-energy collisional activation within the collision chamber, thus producing product ions resulting mainly from cleavage at labile glycosidic bonds without the complexity of additional cross-ring fragmentation (Kui Wong et al., 2003; Moody et al., 2003; North et al., 2006). Materials 1. Quadrupole orthogonal acceleration time of flight (Q-TOF) mass spectrometer (Micromass, UK) fitted with a Z-spray API source (or similar) 2. NanoES spray capillaries (Protana) 3. Methanol 4. External peptide calibration solution (1 ml): 1 pmol/ml solution of [Glu1]-fibrinopeptide B in 1:3 (v/v) acetonitrile/5% acetic acid (v/v) in water Method 1. Calibrate the mass spectrometer with 3 ml of the external peptide calibration solution. 2. Dissolve the dry, permethylated glycan sample in 10 ml of methanol 3. Load 2 ml of the sample into a NanoES capillary, break the tip, and position at the source for analysis 4. Typical ES-MS/MS settings for fragmentation of permethylated glycans Mode: Positive ion mode Drying gas: Nitrogen Collision gas: Argon Collision gas pressure: 10 4 mbar Collision gas energy: 30–80 eV Potential at nanoflow tip: 1.5 kV Desired nanoflow rate: 10–30 nL/min

56


4.6: Preparation of partially methylated alditol acetates Prior to linkage analysis by GC–MS, the permethylated glycans must first be converted into partially methylated monosaccharides which are then converted into their alditol forms (Albersheim et al., 1967). The methyl groups of the permethylated sample act as permanent labels for hydroxyl groups not involved in ring formation or glycosidic bonding. The partially methylated alditols are then O-acetylated, labeling the free hydroxyl groups thus allowing identification of former linkage sites. Materials 1. Hydrolysis solution (1 ml): 2 M TFA 2. Reduction solution (1 ml): 10 mg/ml sodium borodeuteride (NaBD4) in 2 M ammonium solution. 3. Methanolic acetic acid (25 ml): 10% (v:v) acetic acid in methanol 4. Glacial acetic acid 5. Acetic anhydride 6. Chloroform Method 1. Incubate the dry permethylated glycan sample in 200 ml of hydrolysis solution at 121 C for 2 h. This produces partially methylated monosaccharides. Dry under a gentle stream of nitrogen gas. 2. Incubate the dry sample in 200 ml of reduction solution at room temperature for 2 h. 3. Terminate the reaction by addition of a few (3–5) drops of glacial acetic acid, adding dropwise until the effervescence is observed to cease. Dry under a gentle stream of nitrogen gas (samples do not need to be completely dry) 4. Borates introduced during the reduction now need to be removed by coevaporation. Add 1 ml of methanolic acetic acid to the sample, before evaporating under a gentle stream of nitrogen gas at room temperature. Repeat this process four times. 5. Incubate the dry sample in 200 ml of acetic anhydride at 100 C for 1 h in order to acetylate, then evaporate under a gentle stream of nitrogen gas 6. Add 2 ml of chloroform to the dry sample 7. Top up with water to give a total volume of 5 ml and mix vigorously. 8. Allow the mixture to separate into two layers (briefly centrifuge if necessary) 9. Discard upper aqueous layer, and repeat steps 7 and 8 four more times


57

10. Dry the remaining chloroform layer under a gentle stream of nitrogen gas. The partially methylated alditol acetates (PMAA) are now ready for linkage analysis by GC–MS. 4.7: Gas-chromatography–mass spectrometry (GC–MS) Analysis of the PMAA derivatives by GC–MS facilitates the identification of monosaccharide constituents and glycosidic bond positions, generally termed ‘‘linkage analysis.’’ It should be noted that the preparation of the PMAA derivatives (Protocol 4.6) does irreversibly disrupt the glycosidic bonds of the sample, in addition to destroying any sialic acids present due to the nature of the conditions. This should therefore be an experiment to be carried out once MS and MS/MS analyses of the permethylated glycans has been completed, unless there is sufficient sample to take an aliquot. Identification of monosaccharides is achieved by comparison of GC retention time and electron impact mass spectrometry (EI-MS) spectra with standards, while relative quantitation can be carried out by comparison of ion chromatogram peak areas. Particularly powerful information can be obtained by comparing GC–MS data before and after enzymatic/chemical degradations (Dell et al., 2007; Haslam et al., 1997; Kui Wong et al., 2003; North et al., 2010b). An excellent repository of standardized data for reference purposes can be found at the Complex Carbohydrate Research Centre (CCRC) Website—http://www.ccrc.uga.edu/ specdb/ms/pmaa.html. Materials 1. Perkin Elmer Clarus 500 Gas Chromatograph–Mass Spectrometer (GC–MS) or similar 2. RTX-5MS column (30 m 0.25 mm internal diameter, 5% diphenyl/ 95% dimethyl polysiloxane stationary phase, Restek Corp.) 3. Hexanes (Sigma Aldrich). Method 1. Dissolve the partially methylated alditol acetate (PMAA) samples in a small volume (20 ml) of hexanes 2. Inject 1 ml of dissolved sample onto the column at 60 C. Use a linear gradient, increasing to 300 C over 30 min at a rate of 8 C/min.

2.6. Protocol 5: Glycobioinformatics The computational challenges presented by glycobiological analysis and the approaches being taken to tackle them are addressed in a number of excellent recent articles (Aoki-Kinoshita, 2008; von der Lieth et al., 2006;

58


York et al., 2010). It should be emphasized that all of the tools mentioned here are only ever used to supplement or assist a researcher’s analyses, never to replace them entirely. Glycobiological analysis is too complex and variable a subject to rely on a fully automated approach, at least at present. Though it is not really possible to satisfactorily cover the emerging field of glycobioinformatics methods in a short section here, we feel it will be useful for researchers to have a short summary of a few of the tools available at present—where to find them, how best to utilize them, and where to look for support. This is not intended to be a complete or exhaustive list in any way, indeed no two laboratories use the same tools or algorithms, instead tending to favor specific or bespoke programs produced in response to the data processing requirements of their environment. Instead this is more to illustrate which utilities we use to compliment our experimental methodology. 2.6.1. The PARC mass spectrometry viewer (PMSV) Use: This is a very useful and widely applicable tool, specifically designed to enable the interactive viewing of ACSII format MS data. A major issue when trying to work with raw data in MS (like other fields reliant on specialist equipment) is the proprietary and/or poorly documented data formats of the output. This restricts the utility of the data to those who have access to the original analysis software. However, it is usually possible to export such data in ASCII format or similar, and with the use of software designed to display such information, enables any researcher to make use of the raw data files. It is also capable of displaying annotations by use of companion files via Cartoonist (below). Source: http://bio.parc.com/mass_spec (free download) Compatibility: Solaris, Mac OS X, Windows Support: Usage instructions and contact information available from the above site 2.6.2. Cartoonist Use: Cartoonist is an algorithm for the analysis and annotation of MS data from N-linked and O-linked glycan samples, specifically MALDI mass profiles of mammalian samples. It is designed to mimic the approach of a human expert in the annotation of MS data and is used in tandem with the PMSV (above) to allow the nonexpert user to utilize the data presented by the various repositories in a more meaningful way. At present, Cartoonist is available as a free web-based tool, though it is still evolving and may become available as a standalone download in the future. Source: http://bio.parc.com/cartoonist/newrun Compatibility: Solaris, Mac OS X, Windows


59

Support: Usage instructions and contact information available from the above site (Goldberg et al., 2005, 2009; Jang-Lee et al., 2006) 2.6.3. GlycoWorkBench Use: GlycoWorkBench is a suite of software tools designed for rapid drawing of glycan structures and for assisting the process of structure determination from MS data. The graphical interface of GlycoWorkBench provides an environment in which structure models can be rapidly assembled, their mass computed, their fragments automatically matched with MSn data and the results compared to assess the best candidate. GlycoWorkBench can greatly reduce the time needed for the interpretation and annotation of mass spectra of glycans. Source: http://www.glycoworkbench.org/attachment/wiki/WikiStart/Gly coWorkbench.1.1.3480.zip Compatibility: Linux, Mac OS X, Windows Support: http://www.glycoworkbench.org/raw-attachment/wiki/Wiki Start/GlycoWorkbench_short_manual.1.2.4105.pdf (Ceroni et al., 2008)

3. Interpretation of Glycomic Data The MALDI analyses of the permethylated N- and O-linked glycan pools from specific defined systems provide highly sensitive mass profiles. The assignments made within these profiles should be viewed primarily as unequivocal monosaccharide unit compositions for each peak. Knowledge of the specific biosynthetic pathways utilized by the system being analyzed is then used to provide the most informed suggestion for the structural identity of the glycoprotein-derived N-linked and O-linked glycans. For example, since N- and O-glycans are constructed from common pathways, the observation of a specific structural motif leads to the expectation of other related intermediate structures along the same pathway. This can also be indicative of the relative activities of the various glycosyltransferases present in the sample. The symbolic nomenclature used to annotate the resultant spectra is that used by the CFG (Fig. 2.2) (Varki et al., 2009). Even with very well informed biosynthetic knowledge, the possibility of alternative sequences for these putative assignments cannot initially be ruled out. Wherever possible, therefore, ES-MS/MS or MALDI-TOF-TOFMS/MS, in conjunction with additional data derived from experiments such as linkage analysis and enzyme digests, are applied to differentiate alternative possibilities. In most cases, structural cues are obtained from the fragmentation analysis of signals within the spectrum. This can lead,

60


A Glycan residues Monosaccharide

Abbreviation Symbol

Permethylated residue mass

Deoxyhexose Fucose

Fuc

174

Hexose Mannose Galactose Glucose

Man Gal Glc

204 204 204

GalNAc

245

GlcNAc N-Acetylglucosamine Sialic acid NeuAc N-Acetylneuraminic acid N-Glycolylneuraminic acid NeuGc

245

N-Acetylhexosamine N-Acetylgalactosamine

361 391

B Nonreducing and reducing ends Permethylated mass of ends

Glycan structure

H RO

CH2OR

CH2OR

CH2OR

O

O

O

H OR

Non-reducing end H

H

H OR

H O

H

H

H

H

OR H OR

Residues

Reducing end

H

OR

R + OR = 46

O

OR

H

n

H

OR

H

R = H for underivatised glycans R = CH3 for permethyl derivatives

Figure 2.2 (A) Symbol nomenclature as outlined by the Consortium for Functional Glycomics Nomenclature Committee (May 2004) (Varki et al., 2009). Full documentation available from: http://glycomics.scripps.edu/CFGnomenclature.pdf. (B) The mass of a permethylated sugar is calculated by adding the sum of the increment (or residue) masses of the sugars (a) to the sum of the masses of the reducing and nonreducing ends (b).

for example, to the establishment of the propensity for a cell or tissue type to produce antennal LacNAc extensions versus tri- or tetraantennary structures, or fucosylate on the antennae rather than (or as in most cases, as well as) on the N-glycan core. Where such features have been established, they are included in the structural representations. Structural features that remain ambiguous are represented in a manner that conveys this to the reader. Where there are large numbers of antennal fucosylation events, combined with sialic acid capping and multiple LacNAc extensions, it is impossible to give a single definitive glycan structure. Indeed, in our experience such structures inevitably exist in multiple

61


A

B

C ×2

×19

D

E ×1

×3

F

×2 b4

b4 b4 b2 b2

b4 a3

b4 b6

a6 b4 b4 a6

Figure 2.3 Representation of structural ambiguity: (A) Gives a typical representation of a triantennary structure. For convenience, only a single branching pattern is shown. Also, this composition could correspond to a biantennary structure with an extended antenna. Further experiments would be required to distinguish these structural features. (B) Shows an example of the use of brackets to convey both structural identity and variation. The cartoon indicates the presence of multiple (19) polylactosamine extensions but the exact length of an extended antennae is not indicated, nor on which antenna or antennae they are found. (C) Is an example of a structure with both sialic acid capping and LacNAc extensions. The sialic acids are able to cap unextended antennae, as well as the longer polyLacNAc type so in the absence of more definitive structural evidence these moieties are represented as shown. (D) Displays an example of a structure with both antennal fucosylation and sialylation. It is biosynthetically possible that the fucosylation and sialylation are on the same or different antennae. In the absence of any further structural evidence for a specific sample these residues are represented as shown. (E) Shows an example of a structure with antennal fucosylation. Linkages are not indicated on the cartoons nor should they be directly inferred from them. For example, it is not inferred that the fucosylated antennae is exclusively on the 3-linked or 6-linked mannose of the core. (F) Is an example of the use of the nomenclature system to fully annotate a structure where all of the linkage information is known.

isoforms, which are conveyed by use of brackets within the assignments (see Fig. 2.3 for visual examples) with the intention that the reader can better appreciate the potential for heterogeneity within such peaks.

62


4. Example Project: Characterization of Pancreatic Tissue from Wild-Type and Mgat4a Knockout Mice The mannosyl (a-1,3-)-glycoprotein b-1,4-N-acetylglucosaminyltransferase, isozyme A (Mgat4a) gene encodes an enzyme—N-acetylglucosaminyltransferase IVa (GlcNAcT-IVa)—that is responsible for the transfer of a GlcNAc residue in a b1-4 linkage onto the 3-arm of the trimannosyl core during N-glycan biosynthesis. In structural terms, knocking out this gene (together with the second isozyme, Mgat4b) diminishes the system’s ability to make tetraantennary N-glycans (Takamatsu et al., 2010). We present this data as a case study for the purpose of illustrating the application of the glycomic analysis methodology described within this chapter. All the samples analyzed were provided by Jamey Marth, Sanford-Burnham Medical Research Institute at UC Santa Barbara, USA (Ohtsubo et al., 2005).

4.1. MALDI-TOF MS mass fingerprinting The MALDI-TOF spectra—or mass fingerprints—of the N-glyan pool derived from wild-type (panel A) and Mgat4a knockout (panel B) mouse pancreatic tissue are shown in Fig. 2.4. In each case, the N-glycans consist of high mannose and bi-, tri- and potentially tetraantennary glycans. The complex glycans predominantly contain fucose on their cores, with antennal fucose being present in significant quantities. The antennae are terminated by both sialic acids (NeuAc/NeuGc) and Gal-a-Gal epitopes, with the latter being much more abundant. The most prevalent complex structure in the spectrum is a core-fucosylated, biantennary glycan bearing two Gal-a-Gal terminal groups (m/z 2651, Fuc1Hex7HexNAc4). Based upon relative intensities, there is a significant reduction in abundance of the higher mass complex glycans (m/z > 3000) in the knockout spectrum. Interestingly, the peak at m/z 2530, corresponding to either a bisected triantennary or a tetraantennary N-glycan, is completely absent in the knockout. The initial MALDI-TOF profiles seem to indicate that a significant perturbation of glycan production has been caused by the ablation of the Mgat4a gene.

4.2. MALDI-TOF/TOF MS sequencing In order to further clarify the structures present in the two samples, glycan sequencing using tandem MS/MS technology is performed using MALDITOF/TOF MS. These experiments are capable of providing insights into branching patterns, terminal epitopes, and can also inform on linkage

63


A

3071

3101 3142

3246

2040

2418

2448

2478

2489

2693

2897

3026

% Intensity

3306 3348

3551 3433

and/or

3620

2996

3899

and/or

3725

3755

3929

3959

4116

3869

2852 2622

2396

2244

3695

3463

2635

2326

50

3276

2652

100

2081 2966 and/or

2809

4056

2192

4086

4348

4378

4408

2285 2530

0 2000

2500

3000 3500 Mass (m/z)

4000

4500

B 100 2418

2448 2478

2489

2652

2897

3026

3071

3101 3246

3276

3306

% Intensity

2040

2635

50

2693

and/or 2244

2326

3695

2996

2285

3725

3755

3869

3929

2622 2852

3463

and/or 2966 3959 2081

4086

2396 2192 2809 3551

0 2000

2500

3000

3500

4000

4500

Mass (m/z)

Figure 2.4 MALDI-MS spectra of 50% (v:v) aqueous acetonitrile fractions of permethylated N-glycans from wild-type (panel A) and Mgat4a knockout mouse (panel B) pancreas. Structural assignments correspond to tentative structures based on compositions and knowledge of N-glycan biosynthesis.

positions. The spectra in Fig. 2.5 represent the data generated from two such analyses on the ions present at m/z 3755 in wild-type and Mgat4a knockout mouse pancreas. Key fragmentations are shown schematically and the peaks labeled with the corresponding product ions. The composition of this molecular ion (Fuc1Hex10HexNAc6) together with biosynthetic knowledge indicates that it is most likely to correspond

64


A HO

100

HO

1958

HO

3088 3074

2625 2407 1768

690

474

% Intensity

2231

HO HO

50

3292

2421 HO HO

2421

HO HO HO

HO

472

HO

HO

1958

+

[M+Na] 3755

2639

HO

2231

1768

1110

HO

2625

2407

HO HO

0 400

HO

Minor component Major component

HO

690 474

2435

1820 2530 Mass (m/z)

3240

3950

B 2639 1972

100

HO

3088 2407 1768

690 2188

474 Major component Minor component

HO

% Intensity

HO

3074

HO HO

2421

50

2625 HO HO

0 400

HO

2639

[M+Na]+ 3755

3292

1972

HO HO

HO

1768

690

HO

HO HO

2421 2188

1110

1820

2530

3240

3950

Mass (m/z)

Figure 2.5 MALDI-TOF/TOF spectra of the molecular ion m/z 3755 (Fuc1Hex10HexNAc6) from the 50% (v:v) aqueous of permethylated N-glycans from wild-type (panel A) and Mgat4a knockout mouse (panel B) pancreas. Assignments of the fragment ions generated are indicated, with the bold/red numbering indicating diagnostic fragment ions. Structures in each panel indicate the most abundant structure in the respective N-glycan sample.

with two potential structures, one triantennary and one tetraantennary. In each sample, fragmentation of this molecular ion produces major product ions corresponding to the single b-cleavage of HexHexNAc (m/z 3292)


65

and Hex2HexNAc (m/z 3088). Lack of fragment ions at m/z 1291 (FucHex3HexNAc2) and m/z 2669 (corresponding to the single b-cleavage of FucHex3HexNAc2), together with small peaks at m/z 474 indicate that the fucose is attached exclusively to the core of the glycan rather than the antennae in this peak. In addition, a low abundance fragment ion at m/z 2639 that contains one free -OH group is also observed. This corresponds to the loss of Hex3HexNAc2 from the molecular ion, thus implying the presence of the triantennary structure in both the wild-type and Mgat4a knockout mouse pancreas. In the wild-type sample, there is also a signal at m/z 1958 representing a triple independent b-cleavage of two Hex2HexNAc moieties and a HexHexNAc. This supports the presence of the tetraantennary isomer, suggesting that a mixture of both the tri- and tetraantennary structures shown in Fig. 2.5 is present in the wild-type mouse pancreas. In order to determine which of the isomers are more abundant, the peak heights of the signals at m/z 3292 (single b-cleavage of HexHexNAc) and m/z 2639 (single b-cleavage of HexHexNAc and Hex2HexNAc) were compared. From the spectrum, it is evident that the signal at m/z 3239 is higher relative to 2639, implying that the tetraantennary N-glycan is likely to be the more abundant structure in the wild-type mouse pancreas. In addition, to the signals observed in the wild type, additional signals are present in the knockout mouse pancreas. These include fragment ions at m/z 1972 and 2188, which are consistent with the double b-cleavage of Hex2HexNAc-HexHexNAc and Hex2HexNAc, and a single b-cleavage of Hex2HexNAc-HexHexNAc. Comparison of peak heights of signals at m/z 3292 and 2639 showed an increase of m/z 2639 relative to 3292 (Fig. 2.5, panel B). Taken together, these data suggest that the more abundant structure in the Mgat4a knockout mouse pancreas is the triantennary structure. Overall the data from these MS/MS analyses indicates that there is a mixture of isomers in both the wild-type and knockout mouse samples, with the tetraantennary structure in the majority in the wild-type pancreas while the triantennary dominates in the case of the Mgat4a knockout.

4.3. Enzymatic digestion—a-galactosidase treatment To clarify the anomeric configurations and the nature of the terminal epitopes, enzymatic digestions were carried out upon the mouse pancreas samples. In this example, an a-galactosidase digest was carried out. This glycosidase acts to cleave all a-linked, nonreducing terminal galactose residues (Kobata, 1979) in the sample. Once the digest was complete, an aliquot of each sample was permethylated and analyzed by MALDI-TOF MS. Figure 2.6 shows the result of the digests, with the wild-type sample in panel A and the Mgat4a knockout in panel B.

66


A 100

% Intensity

2244

2040 2070

50

2693 and/or and/or

2938

3041

3142

2081 2326

2192

2867 2418 2489 2635

2396

0 2000

2852

2400

3491

3026

3317

2800

3200

3600

4000

3600

4000

Mass (m/z)

B 100

% Intensity

2244

50

2040 2070

2693

and/or

2938

and/or

3142

3317

2081 2326

2418

2867

2489 2192

0 2000

2396

2400

2635

2852

3026 3041

2800

3200 Mass ( m/z)

Figure 2.6 MALDI-TOF profiles of a-galactosidase digestions of the 50% (v:v) aqueous acetonitrile fractions of permethylated N-glycans from wild-type (panel A) and Mgat4a knockout mouse (panel B) pancreas.

Comparison of these spectra with those of the untreated samples (Fig. 2.4) shows that a number of previously abundant signals are now totally absent (e.g., m/z 2448, 2652, 2809, 2987, and 3101). These results confirm the presence of terminal Gal-a-Gal epitopes in both the wild-type and knockout mouse pancreas. The remaining complex-type structures in the spectra were resistant to the digest, while those structures that in the untreated samples displayed Gal-a-Gal terminal structures have been trimmed and mass shifted.

67


4.4. Linkage analysis by GC–MS PMAA derived from the wild-type and knockout mouse pancreas were subjected to GC–MS linkage analyses. The total ion chromatograms (TIC) are displayed in Fig. 2.7 and key features of the data are tabulated in Table 2.1. The TIC and tabulated key features illustrate that the wild-type and the knockout pancreatic N-glycans contain similar components. It is noticeable from the TIC, however, that there has been a decrease in the abundance of 2,4-linked Man, which is consistent with the genotype of this mouse. In addition, a slight increase in 2-linked Man in the knockout mouse pancreas is observed. This corresponds with MALDI-MS data showing that biantennary structures are relatively more abundant than triand tetraantennary structures. This data continues to support the hypothesis that ablation of the Mgat4a gene has an effect on the branching of N-glycans in the pancreas and is consistent with results from the MALDI-MS and MS/MS experiments.

A t-Man 18.75

% Intensity

100

t-Gal 19.02

50

t-Fuc 17.22

2-Man 19.92 3-Gal 20.22

3,6-Man 21.70 2,6-Man 2,4-Man 21.14 6-Gal 21.53 20.74

4-GlcNAc 23.54 t-GlcNAc 22.64

4,6-GlcNAc 24.86

0

17.00 17.50 18.00 18.50 19.00 19.50 20.00 20.50 21.00 21.50 22.00 22.50 23.00 23.50 24.00 24.50 25.00

Elution time (min)

B

t-Man 18.76

100

% Intensity

2-Man 19.93 t-Gal 19.03 50

t-Fuc 17.24

3-Gal 20.23

3,6-Man 21.71 2,6-Man 2,4-Man 6-Gal 21.17 21.54 20.76

4-GlcNAc 23.54 t-GlcNAc 22.66

4,6-GlcNAc 24.86

0 17.00 17.50 18.00 18.50 19.00 19.50 20.00 20.50 21.00 21.50 22.00 22.50 23.00 23.50 24.00 24.50 25.00

Elution time (min)

Figure 2.7 Total ion chromatogram (TIC) of the partially methylated alditol acetates from the 50% acetonitrile N-glycan fractions of wild-type (panel A) and Mgat4a knockout (panel B) mouse pancreas.

Table 2.1 GC–MS linkage analysis of partially methylated alditol acetates derived from the 50% (v:v) aqueous acetonitrile fractions of the permethylated N-glycans derived from the wild-type and Mgat4a knockout mouse pancreas

Elution time (min) (WT)

Elution time (min) (KO)

Characteristic fragment ions

17.224 18.753 19.024 19.921 20.12 20.222 20.742 21.145 21.526 21.699 22.156 22.643 23.536 24.387 24.857

17.237 18.761 19.032 19.938 20.133 20.23 20.755 21.17 21.538 21.707 22.194 22.656 23.541 24.383 24.84

102, 115, 118, 131, 162, 175 102, 118, 129, 145, 161, 205 102, 118, 129, 145, 161, 205 129, 130, 161, 190, 204, 234 118, 129, 161, 202, 234 118, 129, 161, 203, 234, 277 99, 102, 118, 129, 162, 189, 233 87, 88, 99, 113, 130, 190, 233 87, 88, 129, 130, 189, 190 118, 129, 189, 202, 234 118, 139, 259, 333 117, 129, 145, 205, 247 117, 159, 233 117, 159, 346 117, 159, 261

The relative abundances were set against the most abundant sugar residue.

Assignments

Relative abundance (WT)

Relative abundance (KO)

Terminal fucose Terminal mannose Terminal galactose 2-linked mannose 3-linked mannose 3-linked galactose 6-linked galactose 2,4-linked mannose 2,6-linked mannose 3,6-linked mannose 3,4,6-linked mannose t-GlcNAc 4-linked GlcNAc 3,4-linked GlcNAc 4,6-linked GlcNAc

0.24 1.00 0.40 0.47 0.02 0.33 0.02 0.13 0.06 0.67 0.02 0.06 0.33 Trace 0.04

0.39 1.00 0.43 0.87 0.03 0.41 0.04 0.05 0.08 0.62 0.01 0.04 0.27 Trace 0.03


69

4.5. Summary This biological example illustrates the power of the MS-based glycomics approach described within this chapter for the analysis of cells and tissues from mutant mice. From a single murine pancreas, N-glycans can be purified, derivatized, and profiled by MALDI-MS to produce mass fingerprints and inform on the monosaccharide constituents of the glycans. These compositions, together with knowledge of the well-defined N-glycan biosynthetic pathways, allow the assignment of tentative N-glycan structures and structural isomers. Evidence from further analytical methods such as enzymatic digests, MS/MS sequencing and GC–MS linkage analysis enables the refinement of these assignments and incrementally increases the level of confidence and information, depending on how many of these methods are applied to the sample. In this case, both the wild-type and Mgat4a knockouts are observed to carry similar components, though in very different relative abundances. The majority of components are core fucosylated, while LacNAc, Lex and Gala1-3Gal are the major nonreducing end structures. Higher molecular weight components carry these terminal structures on tandem repeats of LacNAc. The major difference between the two samples is the significant decrease in abundance of tetraantennary structures in the Mgat4a knockout pancreas. These observations strongly correlate with results from the enzymatic assay studies showing that GlcNAcT-IVa activity is highest in the pancreas of the wild-type mouse and that the activity in the homozygous knockout is significantly reduced (Ohtsubo et al., 2005). These structural analyses imply that in murine pancreas GlcNAcT-IVa is the main enzyme catalyzing the reaction of transferring a GlcNAc residue via a b1-4 linkage onto the 3-arm of the b-linked mannose of the pentasaccharide core of N-glycans. The presence of 2,4-linked Man in the knockout mouse pancreas indicates that the loss of this enzyme only leads to a partial compensation by GlcNAcT-IVb (Takamatsu et al., 2010).

5. Summary of Glycan Structural Observations in Murine Tissues, Cells, and Knockouts There now follows a series of tables (Tables 2.2–2.5), summarizing in a general fashion the structural observations derived from the analysis of murine tissues and cell isolates, from both wild-type C57Bl/6 and glycosylation-related genetic knockout mice. The data corresponding to these observations is available from the glycan profiling data pages of the CFG—http://www.functionalglycomics.org.

Table 2.2 Summary of glycan-related gene knockout mice studied by the CFG and summarized in this chapter (Araki et al., 1999; Ellies et al., 1998; Hashimoto et al., 1983; Homeister et al., 2001; Kurosawa et al., 1996; Maly et al., 1996; Ponder et al., 1985; Priatel et al., 2000) Original publication name

Gene symbol

C2 GlcNAcT

Gcnt 1

FucT-IV FucT-VII FucT-IV and FucT-VII ST3Gal-I

Fut4 Fut7 Fut7

St6galnac-II

St6galnac2

Mgat4a

Mgat4a

St3gal1

Gene name(s)

Reported function

Primary reference

C2 GlcNAcT, b-1,6 N-acetylglucosaminyltransferase, IGnT, 5630400D21Rik a-1,3 fucosyltransferase-IV a-1,3 fucosyltransferase VII a-1,3 fucosyltransferase IV and a-1,3 fusosyltransferase VII Siat4, ST3GalI, CMP-N-acetylneuraminate: b-galactosidase aSiat7, Siat7b, ST6GalNAc II

Synthesis of Core 2 branch on serine- and threoninelinked O-glycans Synthesis of selectin ligands Synthesis of selectin ligands Synthesis of selectin ligands

Ellies et al. (1998)

Core 1 O-glycan sialylation

Priatel et al. (2000)

Synthesis of the Sialyl-Tn antigen Synthesis of the b1-4 branch of N-glycans

Kurosawa et al. (1996)

Mannoside acetylglucosaminyltransferase 4, isoenzyme A, 9530018I07Rik, GnT-IVa

Homeister et al. (2001) Maly et al. (1996) Homeister et al. (2001)

Araki et al. (1999)

71


Table 2.3 Summary table of N-glycan structures observed in the analysis of tissues and cells from C57Bl/6 wild-type mice

Wehi-3

Neutrophils

Thymus

Testes

Spleen

Small intestine

Pancreas

Ovaries

Lung

Liver

Kidney

Heart

Example of structure

Colon

Structural characteristics

Brain

Observation in C57BI/6 wild-type murine tissues and cell isolates

N-glycosylation (high mannose) a6

High mannose

a6

a3

b4

b4

a3

N-glycosylation (hybrid) a6

Hybrid structures

a6 a3 b2

b4

+/- a6 b4

a3

m m m m m

m m m m m m m m

N-glycosylation (complex)

Heavily truncated structures

a6

b2 b2

b4

b4

b4

+/- a6 b4

b4

+/- a6 b4

a6 b4

+/- a6 b4

m

a3

m

m

m m

m m

b4

Bi-antennary structures Tri-antennary structures

a6

b2 b2

a3

a6

b2 b2

a3

b4

b6

Tetra-antennary structures

b2 b2

a3 b4

Bisecting GlcNAc Sialylation

b4

+/- a6 b4

b2

a3

b2

a6 b4

+/- a6 b4

b2

a3

b2

b4

a6

a6

b2

Gal-a-Gal

b2

a6 b4

b2

a3

Core fucose

b2

a6 b4

b2

a3

Lewis

x/a

(antennal fucose)

Sialyl lewis

x/a

Antennal Fuc (multiple)

b4

a6 b4

b2

a3

b2

a6 b4

b2

a3

b2

a6

m Only

+/- a6 b4

m m m m m m

m m

+/- a6 b4

m

m

a3

b2

b2

m m m

Only

LacNAc extensions

b2

m m m

m

b4 a6

b4

m

a6

m

+/- a6 b4

b4

b4 a6

b4

b4

a3

m

m

b4

Sd

a

b2 b2

a6

a6

m

a3

Structural assignments are based upon MALDI-TOF, MALDI-TOF/TOF, and ESI-MS analyses, with t linkages assigned according to known biosynthetic pathways. ¼ present, x ¼ absent, m ¼ present in minor amounts. The data corresponding to this summary can be found within the CFG databases at http://www.functionalglycomics.org (Babu et al., 2009; Comelli et al., 2006; North et al., 2010a,b).

72


Table 2.4 Summary table of O-glycan structures observed in the analysis of tissues and cells from C57Bl/6 wild-type mice

WeHi-3

Neutrophils

Thymus

Testes

Spleen

Small intestine

Pancreas

Ovaries

Lung

Liver

Kidney

Heart


Colon


Brain

Observation in C57BI/6 wild-type murine tissues and cell isolates

Core-1 O-glycosylation

Core-1

b3

Sialylated core-1

b3

Only

Only

Only

Only

Only

Only

Core-2 O-glycosylation b6

Core-2

b3 b6

Sialylated core-2 x

Lewis (antennal fucose)

b3

b6 b3

Fucosylation and sialylation

b6

m

b3

Sialyl lewis

x

b6 b3

m

b6

LacNAc extensions

Sd

a

b3

b6 b3 b6

Gal-a-Gal

b3

O-mannosyl

O-mannosyl Sialylated O-mannosyl

Structural assignments are based upon MALDI-TOF, MALDI-TOF/TOF, and ESI-MS analyses, with linkages assigned according to known biosynthetic pathways. ✓ ¼ present, x ¼ absent, m ¼ present in minor amounts, ¼ analysis inconclusive due to paucity of data. The data corresponding to this summary can be found within the CFG databases at http://www.functionalglycomics.org (Babu et al., 2009; Comelli et al., 2006; North et al., 2010a,b).

73


Table 2.5 Summary table of N-glycan and O-glycan structures observed in analysis of tissues of C57Bl/6 in the absence of specified glycan-related genes (see Table 2.2). Mutant mouse tissues-changes relative to wild-type Thymus

Testes

Spleen

Small intestine

Kidney

Colon

Brain

Spleen

Kidney

FucT IV + FucT VII Thymus

FucT VII Thymus


Spleen


Kidney

FucT IV

N-glycosylation (high mannose) a6

High mannose

a3

a6 b4

b4

a3

N-glycosylation (hybrid) a6

Hybrid structures

a3 b2

a6 b4

No changes in reltive levels of N- or O-linked glycosylation were observed in any of these mutants

+/- a2 b4

a3

N-glycosylation (complex)

Heavily truncated structures

b2 b2

a6

b4

b4

a3

b4

Mutant mouse tissues-changes relative to wildtype Thymus

Lymph nodes

Kidney

Thymus

Testes

Core 2 GalNAcT Spleen

Small intestine

Kidney

Colon

ST3Gal I Brain


Testes


Kidney

St6GalNAc II

Core-1 O-glycosylation

Core-1

b3

Sialylated core-1

b3

Core-2 O-glycosylation b6

Core-2

b3

Structural assignments are based upon MALDI-TOF, MALDI-TOF/TOF, and ESI-MS analyses, with linkages assigned according to known biosynthetic pathways. Upward arrows indicate a relative increase in abundance, downward arrows indicate a relative decrease in abundance, an equals sign indicates no significant change in abundance and structural archetypes not detected are indicated with a cross. The data corresponding to this summary can be found within the CFG databases at http://www.functionalglycomics.org (Babu et al., 2009; Comelli et al., 2006; North et al., 2010a,b).

ACKNOWLEDGMENTS This work was supported by the Analytical Glycotechnology Core (Core C) of the Consortium for Functional Glycomics Grant GM 62116, The Biotechnology and Biological Sciences Research Council Grants BBF0083091 and B19088, and The Wellcome Trust.

74


REFERENCES Akama, T. O., et al. (2006). Essential and mutually compensatory roles of a-mannosidase II and a-mannosidase IIx in N-glycan processing in vivo in mice. Proc. Natl. Acad. Sci. USA 103, 8983–8988. Albersheim, P., et al. (1967). A method for the analysis of sugars in plant cell-wall polysaccharides by gas–liquid chromatography. Carbohydr. Res. 5, 340–345. Angel, A. S., and Nilsson, B. (1990). Linkage positions in glycoconjugates by periodate oxidation and fast atom bombardment mass spectrometry. Methods Enzymol. 193, 587–607. Aoki-Kinoshita, K. F. (2008). An introduction to bioinformatics for glycomics research. PLoS Comput. Biol. 4, e1000075. Araki, K., et al. (1999). Exchangeable gene trap using the Cre/mutated lox system. Cell. Mol. Biol. (Noisy-le-grand) 45, 737–750. Babu, P., et al. (2009). Structural characterisation of neutrophil glycans by ultra sensitive mass spectrometric glycomics methodology. Glycoconj. J. 26, 975–986. Bax, M., et al. (2007). Dendritic cell maturation results in pronounced changes in glycan expression affecting recognition by siglecs and galectins. J. Immunol. 179, 8216–8224. Bhaumik, M., et al. (1998). Progression of hepatic neoplasms is severely retarded in mice lacking the bisecting N-acetylglucosamine on N-glycans: Evidence for a glycoprotein factor that facilitates hepatic tumor progression. Cancer Res. 58, 2881–2887. Bobbitt, J. M. (1956). Periodate oxidation of carbohydrates. Adv. Carbohydr. Chem. 48, 1–41. Ceroni, A., et al. (2008). GlycoWorkbench: A tool for the computer-assisted annotation of mass spectra of glycans. J. Proteome Res. 7, 1650–1659. Chen, X., and Flynn, G. C. (2007). Analysis of N-glycans from recombinant immunoglobulin G by on-line reversed-phase high-performance liquid chromatography/mass spectrometry. Anal. Biochem. 370, 147–161. Chui, D., et al. (1997). Alpha-mannosidase-II deficiency results in dyserythropoiesis and unveils an alternate pathway in oligosaccharide biosynthesis. Cell 90, 157–167. Comelli, E. M., et al. (2006). A focused microarray approach to functional glycomics: Transcriptional regulation of the glycome. Glycobiology 16, 117–131. Corfield, A. P., et al. (1983). The specificity of viral and bacterial sialidases for alpha(2-3)- and alpha(2-6)-linked sialic acids in glycoproteins. Biochim. Biophys. Acta 744, 121–126. Dell, A. (1987). F.A.B.-mass spectrometry of carbohydrates. Adv. Carbohydr. Chem. Biochem. 45, 19–72. Dell, A. (1990). Preparation and desorption mass spectrometry of permethyl and peracetyl derivatives of oligosaccharides. Methods Enzymol. 193, 647–660. Dell, A., and Ballou, C. E. (1983). Fast-atom-bombardment, negative-ion mass spectrometry of the mycobacterial O-methyl-D-glucose polysaccharide and lipopolysaccharides. Carbohydr. Res. 120, 95–111. Dell, A., et al. (1994). Mass spectrometry of carbohydrate-containing biopolymers. Methods Enzymol. 230, 108–132. Dell, A., et al. (2007). Mass spectrometry of glycoprotein glycans: Glycomics and glycoproteomics. In ‘‘Comprehensive Glycoscience, Vol. 2,’’ ( J. P. Kamerling, ed.), pp. 69–100. Elsevier, UK. Dell, A., et al. (2008). Glycomics and Mass Spectrometry . pp. 2191–2217, Springer-Verlag Berlin Heidelberg New York, Leipzig, Germany. Domon, B., and Costello, C. E. (1988). Structure elucidation of glycosphingolipids and gangliosides using high-performance tandem mass spectrometry. Biochemistry 27, 1534–1543.


75

Ellies, L. G., et al. (1998). Core 2 oligosaccharide biosynthesis distinguishes between selectin ligands essential for leukocyte homing and inflammation. Immunity 9, 881–890. Fenn, J. B., et al. (1989). Electrospray ionization for mass spectrometry of large biomolecules. Science 246, 64–71. Fukuda, M., et al. (1984a). Structure of branched lactosaminoglycan, the carbohydrate moiety of band 3 isolated from adult human erythrocytes. J. Biol. Chem. 259, 8260–8273. Fukuda, M., et al. (1984b). Structure of sialylated fucosyl lactosaminoglycan isolated from human granulocytes. J. Biol. Chem. 259, 10925–10935. Fukuda, M. N., et al. (1985). Embryonal lactosaminoglycan. The structure of branched lactosaminoglycans with novel disialosyl (sialyl alpha 2–9 sialyl) terminals isolated from PA1 human embryonal carcinoma cells. J. Biol. Chem. 260, 6623–6631. Goldberg, D., et al. (2005). Automatic annotation of matrix-assisted laser desorption/ionization N-glycan spectra. Proteomics 5, 865–875. Goldberg, D., et al. (2009). Glycan family analysis for deducing N-glycan topology from single MS. Bioinformatics 25, 365–371. Harvey, D. J. (1999). Matrix-assisted laser desorption/ionization mass spectrometry of carbohydrates. Mass Spectrom. Rev. 18, 349–450. Hashimoto, Y., et al. (1983). Genetic regulation of GM2 expression in liver of mouse. J. Biochem. 93, 895–901. Haslam, S. M., et al. (1997). Characterisation of the phosphorylcholine-containing N-linked oligosaccharides in the excretory-secretory 62 kDa glycoprotein of Acanthocheilonema viteae. Mol. Biochem. Parasitol. 85, 53–66. Homeister, J. W., et al. (2001). The alpha(1, 3)fucosyltransferases FucT-IV and FucT-VII exert collaborative control over selectin-dependent leukocyte recruitment and lymphocyte homing. Immunity 15, 115–126. Ioffe, E., and Stanley, P. (1994). Mice lacking N-acetylglucosaminyltransferase I activity die at mid-gestation, revealing an essential role for complex or hybrid N-linked carbohydrates. Proc. Natl. Acad. Sci. USA 91, 728–732. Jang-Lee, J., et al. (2006). Glycomic profiling of cells and tissues by mass spectrometry: Fingerprinting and sequencing methodologies. Methods Enzymol. 415, 59–86. Kobata, A. (1979). Use of endo- and exoglycosidases for structural studies of glycoconjugates. Anal. Biochem. 100, 1–14. Kui Wong, N., et al. (2003). Characterization of the oligosaccharides associated with the human ovarian tumor marker CA125. J. Biol. Chem. 278, 28619–28634. Kuraya, N., and Hase, S. (1996). Analysis of pyridylaminated O-linked sugar chains by twodimensional sugar mapping. Anal. Biochem. 233, 205–211. Kurosawa, N., et al. (1996). Molecular cloning and genomic analysis of mouse Galbeta1, 3GalNAc-specific GalNAc alpha2, 6-sialyltransferase. J. Biol. Chem. 271, 15109–15116. Lundblad, A., et al. (1976). Mannosidosis. Storage material, alpha-mannosidase specificity and diagnostic methods. Adv. Exp. Med. Biol. 68, 301–311. Maly, P., et al. (1996). The alpha(1, 3)fucosyltransferase Fuc-TVII controls leukocyte trafficking through an essential role in L-, E-, and P-selectin ligand biosynthesis. Cell 86, 643–653. Mason, K. E., et al. (2006). Characterization of sulfated oligosaccharides in mucopolysaccharidosis type IIIA by electrospray ionization mass spectrometry. Anal. Chem. 78, 4534–4542. Mechref, Y., and Novotny, M. V. (2002). Structural investigations of glycoconjugates at high sensitivity. Chem. Rev. 102, 321–369. Metzler, M., et al. (1994). Complex asparagine-linked oligosaccharides are required for morphogenic events during post-implantation development. EMBO J. 13, 2056–2065. Miyatake, T., and Suzuki, K. (1975). Partial purification and characterization of betagalactosidase from rat brain hydrolyzing glycosphingolipids. J. Biol. Chem. 250, 585–592.

76


Montpetit, M. L., et al. (2009). Regulated and aberrant glycosylation modulate cardiac electrical signaling. Proc. Natl. Acad. Sci. USA 106, 16517–16522. Moody, A. M., et al. (2001). Developmentally regulated glycosylation of the CD8alphabeta coreceptor stalk modulates ligand binding. Cell 107, 501–512. Moody, A. M., et al. (2003). Sialic acid capping of CD8beta core 1-O-glycans controls thymocyte-major histocompatibility complex class I interaction. J. Biol. Chem. 278, 7240–7246. Morris, H. R., et al. (1996). High sensitivity collisionally-activated decomposition tandem mass spectrometry on a novel quadrupole/orthogonal-acceleration time-of-flight mass spectrometer. Rapid Commun. Mass Spectrom. 10, 889–896. North, S. J., et al. (2006). Glycomic studies of Drosophila melanogaster embryos. Glycoconj. J. 23, 345–354. North, S. J., et al. (2009). Mass spectrometry in the analysis of N-linked and O-linked glycans. Curr. Opin. Struct. Biol. 19, 498–506. North, S. J., et al. (2010a). Mouse and human glycomes. In ‘‘Handbook of Glycomics,’’ (D. C. Richard and J. M. Pierce, eds.), pp. 263–327. Academic Press, San Diego. North, S. J., et al. (2010b). Glycomics profiling of Chinese hamster ovary cell glycosylation mutants reveals N-glycans of a novel size and complexity. J. Biol. Chem. 285, 5759–5775. Ohtsubo, K., et al. (2005). Dietary and genetic control of glucose transporter 2 glycosylation promotes insulin secretion in suppressing diabetes. Cell 123, 1307–1321. Pang, P. C., et al. (2009). Analysis of the human seminal plasma glycome reveals the presence of immunomodulatory carbohydrate functional groups. J. Proteome Res. 8, 4906–4915. Parry, S., et al. (2006). Glycomics investigation into insulin action. Biochim. Biophys. Acta 1760, 652–668. Parry, S., et al. (2007). Integrated mass spectrometric strategy for characterizing the glycans from glycosphingolipids and glycoproteins: Direct identification of sialyl Le(x) in mice. Glycobiology 17, 646–654. Perdivara, I., et al. (2009). Elucidation of O-glycosylation structures of the beta-amyloid precursor protein by liquid chromatography-mass spectrometry using electron transfer dissociation and collision induced dissociation. J. Proteome Res. 8, 631–642. Ponder, B. A., et al. (1985). An allelic difference determines reciprocal patterns of expression of binding sites for Dolichos biflorus lectin in inbred strains of mice. J. Embryol. Exp. Morphol. 87, 229–239. Priatel, J. J., et al. (2000). The ST3Gal-I sialyltransferase controls CD8þ T lymphocyte homeostasis by modulating O-glycan biosynthesis. Immunity 12, 273–283. Schneider, P., and Ferguson, M. A. (1995). Microscale analysis of glycosylphosphatidylinositol structures. Methods Enzymol. 250, 614–630. Scudder, P., et al. (1983). Isolation and characterization of an endo-beta-galactosidase from Bacteroides fragilis. Biochem. J. 213, 485–494. Sihlbom, C., et al. (2009). Localization of O-glycans in MUC1 glycoproteins using electroncapture dissociation fragmentation mass spectrometry. Glycobiology 19, 375–381. Stoll, M. S., et al. (1990). Microscale sequencing of O-linked oligosaccharides using mild periodate oxidation of alditols, coupling to phospholipid and TLC-MS analysis of the resulting neoglycolipids. Eur. J. Biochem. 189, 499–507. Stone, E. L., et al. (2009). Glycosyltransferase function in core 2-type protein O glycosylation. Mol. Cell. Biol. 29, 3770–3782. Sutton-Smith, M., et al. (2000). A rapid mass spectrometric strategy suitable for the investigation of glycan alterations in knockout mice. Tetrahedron Asymmetry 11, 363–369. Sutton-Smith, M., et al. (2002). MS screening strategies: investigating the glycomes of knockout and myodystrophic mice and leukodystrophic human brains. Portland Press, University of York, Biochemical Society Symposium 69, pp. 105–115.


77

Takamatsu, S., et al. (2010). Physiological and glycomic characterization of N-acetylglucosaminyltransferase-IVa and -IVb double deficient mice. Glycobiology 20, 485–497. Tarentino, A. L., and Plummer, T. H., Jr. (1994). Enzymatic deglycosylation of asparaginelinked glycans: Purification, properties, and specificity of oligosaccharide-cleaving enzymes from Flavobacterium meningosepticum. Methods Enzymol. 230, 44–57. Tarentino, A. L., et al. (1985). Deglycosylation of asparagine-linked glycans by peptide: N-glycosidase F. Biochemistry 24, 4665–4671. Tretter, V., et al. (1991). Peptide-N4-(N-acetyl-beta-glucosaminyl)asparagine amidase F cannot release glycans with fucose attached alpha 1–3 to the asparagine-linked N-acetylglucosamine residue. Eur. J. Biochem. 199, 647–652. Uchida, Y., et al. (1977). Distribution of neuraminidase in Arthrobacter and its purification by affinity chromatography. J. Biochem. 82, 1425–1433. Varki, A., et al. (2009). Symbol nomenclature for glycan representation. Proteomics 9, 5398–5399. Vestal, M. L., and Campbell, J. M. (2005). Tandem time-of-flight mass spectrometry. Methods Enzymol. 402, 79–108. von der Lieth, C. W., et al. (2006). The role of informatics in glycobiology research with special emphasis on automatic interpretation of MS spectra. Biochim. Biophys. Acta 1760, 568–577. Wada, Y., et al. (2007). Comparison of the methods for profiling glycoprotein glycans–HUPO Human Disease Glycomics/Proteome Initiative multi-institutional study. Glycobiology 17, 411–422. Wada, Y., et al. (2010). Comparison of methods for profiling O-glycosylation: HUPO human disease glycomics/proteome initiative multi-institutional study of IgA1. Mol. Cell. Proteomics 4, 719–727. Wang, Y., et al. (2006). Approaches to the study of N-linked glycoproteins in human plasma using lectin affinity chromatography and nano-HPLC coupled to electrospray linear ion trap–Fourier transform mass spectrometry. Glycobiology 16, 514–523. York, W. S., et al. (2010). Integration of glycomics knowledge and data. In ‘‘Handbook of Glycomics,’’ (D. C. Richard and J. M. Pierce, eds.), pp. 177–195. Academic Press, San Diego. Yost, R. A., and Enke, C. G. (1978). Selected ion fragmentation with a tandem quadrupole mass-spectrometer. J. Am. Chem. Soc. 100, 2274–2275. Yu, S. Y., et al. (2006). Distinctive characteristics of MALDI-Q/TOF and TOF/TOF tandem mass spectrometry for sequencing of permethylated complex type N-glycans. Glycoconj. J. 23, 355–369. Zaia, J. (2008). Mass spectrometry and the emerging field of glycomics. Chem. Biol. 15, 881–892.

C H A P T E R

T H R E E

Glycosaminoglycan Characterization by Electrospray Ionization Mass Spectrometry Including Fourier Transform Mass Spectrometry Tatiana N. Laremore,* Franklin E. Leach III,† Kemal Solakyildirim,* I. Jonathan Amster,† and Robert J. Linhardt*,‡ Contents 1. Overview 2. Preparation of Crude PG/GAG 3. Disaccharide Profiling Using Ion-Pairing Reverse-Phase (IP RP) HPLC with MS Detection 4. CS/DS Disaccharide Preparation 4.1. Materials and solutions for CS/DS disaccharide preparation 4.2. Method for CS/DS disaccharide preparation 5. ESI IP RP LC MS Analysis of CS/DS Disaccharides 5.1. Materials for CS/DS disaccharide analysis 5.2. Solutions for CS/DS disaccharide analysis 5.3. Method for CS/DS disaccharide analysis 6. HS Disaccharide Preparation 6.1. Materials and solutions for HS disaccharide preparation 6.2. Method for HS disaccharide preparation 7. ESI IP RP LC MS Analysis of HS Disaccharides 7.1. Materials for HS disaccharide analysis 7.2. Solutions for HS disaccharide analysis 7.3. Method for HS disaccharide analysis 8. Direct Infusion ESI FTMS Analysis of Oligosaccharides and Polysaccharides Separated by Preparative Continuous-Elution PAGE 9. Preparation of Bikunin GAG

81 83 84 84 84 85 85 85 86 87 88 88 89 90 90 91 91

92 93

* Department of Chemistry and Chemical Biology, Rensselaer Polytechnic Institute, Troy, New York, USA Department of Chemistry, University of Georgia, Athens, Georgia Departments of Chemical and Biological Engineering and Biology, Center for Biotechnology and Interdisciplinary Studies, Rensselaer Polytechnic Institute, Troy, New York, USA

{ {

Methods in Enzymology, Volume 478 ISSN 0076-6879, DOI: 10.1016/S0076-6879(10)78003-4

#


79

80

Tatiana N. Laremore et al.

9.1. Materials and solutions for GAG release by b-elimination under reducing conditions 9.2. Method for GAG release by b-elimination under reducing conditions 9.3. Materials and solutions for preparative CE PAGE separation of the bikunin GAG chains 9.4. Method for preparative CE PAGE separation of the bikunin GAG chains 9.5. Materials and solutions for purification of gel-eluted GAG fractions for FTMS analysis 9.6. Method for purification of gel-eluted GAG fractions for FTMS analysis 9.7. ESI FTMS analysis of bikunin GAG chains separated by preparative PAGE 9.8. Materials and solutions for the ESI FTMS analysis of gel-eluted bikunin GAG fractions 9.9. Method for the ESI FTMS analysis of gel-eluted bikunin GAG fractions 10. An Approach to FTMS Data Interpretation 11. Direct Infusion ESI FT-ICR MS Analysis of Bikunin GAG Mixture 11.1. Method for the ESI FT-ICR MS analysis of bikunin GAG mixture 12. Structural Characterization of GAG Oligosaccharides by Tandem Mass Spectrometry 12.1. Method for structural characterization of GAG oligosaccharides by tandem MS 13. Summary Acknowledgments References

93 94 95 96 97 97 99 99 99 100 101 101 102 103 105 106 106

Abstract Electrospray ionization mass spectrometry (ESI MS) is a versatile analytical technique in glycomics of glycosaminoglycans (GAGs). Combined with enzymology, ESI MS is used for assessing changes in disaccharide composition of GAGs biosynthesized under different environmental or physiological conditions. ESI coupled with high-resolution mass analyzers such as a Fourier transform mass spectrometer (FTMS) permits accurate mass measurement of large oligosaccharides and intact GAGs as well as structural characterization of GAG oligosaccharides using information-rich fragmentation methods such as electron detachment dissociation. The first part of this chapter describes methods for disaccharide compositional profiling using ESI MS and the second part is dedicated to FTMS and tandem MS methods of GAG compositional and structural analysis.

ESI MS and FTMS of GAGs

81

1. Overview Electrospray ionization (ESI) is a soft ionization method and is most suitable for mass spectrometric (MS) analysis of glycosaminoglycans (GAGs) (Bielik and Zaia, 2010; Zaia, 2005, 2008, 2009). The spray conditions can be modified to suppress the loss of sulfo groups and enhance the abundance of a certain types of ions, such as, for example, sodium cationized or protonated species and ions with specific charge states (Zaia, 2005). Compatibility of ESI with liquid chromatographic (LC) separation is another advantage of this ionization method for GAG analysis. There are 17 different structures of GAG-derived 4,5-unsaturated disaccharides but only 8 unique masses; therefore, an LC separation step is required to distinguish the isomers. The sensitivity of ESI MS applications in glycomics of GAGs is superior to other methods of ionization and is continuously improving (Flangea et al., 2009; Staples et al., 2009, 2010; Zamfir et al., 2004, 2009). Modern ESI sources with spray emitter diameters on the order of mm support nL/min flow rates (nano-ESI), resulting in very fine sprays and softer, more efficient desolvation conditions, which in turn leads to an order of magnitude enhancement in the analytical sensitivity (Zaia, 2008, 2009). Due to their acidity, GAG oligosaccharides are usually analyzed by ESI MS in the negative ionization mode. However, under certain experimental conditions, GAG oligosaccharides can be detected as positive ions with comparable or even greater sensitivity (Gunay et al., 2003). ESI MS in GAG analysis can be divided into two broad categories: the disaccharide composition analysis and the larger oligosaccharide and polysaccharide analysis. The two categories provide complementary information and aid in elucidation of structure–function relationship of GAGs. This chapter describes the ESI MS methods for chondroitin sulfate/dermatan sulfate (CS/DS) and heparan sulfate (HS) disaccharide composition analysis using ion-pairing reverse-phase (IP RP) LC as the separation method. In our laboratory, the methods here described of disaccharide composition profiling are routinely used in determining the effects of different stimuli on GAG expression in various cells, tissues, and small organisms (Nairn et al., 2007; Sinnis et al., 2007; Warda et al., 2006; Zhang et al., 2009a,b). While the disaccharide compositional profiling is robust and wellestablished approach to GAG analysis, ESI MS analysis of intact GAGs and large oligosaccharides requires high resolving power of the Fourier transform mass spectrometry (FTMS). The challenges in the MS analysis of intact GAGs arise from physicochemical properties of these biopolymers. GAG component of a single PG is a polydisperse, structurally heterogeneous mixture of chains having different sulfation patterns and associated with innumerable combinations of cations through electrostatic interactions in a physiological solution. Thus, the success of an ESI FTMS analysis of intact GAGs depends on the spray solution conditions (pH and ion composition) and on purity of

82


the GAG sample. This chapter describes two methods for analyzing intact GAG component of a PG bikunin over a narrow molecular weight (MW) range. First method uses continuous-elution polyacrylamide gel electrophoresis (CE PAGE) for separation of bikunin GAG prior to the ESI FTMS analysis. In the second method, the GAGs are ionized as a mixture, and a quadrupole mass filter is used in a manner of sliding window to allow a narrow m/z ‘‘band’’ of ions reach the Fourier Transform Ion Cyclotron Resonance (FT-ICR) mass analyzer (Chi et al., 2008). The advantages of the latter method are its speed and the low analyte requirement: the MW and composition of the bikunin GAG was determined in a matter of hours using less than 10 mg of the GAG mixture. However, only five chain lengths with two different sulfation states were identified, representing a small fraction of the bikunin GAG ensemble. The former method requires a significant amount of the analyte (300 mg) and is time-consuming, but it permits the MW and composition analysis over entire MW distribution of bikunin GAG. In addition, the isolated GAG fractions are amenable for further characterization, for example, disaccharide or oligosaccharide compositional and structural analysis. A single-stage high-resolution MS analysis provides information about the oligosaccharide composition, whereas a tandem MS analysis allows for the determination of the type and position of each structural element in the oligosaccharide (Saad and Leary, 2005; Tissot et al., 2008; Wolff et al., 2008a; Zamfir et al., 2002, 2003, 2004, 2009). During a tandem MS experiment, a selected ion, that is, precursor ion is activated to induce its fragmentation. The resulting fragment ions provide information about the number and positions of O-sulfo, N-sulfo, and N-acetyl groups and in some cases help to distinguish the C-5 epimers of uronic acid (Wolff et al., 2007b). The convention for the oligosaccharide fragment ion assignment follows the Domon and Costello nomenclature (Fig. 3.1). For tandem mass analysis, precursor ion activation can be categorized as a threshold or electron-based method. The threshold activation methods, including collision-induced dissociation (CID) and infrared multiphoton dissociation (IRMPD) (Zimmerman et al., 1991), typically cleave the most labile bond in sulfated GAGs, the sulfate half-ester (Zaia, 2005). Selection of a precursor ion in which sulfo groups are ionized or paired with a metal cation minimizes the loss of SO3 (Wolff et al., 2008b; Zaia and Costello, 2003). Recently, electron detachment dissociation (EDD) (Budnik et al., 2001) has been applied in the structural characterization of sulfated GAGs. Compared to the threshold activation methods, EDD results in more abundant cross-ring fragmentation (Wolff et al., 2007a). During this ion activation method, a multiply charged anion is irradiated with electrons of moderate kinetic energy (19 eV) resulting in the formation of an excited state intermediate. The intermediate then undergoes fragmentation due to electron detachment or electronic excitation (i.e., electron-induced dissociation, EID) (Fig. 3.2).

83


Y3 n,m

1,5

Z3

X3

HOH2C

HOOC O

5

4

OH HO

2

OH

1

A1

O OH

O

OH

O OH

NH2

B1 n,m

X0

HOH2C O

0 OH

3

n,m

Z1

HOOC O

OH

O

Y1

X1

0,2

C1

A3

NH2 B3 C3

n,m

A4

Figure 3.1 The Domon-Costello nomenclature for naming hexose fragments. Glycosidic cleavages are denoted with B or C if they contain the nonreducing end (NRE), and Y or Z if they contain the reducing end (RE). Cross-ring cleavages are denoted by A or X with A containing the NRE and X containing the RE. Cross-ring superscripts indicate the cleaved bonds and subscripts indicated the position along the oligosaccharide. Complementary cleavage subscripts add up to the length of the oligosaccharide, for example, B3 þ Y1 indicate a tetrasaccharide.

EID e– n– Pn– 19eV (P )* n≥2 –e– P(n–1)•

EDD

An–, B(n–1)

Even electron C(n–1)•, D(n–1) Odd and even electron

Figure 3.2 A scheme of electron-based activation of multiply charged anions during the EDD experiment.

The experimental conditions for high-resolution tandem MS characterization of GAG oligosaccharides are described later in this chapter.

2. Preparation of Crude PG/GAG Procedures for extraction of proteoglycans (PGs) and GAGs from tissues, cells, or whole organisms vary depending on a type of biological sample and are described in great detail in recent literature (Guimond et al., 2009; Iozzo, 2001). A workflow for preparation of crude PG/GAG sample includes (1) tissue or cell homogenization or disruption in a small volume of a neutral pH buffer; (2) PG extraction in a buffered 4 M guanidinium chloride solution containing 0.5–2% CHAPS; (3) buffer exchange to 8 M urea containing 0.5–2% CHAPS; and (4) anion exchange chromatographic enrichment

84


of PGs. If the protein component, including PG core protein, of the tissue or cell sample has to be identified by proteomics methods, protease inhibitors are added to the homogenization and extraction buffers to prevent nonspecific proteolysis. The 4 M guanidinium chloride buffer must be exchanged to a 6– 8 M urea buffer containing 0.5–2% CHAPS because guanidinium chloride is incompatible with the ion exchange separation. PGs bound to the anionexchange medium are washed with the urea/CHAPS buffer, followed by a low-salt buffer, such as 0.15 M sodium chloride to remove contaminants associated with PGs through electrostatic interactions. The PGs are released from the anion-exchange medium with an increasing salt gradient or with a high-salt buffer, and the resulting crude PG fraction is desalted for subsequent enzymatic treatment and analysis. To further enrich and purify the GAG, peptidoglycans (pGs) are obtained from the crude PG fraction by the digestion with DNAse followed by a nonspecific proteolysis and the pGs are purified from the mixture in the second anion-exchange step. The desalted pG fraction is subjected to further treatment with specific GAG lyases to identify the GAGs in the sample and determine their disaccharide composition. Alternatively, GAGs are released from the protein/peptide core of PGs/pGs by b-elimination under reducing conditions prior to the lyase treatment.

3. Disaccharide Profiling Using Ion-Pairing Reverse-Phase (IP RP) HPLC with MS Detection Disaccharide compositional analysis affords information about the types of disaccharides present in each sample and their relative or absolute amounts and relies on the well-characterized disaccharide standards. It is worth noting that the ion pairing alkyl amines persist in the LC–MS system and their use is best confined to a dedicated instrument.

4. CS/DS Disaccharide Preparation 4.1. Materials and solutions for CS/DS disaccharide preparation 1. Chondroitin ABC lyase (EC 4.2.2.4) from Proteus vulgaris (Seikagaku, Japan; or Associates of Cape Cod, East Falmouth, MA); 10 mU/ 1 mg of CS (Linhardt, 2001). 2. A controlled temperature bath or dry block. 3. Centrifuge capable of achieving 13,000 g. 4. Digestion buffer (optional): 5 mM Tris, 6 mM sodium acetate buffer, pH 8 (adjusted with HCl). Distilled water can be used instead of the


85

buffer for the digestion, especially for low amounts of CS, to avoid buffer salt interference during the LC separation. 5. Centrifugal filter with a 30,000 MWCO membrane (e.g., Millipore Ultracel YM-30, cat. # 42410).

4.2. Method for CS/DS disaccharide preparation 1. Purified, desalted PG or pG sample is reconstituted in distilled water or in the digestion buffer; and an aliquot of solution containing an appropriate amount of the chondroitin ABC lyase (approximately 10 mU/mg substrate) is added to the sample. 2. The digestion is allowed to proceed overnight at 37 C, after which the lyase is inactivated by heating the digestion for 5 min in a boiling water bath. 3. The disaccharides are separated from the enzyme and the non-CS polysaccharides using a centrifugal filter with 30K MWCO membrane. 4. The centrifugal filter flow-through containing the CS disaccharides is directly amenable for LC–MS analysis. However, if the sample volume is significantly greater than the volume necessary for the LC–MS analysis (>20 mL), the volume can be reduced by lyophilizing.

5. ESI IP RP LC MS Analysis of CS/DS Disaccharides 5.1. Materials for CS/DS disaccharide analysis 1. CS/DS 4,5-unsaturated disaccharide standards (Fig. 3.3) are available from Seikagaku (Associates of Cape Cod, East Falmouth, MA) and from Iduron, Manchester, UK. a. DUA-GalNAc b. DUA-GalNAc4S c. DUA-GalNAc6S d. DUA2S-GalNAc e. DUA2S-GalNAc4S f. DUA2S-GalNAc6S g. DUA-GalNAc4S6S h. DUA2S-GalNAc4S6S 2. Ion-pairing reagent, n-hexylamine (HXA, Sigma, St. Louis, MO). 3. Organic modifier, 1,1,1,3,3,3-hexafluoroisopropanol (HFIP, Sigma, St. Louis, MO). 4. Acquity UPLC bridged ethyl hybrid (BEH) C18 column, 2.1 150 mm, 1.7 mm and Acquity BEH C18 VanGuard pre-column, 2.1 5 mm, 1.7 mm (Waters). The use of silica-supported C18 column packing is not

86


OY

OX HO

O

O

O

OZ

OH NHAc

HO2C Shorthand notation

X

Y

Z

0S

H

H

H

379.1

481.2

[M+HXA+H]

2

ΔUA-GalNAc ΔUA2S-GalNAc

2S

SO3H

H

H

459.1

662.2

[M+2HXA+H]+

3

ΔUA-GalNAc6S

6S

H

H

SO3H

459.1

662.2

[M+2HXA+H]

4

ΔUA-GalNAc4S

4S

H

SO3H

H

459.1

662.2

[M+2HXA+H]+

5

ΔUA2S-GalNAc6S

2S6S

SO3H

H

SO3H

539.0

843.3

[M+3HXA+H]+

6

ΔUA2S-GalNAc4S

2S4S

SO3H SO3H

H

539.0

843.3

[M+3HXA+H]+

7

ΔUA-GalNAc4S6S

4S6S

SO3H

SO3H

539.0

843.3

[M+3HXA+H]+

8

ΔUA2S-GalNAc4S6S

2S4S6S

SO3H

619.0

1024.4

[M+4HXA+H]+

CS disaccharides in order of elution 1

a

H SO3H SO3H

Mmono Observed m/z

Major ion +

+

a

Hyaluronan is completely comprised of an isobaric disaccharide ΔUA-GlcNAc having the same 1→3 linkage.

Figure 3.3 Structures, monoisotopic masses, and observed m/z of eight CS/DS disaccharide standards.

recommended for this method because of a limited compatibility of the silica packing with basic mobile phases (pH > 8) at elevated temperatures. 5. 0.2 mm membrane filters for mobile phase filtration (Millipore prod. # JGWP04700 or similar). 6. Glass vials, small volume inserts, and screw caps with silicone/PTFE septa for Agilent 1200 autosampler (MicroSolv, Eatontown, NJ; cat. # 9502S-0CV, 95001-04N, and 9502S-10C-B). 7. Access to an LC–MS instrument equipped with a UV detector and column heater and capable of supporting 100 mL/min flow rate (Agilent 1100 LC–MSD with ion trap mass analyzer and diode array UV detector is used in our laboratory).

5.2. Solutions for CS/DS disaccharide analysis 1. Mobile phase A: 15 mM HXA, 100 mM HFIP in HPLC-grade water. 2. Mobile phase B: 15 mM HXA, 100 mM HFIP, 75% acetonitrile in HPLC-grade water. 3. CS/DS disaccharide standard mixture in HPLC-grade water, 2 mL per injection, containing 10 ng of each disaccharide standard (40 ng/mL total disaccharide concentration). 4. CS/DS digest: >20 ng disaccharides in >5 mL of digestion buffer or HPLC-grade water. Injection volume of 5 mL requires a >5 mL sample volume to avoid introducing air into the column. The requirement for the additional sample volume depends on the autosampler needle off-set and the shape of the vial insert.


87

5.3. Method for CS/DS disaccharide analysis 1. Mobile phase solutions are filtered through 0.2 mm filters. Due to their small particle size, UPLC columns are easily blocked by particulates in the mobile phase. 2. The column temperature is set to 45 C, and the column is allowed to equilibrate with 100% A at 100 mL/min. 3. The UV detector is set to record absorbance at 232 nm. 4. An LC method consisting of a 10-min isocratic segment of 0% B, and a linear gradient segment of 0–50% B over 10–40 min is created. 5. The sample injection volume is set at 5 mL (2 mL for the standards mixture). If a series of disaccharide samples are introduced through an autosampler, an appropriate injection sequence is created including 1 injection of the standards for every 8–10 sample injections to monitor the MS detection sensitivity. 6. MS parameters are set as follows: positive ionization mode, skimmer 40 V, capillary exit 40 V, source temperature 350 C, drying gas (N2) 8 L/min, nebulizing gas (N2) 40 psi. An example of a data set obtained during CS/DS disaccharide analysis is shown in Figs. 3.4 and 3.5. Under the experimental conditions described here, the disaccharides elute in order of increasing number of sulfo groups (Fig. 3.3). The total ion chromatogram (TIC) is a plot of total ion signal as a function of time (Fig. 3.4A). The extracted ion chromatogram (EIC) is a plot of selected ion’s signal as a function of time, and in Fig. 3.4B, the ion signals at m/z 481, 662, 843, and 1024 are extracted from the TIC. The TIC, the EIC, or the UV absorbance chromatogram can be used for constructing the disaccharide compositional profile of a sample in which peak area of an individual disaccharide is a percentage of the sum of peak areas for all detected disaccharides (S/N > 3). In the positive-ion mass spectra, major peaks are singly charged and correspond to the HXA adducts of disaccharides with one HXA per acidic site in the disaccharide (Fig. 3.5). Minor peaks represent various Na/H/ HXA exchange products. Saccharide mass spectra acquired under the described experimental conditions exhibit a characteristic pattern of peaks separated by 79 mass units which can be attributed to the HXA/Na exchange. The pattern can be used as an aid in assigning unusual saccharide peaks, for example, trisaccharides. Absolute quantification is best achieved by constructing a standard curve based on several dilutions of the disaccharide standards and using the resulting linear relationship for calculating the amounts of disaccharides in the sample. Our instrument affords good linearity in both UV absorbance and MS signal intensity (R2 0.97) with standard mixtures containing 20–100 mg/mL total CS/DS disaccharides.

88


Relative intensity %

A

1

5 67

3 2 4

8

0 3

100 EIC


B

100 TIC

5

2 4

6 7 8

1

0 50 UV

Abs 232 nm (mAU)

C

8 1 5 6

7

23 4

0 0

10

30

40

Retention time (min)

Figure 3.4 Elution profiles of CS/DS disaccharides separated by IP RP HPLC: (A) total ion chromatogram, (B) extracted ion chromatogram for m/z 481, 662, 843, and 1024, and (C) absorbance trace at 232 nm.

6. HS Disaccharide Preparation 6.1. Materials and solutions for HS disaccharide preparation 1. Heparin lyase I (EC 4.2.2.7) from Flavobacterium heparinum, heparin lyase II (no EC number) from F. heparinum, and heparin lyase III (EC 4.2.2.8) from F. heparinum can be obtained from Seikagaku, Japan (Associates of Cape Cod, East Falmouth, MA). A 4-mU amount of each heparin lyase (12 mU total) is sufficient to depolymerize 1 mg of HS (Linhardt, 2001). 2. A controlled temperature bath or dry block.

89


700

662.3 685.1

803.0 825.0

800

764.2

Δ79

700

800

500

D

[M+4HXA+H]+

600

700

800

700

Δ79

866.1

787.0

707.8 729.9

641.1

945.3

Δ79

561.2

520.4

464.3 481.3

583.2

[M+2HXA+H]+

1000

900 m/z

662.2

m/z

B

[M+3HXA+H]+

800

m/z

1024.4

600

781.1

601.9

500

710.1

Δ79 534.4

492.3

436.2

464.3

513.3

843.3

C

[M+HXA+H]+

725.9 742.2

481.2

A

Δ79

Δ79

900

1000

m/z

Figure 3.5 Positive-ion ESI mass spectra of CS/DS disaccharides: (A) unsulfated disaccharide, (B) monosulfated disaccharide, (C) disulfated disaccharide, and (D) trisulfated disaccharide. Characteristic loss of 79 mass units can be attributed to HXA/Na exchange, ( 101 þ 22).

3. Centrifuge capable of achieving 13,000 g. 4. Digestion buffer (optional): 10 mM sodium phosphate buffer, pH 7.4 (adjusted with phosphoric acid). Distilled water can be used instead of the buffer for the digestion, especially for low amounts of HS, to avoid buffer salt interference during the LC separation. 5. Centrifugal filter with a 30,000 MWCO membrane (e.g., Millipore Ultracel YM-30, cat. # 42410).

6.2. Method for HS disaccharide preparation 1. Purified, desalted PG or pG sample is reconstituted in distilled water or the digestion buffer; and an aliquot of solution containing an appropriate amount of the heparin lyases I, II, and III (approximately 12 mU/mg substrate) is added to the sample. 2. The digestion is allowed to proceed overnight at 35 C, after which the lyases are inactivated by heating the digestion for 5 min in a boiling water bath.

90


3. The disaccharides are separated from the enzymes and the non-HS polysaccharides using a centrifugal filter with 30,000 MWCO membrane. 4. The centrifugal filter flow-through containing the HS disaccharides is directly amenable for LC–MS analysis. However, if the sample volume is significantly greater than the volume necessary for the LC–MS analysis (>8 mL), the volume can be reduced by lyophilizing.

7. ESI IP RP LC MS Analysis of HS Disaccharides 7.1. Materials for HS disaccharide analysis 1. HS 4,5-unsaturated disaccharide standards (Fig. 3.6) are available from Seikagaku, Japan (Associates of Cape Cod, East Falmouth, MA) and from Iduron, Manchester, UK. a. DUA-GlcNAc b. DUA-GlcNS c. DUA-GlcNAc6S d. DUA2S-GlcNAc e. DUA2S-GlcNS f. DUA-GlcNS6S g. DUA2S-GlcNAc6S h. DUA2S-GlcNS6S 2. Ion-pairing reagent, tributylamine (TrBA). 3. Ammonium acetate(NH4OAc). 4. Acetic acid for adjusting pH. 5. Zorbax SB-C18 column (Agilent Technologies), 0.5 250 mm, 5 mm. The recommended pH range for this column is 1–8. 6. A syringe pump, such as Harvard Apparatus Pump 11 Pico Plus or similar, and a 1-mL glass syringe (e.g., 1-mL, blunt tip, precision glass syringe, National Scientific product # NS600001) for delivering 5 mL/min postcolumn flow of acetonitrile. We have found that the addition of acetonitrile aids in solvent evaporation and analyte ionization, dramatically improving the quality of the resulting mass spectra. 7. 0.2 mm membrane filters for mobile phase filtration (Millipore prod. # JGWP04700 or similar). 8. Glass vials, small volume inserts, and screw caps with silicone/PTFE septa for Agilent 1200 autosampler (MicroSolv, Eatontown, NJ; cat. # 9502S-0CV, 95001-04N, and 9502S-10C-B). 9. Access to an LC–MS instrument equipped with a UV detector and capable of supporting 10 mL/min flow rate (An Agilent 1100 LC– MSD with ion trap mass analyzer and diode array UV detector is used in our laboratory).

91


OZ OX O

HO

O O

OH

HO NHY

HO2C

HS disaccharides in order of elution

Shorthand notation

X

Y

Z

Mmono

Observed m/z Major ion

1

ΔUA-GlcNAc

0S

H

Ac

H

379.1

377.8

[M-H]-

2

ΔUA-GlcNS

NS

H

SO3H

H

417.1

415.8

[M-H]-

3

ΔUA-GlcNAc6S

6S

H

Ac

SO3H

459.1

457.7

[M-H]-

4

ΔUA2S-GlcNAc

2S

SO3H

Ac

H

459.1

457.7

[M-H]-

5

ΔUA-GlcNS6S

NS6S

497.0

495.6

[M-H]-

6

ΔUA2S-GlcNS

SO3H

SO3H

2SNS

7

ΔUA2S-GlcNAc6S

2S6S

8

ΔUA2S-GlcNS6S

2SNS6S

H

SO3H SO3H SO3H

Ac

SO3H SO3H

H

497.0

495.6

[M-H]-

SO3H

539.0

537.7

[M-H]-

SO3H

577.0

575.6

[M-H]-

Figure 3.6 Structures, monoisotopic masses, and observed m/z of eight HS disaccharide standards.

7.2. Solutions for HS disaccharide analysis 1. Mobile phase C: 12 mM TrBA, 38 mM NH4OAc, 15% acetonitrile in HPLC-grade water, pH 6.5 adjusted with acetic acid. 2. Mobile phase D: 12 mM TrBA, 38 mM NH4OAc, 65% acetonitrile in HPLC-grade water, pH 6.5 adjusted with acetic acid. 3. HS disaccharide standard mixture, 2 mL containing 50 ng of each disaccharide standard in HPLC-grade water (200 ng/mL total disaccharide concentration). The sensitivity of the HS disaccharide analysis is lower than that of the CS/DS disaccharides. 4. HS digest: 40 ng disaccharides in >2 mL of digestion buffer or HPLCgrade water. Injection volume of 2 mL requires a >2 mL sample volume to avoid introducing air into the column. The requirement for the additional sample volume depends on the autosampler needle off-set and the shape of the vial insert. Lower than 40 ng amounts of HS disaccharides can be detected, but may result in a noisy UV baseline.

7.3. Method for HS disaccharide analysis 1. Mobile phase solutions are filtered through the 0.2 mm filters. 2. The column is allowed to equilibrate with 100% C at 10 mL/min at room temperature. 3. The UV detector is set to record absorbance at 232 nm. 4. A two-segment LC gradient is programmed as follows: 0% D, 0–20 min; 0–50% D, 20–25 min.

92


5. The 1-mL Hamilton syringe is filled with acetonitrile, connected to the postcolumn part of the LC system through a T-connector, positioned in the syringe pump, and the pump flow rate is set to 5 mL/min. 6. The sample injection volume is set to 2 mL. If a series of disaccharide samples are introduced through an autosampler, an appropriate injection sequence must be created which includes a standard disaccharide mixture injection after every 8–10 sample injections to periodically monitor the MS detection sensitivity. 7. MS parameters are set as follows: negative ionization mode, skimmer 40 V, capillary exit 40 V, source temperature 325 C, drying gas (N2) 5 L/min, nebulizing gas (N2) 20 psi. The data obtained during HS disaccharide analysis in the negative ion mode is straightforward to interpret because the disaccharides appear as [MH] ions (Fig. 3.6). An example of elution profile of HS disaccharide standards is shown in Fig. 3.7. The EIC was plotted based on ion signals at m/z 378, 416, 458, 496, 538, and 576. The TIC recorded using 50 ng of each HS disaccharide standard (400 ng total amount) exhibits S/N < 3 (not shown) and therefore is not usable for the relative quantification analysis. The EIC or the UV absorbance chromatogram can be used for constructing the disaccharide compositional profile of the sample as well as for absolute quantification, provided that a linear relationship between a disaccharide signal and its amount is established using the appropriate standards.

8. Direct Infusion ESI FTMS Analysis of Oligosaccharides and Polysaccharides Separated by Preparative Continuous-Elution PAGE High resolving power and mass accuracy of FTMS are necessary for the molecular weight analysis of GAG polysaccharides and large oligosaccharides due to the formation of multiply charged ions in ESI of these polyanions. As is true for any type of analyte, the sample for MS analysis must be of highest possible purity and homogeneity to obtain a good quality mass spectrum. This requirement is especially important for the GAGs since they produce low ion yields, have a high propensity to form adducts with cations present in solution, and usually exhibit more than one charge state in the ESI mass spectra. Separation of highly polydisperse GAG mixtures can be achieved off-line using preparative PAGE; and the resulting components can be purified from buffer salts and analyzed by ESI FTMS. This method has proven useful in our laboratory for the preparation of heparin oligosaccharide MW markers (Laremore et al., 2010) and heparosan (K5 polysaccharide) MW markers (Ly et al., 2010) as well as in the MW analysis of the proteoglycan bikunin GAG chains. Urinary bikunin is a 16 kDa PG modified

93



A 100

3 EIC

8 7 4 1

2 5

6

0

B

1 250 UV Abs 232 nm (mAU)

8 3 2

7 4

5

6

0 0

10

20 30 Retention time (min)

40

Figure 3.7 Elution profiles of HS disaccharides separated by IP RP HPLC: (A) Extracted negative-ion chromatogram for m/z 378, 416, 458, 496, 538, and 576, and (B) absorbance trace at 232 nm.

with a 5–7 kDa GAG chain on Ser10 and a 2 kDa N-glycan on Asn45 (Chi et al., 2008; Enghild et al., 1999). Bikunin GAG is a CS-A type polysaccharide comprised by GlcA-GalNAc4S and GlcA-GalNAc0S disaccharides (Capon et al., 2003; Enghild et al., 1991, 1999; Fries and Kaczmarczyk, 2003). A method described here is used in our laboratory for MW analysis of intact GAG released from the urinary bikunin PG by b-elimination under reducing conditions. The analytical workflow consists of (1) separation of the GAG mixture using preparative CE PAGE, (2) analysis of gel-eluted fractions by analytical PAGE, and (3) purification and ESI FTMS MW analysis of fractions of interest.

9. Preparation of Bikunin GAG 9.1. Materials and solutions for GAG release by b-elimination under reducing conditions 1. Bikunin proteoglycan, 1 mg (Mochida Pharmaceuticals, Tokyo, Japan). To obtain a sufficient amount of each purified GAG fraction for the ESI FTMS analysis, ca. 300 mg GAG mixture is required for CE PAGE

94

2. 3. 4. 5. 6. 7. 8. 9.


separation. Since the GAG is approximately 30% of the bikunin PG mass, the use of such large quantity of the PG as 1 mg is justified. Amicon Ultra 10,000 MWCO centrifugal filter for volumes 1.0 TB)

PC for data analysis

Figure 7.1 An overview of the lectin microarray system. The system is composed of an evanescent-field fluorescence scanner GlycoStationTM Reader1200, a hard disk for data (> 1 TB) storage, a PC for data analysis, and a lectin microarray chip. Commercialized lectin microarray LecChipTM is purchased from GP Bioscience, Inc.

168

Atsushi Kuno et al.

using both commercial tissue arrays and clinical specimens. The protocol for scratching tissue sections has been updated since the original report published in 2008, and is included here in detail.

2. Differential Analysis of Glycoproteins Derived from One-Dot Sections of Tissue Microarray Recent advances in tissue array technology enable us to use a variety of tissue sections from patients with different diseases (Kononen et al., 1998). All specimens on commercially available tissue array have received approval from the Ethics Committee, and so tissue array has been frequently utilized for histochemical studies, expression analyses by q-PCR, and so on. We describe here a method of differential analysis of tissue glycoproteins extracted from one-dot sections of tissue arrays (Fig. 7.2).

2.1. Methods for preparation of Cy3-labeled glycoprotein from formalin-fixed tissue sections 1. Deparaffinization: Formalin-fixed paraffin-embedded tissue microarrays are deparaffinized by incubating in xylene (2 10 min) and with sequential incubations in (i) 100% EtOH for 10 min, (ii) 95% EtOH for 10 min, (iii) 90% EtOH for 5 min, (iv) 80% EtOH for 5 min, (v) 70% EtOH for 5 min, and (vi) MilliQ water for 5 min. The tissue microarrays are then dried at room temperature (RT). Comment: Contrast of each specimen on the tissue array was enhanced using hematoxylin and eosin (HE) staining before the differential analysis. Two or more sequential tissue array sections were therefore prepared. In our previous report (Matsuda et al., 2008), two sequential tissue arrays comprising three dots for each specimen was purchased from Cybrdi, Inc. (Gaithersburg, MD), with one dot used for lectin staining, and the other two dots used for lectin microarray analysis (Fig. 7.3). 2. Tissue scratching: Each specimen on the array glass slide is viewed using a microscope and scratched using a needle (gauge size: 21 gauge) connected to a disposable syringe. The scratched tissue fragments are collected into a 1.5-mL microtube containing 200 mL of 10 mM citrate buffer (pH 6.0) and incubated for 1 h at 95 C (antigen retrieval). The heat treated solution is centrifuged at 20,000g for 5 min, and the supernatant discarded. Comment: We recently adapted this protocol, using a disposable scalpel (AS ONE Co., Osaka, Japan) in place of the needle because it

169

Tissue Glycan Profiling using Lectin Array

I. Preparation of Cy3-labeled glycoprotein

1. Deparaffinization

2. One-dot tissue scratching

3. Protein extraction

4. Cy3labeling

II. Lectin microarray analysis and data processing

Normal vs. Tumor 1. Sample injection Binding reaction 2. Scanning

3. Data-processing

III. Selection of the best lectin probe P < 0.001

P = 0.002 60 40

40 30 20

20

10

0

0 N T N T 1. Statistical analysis

2. Lectin staining

Figure 7.2 A scheme for manipulation of lectin microarray analysis targeting tissue arrays.

can scratch one-dot tissue with a single stroke. Fragmentation of deparaffinized tissue is prevented by spraying distilled water on the edge of the needle or scalpel. 3. Protein extraction: The pellet is rinsed with 200 mL of PBS, centrifuged at 20,000g for 5 min, and the subsequent pellet resuspended in 20 mL of PBS containing 0.5% Nonidet P-40 (NP-40) with gentle sonication. The suspension is stored on ice for 1 h, and the cell debris is pulled down by centrifugation at 20,000g for 5 min. The supernatant is transferred to a new tube (designated detergent-solubilized glycoprotein extract).

170

Atsushi Kuno et al.

For HE staining

Qualification of tissue sections

Bad

Good Disposable scalpel

For lectin assay Two deparaffinized colon tissue arrays with 63 dots (three dots per one specimen) 1.5 mm in diameter, 5 mm thickness

Figure 7.3

Scratched for lectin microarray Left for lectin histochemistry

How to scratch one-dot tissue section from a colon tissue array.

4. Cy3 labeling: Ten micrograms of Cy3-SE (GE Healthcare, UK) is added to the detergent-solubilized glycoprotein extract and then incubated at RT for 1 h in the dark. The reaction product is applied onto a spin column (BioRad, Hercules, CA) containing Sephadex-G25 (GE Healthcare) to remove the excess fluorescent reagent. The product is eluted by adding 50 mL of probing buffer (1.0% Triton X-100 and 500 mM glycine in Tris-buffered saline (TBS)) to the column and centrifuging at 1500g for 1 min. The volume of the eluant is adjusted to 200 mL with the probing buffer and incubated for 2 h at RT to inactivate any residual fluorescent reagent.

2.2. Methods for lectin microarray analysis and data processing 1. Sample injection and binding reaction: The Cy3-labeled protein solution is diluted with the probing buffer to an appropriate concentration. The analyte solution is applied to the lectin microarray glass slide produced as described previously (Uchiyama et al., 2006, 2008). The glass slide is incubated with gentle rotation at 20 C for over 12 h. Comment: The density and species of the tissue array dots are not always identical to one another, causing significant deviation in signal intensities among the analytes (Fig. 7.4). Thus, various concentrations of Cy3-labeled protein solution should be simultaneously subjected to lectin microarray analysis to obtain results with appropriate signal intensities.

171


A

1/2

1/4

1/8

B

Figure 7.4 Glycan profiles of tissue dots of lung adenocarcinoma (A) and normal lung (B). (Left) HE staining images of tissue dots used in this study. (Right) Scan images of Cy3-labeled glycoprotein solutions. A series of dilute protein solutions were subjected to the lectin microarray. Among them, 1/8 dilution solution of lung adenocarcinoma and 1/2 dilution solution of normal lung were selected and used for comparative analysis.

2. Scanning: After the binding reaction, the glass slide is washed three times with the probing buffer, and then scanned with GlycoStationTM Reader 1200 (GP Bioscience, Inc.), saving the scanned image as a TIFF file for analysis by Array Pro Analyzer version 4.5 (Media Cybernetics, Inc.), and as a JPEG file. The net intensity value is obtained from the mean of three spots after subtraction of the respective background values. Comment: GlycoStationTM Reader 1200 allows acquisition of a series of scan images with different gain conditions. For best results, we recommend that array scanning is performed with at least five gain conditions. The most suitable dilution should be selected for the following data-processing. 3. Data-processing: Data processing described previously (Kuno et al., 2008) is necessary to substantiate differential glycan profiling using clinical samples. To cover a wider dynamic range of signal intensities, a lectin microarray glass slide should be scanned under two different gain conditions; higher gain to ‘‘rescue’’ weak signals below 1000 (IntH ðlectin f Þ ) and lower gain to ‘‘suppress’’ excessively strong signals over 40,000 (IntLðlectin dÞ ). Appropriate lectins are selected by ‘‘merging’’ lectins with the signal intensities ranging from 1000 to 40,000 under both higher and lower gain conditions. A Factor (F) value is determined as the average of higher/lower ratios calculated for individual merging lectins by Eq. (7.1):

172

Atsushi Kuno et al.

L F ¼ AverageðIntH i =Inti Þ

ð7:1Þ

The over-range intensities (>40,000) obtained under the higher gain condition (e.g., IntH ðlectin cÞ ) are replaced with theoretical intensities (IntTðlectin cÞ ) by Eq. (7.2): IntTðlectin cÞ ¼ IntLðlectin cÞ F

ð7:2Þ

For other lectins with no over-range using the higher gain condition, signal intensities are used with no modification. The merged data is normalized relative to the strongest intensity among the positive-spots under the given conditions (this type of normalization is designated ‘‘max-normalization’’). However, normalization procedures vary depending on the purpose of the studies, for example, normalization by the mean of signal intensities of all lectins and by a particular lectin (see also Chapter 8).

2.3. Methods for selection of the best lectin probe 1. Statistical analysis: To select the most diagnostic lectin, all of the processed data obtained above are statistically analyzed between two independent specimens, for example, cancer and non-cancer. Since data distribution of every lectin signal is probably indicated as asymmetric or symmetric with large-tails, nonparametric statistics, such as chi-square test and Mann–Whitney U-test, are applied. 2. Lectin staining: The deparaffinized tissue array for lectin staining is incubated in 10 mM citrate buffer (pH 6.0), and then autoclaved at 121 C for 20 min (antigen retrieval). After cooling and washing the slide with PBS, endogenous peroxidase is blocked by incubating the slide in methanol containing 0.3% H2O2 at RT for 10 min. The tissue section is rinsed with PBS, blocked with PBS containing 1% BSA, and then incubated for 1 h at RT with 2 mg/mL of biotinylated lectin in 10 mM 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES). Detection of the binding lectin is made with Histofine Simple Stain MAX-PO (Nichirei Co., Tokyo, Japan). A more practical example of differential profiling is described using sections from colon cancer patients (16 cases of grade III) and normal colon tissues (12 cases) obtained from Cybrdi, Inc. (Frederick, MD). As a result of differential analysis using the normalized data (Fig. 7.5A), it was found that 11 lectins showed significant alterations in the mean binding signals. Disialyl-T binder MAH, a2,6-Sia binders SSA, SNA and TJA-I, an a1,6-branched mannose

173


A

C

Normal mean

Cancer mean

29.17 18.78 39.96 19.38 13.83 24.46 17.26 22.17 17.33 9.42 2.76

52.17 37.75 77.12 32.38 24.81 11.48 6.06 11.03 7.5 3.53 5.47

SNA SSA TJA-I RCA120 HHL BPL WFA ACA HPA DBA MAH B

SSA

SNA 60

90 80 70 60 50 40 30 20 10 0

50 40 30 20 10

N

G III HHL

50 40 30 20 10 0

N

G III

N

120 100 80 60 40 20 0

G III

40 30 20 10 0

WFA

G III ACA

50 40

30

30

20

20

10

10

0

0

–10

G III

N

G III

40

N

RCA120 50

N

BPL

50 40 30 20 10 0 –10

TJA-I

N

G III

N

G III

Figure 7.5 Differential glycan analysis between colon normal control and adenocarcinoma (grade III) tissues. Twenty-one and 23 cases of individual patients were used for normal colon and adenocarcinoma, respectively. Acquired signal patterns were normalized against a lectin showing the maximum intensity in each analysis, that is, UDA for colon adenocarcinoma and STL for normal control colon, respectively. (A) Differential glycan analysis between adenocarcinoma and normal controls in a colon tissue array using only data for 11 lectins are shown. (B) Box and whisker plots representation of the data obtained by the selected eight lectins, which shows the average (dot), the 75th and 25th percentiles (the top and bottom of the box, respectively), the median (line through the middle of the box), and the maximum and minimum (the top and bottom of the whiskers, respectively). (C) WFA staining of colon normal tissue dot.

174

Atsushi Kuno et al.

binder HHL, and an N-acetyl-lactosamine binder RCA120 showed apparent increase in binding to adenocarcinoma tissues (grade III), while a T/Tnantigen binder BPL, a core 1 binder ACA, a-GalNAc binders HPA and DBA, and an a/b-GalNAc binder WFA showed an apparent decrease in binding. In the next step, signal variance was statistically analyzed using a two-sided w2 test. As shown in Fig. 7.5B, eight lectins, that is, SSA, SNA, TJAI, HHL, RCA120, BPL, ACA, and WFA yielded good statistical scores (P < 0.001). Among them, WFA showed the best score for discrimination between both normal and grades IþII (P < 0.0001) and between normal and grade III. This observation indicates that glycan epitopes recognized by WFA are drastically reduced in carcinogenesis, even in well-differentiated tumor cells. This phenomenon was confirmed by a histochemical approach employing biotin-labeled WFA and normal and grades III specimens, where WFA staining was observed on mucin-producing goblet-like cells in the normal control specimens (Fig. 7.5C), but not in grade III specimens. Substantial disappearance of the goblet-like cells in grade III specimens was noted, in accordance with the tumor progression.

3. Differential Glycan Analysis Between Cancer Lesions and Normal Regions in the Same Tissue Section Archival formalin-fixed, paraffin-embedded surgical specimens are ideal for differential profiling using the above lectin microarray as they allow comparisons of cancerous and normal regions of tissue. Thus, multiple (> 40) lectins can be compared (bottom of Fig. 7.6C), eliminating possible dispersion attributable to individual differences. An actual procedure to dissect specific regions manually from tissue sections is described below (in case there is no microdissection machine).

3.1. Method for dissecting specific regions in tissue sections Two sequential tissue sections are prepared: one is for HE staining (top of Fig. 7.6A) and the other for tissue dissection (bottom of Fig. 7.6A). Cancer and normal regions are first assigned from the HE-stained tissue specimen using a microscope (we recommend to ask opinions from expert pathologists for rigorous clinical assignment). The relevant tissue fragments (corresponding to 1.0 mm2 and 5 mm thickness) are scratched from the sections using a needle (gauge size: 21 gauge) under a microscope and processed as described above for preparation of Cy3-labeled protein solution.

175


B

A

CC

N

CC

Lectins

N

CC N

A B C D E F G H I J K L M N O P Q R S T U V W

Normalized signal C

D

WFA

30.00

N

15.00

0.00

N

CC CC

15.00 10.00 5.00 0.00

N

CC

Figure 7.6 Differential glycan analysis between cholangiocarcinoma (CC) and normal BDE (N). (A) Regions of cholangiocarcinoma (CC) and normal BDE (N) in the same tissue section and scratched are shown. Two sequential tissue sections were prepared for HE staining and tissue dissection. The relevant tissue fragments (corresponding to 1.0 mm2 and 5 mm thickness) were scratched from the glass slide using a needle (gauge size: 21 gauge) under a microscope. (B) A vertical bar graph representation of the data obtained for 23 representative lectins and 10 specimens. (C) A box and whisker plot (top) and a dot graph (bottom) representation of the data obtained for WFA. The data show significant differences in signals between CC and normal BDE lesions in the same specimens. (D) Fluorescence staining with WFA in CC and normal (N) tissues. WFA was visualized by Alexa 488 (green fluorescence). All panels are as viewed by fluorescence microscopy.

176

Atsushi Kuno et al.

An example was recently reported using cancerous lesions and normal bile duct epithelium (BDE) from the CC specimen each derived from the same patient (14 and 10 cases with and without stones, respectively). The results obtained for the cases of patients without stones are summarized in Fig. 7.6. To identify lectins specific for CC, we analyzed 10 cases of paraffinembedded, formalin-fixed CC tissue sections, which included both cancerous lesions and normal BDE. We obtained intense signals of 23 lectin spots on 43lectin microarray and found significant (P < 0.0001) differences in four lectins (Fig. 7.6B), with the highest score obtained for WFA (Fig. 7.6C). To confirm the result of lectin microarray, histochemical lectin staining was performed to visualize the expression of WFA-reactive glycans using biotinylated WFA. Significant WFA staining was detected with high frequency in CC (bottom of Fig. 7.6D), but with much less frequency in normal BDE (top of Fig. 7.6D). Taken together, we have shown for the first time specific expression of WFA-reactive glycans in CC.

4. (Optional)Glycan Profiling of a Target Glycoprotein Extracted from Tissue Specimens If one has already identified a target glycoprotein, specific tissue regions in specimens can be immunohistochemically assigned with assured dissection (initial step in Fig. 7.7). For focused differential glycan analysis, a target glycoprotein is enriched from the protein extract by immunoprecipitation (Fig. 7.7) and lectin microarray analysis is subsequently performed, in this case, by a modified procedure designated the ‘‘antibody-overlay detection’’ method (Kuno et al., 2009).

4.1. Differential glycan analysis of a fixed target glycoprotein in tissue sections by antibody-assisted lectin profiling One of the sequential tissue sections is immunostained using an antibody against the glycoprotein of interest. The area on the counterpart sections is dissected using a disposable scalpel with the aid of the microscope. After extraction of the protein from the obtained tissue fragments, a target protein is enriched preferably using antibody-immobilized resin. The enriched protein solution is directly applied to lectin microarray, and the array is incubated at 20 C for 12 h. After the binding reaction, 20 mg of human serum polyclonal IgG (Sigma) is added to the array followed by 30-min incubation to mask nonspecific binding sites. The reaction solution is discarded and the glass slide is washed three times with PBS containing 1% Triton X-100 (PBSTx). Biotinylated antibody solution (60 mL) dissolved

177


Immuno-staining

Sample injection and binding reaction

Tissue scratch and protein extraction

Crude protein mixture

Enriched protein Immunoprecipitation

Figure 7.7 Differential analysis of podoplanin enriched from tissue specimens of seminoma xenocraft. After immunostaining with an antipodoplanin antibody, a podoplanin-positive area on the counterpart sections was selectively dissected (excised volume estimated to be approximately 3.6 mm3). hPod was enriched with antipodoplanin antibody-immobilized Sepharose, and was analyzed by antibody-overlay lectin microarray.

in PBSTx is applied to the array, and then the reaction is allowed to proceed at 20 C for 1 h. After washing three times with PBSTx, 60 mL of Cy3-labeled streptavidin (GE Healthcare) solution in PBSTx is added, and the array is incubated at 20 C for 30 min. The glass slide is rinsed with PBSTx, and array scanning and data analyses are performed as described above.

ACKNOWLEDGMENTS We thank Nobuko Ohmichi and Sachiko Unno for manipulating lectin microarray analysis. Thanks also to Yoshiko Kubo and Jinko Murakami for supply of lectin microarray. This work was supported by New Energy and Industrial Technology Development Organization (NEDO) in Japan.

REFERENCES Brooks, S. A., and Leathem, A. J. (1991). Prediction of lymph node involvement in breast cancer by detection of altered glycosylation in the primary tumour. Lancet 338, 71–74. Fukuda, M. (1996). Possible roles of tumor-associated carbohydrate antigens. Cancer Res. 56, 2237–2244.

178

Atsushi Kuno et al.

Hirabayashi, J. (2004). Lectin-based structural glycomics: Glycoproteomics and glycan profiling. Glycoconj. J. 21, 35–40. Kakeji, Y., Tsujitani, S., Mori, M., Maehara, Y., and Sugimachi, K. (1991). Helix pomatia agglutinin binding activity is a predictor of survival time for patients with gastric carcinoma. Cancer 68, 2438–2442. Kondo, T., and Hirohashi, S. (2006). Application of highly sensitive fluorescent dyes (CyDye DIGE Fluor saturation dyes) to laser microdissection and two-dimensional difference gel electrophoresis (2D-DIGE) for cancer proteomics. Nat. Protoc. 1, 2940–2956. Kononen, J., Bubendorf, L., Kallioniemi, A., Barlund, M., Schraml, P., Leighton, S., Torhorst, J., Mihatsch, M. J., Sauter, G., and Kallioniemi, O.-P. (1998). Tissue microarrays for high-throughput molecular profiling of tumor specimens. Nat. Med. 4, 844–847. Korekane, H., Shida, K., Murata, K., Ohue, M., Sasaki, Y., Imaoka, S., and Miyamoto, Y. (2007). Evaluation of laser microdessection as a tool in cancer glycomic studies. Biochem. Biophys. Res. Commun. 352, 579–586. Kuno, A., Uchiyama, N., Koseki-Kuno, S., Ebe, Y., Takashima, S., Yamada, Y., and Hirabayashi, J. (2005). Evanescent-field fluorescence-assisted lectin microarray: A new strategy for glycan profiling. Nat. Methods 2, 851–856. Kuno, A., Itakura, Y., Toyoda, M., Takahashi, Y., Yamada, M., Umezawa, A., and Hirabayashi, J. (2008). Development of a data-mining system for differential profiling of cell glycoproteins based on lectin microarray. J. Proteomics Bioinform. 1, 68–72. Kuno, A., Kato, Y., Matsuda, A., Kaneko, M. K., Ito, H., Amano, K., Chiba, Y., Narimatsu, H., and Hirabayashi, J. (2009). Focused differential glycan analysis with the platform antibody-assisted lectin profiling (ALP) for glycan-related biomarker verification. Mol. Cell. Proteomics 8, 99–108. Liu, Y., He, J., Li, C., Benitez, R., Fu, S., Marrero, J., and Lubman, D. M. (2010). Identification and confirmation of biomarkers using an integrated platform for quantitative analysis of glycoproteins and their glycosylations. J. Proteome Res. 9, 798–805. Matsuda, A., Kuno, A., Ishida, H., Kawamoto, T., Shoda, J.-I., and Hirabayashi, J. (2008). Development of an all-in-one technology for glycan profiling targeting formalin-embedded tissue sections. Biochem. Biophys. Res. Commun. 370, 259–263. Matsuda, A., Kuno, A., Kawamoto, T., Matsuzaki, H., Irimura, T., Ikehara, Y., Zen, Y., Nakanuma, Y., Yamamoto, M., Ohkohchi, N., Shoda, J.-I., Hirabayashi, J., et al. (2010). Wisteria floribunda agglutinin-positive MUC1 is a sensitive biliary marker for human intrahepatic cholangiocarcinoma. Hepatology 52, 174–182. Misonou, Y., Shida, K., Korekane, H., Seki, Y., Noura, S., Ohue, M., and Miyamoto, Y. (2009). Comprehensive clinico-glycomic study of 16 colorectal cancer specimens: Elucidation of aberrant glycosylation and its mechanistic causes in colorectal cancer cells. J. Proteome Res. 8, 2990–3005. Narimatsu, H., Sawaki, H., Kuno, A., Kaji, H., Ito, H., and Ikehara, Y. (2010). A strategy for discovery of cancer glyco-biomarkers in serum using newly developed technologies for glycoproteomics. FEBS J. 277, 95–105. Schumacher, U., Higgs, D., Loizidou, M., Pickering, R., Leathem, A., and Taylor, I. (1994). Helix pomatia agglutinin binding is a useful prognostic indicator in colorectal carcinoma. Cancer 74, 3104–3107. Taniguchi, N., Miyoshi, E., Ko, J. H., Ikeda, Y., and Ihara, Y. (1999). Implication of N-acetylglucosaminyltransferases III and V in cancer: Gene regulation and signaling mechanism. Biochim. Biophys. Acta 1455, 287–300. Tho¨m, I., Schult-Kronefeld, O., Burkholder, I., Goern, M., Andritzky, B., Blonski, K., Kugler, C., Edler, L., Bokemeyer, C., Schumacher, U., and Laack, E. (2007). Lectin histochemistry of metastatic adenocarcinomas of the lung. Lung Cancer 56, 391–397.


179

Uchiyama, N., Kuno, A., Koseki-Kuno, S., Ebe, Y., Horio, K., Yamada, M., and Hirabayashi, J. (2006). Development of a lectin microarray based on an evanescentfield fluorescence principle. Methods Enzymol. 415, 341–351. Uchiyama, N., Kuno, A., Tateno, H., Kubo, Y., Mizuno, M., Noguchi, M., and Hirabayashi, J. (2008). Optimization of evanescent-field fluorescence-assisted lectin microarray for high-sensitivity detection of monovalent oligosaccharides and glycoproteins. Proteomics 8, 3042–3050.

C H A P T E R

E I G H T

A Versatile Technology for Cellular Glycomics Using Lectin Microarray Hiroaki Tateno, Atsushi Kuno, Yoko Itakura, and Jun Hirabayashi Contents 1. Introduction 2. Strategy for Systematic Development of Cell Discrimination Procedures Using Lectin Microarray 3. Production of the Lectin Microarray 4. Sample Preparation and Lectin Microarray Hybridization 5. Data Normalization 6. Glycan Profiles of CHO, Lec2, Lec8, and Lec1 7. Unsupervised Clustering and Principal Component Analysis 8. Significance Difference Test 9. Discriminant Analysis 10. Differential Analysis Between CHO and Lec1 11. Validation 12. Concluding Remarks Acknowledgments References

182 182 183 185 185 186 189 190 191 192 193 193 195 195

Abstract All cells in nature are covered with a dense and complex array of glycans. The total glycan repertoire expressed on cells, ‘‘the cellular glycome,’’ varies at every level of biological organization, and in response to intrinsic and extrinsic stimuli. The cellular glycome is often referred to as the ‘‘cell face,’’ which reflects the condition and type of the cell. In other words, cells can be discriminated in detail by characterization of their individual cellular glycome. Based on this concept, we describe our strategy for profiling the cellular glycome using lectin microarray followed by lectin-based cell discrimination using Chinese hamster ovary cells and their glycosylation-defective mutants (Lec1, Lec2, and Lec8) as models. The results add to the understanding and applications of ‘‘Cellular Glycomics.’’ Research Center for Medical Glycoscience, National Institute of Advanced Industrial Science and Technology (AIST), Central 2, Umezono, Ibaraki, Japan Methods in Enzymology, Volume 478 ISSN 0076-6879, DOI: 10.1016/S0076-6879(10)78008-3

#


181

182

Hiroaki Tateno et al.

1. Introduction It is a universal principle that all cells in nature are covered with a dense and complex array of glycans. These cell surface glycans are known to mediate many important biological phenomena, including structural and physical roles as well as specific recognition by both extrinsic and intrinsic lectins (Gagneux and Varki, 1999). Diversity in glycosylation exists at every level of biological organization; between species, and among different tissues, cell types, and molecules within the same organism. Changes in glycosylation also occur in relation to both inner and outer cellular environmental changes such as cell activation, differentiation, inflammation, and malignant transformation. Therefore, the cellular glycome is often referred to as the ‘‘cell face,’’ which reflects cellular conditions. Cells can be discriminated in detail by characterization of their individual cellular glycome. Consistently, many tumor and stem cell markers are glycans such as CA199, SSEA3/4, Tra1-60, and Tra1-81. The complexity and diversity of glycans seem to be greater than that can be explained simply by evolutionary pressure, called the ‘‘Red Queen Effect’’ (Gagneux and Varki, 1999). Before a comprehensive explanation for glycan diversity is possible, full comprehension of the glycan diversity of cellular life is a challenging issue in need of solution. In fact, no information is available on the glycans in many organisms. Although we can presume the importance of changes in cell surface glycans from traces of information obtained by a relatively small set of lectins or antibodies, we still do not know how and to what degree cell surface glycans change. One of the reasons is that there has been no practical technology to survey global changes in the cellular glycome in a highthroughput and sensitive manner. In the previous chapter (Chapter 7), we described development of a lectin microarray for glycan profiling, and its application to the discovery of disease-related biomarkers. Here, we demonstrate the experimental feasibility of lectin microarray, followed by lectin-based cell discrimination procedures without antibodies, in profiling the cellular glycome.

2. Strategy for Systematic Development of Cell Discrimination Procedures Using Lectin Microarray A strategy for development of cell discrimination procedures using a lectin microarray is shown in Fig. 8.1. The strategy comprises four steps: sample preparation (I), lectin microarray (II), statistical analysis (III), and validation (IV). Here, we applied this systematic approach for wild-type

183

Cellular Glycomics Using Lectin Microarray

I: Sample prep.

II: Lectin microarray

Preparation of Cy3-labeled cell membrane fractions or CMRA-labeled live cells

Production of lectin microarray

Lectin microarray analysis

Data normalization

Unsupervised cluster analysis

Principal component analysis III: Statistical analysis Significance difference test

Canonical discriminant analysis

IV: Validation

Cell staining, lectin blotting, etc.

Figure 8.1 A strategy for the systematic development of cell discrimination procedures based on lectin microarray.

(WT) and glycosylation-defective mutants (Lec1, Lec2, and Lec8) of Chinese hamster ovary (CHO) cells used as models. Lec2, Lec8, and Lec1 are deletion mutants of CMP-Sia transporter, UDP-Gal transporter, and MGAT1 (mannosyl (a-1,3-)-glycoprotein-b-1,2-N-acetylglucosaminyltransferase), respectively.

3. Production of the Lectin Microarray The lectin microarray was produced as previously described with minor modifications (Kuno et al., 2005; Uchiyama et al., 2006, 2008). Briefly, lectins were dissolved at a concentration of 0.5 mg/mL in a spotting solution (Matsunami Glass), and spotted onto epoxysilane-coated glass slides (Schott) in triplicate using a noncontact microarray printing robot, MicroSys 4000 (Genomic solutions). The glass slides were incubated at 25 C for 3 h to allow lectin immobilization. The lectin-immobilized

184


glass slides were then washed with probing buffer (25 mM Tris–HCl, pH 7.5, 140 mM NaCl (TBS) containing 2.7 mM KCl, 1 mM CaCl2, 1 mM MnCl2, and 1% Triton X-100), blocked with blocking reagent N102 (NOF Co.) at 20 C for 1 h, and stored in TBS containing 0.02% NaN3 at 4 C until use. The spot quality and reproducibility of the produced microarrays were then checked using a Cy3-labeled 10-mix probe containing 250 mg/mL asialofetuin (Sigma-Aldrich), 25 ng/mL Siaa23Galb1-4GlcNAc-BSA (Dextra), 10 ng/mL Fuca1-2Galb1-3GlcNAcb13Galb1-4Glc-BSA (Dextra), 10 ng/mL bGlcNAc-BSA (Dextra), 10 ng/mL GalNAca1-3(Fuca1-2)Gal-BSA (Dextra), 10 ng/mL Gala1-3Galb14GlcNAc-BSA (Dextra), 10 ng/mL Mana1-3(Mana1-6)Man-BSA (Dextra), 10 ng/mL aFuc-BSA (Dextra), 10 ng/mL aGalNAc-BSA (Dextra), and 10 ng/mL Siaa2-6Galb1-4Glc-BSA (Dextra) in probing buffer. The concentration of each compound was optimized to provide maximum fluorescence without saturation of binding sites of immobilized lectins. A representative image is shown in Fig. 8.2.

Positioning marker (Cy3-BSA)

A LTL

TJAI

GNA

STL

VVA

1

PSA

PHAL

HHL

UDA

DBA

2

LCA

ECA

BSA

PWM

SBA

3

UEAI

RCA120

BSA

Jacalin

GSLI

4

AOL

PHAE

BPL

PNA

PTLI

5

AAL

DSA

TJAII

WFA

MAH

6

MAL

GSLII

EEL

ACA

WGA

7

SNA

NPA

ABA

MPA

GSLI-A4

SSA

ConA

LEL

HPA

GSLI-B4

Rubber sheet

Glass slide

B

Figure 8.2 (A) Lectin spotting pattern. (B) A representative image of lectin microarray data obtained using the Cy3-labeled 10-mix probe for quality check.


185

4. Sample Preparation and Lectin Microarray Hybridization To profile the cellular glycome using lectin microarray (Kuno et al., 2005), two methods are currently available. One is to use cell membrane fractions (Ebe et al., 2006; Kuno et al., 2008) and the other is direct profiling of live cells (Tateno et al., 2007). Here, we applied cell membrane fractions of CHO, Lec2, Lec8, and Lec1 for lectin microarray analysis. Detailed protocols are also available on the website of GP Biosciences (http://www.gpbio.jp/ english/index.html). 1. Cells were cultured in T150 flasks, each cell line in three flasks. On reaching 80% confluence, cells were recovered, washed with PBS three times, and stored at 80 C until use. 2. Hydrophobic fractions were prepared using CelLytic MEM Protein Extraction (Sigma-Aldrich) in accordance with the manufacturer’s procedures with minor modifications. 3. After protein quantification using BCA assay (Thermo Fisher Scientific), hydrophobic fractions were fluorescently labeled with Cy3 MonoReactive dye (GE) and excess Cy3 was removed with Sephadex G-25 desalting columns (GE). 4. After dilution, Cy3-labeled hydrophobic fractions were incubated with lectin microarray at 20 C overnight. 5. After washing with probing buffer, bound fluorescence was scanned using a GlycoStationTM Reader 1200 (GP Biosciences). 6. Data were analyzed with the Array-Pro Analyzer ver.4.5 (Media Cybernetics). To determine the optimal protein concentration, varying concentrations of the hydrophobic fraction (0.016, 0.03, 0.06, 0.125, 0.25, 0.5, and 1 mg/mL) of each cell line were first analyzed by lectin microarray. Fluorescence intensity was increased in a concentration-dependent manner. As the binding curves of some lectins reached saturation at a concentration of 1 mg/mL, we analyzed samples at a concentration of 0.5 mg/mL, which provided maximum fluorescence without saturation of the binding sites of immobilized lectins.

5. Data Normalization The routine application of lectin microarray for profiling the cellular glycome requires establishment of optimized analytical methods to ensure the proper interpretation of the data. Normalization is a common data processing procedure for microarray, which adjusts the data from each microarray to

186


Table 8.1 Normalization methods

Max Mean Particular lectin Median

The signal intensity was multiplied by a normalization factor N for each array, which was calculated by N ¼ 1/m, where m is: The highest signal intensity of all of the lectins on the array The mean of all of the lectins on the array The signal intensity of one selected lectin on the array The median of all of the lectins on the array

account for possible systematic variation in factors such as microarray quality, scanner detection stability, sample preparation reproducibility, and labeling efficiency. Previously, Kuno et al. (2008) developed a gain-merging procedure to increase the dynamic range. Here, we evaluated the effects of four different normalization methods to determine the optimal normalization procedure to process lectin microarray data: ‘‘max,’’ ‘‘mean,’’ ‘‘particular lectin,’’ and ‘‘median’’ (Table 8.1). Hydrophobic fractions from the four different cell lines (CHO, Lec1, Lec2, and Lec8), each in triplicate (total 12 cells), were incubated on a microarray containing 43 lectins. The set of 12 samples was analyzed in quadruplicates using different batches of microarrays that had been printed on different days. The effect of normalization on the reproducibility between replicate data sets was evaluated by examining the coefficients of variation (CV) between replicate experiments. The CV of each lectin (SD divided by average) among the quadruplicate measurements of triplicate samples (total 12 measurements) was calculated for each normalization method. The average CVs of lectins with >3000 signal intensity (approximately 3000 is the lower limit of quantitative detection of fluorescence by the scanner after gain-merging) was compared between the nonnormalized data and each set of normalized data. Since the median of 43 lectins gave ‘‘0’’ in many samples, this normalization method was considered to be unsuitable for the processing of lectin microarray data. Among the three normalization methods, the mean normalization gave significantly lower average CVs (0.18–0.26) in comparison with the nonnormalized data (0.24–0.31) (Fig. 8.3). We therefore adopted the mean normalization to process lectin microarray data in this study. It should be noted, however, that the most adequate normalization procedures vary depending on the purpose of the study.

6. Glycan Profiles of CHO, Lec2, Lec8, and Lec1 Glycan profiles of CHO cells and their glycosylation-defective mutants, Lec2, Lec8, and Lec1, were compared (Fig. 8.4). Representative structures of N- and O-glycans of each cell line are also shown in Fig. 8.4. CHO cells express a broad range of complex-type N-glycans, high-

187

CV average of lectins with >3000 signal intensity


35 30 25 20 15 10 5 0 CHO

Lec1

Lec2

Lec8

Cell Non-normalized Mean-normalized

Max-normalized LEL-normalized

Figure 8.3 Comparison of the effects of normalization procedures. CHO, Lec1, Lec2, and Lec8, each prepared in triplicate, were incubated on a microarray containing 43 lectins. The set of 12 cells was analyzed in quadruplicate using different batches of microarrays. The fluorescence intensity was first processed with the gain-merging procedure (Kuno et al., 2008) and the effect of normalization between replicate data sets was evaluated by examining the coefficients of variation (CV) between the replicate experiments. The CV of each lectin (S.D. divided by average) between the quadruplicate measurements of triplicate samples (total 12 measurements) was calculated for each normalization method. Data are shown as the average CVs of lectins with > 3000 signal intensity after gain-merging.

mannose-type N-glycans, and very few hybrid-type N-glycans (North et al., 2010). Consistently, CHO cells bound to asialo complex-type N-glycan binders (PHAL, ECA, and RCA120) (Itakura et al., 2007) and high-mannose-type N-glycan binders (NPA, ConA, GNA, and HHL) (Mega et al., 1992; Van Damme et al., 2007). Characteristically, CHO cells express a23Sia, but not a2-6Sia. Indeed, the signals of an a2-3Sia-binder (MAL), but not a2-6Sia binding lectins (SNA, SSA, TJAI) were observed (Yabe et al., 2009). Significant signals were also observed for core-fucosylated biantennary N-glycan binders (PSA and LCA) (Tateno et al., 2009) and broader fucose binders (AOL and AAL) (Matsumura et al., 2009), indicating the expression of core-fucosylated biantennary N-glycans. Interestingly, CHO cells bound to PHAE, suggesting the expression of bisecting GlcNAc, which agrees with a recent finding (North et al., 2010). The expression of polylactosamine was also confirmed by the signals of LEL and STL, which are specific for polylactosamine. Regarding O-glycans, major structures are core 1 (Galb1-3GalNAc) and its sialylated forms such as sialyl T (Siaa2-3Galb13GalNAc) and disialyl T (Siaa2-3Galb1-3(Siaa2-6)GalNAc) (North et al., 2010). Consistent with this, CHO cells exhibited significant binding to ABA

188


CHO

700 Asn

Ser / Thr

Asn

Ser / Thr

Asn

Ser / Thr

Asn

Ser / Thr

350

0

Lec2

700

350

0

Lec8

700

350

0

Lec1

700

350

LTL PSA LCA UEAI AOL AAL MAL SNA SSA TJAI PHAL ECA RCA120 PHAE DSA GSLII NPA ConA GNA HHL BPL TJAII EEL ABA LEL STL UDA PWM Jacalin PNA WFA ACA MPA HPA VVA DBA SBA GSLI PTLI MAH WGA GSLIA4 GSLIB4

0

Figure 8.4 Glycan profiles of CHO, Lec2, Lec8, and Lec1. Data are shown as the average of 12 samples for each cell line.

(Nakamura-Tsuruta et al., 2006), Jacalin (Tachibana et al., 2006), ACA, and MPA, which bind to core 1 and sialyl T (Kuno et al., 2009) as well as MAH specific to disialyl T. Lec2 is a CMP-sialic acid Golgi transporter deletion mutant which expresses little or no sialylated glycoconjugates. Consistently, the signals of MAL, the a2-3Sia binder, were abolished, while those of asialo N-glycans (PHAL, ECA, and RCA120) were increased, consistent with the glycosylation phenotype of Lec2. In terms of O-glycans, no binding was observed for MAH, which bind to disialyl T, while the signals of core 1 binders (BPL, ACA, and MPA) were increased correspondingly. Lec8 is a UDP-Gal Golgi transporter deletion mutant, which expresses little or no galactosylated glycoconjugates compared to Lec2. The signals of Gal-binders (ECA, RCA120, BPL, TJA-II, PNA, WFA, and ACA) were clearly decreased, while those of a GlcNAc-binder, GSLII (NakamuraTsuruta et al., 2006), were increased. Lec1 is a MGAT1 deletion mutant which is incapable of synthesizing complex- and hybrid-type N-glycans. Indeed, the signals of high-mannosetype N-glycan binders (NPA, GNA, and HHL) were increased, while little or no binding was observed for Sia-, Gal-, and GlcNAc-terminated complex-type N-glycan binders such as MAL, PHAL, ECA, RCA120, PHAE,

189


and GSLII. However, the signals of PSA, LCA, AOL, and AAL were detected. This might be due to the presence of core-fucosylated highmannose-type N-glycans as reported recently (North et al., 2010).

7. Unsupervised Clustering and Principal Component Analysis The normalized data were analyzed by two multivariable analyses, unsupervised clustering and principal component analysis, to achieve overall classification of the samples without supervision. As shown in Fig. 8.5,

Lec2

Lec1

Lec8

CHO-1-1 CHO-2-1 CHO-3-1 CHO-1-4 CHO2-4 CHO-1-2 CHO-1-3 CHO3-2 CHO-2-3 CHO2-2 CHO-3-3 CHO-3-4 Lec2-2-1 Lec2-1-1 Lec2-1-3 Lec2-1-2 Lec2-4-1 Lec2-2-3 Lec2-4-2 Lec2-3-2 Lec2-2-2 Lec2-3-1 Lec2-3-3 Lec2-4-3 Lec1-1-1 Lec1-3-1 Lec1-3-2 Lec1-1-3 Lec1-3-3 Lec1-2-1 Lec1-4-3 Lec1-1-2 Lec1-2-2 Lec1-4-1 Lec1-2-3 Lec1-4-2 Lec8-2-1 Lec8-1-2 Lec8-1-3 Lec8-1-1 Lec8-4-1 Lec8-2-2 Lec8-2-3 Lec8-3-2 Lec8-3-3 Lec8-3-1 Lec8-4-3 Lec8-4-2

CHO

SNA SSA TJA-I PHA-L PHA-E RCA120 ECA BPL WFA TJA-II SBA PNA AAL PSA LCA ConA UDA Jacalin NPA GNA HHL LEL STL DSA WGA ACA PWM MAL MAH UEA-I AOL ABA MPA EEL LTL WA DBA GSL-II HPA GSL-I PTL-I GSL-IA4 GSL-IB4

Figure 8.5 Unsupervised clustering. Mean-normalized data were analyzed by Cluster 3.0. Positive: red, negative: green. Clustering method: complete linkage. The heat map with clustering was acquired using Java Treeview. A representative N-glycan structure of each cell line is illustrated above.

190


PC2 6 CHO Lec1

Lec1

4

Lec2 CHO 2

Lec8

0

PC1 –6

–4

–2

0

2

4

6

8

–2 Lec8 –4 Lec2 –6

–8

Figure 8.6 Principal component analysis.

samples of the same cell line were grouped into the same cluster. Lec2 was clustered more closely to CHO than to Lec1 and Lec8 as expected, since the difference of Lec2 from CHO is only the absence of a2-3Sia in glycoconjugates, whereas both Lec8 and Lec1 further lack galactose on N-glycans (Fig. 8.4). In principal component analysis, the four cell lines were clearly separated by two components, PC1 and PC2 (Fig. 8.6). These results demonstrate that the four cell lines are clearly discriminated on lectin microarray depending on their glycan structures.

8. Significance Difference Test To select lectins discriminating the four cell lines, the data were first analyzed by the Kruskal–Wallis test (Scheffe´’s method), a nonparametric test for multiple comparisons, which does not assume a normal population. For statistical analysis of lectin microarray data, we used nonparametric tests rather than parametric tests, in case obtained lectin microarray data might not follow normal distribution and variance. Among 43 lectins, 36 were selected as significantly different lectins with a p-value < 0.01 between

191


A

Intensity

500 MAL PHAL GSLII HHL

250

0 CHO

Lec2

Lec8

Lec1

Cells B

Eq.2 6 4

Lec1

CHO

2 Eq.1 –20

CHO Lec1

0 –15

–10 Lec8

–5

–2

0

–4

5

10

15

Lec2 Lec8

Lec2

–6 –8 Eq1:Y = 0.096PHAL + 0.045MAL – 0.047HHL – 0.062GSLII + 7.03 Eq.2: Y = 0.065MAL + 0.018HHL – 0.002PHAL – 0.134GSLII – 4.92

Figure 8.7 (A) Glycan profiles of CHO, Lec2, Lec8, and Lec1 based on four lectins. (B) Canonical discriminant analysis.

certain pairs of the cell lines. Among these 36 lectins, we further selected four, MAL, PHAL, GSLII, and HHL, which differentially interact with the four cell lines. As shown in Fig. 8.7A, the four cell lines gave clearly distinct profiles on the four lectins.

9. Discriminant Analysis Using the four lectins, we developed two discriminant formulas by linear discriminant analysis as follows (Fig. 8.7B): Y ¼ 0:096PHAL þ 0:045MAL 0:047HHL 0:062GSLII þ 7:03 ð8:1Þ Y ¼ 0:065MAL þ 0:018HHL þ 0:002PHAL 0:134GSLII 4:92 ð8:2Þ

192


Using the above two equations, the four cell lines could be clearly differentiated (Fig. 8.7B). The discriminant analysis is indeed a useful statistical procedure to construct cell discrimination formulas based on the data obtained by lectin microarray.

10. Differential Analysis Between CHO and Lec1 Having constructed discriminant formulas of the four cell lines using linear discriminant analysis, we also showed a simpler strategy to construct formulas to distinguish two cell types. First, the mean-normalized data of CHO and Lec1 were analyzed by the Mann–Whitney U-test, a nonparametric test to select significantly different lectins between them (Fig. 8.8). Eight lectins with significantly higher signals and with p-value 0, whereas those of all Lec1 cell samples were Lec1 Lectin

P-value (two-tailed)

Lectin

P-value (two-tailed)

PHAE

0.0000

HHL

0.0000

PHAL

0.0000

GNA

0.0000

MAL

0.0000

NPA

0.0000

DSA

0.0000

PSA

0.0000

RCA120

0.0000

LCA

0.0000

ECA

0.0000

ConA

0.0001

GSLII

0.0000

PWM

0.0005

WGA

0.0000

AOL

0.0003

Figure 8.8

Mann–Whitney U-test.

ð8:3Þ

193


A

CHO

Relative intensity

600

Lec1

PHAE HHL

300

0

1

4

B

7

10

13 16 Sample

CHO

19

22

Lec1

Score

300

0

–300

0

12

24

Y = PHAE – HHL(Eq. 3) Y ³ 0, CHO Y < 0, Lec1

– 600

Sample

Figure 8.9 (A) Relative intensity of PHAE and HHL for CHO and Lec1 cells. (B) Scores of the discrimination formula (Y ¼ PHAE HHL) obtained for CHO and Lec1 cells.

11. Validation The lectin microarray data were then validated by flow cytometry, a conventional analytical method (Fig. 8.10). PHAE, PHAL, MAL, DSA, and RCA120, which gave higher signals to CHO than to Lec1 (Mann–Whitney U-test) (Fig. 8.8), bound more intensely to CHO than to Lec1. Inversely, HHL, GNA, NPA, PSA, and LCA, giving lower signals to CHO than to Lec1, indeed exhibited lower fluorescence intensities to CHO than to Lec1. Therefore, the data of lectin microarray correspond well to those of flow cytometry.

12. Concluding Remarks Here, we described the applications of lectin microarray for profiling the cellular glycome, and systematic and strategic development of cell discrimination procedures using lectins without antibodies. Indeed, the glycan profiles of

194


Cell number

PHAE

0

10

2

3

10

PHAL

4

10

5

10

0

10

2

3

10

2

3

10

4

10

5

10

0

GNA

HHL

0

10

MAL

4

10

5

10

0

10

2

3

10

10

2

3

10

DSA

4

10

5

10

0

NPA

4

10

5

10

0

10

2

3

10

10

2

3

10

RCA120

4

10

5

10

0

PSA

4

10

5

10

0

10

2

3

10

10

2

3

10

4

10

5

10

LCA

4

10

5

10

0

10

2

3

10

4

10

5

10

Lectin : Control

: CHO

: Lec1

Figure 8.10 Flow cytometry. Cells (2 105) suspended in 100 mL PBS/BSA (10 mM phosphate buffer, pH 7.2, 0.15 M NaCl, and 1% BSA) were incubated with 20 mg/mL Cy3-labeled lectins on ice for 1 h and analyzed by flow cytometry.

CHO and its mutants obtained by lectin microarray clearly reflected their glycosylation phenotypes. The data were also supported by flow cytometry, demonstrating that lectin microarray is a useful tool for profiling the cellular glycome. Since lectin microarray requires only 10–100 ng of protein for each analysis (sensitive), and multiple samples (e.g., 100 samples) can be analyzed simultaneously (high-throughput), this technology has obvious potential advantages over others (e.g., mass spectrometry and multiple liquid chromatography mapping) to become a standard technique for analyzing global changes of the cellular glycome. The profiles are quite useful for cell discrimination. To distinguish more than three cell types, canonical discriminant analysis should be a convenient statistical method to construct lectin-based discriminant formulas. For discrimination of two cell types, discriminant formulas can be constructed more easily by the use of two lectins selected from the Mann–Whitney U-test. The procedures described in this chapter could also be widely used for the systematic development of cell discrimination procedures to distinguish a wide a range of cells. The strategy could be readily applicable to the evaluation of properties and conditions of various cell types. For example, the differentiation state and propensity of stem cells might be adequately evaluated by means of lectin microarray. Furthermore, the protocols described here are also applicable to the analysis of the cellular glycomes of various microorganisms, such as fungi, bacteria, and viruses, as described by others (Hsu et al., 2006). This should lead to understanding as yet unknown glycan functions in the light of evolutionary strategies of complex life systems. Research from this aspect is now ongoing.


195

ACKNOWLEDGMENTS We thank Noboru Uchiyama, Keiko Hiemori, Mihoko Fukumura, Yoshiko Kubo, and Jinko Murakami for technical assistance. This work was supported by New Energy and Industrial Technology Development Organization (NEDO) in Japan.

REFERENCES Ebe, Y., et al. (2006). Application of lectin microarray to crude samples: Differential glycan profiling of lec mutants. J. Biochem. 139, 323–327. Gagneux, P., and Varki, A. (1999). Evolutionary considerations in relating oligosaccharide diversity to biological function. Glycobiology 9, 747–755. Hsu, K. L., et al. (2006). Analyzing the dynamic bacterial glycome with a lectin microarray approach. Nat. Chem. Biol. 2, 153–157. Itakura, Y., et al. (2007). Systematic comparison of oligosaccharide specificity of Ricinus communis agglutinin I and Erythrina lectins: A search by frontal affinity chromatography. J. Biochem. 142, 459–469. Kuno, A., et al. (2005). Evanescent-field fluorescence-assisted lectin microarray: A new strategy for glycan profiling. Nat. Methods 2, 851–856. Kuno, A., et al. (2008). Development of a data-mining system for differential profiling of cell glycoproteins based on lectin microarray. J. Proteomics Bioinform. 1, 68–72. Kuno, A., et al. (2009). Focused differential glycan analysis with the platform antibody-assisted lectin profiling for glycan-related biomarker verification. Mol. Cell. Proteomics 8, 99–108. Matsumura, K., et al. (2009). Comparative analysis of oligosaccharide specificities of fucosespecific lectins from Aspergillus oryzae and Aleuria aurantia using frontal affinity chromatography. Anal. Biochem. 386, 217–221. Mega, T., et al. (1992). Characterization of carbohydrate-binding specificity of concanavalin A by competitive binding of pyridylamino sugar chains. J. Biochem. 111, 396–400. Nakamura-Tsuruta, S., et al. (2006). Comparative analysis by frontal affinity chromatography of oligosaccharide specificity of GlcNAc-binding lectins, Griffonia simplicifolia lectin-II (GSL-II) and Boletopsis leucomelas lectin (BLL). J. Biochem. 140, 285–291. North, S. J., et al. (2010). Glycomics profiling of Chinese hamster ovary cell glycosylation mutants reveals N-glycans of a novel size and complexity. J. Biol. Chem. 285, 5759–5775. Tachibana, K., et al. (2006). Elucidation of binding specificity of Jacalin toward O-glycosylated peptides: Quantitative analysis by frontal affinity chromatography. Glycobiology 16, 46–53. Tateno, H., et al. (2007). A novel strategy for mammalian cell surface glycome profiling using lectin microarray. Glycobiology 17, 1138–1146. Tateno, H., et al. (2009). Comparative analysis of core-fucose-binding lectins from Lens culinaris and Pisum sativum using frontal affinity chromatography. Glycobiology 19, 527–536. Uchiyama, N., et al. (2006). Development of a lectin microarray based on an evanescentfield fluorescence principle. Methods Enzymol. 415, 341–351. Uchiyama, N., et al. (2008). Optimization of evanescent-field fluorescence-assisted lectin microarray for high-sensitivity detection of monovalent oligosaccharides and glycoproteins. Proteomics 8, 3042–3050. Van Damme, E. J., et al. (2007). Phylogenetic and specificity studies of two-domain GNArelated lectins: Generation of multispecificity through domain duplication and divergent evolution. Biochem. J. 404, 51–61. Yabe, R., et al. (2009). Engineering a versatile tandem repeat-type alpha2-6sialic acid-binding lectin. Biochem. Biophys. Res. Commun. 384, 204–209.

C H A P T E R

N I N E

Applications of Heparin and Heparan Sulfate Microarrays Jian Yin*,† and Peter H. Seeberger*,† Contents 1. Introduction 2. Preparation of Amino-Functionalized HS/Heparin Oligosaccharides 3. Microarray Analysis of HS/Heparin–FGF Binding 3.1. Materials and equipment 3.2. Fabrication of HS/heparin microarrays 3.3. Incubation with HS/heparin-binding FGFs 3.4. Binding affinities of HS/heparin with FGFs 4. A HS/Heparin Microarray to Determine the Binding Profiles of Heparin Dendrimers to FGF-2 4.1. Materials and equipment 4.2. Preparation of glycodendrimers and amine-functionalized 5 kDa heparin 4.3. Fabrication of HS/heparin microarrays 4.4. Incubation with FGF-binding heparin oligosaccharide dendrimers 4.5. Binding affinities of heparin oligosaccharide dendrimers to FGFs 5. HS/Heparin Interaction with Chemokines as Determined by Microarray Analysis 5.1. Materials and equipment 5.2. Fabrication of HS/heparin microarray and incubation with chemokines 5.3. Binding affinities of HS/heparin with chemokines 6. HS/Heparin Microarray for Determination of Their Interaction with NCRs 6.1. Materials and equipment

198 201 202 204 204 206 206 207 207 209 210 210 211 211 211 212 212 213 213

* Department of Biomolecular Systems, Max Planck Institute of Colloids and Interfaces, Research Campus Potsdam-Golm, Potsdam, Germany Department of Biology, Chemistry and Pharmacy, Institute of Chemistry and Biochemistry, Free University Berlin, Arnimallee, Berlin, Germany

{


#


197

198

Jian Yin and Peter H. Seeberger

6.2. Fabrication of HS/heparin microarray and incubation with chemokines 6.3. Binding affinities of HS/heparin to NCRs 7. Conclusions Acknowledgments References

214 214 214 215 215

Abstract Carbohydrate microarrays have become crucial tools for revealing the biological interactions and functions of glycans, primarily because the microarray format enables the investigation of large numbers of carbohydrates at a time. Heparan sulfate (HS) and heparin are the most structurally complex glycosaminoglycans (GAGs). In this chapter, we describe the preparation of a small library of HS/ heparin oligosaccharides, and the fabrication of HS/heparin microarrays that have been used to establish HS/heparin-binding profiles. Fibroblast growth factors (FGFs), natural cytotoxicity receptors (NCRs), and chemokines were screened to illuminate the very important biological functions of these glycans.

1. Introduction Oligonucleotides (Insight, 2004) and proteins (Insight, 2003) have been the primary focus of most scientific studies on biomacromolecules, and the significance of carbohydrates in biological systems (see Fig. 9.1) has been underappreciated by the scientific community. Recently, however, the number of chemists, biochemists, and biologists active in the GPI-anchored protein

GPI-anchored protein

N-glycoprotein

Extracellular fluid

O-glycoprotein

GPI

Glycan

Glycosphingolipid Proteoglycan

GPI

Lipid bilayer

Cytoplasm

Figure 9.1 Schematic view of a cell membrane with glycosphingolipids, N- and Oglycoproteins, proteoglycans, and glycosylphosphatidylinositol (GPI)-anchored proteins.

199

Heparin/Heparan Sulfate Microarrays

glycosciences has been steadily increasing (Insight, 2007), resulting in new strategies and ideas for the purification, synthesis, analysis, and pharmaceutical applications of carbohydrates (Laughlin and Bertozzi, 2009; Murrey and Hsieh-Wilson, 2008; Seeberger and Werz, 2007). Carbohydrate microarrays (see Fig. 9.2) are a powerful technology as they can reveal carbohydrate–protein interactions. These insights can be exploited to evaluate and identify sugar ligands in both endogenous receptor systems and pathogen–host interactions (Feizi et al., 2003; Love and Seeberger, 2002; Shin et al., 2005). By covalently or noncovalently immobilizing many different carbohydrates at a time on a solid surface, this technology exhibits a clear advantage over conventional strategies, such as surface plasmon resonance (SPR) or isothermal titration calorimetry (ITC) (Dam and Brewer, 2002; Lis and Sharon, 1972). Consequently, the carbohydrate microarray has become a high-throughput analytic tool for elucidating the function of carbohydrates in biological processes (De Paz and Seeberger, 2006; Horlacher and Seeberger, 2008; Liang et al., 2008). Glycosaminoglycan (GAG) microarrays were developed in the last decade and have emerged as an important tool to address the many unanswered scientific questions related to the structure and function of GAGs in biological systems (De Paz and Seeberger, 2008; De Paz et al., 2006a; Gandhi and Mancera, 2008; Laremore et al., 2009). Heparan sulfate (HS) and heparin, the most complex polysaccharides of the GAG family, play an important role in fundamental physiological processes, such as cell growth and differentiation, blood coagulation, and inflammatory responses (Capila and Linhardt, 2002; Gama and HsiehWilson, 2005; Noti and Seeberger, 2005). The highly negatively charged HS and heparin sequences have been studied extensively due to their well understood role in anticoagulation (Lindahl and Li, 2009). HS and heparin are structurally related polysaccharides (see Fig. 9.3), and the differences between the two polysaccharides are widely regarded as quantitative and

Synthetic or isolated carbohydrates

NH2 NH2 NH2

O

N O O O

Addition of protein

Immobilization A

HN HN HN O O O

B

HN HN HN O O O

NHS-derivatized glass slide

Figure 9.2 General strategy for the preparation of microarrays containing synthetic or isolated carbohydrates.

200


OH

OH O

O

Heparan sulfate

HO2C

O

OH O

O

O

HO2C

OX

O

HO HN O Ac

HO HN O Y

X = H or SO−3 Y = H, SO−3 or Ac

OH

OX

Major aptotic disaccharide repeating unit

Minor variable disaccharide repeating unit

OSO−3 OX O

Heparin

HO2C

OH O HO O

HN Y

O

OX

Minor variable disaccharide repeating unit

HO2C O

OH O

O

O HO HN O O3SO

− −

X = H or SO−3 Y = H, SO−3 or Ac

OSO3

Major aptotic disaccharide repeating unit

Figure 9.3 Major and minor disaccharide repeating units in heparan sulfate and heparin.

not qualitative (Lindahl and Kjellen, 1991). HS chains often contain extended domains with a level of sulfation greater than that found in heparin sequences, as well as a higher level of acetylated glucosamines. Moreover, heparin is only biosynthesized and stored in mast cells, whereas HS is expressed on cell surfaces and in the extracellular matrix of proteoglycans (Varki et al., 1999). HS/heparin microarrays were used to determine the binding profiles of these polysaccharides with different proteins (Marson et al., 2009; Mercey et al., 2008; Shipp and Hsieh-Wilson, 2009). Our group has developed several different carbohydrate microarrays to elucidate the function of carbohydrates in various biological systems, such as the identification of human immunodeficiency virus (HIV) vaccine candidate antigens (Adams et al., 2004), the detection of pathogenic bacteria (Disney and Seeberger, 2004a), and the evaluation of aminoglycoside antibiotics (Disney and Seeberger, 2004b; Disney et al., 2004). Synthetic HS/ heparin sequences (Orgueira et al., 2003) were employed to fabricate HS/ heparin microarrays to determine binding profiles of fibroblast growth factors (FGFs) (De Paz et al., 2006b,c; Noti et al., 2006), chemokines (De Paz et al., 2007a), and natural cytotoxicity receptors (NCRs) (Hecht et al., 2009). Heparin dendrimers (De Paz et al., 2007b) served to compare the efficiency


201

of heparin binding to FGF-2. Here, we describe the preparation of heparin microarrays using synthetic heparin oligosaccharides. A summary of binding profiles with several biologically significant proteins is provided.

2. Preparation of Amino-Functionalized HS/Heparin Oligosaccharides The synthesis of complex carbohydrates is difficult, but the synthesis of HS/heparin is even more challenging, since the variability in monomer sulfation results in highly complex structures (see Fig. 9.4). Since the early 1980s, the synthesis of HS and heparin sequences has aroused the keen interests of organic chemists, and different approaches have been developed (Arungundram et al., 2009; Polat and Wong, 2007; Zhang et al., 2008). Our approach relies on GlcN-IdoA disaccharides, which are designed to serve as the repeating unit of the major sequence of HS/heparin fragments. Differentially protected glucosamines as well as iduronic acids were synthesized to prepare disaccharide modules (Lohman et al., 2003; Orgueira et al., 2003). Following methods established in our laboratory, the glucosamine building blocks were readily available in sufficient quantities, and a ‘‘de novo’’ approach allowed for the synthesis of glucuronic and iduronic acid units via relatively short routes from noncarbohydrate precursors (Adibekian et al., 2007; Timmer et al., 2005). Acyl groups served to acetate esters mark the hydroxyl groups to be sulfated, while benzyl ethers masked hydroxyl groups that will not be modified. The amine group of glucosamine required the installation of different protecting groups, such as azides. In succession, 2-azidoglucopyranose trichloroacetimidates served as glycosylating agent and iduronic acid units served as nucleophiles for the stereoselective preparation of 1,2-cis glycosidic linkages. One of two different methods was used to place an amine-terminated linker at the reducing end of HS/heparin oligosaccharides ranging in length from di- to hexamers and containing different sequences and sulfation patterns (Noti et al., 2006). One strategy involved conversion of a glycosyl trichloroacetimidate to an n-pentenyl glycoside by coupling with n-pentenyl alcohol, followed by radical elongation of the pentenyl moiety using 2-(benzyloxycarbonylamino)-1-ethanethiol. Saponification with lithium hydroperoxide and potassium hydroxide solution completed the conversion of sulfide to sulfone without detectable sulfoxide intermediates. Alternatively, to avoid elaborate linker modifications at the late stage of oligosaccharide synthesis, and to improve upon the moderate yields observed during the radical elongation of the pentenyl moiety of tetra- and hexasaccharides, an amineprotected pentyl linker, such as n-benzyloxycarbonyl-5-aminopentane-1-ol, was installed at the reducing end prior to oligosaccharide construction. Both amine-terminated linkers allowed for immobilization onto

202


–

OSO3 O

O – O3SO

AcHN O HO

CO2H O

OSO3–

O OH HO

O AcHN

48 modified decasaccharide building blocks

OH O

HO2C O

255 million decasaccharides 5.3 million octasaccharides 11,592 hexasaccharides

–

O3SO

OPG1 PGO PG1O

O PG4N O 2 PG O

CO2PG5 O

OPG1 X

O

PGO PG1O

3

OPG

CO2PG O

N3 O

O 1 PG O 3 OPG

PG2O

PGO PG1O

2 PG O

PGO 1 PG O

CO2Me O OPG

O X N3

X 3

CO2PG

OPG2

OPG2 O Y

PGO PG2O

O Y

N3

N3

OPG1 O

OPG1 Y

N3

1

O N3 O

PGO PG1O

OPG

32 trisaccharide building blocks

OPG1

8 decasaccharide building blocks

5

PGO 2 PG O

O N3

PGO PG2O

OPG

PG2O 5

Y

5

O

PG O2C PGO

X

1 glucuronic acid building block

X

O OPG

1 iduronic acid building block

4 glucosamine building blocks

Figure 9.4 Retrosynthetic analysis of a general, modular approach to the preparation of heparan sulfate and heparin oligosaccharides.

N-hydroxysuccinimide (NHS)-activated glass slides, and the creation of HS/heparin microarrays. In this way, a collection of 13 synthetic HS/ heparin oligosaccharides (see Fig. 9.5) were prepared that was ready for HS/heparin microarray fabrication.

3. Microarray Analysis of HS/Heparin–FGF Binding FGFs bind to the extracellular matrix of target tissues by interacting with heparin-like glycosaminoglycans (HLGAGs) (Mohammadi et al., 2005; Noti and Seeberger, 2005). The FGF family of proteins contains 23

OSO3–

OSO3– O HO2C O HO NH – HO O3SO – O3SO

O HO2C OH O O O NH – HO O3SO – O3SO

OSO3– O HO2C OH O O O OH O NH – O HO O3SO – O3SO

O 3

O S

NH2

OSO3– O HO2C O HO2C OH O O O NH –O SO HO O 3 HO NH –O SO HO Ac 3 2 Ac

1

OH HO2C OH O O O OH O OH O HO C O 2 OH NH O OH HO HO – O HO2C O O O3SO O NH HO O HO HO – O3SO NH HO HO – O3SO 3 OH O O OH OH O O HO2C OH NH O HO OH O HO2C O O – O3SO O OH O HO2C – NH O SO HO O O 3 – NH –O SO HO HO O3SO 3 – – O3SO O3SO

NH2

3

3

NH2

5

OSO3– O HO2C O HO NH –O SO HO 3 – O3SO

OSO3– O HO2C OH O O O OH O NH – O HO O3SO – O3SO

7

OSO3–

S

OH O O OH OH O O HO2C NH O HO OH O HO2C O Ac O NH HO HO HO 9 Ac HO

NH2

3

OH O O OH OH O O HO2C NH O HO OH O HO2C O – O O3SO NH HO HO HO – HO O3SO

3

NH2

8

OSO3– O

HO NH HO – O3SO

O

O

3

O S

NH2

HO2C HO

10

HO

OH O O

S 3

11

HO2C

O

O

NH2

3

OH O HO2C OH O OH O O OH O HO C O 2 OH NH O HO HO OH O O HO2C O Ac O NH O HO HO HO NH HO HO Ac 4 Ac OH O O OH OH HO C O 2 O OH NH O HO OH O O HO2C O Ac O OH HO2C NH – O O HO O3SO O NH –O SO HO HO Ac 3 – Ac O3SO 6

O

O

OSO3– HO2C OH O O O O OH O NH – O HO O3SO Ac

–

O3SO –

O3SO

OH O O

O

O S 3

OSO3– HO2C OH O O O O HO NH – HO O3SO – O3SO 13

O 3

NH2 3

3

O 3

O S

NH2

12

Figure 9.5 A small library of heparan sulfate and heparin-containing 13 synthetic oligosaccharides.

NH2

NH2

NH2

O S

NH2

204


different members and is involved in developmental and physiological processes (Noti and Seeberger, 2005). The best-studied members of the FGF family are FGF-1 (acidic FGF) and FGF-2 (basic FGF). Recent studies explained the complexity of the molecular mechanism involved in HS/ heparin-mediated FGF signaling; however, the HS/heparin sequences used for those studies were obtained through either enzymatic or chemical depolymerization methods (Pellegrini, 2001; Powell et al., 2004). Therefore, it was necessary to further investigate FGF signaling using carbohydrate microarrays assembled with chemically defined HS/heparin oligosaccharides for more precise structure–activity relationship studies. We developed a technique for the preparation and use of a microarray format containing synthetic HS/heparin oligosaccharides with varying sulfation patterns to determine HS/heparin–FGF binding affinities (see Fig. 9.6) (De Paz et al., 2006b; Noti et al., 2006).

3.1. Materials and equipment Sodium phosphate buffer (pH 9.0, 50 mM) Automated arraying robot (Perkin Elmer) NHS-activated slides (CodeLink, Amersham Biosciences) FGF-1, FGF-2, FGF-4, anti-FGF-2, and anti-FGF-4 (PeproTech EC) Anti-FGF-1 (Santa Cruz Biotechnology, Inc.) AlexaFluro-546-labled anti-rabbit IgG (Molecular Probes) PBS buffer (pH 7.5, 10 mM) HybriSlip hybridization covers (Grace Bio-Labs) Nanopure water Fluorescence reader (LS400, Tecan) Scan Array Express software (Perkin Elmer) Gene Spotter software (Microdiscovery GmbH)

3.2. Fabrication of HS/heparin microarrays The HS/heparin microarrays were prepared using the following protocol (De Paz et al., 2006b; Noti et al., 2006): HS/heparin oligosaccharides were dissolved in sodium phosphate buffer (pH 9.0, 50 mM), and were arrayed onto NHS-activated CodeLink slides (Blixta et al., 2004) using an automated arraying robot. Slides were printed in 50% relative humidity at 22 C, followed by incubation overnight in a saturated NaCl chamber that provides an environment of 75% relative humidity. The robot delivered 1 nL of sugar solutions at four different concentrations (2 mM, 400, 80, and 16 mM), and the resulting spots had an average diameter of 200 mm with a distance of 500 mm between the centers of adjacent spots. All samples were printed in replicates of 16. Slides were then washed with water (three times) to remove

-

OSO3 O

OSO3

OH O

HO2C O HO NH O HO O3SO O3SO

OAc O BnO BnO

Solution phase synthesis

O NH O HO O3SO

OAc O MeO2C

N3 O

OBn O BnO O

HO2C -

O3SO

OH S O O 5 O O

MeO2C OBn O O

N3 O

NH2

Immobilization of oligosaccharides on NHS slides HS/heparin microarrays

3

OAc O

PivO BnO BnO

AcO

N3 O

MeO2C

OBn O O BnO

OAc O

MeO2C

N3 O

OBn O O

PivO

AcO

Solid phase synthesis OAc O BnO BnO

OAc O MeO2C

N3 O

OBn O BnO O

MeO2C

–

HO HO NH O − O3SO

NBnZ 3

PivO

N3 O

AcO

OSO3 O

-

OSO3 O

OBn O O

HO2C −

OH O

O3SO

O HO NH O − O3SO

HO2C −

OH O O

O3SO

3

NH2

Immobilization of oligosaccharides on NHS slides HS/heparin microarrays

Figure 9.6 A general method for the preparation of microarrays containing synthetic HS/heparin oligosaccharides.

206


the remaining unbound carbohydrates from the chip. Remaining succinimidyl groups were quenched by placing slides in a solution preheated to 50 C that contained 100 mM ethanolamine in sodium phosphate buffer (pH 9.0, 50 mM) for 1 h. Slides were rinsed several times with distilled water, dried by centrifugation, and stored in a desiccator before proceeding with binding studies.

3.3. Incubation with HS/heparin-binding FGFs The binding assay, as well as image acquisition and signal processing, was performed using the protocol described in Noti et al. (2006). Solutions used for chip hybridization were sterile filtered through a 0.2 mm syringe filter before use. The FGF hybridization solutions were prepared by diluting the stock solutions to a concentration of 4–20 mg/mL with PBS buffer (pH 7.5, 10 mM) containing BSA (1%). Array incubations were performed as follows (Noti et al., 2006): FGF solution (100 mL) was placed between array slides and plain coverslips and incubated for 1 h at room temperature. The arrays were washed with PBS (pH 7.5, 10 mM) containing 1% Tween 20 and 0.1% BSA, two times with water, and then centrifuged for 5 min to ensure dryness. For detection of bound FGF, arrays were incubated with antihuman FGF polyclonal antibody (4–20 mg/mL) and then washed in a similar manner as above. Finally, AlexaFluor-546-labeled anti-rabbit IgG (20 mg/mL) secondary antibody was used, and the slides were again washed. All arrays were scanned using an LS400 scanner, and fluorescence spotter intensities were integrated using appropriate software.

3.4. Binding affinities of HS/heparin with FGFs The binding affinities of 12 different HS/heparin oligosaccharides (see Fig. 9.5, 1–8 and 10–13) for FGFs were assayed using the microarray. FGF-2 bound strongly to hexasaccharides 1, 2, and 5, tetrasaccharide 7, and had weak affinity for disaccharide 10 and monosaccharide 12. No binding was observed for hexasaccharides 3, 4, and 6. These findings indicated at least two sulfate groups on each disaccharide are required for oligosaccharide recognition by FGF-2. Meanwhile, further experiments indicated that the sulfate group at position 6 of the glucosamine unit was unnecessary for recognition, since the binding affinities of FGF-2 for hexasaccharides 1 and 5 were similar, a finding that supports the work of Pellegrini (2001). Incubation of FGF-1 with the microarray showed that FGF-1 binds hexasaccharides 1, 2, and 5 weakly suggesting that a higher negative charge density, with three sulfate groups on each disaccharide, is required for FGF-1 binding. The monosaccharide 12, which exhibited the strongest binding affinity for FGF-1 of all the synthetic oligosaccharides tested, was previously reported to be a potential inhibitor of angiogenesis


207

with anticancer properties (Cochran et al., 2003; Karoli et al., 2005). Our analysis indicated that the nonsulfated oligosaccharides also bound FGF-1 tightly, as reported by Ornitz et al. (1995). These findings suggested that FGF can specifically recognize structural features of nonsulfated HS/heparin backbone instead of ionic interactions with sulfate groups. Finally, our microarray screen demonstrated that FGF-4 bound hexasaccharide 1 more tightly than oligosaccharides 2, 5, and 7. Comparison of structures 1 and 7 shows that, while the two oligosaccharides are equally sulfated, oligosaccharide 7 is shorter, implying that the distribution of the sulfate groups has a significant effect on FGF-4 binding.

4. A HS/Heparin Microarray to Determine the Binding Profiles of Heparin Dendrimers to FGF-2 Dendrimers are hyperbranched synthetic macromolecules of defined structure and molecular weight that have been evaluated for biomedical applications (Haag and Kratz, 2006). Multivalent presentation of sugar epitopes on an appropriate macromolecular scaffold increases conjugate binding due to a cluster effect, and enhances carbohydrate–protein interactions (Suda et al., 2006). Installation of terminal sugar residues on dendrimers was reported to create a multivalent display that mimicked cell-surface glycans (Rele et al., 2005). Recently, a HS/heparin oligosaccharide microarray was employed by our group to determine the inhibition of heparin– protein interactions by heparin dendrimers (De Paz et al., 2007b). In this study, anionic polyamidoamine (PAMAM) dendrimers were employed to prepare multivalent conjugates of synthetic heparin oligosaccharides (see Fig. 9.7). Dendrimer binding to FGF-2 was analyzed by HS/heparin microarrays and SPR measurements on gold chips.

4.1. Materials and equipment Sodium phosphate buffer (pH 9.0, 50 mM) sciFLEXARRAYER noncontact printer (Scienion AG) NHS-activated slides (CodeLink, Amersham Biosciences) FGF-2, anti-FGF-2 (PeproTech EC) AlexaFluro-546-labled goat anti-rabbit IgG (Invitrogen, Carlsbad) HBS-EP buffer (pH 7.4, 10 mM, 150 mM NaCl, 3 mM EDTA, 0.005% (v/v) surfactant P20) (BIAcore) HybriSlip hybridization covers (Grace Bio-Labs) Nanopure water Fluorescence reader (LS400, Tecan)

ONa

O

NaO

O

R NaO

O ONa

N

N

O

N

N HN

O

O N N H

NaO

O

R O NaO

N

N HO

HN

NH

O

O

ONa

NH

O N H

O N

NaO O

R

O

ONa

R

N

O

NaO

O NaO

O

ONa O NaO

HN

N

N

HO3C OH O O

O 3

O S

NH2

-

NaO O

O

NH2

O3SO

O3SO

N

O

O

O S

10

-

R

O

N

N

1 10 12 13 ONa

O

HN

NH

= = = = =

ONa

O

O

O

NH O

N

HN N

N

N

NaO

O

HN

N H

NH

N

R R R R R

N

O O

NH

O

O

N H

HOD-1: HOD-2: HOD-3: HOD-4: HOD-5:

O ONa

N

O O

O N

O N

O

R

O 3

HO NH HO O3SO

NH

O

HN

NH O

O

H N

OSO3O O

O ONa

O

N

N

N H

O N

O

O

O

HN

O

N

O

O N

HN

N

Heparin oligosaccharide dendrimers (HOD)

ONa

O

R

NaO

N

O

N

NaO

NaO

O

HN

NH

O

O

O

ONa

NH

HN NaO

R

O

O

OSO3O HO2C OH O O O HO NH HO O3SO O3SO

O 3

12

O S

NH2

13

OSO3O HO2C OH O OSO3O HO2C OH O O OSO3O NH -O SO O HO HO2C OH 3 O O O O O3SO NH -O SO HO O HO 3 NH -O SO HO O3SO 3 O3SO

1

Figure 9.7 Structure of heparin oligosaccharide dendrimers (HOD).

O

O S

3

NH2


209

Gene Spotter software (Microdiscovery GmbH) Starburst PAMAM dendrimer generation 2.5 containing 32 sodium carboxylate surface groups (Sigma-Aldrich) Microcon centrifugal filter units (Millipore, Billerica) Deaminated heparin (5 kDa) (Sigma-Aldrich) SPR measurements device (BIAcore 3000, BIAcore) SPR measurements software (BIAcore) CM5 chips (BIAcore)

4.2. Preparation of glycodendrimers and aminefunctionalized 5 kDa heparin The terminal carboxylate groups on PAMAM dendrimer were activated using the following protocol (De Paz et al., 2007b): Commercially available methanolic PAMAM solution was coevaporated with CH2Cl2 twice under reduced pressure. The residue (20 mg, 3.2 mmol) was dissolved in anhydrous DMSO (2 mL). EDC (23.6 mg, 122 mmol, 1.2 equivalents per carboxylate group) and NHS (14.2 mg, 122 mmol, 1.2 equivalents per carboxylate group) were added, and the reaction mixture stirred under argon atmosphere for 24 h. Four heparin oligosaccharide dendrimers (HOD) were synthesized using the protocol (De Paz et al., 2007b) exemplified by the synthesis of HOD-4. Triethylamine (10 mL) and monosaccharide 10 (3.0 mg, 5.6 mmol, 1.1 equivalents per carboxylate group) were added to an aliquot of the activated PAMAM solution (100 mL, 0.16 mmol). DMSO (100 mL) was then added to dissolve the sugar. The reaction mixture was stirred under argon for 24 h and then lyophilized to remove the DMSO. The residue was dissolved in water, purified by centrifugal ultrafiltration (Microcon 3 kDa, 40 min, 14,000 rpm) and washed twice with water. Lyophilization in water afforded the corresponding dendrimer HOD-4 as a white powder (1.4 mg). Some batches were submitted to an additional purification step through Sephadex G-25 (prepacked PD-10 column; Amersham Biosciences) in order to remove the glycerin which coats Microcon ultrafiltration membranes and may eventually contaminate the dendrimer sample. The ratio of sugar to dendrimer was 8.6 (27% loading of monosaccharide 10 on the dendrimer) based on the 1H NMR spectrum in D2O (300 MHz). The molecular weight of HOD-4 (10.5 kDa) was estimated based on NMR integration. To estimate the molecular weight, it was presumed that all the sulfate groups of the sugar and the carboxylic acid groups of the dendrimer were presented as sodium salts. Three other HODs were also prepared by using this same protocol. Amine-functionalized 5 kDa heparin was prepared as follows (De Paz et al., 2007b). Deaminated 5 kDa heparin (2 mg, 0.4 mmol) was dissolved in

210


MES buffer (0.4 mL, 50 mM, pH 6.8). 1,11-Diamino-3,6,9-trioxaundecane (0.8 mg, 4 mmol) in DMF (25 mL) and NaCNBH3 (50 mL, 0.16 M solution in water) were added and the reaction mixture was stirred at room temperature for 18 h. To remove excess linker, the solution was submitted to centrifugal ultrafiltration (Microcon 3 kDa, 40 min, 14,000 rpm). After washing three times with water the precipitate was diluted with water and lyophilized to give a white solid (2 mg) which was stored at 20 C prior to use.

4.3. Fabrication of HS/heparin microarrays The HS/heparin microarrays were prepared using the following protocol (De Paz et al., 2007b): Amine-functionalized heparin (average molecular weight 5 kDa) was dissolved in sodium phosphate buffer (50 mM, pH 9.0) and spatially arrayed onto NHS-activated CodeLink slides by an automated arraying robot. Each glass slide was subdivided into eight different sections with heparin printed on each section, at two different concentrations (0.5 and 1 mM) in 17 replicates. Slides were printed in 50% relative humidity at 22 C, followed by incubation overnight in a saturated aqueous NaCl solution chamber that provides a 75% relative humidity environment. The robot delivered approximately 1 nL heparin solution and the resulting spots had an average diameter of 200 mm with a distance of 500 mm between the centers of adjacent spots. Slides were then washed three times with water to remove unbound carbohydrate from the surface. Remaining succinimidyl groups were quenched by placing slides in a solution preheated to 50 C that contained 100 mM ethanolamine in sodium phosphate buffer (50 mM, pH 9.0) for 1 h. Slides were rinsed several times with distilled water, dried by centrifugation, and stored in a dessicator before proceeding with binding experiments.

4.4. Incubation with FGF-binding heparin oligosaccharide dendrimers After attaching an 8-well hybridization chamber to the slide, each block was incubated with 40 mL of a mixture of FGF-2 (29 nM) and a competitor in PBS buffer (10 mM, pH 7.5) containing BSA (1%) for 1 h at room temperature. Glycodendrimers HOD-1–HOD-4, PAMAM dendrimer 5, oligosaccharides 1, 10, 12, and 13, sucrose octasulfate, and deaminated 5 kDa heparin were added as competitors at concentrations ranging from 0.5 nM to 2.5 mM. A solution of FGF-2 (29 nM) without competitor was used as positive control. The arrays were washed twice with PBS (10 mM, pH 7.5) containing 1% Tween 20 and 0.1% BSA, twice with water, and then centrifuged for 3 min to ensure dryness. For detection of bound FGF2, arrays were incubated with anti-FGF-2 (20 mg/mL) and then washed in a similar manner as above. Afterwards, 100 mL of antibody solution was


211

placed between the array slide and a plain coverslip and incubated for 1 h at room temperature. Alexa Fluor 546-labeled anti-rabbit IgG (20 mg/mL) was applied to detect bound rabbit primary antibodies. After washing, the heparin arrays were scanned with an LS400 scanner, and fluorescence spotter intensities were integrated using appropriate software. All the competition assays were carried out at least in duplicate.

4.5. Binding affinities of heparin oligosaccharide dendrimers to FGFs The binding affinity of FGF-2 to the dendrimers was analyzed by determining the IC50 values of soluble dendrimers with heparin-coated microarrays. Dendrimer HOD-1 exhibited more effective binding than monovalent oligosaccharide 1. Significant binding affinities were found with dendrimers HOD-3 and HOD-4 instead of negligible affinities with monosaccharide 12 and disaccharide 13. These findings indicated that HS/ heparin derivatized dendrimers were bound stronger by FGF-2 than monovalent oligosaccharides.

5. HS/Heparin Interaction with Chemokines as Determined by Microarray Analysis Chemokines are a family of small secreted proteins which are of extraordinary importance in lymphocyte migration and the recruitment of leukocyte subsets to sites of inflammation (Rot, 1992; Rot and von Andrian, 2004). Chemokines are released from a wide range of cells and function predominantly as chemoattractants for leukocytes; recruiting them from the blood to sites of infection or inflammation. Generating a profile of chemokine binding to defined HS/heparin sequences was intended to provide insight into chemokine function at the molecular level. Consequently, binding analyses of eight different chemokines (CCL21, IL-8, CXCL12, CXCL13, CCL19, CCL25, CCL28, and CXCL16) and a small library of HS/heparin oligosaccharides were conducted using HS/ heparin microarrays and SPR measurements (De Paz et al., 2007a).

5.1. Materials and equipment Sodium phosphate buffer (pH 9.0, 50 mM) Automated arraying robot (Perkin Elmer) NHS-activated slides (CodeLink, Amersham Biosciences) CXCL12, CCL19, CCL21, CXCL13, CCL28, CCL25, CXCL16, and anti-CXCL16 (PeproTech EC)

212


Anti-CXCL12 (Aviva Systems Biology) Anti-CXCL13 (eBioscience) Anti-CCL19 (Abgent) Anti-CCL21 (Abcam) Anti-CCL28 and anti-CCL25 (R&D System) IL-8 and anti-IL-8 (Novartis) AlexaFluro-546-labled anti-rabbit IgG and anti-goat IgG (Molecular Probes) HybriSlip Hybridization covers (Grace Bio-Labs) Nanopure water Fluorescence reader (LS400, Tecan) Scan Array Express software (Perkin Elmer) Gene Spotter (Microdiscovery GmbH) SPR measurement with BIAcore 3000 (BIAcore, Uppsala, Sweden) SPR express software (BIA control software) HBS-EP buffer (10 mM HEPES, pH 7.4, 150 mM NaCl, 3 mM EDTA, 0.005% (v/v) surfactant P20) and CM5 chips (BIAcore) Starburst PAMAM dendrimer generation 2.5 containing 32 sodium carboxylate surface groups and deaminated heparin (5 kDa) (Sigma-Aldrich)

5.2. Fabrication of HS/heparin microarray and incubation with chemokines An identical protocol to that described previously (Noti et al., 2006) was employed (De Paz et al., 2007a).

5.3. Binding affinities of HS/heparin with chemokines Significant binding of CCL21 to hexasaccharides 1, 2, and 5, tetrasaccharide 7, and monosaccharide 12 was observed. In contrast, binding to hexasaccharide 6 and disaccharide 13 was rather weak. CXCL13 displayed decreased affinity to all oligosaccharides compared with CCL21. Incubation of CXCL12 or CCL19 with the HS/heparin microarray demonstrated that these chemokines bound the synthetic HS/heparin oligosaccharides only weakly or not at all. A 5 kDa heparin sample was used as a control to confirm the known binding profiles of heparin with the chemokines. Our results suggest that synthetic HS/heparin oligosaccharides, hexamer length or less, which lack certain sulfation patterns, were dispensable for CXCL12 and CCL19 activation. Further studies with the SPR technology confirmed the above observations. Four other chemokines were tested: IL-8, CCL25, CCL28, and CXCL16. The IL-8 profile indicated that binding to sulfated oligosaccharides was not as a result of nonspecific charge–charge interactions, but that the 2-O-sulfate groups on the IdoA units were crucial for this specific


213

interaction. CCL25 and CCL28, two closely related chemokines expressed in epithelial cells, had similar binding profiles compared with the profiles generated for IL-8 and CCL21. This suggests that tetra- and hexasaccharides with a minimum of two sulfate groups on each disaccharide unit constitute a general recognition motif for these chemokines. Furthermore, this work indicated that N-sulfate and 6-O-sulfate groups played important roles in HS/heparin binding to CCL16. Chemotaxis assays that showed the effect of dendrimers on cells in vitro, suggested that the structure of HS/heparin-containing dendrimers could influence the design of chemokine-modulating agents (De Paz et al., 2007a).

6. HS/Heparin Microarray for Determination of Their Interaction with NCRs Natural killer (NK) cells are a type of cytotoxic lymphocyte that constitute a major component of the innate immune system, and play an important role in rejecting tumors and cells infected by viruses (Vivier et al., 2008). NK cell regulation is mediated by activating and inhibiting receptors on NK cell surfaces. The NCRs, NKp30, NKp44, and NKp46, are key for NK cell regulation, and are central to triggering the tumor cell recognition pathway (Moretta et al., 2001). Since carbohydrate–protein interactions are known to be important for NK cell targeting, we again used microarray technology (Hecht et al., 2009), and SPR measurements to investigate the potential for NKp30, NKp44, and NKp46 to bind synthetic HS/heparin.

6.1. Materials and equipment Sodium phosphate buffer (pH 9.0, 50 nM) Piezoelectric spotting robot (S11, Scienion) NHS-activated slides (CodeLink, Amersham Biosciences) AlexaFluro-647-labled a goat anti-human IgG (HþL) antibody (Invitrogen) HBS-N buffer; surfactant P20 and CM5 chips (BIAcore) HybriSlip Hybridization covers (Grace Bio-Labs) Array-ProAnalyzer software (MediaCybernetics) Nanopure water Fluorescence reader (LS400, Tecan) Gene Spotter (Microdiscovery GmbH) SPR measurement with BIAcore T100 (BIAcore) NCRs: NCRs were expressed as recombinant soluble IgG Fc chimeras. NKp30 and NKp44 consisted of a single IgG-like V-type domain fused to human IgG1. NKp46D2 was an IgG1 fusion protein containing only

214


the membrane-proximal D2 domain of NKp46. NKp30, NKp44, and NKp46D2 were expressed in CHO cells.

6.2. Fabrication of HS/heparin microarray and incubation with chemokines A protocol identical to that described previously was used (Hecht et al., 2009; Noti et al., 2006).

6.3. Binding affinities of HS/heparin to NCRs Both NKp30 and NKp46D2 bound the highly sulfated oligosaccharides 1, 2, 5, 7, and 10, which were all synthetic HS/heparin molecules with two to three sulfate groups on each disaccharide unit. All three NCRs bound to iduronic acid containing O-sulfate groups at the C2 and C4 position (monosaccharide 12). However, the NCRs did not bind to the remaining monosaccharides 10 or 11, showing that the NCRs prefer highly sulfated HS/heparin structures as binding partners. Additional SPR measurements suggested that HS/heparin oligosaccharides represent only fragments of the natural NCR binding epitopes. This was supported inhibition studies of NCR binding to tumor cells. Natural HS/heparin was a significantly more potent inhibitor of NCR binding than even the synthetic HS/heparin sequences that tightly bound the receptors in the microarray studies, indicating that the synthetic HS/heparin oligosaccharides were not identical to natural HS/heparin ligands.

7. Conclusions Carbohydrate microarrays have become important tools for glycomics as they save both time and materials in determining carbohydrate–protein interactions. The limiting factor for all glycan microarray studies is access to defined oligosaccharides. Access to a small collection of 13 synthetic HS/ heparin oligosaccharides provided a first example of the power of heparin microarrays. In a short time period, the binding specificities of several heparin-binding proteins were elucidated. Based on information gathered from the microarrays, novel multivalent displays such as heparin dendrimers were designed to modulate heparin activity in vivo. Currently, a host of other proteins are being screened for heparin oligosaccharide binding and the insights will allow us to correlate binding with biological function. While much has been achieved with the first set of heparin arrays, a more diverse set of synthetic GAG oligosaccharides is needed. To this end, we are currently advancing the automated synthesis of such complex


215

oligosaccharides. With more diverse structures in hand, screening by using the methods established in this chapter will yield more in-depth details about GAG–protein interactions and their biological implications.

ACKNOWLEDGMENTS We thank the Max Planck Society for very generous support and the European Research Council (ERC Advanced Grant to PHS). We thank all present and past members of Seeberger group who contributed to the results reported in this chapter.

REFERENCES Adams, E. W., Ratner, D. M., Bokesch, H. R., McMahon, J. B., O’Keefe, B. R., and Seeberger, P. H. (2004). Oligosaccharide and glycoprotein microarrays as tools in HIV glycobiology: Glycan-dependent gp120/protein interactions. Chem. Biol. 11, 875–881. Adibekian, A., Bindschaedler, P., Timmer, M. S. M., Noti, C., Schuetzenmeister, N., and Seeberger, P. H. (2007). De novo synthesis of uronic acid building blocks for assembly of heparin oligosaccharides. Chem. Eur. J. 13, 4510–4522. Arungundram, S., Al-Mafraji, K., Asong, J., Leach, III, F. E., Amster, I. J., Venot, A., Turnbull, J. E., and Boons, G.-J. (2009). Modular synthesis of heparan sulfate oligosaccharides for structure–activity relationship studies. J. Am. Chem. Soc. 131, 17394–17405. Blixta, O., Head, S., Mondala, T., Scanlan, C., Huflejt, M. E., Alvarez, R., Bryan, M. C., Fazio, F., Calarese, D., Stevens, J., Razi, N., Stevens, D. J., et al. (2004). Printed covalent glycan array for ligand profiling of diverse glycan binding proteins. Proc. Natl. Acad. Sci. USA 101, 17033–17038. Capila, I., and Linhardt, R. J. (2002). Heparin–protein interactions. Angew. Chem. Int. Ed. Engl. 41, 391–412. Cochran, S., Li, C. P., Fairweather, J. K., Kett, W. C., Coombe, D. R., and Ferro, V. (2003). Probing the interactions of phosphosulfomannans with angiogenic growth factors by surface plasmon resonance. J. Med. Chem. 46, 4601–4608. Dam, T. K., and Brewer, C. F. (2002). Thermodynamic studies of lectin–carbohydrate interactions by isothermal titration calorimetry. Chem. Rev. 102, 387–429. De Paz, J. L., and Seeberger, P. H. (2006). Recent advances in carbohydrate microarrays. QSAR Comb. Sci. 25, 1027–1032. De Paz, J. L., and Seeberger, P. H. (2008). Deciphering the glycosaminoglycan code with the help of microarrays. Mol. BioSyst. 4, 707–711. De Paz, J. L., Horlacher, T., and Seeberger, P. H. (2006a). Oligosaccharide microarray to map interaction of carbohydrates in biological system. Methods Enzymol. 415, 269–292. De Paz, J. L., Noti, C., and Seeberger, P. H. (2006b). Microarrays of synthetic heparin oligosaccharides. J. Am. Chem. Soc. 128, 2766–2767. De Paz, J. L., Spillmannb, D., and Seeberger, P. H. (2006c). Microarrays of heparin oligosaccharides obtained by nitrous acid depolymerization of isolated heparin. Chem. Commun. 3116–3118. De Paz, J. L., Moseman, E. A., Noti, C., Polito, L., von Andrian, U. H., and Seeberger, P. H. (2007a). Profiling heparin–chemokine interactions using synthetic tools. ACS Chem. Biol. 2, 735–744.

216


De Paz, J. L., Noti, C., Bo¨hm, F., Werner, S., and Seeberger, P. H. (2007b). Potentiation of fibroblast growth factor activity by synthetic heparin oligosaccharide glycodendrimers. Chem. Biol. 14, 879–887. Disney, M. D., and Seeberger, P. H. (2004a). The use of carbohydrate microarrays to study carbohydrate–cell interactions and to detect pathogens. Chem. Biol. 11, 1701–1707. Disney, M. D., and Seeberger, P. H. (2004b). Aminoglycoside microarray to explore interactions of antibiotics with RNAs and proteins. Chem. Eur. J. 10, 3308–3314. Disney, M. D., Magent, S., Blanchard, J. S., and Seeberger, P. H. (2004). Aminoglycoside microarray to study antibiotic resistance. Angew. Chem. Int. Ed. Engl. 43, 1591–1594. Feizi, T., Fazio, F., Chai, W., and Wong, C.-H. (2003). Carbohydrate microarray: A new set of technologies at the frontiers of glycomics. Curr. Opin. Struct. Biol. 13, 637–645. Gama, C. I., and Hsieh-Wilson, L. C. (2005). Chemical approaches to deciphering the glycosaminoglycan code. Curr. Opin. Chem. Biol. 9, 609–619. Gandhi, N. S., and Mancera, R. L. (2008). The structure of glycosaminoglycans and their interactions with proteins. Chem. Biol. Drug Des. 72, 455–482. Haag, R., and Kratz, F. (2006). Polymer therapeutics: Concepts and applications. Angew. Chem. Int. Ed. 45, 1198–1215. Hecht, M.-L., Rosental, B., Horlacher, T., Hershkovitz, O., De Paz, J. L., Noti, C., Schauer, S., Porgador, A., and Seeberger, P. H. (2009). Natural cytotoxicity receptors NKp30, NKp44 and NKp46 bind to different heparan sulfate/heparin sequences. J. Proteome Res. 8, 712–720. Horlacher, T., and Seeberger, P. H. (2008). Carbohydrate arrays as tools for research and diagnostics. Chem. Soc. Rev. 37, 1414–1422. Insight (2003). Proteomics. Nature 422, 191–237. Insight (2004). Human genomics and medicine. Nature 429, 439–481. Insight (2007). Glycochemistry & glycobiology. Nature 446, 999–1051. Karoli, T., Liu, L. G., Fairweather, J. K., Hammond, E., Li, C. P., Cochran, S., Bergefall, M., Trybala, E., Addison, R. S., and Ferro, V. (2005). Synthesis, biological activity, and preliminary pharmacokinetic evaluation of analogues of a phosphosulfomannan angiogenesis inhibitor (PI-88). J. Med. Chem. 48, 8229–8236. Laremore, T. N., Zhang, F., Dordick, J. S., Liu, J., and Linhardt, R. J. (2009). Recent progress and applications in glycosaminoglycan and heparin research. Curr. Opin. Chem. Biol. 13, 633–640. Laughlin, S. T., and Bertozzi, C. R. (2009). Imaging the glycome. Proc. Natl. Acad. Sci. USA 106, 12–17. Liang, P. H., Wu, C. Y., Greenberg, W. A., and Wong, C. H. (2008). Glycan arrays: Biological and medical applications. Curr. Opin. Chem. Biol. 12, 86–92. Lindahl, U., and Kjellen, L. (1991). Heparin or heparan sulfate—What is the difference? Thromb. Haemost. 66, 44–48. Lindahl, U., and Li, J.-P. (2009). Interaction between heparin sulfate and proteins—Design and functional implication. Int. Rev. Cell Mol. Biol. 276, 105–159. Lis, H., and Sharon, N. (1972). Lectins: Cell-agglutinating and sugar-specific proteins. Science 177, 949–959. Lohman, G. J. S., Hunt, D. K., Hoegermeier, J. A., and Seeberger, P. H. (2003). Synthesis of iduronic acid building blocks for the modular assembly of glycosaminoglycans. J. Org. Chem. 68, 7559–7561. Love, K. R., and Seeberger, P. H. (2002). Carbohydrate arrays as tools for glycomics. Angew. Chem. Int. Ed. 41, 3583–3586. Marson, A., Robinson, D. E., Brookes, P. N., Mulloy, B., Wiles, M., Clark, S. J., Fielder, H. L., Collinson, L. J., Cain, S. A., Kielty, C. M., McArthur, S., Buttle, D. J., et al. (2009). Development of a microtiter plate-based glycosaminoglycan array for the investigation of glycosaminoglycan–protein interactions. Glycobiology 19, 1537–1546.


217

Mercey, E., Sadir, R., Maillart, E., Roget, A., Baleux, F., Lortat-Jacob, H., and Livache, T. (2008). Polypyrrole oligosaccharide array and surface plasmon resonance imaging for the measurement of glycosaminoglycan binding interactions. Anal. Chem. 80, 3476–3482. Mohammadi, M., Olsen, S. K., and Goetz, R. (2005). A protein canyon in the FGF–FGF receptor dimer selects from an a` la carte menu of heparan sulfate motifs. Curr. Opin. Struct. Biol. 15, 506–516. Moretta, A., Bottino, C., Vitale, M., Pende, D., Cantoni, C., Mingari, M. C., Biassoni, R., and Moretta, L. (2001). Activating receptors and coreceptors involved in human natural killer cell-mediated cytolysis. Annu. Rev. Immunol. 19, 197–223. Murrey, H. E., and Hsieh-Wilson, L. C. (2008). The chemical neurobiology of carbohydrates. Chem. Rev. 108, 1708–1731. Noti, C., and Seeberger, P. H. (2005). Chemical approaches to define the structure–activity relationship of heparin-like glycosaminoglycans. Chem. Biol. 12, 731–756. Noti, C., De Paz, J. L., Polito, L., and Seeberger, P. H. (2006). Preparation and use of microarrays containing synthetic heparin oligosaccharides for the rapid analysis of heparin–protein interactions. Chem. Eur. J. 12, 8664–8686. Orgueira, H. A., Bartolozzi, A., Schell, P. H., Litjens, R., Palmacci, E. R., and Seeberger, P. H. (2003). Modular synthesis of heparin oligosaccharides. Chem. Eur. J. 9, 140–169. Ornitz, D. M., Herr, A. B., Nilsson, M., Westman, J., Svahn, C. M., and Waksman, G. (1995). FGF binding and FGF receptor activation by synthetic heparan-derived di- and trisaccharides. Science 268, 432–436. Pellegrini, L. (2001). Role of heparan sulfate in fibroblast growth factor signalling: A structural view. Curr. Opin. Struct. Biol. 11, 629–634. Polat, T., and Wong, C.-H. (2007). Anomeric reactivity-based one-pot synthesis of heparinlike oligosaccharides. J. Am. Chem. Soc. 129, 12795–12800. Powell, A. K., Yates, E. A., Fernig, D. G., and Turnbull, J. E. (2004). Interactions of heparin/heparan sulfate with proteins: Appraisal of structural factors and experimental approaches. Glycobiology 14, 17R–30R. Rele, S. M., Cui, W. X., Wang, L. C., Hou, S. J., Barr-Zarse, G., Tatton, D., Gnanou, Y., Esko, J. D., and Chaikof, E. L. (2005). Dendrimer-like PEO glycopolymers exhibit antiinflammatory properties. J. Am. Chem. Soc. 127, 10132–10133. Rot, A. (1992). Endothelial-cell binding of Nap-1/II-8-role in neutrophil emigration. Immunol. Today 13, 291–294. Rot, A., and von Andrian, U. H. (2004). Chemokines in innate and adaptive host defense: Basic chemokinese grammar for immune cells. Annu. Rev. Immunol. 22, 891–928. Seeberger, P. H., and Werz, D. B. (2007). Synthesis and medical applications of oligosaccharides. Nature 446, 1046–1051. Shin, L., Park, S., and Lee, M. (2005). Carbohydrate microarrays: An advanced technology for functional studies of glycans. Chem. Eur. J. 11, 2894–2901. Shipp, E. L., and Hsieh-Wilson, L. C. (2009). Profiling the sulfation specificities of glycosaminoglycan interactions with growth factors and chemotactic proteins using microarrays. Chem. Biol. 12, 731–756. Suda, Y., Arano, A., Fukui, Y., Koshida, S., Wakao, M., Nishimura, T., Kusumoto, S., and Sobel, M. (2006). Immobilization and clustering of structurally defined oligosaccharides for sugar chips: An improved method for surface plasmon resonance analysis of protein– carbohydrate interactions. Bioconjug. Chem. 17, 1125–1135. Timmer, M. S. M., Adibekian, A., and Seeberger, P. H. (2005). Short de novo synthesis of fully functionalized uronic acid monosaccharides. Angew. Chem. Int. Ed. 44, 7605–7607. Varki, A., Cummings, R. D., Esko, J. D., Freeze, H. H., Stanley, P., Bertozzi, C. R., Hart, G. W., and Etzler, M. E. (1999). Essentials of Glycobiology Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

218


Vivier, E., Tomasello, E., Baratin, M., Walzer, T., and Ugolini, S. (2008). Functions of natural killer cells. Nat. Immunol. 9, 503–510. Zhang, Z., McCallum, S. A., Xie, J., Nieto, L., Corzana, F., Jimenez-Barbero, J., Chen, M., Liu, J., and Linhardt, R. J. (2008). Solution structures of chemoenzymatically synthesized heparin and its precursors. J. Am. Chem. Soc. 130, 12998–13007.

C H A P T E R

T E N

Measurement of Glycan-Based Interactions by Frontal Affinity Chromatography and Surface Plasmon Resonance Chihiro Sato, Nao Yamakawa, and Ken Kitajima Contents 1. Introduction 2. Frontal Affinity Chromatography (FAC) as a Tool for the Measurement of Glycan-Based Interactions 2.1. FAC principle 2.2. Materials and equipment 2.3. Preparation of affinity adsorbents 2.4. Preparation of neurotransmitters 2.5. Operation of frontal affinity chromatography 2.6. Interaction between polySia and neurotransmitter 3. Surface Plasmon Resonance (SPR)-Based Biosensors as a Tool for the Measurement of Glycan-Based Interactions 3.1. Materials and equipment 3.2. Preparation of biotinylated glycans 3.3. Immobilization of biotinylated glycans on an Au sensor chip 3.4. Immobilization of BDNF on a CM5 sensor chip 3.5. Biacore analysis 4. Conclusions Acknowledgments References

220 221 221 224 224 225 225 226 227 227 228 228 229 229 230 231 231

Abstract Proteins and lipids are often modified with glycan chains, which due to their large hydration effect and structural heterogeneity, significantly alter the surface physicochemical properties of proteins and biomembranes. This ‘‘glycoatmosphere’’ also serves as a field for interactions with various molecules, Bioscience and Biotechnology Center, Graduate School of Bioagricultural Sciences, Nagoya University, Nagoya, Japan Methods in Enzymology, Volume 478 ISSN 0076-6879, DOI: 10.1016/S0076-6879(10)78010-1

#


219

220

Chihiro Sato et al.

including other glycans, lipids, peptides, proteins, and small molecules such as neurotransmitters and drugs as well as lectins. Therefore, sensitive techniques for measuring these glycan-based interactions are becoming more and more necessary, with the appropriate method largely depending on the interacting molecules. In this chapter, we focus on frontal affinity chromatography (FAC) and surface plasmon resonance (SPR) for examining polysialic acid-involved interactions with neurotransmitters and neurotrophins. FAC is characterized by its applicability to analyze weak interactions that are difficult to measure using conventional methods, and by the ease of principle and experimental procedures. SPR is advantageous due to the availability of suitable surface materials and for real-time monitoring with nonlabeled analytes.

1. Introduction Glycosylation is one of the major modifications of proteins and lipids, and a wide variety of glycosylations have been reported. However, the phenomenon of glycosylation remains somewhat mysterious due to a lack of appropriate methodologies for determining glycan structures as well as their biological functions. Recent methods of genetic perturbation for glycan-related enzymes have greatly impacted the understanding of the biological significance of glycans, even if the results are severe or benign. The underlying mechanisms which link gene expression to the resultant phenotypes, however, remain unknown. In this respect, it is important to understand glycan-based interactions with cellular components containing not only proteins, but also glycans, lipids, and other natural substances using appropriate analytical methods. In this chapter, we focus on the measurement of new polysialic acid (polySia)-based interactions using frontal affinity chromatography (FAC) and a surface plasmon resonance (SPR)-based Biacore instrument. PolySia is a polymerized structure of sialic acid present on neural cell adhesion molecules (NCAMs) as a posttranslational modification (Sato, 2004, 2010; Troy, 1996). PolySia-modified NCAM has been well studied in the development of the nervous system since the modification is spatiotemporally regulated (Bonfanti, 2006; Rutishauser, 2008). PolySia is expressed in embryonic brains during neural differentiation and mostly disappears in adult brains, although NCAM expression levels remain unchanged. Based on its bulky polyanionic nature, polySia is thought to function as an antiadhesive molecule against cell–cell and extracellular matrix interactions (Angata et al., 2006; Bonfanti, 2006; Rutishauser, 2008). However, we recently demonstrated by native PAGE analysis that an important neurotrophin in the brain, brain-derived neurotrophic factor (BDNF), interacts with polySia directly to form a large complex (Kanato et al., 2008). Therefore, polySia likely serves an important role as a

Measurement of Glyca-Based Interactions by Frontal Affinity

221

neurogenetic regulator similar to glycosaminoglycans that have been shown to bind growth factors and signaling molecules, such as Wnt (Schwartz and Domowicz, 2004). We thus proposed the hypothesis that, as a novel functional role, polySia may serve as a reservoir of neuroactive molecules, such as neurotrophins, growth factors, and neurotransmitters (Kanato et al., 2008, 2009). To gain insight into this potential novel function of polySia, we applied FAC and SPR for measuring polySia–neurotransmitter and polySia–neurotrophin interactions, respectively.

2. Frontal Affinity Chromatography (FAC) as a Tool for the Measurement of Glycan-Based Interactions Affinity chromatography is a commonly used and powerful tool for the purification of molecules that specifically interact with a target counterpart molecule. In the 1970s, Kasai and Ishii were the first to demonstrate that affinity chromatography is applicable to quantitative analysis for the estimation of dissociation constants for protein–ligand interactions (Kasai and Ishii, 1973). Kasai and his colleagues also applied FAC for the measurement of glycan–lectin interactions (Arata et al., 1997; Oda et al., 1981), which represented the first estimation of Kd. Hindsgaul et al. improved this method through the use of microcolumns and MS as a detector (Ng et al., 2005), allowing materials to be conserved and shortening the time required for analysis. This LC–MS system is particularly applicable for high-throughput screening. At the same time, Hirabayashi also developed an enhanced, high-throughput lectin–glycan interaction-analysis system between immobilized lectins and soluble fluorescent (PA)-labeled glycans (Kuno et al., 2005). Owing to improvements of these FAC-systems, FAC is a widely accepted technique for the analysis of lectin–glycan interactions. The FAC system has been well established and a number of wellwritten reviews are available (Hirabayashi et al., 2003; Kasai et al., 1986; Tateno et al., 2007). In this book, Kamiya and Kato also describe a lectin– glycan interaction revealed using a FAC system; in this chapter, we therefore focus on the general principle and methods for analyzing glycan-polymer and small molecule interactions.

2.1. FAC principle In frontal analysis, the elution front of an analyte and its concentration at a plateau level are measured when the analyte solution at various concentrations is eluted. For the analysis, the volume of the analyte solution should exceed the column volume. Under the assay conditions, as the free analyte

222

Chihiro Sato et al.

concentration is always fixed and equal to the initial concentration of analyte, dynamic equilibrium is achieved in the plateau region. Prior to the analysis, the column is packed with a resin on which the ligand of interest (termed B) is immobilized. The column is then isocratically eluted with an excess column volume of analyte (termed A) and the resulting elution curve of A is monitored. Based on the starting point of analyte elution (the elution front), the interaction between B and A can be detected. The retardation of elution (see line 2 in Fig. 10.1) indicates that analyte A interacts with B (Fig. 10.1). In contrast, an analyte which does not interact with B is eluted in the void volume of the column (see line 1 in Fig. 10.1). The generation of a retard volume (VV0) means a volume of an A–B complex formed as a result of specific interaction. Therefore, the area, [A]0(V V0), represents the amount of the A–B complex. Provided the column volume is u, the equation [A]0(VV0) ¼ u [AB] is realized. The dissociation constant (Kd) can be determined from Eq. (10.1), which is based on the equation: A þ B , AB. Kd ¼ ½A½B=½AB ¼ ½A0 ½B0 ½AB =½AB ¼ u½B0 =ðV V0 Þ ½A0 ð10:1Þ

Concentration of analyte

As [B]0 is the concentration of the immobilized ligand, and the effective ligand content can be obtained with the equation Bt ¼ u [B]0. Therefore, Eq. (10.1) can be followed by Eq. (10.2).

1 (no interaction with B) 2 (interaction with B)

[A]0

[A]0(V−V0)

0

V0

V

Elution volume

Figure 10.1 Schematic elution profiles of FAC. The immobilized ligand (B) is packed into a column and a volume in excess of the total column volume of analyte (A) is then eluted. Line 1 is the curve for the analyte that does not interact with B. The elution volume represents the void volume (V0) of the column. Line 2 is the curve for the analyte that interacts with B. The retard (VV0) indicates a specific interaction between A and B. The shaded area, [A]0(VV0), is the amount of the complex [AB].


Kd ¼ Bt=ðV V0 Þ ½A0

223

ð10:2Þ

The value for Bt is obtained from analyte concentration-dependent experiments using either a Lineweaver–Burk type plot, that is, 1/[A]0 versus 1/(V V0)[A]0, or a Woolf–Hofstee type plot, that is, (V V0) versus (V V0) [A]0. Equation (10.2) can be simplified to Eq. (10.3) where [A]0 (10 8 M) is negligible for the Kd value (e.g., >10 6 M). Kd ¼ Bt=ðV V0 Þ;

if ½A0 Kd

ð10:3Þ

The elution volume of an analyte (V ) is determined graphically as the volume corresponding to the elution point, which occurs at the half value of the plateau of the elution curve. V0 is determined as the V of the appropriate control sample without affinity for the immobilized ligand. The values of Bt and Kd are determined from the intercept of the axis and the slope of the fitted curves (Woolf–Hofstee-type plots, (V V0) vs. (V V0)[A]0) (Fig. 10.2), respectively, and Eq. (10.3) can be changed to Eq. (10.4). ½A0 ðV V0 Þ ¼ KdðV V0 Þ þ Bt

ð10:4Þ

Bt [A]0(V−V0)

Amount of complex

It should be noted that strong interactions cannot be measured with this system, as this method is based only on the retardation of elution and not the

Slope = −Kd

0

(V–V0) Retardation of analyte

Figure 10.2 The Woolf–Hofstee plot for the determination of Bt and Kd. Based on Eq. (10.4), [A]0 (V V0) ¼ Kd (V V0) þ Bt, the obtained data can be plotted. The x- and y-axes represent the retarded elution of analyte A (V V0) and the amount of A–B complex ([A]0 (V V0)), respectively. Therefore, from the intercept of y-axis, the Bt (effective ligand content) is calculated. The Kd value is obtained from the slope (¼ Kd).

224

Chihiro Sato et al.

complete loss of the analyte from the eluate. Therefore, this method is suited for the measurement of relatively weak interactions that are occasionally encountered, as in the case of glycan-based interactions.

2.2. Materials and equipment Polysialic acid (colominic acid) (Wako) Affigel 102 or Affigel Hz (Bio-rad) 1-ethyl-3-(3-diethylaminopropyl)-carbodiimide (Bio-rad) Phosphate-buffered saline (PBS) (pH 7.2; 10 mM; 137 mM NaCl; 2.7 mM KCl) 5. Neurotransmitters (epinephrine and dopamine) 6. Column (4.0 mm 10 mm, 126 ml, GL Science) 7. HPLC system (Rheodyne injector equipped with a 2 ml-PEEK sample loop and a UV-detector connected to an integrator) 1. 2. 3. 4.

2.3. Preparation of affinity adsorbents For the FAC analysis, two types of polySia (colominic acid)-immobilized resins are used which differed on whether polySia is immobilized through the reducing (Fig. 10.3A) or nonreducing (Fig. 10.3B) terminal end. To prepare the resin with reducing end-immobilized polySia, 5 ml (packed volume) of Affigel 102 in 25 mM Na2HPO4 (pH 6.0) are added to 2 ml of 50 mg/ml colominic acid, and the pH of the resulting solution is adjusted to 5.0 with 1 N HCl. After the addition of 8 mg 1-ethyl-3-(3-diethylaminopropyl)-carbodiimide, the reaction mixture is kept at 4 C for 4 h with gentle mixing by rotation. After washing with PBS, the gel is blocked with acetic anhydrite at room temperature for 30 min and washed with PBS and 1 M NaCl. Based on the amount of unbound polySia measured by the resorcinol method, the extent of immobilization of polySia is estimated to be 0.8 mmol/ml. To prepare the resin to which polySia is immobilized through the nonreducing terminal end (Fig. 10.3B), 5 ml (packed volume) of Affigel Hz in 50 mM sodium acetate buffer (pH 5.5) are added to 2 ml of 48 mg/ml periodate-oxidized colominic acid (prepared by incubation with 25 mM sodium periodate in 100 mM sodium acetate (pH 5.5) followed by desalting), and incubated at 25 C overnight with gentle mixing. After washing the gel with PBS, unbound colominic acid is measured by the resorcinol method to estimate the amount of immobilized colominic acid. The immobilization extent of colominic acid is estimated to be 0.7 mmol/ ml. The two types of resins are each packed into an empty column (4.0 mm 10 mm, 126 ml, GL Science) using a syringe. The prepared columns can be stored at 4 C until use.

225


A

H3C

O

H

C

N

O OH

COO− H3C

OH

O

OH

B

O H3C

C

Affi-Gel Hz

H

C

N

H N

O CH

O

H

OH

C

N N

OH

OH

O

N

Affi-Gel 102

OH

n

N

NH NH

O

COO−

H3C

O

H

C

N

O

O

OH

OH

COO−

OH

O

OH

n

NH

Figure 10.3 PolySia-immobilized beads used for FAC analysis. (A) polySia-affigel 102. (B) Affigel Hz-polySia.

2.4. Preparation of neurotransmitters A variety of neurotransmitters are commercially available. In this study, we selected epinephrine and dopamine for FAC analyses with the immobilized polySia ligand. For the analysis, each neurotransmitter is dissolved in PBS or an appropriate buffer at concentrations ranging from 0 to 30 nM.

2.5. Operation of frontal affinity chromatography A FAC system consists of a pump ( JASCO PU-980i), an injector with a sample loop, a column, and a UV detector ( JASCO 875-UV) connected to a chromato-PRO integrator (Run Time Corporation, Kanagawa, Japan). The 2 ml-sample loop (PEEK) and the column are either kept in an oven (CTO-6A, Shimadzu) or water bath at 25 or 37 C. Prior to analysis, the column is equilibrated with PBS at a flow rate of 0.125 ml/min until a flat baseline of absorbance was achieved. The sample injector is turned to the load position, and 10–20 ml of air is injected using a syringe to empty the sample loop completely. The analyte solution dissolved in PBS is then injected to fill the 2 ml-sample loop. Note that 50–100 ml of excess analyte

226

Chihiro Sato et al.

solution is required to completely fill the 2 ml-sample loop. The injector is then turned to the inject position to flow the analyte into the column at a rate of 0.125 ml/min. The analyte eluted from the column is monitored with a UV-detector and the elution curve is recorded for 8 min. After importing the recorded data into Microsoft Excel, the elution volume of the analyte (V) can be calculated, which represents the volume at which half of the plateau concentration is attained. To calculate V0, the elution data from the noninteracting control material (acetylcholine) is used.

2.6. Interaction between polySia and neurotransmitter Using either polySia-affigel 102 or affigel Hz-polySia, the interaction between polySia and the catecholamine neurotransmitters, i.e., epinephrine and dopamine could be observed. Typical elution profiles for epinephrine with polySia-affigel 102 at 37 C are shown in Fig. 10.4A. For the analyses, acetylcholine was used as a noninteracting neurotransmitter with polySia for the determination of V0. The VV0 values were measured for analyte concentrations ranging from 10 to 30 nM. Based on Eq. (10.4), the Kd value determined for epinephrine was 3.1 10 5 (M). Using the FAC, the interaction between two molecules can be examined under different conditions. For example, the effect of pH on the dopamine–polySia interaction was examined by varying the pH of the equilibration buffer and the analyte solution. The Kd value was affected by pH (Fig. 10.4B), indicating that the microenvironmental pH of the cell surface is important for the interaction between dopamine and polySia.

30 nM Acetylcholine

100

20 nM 10 nM

B 3.0

80

Kd(10−5)

% of plateau

A

60 40

2.5

20 0

0

0.2

0.4 0.6 V (ml)

0.8

1.0

2.0 5.5

6.0

6.5

7.0

7.5

8.0

8.5

pH

Figure 10.4 The interaction between polySia and catecholamine neurotransmitters as analyzed by FAC. (A) Typical elution profiles for epinephrine. The neurotransmitter was dissolved in PBS at a concentration of 10, 20, and 30 nM and 2 ml of each solution was applied to the column (126 ml) through the 2 ml-sample loop at a flow rate of 0.125 ml/min at 37 C. Each elution curve for epinephrine was superimposed on that of acetylcholine. The observed retardation of elution was dependent upon the concentration of epinephrine. (B) The Kd values for the interaction of dopamine–polySia at different pHs were calculated by FAC analysis. The Kd value for this interaction was dependent upon the pH of the solution.


227

3. Surface Plasmon Resonance (SPR)-Based Biosensors as a Tool for the Measurement of Glycan-Based Interactions An SPR-based biosensor was first reported in 1983 (Liedberg et al., 1983) and led to the development of the Biacore instrument. Although Biacore is the most commonly used, several other SPR-based instruments exist for analyzing molecular interactions. The principle of SPR is well documented in other reviews, including a recent one (Willander and Al-Hilli, 2009); thus, in this chapter, we focus on experimental procedures using a Biacore instrument to examine interactions between polySia and a neurotrophin. In order to reliably detect interactions by SPR, we recommend examining several sensor chips with respect to immobilization of the target ligand and the specificity of interaction. Biacore provides numerous types of sensor chips. For example, the CM5 chip is coated with carboxymethyldextran (100 nm thickness of dextran layer), and is the most commonly used because it effectively immobilizes ligands and low nonspecific binding can be achieved. Two other dextran-coated chips, CM4 and CM3, are available and contain a low amount of carboxyl groups and short chain-length dextran (30 nm thickness of dextran), respectively. The carboxyl groups in dextran are activated with either N-ethyl-N0 -[3-(dimethylamino)propyl] carbodiimide) (EDC) or N-hydroxysuccinimide (NHS) to allow conjugation with ligands through their -NH2, -SH, -COOH, and -CHO groups after the addition of appropriate reagents. Other special sensor chips are also available for the immobilization of tagged molecules. These include the SA chip, which is coated with a streptavidin-conjugated dextran for the immobilization of biotinylated ligands, and the NTA chip, which is coated with a NTA-conjugated dextran for immobilization of His-tagged ligands. Although CM5 is widely used, as the carboxymethyldextran coating on this chip contains many anions, it is important to confirm that observed interactions are specific and not due to nonspecific electrostatic forces. When nonspecific interactions are suspected, an Au chip without any coating should be used. In this chapter, we describe the use of the CM5 and Au chips for the immobilization of proteins and glycans, respectively.

3.1. Materials and equipment 1. 2. 3. 4.

Polysialic acid (polySia) (colominic acid) (Wako) Heparan sulfate (HS) (Seikagaku Co.) Tri-N-acetyl-chitotriose (!1GlcNAcb4!)3 (Seikagaku co.) Sensor Chip CM5 or Au (GE Healthcare)

228

Chihiro Sato et al.

Biotin-LC-hydrazide (Pierce) Sephadex G-25 (GE Healthcare) 4,4-thio-dibutylic acid (DBA, Aldrich) N-Ethyl-N0 -[3-(dimethylamino)propyl]carbodiimide) (EDC) N-Hydroxysuccinimide (NHS) Streptavidin HBS-EP (0.01 M HEPES, 0.15 M NaCl, 3 mM EDTA, 0.005% polysorbate 20 (v/v)), pH 7.4) 12. Brain-derived neurotrophic factor (BDNF, Almone) 13. Biacore 2000 (GE Healthcare) 5. 6. 7. 8. 9. 10. 11.

3.2. Preparation of biotinylated glycans To prepare biotinylated glycans, chitotriose (GlcNAc)3 (2 mg/ml), polySia (10 mg/ml), or heparan sulfate (HS) (10 mg/ml) in 50 mM sodium acetate buffer (pH 5.5) are mixed with biotin-LC-hydrazide (final concentration, 5 mM) dissolved in DMSO. After incubation at 50 C for 2 h, NaBH3CN in methanol (22.4 mM final concentration) is added to the reaction mixture. The biotinylated glycans are then applied to a Sephadex G-25 column (1.2 cm 60 cm) and eluted with water to remove free biotin.

3.3. Immobilization of biotinylated glycans on an Au sensor chip The Au sensor surface is washed once with acetone and after drying, the chip is immersed in 10 mM DBA in ethanol to form a self-assembly membrane (SAM) on the Au surface. After gently shaking for 30 min at room temperature, the sensor surface is washed with ethanol three times and allowed to dry. The chip is then placed in a solution of EDC and NHS (a 1:9 mixture of 130 mM EDC in water and 144 mM NHS in 1,4-dioxane) at room temperature for 30 min with gentle shaking to activate the SAM on the Au surface. After adding water, the surface is incubated for 5 min, and then washed the Au surface. The Au chip containing surface-activated SAM is placed on the sensor chip support using the sensor chip assembly unit, and is set in a Biacore 2000 instrument. After priming with water for 7 min, a 0.1 mg/ml streptavidin solution is loaded twice, each time for 7 min at a flow rate of 10 ml/min. The immobilized streptavidin is monitored by the resonance unit (RU) value and typically reaches 490–580 RU. To destroy the excess activated groups, 1 mM ethanolamine is injected into the system for 7 min. After washing with HBS-EP, the target biotinylated glycan (0.1 mg/ml in 500 mM HBS-EP) is injected to allow immobilization on the Au surface (Fig. 10.5). The captured glycans can be monitored and reach around 30 RU for (GlcNAc)3, 120 RU for polySia, and 120 RU for HS.

229


O H

O H H3 C

C N

OH OH OH

O

H COO− C 3

C N

O OH OH

O n

O

H

C

N

OH

O N H

H N

Bio tin Streptavidin

O

OH

H N

O C DBA*

S S Au sensor chip

Figure 10.5 PolySia-immobilized Au sensor chip using 40 -dithiodibutyric acid (DBA). DBA was incubated on an Au surface to make self-assembly membrane (SAM). The SAM was activated with NHS and EDC, and then streptoavidin was conjugated. By flowing biotinylated polySia, polySia-immobilized Au surface was applicable to measure interaction.

3.4. Immobilization of BDNF on a CM5 sensor chip To immobilize the neurotrophin BDNF, a research grade CM5 chip is set in a Biacore 2000. After washing with 40% glycerol, activation of the sensor chip surface is performed with a mixture of 400 mM EDC and 100 mM NHS for 7 min at a flow rate of 10 ml/min. Immediately after activation, a BDNF solution (5 ng/ml) in sodium acetate buffer (pH 5.0) is added. After the RU reaches an appropriate value, 1 mM ethanolamine is flowed for 7 min to destroy activated residues (Kanato et al., 2009).

3.5. Biacore analysis The interactions between BDNF and several glycans can be measured using a Biacore 2000 instrument. For the interaction of immobilized glycans with BDNF, varying concentrations of BDNF (0–220 nM) in HBS-EP are injected over the glycan-immobilized sensor chips at a flow rate of 20 ml/min. For the analysis of the interactions between immobilized BDNF and glycans, varying concentrations of polySia (0–80 mM) and HS (0–36 mM) in HBS-EP are injected over the BDNF-immobilized sensor chip at a flow rate of 20 ml/min. After 120 s, HBS-EP is flowed over the sensor surface to monitor the dissociation phase. Following 180 s of dissociation, the sensor surface is fully regenerated by the injection of 10 ml of 3 M NaCl. Using a range of polySia concentrations (0–80 mM) as the analyte and the BDNF-immobilized CM5 sensor chip, several sensorgrams can be obtained (Fig. 10.6). The polySia is flowed for 120 s at 20 ml/min for the association phase, and HBS-EP is then flowed for 120–300 s to monitor the dissociation phase. The sensorgrams allow not only the Kd value, but also the ka (M 1s 1) and kd (s 1) values to be calculated. The Kd value of polySia toward BDNF is 9.1 10 6 (M), whereas that of HS toward BDNF is 1.5 10 9 (M) (Kanato et al., 2009). Interestingly, HS displays nearly the identical affinity to BDNF and polySia as determined by gel-shift assays

230

Chihiro Sato et al.

Analyte

Buffer

200 150 80 mM

RU

100

40 mM 20 mM 10 mM 5 mM

50 0

0

50

100 150 200 250 300 Time (s)

Figure 10.6 SPR sensorgrams for the interaction between polySia and BDNF. BDNF was immobilized on a CM5 sensor chip. Several concentrations of polySia (5–80 mM) were flowed over the chip, and the sensorgrams were monitored. BDNF in HBS-EP (analyte) flowed for 0–120 s; HBS-EP (buffer) flowed for 120–300 s.

using native PAGE (Kanato et al., 2008, 2009). It is clearly indicated that polySia requires amine-groups of BDNF for binding and that HS does not require amine group of BDNF for binding. The binding modes of BDNF toward polySia and HS are different. The reverse mode of interaction, that is, immobilized glycans and flowing BDNF, can be also measured. We usually adopt the Au sensor chip for immobilization of polySia to exclude the relatively high affinity of BDNF for the dextran matrix on the CM5 chip. For both the polySiaand HS-immobilized Au sensor chips, BDNF (0–220 mM) is flowed at 20 ml/min for 120 s, followed by elution with HBS-EP. Based on the sensorgram of poySia or HS subtracted with that of (GlcNAc)3, the Kd values of polySia and HS were calculated to be 6.4 10 9 (M) and 2.5 10 9 (M), respectively. Notably, the dissociation constants of polySia obtained using glycan-immobilized and BDNF-immobilized chips are 1000 times different in magnitude, while those of HS are much the same. These results suggest that the binding mechanism is different between the BDNF-polySia and the BDNF-HS complexes.

4. Conclusions FAC has been widely used for analyzing lectin–carbohydrate interactions. The principle and the system of FAC are very simple and it has a merit for measurements of relatively low-affinity interactions that are often the case with


231

glycan-based interactions. This chapter features a new application of FAC to a measurement of interaction between glycans and small molecules. Indeed, the polySia–neurotransmitter interaction is demonstrated by this method. Thus, FAC can be applied to non-protein-based interactions that remain unnoticed and low-affinity interactions that are difficult to measure by other methods. With the FAC system, in addition to Kd value, its subtle changes depending on environmental conditions can be easily determined. In FAC, interactions are measured in the flow of analytes, and thus can mimic those in natural flow system such as bloodstream by changing the flow rate using appropriate HPLC pump. This chapter also features an improved protocol in Biacore to measure glycan-based interactions. Biacore has been widely used for glycan-based interactions using glycans immobilized onto the surface of CM-coated chip or flowing glycans as analytes. However, it is a problem that analytes sometimes bind to CM surface nonspecifically. In the improved protocol, we adopted a noncoated Au surface instead of the CM surface. Indeed, with this method, a novel specific interaction between polySia and BDNF together with the Kd value is demonstrated. In the fields of glycomics, researchers are searching for new glycan-based interactions with microarrays and other methods described in this book. Therefore, methodologies have become more and more important to understand quantitatively how specific and how strong the interaction occurs. Furthermore, not only binding but also releasing processes that may be regulated by the microenvironment of cell surface are crucial for functional regulation of the interacting molecule. To understand precise conditions of functioning of the glycan-based interactions, high-throughput and quantitative methods that can be applied under various conditions (flow or static, pH, salt, cations, and so on) would be important.

ACKNOWLEDGMENTS This research was supported in part by Grants-in-Aid for Scientific Research (C) (20570107) (to C. S.) from the Ministry of Education, Science, Sports and Culture and Grants-in-Aid for the Global COE Program: Advanced Systems Biology (to K. K. and N. Y.). We also thank Mr. Ryo Isomura and Miss Sayaka Ono for the results presented in this chapter.

REFERENCES Angata, K., Lee, W., Mitoma, J., Marth, J., and Fukuda, M. (2006). Cellular and molecular analysis of neural development of glycosyltransferase gene knockout mice. Methods Enzymol. 417, 25–37. Arata, Y., Hirabayashi, J., and Kasai, K. (1997). The two lectin domains of the tandemrepeat 32-kDa galectin of the nematode Caenorhabditis elegans have different binding properties. Studies with recombinant protein. J. Biochem. 121, 1002–1009.

232

Chihiro Sato et al.

Bonfanti, L. (2006). PSA-NCAM in mammalian structural plasticity and neurogenesis. Prog. Neurobiol. 80, 129–164. Hirabayashi, J., Arata, Y., and Kasai, K. (2003). Frontal affinity chromatography as a tool for elucidation of sugar recognition properties of lectins. Methods Enzymol. 362, 353–368. Kanato, Y., Kitajima, K., and Sato, C. (2008). Direct binding of polysialic acid to a brainderived neurotrophic factor depends on the degree of polymerization. Glycobiology 18, 1044–1053. Kanato, Y., Ono, S., Kitajima, K., and Sato, C. (2009). Complex formation of a brainderived neurotrophic factor and glycosaminoglycans. Biosci. Biotechnol. Biochem. 73, 2735–2741. Kasai, K., and Ishii, S. (1973). Unimportance of histidine and serine residues of trypsin in the substrate binding function proved by affinity chromatography. J. Biochem. 74, 631–633. Kasai, K., Oda, Y., Nishikata, M., and Ishii, S. (1986). Frontal affinity chromatography: Theory for its application to studies on specific interactions of biomolecules. J. Chromatogr. 376, 33–47. Kuno, A., Uchiyama, N., Koseki-Kuno, S., Ebe, Y., Takashima, S., Yamada, M., and Hirabayashi, J. (2005). Evanescent-field fluorescence-assisted lectin microarray: A new strategy for glycan profiling. Nat. Methods 2, 851–856. Liedberg, B., Nylander, C., and Lundstro¨m, I. (1983). Surface plasmon resonance for gas detection and biosensing. Sens. Actuators 4, 299–304. Ng, E., Yang, F., Kameyama, A., Palcic, M., Hindsgaul, O., and Schriemer, D. (2005). High-throughput screening for enzyme inhibitors using frontal affinity chromatography with liquid chromatography and mass spectrometry. Anal. Chem. 77, 6125–6133. Oda, Y., Kasai, K., and Ishii, S. (1981). Studies on the specific interaction of concanavalin A and saccharides by affinity chromatography. Application of quantitative affinity chromatography to a multivalent system. J. Biochem. 89, 285–296. Rutishauser, U. (2008). Polysialic acid in the plasticity of the developing and adult vertebrate nervous system. Nat. Rev. Neurosci. 9, 26–35. Sato, C. (2004). Chain length diversity of sialic acids and its biological significance. Trends Glycosci. Glycotech. 14, 331–344. Sato, C. (2010). Polysialic Acid. Bentham Science, UAE. (in press). Schwartz, N., and Domowicz, M. (2004). Proteoglycans in brain development. Glycoconj. J. 21, 329–341. Tateno, H., Nakamura-Tsuruta, S., and Hirabayashi, J. (2007). Frontal affinity chromatography: Sugar–protein interactions. Nat. Protoc. 2, 2529–2537. Troy, F. A. II. (1996). Sialobiology and the polysialic acid glycotope. In ‘‘Biology of the Sialic acid,’’ (Rosenerg, ed.). pp.95–144. Plenum press, New York. Willander, M., and Al-Hilli, S. (2009). Analysis of biomolecules using surface plasmons. Methods Mol. Biol. 544, 201–229.

C H A P T E R

E L E V E N

Detection of Weak-Binding Sugar Activity Using Membrane-Based Carbohydrates Kazuo Yamamoto and Norihito Kawasaki Contents 1. Introduction 2. Construction of Plasmids for Biotinylated Soluble Lectins 2.1. Materials 2.2. Methods 3. Preparation of Biotinylated Soluble Lectins and PE-Labeled Lectin Tetramer 3.1. Materials 3.2. Methods 4. Construction of Plasmids for the Fc-Fusion Protein and Purification of the Lectin–Fc Fusion Protein 4.1. Materials 4.2. Methods 5. Binding Assay for PE-Labeled Lectin Tetramer Using Flow Cytometry 5.1. Materials 5.2. Methods 6. Cells with Altered Glycans or Modification of Cell-Surface Glycans 6.1. Materials 6.2. Methods References

234 234 234 235 236 236 236 237 237 237 238 238 238 239 239 239 240

Abstract Protein–sugar interactions underlie many biological events. Although protein– sugar interactions are weak, they are regulated in physiological conditions including clustering, association with other proteins, pH condition, and so on. The elucidation of the precise specificities of sugar-binding proteins is essential for understanding their biological functions. To detect the weak-binding activity of carbohydrate-binding proteins to sugar ligands, we studied lectin tetramer binding Department of Integrated Biosciences, Graduate School of Frontier Sciences, The University of Tokyo, Chiba, Japan Methods in Enzymology, Volume 478 ISSN 0076-6879, DOI: 10.1016/S0076-6879(10)78011-3

#


233

234

Kazuo Yamamoto and Norihito Kawasaki

to cell-surface carbohydrates by flow cytometry. Tetramerization of lectins enhanced their avidity for sugar ligands, and sugar chains displayed on the cell surfaces were easily accessible to such soluble lectins. In this chapter, we describe methods to (1) prepare biotinylated soluble lectin, (2) obtain R-phycoerythrinlabeled lectin tetramer, and (3) measure tetramer binding to various lectin-resistant cell lines or cells treated with sugar-processing inhibitors. This approach enabled us to detect the weak sugar-binding activity of lectins (Ka 104 M 1), especially those from animals, and also to elucidate their specificity for sugar ligands.

1. Introduction Lectin is a protein that causes cell aggregation via surface sugar chains. Many plant lectins have been classified based on their sugar-binding specificities and they are a widely used tool for the purification and characterization of many glycoproteins, glycolipids, and some glycoconjugates. It is well known that many kinds of sugar-binding proteins are also present in animals. These proteins act as receptors for sugar-containing ligands, although they cannot induce cell aggregation. The elucidation of animal lectin ligands is essential for understanding various biological functions, such as sugar-mediated signaling. However, animal lectins are quite different from classical plant lectins in their sugar-binding ability; the Ka values of plant lectin–sugar interactions range approximately from 106 to 107 M 1, while those of animal lectin–sugars are much weaker (almost 104 M 1). Furthermore, although one can easily prepare large amounts of plant lectins, animal lectins cannot be detected without an antibody against them. Thus, several methods used to analyze plant lectins are not suitable for animal lectins. We have established a highly sensitive method to monitor the interaction of animal lectins and sugars (Hu et al., 2009; Kawasaki et al., 2007, 2008; Mikami et al., 2010; Yamaguchi et al., 2007).

2. Construction of Plasmids for Biotinylated Soluble Lectins 2.1. Materials Quikchange II site-directed mutagenesis kit (Qiagen) pBluescript II SK(þ) vector (Stratagene) pET-3c vector (New England Biolabs) Plasmid harboring lectin cDNA Primers for PCR DNA polymerase for PCR

235

Detection of Weak Lectin Activity

2.2. Methods To construct plasmids of a soluble lectin domain with an enzymatic biotinylation sequence, cDNA encoding a lectin domain was amplified by polymerase chain reaction (PCR) using appropriate primers and a lectin cDNA as a template. The amplified DNA was inserted between the Sma I sites of the pBluescript SK II vector. To introduce several amino acid substitutions to obtain sugar-binding defective mutants, the respective nucleotide(s) were substituted by PCR-based mutagenesis using a Quikchange II site-directed mutagenesis kit and mutated primers. A cDNA encoding the enzymatic biotinylation sequence GGGLNDIFEAQKIEWHE was introduced between Nde I and BamH I sites of pET-3c expression vector in Escherichia coli cells by ligation with a synthetic DNA of 50 -ggaattccatatggaattcccgggggcggtctgaacgacatcttcgaagctcagaaaatcgaatggcacgaataaggatccgcg-30 (Nde I, Sma I, and BamH I sites are underlined) (Wada et al., 2004) (pET-3cbio, Fig. 11.1). The cDNAs encoding a lectin domain were again PCR-amplified and ligated into between Nde I and Sma I sites of pET-3cbio (Fig. 11.1).

Nde I

Nde I

Lectin domain

Bio tag BamH I Nde I Sma I

Sma I

PCR Nde I

BamH I

Sma I

pET-3c

I Bam H

pET-3cbio

N

de

I

Sma I

Sma I

Lectin domain

pET-3cbioLec

Figure 11.1 Preparation of plasmids encoding a soluble lectin domain with a biotinylation tag. PCR-amplified or synthetic biotin-tagged DNA with Nde I and BamH I sites at 50 - and 30 -end, respectively, is inserted between the Nde I and BamH I sites of pET-3c (pET-3cbio). After digestion with both Nde I and Sma I, the PCR-amplified lectin domain cDNA is inserted between the Nde I and Sma I sites of pET-3cbio (pET3cbioLec).

236


3. Preparation of Biotinylated Soluble Lectins and PE-Labeled Lectin Tetramer 3.1. Materials BL21(DE3)pLysS E. coli cells Isopropyl b-thiogalactopyranoside Solubilization buffer: 50 mM Tris–HCl, pH 8.0, containing 6 M guanidine, 1 mM DTT, and 0.1 mM EDTA Refolding buffer: 100 mM Tris–HCl, pH 7.5, containing 0.4 M L-arginine, 5 mM reduced glutathione, 0.5 mM oxidized glutathione, and 0.5 mM phenylmethanesulfonyl fluoride (PMSF) Dialysis buffer: 20 mM Tris–HCl, pH 7.5, containing 25 mM NaCl and 0.1 mM EDTA UNO Q-6 ion-exchange column (12 mm 53 mm; Bio-Rad) UNO Q buffer A: 20 mM Tris–HCl, pH 7.5, containing 25 mM NaCl UNO Q buffer B: 20 mM Tris–HCl, pH 7.5, containing 500 mM NaCl Biotin ligase BirA (Avidity) Superdex-75 10/300 GL column (10 mm 300 mm; GE Healthcare BioSciences) HBS: 20 mM HEPES–NaOH, pH 7.4, containing 150 mM NaCl R-phycoerythrin (PE)-conjugated streptavidin (BD Biosciences)

3.2. Methods The soluble lectin domain with a C-terminal biotinylation tag was expressed in the BL21(DE3)pLysS strain of E. coli in the presence of 1 mM isopropyl b-thiogalactopyranoside, and recovered as soluble proteins or inclusion bodies. Recovered inclusion bodies were solubilized in solubilization buffer, diluted with refolding buffer to a protein concentration of 6 mM, and refolded in vitro by dialysis against dialysis buffer at 4 C for 24 h. Instead of EDTA, several metal ions were occasionally added to the refolding and dialysis buffers to enhance the correct folding of denatured polypeptides. The pH condition was critical to obtaining large amounts of recombinant proteins. The dialyzed fraction was applied to a UNO Q-6 column, and then equilibrated with 20 mM Tris–HCl, pH 7.5, containing 25 mM NaCl. The column was eluted with 18 ml of a linear gradient of NaCl from 25 to 500 mM in the same buffer. The purity of each fraction was analyzed by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) according to the method of Laemmli. Purified proteins fused to a C-terminal biotinylation tag were biotinylated with a biotin ligase, BirA (Fig. 11.2). The remaining free biotin was removed by gel filtration using a

237


Biotin ligase PE-labeled SA Biotinylation tag

Biotin

Figure 11.2 Strategies for preparing PE-labeled lectin tetramer. The purified recombinant soluble lectin domain with a C-terminal biotinylation tag is enzymatically biotinylated with BirA, and then tetramerized with PE-labeled streptavidin (PE-SA).

Superdex-75 10/300 GL column, and eluted with 20 mM HEPES–NaOH, pH 7.4, containing 150 mM NaCl (HBS). Biotinylation was confirmed with a gel-shift assay using SDS-PAGE of the sample without boiling (Kawasaki et al., 2007). To prepare the R-PE-labeled soluble lectin tetramer, biotinylated lectin was mixed with PE-conjugated streptavidin (PE-SA) at a molar ratio of 4:1 for 1 h on ice (Fig. 11.2). If it was difficult to prepare the soluble lectin domain in E. coli cells, then a lectin domain fused with human IgG-Fc was used instead (Yamaguchi et al., 2010), although it should be noted that the avidity of the dimeric Fc-fusion protein is weaker than that of tetramer complexed with streptavidin (Knibbs et al., 1998).

4. Construction of Plasmids for the Fc-Fusion Protein and Purification of the Lectin–Fc Fusion Protein 4.1. Materials pRc/CMV vector (Invitrogen) HEK293T cell Lipofectamine 2000 (Invitrogen) Protein G-Sepharose (GE Healthcare) Equilibration buffer: 20 mM NaOAc, pH 5.6, containing 150 mM NaCl Elution buffer: 100 mM glycine–HCl, pH 2.8 Anti-human Fc antibody (Zymed)

4.2. Methods cDNA fragment encoding the Fc segment of human IgG1 was cloned into a pRC/CMV vector (pRc/CMV-Fc). The cDNA corresponding to the lectin domain was PCR-amplified using primers and a lectin cDNA as a

238


template. The PCR product was then inserted into a pRc/CMV-Fc to generate a lectin–Fc fusion protein (pRc/CMV-LecFc). HEK293T cells were transfected with pRc/CMV-LecFc using Lipofectamine 2000, and culture supernatants were collected. Expression of lectin–Fc fusion protein in the culture supernatant was confirmed by Western blotting using an antihuman Fc antibody. To purify the lectin–Fc fusion protein, the culture supernatant was applied to a Protein G-Sepharose column equilibrated with equilibration buffer, washed with the same buffer, and eluted with elution buffer. The eluted fraction was neutralized with 1 M Tris–HCl, pH 9.5.

5. Binding Assay for PE-Labeled Lectin Tetramer Using Flow Cytometry 5.1. Materials HEPES buffered saline (HBS): 20 mM HEPES–NaOH, pH 7.4, containing 150 mM NaCl and 1 mM EDTA BSA-containing HBS (HBSB): HBS containing 0.1% NaN3 and 0.1% bovine serum albumin U bottom 96-well plate (Millipore) Propidium iodide (PI) FACSCalibur Flow Cytometer (BD Biosciences) Monosaccharides and oligosaccharides for inhibition

5.2. Methods Cultured mammalian cells were harvested, washed with HBS, and suspended in HBSB at a concentration of 2 107 cells/ml. Ten microliters of the cells were incubated with 10–100 mg/ml PE-labeled soluble lectin tetramer in HBSB at 25 C for 30 min in a 96-well plate. After washing twice with HBS, cells were suspended in 200 ml of HBS containing 1 mg/ml PI. The fluorescence of stained cells was measured using flow cytometry. The cell-surface fluorescence at 575 nm associated with PE was then recorded (Fig. 11.3). In total, 104 live cells gated by forward and side scattering and PI exclusion were acquired for analysis. To determine the divalent cation dependency of the tetramer binding, 1 mM of metal ions was added to HBSB instead of 1 mM EDTA. To test the effect of exogenous mono- and oligosaccharides on the binding of PE-labeled soluble lectin tetramer to the cells, tetramer was preincubated with various concentrations of mono- or oligosaccharides at 25 C for 30 min before being added to the cells.

239


KIF

KIF/endo H

Cell number

None

100

101

102

103

104

100 101 102 103 104 100 Fluorescence intensity

101

102

103

104

Figure 11.3 Soluble VIP36 tetramer bound to high molecular weight high mannosetype glycans on the cell surfaces. HeLaS3 cells (none), cells treated with 2 mg/ml kifunensine for 24 h (KIF), or kifunensine-treated cells subjected to endo H digestion (KIF/endo H) were incubated with 10 mg/ml PE-labeled soluble VIP36 tetramer (filled histogram) or PE-SA as a control (thin line) in the presence of 1 mM CaCl2, and then analyzed by flow cytometry.

6. Cells with Altered Glycans or Modification of Cell-Surface Glycans 6.1. Materials Cultured cell lines (CHO, Lec1, Lec2, Lec4, Lec8, pgsA-745, pgsB-618, pgsC-605, BHK, RicR14, RicR15, and RicR21) (American Type Culture Collection) Castanospermine (CST) (Sigma-Aldrich) Deoxynojirimycin (DNJ) (Sigma-Aldrich) Deoxymannojirimycin (DMJ) (Sigma-Aldrich) Kifunensine (KIF) (Calbiochem) Swainsonine (SW) (Calbiochem) Endo-b-N-acetylglucosaminidase H (endo H, New England Biolabs)

6.2. Methods Several kinds of cells with defects in sugar chain processing have been established. Lec1, Lec2, Lec4, and Lec8 cells are lectin-resistant CHO mutants defective in GlcNAc-TI, CMP-sialic acid transporter, GlcNAcTV, and UDP-Gal transporter, respectively. The pgsA-745, pgsB-618, and pgsC-605 mutants of CHO cells have defects in xylosyltransferase I, galactosyltransferase I, and the sulfate transporter involved in galactosaminoglycan synthesis, respectively. The RicR14, RicR15, and RicR21 mutants of BHK cells are defective in GlcNAc-TI, a-mannosidase II, and GlcNAc-TII,

240


respectively. These are useful target cells to measure binding and to determine the sugar-binding specificity of lectin. To modify cell-surface N-glycans, cells were treated for 24 h with 1 mM CST, 1 mM DNJ, 1 mM DMJ, 2 mg/ml KIF, or 10 mg/ml SW (Fig. 11.3). To remove high mannose-type N-glycans from the cell surface, endo-b-Nacetylglucosaminidase H (endo H, 5.0 103 U) was added to 2 106 cells suspended in 500 ml of HBSB and incubated at 37 C for 3 h (Fig. 11.3). Alternatively, cells expressing specific sugar chains are available by transfection with cDNA encoding carbohydrate-modifying enzyme (Mitoma and Fukuda, 2006). Using these kinds of cells, it is possible to determine the sugar-binding specificity of the lectin.

REFERENCES Hu, D., Kamiya, Y., Totani, K., Kamiya, D., Kawasaki, N., Yamaguchi, D., Matsuo, I., Matsumoto, N., Ito, Y., Kato, K., and Yamamoto, K. (2009). Sugar-binding activity of the MRH domain in ER-glucosidase II b subunit is important for efficient glucose trimming. Glycobiology 19, 1127–1135. Kawasaki, N., Matsuo, I., Totani, K., Nawa, D., Suzuki, N., Yamaguchi, D., Matsumoto, N., Ito, Y., and Yamamoto, K. (2007). Detection of weak sugar binding activity of VIP36 using VIP36–streptavidin complex and membrane-based sugar chains. J. Biochem. 141, 221–229. Kawasaki, N., Ichikawa, Y., Matsuo, I., Totani, K., Matsumoto, N., Ito, Y., and Yamamoto, K. (2008). The sugar-binding ability of ERGIC-53 is enhanced by its interaction with MCFD2. Blood 111, 1972–1979. Knibbs, R. N., Takagaki, M., Blake, D. A., and Goldstein, I. J. (1998). The role of valence on the high-affinity binding of Griffonia simplicifolia isolectins to type A human erythrocytes. Biochemistry 37, 16952–16957. Mikami, K., Yamaguchi, D., Tateno, H., Hu, D., Qin, S., Kawasaki, N., Yamada, M., Matsumoto, N., Hirabayashi, J., Ito, Y., and Yamamoto, K. (2010). The sugar-binding ability of human OS-9 and its involvement in ER-associated degradation. Glycobiology 20, 310–321. Mitoma, J., and Fukuda, M. (2006). Expression of specific carbohydrates by transfection with carbohydrate modifying enzymes. Methods Enzymol. 416, 293–304. Wada, H., Matsumoto, N., Maenaka, K., Suzuki, K., and Yamamoto, K. (2004). The inhibitory NK cell receptor CD94/NKG2A and the activating receptor CD94/ NKG2C bind the top of HLA-E through mostly shared but partly distinct sets of HLA-E residues. Eur. J. Immunol. 34, 81–90. Yamaguchi, D., Kawasaki, N., Matsuo, I., Totani, K., Tozawa, H., Matsumoto, N., Ito, Y., and Yamamoto, K. (2007). VIPL has sugar-binding activity specific for high-mannnosetype N-glycans, and glucosylation of the a1, 2 mannotriosyl branch blocks its binding. Glycobiology 17, 1061–1069. Yamaguchi, D., Hu, D., Matsumoto, N., and Yamamoto, K. (2010). Human XTP3-B binds to a1-antitrypsin variant null Hong Kong via the C-terminal MRH domain in a glycan-dependent manner. Glycobiology 20, 348–355.

C H A P T E R

T W E LV E

Fluorescence-Based Solid-Phase Assays to Study Glycan-Binding Protein Interactions with Glycoconjugates ¨nen* and Richard D. Cummings† Anne Leppa Contents 1. Overview 2. Biotinylation of Glycopeptides, Oligosaccharides, and Cells 2.1. Biotinylation of glycopeptides through cysteine residues 2.2. Biotinylation of glycopeptides through primary amines 2.3. Biotinylation of oligosaccharides 2.4. Biotinylation and fixation of HL-60 cells 3. Fluorescence Labeling of GBPs and Cells 3.1. Labeling of Gal-1 through primary amines 3.2. Labeling of Gal-1 through cysteines 3.3. Labeling of tomato (LEA) lectin through carbohydrates 3.4. Labeling of human T lymphocytes 4. P- and L-Selectin Binding to Immobilized Glycosulfopeptides and Determination of Apparent Binding Affinity 4.1. An assay with recombinant P- and L-selectin 4.2. Determination of apparent binding affinity for L-selectin 4.3. An assay with T lymphocytes 5. Galectin-1 Binding to Immobilized Glycopeptides and Glycans and Determination of Apparent Binding Affinity 5.1. An assay with recombinant galectin-1 5.2. Determination of apparent binding affinity 6. Galectin-1 Binding to Immobilized HL-60 Cells and Determination of Apparent Binding Affinity 6.1. An assay with recombinant galectin-1 6.2. Determination of apparent binding affinity Acknowledgments References

242 244 245 245 245 247 247 247 248 248 248 249 249 250 252 254 255 257 257 257 259 262 262

* Department of Biosciences, Division of Biochemistry, University of Helsinki, Viikinkaari, Helsinki, Finland Department of Biochemistry, Emory University School of Medicine, Atlanta, Georgia, USA

{


#


241

242

¨nen and Richard D. Cummings Anne Leppa

Abstract Development of glycan microarray technologies have recently revealed many new features in the binding specificities of glycan-binding proteins (GBPs) including animal and plant lectins, antibodies, toxins, and pathogens, including viruses and bacteria. Printed glycan microarrays are very sensitive, robust, and require very small quantities of glycans and GBPs. However, glycan arrays have been limited mostly to chemoenzymatically synthesized oligosaccharides and Nglycans isolated from natural glycoproteins. O-Glycans and more complex glycoconjugates, such as glycopeptides or whole cells, are generally lacking from most types of glycan microarrays. Certain GBPs such as selectins, that have more complex binding specificity, require peptide components besides the glycan structure for high-affinity binding to the ligand. GBP binding assays on glycan microarrays will provide only partial information about the specificity and high-affinity ligands for those GBPs. Therefore, more ‘‘natural’’ glycoconjugate arrays are required to study more complex GBP–glycoconjugate interactions. We have utilized a simple fluorescence-based solid-phase assay on a microplate format to study GBP–glycoconjugate interactions. The method utilizes commercial streptavidin-coated microplates, where various biotinylated ligands, such as glycopeptides, oligosaccharides, and whole cells, can be immobilized at a defined density. The binding of GBPs to immobilized ligands can be studied using fluorescently labeled GBPs or cells, or bound GBPs can be detected using fluorescently labeled anti-GBP antibodies. Our approach utilizing biotinylated and fixed cells in a solid-phase assay is a versatile method to study binding of GBPs to natural cell-surface glycoconjugates. Not only mammalian cells, but also microorganisms can be biotinylated and fixed, and adhesion of fluorescently labeled GBPs and antibodies to immobilized cells can be studied using standard streptavidin-coated microplates. Here, we present examples of fluorescence-based solid-phase assays to study P- and L-selectin and galectin-1 binding to immobilized glycopeptides, oligosaccharides, and cells. It should be noted that with the availability of complex glycoconjugates containing available primary amine groups, such as semisynthetic glycopeptides described here, that these could also be printed on covalent microarrays for interrogation by GBPs.

1. Overview Glycan microarray technologies have developed rapidly within recent years and at the same time knowledge on the binding specificity of GBPs has increased enormously (Blixt et al., 2004; de Boer et al., 2007; Song et al., 2008, 2009b,c; Xia et al., 2005). Glycan microarray binding data has revealed many new features on the binding specificity of mammalian GBPs such as galectins (Song et al., 2009d), siglecs (Blixt et al., 2008) and P-type lectins (Song et al., 2009a) and helped to identify novel GBPs, such as malectin (Schallus et al.,

Fluorescence-Based Solid Phase Assays

243

2008), and novel biological activities of GBPs, such as the ability of some tandem-repeat galectins (galectin-4 and -8) to bind blood group antigens on bacteria and directly kill them (Stowell et al., 2010). Printed glycan microarrays provide a high-throughput method to study GBP–glycan interactions with very small quantities of materials. However, glycan microarrays have been limited mostly to chemoenzymatically synthesized oligosaccharides containing relatively small sizes and often simple structures with only the most terminal monosaccharide residues of natural ligands, and thus lacking the backbone and core structures (Blixt et al., 2004). Because GBPs often recognize glycan backbone components besides terminal residues, the binding data on relatively simple glycan microarrays may not reveal the specificity and nature of highaffinity ligands. Therefore, there is a need for more natural glycan microarrays representing the display of glycans in a more relevant presentation, such as glycopeptides and glycolipids. Recently, first ‘‘natural’’ glycan arrays have been developed to contain N-glycans isolated from natural glycoproteins (de Boer et al., 2007; Song et al., 2009a,b,c,d), or natural glycolipids (de Boer et al., 2007; Liu et al., 2009). Natural glycan arrays have, for example, revealed differences in the specificity of galectins for complex N-glycans (Song et al., 2009d). However, O-glycans are often lacking from typical glycan microarrays, likely because O-glycans cannot be released from glycoproteins enzymatically and chemical methods are required for cleavage. Moreover, only those methods that release O-glycans in reducing form are applicable because the reducing monosaccharide unit must preserve the intact ring structure for coupling purposes. Some GBPs, such as selectins, recognize O-glycan structures on their glycoprotein ligands. For example, P-selectin has a very complex binding mechanism and requires peptide components besides a specific O-glycan at the extreme N-terminus of P-selectin glycoprotein ligand-1 (PSGL-1) for high-affinity binding (Leppa¨nen et al., 2000; Somers et al., 2000). We have utilized a simple fluorescence-based solid-phase assay on a microplate format to study more complex GBP–glycoconjugate interactions. Synthetic biotinylated glycopeptides, oligosaccharides, and cells were immobilized onto the streptavidin-coated microplates and probed with fluorescently labeled GBPs, and in some cases with fluorescently labeled cells. For example, a panel of biotinylated synthetic glyco(sulfo)peptides (GSPs) modeled after N-terminus of PSGL-1 was utilized to study the role of O-glycosylation and tyrosine sulfation for binding to P- and L-selectin (see below and Fig. 12.1A). As another example, the role and the mode of presentation of poly-N-acetyllactosamine (poly-LN) structures for galectin-1 binding were studied using biotinylated O-glycopeptides, oligosaccharide, and cells (see below and Fig. 12.1B). Examples of basic protocols, including methods to biotinylate glycopeptides, oligosaccharides, and cells, and methods to label GBPs and cells with fluorescent probes will be presented first. Please also read Chapter 19 in Volume 480 of this series, ‘‘Use of glycan microarrays to explore specificity of glycan-binding proteins’’ by David Smith, Xuezheng Song, and Richard Cummings.

244


A

L-/P-selectin binding to immobilized glycosulfopeptides Fluorescently-labeled anti-human IgG L-/P-selectin-Ig

Fluor. labeled T cell

L-selectin

Biotin with spacer Immobilized glycosulfopeptides B

Immobilized glycosulfopeptides

Galectin-1 binding to immobilized oligosaccharides and cells Fluorescently-labeled galectin-1 Fluorescently-labeled galectin-1

Biotin with spacer Immobilized oligosaccharides

HL-60 cell

Immobilized HL-60 cells

Figure 12.1 Examples of fluorescence-based solid-phase assays to study GBP– glycoconjugate interactions on streptavidin-coated microplates. (A) L-/P-selectin binding to immobilized glycosulfopeptides; (B) Galectin-1 binding to immobilized glycans and cells.

2. Biotinylation of Glycopeptides, Oligosaccharides, and Cells Various types of glycoconjugates can be biotinylated quantitatively using different types of coupling chemistries. For example, glycopeptides can be biotinylated through primary amine groups present at the amino terminus of each peptide chain and on lysine side chains. Glycopeptides can also be biotinylated through reduced cysteine residues, or carboxyl groups present at the carboxy termini of each peptide chain and on aspartate and glutamate residues. Reducing glycans can be biotinylated by reductive amination through reactive aldehyde at the reducing end under conditions


245

that favor opening of the reducing end monosaccharide residue. Cellsurface proteins of intact cells are usually biotinylated through primary amines. Examples of biotinylation protocols for each type of glycoconjugate will be given below.

2.1. Biotinylation of glycopeptides through cysteine residues Synthetic GSPs modeled after N-terminus of human PSGL-1 were biotinylated through C-terminal cysteine residue (Fig. 12.2) (Leppa¨nen et al., 2002). Reduced GSPs were biotinylated by incubating peptides with a 2–20-fold molar excess of EZ-Link Biotin-HPDP (Pierce) in 100 ml of PBS containing 1 mM EDTA for 1–2 h at room temperature or overnight at þ4 C. The completeness of the reaction was confirmed by HPLC in an analytical reversed-phase C-18 column under eluent conditions that clearly separated biotinylated peptide product from nonbiotinylated peptide and free Biotin-HPDP. Biotinylated peptides were separated from excess Biotin-HPDP by HPLC and dried in vacuo. Biotinylated GSPs were dissolved in physiological buffer and the concentration of each peptide solution was determined by UV absorbance at 215 nm of a sample subjected to HPLC.

2.2. Biotinylation of glycopeptides through primary amines Poly-LN O-glycopeptides modeled after N-terminus of human PSGL-1 were biotinylated through the N-terminal primary amine group (Fig. 12.2) (Leppa¨nen et al., 2005). Glycopeptides were dissolved in PBS and incubated with a 10-fold molar excess of EZ-Link NHS-LC-LC-Biotin (Pierce) overnight at room temperature. Biotinylated glycopeptides were purified by reversed-phase HPLC as described above.

2.3. Biotinylation of oligosaccharides Reducing glycans can be biotinylated through their reducing terminus by reductive amination. Oligosaccharides (final concentration 1 mM) were incubated with 12.5 mM EZ-Link Biotin-LC-hydrazide (Pierce) and 0.25 M sodium cyanoborohydride in acetic acid/DMSO (3:7, v/v) overnight at þ60 C. Biotinylated oligosaccharides were purified by reversedphase HPLC under eluent conditions that clearly separated biotinylated oligosaccharide products from free Biotin-LC-hydrazide. Glycans in Fig. 12.2 were biotinylated using 2-azidoethyl glycoside derivatives of oligosaccharides as described (Leppa¨nen et al., 2005).

246


PSGL-1 O-glyco(sulfo)peptides

Poly-LN oligosaccharides b4

a3 – – – SO3 SO3 SO3

GSP-6

b3 a1

b6

E Y E Y L D Y D F L P E T E P P E C

46 48

51

57

b3 b4

b4 b3 b4

b3 b4

LN3

a3 b4 b3 b4

b3 b4

S3LN3

b4

a3

GSP(46)-6

Biotin b4

LN2 -Biotin

a3

SO–3

b3 a1

b6


46

LNnT

GSP(48)-6


48

-Biotin

57 b4

SO–3

GSP(51)-6

b3 a1

b6


51

GSP(46,48)-6

46 48

57

SO–3

E Y E Y L D Y D F L P E T E P P E M

b3 b4

-Biotin


b4 b3

-Biotin

b4

57

b3 b4

a3 a3

SO–3

GSP(48,51)-6

SO–3

b3 a1


48

51

GP-4 Biotin


-Biotin

b4

57

b3

a3 a3

b4 b3

b4 b3 a1

b6

GP-6


b4 b3

-Biotin

b4

57 b4 b6

GP-4

b6 b3 a1


GP-4 Biotin

a1


57

b3 a1


-Biotin Key

57

GP-1

b3 a1

b6

b4 b6

a1


b4 b3 a1

b3

b6

GP-4 Biotin

b6

51

b3 a1

b6

GP-4 Biotin

a3

46

b4

b4 b3 a1

a3

GSP(46,51)-6

Biotin

b4 b6


SO–3

Biotin b3 b4

Poly-LN O-glycopeptides

a3

SO–3

NGLN3

-Biotin

57 a3

SO–3

Biotin

b3 b4 b3 b4

NGLNnT

a3 a3

Biotin

b3 b4

NGLN2 b3 a1

Biotin

Degalactosylated poly-LN oligosaccharides

b4 b6

Biotin

-Biotin

57 a3

Biotin

b4 b3 b4

a3 – SO3

b4

LN

a3

-Biotin

GalNAc

Gal

GlcNAc

Glc

NeuNAc

Man

Fuc SO3

Sulfate

Figure 12.2 Structures of biotinylated glyco(sulfo)peptides (GSPs) and oligosaccharides. This research was originally published in the Journal of Biological Chemistry (Lepp€anen et al., 2003, 2005). # The American Society for Biochemistry and Molecular Biology.


247

2.4. Biotinylation and fixation of HL-60 cells Cell-surface proteins of HL-60 cells were biotinylated using EZ-Link Sulfo-NHS-LC-Biotin (Pierce) according to manufacturer’s instructions. After biotinylation, the cells were fixed for 30 min with 2% paraformaldehyde in PBS at room temperature, and fixed cells were washed three times with PBS and counted. A portion of biotinylated and fixed HL-60 cells were treated with glycosidases as described in Leppa¨nen et al. (2005), before immobilization on streptavidin-coated microtiter plates.

3. Fluorescence Labeling of GBPs and Cells GBPs can be labeled using a variety of fluorescent probes that react with different functional groups of a protein. Fluorescent probes can be covalently attached in a variety of ways, including via primary amines, reduced cysteine residues, or oxidized carbohydrate residues. The fluorescent probe should be selected carefully, because covalent derivatization of amino acid residues involved in ligand binding may result in inactivation of the GBP. Therefore, it is important to test the activity of the labeled GBP before using it in an assay. As an example, labeling of galectin-1 through cysteine residues preserves the activity of the protein better than labeling through primary amines. Selectin binding to the ligand utilizes certain lysine residues at the carbohydrate-binding domain and therefore, selectins should not be derivatized through primary amines. If a monoclonal antibody is available to a GBP under study, an antibody can be fluorescently labeled and used to detect the bound GBP in the assay. Alternatively, if a recombinant GBP has been produced as an IgG or IgM fusion protein, commercial fluorescently labeled anti-IgG (or IgM) monoclonal antibodies can be used for detection. Glycoproteins can also be labeled through oxidized carbohydrate residues with hydrazide compounds, if carbohydrates are not required for biological activity. Live cells can be labeled with fluorescent compounds that are taken up by cells. For example, nonfluorescent Calcein-AM is membrane permeable and after taken up by live cells it is hydrolyzed to fluorescent compound (calcein) by intracellular esterases. Membrane permeable compounds do not react with cell surface macromolecules and thus do not change adhesive properties of cells. Examples of protocols on fluorescence labeling of GBPs and cells will be presented below.

3.1. Labeling of Gal-1 through primary amines Human recombinant dimeric galectin-1 (Gal-1) was labeled through primary amines by incubating 1–2 mg of Gal-1 with Alexa Fluor 488 carboxylic acid succinimidyl ester (Molecular Probes, Inc.) in PBS containing

248


0.1 M lactose for 1 h at room temperature, and incubation was continued overnight at þ4 C under stirring. Free dye and lactose were removed from the labeled Gal-1 using a PD-10 column (Amersham Biosciences) in PBS containing 14 mM b-mercaptoethanol (2-mercaptoethanol). Labeled Gal-1 was chromatographed on a small lactosyl-agarose column (1–2 ml volume) in PBS to separate functionally inactive and active proteins. Bound (active) Gal-1 was eluted with 0.1 M lactose in PBS. Before each experiment, lactose was removed using a PD-10 column in PBS containing 14 mM b-mercaptoethanol.

3.2. Labeling of Gal-1 through cysteines Human recombinant Gal-1 was labeled through cysteines by incubating 1– 1.5 mg of Gal-1 with 10-fold molar excess of thiol reactive Alexa Fluor 488 C5-maleimide (Molecular Probes, Inc.) in PBS containing 0.1 M lactose overnight at þ4 C under stirring. Free dye and lactose were removed from the labeled Gal-1 using a PD-10 column in PBS containing 14 mM b-mercaptoethanol. Labeled Gal-1 was chromatographed on a small lactosyl-agarose column in PBS and bound Gal-1 was eluted with 0.1 M lactose in PBS. Before each experiment, lactose was removed using a PD-10 column in PBS containing 14 mM b-mercaptoethanol. Gal-1 labeled with Alexa Fluor 488 C5-maleimide was more stable during long-term storage than Gal-1 labeled with Alexa Fluor 488 carboxylic acid succinimidyl ester.

3.3. Labeling of tomato (LEA) lectin through carbohydrates Lycopersicon esculentum (tomato) agglutinin (LEA) (Vector Laboratories) (4 mg/ml in PBS) was first treated with 100 mM sodium m-periodate for 30 min at room temperature in the dark to oxidize cis-diols of carbohydrates to aldehydes. Sodium m-periodate was removed using a PD-10 gel filtration column in PBS. Oxidized LEA was incubated with Alexa Fluor 488 hydrazide (100 mg/mg lectin) (Molecular Probes, Inc.) for 1.5–2 h at room temperature under stirring. Free dye was removed using a PD-10 column in PBS. The degree of labeling of LEA was significantly higher using labeling through carbohydrates than in commercial fluorescently labeled LEA.

3.4. Labeling of human T lymphocytes Purified human T lymphocytes (5–10 106 cells/ml) were labeled with 1 mM Calcein-AM (acetoxymethyl) (Molecular Probes, Inc.) in PBS at 37 C for 30 min. Labeled cells were washed three times with PBS and suspended into Hank’s balanced salt solution (HBSS, with 1.3 mM Ca2þ and 0.9 mM Mg2þ) containing 1% BSA, and counted before being used in the assay.


249

4. P- and L-Selectin Binding to Immobilized Glycosulfopeptides and Determination of Apparent Binding Affinity Selectins are a family of C-type lectins that are involved in leukocyte trafficking to inflamed tissues and lymphoid organs (McEver, 2002). P- and E-selectin are expressed on activated endothelial cells, P-selectin is also expressed on activated platelets, and L-selectin is expressed on leukocytes. All selectins bind to PSGL-1 expressed on leukocytes and in some cases on endothelial cells (Carlow et al., 2009; McEver and Cummings, 1997). Pand L-selectin bind to the extreme N-terminus of PSGL-1 specifically recognizing at least one of three tyrosine sulfate residues (Y46, Y48, Y51) and a nearby core 2-based O-glycan containing a sialyl Lewis x epitope (C2SLex) at T57 (Leppa¨nen et al., 1999, 2000, 2003; Somers et al., 2000). We have studied the site-specific role of each tyrosine sulfate residue (Tyr-SO3) and the role of O-glycan for binding to P- and L-selectin utilizing synthetic glycosulfopeptides (GSPs) modeled after the N-terminal sequence of mature PSGL-1 (Leppa¨nen et al., 2003). We have used different methods to compare binding affinity of P- and L-selectin to different GSPs. The methods include affinity chromatography, equilibrium gel filtration, and fluorescence-based solid-phase assay. Quantitative equilibrium binding data obtained from equilibrium gel filtration matched the binding data obtained from affinity chromatography and fluorescence-based solid-phase assay. Therefore, affinity chromatography and fluorescence-based solid-phase assay can be used as semiquantitative methods to determine relative binding affinity of a GBP to various immobilized ligands. Here, we present examples of fluorescence-based solid-phase assay to study the binding specificity of P- and L-selectin to immobilized GSPs.

4.1. An assay with recombinant P- and L-selectin Synthetic GSPs modeled after N-terminus of human PSGL-1 were biotinylated through C-terminal cysteine as described above (see Fig. 12.2 for structures). Reacti-BindTM streptavidin high binding capacity coated black 96-well plates (Pierce) were washed three times with 200 ml of 20 mM MOPS, pH 7.5, 150 mM NaCl, 2 mM CaCl2, 2 mM MgCl2, 0.02% NaN3 (buffer A) or 20 mM MOPS, pH 7.5, 150 mM NaCl, 5 mM EDTA, 0.02% NaN3 (buffer B). Equimolar amount of each biotinylated GSP (1 pmol/well with P-sel-Ig and 10 pmol/well with L-sel-Ig) was captured on the plate for 1.5 h in 50 ml of buffer A or B. After washing, the wells were incubated for 1 h with 50 ml of recombinant P-selectin IgG chimera (P-sel-Ig, 1 mg/ml) or L-selectin IgG chimera (L-sel-Ig, 10 mg/ml) in buffer A or B containing

250


1% BSA and 0.05% Tween-20. The wells were washed and subsequently incubated for 1 h with 50 ml of Alexa FluorTM 488 goat antihuman IgG (HþL) (Molecular Probes) (10 mg/ml) in buffer A or B containing 1% BSA and 0.05% Tween-20. All incubations were performed at room temperature, and the wells were washed three times with 300 ml of buffer A or B containing 0.05% Tween-20 between incubations. After final washing, 100 ml of buffer A or B was added to each well and the fluorescence was measured using a microtiter plate reader with excitation wavelength at 485 nm and emission wavelength at 535 nm. Background fluorescence reading without immobilized peptide was subtracted from each sample. The results show that both P- and L-sel-Ig bound with high affinity to GSP-6 containing three tyrosine sulfate residues and C2-SLex O-glycan at Thr57 (Fig. 12.3A and B). P- and L-sel-Ig bound very weakly to nonsulfated GP-6, but showed better binding to monosulfated and disulfated GSPs, with stronger binding to disulfated GSPs than to monosulfated GSPs. P-sel-Ig preferred to bind to GSPs containing tyrosine sulfate at position 48, but L-sel-Ig did not show clear preferential binding to any isomer of the mono- or disulfated GSPs, except for GSP(48,51)-6. Binding of P- and L-sel-Ig to all GSPs was strictly Ca2þ-dependent and quantitatively inhibited by including EDTA. The binding experiments with Pselectin and GSPs have also been carried out using equilibrium gel filtration and results from fluorescence-based solid-phase assay are in good agreement with equilibrium gel filtration data (Leppa¨nen et al., 2000).

4.2. Determination of apparent binding affinity for L-selectin Experiments shown in Fig. 12.3 were carried out using a single concentration of GBP. To identify high-affinity and low-affinity ligands, different concentrations of GBPs should be used in the binding assays. For example, a wide range of GBP concentrations is used to determine an apparent dissociation constant (Kd) for GBP binding to an immobilized ligand. We measured an apparent Kd for L-selectin and relative binding to immobilized GSP-6 and GP-6. Reduced salt concentration (50 mM NaCl) was used because it increases the binding affinity of selectins to ligands and therefore increases the sensitivity of binding (Koenig et al., 1997; Leppa¨nen et al., 2000, 2003). Various concentrations of L-sel-Ig were incubated with the immobilized peptides (10 pmol/well) in low salt buffer (50 ml/well in 20 mM MOPS, pH 7.5, 50 mM NaCl, 2 mM CaCl2, 2 mM MgCl2, 0.02% NaN3 containing 1% BSA and 0.05% Tween-20). The wells were subsequently incubated for 1 h with 50 ml of Alexa FluorTM 488 goat antihuman IgG (HþL) (Molecular Probes) in buffer (40 mg/ml). All incubations were performed at room temperature, and the wells were washed with buffer (without BSA) between incubations. Binding isotherms were obtained and apparent dissociation constants were derived from the binding

251


A

25,000

P-sel-Ig binding Ca2+

20,000

5 mM EDTA

RFU

15,000

10,000

5000

P4 G

P6

51 8, (4

G

)-6

)-6 51 SP

SP

G

G

G

SP

(4

6, (4

SP G

6,

48

)-6

1) -6 (5

8) -6 (4 SP

G

G

SP

G

(4

SP

6) -6

-6

0

B

L-sel-Ig binding 15,000

Ca2+

RFU

5 mM EDTA

10,000

5000

P4 G

P6 G

)-6

G

SP (4

8,

51

)-6 51 6,

SP (4 G

SP (4

6,

48

)-6

-6 G

G SP (5

1)

-6

G SP (4

8)

-6 6)

G SP (4

G

SP -6

0

Figure 12.3 Binding of P-sel-Ig and L-sel-Ig to immobilized glyco(sulfo)peptides (GSPs). Biotinylated GSPs were immobilized on streptavidin-coated microtiter well (A, 1 pmol/well; B, 10 pmol/well). P-sel-Ig (A, 1 mg/ml) or L-sel-Ig (B, 10 mg/ml) was incubated with the immobilized GSPs in Ca2þ-containing buffer A (light gray bars) or in EDTA-containing buffer B (dark gray bars). Fluorescently labeled antihuman IgG was used for detection. All assays were performed in triplicate and the results represent the mean SD of three determinations. This research was originally published in the Journal of Biological Chemistry (Lepp€anen et al., 2003). # The American Society for Biochemistry and Molecular Biology.

252


curves using nonlinear curve fitting with one site saturation equation. The apparent Kds were 20.4 nM for GSP-6 and 233 nM for GP-6, which indicates that L-sel-Ig binds to fully sulfated GSP-6 with a 12-fold higher affinity than to nonsulfated GP-6 (Fig. 12.4A and B). The data is consistent with the results obtained by fluorescence-based solid-phase assay using a single concentration of L-sel-Ig under physiologic and low salt conditions (not shown).

4.3. An assay with T lymphocytes To compare the binding specificity of recombinant L-sel-Ig and natural L-selectin present at the surface of resting human T lymphocytes, we used fluorescently labeled human T lymphocytes in solid-phase assay. The specificity of binding was controlled using EDTA and a function blocking mAb to L-selectin. Biotinylated GSP-6, GP-6, and GP-1 were immobilized onto streptavidin-coated microtiter plates at density of 5 pmol/well. Labeled cells (100,000 cells/well) were incubated with immobilized GSPs in physiologic Hank’s balanced salt solution (HBSS, with 1.3 mM Ca2þ and 0.9 mM Mg2þ) containing 1% BSA (50 ml/well) at room temperature for 30 min. Microtiter wells were washed four times with HBSS containing 1% BSA and bound fluorescence was measured using a microtiter plate reader. Parallel control experiments with EDTA and a monoclonal anti-L-selectin antibody DREG-56 (Pharmingen) were performed by preincubating cells for 15 min with 5 mM EDTA or DREG-56 (20 mg/ml), respectively, before incubating the cells with the immobilized ligands. Control wells with EDTA were washed four times with HBSS containing 1% BSA and 5 mM EDTA, and control wells with DREG-56 were washed four times with HBSS containing 1% BSA. T lymphocytes showed high-affinity binding to fully sulfated GSP-6, weak binding to nonsulfated GP-6, and no detectable binding to nonsulfated GP-1 containing only an a-linked GalNAc unit at Thr57 (Fig. 12.5). T lymphocyte binding to GSP-6 and GP-6 was completely inhibited by EDTA and DREG-56 showing that interaction was strictly dependent on Ca2þ and L-selectin. These results are in good agreement with the results obtained using recombinant L-sel-Ig (Figs. 12.3B and 12.4), indicating that natural L-selectin on T lymphocytes and recombinant L-sel-Ig have the same binding specificity to immobilized GSPs. Other types of leukocytes, such as acute promyelocytic HL-60 cells and acute monocytic leukemia THP-1 cells, did not show any detectable binding to immobilized GSP-6 (not shown). The results were consistent with cell-surface expression of L-selectin as verified using an anti-L-selectin monoclonal antibody in flow cytometry. Binding of fluorescently labeled cells to immobilized glycans/glycoconjugates can provide new information on the specificity of cell-surface GBPs

253


A 140,000 GSP-6 120,000

RFU

100,000 80,000 60,000 40,000 App. Kd = 20.4 +/–2.4 nM 20,000 0 0 B

20

40

60

80 100 L-Ig (nM)

120

140

160

60,000 GP-6 50,000

RFU

40,000 30,000 20,000 App. Kd = 233 +/–30 nM

10,000 0 0

500

1000

1500 L-Ig (nM)

2000

2500

3000

Figure 12.4 Equilibrium binding affinity of L-sel-Ig for immobilized GSP-6 and GP-6 at low salt buffer. Biotinylated GSPs were immobilized on streptavidin-coated microtiter wells (10 pmol/well). Various concentrations of L-sel-Ig were incubated with the immobilized (A) GSP-6 and (B) GP-6 in low salt buffer (20 mM MOPS, pH 7.5, 50 mM NaCl, 2 mM CaCl2, 2 mM MgCl2, 1% BSA, 0.05% Tween 20, 0.02% NaN3). Fluorescently labeled antihuman IgG (40 mg/ml) was used to detect the bound L-sel-Ig. Assays were performed in triplicate and the results represent the mean standard error of the mean. Experiments shown are representative of three independent experiments. Modified, with permission, from Glycobiology Online (http://glycob.oxfordjournals.org/) (Lepp€anen et al., 2010). Copyright Oxford University Press.

254


80,000 T lymphocyte binding No inhibitor DREG-56 5 mM EDTA

RFU

60,000

40,000

20,000

0

GSP-6

GP-6

GP-1

Figure 12.5 Binding of human T lymphocytes to immobilized glyco(sulfo)peptides at physiologic buffer. Biotinylated GSPs were immobilized on streptavidin-coated microtiter wells (10 pmol/well). Purified, fluorescently labeled human T lymphocytes ( 100,000 cells/well) were incubated with the immobilized ligands in Hank’s balanced salt solution (with Ca2þ and Mg2þ) containing 1% BSA (light gray bars). In control experiments a function blocking mAb to L-selectin, DREG-56 (20 mg/ml) (medium gray bars), or 5 mM EDTA (dark gray bars) were preincubated with the cells in HBSS before adding to the wells. All assays were performed in triplicate, and the results represent the mean SD of three determinations. Modified, with permission, from Glycobiology Online (http://glycob.oxfordjournals.org/) (Lepp€anen et al., 2010). Copyright Oxford University Press.

or reveal the presence of novel GBPs. GBPs in their natural environment at the cell surface likely provides more ‘‘biologically’’ relevant information on their binding specificity than recombinant forms of the GBP, because GBP density and presentation is optimal at the cell surface.

5. Galectin-1 Binding to Immobilized Glycopeptides and Glycans and Determination of Apparent Binding Affinity Galectins are a family of soluble b-galactoside-binding GBPs and many members require reducing free thiols for activity (Leffler et al., 2004). Galectin-1 (Gal-1) is widely expressed in animals and has immunoregulatory functions (Rabinovich and Ilarregui, 2009). Gal-1 recognizes a2-3-sialylated and nonsialylated extended poly-N-acetyllactosamine


255

(poly-LN) containing glycoconjugates (Leppa¨nen et al., 2005; Stowell et al., 2004, 2008). Our recent results indicated that Gal-1 recognizes terminal LN units on poly-LN structures and on other extended carbohydrate backbones, and binding depends on their mode of presentation (Leppa¨nen et al., 2005). In solid-phase assays, where glycans were immobilized on the surface of microtiter wells or at the cell surface, Gal-1 preferred to bind to extended poly-LN-type structures. However, in solution, Gal-1 bound to nonextended single LN unit similarly than to extended poly-LN structures (Ahmad et al., 2004; Hirabayashi et al., 2002; Leppa¨nen et al., 2005; Stowell et al., 2008). Here, we present some of the data on Gal-1 binding to immobilized glycopeptides, oligosaccharides, and cells, as another example of fluorescence-based solid-phase assay to study GBP–glycoconjugate interactions. The fluorescently labeled plant lectins Ricinus communis agglutinin I (RCA-I) and tomato lectin (LEA) were used as controls.

5.1. An assay with recombinant galectin-1 The contribution of the length of the poly-LN chain for binding to recombinant human dimeric Gal-1 was studied using immobilized synthetic glycopeptides and oligosaccharides. A series of glycopeptides (GP-4, GP-40 , GP-400 , GP-4000 ) modeled after N-terminus of PSGL-1 were synthesized to contain an extended core-2-based O-glycan at Thr57 (Fig. 12.2). Biotinylated glycopeptides and oligosaccharides were immobilized on streptavidincoated microtiter wells (50 pmol/well) in 50 ml of PBS for 1.5 h at room temperature. The wells were washed three times with 200 ml of PBS containing 0.05% Tween-20 and successively incubated for 1 h at room temperature with fluorescently labeled Gal-1 (40 mg/ml), RCA-I (120 mg/ ml, Vector laboratories), and LEA (50 mg/ml) in PBS containing 0.05% Tween-20 and 1% BSA. Bound Gal-1 and RCA-I were removed using 0.2 M lactose in PBS and 0.05% Tween-20 before incubating with the next lectin. After incubating with the lectins, wells were washed four times with 300 ml of PBS containing 0.05% Tween-20 and 100 ml of PBS was added to each well and fluorescence was measured using a microtiter plate reader with excitation and emission wavelengths at 485 and 535 nm, respectively. The background fluorescence reading after lactose washing was measured between incubations and subtracted from each sample. The results show that Gal-1 shows preferential binding to extended poly-LN structures on O-glycopeptides (GP-40 , GP-400 , and GP-4000 ) and on peptide-free glycans (LN2, LN3, and LNnT) (Fig. 12.6A). The degree of binding increased as the number of LN repeats increased. Gal-1 did not show detectable binding to degalactosylated NGLN2, NGLN3, and NGLNnT indicating that terminal galactose is highly important for binding. Bound Gal-1 was removed by washing with lactose and the wells were incubated with fluorescently labeled RCA-I that recognizes terminal

256


A

Gal-1 4000

RFU

3000

2000

1000

LN LN 2 LN LN 3 N nT G N LN G 2 N LN G LN 3 nT

G PG 4 P4 G PG 4 P4

0

B 150,000

RCA-I

RFU

100,000

50,000

LN LN 2 LN LN 3 N nT G N LN G 2 N LN G LN 3 nT

P4 G PG 4 P4

G

G

P-

4

0

C

LEA 80,000

RFU

60,000

40,000

20,000

LN 2 LN 3 LN N nT G L N N2 G N LN G 3 LN nT

G PG 4 P4 G P4 G P4 LN

0

Figure 12.6 Binding of human Gal-1, RCA-I, and LEA to immobilized glycopeptides and glycans. Biotinylated glycopeptides and glycans were immobilized on streptavidincoated microtiter wells (50 pmol/well). Fluorescently labeled (A) Gal-1 (40 mg/ml), (B) RCA-I (120 mg/ml), and (C) LEA (50 mg/ml) were successively incubated with the immobilized ligands. The data are representative of two independent experiments. All assays were performed in triplicate and the results are the mean SD of three determinations. This research was originally published in the Journal of Biological Chemistry (Lepp€anen et al., 2005). # The American Society for Biochemistry and Molecular Biology.


257

galactose residues (Baenziger and Fiete, 1979). RCA-I showed comparable binding to all ligands containing terminal Gal residues, indicating that the ligands were equivalently immobilized on the plate and were equally accessible for lectin binding (Fig. 12.6B). After removal of bound RCA-I, the wells were incubated with fluorescently labeled tomato lectin (LEA) that binds to long PL chains (Merkle and Cummings, 1987). LEA showed increased binding to ligands with increasing amount of LN repeats, including degalactosylated ligands (Fig. 12.6C). Control experiments with RCA-I and LEA verified that all ligands were equivalently immobilized and accessible for lectin binding. Our experiments also show that the same immobilized ligands on microtiter wells can be reused several times, if bound GBPs can be removed quantitatively by washing with a hapten sugar. This is helpful in conducting parallel experiments with different concentrations of the same GBP or probing with other GBPs.

5.2. Determination of apparent binding affinity The contribution of a2,3-sialic acid on S3LN3 for Gal-1 binding was examined by determination of apparent Kd for Gal-1 binding to immobilized S3LN3 and LN3. Various concentrations of fluorescently labeled Gal1 were incubated with immobilized LN3 and S3LN3 as described above. The apparent Kd derived from the binding curves were 3.5 mM for LN3 and 4.3 mM for S3LN3 indicating that Gal-1 binds to a2,3-sialylated and nonsialylated poly-LN structures with similar affinity (Fig. 12.7A and B). The result obtained with a single concentration of Gal-1 (40 mg/ml) was in good agreement with the apparent Kd values (Fig. 12.7A, the inset).

6. Galectin-1 Binding to Immobilized HL-60 Cells and Determination of Apparent Binding Affinity 6.1. An assay with recombinant galectin-1 The possibility that Gal-1 recognizes cell-surface poly-LN structures with higher affinity than immobilized monomeric poly-LN structures due to either differential presentation of poly-LN structures at the cell surface and/ or presence of higher affinity ligands at the cell surface was explored using immobilized cells in solid-phase assay. Human promyelocytic HL-60 cells expressing sialylated and nonsialylated cell-surface poly-LN glycans were biotinylated and fixed, and a portion of cells were treated with Arthrobacter ureafaciens neuraminidase. Untreated and neuraminidase-treated HL-60 cells were then digested with Escherichia freundii endo-b-galactosidase to cleave poly-LN structures, or alternatively jack bean b-galactosidase to digest

258


A 20,000

LN3 8000

15,000

RFU

10,000

4000 2000

Kd = 3.5 ± 0.8 mM 0

0

4

8

12

0 LN 3 LN 3

5000

S3

RFU

6000

16

20

16

20

B 10,000 S3LN3

RFU

8000 6000 4000 2000 Kd = 4.3 ± 0.7 mM 0 0

4

8 12 Gal-1 (mM)

Figure 12.7 Binding affinity of Gal-1 for immobilized LN3 and S3LN3. Biotinylated LN3 (A) and S3LN3 (B) were immobilized on streptavidin-coated microtiter wells (50 pmol/well). Various concentrations of fluorescently labeled Gal-1 were incubated with immobilized ligands. Assays were performed in duplicate and the results are the average of two determinations. This research was originally published in the Journal of Biological Chemistry (Lepp€anen et al., 2005). # The American Society for Biochemistry and Molecular Biology.

terminal b-galactosyl residues. Parallel binding experiments were performed with tomato lectin (LEA) and Griffonia simplicifolia lectin II (GS-II) to control the effect of endo-b-galactosidase and b-galactosidase on HL-60 cells. Glycosidase digested and nondigested HL-60 cells were immobilized on streptavidin-coated microtiter wells at equivalent densities (100,000 cells/well) in 50 ml of PBS for 1.5 h at room temperature. The wells were washed three times with 200 ml of PBS containing 1% BSA and incubated with 50 ml of fluorescently labeled Gal-1 (40 mg/ml), LEA (100 mg/ml), or GS-II (100 mg/ml, EY Laboratories, Inc.) in PBS


259

containing 1% BSA for 1 h at room temperature. The wells were washed four times with PBS containing 1% BSA and 100 ml of PBS was added to each well and the fluorescence was measured. Gal-1 bound to nontreated and neuraminidase-treated HL-60 cells but binding to nontreated cells was twofold less suggesting that a2-6-sialic acid residues on HL-60 cells may inhibit Gal-1 binding (Fig. 12.8A). Endo-bgalactosidase and b-galactosidase treatments of desialylated HL-60 cells reduced Gal-1 binding by 50% indicating that poly-LN structures and terminal galactose residues are important for Gal-1 binding. Endo-b-galactosidase and b-galactosidase treatments of parent sialylated HL-60 cells also reduced Gal-1 binding, but the effect of b-galactosidase treatment was less significant suggesting that most of the glycan structures on HL-60 cells are sialylated and unable to cleave with b-galactosidase. Bound Gal-1 was removed by incubating cells with 0.2 M lactose indicating the specificity of Gal-1 binding to HL-60 cells (not shown). Parallel binding experiment with LEA showed that endo-b-galactosidase treatment of parent sialylated and desialylated HL-60 reduced LEA binding by 50% indicating that 50% of poly-LN structures were removed by endo-b-galactosidase (Fig. 12.8B). By contrast, b-galactosidase treatment did not have a significant effect on LEA binding indicating that terminal b-galactosyl residues are not important for LEA binding, consistent with results obtained using immobilized glycans. GS-II recognizes terminal b-GlcNAc residues (Lyer et al., 1976) that are exposed when glycans are cleaved with b-galactosidase and endo-b-galactosidase. Endo-b-galactosidase treatment of parent sialylated and desialylated HL-60 cells significantly increased GS-II binding indicating that enzyme treatment was successful (Fig. 12.8C). b-Galactosidase treatment had a significant increase on GS-II binding only with desialylated HL-60 cells, but not with parent HL-60 cells indicating that most of the terminal b-galactosyl residues are penultimate to sialic acid in HL-60 cells and that sialidase treatment was successful.

6.2. Determination of apparent binding affinity The binding affinity of Gal-1 for neuraminidase-treated and nontreated immobilized HL-60 cells was determined by incubating various concentrations of fluorescently labeled Gal-1 with immobilized cells in the presence or absence of 20 mM lactose. The apparent Kds derived from the binding curves were 2.6 mM for desialylated HL-60 cells and 5.9 mM for nontreated HL-60 cells (Fig. 12.9A and B). Lactose concentration of 20 mM inhibited >70% of the binding in both cases. The data is consistent with results obtained by using a single concentration of Gal-1 (Fig. 12.8A). In conclusion, Gal-1 bound cell-surface poly-LN structures and immobilized synthetic poly-LN glycans (S3LN3 and LN3) with comparable affinity

260


A

40,000

Gal-1 No treatment

30,000

Endo-b-galactosidase

RFU

b-galactosidase 20,000

10,000

B

0 LEA

RFU

200,000

150,000

100,000

50,000

C

0 GS-II 100,000

RFU

80,000 60,000 40,000 20,000 0 Desialylated HL-60 cells

HL-60 cells

Figure 12.8 Binding of Gal-1, LEA, and GS-II to immobilized desialylated and nontreated HL-60 cells. A Portion of biotinylated and fixed HL-60 cells first were desialylated. Nontreated and desialylated HL-60 cells were treated with endo-b-galactosidase or b-galactosidase, and immobilized on streptavidin-coated microtiter wells (100,000 cells/well). Fluorescently labeled (A) Gal-1 (40 mg/ml), (B) LEA (100 mg/ ml), and (C) GS-II (100 mg/ml) were incubated with the immobilized cells. All assays were performed in triplicate and the results are the mean SD of three determinations. This research was originally published in the Journal of Biological Chemistry (Lepp€anen et al., 2005). # The American Society for Biochemistry and Molecular Biology.

261


A

40,000 Desialylated HL-60 cells

30,000

RFU

– Lactose + Lactose

20,000

Kd = 2.6 ± 0.3 mM

10,000

0 0

B

15,000

4

8 12 Gal-1 (mM)

16

20

HL-60 cells

– Lactose + Lactose

RFU

10,000

Kd = 5.9 ± 0.7 mM

5000

0 0

4

8 12 Gal-1 (mM)

16

20

Figure 12.9 Binding affinity of Gal-1 for immobilized desialylated and nontreated HL-60 cells. Biotinylated, fixed, desialylated HL-60 cells (A) and biotinylated, fixed, nontreated HL-60 cells (B) were immobilized on streptavidin-coated microtiter wells (100,000 cells/well). Various concentrations of Gal-1 were incubated with the immobilized cells in buffer with or without 20 mM lactose. All assays were performed in duplicate and the results are the average of two determinations. This research was originally published in the Journal of Biological Chemistry (Lepp€anen et al., 2005). # The American Society for Biochemistry and Molecular Biology.

262


indicating that the presentation of poly-LN structures at the cell surface does not improve the binding affinity. Our results indicate that fluorescence-based solid-phase assay is a versatile method to study binding of purified or recombinant GBPs to cellsurface glycoconjugates. Not only mammalian cells, but also microorganisms can be biotinylated and fixed, and adhesion of fluorescently labeled GBPs and antibodies can be studied using standard streptavidin-coated microplates. Additional binding experiments with plant lectins can provide valuable structural information of cell-surface glycoconjugates, because the specificity of many plant lectins has been well defined. If combined with highly specific glycosidase digestions, the binding assays can provide more structural information on cell-surface glycoconjugates. In the present experiments, we treated biotinylated and fixed cells with glycosidases before capturing cells on streptavidin plates. We did not observe significant differences in the binding assays, if glycosidase treatments were performed on the plate after capturing biotinylated and fixed cells (not shown). GBP binding assays with immobilized cells may reveal new high-affinity ligands for a given GBP. It is likely that GBPs may recognize natural ligands present at the cell surface differently than individual isolated glycans. The approaches we have described here allow researchers to prepare glycans from target cells and to produce natural glycan microarrays that reflect glycan presentations on natural glycoconjugates.

ACKNOWLEDGMENTS This work was supported by grants from the Academy of Finland (no. 106908 and 118469 [to A. L.]), from Magnus Ehrnrooth Foundation (Helsinki, Finland [to A. L.]), and from the National Institutes of Health, USA (no. HL085607 [to R. D. C.]).

REFERENCES Ahmad, N., Gabius, H., Sabesan, S., Oscarson, S., and Brewer, C. F. (2004). Thermodynamic binding studies of bivalent oligosaccharides to galectin-1, galectin-3, and the carbohydrate recognition domain of galectin-3. Glycobiology 14, 817–825. Baenziger, J. U., and Fiete, D. (1979). Structural determinants of Ricinus communis agglutinin and toxin specificity for oligosaccharides. J. Biol. Chem. 254, 9795–9799. Blixt, O., Head, S., Mondala, T., Scanlan, C., Huflejt, M. E., Alvarez, R., Bryan, M. C., Fazio, F., Calarese, D., Stevens, J., Razi, N., Stevens, D. J., et al. (2004). Printed covalent glycan array for ligand profiling of diverse glycan binding proteins. Proc. Natl. Acad. Sci. USA 101, 17033–17038. Blixt, O., Han, S., Liao, L., Zeng, Y., Hoffmann, J., Futakawa, S., and Paulson, J. C. (2008). Sialoside analogue array for rapid identification of high affinity siglec ligands. J. Am. Chem. Soc. 130, 6680–6681. Carlow, D. A., Gossens, K., Naus, S., Veerman, K. M., Seo, W., and Ziltener, H. J. (2009). PSGL-1 function in immunity and steady state homeostasis. Immunol. Rev. 230, 75–96.


263

de Boer, A. R., Hokke, C. H., Deelder, A. M., and Wuhrer, M. (2007). General microarray technique for immobilization and screening of natural glycans. Anal. Chem. 79, 8107–8113. Hirabayashi, J., Hashidate, T., Arata, Y., Nishi, N., Nakamura, T., Hirashima, M., Urashima, T., Oka, T., Futai, M., Muller, W. E. G., Yagi, F., and Kasai, K. (2002). Oligosaccharide specificity of galectins: A search by frontal affinity chromatography. Biochim. Biophys. Acta 1572, 232–254. Koenig, A., Jain, R., Vig, R., Norgard-Sumnicht, K. E., Matta, K. L., and Varki, A. (1997). Selectin inhibition: Synthesis and evaluation of novel sialylated, sulfated and fucosylated oligosaccharides, including the major capping group of GlyCAM-1. Glycobiology 7, 79–93. Leffler, H., Carlsson, S., Hedlund, M., Qian, Y., and Poirier, F. (2004). Introduction to galectins. Glycoconj. J. 19, 433–440. Leppa¨nen, A., Mehta, P., Ouyang, Y. B., Ju, T., Helin, J., Moore, K. L., van Die, I., Canfield, W. M., McEver, R. P., and Cummings, R. D. (1999). A novel glycosulfopeptide binds to P-selectin and inhibits leukocyte adhesion to P-selectin. J. Biol. Chem. 274, 24838–24848. Leppa¨nen, A., White, S. P., Helin, J., McEver, R. P., and Cummings, R. D. (2000). Binding of glycosulfopeptides to P-selectin requires stereospecific contributions of individual tyrosine sulfate and sugar residues. J. Biol. Chem. 275, 39569–39578. Leppa¨nen, A., Penttila¨, L., Renkonen, O., McEver, R. P., and Cummings, R. D. (2002). Glycosulfopeptides with O-glycans containing sialylated and polyfucosylated polylactosamine bind with low affinity to P-selectin. J. Biol. Chem. 277, 39749–39759. Leppa¨nen, A., Yago, T., Otto, V. I., McEver, R. P., and Cummings, R. D. (2003). Model glycosulfopeptides from P-selectin glycoprotein ligand-1 require tyrosine sulfation and a core 2-branched O-glycan to bind to L-selectin. J. Biol. Chem. 278, 26391–26400. Leppa¨nen, A., Stowell, S., Blixt, O., and Cummings, R. D. (2005). Dimeric galectin-1 binds with high affinity to a2, 3-sialylated and non-sialylated terminal N-acetyllactosamine units on surface-bound extended glycans. J. Biol. Chem. 280, 5549–5562. Leppa¨nen, A., Parviainen, V., Ahola-Iivarinen, E., Kalkkinen, N., and Cummings, R. D. (2010). Human L-selectin preferentially binds synthetic glycosulfopeptides modeled after endoglycan and containing tyrosine sulfate residues and sialyl Lewis x in core 2 O-glycans. Glycobiology doi: 10.1093/glycob/cwq083. Liu, Y., Palma, A. S., and Feizi, T. (2009). Carbohydrate microarrays: Key developments in glycobiology. Biol. Chem. 390, 647–656. Lyer, P. N., Wilkinson, K. D., and Goldstein, I. J. (1976). An N-acetyl-D-glucosamine binding lectin from Bandeirea simplicifolia seeds. Arch. Biochem. Biophys. 177, 330–333. McEver, R. P. (2002). Selectins: Lectins that initiate cell adhesion under flow. Curr. Opin. Cell. Biol. 14, 581–586. McEver, R. P., and Cummings, R. D. (1997). Perspectives series: Cell adhesion in vascular biology. Role of PSGL-1 binding to selectins in leukocyte recruitment. J. Clin. Invest. 100, 485–491. Merkle, R. K., and Cummings, R. D. (1987). Relationship of the terminal sequences to the length of poly-N-acetyllactosamine chains in asparagine-linked oligosaccharides from the mouse lymphoma cell line BW5147: Immobilized tomato lectin interacts with high affinity with glycopeptides containing long poly-N-acetyllactosamine chains. J. Biol. Chem. 262, 8179–8189. Rabinovich, G. A., and Ilarregui, J. M. (2009). Conveying glycan information into T-cell homeostatic programs: A challenging role for galectin-1. Immunol. Rev. 230, 144–159. Schallus, T., Jaeckh, C., Feher, K., Palma, A. S., Liu, Y., Simpson, J. C., Meckeen, M., Stier, G., Gibson, T. J., Feizi, T., Pieler, T., and Muhle-Goll, C. (2008). Malectin: A

264


novel carbohydrate-binding protein of the endoplasmic reticulum and a candidate player in the early steps of protein N-glycosylation. Mol. Biol. Cell 19, 3404–3414. Somers, W. S., Tang, J., Shaw, G. D., and Camphausen, R. T. (2000). Insights into the molecular basis of leukocyte tethering and rolling revealed by structures of P- and E-selectin bound to SLex and PSGL-1. Cell 103, 467–479. Song, X., Xia, B., Lasanajak, Y., Smith, D. F., and Cummings, R. D. (2008). Quantifiable fluorescent glycan microarrays. Glycoconj. J. 25, 15–25. Song, X., Lasanajak, Y., Olson, L. J., Boonen, M., Dahms, N. M., Kornfeld, S., Cummings, R. D., and Smith, D. F. (2009a). Glycan microarray analysis of P-type lectins reveals distinct phosphomannose glycan recognition. J. Biol. Chem. 284, 35201–35214. Song, X., Lasanajak, Y., Rivera-Marrero, C., Luyai, A., Willard, M., Smith, D. F., and Cummings, R. D. (2009b). Generation of a natural glycan microarray using 9-fluorenylmethyl chloroformate (FmocCl) as a cleavable fluorescent tag. Anal. Biochem. 395, 151–160. Song, X., Lasanajak, Y., Xia, B., Smith, D. F., and Cummings, R. D. (2009c). Fluorescent glycosylamines produced by microscale derivatization of free glycans for natural glycan microarrays. ACS Chem. Biol. 4, 741–750. Song, X., Xia, B., Stowell, S. R., Lasanajak, Y., Smith, D. F., and Cummings, R. D. (2009d). Novel fluorescent glycan microarray strategy reveals ligands for galectins. Chem. Biol. 16, 36–47. Stowell, S. R., Dias-Baruffi, M., Penttila¨, L., Renkonen, O., Nyame, A. K., and Cummings, R. D. (2004). Human galectin-1 recognition of poly-N-acetyllactosamine and chimeric polysaccharides. Glycobiology 14, 157–167. Stowell, S. R., Arthur, C. M., Mehta, P., Slanina, K. A., Blixt, O., Leffler, H., Smith, D. F., and Cummings, R. D. (2008). Galectin-1, -2, and -3 exhibit differential recognition of sialylated glycans and blood group antigens. J. Biol. Chem. 283, 10109–10123. Stowell, S. R., Arthur, C. M., Dias-Baruffi, M., Rodrigues, L. C., Gourdine, J. P., Heimburg-Molinaro, J., Ju, T., Molinaro, R. J., Rivera-Marrero, C., Xia, B., Smith, D. F., and Cummings, R. D. (2010). Innate immune lectins kill bacteria expressing blood group antigen. Nat. Med. 16, 295–301. Xia, B., Kawar, Z. S., Ju, T., Alvarez, R. A., Sachdev, G. P., and Cummings, R. D. (2005). Versatile fluorescent derivatization of glycans for glycomic analysis. Nat. Methods 2, 845–850.

C H A P T E R

T H I R T E E N

Multifaceted Approaches Including Neoglycolipid Oligosaccharide Microarrays to Ligand Discovery for Malectin Angelina S. Palma,*,1 Yan Liu,* Claudia Muhle-Goll,†,2 Terry D. Butters,‡ Yibing Zhang,* Robert Childs,* Wengang Chai,* and Ten Feizi* Contents 266 268 268

1. Overview 2. Preparation of Recombinant Soluble Human Malectin 2.1. Materials and equipment 2.2. Generation of plasmids containing the His6-tagged malectin globular domain 2.3. Expression and purification of His6-tagged malectin for microarray analysis 3. Preparation of Glucan Oligosaccharides 3.1. Materials and equipment 3.2. Glucan oligosaccharides di- to heptasaccharides 4. Preparation of Glucosylated High-Mannose N-Glycans 4.1. Triglucosylated high-mannose N-glycans 4.2. Diglucosylated high-mannose N-glycans 5. Preparation of NGL Probes 5.1. Materials and equipment 5.2. Preparation of AO-NGLs of glucan oligosaccharides 5.3. Preparation of AO-NGLs of the glucosylated N-glycans

269 270 271 271 272 274 274 276 278 279 279 279

* Glycosciences Laboratory, Faculty of Medicine, Imperial College London, Northwick Park Hospital Campus, Harrow, Middlesex, United Kingdom { European Molecular Biology Laboratory, Heidelberg, Germany { Department of Biochemistry, Oxford Glycobiology Institute, University of Oxford, Oxford, United Kingdom 1 Present address: REQUIMTE, Departamento de Quı´mica, Centro de Quı´mica Fina e Biotecnologia, Faculdade de Cieˆncias e Tecnologia, Universidade Nova de Lisboa, Caparica, Portugal 2 Present address: Karlsruhe Institute of Technology (KIT), Institut fu¨r Biologische Grenzfla¨chen (IGB-2), Karlsruhe, Germany Methods in Enzymology, Volume 478 ISSN 0076-6879, DOI: 10.1016/S0076-6879(10)78013-7

#


265

266

Angelina S. Palma et al.

6. Carbohydrate Microarray Analysis of Human Malectin 6.1. Materials and equipment 6.2. Microarray printing 6.3. Probing the microarrays 7. Conclusions Acknowledgments References

281 281 282 283 284 284 284

Abstract In this chapter, we describe the key procedures for isolation of the oligosaccharides and the preparation of neoglycolipid probes together with expression of malectin that have enabled the discovery of the highly selective binding of this newly described protein in the endoplasmic reticulum (ER) to a diglucosyl high-mannose N-glycan. This is the first indication of a bioactivity for a diglucosyl high-mannose N-glycan of the type that occurs in the ER of eukaryotic cells and which is an intermediate in the early steps of the N-glycosylation pathway of nascent proteins. The malectin story is an example of a powerful convergence of disciplines in biological sciences: (i) developmental biology, (ii) bioinformatics, (iii) recombinant protein expression, (iv) protein structural studies, (v) glucan biochemistry, and (vi) drug-assisted engineering of oligosaccharide biosynthesis, culminating in (vii) oligosaccharide ‘‘designer’’ microarrays, to clinch the remarkable selectivity of the binding of this newly discovered ER protein. Thus, the way is open to the identification of the role of malectin in the N-glycosylation pathway.

1. Overview The malectin protein gene was originally identified in Xenopus laevis in the search for proteins that are developmentally regulated in the pancreas. However, it was soon found to be broadly expressed in embryonic and adult X. laevis, and moreover detected in all tissues examined (Schallus et al., 2008). Although possibly disappointing initially that this was not a pancreatic developmental marker, bioinformatic studies with the deduced amino acid sequence revealed malectin as a highly conserved protein in the animal kingdom (Fig. 13.1) pointing to an important biological function. A clue for possible ligands for malectin came from its three-dimensional structure resolved by NMR, which showed that the highest hits for fold homologues were microbial carbohydrate-binding modules (CBMs) that recognize glucan polysaccharides (Schallus et al., 2008). Armed with this knowledge, we performed NMR-based ligand-screening studies, using glucose containing oligosaccharides. Maltose was the first disaccharide that was observed to be bound by the xenopus protein, hence its designation ‘‘malectin.’’ The glucan binding property of the xenopus malectin was further

267

Carbohydrate Microarrays and Ligands for Malectin

SP 1

60

77

70

Expression construct start 27

20

10

90

110

100

50

40

120

130

140

Expression construct stop 150

170

160

180

190

210 213

200

TM 220

230

240

250

270

Figure 13.1 Sequence alignment of malectin protein in animals. The beginning and the end of the expression construct are indicated. The secondary structure elements of the globular domain are shown on top of the amino acid sequence, and the four aromatic residues (Y67, Y89, Y116, and F117) and D186 mediating the carbohydrate interaction are marked by crosses. SP, signal peptide; TM, C-terminal transmembrane helix; Xen, Xenopus laevis; Hum, Homo sapiens; Mou, Mus musculus; Hen, Gallus gallus; Fly, Drosophila melanogaster; Aed, Aedes aegyptii; Cae, Caenorhabditis elegans; Sch, Schistosoma japonicum; Nem, Nematostella vectensis. (The figure was reprinted with kind permission from MBC (Schallus et al., 2008).)

corroborated using microarrays of glucan oligosaccharide probes (Schallus et al., 2008). With the finding soon, thereafter, that malectin is localized in the endoplasmic reticulum (ER) of mammalian cells, we populated our neoglycolipid (NGL)-based oligosaccharide microarrays with the glucosylated high-mannose N-glycans of the type that occur in the ER, having one, two, or three terminal glucosyl residues at the D1 mannosyl branch. These experiments revealed a high selectivity of malectin for a high-mannose N-glycan with two terminal glucose residues (Schallus et al., 2008).

268


The NGL technology and the NGL-based microarray system that have come to play in this study have been the subject of chapters in previous volumes of this series (Chai et al., 2003; Feizi and Childs, 1994; Feizi et al., 1994; Liu et al., 2006). Major advantages of the NGL approach are: its inherent flexibility, generation of multivalent probes with increased sensitivity of detection, and the ability to rapidly prepare and populate the arrays with probes containing novel oligosaccharide sequences, especially those of oligosaccharides isolated from biological sources and available only in minute amounts. In this chapter, we dwell in some detail on the multifaceted technical procedures that have been essential to carbohydrate ligand discovery for malectin. These include the procedures for generation of recombinant His-tagged malectin, focusing on how to prepare the human malectin homologue; the preparation of glucan oligosaccharides and glucosylated N-glycans from polysaccharides and drug-treated mammalian cell cultures, respectively; the microscale conjugation of these oligosaccharides to lipid for preparation of NGL probes, followed by NGL-based microarray analysis of recombinant human malectin. Please also read Chapter 19 in Volume 480 of this series, ‘‘Use of glycan microarrays to explore specificity of glycan-binding proteins’’ by David Smith, Xuezheng Song, and Richard Cumming.

2. Preparation of Recombinant Soluble Human Malectin Malectin contains a single highly conserved globular domain (Fig. 13.1; Homo sapiens AAs 42–228; Schallus et al., 2008). Protein domain databases such as SMART (Letunic et al., 2006) and Pfam (Finn et al., 2006) predict an Nterminal signal peptide (H. sapiens: AAs 1–28) and a C-terminal transmembrane helix (H. sapiens: 271–290). Between the globular domain and the C-terminal membrane anchor, a highly charged sequence segment is found, that extends over eight contiguous glutamate residues, for example, in human malectin. The lectin function resides in the globular domain of malectin, which was used for microarray screening of putative ligands. It was cloned into the pET M10 vector of the M-series generated by Gunter Stier, EMBL (Bogomolovas et al., 2009). This vector contains an N-terminal hexahistidine (His6)-sequence that is used as a tag for affinity purification and enables detection of the malectin construct using an anti-His antibody. Further information on the vector can be found at http://www.embl.de/ pepcore/pepcore_services/strains_vectors/index.html.

2.1. Materials and equipment 1. Taq DNA polymerase supplied with appropriate buffer (e.g., AmpliTaq DNA Polymerase, Applied Biosystems)


269

2. Sense and reverse primers 3. PCR Nucleotide mix (dATP, dGTP, dCTP, and dTTP; Roche Applied Science) 4. Genomic DNA (e.g., Qiagen) 5. Deionized and sterilized H2O 6. Restriction enzymes (NcoI, Acc65I, supplied with respective buffers; Fermentas) 7. T4 DNA Ligase, supplied with respective buffer (Fermentas) 8. Kits for PCR fragment extraction or plasmid purification (Qiagen) 9. Expression vector (recommended pET system) 10. Escherichia coli BL21[DE3] DH5a (New England Biolabs) for cloning 11. E. coli BL21[DE3] cells (New England Biolabs) for expression 12. LB-medium (10 g/l bactotryptone, 5 g/l yeast extract, 10 g/l NaCl, pH 7.4, supplemented with appropriate antibiotic (e.g., 50 mM kanamycin) and 200 mM isopropylthio-b-galactoside (IPTG) for induction) 13. Lysis buffer (20 mM Tris–HCl, pH 8.0,150 mM NaCl,10 mM Imidazole, 2 mM b-mercaptoethanol, protease inhibitors, for example, Complete Protease Inhibitor Cocktail, Roche) 14. Lysozyme (Sigma) (300 mg/ml) 15. DNase I (Roche Applied Science, 1 mg/ml) þ MgCl2 16. High salt wash buffer (lysis buffer containing 1 M NaCl) 17. Elution buffer (lysis buffer þ 250 mM Imidazole, pH 8.0) 18. Ni NTA resin (Qiagen) 19. Millipore Amicon Ultra-4 centrifugal filter units, 10 kDa cutoff 20. Gel filtration buffer (20 mM Tris–HCl, pH 7.2, 150 mM NaCl, 2 mM 1,4 dithiothreitol (DTT), 1% (w/v) maltose (Sigma) 21. Gel filtration column, for example, Superdex 200 HiLoad 16/60 (GE Healthcare) 22. Dialysis buffer (20 mM Tris–HCl, pH 7.2, 150 mM NaCl, 2 mM DTT) 23. Desalting columns PD10 (Amersham Biosciences) 24. Sonicator (e.g., Sonopuls HD 2070, Bandelin (Berlin)) 25. Ultracentrifuge, for example, Beckmann OptimaTM L-90 26. Spectrophotometer, nanodrop ND-1000, PEQLAB

2.2. Generation of plasmids containing the His6-tagged malectin globular domain The coding sequence for the human malectin globular domain was amplified by PCR from a plasmid carrying the human genomic sequence using the following primers: sense primer MalHsen TTG CCA TGG CCGGG CTG CCCGAGAG and reverse primer MalHrev TTGCGGTACCTT ACTCCAATCCCGGATGAGGCTG.

270


The sense primer carries the restriction site for NcoI (CCAGGT). The reverse primer is designed with a stop codon UAA before the Acc65I restriction site (GGTACC). The xenopus malectin domain can, in principle, be amplified using the homologous primers, but an internal NcoI site within the first 40 amino acids of the globular domain should be mutated prior to PCR amplification to avoid internal cleavage. Five PCR cycles were performed at lower annealing temperature of 57 C, which were followed by 20 cycles at 62 C, and the elongation was done at 72 C for 1 min. When the PCR reaction is performed with genomic DNA as a template, 25 cycles at 62 C are recommended to generate enough of the PCR product. The PCR product was cloned using the 30 -Acc65I and 50 -NcoI restriction sites into the pET M10 vector referred above. Plasmids were amplified in E. coli BL21[DE3] DH5a and the kanamycin resistance conferred by the plasmid was used for clone selection. Plasmid purification was achieved using the standard protocols and kits of the Qiagen plamid purification kit. DNA sequencing using T7 sense and reverse primers was used to verify the sequence. The plasmids were transformed into E. coli BL21[DE3] cells for expression.

2.3. Expression and purification of His6-tagged malectin for microarray analysis Five milliliters of an overnight culture of E. coli BL21[DE3] cells with the plasmid containing the gene for the globular segment of human His6malectin gene were diluted in 1 l LB medium containing 50 mg/ml kanamycin. The cells were grown at 37 C to an OD600 of 0.6 (approximately 6–8 h). Expression was induced by adding 0.2 mM isopropyl-b-D-thiogalactopyranoside overnight at 18 C. The cells were harvested by centrifugation (3–5 g/l). At this point, the cell pellet could be stored at 20 C, a step that also facilitated cell rupture. For protein extraction, the cell pellet was resuspended in lysis buffer (10 ml/g wet cell pellet), chilled to 4 C, complemented with lysozyme, and incubated on ice until cell rupture became visible. To digest genomic DNA, DNase I and 5 mM MgCl2 were added and the cell suspension was incubated at ambient temperature for another 10 min. A sonication step (30–50 short pulses of 5–10 s with pauses 10–20 s) was employed to complete cell rupture. Following ultracentrifugation at 125,000g for 25 min at 4 C (e.g., 40,000 RPM using a Beckmann 45TI rotor), the supernatant was applied onto a Ni-NTA agarose column at ambient temperature (2 ml Ni-NTA resin /1 l cell culture were used). The column was washed with 10–20 column volumes of lysis buffer, followed by 10 column volumes of high salt wash buffer and with 10 volumes of lysis buffer. The His6-tagged proteins were eluted with 5–7 volumes of elution buffer.


271

Purity after this step was already high as judged by SDS-PAGE analysis. Nevertheless, a size exclusion purification step on a gel filtration column (Superdex 200) was added to remove spurious copurified proteins that bind nonspecifically to the Ni-NTA column, and might be present at concentrations below the detection limit of SDS-PAGE analysis. On one occasion, we noticed an unidentified glycosidase activity in a protein preparation purified only with Ni-NTA chromatography. This was detected by a change in the color of the protein solution to yellow upon addition of the nitrophenylmaltoside compound, indicating that the nitrophenyl group was cleaved off. The size exclusion matrix material is composed of polymers of dextran. We observed that malectin binds tightly to the column and could not be readily eluted off using standard elution buffers. The inclusion of 1% (w/v) maltose in the gel filtration buffer was sufficient to overcome this problem and the protein eluted as the predicted molecular mass of 18 kDa, which was confirmed by NMR 1H-T2 measurements and dynamic light scattering. Finally, a buffer exchange step was performed by applying the protein to a small size exclusion, desalting column (e.g., PD10) or by dialysis against maltose-free Tris buffer. Up to 20 mg of malectin could be produced from 1 l of bacterial culture following this protocol. Protein concentration was measured by UV measurements at 280 nm using an extinction coefficient of 20,400 M 1 cm 1. The protein was stable under these conditions at 4 C for several weeks at concentrations below 4 mg/ml (20 mM), in the presence of 0.02% sodium azide. The recombinant malectin is soluble up to 20 mg/ml, but on storage tends to partially precipitate at concentrations higher than 10 mg/ml. For long-term storage, the protein can be lyophilized in suitable aliquots with retention of carbohydrate-binding activity and specificity.

3. Preparation of Glucan Oligosaccharides A series of glucan oligosaccharides, in the range of di- to heptasaccharides, with a1,4-, a1,6-, b1,3-, b1,4-, b1,6-linkages and the disaccharides with a1,2- and a1,3-linkages, were either from commercial sources or prepared from polysaccharides after partial depolymerization by chemical means, as described below.

3.1. Materials and equipment 1. Disaccharides nigerose (a1,3-linked) from Wako Chemicals, cellobiose (b1,4-linked) and kojibiose (a1,2-linked) from Sigma 2. Laminarin (b1,3-linked) di-, tri-, and tetrasaccharides from Dextra, penta- and hexasaccharides from Megazyme, and heptasaccharide from Seikagaku America

272


3. Malto-di- to heptasaccharides (a1, 4 linked) from Sigma 4. Cello-oligosaccharide (b1,4-linked) mixture obtained by acid hydrolysis of cellulose from Megazyme 5. Glucan polysaccharides pustulan from Umbilicaria papullosa (b1,6linked) (de la et al., 1995) from Calbiochem and dextran (MW 500 kDa, a1,6-linked with 5% a1,3-branches) (de Belder, 1993) from Amersham Biosciences 6. Deionised water 7. HCl and NaOH solutions (0.1 M and 0.2 M) 8. Solvent n-propanol:water (8:3, v/v) 9. Gel filtration columns: Sephadex G10 (1.630 cm, Amersham Biosciences) and Bio-Gel P4 (1.690 cm, Bio-Rad), with an on-line refractive index detector and auto sample collector 10. Silica gel high-performance (HP) TLC plates with aluminium-backing (Merck) 11. Orcinol staining reagent (Chai et al., 2003) 12. MALDI-TOF mass spectrometer, Tof Spec-2E (Waters)

3.2. Glucan oligosaccharides di- to heptasaccharides Partial depolymerization of dextran and pustulan were carried out by acid hydrolysis. Dextran (100 mg) was treated with 0.1 M HCl, at a concentration of 20 mg polysaccharide/ml, at 100 C for 4 h. Pustulan (100 mg) was treated with 0.2 M HCl at a concentration of 10 mg polysaccharide/ml at 100 C for 8 h. The reaction was stopped by neutralization with aqueous NaOH solution. Acid hydrolysis was selected as the depolymerization procedure for pustulan upon our observation that the b-linked oligosaccharide series prepared by acetolysis of the parent polysaccharide contained a percentage of a-anomers. The dextran and pustulan hydrolysates were desalted using a short column (1.630 cm) of Sephadex G10 eluted with deionized water at a flow rate of 20 ml/h monitored on-line by a refractive index detector. The oligosaccharide fraction was collected at the void volume and lyophilized. Oligosaccharide fractions of dextran, pustulan, and cellulose were obtained from gel filtration chromatography on a column (1.690 cm) of Bio-Gel P4. The Bio-Gel P4 column was first equilibrated with deionized water and calibrated with an analytical mixture of dextran hydrolysate by eluting with deionized water at a flow rate of 15 ml/h. The desalted oligosaccharide mixtures (typically 1–2 ml of clear solution, the concentration applied varied according to solubility) were applied to the column and eluted under the same conditions. The eluate was monitored on-line by refractive index and the respective di- to heptasaccharide peaks were pooled according to their predominant glucose units (Fig. 13.2A) and lyophilized.

273


A

Refractive index

1 Origin 1 2 3 4 5 6 7 8

5

6 V0 13 12

3

4

5

6

11 10

9

8

7 8 Time (h)

3

4

2

7

9

10

11

12

B 1175 Glc7Na+

%

100

Glc6Na+ 1013

Glc8Na+ 1337

0 600 650 700 750 800 850 900 950 1000 1050 1100 1150 1200 1250 1300 1350 1400 1450

m/z

Figure 13.2 (A) Gel filtration chromatography of the dextran hydrolysate on a column of Bio-Gel P4; the column was loaded with 4 mg hexose of the hydrolysate mixture in 1 ml deionized water (the dextran polysaccharide, 0.1 mg hexose, was added to the mixture to mark exclusion volume, Vo; the glucose units are indicated for each peak); the inset shows the HPTLC analysis of the isolated dextran oligosaccharide fractions (mono- to heptasaccharide, lanes 1–7, 2 mg hexose per lane) and of the hydrolyzed dextran mixture (total 40 mg hexose) before fractionation (lane 8); (B) MALDI-MS analysis of dextran heptasaccharide fraction. The molecular masses of the sodiated molecules are indicated (major component is the heptasaccharide).

Quantitation of the oligosaccharide fractions, after their reconstitution with deionized water, was carried out by TLC-based orcinol assay using glucose (0.05–1 mg/ml in deionized water) as standard, as described (Chai et al., 1997). Stock solutions were then prepared typically at 5 mg/ml.

274


Overall, the solubility in water of fractions up to heptasaccharides is good with the exception of the cello-oligosaccharides with degree of polymerization five or higher, for which the stock solutions were prepared at 1 mg/ml. Aliquots of each fraction (2 mg hexose) were analyzed by HPTLC on silica gel plates, using a solvent system of n-propanol/water, 8:3 (by volume), and stained with orcinol reagent (inset Fig. 13.2A). The molecular masses of the main components of oligosaccharide fractions from gel filtration were corroborated by MALDI-MS (Fig. 13.2B). For MALDI-MS, oligosaccharides solutions were diluted in methanol, at an estimated concentration of 10–20 pmol/ml, and 0.5 ml was deposited on the sample target together with a matrix of 2-(4-hydroxyphenylazo)benzoic acid. The linkage of the oligosaccharide fractions was corroborated by methylation analysis or 1H NMR.

4. Preparation of Glucosylated High-Mannose N-Glycans The monoglucosylated high-mannose N-glycan, Glc1Man9GlcNAc2, was isolated from hen egg yolk IgY using a similar procedure to that previously reported by Ohta et al. (1991). The triglucosylated N-glycan, Glc3Man7(D1) GlcNAc2, was isolated from the recombinant-expressed glycoprotein HIVIIIB gp120, secreted by cells that were treated with an ER-a-glucosidase inhibitor (Petrescu et al., 1997). The diglucosylated N-glycan, Glc2Man7(D1) GlcNAc2, was derived from this Glc3 analogue by digestion with ER-aglucosidase I (Alonzi et al., 2008). The general methodologies for preparation of the tri- and diglucosylated N-glycans are described below and the reader is advised to refer to primary papers cited.

4.1. Triglucosylated high-mannose N-glycans Triglucosylated N-glycans can be obtained from two major sources using tissue cultured cells treated with an ER-a-glucosidase inhibitor: (i) release of the oligosaccharides from secreted glycoproteins; (ii) isolation of free glucosylated oligosaccharides from the cells. 4.1.1. Materials and equipment 1. Chinese hamster ovary (CHO) cell line (Petrescu et al., 1997) stably transfected with highly glycosylated glycoproteins, eg., HIV gp 120 2. a-Glucosidase inhibitor N-butyl-deoxynojirimycin (NB-DNJ) (2–5mM, Toronto Research Chemicals, Inc.) 3. Concanavalin A-Sepharose beads (Sigma) 4. a-Methyl-mannoside (Sigma) 5. Hydrazine reagent (Ludger)


275

6. Bacterial peptide N-glycosidase F (PNGase) from NEB or Ludger 7. High-performance anion-exchange chromatography (HPAEC, Dionex) 8. Normal-phase high-performance liquid chromatography NP-HPLC (Anachem) 9. Bio-Gel P4 columns (Bio-Rad) 10. In-line electrochemical detector (Dionex) 4.1.2. Isolation of glycoproteins Tissue cultured cells are grown in the presence of a suitable a-glucosidase inhibitor, such as castanospermine, deoxynojirimycin (DNJ), or NB-DNJ. These are available commercially and DNJ can be conveniently chemically synthesized from inexpensive starting materials such as D-glucuronolactone (Best et al., 2010). N-butylation of DNJ by reductive amination using sodium cyanoborohydride (Mellor et al., 2002) increases the a-glucosidase inhibitory efficacy in cells significantly (Alonzi et al., 2009). CHO cells are grown for 3–4 days in medium containing 2–5 mM sterile filtered NB-DNJ. The choice of CHO is critical for maximizing the synthesis of glucosylated products as it lacks an endomannosidase-mediated salvage pathway that deglucosylates proteins destined for secretion (Spiro, 2004). CHO cells stably transfected with highly glycosylated glycoproteins, for example, HIV gp 120, allow easy isolation of the highly expressed protein by antibody affinity column chromatography, for isolation of glycans. The abundance of glucosylated glycan is increased using lectin-resistant cells that are deficient in complex glycan biosynthesis (Butters et al., 1999). Culture medium that is rich in glucosylated glycoprotein is also used as a convenient source of oligosaccharide. Oligomannosidic glycoproteins can be bound to Concanavalin A-Sepharose beads and subsequently eluted with a-methyl-mannoside (0.5 mM in water). Following dialysis against water (or 0.1% TFA to reduce precipitation) to remove the methyl-mannoside, the glycoprotein-rich extract is lyophilized and subjected to glycan release (Petrescu et al., 1997). 4.1.3. Release and purification of triglucosylated N-glycans from glycoproteins Anhydrous hydrazine is used to cleave the glycan between the terminal nonreducing N-acetylglucosamine residue and the asparagine amino acid. This method is relatively quantitative and preserves the glycan intact. At least 1 mg of protein can be treated with commercially available kit forms of the hydrazine reagent (available from Ludger, UK see http://www.ludger.com/). An alternative method for release is to use bacterial PNGase. For this, the protein is denatured in SDS and disulphide reducing agent to allow enzyme access to all N-glycosylation sites, before dilution and addition of Triton X-100 to preserve PNGase enzymatic activity and incubation at 37 C overnight. For large

276


amounts of protein (> 1 mg), the sequential addition of small amounts of enzyme assists hydrolysis. The released oligosaccharides are analyzed by a suitable method to determine structure, such as mass spectrometry and HPAEC for unlabeled glycans, or NP-HPLC using fluorescently labeled glycans (Butters and Neville, 2008). For large-scale purification, Biogel P4 columns are used but smaller amounts are readily isolated using standard HPLC columns where either in-line electrochemical detection or postcolumn labeling and analysis are used to identify and isolate glucosylated glycans (Petrescu et al., 1997). The biosynthesis of glucosylated glycans in CHO cells in the presence of NB-DNJ favors the Glc3Man7(D1)GlcNAc2 oligosaccharide (Petrescu et al., 1997). Smaller amounts of Glc3Man8GlcNAc2 and Glc3Man9GlcNAc2 are also generated. 4.1.4. Triglucosylated N-glycans production and isolation from free glycans As an alternative to using proteins as a source of glucosylated oligosaccharide, treatment of cells with NB-DNJ results in the synthesis of free glycans as a product of glycoprotein degradation following misfolding. The glycan material can be harvested in cell-free extracts from MDBK cells where nonproteasomal hydrolysis in the ER produces Glc3Man7GlcNAc2 in high abundance (Butters and Alonzi, unpublished data). This material does not require chemical or enzyme-mediated release, and can be isolated using HPLC methods described above.

4.2. Diglucosylated high-mannose N-glycans The diglucosylated N-glycan was prepared by enzymatic digestion of Glc3Man7(D1)GlcNAc2 oligosaccharide (10–100 mg), prepared from glucosidase inhibitor treated CHO cells described above, using purified preparations of ER-a1,2-glucosidase I. The enzyme is purified in a single ligand affinity step from detergent extracts of porcine or rat liver, using DNJ linked to chromatography beads. The preparation of the affinity column and general procedures in the affinity chromatography are described below and the reader is advised to refer to primary papers cited. 4.2.1. Materials and equipment 1. N-Carboxypentyl-DNJ (CPDNJ) from Toronto Chemicals, Inc. or chemically synthesized as described below 2. Affi-Gel 102 (Bio-Rad) 3. 1-ethyl-3-(3-dimethylaminopropyl)-carbodiimide (EDC) (Sigma) 4. Porcine or rat liver 5. Ultraturrax homogenizer (Janke & Kunkle)


277

6. Homogenisation buffer: 50 mM Tris/HCl buffer, pH 7.2, containing 0.25 M sucrose, 5 mM leupeptin, 15 mM pepstatin A, 0.5 mM PMSF, and 1 mM 6-aminohexanoic acid 7. Buffer A (0.1 M sodium phosphate buffer, pH 7.0, containing 0.8% Lubrol PX) described above. 8. Sep-Pak C18 cartridge (1 cc, Waters) 9. NP-HPLC (Anachem) 4.2.2. Preparation of a-glucosidase I affinity ligand Coupling DNJ to chromatography beads provides a suitable ligand affinity support. CPDNJ is available from Toronto Chemicals, Inc., Ontario, Canada or can be synthesized by reacting DNJ with an excess of 6-bromohexanoic acid (Kaushal and Elbein, 1994). An amino-derivatized support such as Affi-Gel 102 allows simple coupling using EDC according to the following protocol. Aqueous solutions of CPDNJ (250 mmol) are adjusted to pH 4.8 with 1 M HCl and added to 5 ml of washed, packed Bio-Rad Affi-Gel 102. The gel is resuspended, adjusted to 9 ml with water, and 1 ml of 100 mM EDC is added. After maintaining the pH at 4.8 for 30 min at room temperature by the addition of 1 M HCl, the coupling reaction is allowed to proceed with gentle mixing for 16 h at 4 C. The gel is filtered using a sintered glass funnel and the filtrate is retained for the estimation of unbound ligand. The derivatized gel is washed sequentially with 100 ml of 50 mM NaOAc buffer, pH 4.5, containing 0.5 M NaCl and with 50 mM Tris/HCl buffer, pH 8.0, containing 0.5 M NaCl. This wash cycle is repeated twice before equilibrating the gel in an appropriate running buffer. The gel can be stored at 4 C in the presence of 0.02% azide and reused several times. To measure the concentration of coupled ligand, a sample of the gel filtrate containing alkylated imino-sugar is taken before and after coupling and subjected to HPAEC using an eluant of 150 mM NaOH/30 mM NaOAc (Dionex BioLC) and a CarboPac column. Alternatively, a Dionex CS10 column eluted with 50 mM sodium sulfate containing 5% (v/v) acetonitrile and in-line micromembrane suppression and electrochemical detection (Mellor et al., 2000) can be used. Typically, values of 20–25 mmol of ligand/ml of gel are obtained. 4.2.3. Isolation of a-glucosidase I The following procedures for extraction and purification of a-glucosidase I are performed at 4 C. Freshly obtained porcine or rat liver (300 g) is homogenized for 2–3 min (Ultraturrax homogenizer) with 0.8 l of 50 mM Tris/HCl buffer, pH 7.2, containing 0.25 M sucrose, 5 mM leupeptin, 15 mM pepstatin A, 0.5 mM PMSF, and 1 mM 6-aminohexanoic acid. The homogenate is filtered through cheesecloth and the filtrate is centrifuged at 15,000g for 30 min. The supernatant is recovered and centrifuged at 150,000g for 60 min

278


and the pelleted material is washed with 120 ml of homogenization buffer. The pellet is recovered by centrifugation and resuspended in 75 ml of 10 mM sodium phosphate buffer, pH 6.8, containing 0.5% v/v Triton X-100. The suspension is stirred for 30 min and the membrane fraction is recovered by centrifugation at 150,000g for 30 min. The Triton-extracted pellet is suspended in 120 ml of 0.2 M sodium phosphate buffer, pH 6.8, containing 0.8% Lubrol PX and stirred for 60 min. A glucosidase I-enriched supernatant is recovered by centrifugation at 150,000g for 90 min. 4.2.4. Affinity chromatography of a-glucosidase I For affinity chromatography of a-glucosidase I, the Lubrol PX-extract is mixed with 5 ml of CPDNJ-Affi-Gel (25 mmol ligand/ml gel), prepared as described above, for 18 h and the gel is recovered by low-speed centrifugation. The gel is washed with 250 ml of 0.1 M sodium phosphate buffer, pH 7.0, containing 0.8% Lubrol PX (buffer A) and then eluted with 25 ml of buffer A containing 100 mM NB-DNJ. The column eluate is pooled and the enzyme is stored at 4 C. Before use, an aliquot is dialyzed against 41 l of buffer A to remove NB-DNJ. Enzyme activity and purity is assessed using [14C-Glc]radiolabeled Glc3Man9GlcNAc2 as described (Jacob and Scudder, 1994) or using NP-HPLC with fluorescently labeled substrates (Alonzi et al., 2008). 4.2.5. Hydrolysis of Glc3Man7GlcNAc2 and isolation of Glc2Man7GlcNAc2 Purified Glc3Man7(D1)GlcNAc2 oligosaccharide (10–100 mg) was dried under vacuum and resuspended in 20 ml of a-glucosidase I in buffer A (25 U) and incubated at 37 C for 16 h. An aliquot (1 ml) was taken, labeled with anthranilic acid, 2-AA (Neville et al., 2004) and analyzed by NPHPLC to determine reaction completion. When all the substrate was hydrolyzed to Glc2Man7(D1)GlcNAc2 the enzyme and detergent was removed using a Sep-Pak C18 cartridge and the nonbound eluate containing diglucosylated oligosaccharide was used for further derivatization.

5. Preparation of NGL Probes The glucan oligosaccharides and the glucosylated high-mannose N-glycans prepared above were converted into oxime-linked NGLs (AONGLs) for recognition studies in microarrays. The step by step procedures for the preparation, purification, and quantitation of the AO-NGLs are essentially as described previously (Liu et al., 2006, 2007).


279

5.1. Materials and equipment 1. Reducing glucan oligosaccharides and glucosylated high-mannose N-glycans 2. Aminooxy-functionalized 1,2-dihexadecyl-sn-glycero-3-phosphoethanolamine (AOPE), step by step synthesis procedure described in Liu et al. (2006) 3. Glass microvials with Teflon-lined caps (Chromacol) 4. Aluminium-backed HPTLC plates (Merck) 5. Silica cartridges (Waters or Phenomenex) 6. Primulin and orcinol staining reagents (Chai et al., 2003) 7. Solvents and solutions: methanol and chloroform are of HPLC grade; ammonium acetate solution (0.2 M in deionized water); chloroform/ methanol/water (C/M/W).

5.2. Preparation of AO-NGLs of glucan oligosaccharides In brief, 100 nmol of each glucan oligosaccharide or oligosaccharide fraction (di- to heptasaccharide) was used for conjugation in the presence of 200 nmol of N-aminooxyacetyl-DHPE (AOPE) in 50 ml C/M/W (10:10:1, by volume). After incubation at ambient temperature 16 h, the reaction mixtures were evaporated slowly to dryness at 60 C and reconstituted in 100 ml of C/M/W (25:25:8). The reaction completion was determined by HPTLC analysis (1 ml of the solution applied; developed with C/M/W (60:35:8)) using primulin and then orcinol staining (Chai et al., 2003). The conjugation yields were greater than 80%. AO-NGLs of disaccharides and trisaccharides were purified by semipreparative HPTLC, and those of tetra to heptasaccharides were purified using silica cartridges (stepwise procedures were described in Liu et al. (2006)). Purified AO-NGLs are dissolved in C/M/W (25:25:8) to give an approximate concentration of 100 pmol/ml for HPTLC and MALDI-MS analyses, quantitation (Liu et al., 2006), and for storage (at 20 C). HPTLC analysis of purified AO-NGLs of disaccharides nigerose and kojibiose, and of dextran oligosaccharide fractions is shown in Fig. 13.3.

5.3. Preparation of AO-NGLs of the glucosylated N-glycans The procedures for preparing AO-NGLs of the three glucosylated N-glycans (Fig. 13.4) are similar to those described above, except that a smaller amount, 10 nmol of each N-glycan, was used for conjugation and 200 nmol AOPE was applied. The incubation time was prolonged to 24 h. Aliquots of the reaction mixtures (1/50) were analyzed by HPTLC (developed with C/M/W (55:45:10)) using primulin and orcinol staining.

Origin 1

2

3

4

5

6

7

8

1

2

3

4

5

6

7

8

Figure 13.3 HPTLC analysis of purified AO-NGLs of kojibiose (lane 1), nigerose (lane 2), and dextran oligosaccharide fractions di- to heptasaccharide (lanes 3–8). The left panel shows primulin staining, and the right panel orcinol staining. For each NGL, 500 pmol was applied.

A 100

2762 [M−H]

Mana 2Mana 6 Mana 3

%

Mana 2Mana

−

6 Manb 4GlcNAcb 4GlcNAc AO

3

Glca 3Mana 2Mana 2Mana

0 1200 1300 1400 1500 1600 1700 1800 1900 2000 2100 2200 2300 2400 2500 2600 2700 2800 2900 m/z

B 100

2559 [M−H]−

Mana 6 Mana 3

%

Mana

6 Manb 4GlcNAc AO 3

Glca 2Glca 3Glca 3Mana 2Mana 2Mana

2762 0 1200 1300 1400 1500 1600 1700 1800 1900 2000 2100 2200 2300 2400 2500 2600 2700 2800 2900 m/z

C 100

2397 [M−H]−

Mana 6 Mana 3

%

Mana

6 Manb 4GlcNAc AO 3

Glca 3Glca 3Mana 2Mana 2Mana

2600 2559 0 1200 1300 1400 1500 1600 1700 1800 1900 2000 2100 2200 2300 2400 2500 2600 2700 2800 2900 m/z

Figure 13.4 Sequences and negative-ion MALDI mass spectra of the three glucosylated high-mannose N-glycan AO-NGLs, Glc1Man9GlcNAc2 (A), Glc3Man7(D1) GlcNAc (B), and Glc2Man7(D1)GlcNAc (C). The [MH] ions observed are in accord with their expected values.


281

The AO-NGL products were purified by 1 cc silica cartridges (Liu et al., 2006). The purified NGLs can be analyzed by MALDI-MS. For this, the NGLs are dissolved in chloroform/methanol/water (25:25:8) at a concentration of 10 pmol/ml; 1 ml is deposited on the sample target together with a matrix of 2-(4-hydroxyphenylazo)benzoic acid. Negative-ion MALDI spectra of the AO-NGLs of Glc1Man9GlcNAc2, Glc3Man7(D1)GlcNAc, and Glc2Man7(D1)GlcNAc are shown in Fig. 13.4. It should be noted here that the lack of a GlcNAc residue at the chitobiose core in the AO-NGLs of tri- and diglucosylated oligosaccharides (Fig. 13.4B and C) was from the original oligosaccharide materials. The triglucosylated N-glycan material for NGL preparation was recovered from a NMR sample of Glc3Man7(D1) GlcNAc2 which contained filamentous bacteriophages. The loss of the GlcNAc residue was attributed (Schallus et al., 2008) to residual endoglycosidase activity present in the bacteriophages. As the diglucosylated N-glycan Glc2Man7(D1)GlcNAc was prepared from the triglucosylated analogue by enzymatic digestion, it also lacks the core GlcNAc residue.

6. Carbohydrate Microarray Analysis of Human Malectin The microarray used for studies of human malectin encompassed the AO-NGLs of the glucose oligosaccharide sequences and those from the tri-, di-, and monoglucosyl-high-mannose N-glycans (Fig. 13.5). In addition, the microarray included a diverse range of mammalian-type sequences, all lipid-linked: N-glycans of high-mannose and of neutral and sialylated complex-type; O-glycans, blood group-related sequences (A, B, H, Lewisa, Lewisb, Lewisx, and Lewisy) on linear or branched backbones and their sialylated and/or sulfated analogues, gangliosides, and oligosaccharide fragments of glycosaminoglycans and polysialic acid. Also included were homo-oligomers of other monosaccharides.

6.1. Materials and equipment 1. 16 pad nitrocellulose-coated glass slides (Whatman FAST slides, available from Sigma) 2. Noncontact arrayer (Piezorray, Perkin-Elmer LAS) 3. Cy3 mono NHS ester (Amersham Biosciences) 4. Mouse monoclonal anti-polyhistidine (Ab1) (Sigma) 5. Biotinylated antimouse IgG antibody (Ab2) (Sigma) 6. Saline buffer, for example, HBS (5 mM Hepes, pH 7.4, 150 mM NaCl) 7. Blocking solution: HBS containing 3% (w/v) bovine serum albumin (Sigma) and 5 mM CaCl2 8. Alexa Fluor-647-labeled streptavidin (Molecular Probes)

282


Mana-6 Mana-6 Mana-3

35,000

Manb-4GlcNAc-AO

Glca-3Glca-3Mana-2Mana-2Mana-3

Fluorescence intensity

45,000 40,000 35,000

Fluorescence intensity

Glc2-N-glycan

30,000

Glc2-N-glycan

30,000

25,000

20,000

15,000

10,000

Glucose oligomers 5000

0

25,000

1

21 41 61 81 101 121 141 161 181 201 221 241 261 281 301 321 341 361 381 401 421 441 461

Mammalian-type 20,000

Homo- Misc. oligomers

15,000 10,000 5000

,3

a1

a1

,2

2 2 2 3 4 5 6 7 2 3 4 5 6 7 2 3 4 5 6 7 2 3 4 5 6 7 2 3 4 5 6 7

a1,4

G

lu

co N syl -g at ly ed ca ns

G 3N G 2N G 1N

0

a1,6

b1,3

b1,4

b1,6

Oligosaccharide probes (dp)

Figure 13.5 Microarray analyses of the interactions of malectin using Glc3-, Glc2-, and Glc1-high-mannose N-glycans and glucan oligosaccharide probes in the context of a full microarray containing more than 370 mammalian-type sequences and homo-oligomers of other monosaccharides (inset). The asterisk indicates that the binding signal for the Glc2-N-glycan probe was too high to be accurately quantified, using the imaging conditions selected to highlight the binding of malectin to the glucan oligosaccharide sequences. Abbreviations G3N, G2N, and G1N designate Glc3Man7(D1)GlcNAc, Glc2Man7(D1)GlcNAc, and Glc1Man9GlcNAc2 N-glycan probes, respectively; dp, degree of polymerization of the glucan oligosaccharides. The probes and their sequences were described in the Supplementary Table 6 of the publication describing the xenopus malectin (Schallus et al., 2008).

9. Fast frame multislide plate and silicone incubation chambers (16 wells) (Whatman) 10. Fluorescence microarray slide scanner (ProScanArray) and ScanArrayExpress software (Perkin-Elmer LAS)

6.2. Microarray printing The lipid-linked oligosaccharide probes were robotically printed on 16 pad nitrocellulose-coated glass slides using a noncontact arrayer with a spot delivery volume of approximately 330 pl (Palma et al., 2006); further details


283

to be described elsewhere. Each probe was printed in duplicate at two levels, 2 and 5 fmol/spot. The Cy3 dye was included in the probe array solution for quality control of sample delivery while arraying and spot visualization while performing the quantitation analysis.

6.3. Probing the microarrays The microarray analysis with his-tagged malectin was performed with the protein precomplexed with mouse monoclonal anti-polyhistidine (Ab1) and biotinylated antimouse IgG antibodies (Ab2) in a ratio of 1:3:3 (by weight). The malectin–antibody complexes were prepared by preincubating Ab1 with Ab2 for 15 min at ambient temperature, followed by addition of malectin and incubation for a further 15 min. The arrayed slides were prewetted with water and blocked for 1 h with blocking solution, rinsed with HBS, and overlaid for 1.5 h with malectin–antibody complexes diluted in the blocking solution, to give a final malectin concentration of 5 mg/ml. Although calcium is not required for binding, we observed an enhancement of the binding signals elicited in the presence of the cation, especially toward the relative low avidity binders, for example, the glucan oligomers. The precomplexation of malectin with the detection antibodies, in order to increase the valency of the interaction, also resulted in enhancement of the binding signals elicited. Binding was detected using Alexa Fluor-647-labeled streptavidin for 45 min at 1 mg/ml in blocking solution. After each overlay step, slides were rinsed with HBS and an additional rinse with water was performed at the end of the binding experiment. The slides were dried and kept in the dark before scanning. The slides were scanned for Alexa Fluor-647 using a fluorescence microarray slide scanner and the spot fluorescence was quantified after background subtraction using the ScanArrayExpress software. Data analysis after quantitation and presentation was performed with a dedicated software developed by Mark S. Stoll of the Glycosciences Laboratory (Stoll and Feizi, 2009). The microarray results presented in Fig. 13.5 are the means of fluorescence intensities of duplicate spots, printed at 5 fmol. The error bars represent half of the difference between the two values. Human malectin, like its xenopus homologue (Schallus et al., 2008), has the property to bind glucan oligosaccharides, predominantly with a1,3-, a1,4-, and a1,6-linked glucose sequences, and shows an intense and highly selective binding to a diglucosylated high-mannose N-glycan (Fig. 13.5), among a diverse range of mammalian-type sequences that were included in the microarray (highlighted in the inset of Fig. 13.5).

284


7. Conclusions In conclusion, we illustrate here the flexibility of NGL-based microarrays for generating oligosaccharide probes from desired sources, polysaccharides, and ER N-glycans in this instance. The preparation of native glucosylated glycans from cells offers a number of advantages for the experimentalist. The use of selective metabolic inhibitors for glucosidase and mannosidase enzymes in the ER (e.g., kifunensine) increases the yield of glycoforms that would otherwise be difficult to isolate in amounts sufficient for analyses of bioactivities in recognition systems involved in glycan processing and recognition. Among the oligosaccharide probe sequences analyzed, the high-selective malectin binding is to a diglucosyl high-mannose N-glycan, which is an intermediate oligosaccharide in the N-glycosylation pathway in the ER. This now opens the way to investigations of the possible sites of action of malectin in the ER. In our investigations of the xenopus and human malectin, until now, we have used the truncated diglucosyl N-glycan with seven mannose residues (Man7). This is because the Glc3Man7(D1) GlcNAc2 is the most abundant analogue which accumulates on the glycoproteins of the drug-treated cells (Petrescu et al., 1997). Work is under way to determine which Glc2-high-mannose analogues, Man9, Man8 (two isoforms), Man7, Man4, give the highest binding signals with malectin. These upcoming studies on the size of the determinant may shed light on where the malectin acts in the N-glycosylation pathway.

ACKNOWLEDGMENTS We gratefully acknowledge contributions of our colleagues in the Glycosciences Laboratory: Maria Campanero-Rhodes, Mark Stoll, Alex Lawson, and Colin Hebert; and Thomas Schallus from the EMBL Heidelberg. The Glycosciences Laboratory acknowledges with gratitude the collaborators with whom our microarray probes were studied over the years. For grant support, we acknowledge the U.K. Medical Research Council, the U.K. Research Councils Basic Technology Grant (GR/S79268, ‘‘Glycoarrays’’), Engineering and Physical Research Councils Translational Grant EP/G037604/1, and the NCI Alliance of Glycobiologists for Detection of Cancer and Cancer Risk (U01 CA128416). A. S. P. is a fellow of the Fundac¸a˜o para a Cieˆncia e Tecnologia (SFRH/BPD/26515/2006, Portugal).

REFERENCES Alonzi, D. S., Neville, D. C., Lachmann, R. H., Dwek, R. A., and Butters, T. D. (2008). Glucosylated free oligosaccharides are biomarkers of endoplasmic-reticulum alphaglucosidase inhibition. Biochem. J. 409, 571–580.


285

Alonzi, D. S., Dwek, R. A., and Butters, T. D. (2009). Improved cellular inhibitors for glycoprotein processing alpha-glucosidases: Biological characterisation of alkyl- and arylalkyl-N-substituted deoxynojirimycins. Tetrahedron Asymmetry 20, 897–901. Best, D., Wang, C., Weymouth-Wilson, A. C., Clarkson, R. A., Wilson, F. X., Nash, R. J., Miyauchi, S., Kato, A., and Fleet, G. W. J. (2010). Looking glass inhibitors: Scalable syntheses of DNJ, DMDP, and (3R)-3-hydroxy-l-bulgecinine from D-glucuronolactone and of l-DNJ, l-DMDP, and (3S)-3-hydroxy-d-bulgecinine from l-glucuronolactone. DMDP inhibits b-glucosidases and b-galactosidases whereas l-DMDP is a potent and specific inhibitor of a-glucosidases. Tetrahedron Asymmetry 21, 311–319. Bogomolovas, J., Simon, B., Sattler, M., and Stier, G. (2009). Screening of fusion partners for high yield expression and purification of bioactive viscotoxins. Protein Expr. Purif. 64, 16–23. Butters, T. D., and Neville, D. C. A. (2008). In ‘‘Molecular Biomethods Handbook,’’ (J. M. Walker and R. Rapley, eds.), p. 495. Humana Press, Totowa, NJ. Butters, T. D., Sparks, L. M., Harlos, K., Ikemizu, S., Stuart, D. I., Jones, E. Y., and Davis, S. J. (1999). Effects of N-butyldeoxynojirimycin and the Lec3.2.8.1 mutant phenotype on N-glycan processing in Chinese hamster ovary cells: Application to glycoprotein crystallization. Protein Sci. 8, 1696–1701. Chai, W., Feizi, T., Yuen, C.-T., and Lawson, A. M. (1997). Nonreductive release of O-linked oligosaccharides from mucin glycoproteins for structure/function assignments as neoglycolipids: Application in the detection of novel ligands for E-selectin. Glycobiology 7, 861–872. Chai, W., Stoll, M. S., Galustian, C., Lawson, A. M., and Feizi, T. (2003). Neoglycolipid technology—Deciphering information content of glycome. Methods Enzymol. 362, 160–195. de Belder, A. N. (1993). Dextran. In ‘‘Industrial Gums,’’ (R. L. Whistler and J. BeMiller, eds.), pp. 399–426. Academic Press, New York. de la, C. J., Pintor-Toro, J. A., Benitez, T., and Llobell, A. (1995). Purification and characterization of an endo-beta-1,6-glucanase from Trichoderma harzianum that is related to its mycoparasitism. J. Bacteriol. 177, 1864–1871. Feizi, T., and Childs, R. A. (1994). Neoglycolipids: Probes in structure/function assignments to oligosaccharides. Methods Enzymol. 242, 205–217. Feizi, T., Stoll, M. S., Yuen, C.-T., Chai, W., and Lawson, A. M. (1994). Neoglycolipids: Probes of oligosaccharide structure, antigenicity and function. Methods Enzymol. 230, 484–519. Finn, R. D., Mistry, J., Schuster-Bockler, B., Griffiths-Jones, S., Hollich, V., Lassmann, T., Moxon, S., Marshall, M., Khanna, A., Durbin, R., Eddy, S. R., Sonnhammer, E. L., and Bateman, A. (2006). Pfam: Clans, web tools and services. Nucleic Acids Res. 34, D247–D251. Jacob, G. S., and Scudder, P. (1994). In ‘‘Methods Enzymology,’’ (W. J. Lennarz and G. W. Hart, eds.), p. 280. Academic Press, San Diego. Kaushal, G. P., and Elbein, A. D. (1994). In ‘‘Methods Enzymology,’’ (W. J. Lennarz and G. W. Hart, eds.), p. 316. Academic Press, San Diego. Letunic, I., Copley, R. R., Pils, B., Pinkert, S., Schultz, J., and Bork, P. (2006). SMART 5: Domains in the context of genomes and networks. Nucleic Acids Res. 34, D257–D260. Liu, Y., Chai, W., Childs, R. A., and Feizi, T. (2006). Preparation of neoglycolipids with ring-closed cores via chemoselective oxime-ligation for microarray analysis of carbohydrate–protein interactions. Methods Enzymol. 415C, 326–340. Liu, Y., Feizi, T., Campanero-Rhodes, M. A., Childs, R. A., Zhang, Y., Mulloy, B., Evans, P. G., Osborn, H. M., Otto, D., Crocker, P. R., and Chai, W. (2007). Neoglycolipid probes prepared via oxime ligation for microarray analysis of oligosaccharide– protein interactions. Chem. Biol. 14, 847–859.

286


Mellor, H. R., Adam, A., Platt, F. M., Dwek, R. A., and Butters, T. D. (2000). Highperformance cation-exchange chromatography and pulsed amperometric detection for the separation, detection, and quantitation of N-alkylated imino sugars in biological samples. Anal. Biochem. 284, 136–142. Mellor, H. R., Nolan, J., Pickering, L., Wormald, M. R., Platt, F. M., Dwek, R. A., Fleet, G. W., and Butters, T. D. (2002). Preparation, biochemical characterization and biological properties of radiolabelled N-alkylated deoxynojirimycins. Biochem. J. 366, 225–233. Neville, D. C., Coquard, V., Priestman, D. A., te Vruchte, D. J., Sillence, D. J., Dwek, R. A., Platt, F. M., and Butters, T. D. (2004). Analysis of fluorescently labeled glycosphingolipid-derived oligosaccharides following ceramide glycanase digestion and anthranilic acid labeling. Anal. Biochem. 331, 275–282. Ohta, M., Hamako, J., Yamamoto, S., Hatta, H., Kim, M., Yamamoto, T., Oka, S., Mizuochi, T., and Matsuura, F. (1991). Structures of asparagine-linked oligosaccharides from hen egg-yolk antibody (IgY). Occurrence of unusual glucosylated oligo-mannose type oligosaccharides in a mature glycoprotein. Glycoconj. J. 8, 400–413. Palma, A. S., Feizi, T., Zhang, Y., Stoll, M. S., Lawson, A. M., Diaz-Rodrı´guez, E., Campanero-Rhodes, A. S., Costa, J., Brown, G. D., and Chai, W. (2006). Ligands for the beta-glucan receptor, Dectin-1, assigned using ‘designer’ microarrays of oligosaccharide probes (neoglycolipids) generated from glucan polysaccharides. J. Biol. Chem. 281, 5771–5779. Petrescu, A. J., Butters, T. D., Reinkensmeier, G., Petrescu, S., Platt, F. M., Dwek, R. A., and Wormald, M. R. (1997). The solution NMR structure of glucosylated N-glycans involved in the early stages of glycoprotein biosynthesis and folding. EMBO J. 16, 4302–4310. Schallus, T., Jaeckh, C., Feher, K., Palma, A. S., Liu, Y., Simpson, J. C., Mackeen, M., Stier, G., Gibson, T. J., Feizi, T., Pieler, T., and Muhle-Goll, C. (2008). Malectin—A novel carbohydrate-binding protein of the endoplasmic reticulum and a candidate player in the early steps of protein N-glycosylation. Mol. Biol. Cell 19, 3404–3414. Spiro, R. G. (2004). Role of N-linked polymannose oligosaccharides in targeting glycoproteins for endoplasmic reticulum-associated degradation. Cell. Mol. Life Sci. 61, 1025–1041. Stoll, M. S., and Feizi, T. (2009). Software tools for storing, processing and displaying carbohydrate microarray data. (C. Kettner, ed.). Proceeding of the Beilstein Symposium on Glyco-Bioinformatics, 4–8 October, 2009, Potsdam, Germany, Beilstein Institute for the Advancement of Chemical Sciences, Frankfurt, Germany.

C H A P T E R

F O U R T E E N

Imaging Mass Spectrometry of Glycolipids Naoko Goto-Inoue, Takahiro Hayasaka, and Mitsutoshi Setou Contents 288 289 289 290 291 292 293 294 296 297 299 299

1. Overview 2. Preparation of Tissue Sections 2.1. Methods for preparing the cryosections 3. Matrix Selection and Applying Matrix 3.1. Methods for applying matrix 4. Measurements by Imaging Mass Spectrometry 5. Data Analyses 6. Identification of Molecules 7. Application of IMS 7.1. Methods for TLC-Blot-MALDI-IMS Acknowledgments References

Abstract Mass spectrometry (MS) is an analytical technique that separates ionized molecules using differences in their mass, and can be used to determine the structure of the molecules. Matrix-assisted laser desorption/ionization (MALDI) is one of the most commonly used ionization methods for this procedure. A new technical method, imaging mass spectrometry (IMS), which is a two-dimensional MS, enables molecular imaging of tissue sections by the use of the MALDI-MS method. In this chapter, we briefly discuss available methods for analyzing glycolipids by IMS. We describe sample detection strategies, and also introduce a representative example of its research application.

Department of Molecular Anatomy, Hamamatsu University School of Medicine, Handayama, Higashi-ku, Hamamatsu, Shizuoka, Japan Methods in Enzymology, Volume 478 ISSN 0076-6879, DOI: 10.1016/S0076-6879(10)78014-9

#


287

288

Naoko Goto-Inoue et al.

1. Overview Mass spectrometry (MS) is an analytical technique that separates ionized molecules by using differences in the ratios of their mass to charge (mass/charge; m/z), and can also be used to determine the structure of the molecules. Matrix-assisted laser desorption/ionization (MALDI) is one of the ionization methods most widely used with MS (Karas and Hillenkamp, 1988). MALDI allows the analysis of large numbers of biomolecules ranging from small metabolite molecules (m/z < 1000) (Harvey, 2008; Schiller et al., 2004) to much larger proteins with molecular weights of 100,000 daltons (Da) (Yates, 1998). In the procedure, molecules are covered with a matrix and ionized using a pulse laser beam. Owing to the widespread applicability of this method, MALDI is commonly used in various fields, such as proteomics (Li, 2009), lipidomics ( Jackson et al., 2007; Schiller et al., 2007), metabolomics (Szpunar, 2005), and glycomics (An and Lebrilla, 2010; Morelle and Michalski, 2007; Wada et al., 2007). Imaging mass spectrometry (IMS) is a two-dimensional mass spectrometry to visualize the spatial distribution of biomolecules (McDonnell and Heeren, 2007). There are several kinds of detection methods using IMS; the combination of MALDI and IMS is one of the best established (Cornett et al., 2007; Zaima et al., 2009a). MALDI-IMS does not require separation or purification of the target molecules, and enables researchers not only to identify unknown molecules but also to localize numerous molecules simultaneously in cells and tissue sections. Furthermore, tandem mass spectrometry in tissue, which is referred to as MSn, enables the structural analysis of a molecule detected by the first step of IMS (Hayasaka et al., 2009; Shimma et al., 2008). Because of such versatility, the optimization of experimental protocols for sample preparation and of conditions for MS measurements, and the choice of a data analysis method are important issues to consider. IMS was initially used for studying proteins or peptides (Chaurand et al., 2002; Kaletas et al., 2009; Lemaire et al., 2006b). At present, targets of IMS research have expanded to small endogenous metabolites such as phospholipids (Hayasaka et al., 2008; Sugiura et al., 2009), neutral lipids (Hayasaka et al., 2009), glycolipids (Ageta et al., 2009; Goto-Inoue et al., 2009b; Sugiura et al., 2008), and other endogenous metabolites (Khatib-Shahidi et al., 2006). Notably, the number of reports regarding IMS of small compounds has gradually increased. In this chapter, we briefly introduce strategies for the analysis of glycolipids that are currently possible by IMS, and describe the major sample detection strategy using IMS in more detail. We also introduce a representative example of its research application.

Imaging Mass Spectrometry of Glycolipids

289

2. Preparation of Tissue Sections This section is devoted to the preparation of tissue sections subjected to IMS measurements. Fresh frozen tissue sections should be prepared as usual, beginning with the freezing of tissue using powdered dry ice, liquid nitrogen, 2-methylbuthane, etc. In choosing the method, it is important to ensure that the shape of the tissue be well maintained. The process of preparing sections of the frozen tissue for IMS measurement is essentially similar to that used in the preparation of frozen sections for immunostaining or dye-staining. The direct subjection of sections to MS measurement in this method introduces a complication, however. Embedding with an optimal cutting temperature (OCT) compound usually allows samples to retain their shape and facilitates the cutting process, but in the preparation of tissue sections for IMS measurements, the use of embedding agents such as OCT compound must be avoided because the attachment and penetration of such polymer molecules in tissues causes serious inhibition of biomolecule ionization (Schwartz et al., 2003). Such polymer-like resin compounds generally have high ionization efficiency; this leads to a decrease in the detection sensitivity of other molecules. For this reason, when preparing sections for IMS measurement, an OCT compound should be used only for ‘‘supporting’’ the tissue blocks so that it does not directly attach to the tissue sections being analyzed. Unfortunately, without this embedding process, difficulties may be encountered in cutting certain tissues into thin sections. In such cases, Stoeckli et al. (2006) used a precooled semiliquid gel of 2% sodium carboxymethylcellulose (CMC) as an alternative embedding compound that does not interfere with the detection sensitivity of biomolecules by MS analysis. It is also important to regulate the thickness of the sections because ionization efficiency is partly dependent on the thickness of tissue sections (Sugiura et al., 2006). In general, 5- to 15-mm-thick slices are appropriate for IMS.

2.1. Methods for preparing the cryosections The frozen sections are sliced at an appropriate thickness with a cryostat; we use the Leica CM 1950 (Leica, Wetzlar, Germany). Frozen sections are then thaw-mounted on indium-tin oxide (ITO)-coated glass slides (Bruker Daltonics). The following steps are necessary in preparing frozen tissue sections: 1. The blade should be wiped with ethanol and attached to the cryostat beforehand. 2. Tissue, forceps, and brushes must be precooled to the temperature inside the cryostat.

290


3. Optimal conditions for thin sectioning should be determined: that is, the angle of the tissue, temperature, and thickness (5–15 mm). 4. As thin sectioning is performed, sections must be held with forceps or brushes. 5. Each slice should be immediately mounted on a glass slide. The glass slide must be electrically conductive to be analyzed by MS. ITO-coated glass slides are commercially available for IMS and are commonly used. 6. The section should be dried by a dryer or by transferring the section to a vacuum container, and then stored in a sealed case at 20 C until use. 7. When the sections are to be analyzed, they should be removed from the case and dried immediately. 8. The dried slide can then be subjected to the next process.

3. Matrix Selection and Applying Matrix In MALDI-MS analyses, choosing the appropriate matrix is the most important step. General recommendations for each type of biomolecule have been established in traditional MALDI-MS. These recommendations apply also to MALDI-IMS. Sinapic acid (SA) is frequently used as a suitable matrix for large protein measurement. On the other hand, 2,5-dihydroxybenzoic acid (DHB) is generally used as a suitable matrix for glycolipids (An and Lebrilla, 2010). In IMS, analytes must be extracted from tissue by the solvent and cocrystallized with matrices to be ionized. Therefore, the solvent condition is very important to the extraction. To enhance the protein/peptide extraction, the addition of detergents to the matrix solution is expected to increase analyte ion signals (Lemaire et al., 2006a). Nevertheless, a preliminary test for the optimization of matrix conditions, such as the concentration of matrix, amount of matrix, and composition of solution, is recommended for successful analyte detection. Additionally, there are various methods for the coating of the matrix to the section, such as spraying and deposition. The method of matrix application also influences analyte extraction efficiency. Compared to spraying, the deposition of matrix solution increases the signal sensitivity (Aerni et al., 2006), but decreases the spatial resolution. Spraying is one of the most frequently used methods in IMS (Agar et al., 2007). By the use of this method, an entire tissue section can be coated with relatively small crystals homogeneously in a short time without special equipment. For this operation, several instruments including TLC sprayers and artistic airbrushes are available; we use a metal airbrush with a 0.2-mm nozzle because of its simple and easy-to-handle design. Although this method seems to present few technical challenges, it nonetheless requires skillful operation because of the numerous parameters of the hand-operation of the airbrush.


291

If there is an excess of matrix solution on the tissue, an inhomogeneous crystal can be formed with analytes that have migrated from their original location; on the other hand, if not enough matrix solution is sprayed and it evaporates without sufficiently moisturizing the tissue section, analytes cannot be adequately extracted from the tissue section. The operation should be performed at a constant room temperature and humidity. Beginners are recommended to practice spraying until homogeneous matrix spraying can be achieved reproducibly. One of the critical limitations of the spatial resolution of MALDI-IMS is the size of the organic matrix crystal and the analyte migration during the matrix application process. To overcome these problems, our group reported nanoparticle-assisted laser desorption/ionization-based IMS of glycolipids (Ageta et al., 2009). Some of these nanoparticles could have selective ionization efficiency for glycolipids. In the future, we expect to develop a new highly effective matrix for glycolipids.

3.1. Methods for applying matrix All solvents used for MS were of HPLC grade. We used DHB (Bruker Daltonics, Leipzig, Germany) as the matrix, and an airbrush with a 0.2-mm caliber nozzle (Procon Boy FWA Platinum; Mr. Hobby, Tokyo, Japan) for the sprayer. To apply the matrix solution by the spraying method, the following procedure should be used: 1. For the detection of glycolipids, prepare 1 ml of matrix solution (50 mg/ml DHB in 70% methanol) in a 1.5-ml tube. 2. Matrix solution must be entirely dissolved by vortexing or brief sonication. 3. The prepared matrix solution should be transferred into the bottle of an airbrush. 4. The size of the droplet, the amount of mist, the angles, and distances between the nozzle (approximately 15 cm is best) and the sample should be optimized beforehand. 5. Approximately 0.5–1.0 ml of matrix solution should be sprayed on each glass slide, with the flow rate of approximately 0.1 ml/min. To dry the surface, it is possible to simply spray air onto the section. 6. After spraying, the sample must be dried in a cooling dryer. 7. The IMS measurement should be performed as soon as possible to minimize the progression of sample damage. Figure 14.1A shows good conditions for matrix spraying, and Fig. 14.1B shows examples of good and unfavorable results from spraying DHB matrix solution (50 mg/ml, 70% methanol) to a 1 1 cm area. In this case, approximately 0.5 ml of matrix solution was the best condition.

292


A

Angle Distance

Air brush • Mist size • Air pressure • Movement • Flow rate

Tissue ITO-slide glass

B

HE staining image

0 mL

Optical image

0.1 mL

1 mm

0.5 mL

0.1 mm

1 mL

0.1 mm

0.1 mm

Figure 14.1 (A) A schematic image of the matrix application. We had to carefully control the condition of mist size, air pressure, movement, and flow rate, which were optimized manually by an air brush. (B) Optical images of sprayed sections with various amounts (0, 0.1, 0.5, and 1 ml) of DHB solution applied to a 1 1 cm area. Small rectangles in the left lane show the localization of enlarged images (right lane). We found that 0.1 ml of matrix is insufficient but 1 ml is too much. Finally, we determined that 0.5 ml of matrix for spraying is the best condition for the sample.

4. Measurements by Imaging Mass Spectrometry IMS measurements should be performed as soon as possible after matrix application, regardless of the coating method. To obtain a good spectrum in IMS measurements, the procedure is almost the same as in traditional MALDI-MS measurements, so we have to optimize the mass range, detector gain, and laser power. From the mechanical setting perspective, there are two differences between MALDI-MS and MALDI-IMS.


293

One difference is that in IMS measurement, a two-dimensional region must be selected for analysis. In addition, the scan pitch, which decides the spatial resolution of the image, must be fixed because MALDI-IMS can ionize the molecules by laser pulse. The scan pitch, which means the distance between scans, depends on the laser size and mechanical movement control. At this moment, the commercially available instrument can analyze with a laser of approximately 25 mm (Goto-Inoue et al., 2009b). Moreover, we have developed a new instrument, which can move the sample stage by 1 mm, and the finest laser size is approximately 10 mm (Harada et al., 2009). This machine achieves the finest images obtainable by MALDI-IMS. However, these results are not sufficiently fine to detect cell-specific localization of molecules. When even finer images are needed for analysis, another ionization method, such as time-of-flight secondary ion mass spectrometry (TOF-SIMS), which is able to visualize with nm-order imaging, can be adapted to IMS. TOF-SIMS is a technique by which the target is directly submitted to a focused ion beam without matrix application. However, this technique is only applicable for the analysis of small molecules ( 0.1 s 1), the rate constant is most easily determined from a saturation-transfer experiment (Spera et al., 1991). For slower rates (kex < 0.01 s 1), exchange is measured by rapidly transferring the protein from H2O into D2O solution, and repeatedly acquiring 1 H–15N HSQC spectra to observe the exponential decrease in intensities with time (Fairbrother et al., 1991). Initial attempts to monitor the exchange of amide protons by transferring Fc into D2O solution failed because of their rapid exchange with the D2O solvent. Hence, the GlcNAc amide exchange rates were determined from saturation-transfer-type experiments

311

Stable Isotope-Assisted NMR Analyses of Glycoconjugates

A

118 120

GlcNAc-5¢ GlcNAc-5

15N

122

GlcNAc-2

(ppm)

124

9.0

8.0 1H (ppm)

7.0

126

B 15N

T1 (ms)

1000

15N

T2 (ms)

200

800

150

600

cN lc

lc G

G

{1H}-15N NOE

A

cA N

N lc

lc G

5¢

2 A

A N

A N lc G

c-

c-

5¢ c-

cA N lc G

G

0

5

50

0

2

200

5

100

400

kex(s–1)

1.0

5.0

0.5

4.0 3.0

0.0

2.0

–0.5

1.0

–1.0

c5 N lc G

A N lc G

A

c5¢

2 G

lc N

A

c-

5 A clc N G

lc N G

G

lc

N

A

A c-

5¢

c2

0.0

Figure 15.2 Dynamics and solvent accessibility of Fc glycan obtained from 15N NMR parameters and amide exchange rates of GlcNAc amide groups. (A) 1H–15N HSQC spectrum of mouse IgG2b-Fc labeled with 15N at the acetamide groups of GlcNAc residues (Kato and Yamaguchi, 2008; Kato et al., 2010). The HSQC peak of GlcNAc-1 was barely detectable due to severe line broadening. (Partially adapted from Kato and Yamaguchi, 2008.) (B) 15N relaxation and NOE parameters (15N T1, T2 and {1H}–15N NOE) and amide proton exchange rates (kex) of GlcNAc residues on IgG-Fc are shown (Yamaguchi, 2009). The kex values were obtained by a comparison of the 1H–15N HSQC peak intensities between the spectra measured with and without presaturation of the water resonance.

312

Yoshiki Yamaguchi and Koichi Kato

(Yamaguchi, 2009). The intensities of the GlcNAc amide signals, with and without saturation of the water signal, were measured to calculate exchange rates (Fig. 15.2B).

3.1. Determination of amide exchange rates from saturation-transfer experiments 1. Culture the IgG-producing cells in a modified medium (Table 15.1) in which 0.2 g/L of D-[15N]GlcNHCl and 200 mg/L of L-Ala were added and D-glucose content was reduced by half whereas contents of sodium succinate, succinic acid, and sodium pyruvate were increased twice. 2. Prepare Fc fragment by proteolytic fragmentation of the 15N-labeled IgG (ut supra). 3. Measure the 1H–15N HSQC spectra of this Fc sample with and without presaturation of the water signal. 4. Perform the experiment at different pH conditions to distinguish resonance attenuation caused by hydrogen exchange from attenuation caused by cross relaxation. 5. Estimate the T1 of amide 1H. 6. Amide exchange rate constant kex (s 1) is calculated as: kex ðpH1 Þ ¼ ½M0 =Mps ðpH2 Þ M0 =Mps ðpH1 Þ=½ð10ðpH2 pH1 Þ 1Þ T1 in which kex (pH1) is the amide hydrogen exchange rate constant at pH1, T1 is the longitudinal relaxation time of the amide proton, M0 is the resonance intensity without presaturation, and Mps is resonance intensity with presaturation. Solvent exchange data show that the GlcNAc-2 acetamide proton is relatively shielded from the solvent as compared with those of GlcNAc-5 and -50 , again consistent with the crystal structure of IgG-Fc (Kolenko et al., 2009). Deuterium-induced isotope shifts of 13C resonances also offer useful information on proton exchange rates at adjacent sites (Kainosho et al., 1987; Takeda et al., 2009, 2010). For example, the carbonyl 13C resonance originating from the GlcNAc acetamide group in 50% 1H2O/50% D2O is observed as either a coalescent singlet or as a doublet with a chemical shift difference of 0.1 ppm in regimes of extremely fast and slow exchange rates of the acetamide proton, respectively. The results obtained using IgG-Fc are thus consistent with what was concluded from saturationtransfer data (Kato et al., 2010). Similar experiments can be designed for estimating the exchange rates of hydroxyl protons of carbohydrates (Christofides and Davies, 1983; Kato et al., 2008). This line of study will


313

be important for characterizing accessibility and hydrogen bonding of carbohydrate moieties, these being rich in hydroxyl groups.

4. Identification of Binding Surfaces in Glycoprotein/Ligand Complexes Once assignments have been made, NMR peaks provide valuable information on the interactions of glycoconjugates with their cognate ligands. Stable-isotope-assisted NMR spectroscopy has assisted in investigating IgG-Fc glycoprotein–ligand interactions for FcgR and bacterial IgGbinding proteins such as protein A and protein G (Kato et al., 1993, 1995, 2000). A more recent example is the identification of the binding site of an RNA aptamer on human IgG-Fc (Miyakawa et al., 2008). Aptamers are short, folded DNA or RNA molecules that can be selected in vitro on the basis of their high affinity for a target molecule. An optimized 23-nucleotide aptamer, Apt8-2, binds to the Fc region of human IgG with high affinity, but not to other IgGs. Apt8-2 competes with protein A, but not with the Fcg receptor, for binding to IgG (Fig. 15.3).

4.1. Uniform 13C/15N-labeling for NMR assignments of the polypeptide backbone of glycoproteins 1. Prepare a modified Nissui NYSF 404 medium in which glucose, sodium pyruvate, succinic acid, and amino acids are replaced by 2 g/L D-[13C6] glucose, 110 mg/L [13C3]pyruvic acid sodium salt, 59 mg/L [13C4] succinic acid, and 1 g/L [15N/13C]algal amino acid mixture supplemented with 149 mg/L L-[13C6,15N4]ArgHCl, 42.5 mg/L L-[13C4,15N2] AsnH2O, 24 mg/L L-[13C3,15N]Cys, 450 mg/L L-[13C5,15N2]Gln, 17 mg/L L-[13C6,15N3]HisHClH2O, 27 mg/L L-[13C9,15N]Tyr, and 7 mg/L L-[13C11,15N2]Trp. 2. Cultivate CHO cells producing human IgG1 in the Nissui NYSF 404 medium supplemented with 2% dialyzed fetal bovine serum. 3. After cell growth, concentrate the supernatant with a Millipore Pelicon ultrafiltration system and then purify using Affi-gel protein A column (ut supra). 4. Digest the IgG with papain and purify its Fc fragment. The assignments of the 1H–15N HSQC peaks originating from the Fc backbone are carried out using uniformly 13C/15N-labeled human IgG1Fc. By use of triple resonance experiments and a selective 13C–15N double labeling method (Kainosho and Tsuji, 1982; Kato et al., 1989, 1991a,b), 131 amide peaks can be assigned, which correspond to 66% of the amino acid

A 100 T250

105

G402

110 115

15N

(ppm)

S400

L432 L251

F404

120

V264

125 A339

130 135 11.0

10.0

9.0 8.0 1H(ppm)

B

7.0

6.0

C 327 328 234 329 235 330 236 331 237 332 238 239

264

269 268 237 297 267 238 266 298 265 299 239

281

CH2

338 339 340 246 397

403 404 400 402 372 344 405 406 371

316 314 250 251 431 432

309 310 315 311 254 251 253 430 436 252 314

435 432 433 434

CH3

Human IgG1-Fc

Figure 15.3 Determination of an aptamer-binding interface on IgG-Fc. (A) Superposition of the 1H–15N HSQC spectra of the 15N-G, I, L, K, T, V, A, M, C, H, W, Y, F, S-labeled Fc fragment (black) and that of the aptamer-bound Fc fragment (red). Relevant amino acid positions are assigned. (B) Spatial localization of amino acids perturbed upon addition of the aptamer. Amino acid residues that showed NMR chemical shifts upon binding to the aptamer are indicated in red for large shifts, orange for medium shifts, and purple for residues that were perturbed but not quantitatively assigned due to overlap or line broadening of the signals. The assigned residues are marked only on one chain of the CH2 and CH3 domains for clarity. (C) Amino acids responsible for binding FcgR and protein A are shown in cyan and green, respectively (Deisenhofer, 1981; Kato et al., 2000; Sondermann et al., 2000). These figures are adapted from Miyakawa et al., 2008.


315

residues except for prolines and the N-terminal residue, and are used as spectroscopic probes (Yamaguchi et al., 2006). To identify the aptamer-binding site on the Fc fragment, we performed HSQC titration experiments using human IgG1-Fc labeled with 15N at selected amino acid residues (Miyakawa et al., 2008). Several chemical-shift changes are observed upon titration with an Apt8-2 derivative (Fig. 15.2A). For example, the Thr250 resonance completely disappears from its original position and appears at another position in the spectrum when a two molar equivalent of aptamer is added to Fc, suggesting a 1:2 (Fc:aptamer) stoichiometric interaction. The observed chemical-shift changes were quantified for each residue according to the equation [((DdHN)2 þ (DdN/5)2)/2]1/2, where DdHN and DdN represent the difference in proton and nitrogen chemical-shift changes between the free and the aptamer-bound forms, respectively (Pellecchia et al., 1999). The NMR chemical-shift analyses successfully localize the aptamer-binding sites on the Fc fragment (Fig. 15.2B), which partially overlaps the protein A-binding site but not the Fcg receptorbinding site (Fig. 15.2C), in line with the binding property of this aptamer. The amino acid residues constituting the recognition sites thus identified on human IgG-Fc are not conserved in IgG from other species; this, in part, accounts for the high specificity of the selected aptamer. Interestingly, an extensive portion of the surface of the CH3 domain is involved in aptamer binding and this mode of interaction has never before been shown with natural ligands of IgGs.

5. NMR-Based Screening of Glycopeptides Reactive with Lectin Isotope-assisted NMR analyses are also applicable to glycopeptide/ lectin complexes. This is exemplified by our analyses of the ligand binding of Fbs1. Fbs1 is a cytosolic lectin putatively operating as a chaperone as well as a substrate-recognition subunit of the SCFFbs1 ubiquitin ligase complex (Yoshida et al., 2002, 2007). Previous X-ray crystallographic and stableisotope-assisted NMR studies using chitobiose and Man3GlcNAc2 as ligands show that the sugar-binding domain (SBD) of Fbs1 is composed of a 10-stranded b-sandwich fold with two a-helices, and interacts with the Man3GlcNAc2 portion of oligosaccharides through the loops connecting the b-strands (Mizushima et al., 2004). To provide a structural and a functional basis for the preferential binding of Fbs1 to malfolded glycoproteins as targets of the ubiquitin/proteasome-mediated protein degradation system, we examined possible interactions of Fbs1 with glycopeptides derived from isotopically labeled IgG-Fc (Yamaguchi, 2009).

316


5.1. Preparation of glycopeptides for NMR study 1. Dissolve isotopically labeled Fc in 0.5 M Tris–HCl buffer, pH 8.5, containing 6 M guanidium chroride, 16 mM dithiothreitol, and 2 mM EDTA. 2. After addition of 32 mM iodoacetic acid, dialyze the reaction mixture against 50 mM NH4HCO3, pH 7.9. 3. Incubate the reduced and alkylated Fc sample with V8 protease at an enzyme/substrate molar ratio of 1:30 at 37 C for 12–24 h. 4. Load the reaction mixture onto an ODS reverse-phase column and fractionate the peptides. 5. Identify isolated glycopeptides by MALDI-TOF-MS. Systematically modify glycoforms by use of appropriate glycosidases, b-galactosidase, N-acetylhexosaminidase, and a-fucosidase. Dissolve glycopeptide in 0.1 M citrate-phosphate buffer, pH 4.5, at a concentration of 1 mg/ mL and incubate in the presence or absence of jack bean b-galactosidase (0.5 U/mL), N-acetylhexosaminidase (0.5 U/mL), and bovine kidney a-L-fucosidase (0.5 U/mL) at 16 C for 12 h. 6. Perform further trypsin digestion to trim the peptide portion. Dissolve the isolated glycopeptide in 50 mM NH4HCO3 (pH 7.9) at a concentration of 0.2 mg/mL and incubate with trypsin (25 mg/mL). 7. Isolate glycopeptides by the ODS column. To characterize the ligand-binding specificity of Fbs1 by NMR spectroscopy, a series of 13C-labeled glycopeptides is prepared (Fig. 15.4A). Upon addition of Fbs1-SBD, HSQC peaks were significantly perturbed for A

4¢ M

M-GN-GN 3 2 1

M 4

4¢ M M 4

F

F

M-GN-GN 3 2 1

F 5¢ 4¢ GN-M

M-GN-GN 3 2 1

GN-M 4 5

5 ¢ 4¢ GN-M

F

M-GN-GN 3 2 1

GN-M 4 5

75

1

80

1

1

1

4¢

4¢ F

85 13

C (ppm)

90 95

4¢ 100 105 6.0

4 5.0 1

3

4¢ F 3 2 4

2 4.0

H(ppm)

3.0 6.0

5.0 1

4 4.0

H(ppm)

3.0 6.0

3

5, 5¢

5.0 1

4

3

2 4.0

H(ppm)

Figure 15.4 (Continued)

3.0 6.0

5.0 1

5, 5¢

2 4.0

H(ppm)

3.0


B 4⬘ Mana 1 Mana 1 4

317

6 Manb1 - 4GlcNAcb1 - 4GlcNAcb1 3 2 1 3 Asp1-Tyr2-Asn3-Ser4-Thr5-Ile6-Arg7 ppm GlcNAc-1 Ac

0 1 2

Asn3 Hb

3 4 F1 (1H) 5

GlcNAc-1 H1

6 Tyr279 He

7 8

≠

9

Asn3 Hd

10 8.4

8.2

ppm

F3 (1H)

Figure 15.4 NMR analyses of glycopeptides/Fbs1-SBD interactions. (A) Binding ability of Fbs1 with a variety of 13C-labeled glycopeptides. 1H–13C HSQC spectra of glycopeptides in the presence (red) and absence (black) of the sugar-binding domain of Fbs1. Each glycan structure is shown with the corresponding NMR spectrum. (Adapted from Yamaguchi, 2009.) (B) NMR analysis of glycopeptides/Fbs1-SBD complex by observing intermolecular NOEs. Part of 15N-edited NOESY spectrum of the 15Nlabeled glycopeptide bound to Fbs1-SBD with a molecular mass of 20 kDa (F2(15N) ¼ 129.6 ppm corresponding to the Asn3 Nd of the glycopeptide). (Partially adapted from Yamaguchi et al., 2007a).

318


peptides carrying Man3GlcNAc2, Man3GlcNAc2Fuc1, and Man3GlcNAc4. Interestingly, the glycopeptide carrying a fucosylated complex type oligosaccharide exhibits little spectral change.

6. NOE Analyses of Lectin–Ligand Interactions For weakly binding sugar ligands, two NMR techniques are widely employed, that is, saturation-transfer difference-NMR and transferred NOE (Angulo et al., 2006; Hanashima et al., 2010) to characterize the interactions with the cognate lectins. For medium to strong binding, glycan–lectin intermolecular NOE connectivities are crucial for obtaining the structure of a sugar–lectin complex at atomic resolution. Stable isotope labeling facilitates distinguishing between intermolecular and intramolecular NOEs. Figure 15.4B shows the 3D 15N-edited NOESY spectrum of the 15N-labeled glycopeptides interacting with unlabeled Fbs1-SBD (Yamaguchi et al., 2007a). The resonances originating from the glycopeptide bound to Fbs1-SBD are assigned using 3D HNCA, HN(CO)CA, HN (CA)NNH, and 15N-edited NOESY spectra. In addition to inter-residue NOE connectivities within glycopeptides, there are intermolecular NOE connectivities between the asparagines–GlcNAc junction part and the aromatic ring of Tyr279 of Fbs1-SBD (Fig. 15.4B). Fbs1 interacts with sugar–polypeptide junctions; these are usually shielded in native proteins, but accessible in malfolded glycoproteins. In fact, surface plasmon resonance data show that the binding constants of Fbs1-SBD are higher for the peptide-linked Man3GlcNAc2 than for the pentasaccharide alone suggesting that the peptide moiety of the substrate contributes to the affinity for Fbs1SBD (Yamaguchi et al., 2007a).

7. Perspective In this chapter, we describe applications of stable-isotope-assisted NMR to characterize the dynamics and interactions of glycoconjugates mainly based on the conventional data of chemical shift, NOE, and relaxation. These approaches can be extended to more quantitative and in-depth analyses of glycoconjugate structures by sophisticated experimental designs. For example, by introducing lanthanide probes into carbohydrates and their complexes with proteins, paramagnetic relaxation enhancement (PRE) and pseudocontact shift can be utilized as sources of long-distance geometric information (Clore et al., 2007; Shapira and Prestegard, 2010). Spin-labeling techniques can be applied to observe PREs for characterization of carbohydrate–lectin (Zhuang et al., 2008) and peptide–glycolipid interactions (Yagi-Utsumi et al., 2010).


319

Use of residual dipolar coupling is also a promising approach for detailed analyses of conformations and dynamics of glycoconjugates (Bax, 2003; Lange et al., 2008). In addition, trans-hydrogen bond scalar coupling constants can be used for the unambiguous identification of the acceptor group in hydrogen bonds, something not possible with amide exchange measurements (Wang et al., 1999). In most cases, the biological functions of glycoconjugates are expressed through clustering of their glycans. Therefore, it is also envisioned that techniques will be developed to enable us to design glycan assemblies for stable-isotope-assisted NMR purposes. These include the preparation of desired glycoforms of glycoproteins by using bacterial glycoprotein expression systems combined with in vitro chemoenzymatic synthesis (Schwarz et al., 2010), segmental isotope labeling methods (Skrisovska et al., 2010), and construction of physicochemically controllable glycan clusters (Santos et al., 2009). Concomitant developments in stable-isotope-assisted NMR spectroscopy and chemical, biochemical, and molecular biology methodology will lead to a new level of understanding of structure–function relationships in glycoconjugates.

ACKNOWLEDGMENTS We thank Dr. Mitsuo Sato and Dr. Kenya Shitara (Kyowa Hakko Kirin Co., Ltd.) for providing CHO cells producing the human IgG1 antibody used in these studies, and Dr. Markus Wa¨lchli (Bruker Biospin) for help in NMR measurements. Work on the Fc-binding RNA aptamer is a collaborative effort, and we thank Dr. Shin Miyakawa (Ribomic, Inc.), Dr. Taiichi Sakamoto (Chiba Institute of Technology), Dr. Yusuke Nomura (Chiba Institute of Technology), and Prof. Yoshikazu Nakamura (The University of Tokyo). We thank Dr. Keiji Tanaka and Dr. Yukoko Yoshida (Tokyo Metropolitan Institute of Medical Science) with whom we collaborate on the Fbs1–glycopeptide interactions. This work was supported by Grants-in-Aid for Scientific Research and the Nanotechnology Network Project from the Ministry of Education, Culture, Sports, Science and Technology, Japan, by the Program for Promotion of Fundamental Studies in Health Sciences of the National Institute of Biomedical Innovation (NIBIO) and by Core Research for Evolution Science and Technology (CREST) from the Japan Science and Technology Agency.

REFERENCES Angulo, J., et al. (2006). NMR analysis of carbohydrate–protein interactions. Methods Enzymol. 416, 12–30. Arnesano, F., et al. (2005). NMR structures of paramagnetic metalloproteins. Q. Rev. Biophys. 38, 167–219. Bax, A. (2003). Weak alignment offers new NMR opportunities to study protein structure and dynamics. Protein Sci. 12, 1–16. Bertini, I., et al. (2005). NMR spectroscopy of paramagnetic metalloproteins. Chembiochem 6, 1536–1549. Bundle, D. R., et al. (1994). Solution structure of a trisaccharide–antibody complex: Comparison of NMR measurements with a crystal structure. Biochemistry 33, 5183–5192.

320


Christofides, J. C., and Davies, D. B. (1983). Secondary isotope multiplet NMR spectroscopy of partially labeled entities. Carbon-13 SIMPLE NMR of carbohydrates. J. Am. Chem. Soc. 105, 5099–5105. Clore, G. M., et al. (2007). Elucidating transient macromolecular interactions using paramagnetic relaxation enhancement. Curr. Opin. Struct. Biol. 17, 603–616. Deisenhofer, J. (1981). Crystallographic refinement and atomic models of a human Fc fragment and its complex with fragment B of protein A from Staphylococcus aureus at 2.9- and 2.8-A˚ resolution. Biochemistry 20, 2361–2370. Fairbrother, W. J., et al. (1991). Polypeptide backbone resonance assignments and secondary structure of Bacillus subtilis enzyme IIIglc determined by two-dimensional and threedimensional heteronuclear NMR spectroscopy. Biochemistry 30, 6896–6907. Felli, I. C., and Brutscher, B. (2009). Recent advances in solution NMR: Fast methods and heteronuclear direct detection. Chemphyschem 10, 1356–1368. Foster, M. P., et al. (2007). Solution NMR of large molecules and assemblies. Biochemistry 46, 331–340. Hanashima, S., et al. (2010). Synthesis and binding analysis of unique AG2 pentasaccharide to human Siglec-2 using NMR techniques. Bioorg. Med. Chem. 18, 3720–3725. Kainosho, M., and Tsuji, T. (1982). Assignment of the three methionyl carbonyl carbon resonances in Streptomyces subtilisin inhibitor by a carbon-13 and nitrogen-15 doublelabeling technique. A new strategy for structural studies of proteins in solution. Biochemistry 21, 6273–6279. Kainosho, M., et al. (1987). Local structural features around the C-terminal segment of Streptomyces subtilisin inhibitor studied by carbonyl carbon nuclear magnetic resonances three phenylalanyl residues. Biochemistry 26, 1068–1075. Kato, K., and Yamaguchi, Y. (2008). Structural analyses of glycoconjugates by NMR. Experimental Glycoscience Glycochemistry Springer, Japan, pp. 45–50. Kato, K., et al. (1989). Application of 13C nuclear magnetic resonance spectroscopy to molecular structural analyses of antibody molecules. J. Biochem. 105, 867–869. Kato, K., et al. (1991a). Complete assignment of the methionyl carbonyl carbon resonances in switch variant anti-dansyl antibodies labeled with [1-13C]methionine. Biochemistry 30, 270–278. Kato, K., et al. (1991b). Carbon-13 NMR study of switch variant anti-dansyl antibodies: Antigen binding and domain–domain interactions. Biochemistry 30, 6604–6610. Kato, K., et al. (1993). 13C NMR study of the mode of interaction in solution of the B fragment of staphylococcal protein A and the Fc fragments of mouse immunoglobulin G. FEBS Lett. 328, 49–54. Kato, K., et al. (1995). Model for the complex between protein G and an antibody Fc fragment in solution. Structure 3, 79–85. Kato, K., et al. (2000). Structural basis of the interaction between IgG and Fcg receptors. J. Mol. Biol. 295, 213–224. Kato, K., et al. (2008). 920 MHz ultra-high field NMR approaches to structural glycobiology. Biochim. Biophys. Acta 1780, 619–625. Kato, K., et al. (2010). Stable-isotope-assisted NMR approaches to glycoproteins using immunogloblin G as a model system. Prog. Nucl. Magn. Reson. Spectrosc. 56, 346–359. Kolenko, P., et al. (2009). New insights into intra- and intermolecular interactions of immunoglobulins: Crystal structure of mouse IgG2b-Fc at 2.1-A˚ resolution. Immunology 126, 378–385. Lange, O. F., et al. (2008). Recognition dynamics up to microseconds revealed from an RDC-derived ubiquitin ensemble in solution. Science 320, 1471–1475. Matsumiya, S., et al. (2007). Structural comparison of fucosylated and nonfucosylated Fc fragments of human immunoglobulin G1. J. Mol. Biol. 368, 767–779.


321

Miyakawa, S., et al. (2008). Structural and molecular basis for hyperspecificity of RNA aptamer to human immunoglobulin G. RNA 14, 1154–1163. Mizushima, T., et al. (2004). Structural basis of sugar-recognizing ubiquitin ligase. Nat. Struct. Mol. Biol. 11, 365–370. Nezlin, R. S. (1998). The Immunoglobulins: Structure and Function. Academic Press, USA. Pellecchia, M., et al. (1999). Pilus chaperone FimC–adhesin FimH interactions mapped by TROSY-NMR. Nat. Struct. Biol. 6, 336–339. Santos, J. I., et al. (2009). Assessing carbohydrate–carbohydrate interactions by NMR spectroscopy: The trisaccharide epitope from the marine sponge Microciona prolifera. Chembiochem 10, 511–519. Schwarz, F., et al. (2010). A combined method for producing homogeneous glycoproteins with eukaryotic N-glycosylation. Nat. Chem. Biol. 6, 264–266. Shapira, B., and Prestegard, J. H. (2010). Electron–nuclear interactions as probes of domain motion in proteins. J. Chem. Phys. 132, 115102. Skrisovska, L., et al. (2010). Recent advances in segmental isotope labeling of proteins: NMR applications to large proteins and glycoproteins. J. Biomol. NMR 46, 51–65. Sondermann, P., et al. (2000). The 3.2-A˚ crystal structure of the human IgG1 Fc fragment– FcgRIII complex. Nature 406, 267–273. Spera, S., et al. (1991). Measurement of the exchange rates of rapidly exchanging amide protons: Application to the study of calmodulin and its complex with a myosin light chain kinase fragment. J. Biomol. NMR 1, 155–165. Takeda, M., et al. (2009). Hydrogen exchange rate of tyrosine hydroxyl groups in proteins as studied by the deuterium isotope effect on Cz chemical shifts. J. Am. Chem. Soc. 131, 18556–18562. Takeda, M., et al. (2010). Detection of the sulfhydryl groups in proteins with slow hydrogen exchange rates and determination of their proton/deuteron fractionation factors using the deuterium-induced effects on the 13Cb NMR signals. J. Am. Chem. Soc. 132, 6254–6260. Wang, Y. X., et al. (1999). Measurement of 3hJNC’ connectivities across hydrogen bonds in a 30 kDa protein. J. Biomol. NMR 14, 181–184. Yagi-Utsumi, M., et al. (2010). NMR characterization of the interactions between lysoGM1 aqueous micelles and amyloid b. FEBS Lett. 584, 831–836. Yamaguchi, Y. (2009). Development and applications of stable-isotope-labeling methods oriented to strucutral glycobiology. Trends Glycosci. Glycotechnol. 20, 117–130. Yamaguchi, Y., and Kato, K. (2007). Structural glycobiology by stable-isotope-assisted NMR spectroscopy. (G. A. Webb, ed.), Vol. 1, pp. 219–225. Springer-Verlag, The Netherlands. Yamaguchi, Y., et al. (1995). Proteolytic fragmentation with high specificity of mouse immunoglobulin G. Mapping of proteolytic cleavage sites in the hinge region. J. Immunol. Methods 181, 259–267. Yamaguchi, Y., et al. (1998). Dynamics of the carbohydrate chains attached to the Fc portion of immunoglobulin G as studied by NMR spectroscopy assisted by selective 13C labeling of the glycans. J. Biomol. NMR 12, 385–394. Yamaguchi, Y., et al. (2006). Glycoform-dependent conformational alteration of the Fc region of human immunoglobulin G1 as revealed by NMR spectroscopy. Biochim. Biophys. Acta 1760, 693–700. Yamaguchi, Y., et al. (2007a). Fbs1 protects the malfolded glycoproteins from the attack of peptide:N-glycanase. Biochem. Biophys. Res. Commun. 362, 712–716. Yamaguchi, Y., et al. (2007b). Molecular interactions: Antibody structures and functions. ( J. P. Kamerling, ed.), Vol. 3, pp. 745–763. Elsevier, Oxford. Yamaguchi, Y., et al. (2009). A 13C-detection NMR approach for large glycoproteins. Carbohydr. Res. 344, 535–538.

322


Yokoyama, S. (2003). Protein expression systems for structural genomics and proteomics. Curr. Opin. Chem. Biol. 7, 39–43. Yoshida, Y., et al. (2002). E3 ubiquitin ligase that recognizes sugar chains. Nature 418, 438–442. Yoshida, Y., et al. (2007). A neural-specific F-box protein Fbs1 functions as a chaperone suppressing glycoprotein aggregation. J. Biol. Chem. 282, 7137–7144. Zhuang, T., et al. (2008). Structure determination of a Galectin-3-carbohydrate complex using paramagnetism-based NMR constraints. Protein Sci. 17, 1220–1231.

C H A P T E R

S I X T E E N

Self and Nonself Recognition with Bacterial and Animal Glycans, Surveys by Synthetic Chemistry Yukari Fujimoto, Katsunori Tanaka, Atsushi Shimoyama, and Koichi Fukase Contents 1. Overview 2. Bacterial Glycoconjugates for Nonself Recognition—Lipopolysaccharide (LPS) 3. Synthesis of H. pylori Kdo–Lipid A Backbone 4. Glycosylation with Kdo Donor 11 5. Cytokine (IL-6) Induction in Human Peripheral Whole-Blood Cell Cultures 6. Bacterial Glycoconjugates for Nonself Recognition—Peptidoglycan (PGN) 7. Synthesis of Disaccharide Moiety 15 in Tracheal Cytotoxin with b-Selective Glycosylation 8. Synthesis of Tracheal Cytotoxin 20 and Its Fragment 21 9. Immunostimulatory Activities of DAP Containing PGN Fragments 10. Visualizing the In Vivo Dynamics of Animal N-Glycans 11. PET Imaging of Glycoproteins 12. Method for 68Ga-DOTA Labeling and MicroPET Imaging in Rabbit 13. PET Imaging of Glycoclusters 14. Method for Preparation of N-Glycan Clusters and PET Imaging in Mouse Acknowledgments References

324 325 327 327 328 329 331 332 333 333 334 335 337 338 339 340

Abstract In this chapter, we describe synthetic studies on partial structures of lipopolysaccharide (LPS) and peptidoglycan (PGN), which work as tags for nonself recognition in innate immune system. Our previous studies demonstrated that Department of Chemistry, Graduate School of Science, Osaka University, Machikaneyama, Toyonaka, Osaka, Japan Methods in Enzymology, Volume 478 ISSN 0076-6879, DOI: 10.1016/S0076-6879(10)78016-2

#


323

324

Yukari Fujimoto et al.

lipid A is the endotoxic principle of LPS. The synthetic homogeneous preparations have enabled not only precise structure–activity relationships, but also recognition mechanisms of LPS with innate immune receptor complexes, including the TLR4/MD-2 complex, to be studied. Synthetic studies of lipid A and Kdo–lipid A from parasitic Helicobacter pylori revealed their low inflammatory activities, suggesting the molecular evolution to escape from the host immune system. A synthetic study of the partial structures of PGN has also contributed to the understanding of the innate immune mechanism. The biological activities of the synthetic fragments have revealed that the intracellular receptor Nod2 recognizes partial structures containing the muramyl dipeptide (MDP) moiety. The PGN of Gram-negative bacteria and some Gram-positive bacteria contain meso-diaminopimelic acid (meso-DAP), and recent studies have revealed that the intracellular receptor Nod1 recognizes DAP-containing peptides. We have synthesized DAP-containing PGN fragments, including the first chemical synthesis of tracheal cytotoxin (TCT). The ability of these fragments to stimulate human Nod1 as well as differences in Nod1 recognition for various synthesized ligand structures was elucidated. Cell-surface glycans such as N-glycans and O-glycans on glycoproteins and glycoconjugates work as signaling molecules for self-recognition and control immune system. Our new strategy using glycan-imaging in whole-body system is expected to unveil the dynamics of glycans in the body. Positron emission tomography (PET) is a noninvasive method that visualizes the locations and levels of radiotracer accumulation. We developed the facile labeling of peptides and proteins for PET imaging. The labeled glycoproteins and glycoclusters were then subjected to PET imaging in order to examine their in vivo dynamics, visualizing the differences in the circulatory residence of glycoproteins and glycoclusters in the presence or absence of sialic acid residues.

1. Overview Self and nonself recognition is a fundamental function to maintain the life of a multicellular organism with preventing the invasion of microorganisms, components from other species/individuals, or problematic materials. In order to distinguish the difference between self and nonself, glycan structures or glycoconjugates on the cell surface are often used as the signal motifs. One of significant roles in nonself recognition is to recognize microorganism including bacteria, which has recently been exhibited to have close relationships with the innate immunity activation. On the other hand, the cell/cell and cell/protein interactions mediated by the N-glycans, for example, stimulating the immunosuppressive signals through Siglec families, constitute the significant roles in self-recognition process. In this chapter, we show key chemical synthesis methods to build a compound library of bacterial glycoconjugates, and the immunostimulating functions.

Self and Non-Self Recognition with Glycans

325

In addition, the methods for visualizing the in vivo dynamics of N-glycans and glycoproteins, by means of the noninvasive molecular imaging as the new tool for investigating the oligosaccharides functions, will be described.

2. Bacterial Glycoconjugates for Nonself Recognition—Lipopolysaccharide (LPS) LPS of Gram-negative bacteria is one of major signaling motifs for the recognition of bacteria invading to the host organism. The compound is also known as endotoxin due to its potent immunostimulation and the toxicity. LPS is recognized with Toll-like receptor 4 (TLR4)/MD-2 complex on cell surface to produce mediators, for example, cytokines, prostagrandins, the platelet activating factor, oxygen-free radicals, and NO, which all activate and modulate the immune system (Kusumoto and Fukase, 2006; Raetz and Whitfield, 2002). LPS consists of a glycolipid component named lipid A, which is covalently connected to the polysaccharide part. Lipid As from various bacteria have common structural features, that is, GlcNAcb(1-6)GlcNAc possessing phosphono groups at the reducing end and 4-positon of the nonreducing glucosamine, and long-chain acyl groups at 2, 20 , 3, and 30 positions. The total synthesis of Escherichia coli lipid A 1 (synthetic 1 is termed 506) confirmed that lipid A is the chemical entity responsible for the innate immunostimulatory activity of LPS (Fig. 16.1) (Kusumoto and Fukase, 2006). Various lipid A structures are known such as compound 2, 3, 4, and 5 (Fig. 16.1) (Kusumoto et al., 1999, 2009; Takada and Kotani, 1992), and we have synthesized various natural or designed lipid A structures to investigate the structure–activity relationships and analyze the action mechanisms (Fujimoto et al., 2005). One of the characteristic groups of lipid As are the compounds from parasitic bacteria such as Helicobacter pylori and Porphyromonas gingivalis. H. pylori is one of Gram-negative bacteria and an etiological agent of gastroduodenal diseases, including chronic gastritis, gastroduodenal ulcers, and gastric cancer. H. pylori LPS shows a lower endotoxic activity compared to other enterobacterial preparations such as E. coli LPS, but it is considered to have relationships to the chronic inflammation. H. pylori LPS has characteristic lipid A structures; namely it has fewer but longer acyl groups, and does not have the 40 -phosphate group, and the glycosyl phosphate often has an ethanolamine group (8b; Fig. 16.1). H. pylori LPS has one Kdo as a linkage between lipid A and the polysaccharide part. We have synthesized triacyl type H. pylori lipid A with or without ethanolamine (8a, 8b) (Fujimoto et al., 2007; Sakai et al., 2000), and also H. pylori lipid A connecting to one Kdo residue (9) (Fujimoto et al., 2007), in order to elucidate their biological activities.

326


OH O

O (HO)2P O O O O

O HO O

NH O O O

O

O O NH O P(OH)2 OH O OH

(C14)

(C14) (C14) (C14) (C14) (C12)

O

O O NH O P(OH) 2 OH O OH

O

O NH O OH O

O O NHO P(OH) 2 OH O O

O

(C10)

(C14)

NH O O OH O

Antagonistic O

O O NH O P(OH) 2 OH O O

O

(C10)

(C14) 5 : E5564 OH

O

HO O

NH O OH O

(C10)

(C12)

OH

O O

O

O HNO P OH 1 OR O

HO HO

OH

O HN O

O HO HO

O

(C18) (C18)

O O HNO P OH OH O OH

(C18) (C18) (C18)

Figure 16.1

(C12)

Helicobacter pylori lipid A and Kdo-lipidA

R1 = –H, R2 = –H: 8a antagonistic R2 = –H: 8b immunostimulative

(C10)

CO2H

O

O

R1 = –CH2CH2–NH2,

O

OH Kdo

HO O

O

O

CO2H

7 : Ru. gelatinosus CM-analog immunostimulative (Limulus activity +)

HO

HN

O

(C12)

(C12)

O HO R2O

O NH O OH O

(C10)

(C10)

6 : Ru. gelatinosus lipid A antagonistic (Limulus activity ++)

O

O

HO O

O

(C10)

OH

O

O

(HO)2P O O

O

(C10) (C10)

HO HO

O O NHO P(OH) 2 O O

(C18)

O

O

O HO O

(C10)

(C14) 4 : RSLA Antagonistic OH

(C16)

3

(C10)

(C14)

O

O O O NH O P(OH)2 OH O O O

Salmonella typhimurium lipid A immunostimulative

NH OMe O

(C10)

NH O O O

(C14) (C14) (C14) (C12) (C14)

OMe O

O (HO)2P O O

O HO O

O HO O

(C14) (C14)

Escherichia coli biosynthetic precursor lipid IVa antagonistic

OH O

(HO)2P O O

O O

2

1 Escherichia coli lipid A strongly immunostimulative

O

(HO)2P O O

(C14)

(C14)

OH O

O

O HO O

O NH O OH OH

(C14)

O (HO)2P O O

OH O

O (HO)2P O O

O

(C18)

9 Antagonistic

Structures of various lipid A and LPS partial structures.

327


BnO TBSO R5O BnO R3 O

O O

NH 2

R

O BnO R4O

BnO TBSO

O O

NH OAllyl

+

R1

10 (R3 = Bn, R4 = Bn, R5 = TES)

F

CO2Bn

TBSO CO2Bn

TBSO 11

O

OBn O

OBn

O BnO R3O

O O

NH R2

O BnO R4O 12

O O

NH OAllyl 1

R

BnO H. pylori Kdo-lipid A 9

R1 = CH3(CH2)14 O R2 = CH3(CH2)16 O CH3(CH2)14

Figure 16.2 Synthesis of Helicobacter pylori Kdo–lipid A backbone.

3. Synthesis of H. pylori Kdo–Lipid A Backbone For the synthesis of Kdo–lipid A, the lipid A backbone (10) was first prepared, and connected with Kdo (11) with an a-selective glycosylation (Fig. 16.2) (Fujimoto et al., 2007; Yoshizaki et al., 2001). The molecular sieve 5A (MS5A) was found to be more effective for this reaction than MS4A, presumably because MS5A contains calcium, which traps fluoride anions and promotes glycosylation. The stereochemistry at the anomeric position was determined with the chemical shift of the protons at the 30 -position from 1H NMR in comparison with the previously reported data (Fujimoto et al., 2007; Yoshizaki et al., 2001).

4. Glycosylation with Kdo Donor 11 1. To a mixture of 10 (48.8 mg, 0.0275 mmol), Kdo-fluoride 11 (52.7 mg, 0.0713 mmol) and MS5A˚ in dry CHCl3 (2 mL) is added BF3OEt2 (0.052 ml, 0.288 mmol) at –20 C and the mixture is stirred at –20 C for 1.5 h. 2. After addition of saturated NaHCO3, the mixture is extracted with CHCl3. The organic layer is washed with saturated NaHCO3 and brine, dried over anhydrous Na2SO4, filtered, and concentrated under reduced pressure. 3. The residue is purified by column chromatography (silica gel 50 g, CHCl3/acetone ¼ 30/1) to give 12 as a white solid (55.3 mg, 85%). Rf (CHCl3/acetone ¼ 30/1) ¼ 0.25; ESI-MS (positive) m/z 2402.51 [(MþNa)þ]; 1H NMR (600 MHz, CDCl3) d ¼ 7.35–7.20 (m, 40H, PhCH2 8), 6.24 (d, J ¼ 9.0 Hz, 1H, NH), 5.96 (d, J ¼ 7.2 Hz, 1H, NH0 ), 5.68 (m, 1H, OCH2CH¼CH2), 5.17 (d, J ¼ 10.8 Hz, 1H,

328


PhCH2OCO), 5.15–5.13 (m, 1H, OCH2CH¼CH2), 5.12–5.07 (m, 1H, OCH2CH¼CH2), 5.04 (d, J ¼ 10.8 Hz, 1H, PhCH2OCO), 5.01 (d, J ¼ 6.0 Hz, 1H, b-CH of acyloxyacyl), 4.97 (d, J ¼ 7.8 Hz, 1H, H10 ), 4.77 (d, J ¼ 10.8 Hz, 1H, PhCH2), 4.74–4.72 (m, 1H, PhCH2), 4.70 (d, J ¼ 3.7 Hz, 1H, H1), 4.69–4.67 (m, 1H, PhCH2), 4.63 (d, J ¼ 10.8 Hz, 1H, PhCH2), 4.61 (d, J ¼ 10.8 Hz, 1H, PhCH2), 4.57 (d, J ¼ 11.4 Hz, 1H, PhCH2), 4.55 (d, J ¼ 11.4 Hz, 2H, PhCH2), 4.53 (d, J ¼ 12.0 Hz, 1H, PhCH2), 4.52 (d, J ¼ 10.8 Hz, 1H, PhCH2), 4.50 (d, J ¼ 11.4 Hz, 1H, PhCH2), 4.48 (d, J ¼ 10.8 Hz, 1H, PhCH2), 4.46 (d, J ¼ 11.4 Hz, 1H, PhCH2), 4.39 (d, J ¼ 12.0 Hz, 1H, PhCH2), 4.31–4.28 (m, 1H, H2), 4.24 (t, J ¼ 8.5 Hz, 1H, H30 ), 4.13–4.07 (m, 4H, H400 , H500 , H600 , OCH2CH¼CH2), 3.99–3.94 (m, 2H, H4, H700 ), 3.89–3.87 (m, 2H, H60 , H800 ), 3.83 (d, J ¼ 9.0 Hz, 1H, H5), 3.80–3.77 (m, 1H, b-CH of acyl), 3.73–3.63 (m, 6H, H3, H6 2, H60 , H800 , OCH2-CH¼CH2), 3.59–3.56 (m, 1H, H50 ), 3.39 (t, J ¼ 8.5 Hz, 1H, H40 ), 3.25 (q, J ¼ 9.0 Hz, 1H, H20 ), 2.34 (dd, J ¼ 15.6, 7.2 Hz, 2H, a-CH2 of acyloxyacyl, a-CH2 of acyl), 2.25 (dd, J ¼ 15.0, 7.8 Hz, 1H, a-CH2 of acyl), 2.18 (dd, J ¼ 15.6, 4.8 Hz, 1H, a-CH2 of acyloxyacyl), 2.13–2.07 (m, 3H, OCOCH2 of acyloxyacyl 2, H300 ), 1.96 (dd, J ¼ 12.0, 3.6 Hz, 1H, H300 ), 1.58–1.53 (m, 1H, g-CH2 of acyl), 1.48–1.45 (m, 2H, g-CH2 of acyl, OCOCH2CH2 of acyloxyacyl), 1.38 (q, J ¼ 6.6 Hz, 1H, g-CH2 of acyloxyacyl), 1.31–1.04 (m, 82 H, CH2 of acyl 82), 0.92–0.82 (m, 27H, CH3 of acyl 9, t-BuSi 2), 0.11–0.01 (m, 12H, CH3Si 12). For obtaining the final compound, the anomeric position of trisaccharide was changed to the phosphate, and all the protecting groups were cleaved by hydrogenation (Fujimoto et al., 2007).

5. Cytokine (IL-6) Induction in Human Peripheral Whole-Blood Cell Cultures 1. The synthetic samples (lipid A 8a and Kdo–lipid A 9 with noted concentrations in 25 mL of saline) and heparinized human peripheral whole-blood (HWBC) (25 mL) collected from an adult volunteer in RPM1 1640 medium (75 mL; Flow Laboratories, Irvine, Scotland, UK) are incubated in triplicate in a 96-well plastic plate at 37 C in 5% CO2. An LPS specimen prepared by Westphal method from E. coli O111:B4 (Sigma Chemicals Co.) is used as a positive control at the concentration of 0.5 ng/mL. 2. After 24 h of the incubation, the plate is centrifuged at 300g for 2 min and cytokines in the supernatant are assayed (Suda et al., 1995). 3. The levels of IL-6 induced by stimulating HWBC cultures with test samples are measured by means of an enzyme-linked immunosorbent assay (ELISA) using Human IL-6 ELISA kit (eBioscience, San Diego, CA, USA).

329


1.4

Fold (vs. LPS)

1.2

50 ng / ml 500 ng / ml

1

5000 ng / ml 0.8 0.6 0.4 0.2 0

9

9 LPS (E.coli ) + LPS (E.coli )

8a

8a LPS (E.coli ) + LPS (E.coli )

Figure 16.3 IL-6 induction and inhibition by Helicobacter pylori Kdo–lipid A 9 and lipid A 8a with heparinized human peripheral whole-blood (HWBC). E. coli (O111:B4) LPS is used as a positive control at the concentration of 0.5 ng/mL.

Data represent averages of three repeated assays with standard deviations from individual experiments. H. pylori Kdo–lipid A 9 and lipid A 8a inhibited IL-6 induction by E. coli LPS. Kdo–lipid A 9 showed more potent antagonistic activity than 8a (Fig. 16.3) (Fujimoto et al., 2007). On the other hand, Lipid A 8b having ethanolamine induced low levels of cytokines such as IL-18 and TNF-a via TLR4/MD-2 complex (Ogawa et al., 2003). These results demonstrated that the number of anionic charges influences the biological activity of lipid A and LPS. Similar charge effects to biological activities were also observed in our studies of Ru. gelatinosus lipid A and lipid A analogs containing acidic amino acid residues (Fujimoto et al., 2005). Immunostimulating or antagonistic activity can be controlled by the anionic charges in these analogs and H. pylori lipid A. The present study also explains why some strains of H. pylori LPS show weak immunostimulating activity, while others show antagonistic activity. Weak immunostimulating activity may be correlated with the ability of H. pylori to induce chronic inflammation, whereas the antagonistic activity should be important for suppressing the innate immune response and survival as a parasite.

6. Bacterial Glycoconjugates for Nonself Recognition—Peptidoglycan (PGN) PGN is a component of bacterial cell walls, and has conserved structural characteristics, which makes PGN fragment structures to be good motifs for nonself recognition. PGN has polysaccharide chains linked to

330


MurNAc

GlcNAc

O

CH3

HO

HO

HO

O

O

O

CH3

O

O

O

NHAc

NHAc D C L-Ala-D-Glu-(NH2)n g O

HO

O

O

O HO

O

HO NHAc

D C O

NHAc

L-Ala-D-Glu-(NH2)n g AA-D-Ala-(peptide)

AA-D-Ala-(peptide) D-Ala HO O

HO

O

O O

HO NHAc CH3

HO O

O

HO NHAc

D C O

O O

AA-D-Ala-(peptide)

L CH CO2H CH2

NHAc

L-Ala-D-Glu-(NH2)n g

D-Ala

H2N

CH2 CH2 H2N

CH CO2H D meso-Diaminopimelic acid (meso-DAP)

Lys-type peptidoglycan of Staphylococcus aureus: n = 1, AA: L-Lys, peptide: (L-Gly)5 DAP-type peptidoglycan of Escherichia coli: n = 0 or 1, AA: meso-DAP, peptide: D-Ala

Figure 16.4 Schematic structures of bacterial cell wall peptidoglycan; Lys-type peptidoglycan of Staphylococcus aureus, and DAP-type peptidoglycan of Escherichia coli.

peptides, and the polysaccharide is composed of alternating b(1!4) linked N-acetylglucosamine (GlcNAc) and N-acetylmuramic acid (MurNAc) glycan. At the branched position of the peptide, there is a diaminocarboxylic acid such as L-Lys (in many of Gram-positive bacteria) or mesodiaminopimelic acid (meso-DAP, in Gram-negative bacteria and some Gram-positive bacteria) as shown in Fig. 16.4. The structure of the dipeptide (L-Ala-D-Glu) linked MurNAc (MDP) is highly conserved in bacterial species (Schleifer and Kandler, 1972). PGN has been known as a potent immunostimulator and an immune adjuvant, but the mechanism of the immune stimulation had not been unveiled until the recent findings of the receptors. In 2003, two groups independently found that the intracellular protein Nod1, which is the founding member of the NLR protein family, is a receptor of PGN fragments (Chamaillard et al., 2003; Girardin et al., 2003a). Philpott and coworkers reported that Nod1 senses DAP containing muropeptides, such as GlcNAc-MurNAc-L-Ala-g-D-Glu-meso-DAP (Girardin et al., 2003a), whereas Inohara and coworkers found a DAP-containing smaller peptide, iE-DAP (g-D-glutamyl diaminopimelic acid), activates Nod1 using PGN synthetic peptide fragments (Chamaillard et al., 2003). It was also shown

331


HO O HO OH HO AcNH CH CHCO L-Ala-D-isoGln 3 MDP

HO HO O

HO HO O

O OH AcNH CH CHCO peptide 3

HO HO HO

HO HO HO

HO HO HO

O HO O AcNH O

O HO O AcNH HO CH CHCO peptide 3

O HO O AcNH O

O HO O AcNH HO

O HO O AcNH O

HO O O O m AcNH OR' NHAc CH 3 C L-Ala-D-isoGln L-Lys(R)-D-ALa O Peptides: L-Ala-D-isoGln HO HO L-Ala-D-isoGln-L-Lys O O HO L-Ala-D-isoGln-L-Lys(Ac) O O HO L-Ala-D-isoGln-L-Lys-D-Ala m AcNH NHAc OR' L-Ala-D-isoGln-L-Lys(Ac)-D-Ala CH3 C L-Ala-D-isoGln L-Lys L-Ala-D-isoGln-L-Lys-D-Ala-D-Ala O L-Ala-D-isoGln-L-Lys(Ac)-D-Ala-D-Ala

O HO O AcNH O

O OR AcNH CH3CHCO peptide

HO HO HO

O

OR AcNH CH CHCO peptide 3

O HO O AcNH O

O HO O HO O O HO O HO O HO O O AcNH AcNH O O CH3CHCO-L-Ala-D-isoGln AcNHHO OR CH3CHCO-L-Ala-D-isoGln AcNH O AcNH CH3CHCO-L-Ala-D-isoGln CH3CHCO-L-Ala-D-isoGln HO O O O Repeating unit of DAP-type PGN OH n O (n = 1, R = D-Ala): DS-4PDAP NHAc NHAc CH L-Ala-D-Glu H3C HO HN CH CO R (n = 1, R = OH) : DS-3PDAP O (CH2)3 (n = 0, R = D-Ala): MS-4PDAP HO HO H2N CH CO H (n = 0, R = OH) : MS-3PDAP 2 DAP

O

Linked structures

O H3C O O O AcNH n

O

L-Ala-D-Glu O HN CH CO R Tracheal cytotoxin (TCT) (CH2)3 (n = 1, R = D-Ala): DS(anh)-4PDAP H2N CH CO H (n = 1, R = OH) : DS(anh)-3PDAP 2 NHAc DAP (n = 0, R = D-Ala) : MS(anh)-4P DAP (n = 0, R = OH) : MS(anh)-3PDAP

Figure 16.5 Chemically synthesized PGN fragment library (Inamura et al., 2001, 2006; Kawasaki et al., 2008; Kusumoto et al., 2009).

that the MDP is a ligand of Nod2, which is another NLR family of intracellular proteins (Girardin et al., 2003b; Inohara et al., 2003). We constructed the PGN fragments library (Fig. 16.5), which included Lys-type linear and linked fragments, and also DAP-type fragments such as a repeating unit of the PGN fragment (GlcNac(b1-4)MurNAc-L-Ala-g-DGlu-meso-DAP-D-Ala), tracheal cytotoxin (TCT; GlcNac(b1-4)MurNAc (anh)-L-Ala-g-D-Glu-meso-DAP-D-Ala), and their fragments (Kawasaki et al., 2008). The synthesis of TCT is shown in Fig. 16.6, and the one of the key reactions is the b-selective glycosylation as used to prepare 15 from glycosyl donor 13 and glycosyl acceptor 14.

7. Synthesis of Disaccharide Moiety 15 in Tracheal Cytotoxin with b-Selective Glycosylation 1. TMSOTf (5 ml, 0.030 mmol) is added to the solution of glucosamine imidate 13 (200 mg, 0.296 mmol), 14 (193.6 mg, 0.443 mmol), and MS4A˚ in dry CH2Cl2 (3 mL) at 17 C, and the mixture is stirred for 30 min under Ar atmosphere. 2. The reaction is quenched with saturated aqueous NaHCO3. 3. The organic layer is washed with brine, dried over Na2SO4 and concentrated in vacuo. 4. The residue is purified by silica-gel flash column chromatography (toluene/AcOEt ¼ 7/1) to give 15 (182 mg, 65%).

332


(R) CH3 CH COOEt O O O OH

13

NHTroc

Ph

O

O

O BnO

O CCl3 NHTroc

14

Ph

WSCD HCl, HOBt, Et3N, DMF

HO HO HO

O

O

O

81%

NHTroc

NHTroc

15

(R) CH3 CH CO L-Ala-D-Glu-OBn L (S) O O HN CH COR1 O (CH2)3 O

NHAc

ZHN CH COOBn D (R)

1 18 92%: R = D-Ala-OBn HCl H-L-Ala-D-Glu-OBn 19 85%: R1= OBn L H2N CH COOBn (CH2)3 L-Ala-D-Glu-OH

O NHAc

O

O BnO

O O BnO

Zn-Cu AcOH / THF / Ac2O = 1/1/1

O

O

NHAc

(R) CH3 CH CO O O O

Pd(OH)2, H2 (20 kg / cm2) THF

Ph

TMSOTf, MS4A, CH2Cl2, –15°C, 65%

Tripeptide 16 LiOH or THF / 1, 4-dioxane / tetrapeptide 17 H2O = 4 / 2 / 1 Quant.

(R) CH3 CH COOEt O O

NH

NHAc

L (S)

2 HN CH COR (CH2)3

ZHN CH COOBn D

16 HCl H-L-Ala-D-Glu-OBn

H2N CH COOH D (R)

2 20 Quant. (from 18), R = D-Ala; Tracheal cytotoxin (TCT) 21 Quant. (from 19), R2 = H

L

HN CH CO D-Ala-OBn

(CH2)3 ZHN CH COOBn D

17

Figure 16.6 Synthesis of Tracheal cytotoxin (TCT) and its fragments. Abbreviations: Bn, benzyl; Tf, trifluoromethanesulfonyl; Troc, 2,2,2-trichloroethoxycarbonyl; TMS, trimethylsilyl; WSCDHCl, 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide hydrochloride; HOBt, 1-hydroxybenzotriazole; Z, benzyloxycarbonyl.

ESI-TOF-MS (positive) m/z 949.17 [MþH]þ; 1H NMR (500 MHz, CDCl3) d ¼ 7.52–7.18 (10H, m, ArH 2), 5.60 (1H, s, Ph-CH¼ O2), 5.33 (1H, s, Hanh-1), 5.03 (1H, d, J ¼ 7.7 Hz, NH), 4.91 (1H, d, J ¼ 12 Hz, -O-CH2-Ph), 4.82 (1H, d, J ¼ 8 Hz, H-1), 4.76 (2H, dd, J ¼ 12 Hz, 20 Hz, -CH2-CCl3), 4.7 (1H, d, J ¼ 12 Hz, -O-CH2-Ph), 4.60 (1H, d, J ¼ 12 Hz, NH), 4.49 (1H, d, J ¼ 5.7 Hz, Hanh-5), 4.33 (1H, dd, J ¼ 5 Hz, 10 Hz, H-6), 4.22 (1H, q, J ¼ 3.6 Hz, Lac-aH), 4.20–4.14 (3H, m, -CH2CH3, Hanh-60 ), 3.95 (1H, d, J ¼ 9.8 Hz, Hanh-2), 3.86– 3.76 (3H, m, H-3, H-6, H-4), 3.74–3.72 (2H, m, Hanh-60 , Hanh-4), 3.58 (1H, brs, Hanh-3), 3.52–3.42 (2H, m, H-2, H-5). Found: C, 47.02; H, 4.42; N, 2.96%. Calcd for C37H42Cl6N2O14: C, 46.71; H, 4.45; N, 2.94%.

8. Synthesis of Tracheal Cytotoxin 20 and Its Fragment 21 After preparation of the disaccharide 15, 2,2,2-trichloroethoxycarbonyl (Troc) groups were changed to acetyl groups via cleavage of Troc group using Zn–Cu in the presence of acetic anhydride. The ethyl ester was cleaved


333

with LiOH, and then the liberated carboxyl group was connected to the DAP-containing peptides 16 or 17 by using 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide (WSCD; water-soluble carbodiimide), 1-hydroxybenzotriazole (HOBt), and triethylamine (Et3N) in DMF. After the coupling, all benzyl and benzyloxycarbonyl groups of 18 and 19 were removed by catalytic hydrogenation with Pd(OH)2 and H2 to give 20 (TCT) and 21 (Kawasaki et al., 2008).

9. Immunostimulatory Activities of DAP Containing PGN Fragments The human Nod1 stimulating activity of the DAP-type synthetic PGN fragments have been evaluated by HEK293T bioassay expressed human Nod1 as previously described (Chamaillard et al., 2003). In these compounds, 20 (TCT) shows only very weak human-Nod1 stimulatory activity, whereas 21 (DS(anh)-3PDAP) shows approximately 10-fold higher activity than that of known ligand A-iE-DAP (Kawasaki et al., 2008). These results are consistent with a report using TCT from a natural source (Magalhaes et al., 2005), and demonstrate that a free carboxyl group at the 2-position of DAP is favorable for the human Nod1 recognition. In case of human Nod1, TCT is a only weak stimulant, but it plays a fundamental role in innate immune systems of other species such as Drosophila with the activation of PGRP-LC (Kaneko et al., 2004; Stenbak et al., 2004). It has also been reported that recognition of DAP-type PGN by PGRP-LE in Drosophila is crucial for the induction of autophagy, which prevents intracellular growth of Listeria monocytogenes and promotes host survival after an infection (Yano et al., 2008). The structurally defined synthesized PGN fragments have been fundamental to understand the activation mechanism of innate immune system.

10. Visualizing the In Vivo Dynamics of Animal N-Glycans Among the various types of oligosaccharide structures, asparagine-linked oligosaccharides (N-glycans) are the most prominent in terms of diversity and complexity. In particular, N-glycans containing sialic acid residues are involved in a variety of important physiological events, including cell–cell recognition, adhesion, signal transduction, and quality control (Kamerling et al., 2007). Moreover, it has long been known that the sialic acids in N-glycans on soluble proteins or peptides enhance circulatory residence (Morell et al., 1968), that is, N-glycan-engineered erythropoietin (EPO) (Elliott et al., 2003) or insulin

334


(Sato et al., 2004) exhibits a remarkably higher stability in serum, which effects the prolonged bioactivity. Antibody-dependent cellular cytotoxicity (ADCC) and/or complement-dependent cytotoxicity (CDC) has also been proposed to be modulated by the sialic acids of N-glycans in immunoglobulin (IgG) through Siglec interactions by glycosylating or removing the sialic acids (Kaneko et al., 2006). However, these important findings and previous efforts in investigating N-glycan functions have been mostly based on in vitro experiments using isolated lectins, cultured cells, and dissected tissues. Recently, interest has shifted to the dynamics of these glycoproteins and/or glycans in vivo, that is, how the function and/or interaction of the individual N-glycan works synergistically through dynamic processes in the body to eventually exhibit biological phenomena. Molecular imaging (Tanaka and Fukase, 2008) is the most promising tool to visualize the ‘‘on-time’’ N-glycan dynamics in vivo. Although fluorescence imaging is the method of choice due to the convenient experimental and detection procedures at the small animal levels, magnetic resonance (MR), and more preferably, the positron emission tomography (PET) imaging, which have technologically improved sensitivity and resolution, are well suited for diagnostic applications. Nevertheless, molecular imaging of glycans has not been thoroughly examined, except for the 18F-FDG tracer (technically a monosaccharide) and the very limited examples of liposome-conjugated oligosaccharides (Chen et al., 2008; Hirai et al., 2007). This is due to the lack of the efficient labeling methods of glycoproteins, and the bioactivity of the oligosaccharides is affected by the multivalency and/or heterogeneous environment, that is, on cell surfaces that are composed of oligosaccharide clusters along with other biomolecules. A single molecule of the N-glycan, either obtained from a natural or synthetic source, is readily excreted from the body (Vyas et al., 2001). Thus, efficiently labeling and mimicking such a N-glycan-involved bioenvironment, for example, by conjugating the N-glycans, to the liposomes, or to the clusters, may provide information on the ‘‘in vivo dynamics’’ of N-glycans. Below we discuss the methods for microPET imaging of (1) glycoproteins and (2) dendrimertype glycoclusters by making much use of the multivalency effects of 16 molecules of N-glycans.

11. PET Imaging of Glycoproteins In order to investigate the effects of N-glycans, especially the sialic acid residue at the nonreducing end of the glycans, on the metabolic stability of the proteins, a microPET of glycoproteins, orosomucoid, and asialoorosomucoid could be investigated (Tanaka et al., 2008). Based on the recently developed ‘‘azaelectrocyclization protocol’’ (Fig. 16.7A), glycoproteins available in only small amounts (62 mg of orosomucoid and

335


A

HO2C

N-glycan chain NH2

N N

HO2C

NH2

CO2Et O

H N

N N

NH2

Orosomucoid

CO2H

CHO

N H

O

2 Molecules of DOTA

+

22 (STELLA kit) NH2

1) 24 °C, 30 min 0.1 M phosphate buffer (pH 7.4)

NH2 NH2

Asialoorosomucoid –6 62 mg, 4.5×10 M

HO

2)

OH OH

:

CO2

O

AcHN

68

O

HO

Sialic acid

GaCl3

3 Molecules of DOTA

:

HO2C

OH OH O O

OH

:

O

HO C 2

Galactose

B

CO H 2 N N O 68Ga3+ NH N N HN O N EtO C 2

0 –10 min

20–30 min

60 –80 min

180 –240 min

Liver

Lung Liver Gall bladder Spleen Kidney

Figure 16.7 (A) Labeling of glycoproteins by DOTA by STELLAþ kit (B) Dynamic microPET images of [68Ga]DOTA-glycoproteins in rabbits. Time course of accumulation of [68Ga]DOTA-orosomucoid (upper) and [68Ga]DOTA-asialoorosomucoid (lower) in some peripheral organs (axial views). These PET images were fused to anatomical images obtained by using CT.

asialoorosomucoid) were labeled with the incorporation of 2–3 units of DOTA (1,4,7,10-tetraazacyclodecane-1,4,7,10-tetraacetic acid) by incubating the respective protein with aldehyde probe 22 (STELLAþ, a kit available from Kishida, Co., Ltd.) for 30 min (Tanaka et al., 2010). The DOTA-labeled glycoproteins were subsequently radiometallated with 68Ga and their in vivo kinetics were analyzed in rabbit by means of microPET.

12. Method for 68Ga-DOTA Labeling and MicroPET Imaging in Rabbit 1. PBS solutions of orosomucoid and asialoorosomucoid (62 mg, 1.4 nmol, 295 mL, pH ¼ 7.4) are reacted with a DMF solution of probe 22 (STELLAþ kit, 14 nmol, 1.5 mL) at room temperature for 30 min

336

2. 3. 4. 5.

6. 7.

8.

9.


(reaction concentrations: 4.5 10 6 M for orosomucoid, 4.5 10 5 M for 22). DOTA-labeled proteins are purified by centrifugal filtration using MicroconÒ (Millipore, 30,000 cut). DOTA-proteins prepared above (10 mg) are incubated with 68GaCl3 solution (pH ¼ 7.0, 1.68 mCi, 500 mL) obtained from 68Ge/68Ga radionuclide generator, at 40 C for 10 min. A solution of DOTA (1.0 mmol, 10 mM in H2O) is added in order to chelate and excrete the excess 68Ga from the body during the PET study. 68 Ga-DOTA-glycoproteins at a dose of 15.8–16.1 MBq in 2.2 mL are injected via an ear vein of female Japanese white rabbits weighing 2.1–2.2 kg at 13 weeks of age (Japan SLC, Inc., Hamamatsu, Japan) under a general anesthesia with ketamine (60 mg/kg, KetalarÒ, Sankyo, Tokyo, Japan) and xylazine (6 mg/kg, SelactarÒ, Bayer Yakuhin, Tokyo, Japan). During the imaging experiments, the rabbits are sedated continuously with intravenously administered a mixture of ketamine (60 mg/kg/h) and xylazine (6 mg/kg/hr). PET images are obtained by using a small animal PET scanner, the microPET P4 system (Siemens Medical Solutions Inc., Knoxville, TN, USA), and the emission data is collected for 240 min postinjection as 12 frames (6 400 s, 3 1000 s, 2 1600 s, and 1 3400 s), and is acquired with an energy window of 400–650 keV and a coincidence timing window of 6 ns. The images are reconstructed from 120 to 240 min after injection of 68Ga-DOTA-orosomucoid or asialoorosomucoid by an ordered subset expectation maximization (OSEM) algorithm with attenuation correction using CT data or no scatter correction, and are smoothed by using a Gaussian kernel with an FWHM of 3 mm in the all directions. Quantitative analysis is performed using ASIPro VM version 6.3.3.0 software (Siemens Medical Solutions, Inc., Knoxville, TN, USA). Regions of interest (ROIs) are placed on the tissue region.

MicroPET images of 68Ga-DOTA-orosomucoid and asialoorosomucoid in Fig. 16.7B detected the asialo-glycoprotein being cleared throughkidney faster than orosomucoid through the well-known asialoglycoprotein receptor (Morell et al., 1968), thus achieving the visualization of sialic acid-dependent circulatory residence of glycoproteins. PET images also detected another clearance pathway of asialo-glycoprotein through the gallbladder, that is, intestinal excretion pathway, as well the accumulation to the lung and spleen. These promising PET images of glycoproteins suggest future uses for the glycoproteins in pharmacological and/or clinical applications.

337


13. PET Imaging of Glycoclusters In order to image the ‘‘in vivo dynamics’’ of glycans, it is necessary to mimic the ‘‘glycan-cluster environment,’’ as well as to make advantage of the glycan multivalency effects for the stronger interaction with lectins. The polylysine-based dendrimer-type glycoclusters 23a-c with 16 molecules of glycans (Fig. 16.8) were developed as the excellent templates for investigating the N-glycan dynamics in vivo (Tanaka et al., unpublished results); thus, the dendrimer core 23 with the N-benzyl histidine and the terminal acetylene embodied in the propargyl glycine residue (Fig. 16.8A) could be smoothly reacted with the 16 molecules of the azide-containing N-glycans with large and complex structures, that is, many hydroxyls and molecular weight of ca. 1500, based on the ‘‘self-activating’’ Huisgen 1,3-dipolar

N im-Bn-His

A

1)

Acetylene

N3 20 eq CuSO4 (16 eq), Na ascorbate (48 eq), DIPEA (16 eq), DMF, Ar, 40 min then, DOTA (32 eq), 40 min O

2)

OEt

O

Y

N H

23

N H

H

O

23a

H2O, rt, 30 min +

(STELLA kit)

Y=

O

O

a

NeuNAca2,6Galb1,4GlcNAcb1,2Mana1,6

HO O

Manb1,4GlcNAcb1,4GlcNAcbAsp-linker NeuNAca2,6Galb1,4GlcNAcb1,2Mana1,3

B

SO2H

O

30 min

1h

2h

O 68 Ga-DOTA

H

O

O O

Cy5

GB

K B

B Galb1,4GlcNAcb1,2Mana1,6

Manb1,4GlcNAcb1,4GlcNAcbAsp-linker Galb1,4GlcNAcb1,2Mana1,3

30 min

1h

2h

4h

L

NeuNAca2,3Galb1,4GlcNAcb1,2Mana1,6

c

Manb1,4GlcNAcb1,4GlcNAcbAsp-linker NeuNAca2,3Galb1,4GlcNAcb1,2Mana1,3

30 min

1h

2h

4h

L H

H K B

O N

O

4h L

b

N

N N 68Ga N N

K B

B

Figure 16.8 (A) Preparation of glycoclusters. (B) Dynamic PET imaging of glycoclusters 23a-c in normal BALB/c nude mice. 68Ga-DOTA-Labeled glycoclusters (10 MBq) were administered from the tail vein of the mice (n ¼ 3, 500 pmol, 100 mL/mouse) and the whole body was scanned by a small animal PET scanner, microPET Focus 220 (Siemens Medical Solutions, Inc., Knoxville, TN, USA), over 0–4 h after injection; H, heart; K, kidney; L, liver; B, urinary bladder; GB, gallbladder. (a) glycocluster 23a, (b) glycocluster 23b, (c) glycocluster 23c.

338


cycloaddition.’’ The clusters were designed to have a terminal lysine e-amino group so that they could be efficiently labeled by 68Ga-DOTA as the PET radiolabel, and if required, fluorescent groups, in the presence of numerous hydroxyls by labeling kit ‘‘STELLAþ’’ under mild conditions (Tanaka et al., unpublished results).

14. Method for Preparation of N-Glycan Clusters and PET Imaging in Mouse 1. Acetylene-containing polylysine 23 (16-mer, 158 mg, 2.0 10 5 mmol) is reacted with azide-containing N-glycans (1.0 mg, 4.0 10 4 mmol) in DMF (50 mL) and H2O (50 mL) at room temperature in the presence of CuSO4 (64 mg, 3.2 10 4 mmol), sodium L-ascorbate (238 mg, 1.2 10 3 mmol), and diisopropylethylamine (74 nL). 2. Excess copper ion is removed by chelating with DOTA (647 mg, 1.56 10 3 mmol) for 40 min at room temperature, and low-molecular weight compounds are filtered off using the centrifugal filtration by MicroconÒ (10,000 cut, Millipore). 3. Lyophilization of the aqueous solution and the purification by reversephase HPLC provide the desired glycoclusters. 4. Labeling by DOTA and 68Ga, and MicroPET imaging were performed as described above, except using BALB/c mice for imaging (see Fig. 16.8 Caption). Figure 16.8B shows the microPET in mice of the N-glycoclusters (Y ¼ 68Ga-DOTA) with the glycan structure of bis-Neua(2-6)Gal glycan 23a, asialo-glycan 23b, and bis-Neua(2-3)Gal glycan 23c. 68Ga-radioactivity derived from 23a was retained after 4 h in the liver (Fig. 16.8B (a)), and was excreted slowly from the kidney/urinary bladder and from the gallbladder (intestinal excretion pathway). On the other hand, asialoglycan cluster 23b rapidly cleared through the kidney to the bladder (Fig. 16.8B(b)), although some accumulation was observed in the liver because the asialoglycoproten receptors are highly expressed in this organ (Morell et al., 1968). The results are consistent with the PET analyses of glycoproteins discussed above (Tanaka et al., 2008), where the asialocongener is more rapidly excreted than orosomucoid through the kidney. However, the a-linking to the 3-OH of galactose in glycocluster 23c, which also contains sialic acid, was readily excreted through the kidney/ urinary bladder as shown in Fig. 16.8B(c). These PET results on the 16mer glycoclusters 23a-c suggest that the specific sialoside linkage to galactose, that is, Neua(2-6)Gal linkage, in N-glycan structures plays an important role in the circulatory residence of N-glycans, which in turn


339

results in uptake of 23a in the liver. In addition, this specific sialoside linkage markedly differentiates the excretion mechanism from those of the asialo- and Neua(2-3)Gal cases, which proceed via a biofiltration pathway through the kidney. The notable difference in the serum stability due to the sialoside bond linkages to the galactose, that is, the a(2-6)- versus a(2-3)-linkages, is an intriguing observation. These dynamic PET images suggest a new receptor-mediated excretion mechanism for Neua(2-3)Gal-containing glycans. Namely, Neua(2-3)Gal-cluster 23c, which usually cannot be found in serum, is probably recognized as an invader and smoothly excreted by the vascular endothelial cells, erythrocytes, leucocytes, and via the phagocytosis by a macrophage; the smaller sized degradation products may be filtered in the kidney. Alternatively, the ‘‘excretion-escaping’’ mechanism by stimulating the immunosuppressive signals through the ITIM (immunoreceptor tyrosine-based inhibitory motif) molecules via Siglec families (Varki and Angata, 2006), may account for the higher stability of Neua(26)Gal-glycan. It is reported that the Neua(2-6)Gal-containing BSA reduces but does not prevent binding to the asialoglycoprotein receptor, while the Neua(2-3)Gal-congener abolishes the binding (Park et al., 2005). Therefore, the prolonged half-life coupled to uptake by the asialoglycoprotein receptor account for the high accumulation of 23a in the liver (Fig. 16.8B(a)). Biantennary Neua(2-6)Gal-chains especially reduce binding in comparison with tri- and tetraantennary glycans (Lee et al., 1983), nevertheless, the combination of high valency and long circulatory half life (reduced clearance) likely leads to the uptake of 23a by hepatocytes via the Gal/GalNAc lectin receptor. Note that the slow clearance of bis-Neua(2-6)Gal cluster 23a through the gallbladder may be due to the ‘‘polar transport mechanism’’ (Nakagawa et al., 2006), which ‘‘tags’’ the fucose to specific N-glycans in the liver. Elucidation of a detailed mechanism will be the subject of future investigations.

ACKNOWLEDGMENTS This work was supported in part by Grants-in Aid for Scientific research (No. 19310144, 19681024, 19651095, and 20241053) from Japan Society for the Promotion of Science, by grants from the Institute for Fermentation, Osaka (IFO), Collaborative Development of Innovative Seeds from Japan Science and Technology Agency ( JST), New Energy and Industrial Technology Development Organization (NEDO, project ID: 07A01014a), Research Grants from Yamada Science Foundation as well as Molecular Imaging Research Program, Grants-in-Aid for Scientific Research from the Ministry of Education, Culture, Sports, Science and Technology (MEXT) of Japan.

340


REFERENCES Chamaillard, M., Hashimoto, M., Horie, Y., Masumoto, J., Qiu, S., Saab, L., Ogura, Y., Kawasaki, A., Fukase, K., Kusumoto, S., Valvano, M. A., Foster, S. J., et al. (2003). An essential role for NOD1 in host recognition of bacterial peptidoglycan containing diaminopimelic acid. Nat. Immunol. 4, 702–707. Chen, W. C., Completo, G. C., and Paulson, J. C. (2008). Targeting of human B lymphoma with cytotoxic liposomes containing glycan ligands of CD22. Glycobiology 18, 963. Elliott, S., Lorenzini, T., Asher, S., Aoki, K., Brankow, D., Buck, L., Busse, L., Chang, D., Fuller, J., Grant, J., Hernday, N., Hokum, M., et al. (2003). Enhancement of therapeutic protein in vivo activities through glycoengineering. Nat. Biotechnol. 21, 414–421. Fujimoto, Y., Adachi, Y., Akamatsu, M., Fukase, Y., Kataoka, M., Suda, Y., Fukase, K., and Kusumoto, S. (2005). Synthesis of lipid A and its analogues for investigation of the structural basis for their bioactivity. J. Endotoxin Res. 11, 341–347. Fujimoto, Y., Iwata, M., Imakita, N., Shimoyama, A., Suda, Y., Kusumoto, S., and Fukase, K. (2007). Synthesis of immunoregulatory Helicobacter pylori lipopolysaccharide partial structures. Tetrahedron Lett. 48, 6577–6581. Girardin, S. E., Boneca, I. G., Carneiro, L. A., Antignac, A., Jehanno, M., Viala, J., Tedin, K., Taha, M. K., Labigne, A., Zahringer, U., Coyle, A. J., DiStefano, P. S., et al. (2003a). Nod1 detects a unique muropeptide from gram-negative bacterial peptidoglycan. Science 300, 1584–1587. Girardin, S. E., Boneca, I. G., Viala, J., Chamaillard, M., Labigne, A., Thomas, G., Philpott, D. J., and Sansonetti, P. J. (2003b). Nod2 is a general sensor of peptidoglycan through muramyl dipeptide (MDP) detection. J. Biol. Chem. 278, 8869–8872. Hirai, M., Minematsu, H., Kondo, N., Oie, K., Igarashi, K., and Yamazaki, N. (2007). Accumulation of liposome with Sialyl Lewis X to inflammation and tumor region: Application to in vivo bio-imaging. Biochem. Biophys. Res. Commun. 353, 553–558. Inamura, S., Fukase, K., and Kusumoto, S. (2001). Synthetic study of peptidoglycan partial structures. Synthesis of tetrasaccharide and octasaccharide fragments. Tetrahedron Lett. 42, 7613–7616. Inamura, S., Fujimoto, Y., Kawasaki, A., Shiokawa, Z., Woelk, E., Heine, H., Lindner, B., Inohara, N., Kusumoto, S., and Fukase, K. (2006). Synthesis of peptidoglycan fragments and evaluation of their biological activity. Org. Biomol. Chem. 4, 232–242. Inohara, N., Ogura, Y., Fontalba, A., Gutierrez, O., Pons, F., Crespo, J., Fukase, K., Inamura, S., Kusumoto, S., Hashimoto, M., Foster, S. J., Moran, A. P., et al. (2003). Host recognition of bacterial muramyl dipeptide mediated through NOD2. Implications for Crohn’s disease. J. Biol. Chem. 278, 5509–5512. Kamerling, J. P., Boons, G. J., Lee, Y. C., Suzuki, A., Taniguchi, N., and Voragen, A. G. J. (2007). Analysis of glycans, polysaccharide functional properties & biochemistry of glycoconjugate glycans, carbohydrate-mediated interactions. Comprehensive Glycoscience, From Chemistry to Systems Biology, Vol II & III , Elsevier, UK. Kaneko, T., Goldman, W. E., Mellroth, P., Steiner, H., Fukase, K., Kusumoto, S., Harley, W., Fox, A., Golenbock, D., and Silverman, N. (2004). Monomeric and polymeric gram-negative peptidoglycan but not purified LPS stimulate the Drosophila IMD pathway. Immunity 20, 637–649. Kaneko, Y., Nimmerjahn, F., and Ravetch, J. V. (2006). Anti-inflammatory activity of immunoglobulin G resulting from Fc sialylation. Science 313, 670–673. Kawasaki, A., Karasudani, Y., Otsuka, Y., Hasegawa, M., Inohara, N., Fujimoto, Y., and Fukase, K. (2008). Synthesis of diaminopimelic acid-containing peptidoglycan fragments and tracheal cytotoxin (TCT) for investigation of their biological functions. Chem. Eur. J. 14, 10318–10330.


341

Kusumoto, S., and Fukase, K. (2006). Synthesis of endotoxic principle of bacterial lipopolysaccharide and its recognition by the innate immune systems of hosts. Chem. Rec. 6, 333–343. Kusumoto, S., Fukase, K., and Oikawa, M. (1999). The chemical synthesis of lipid A. Endotoxin Health Dis. 243–256. Kusumoto, S., Fukase, K., and Fujimoto, Y. (2009). Chemical synthesis of bacterial lipid A. Microb. Glycobiol.: Struct. Relevance Appl. 415–427. Lee, Y. C., Townsend, R. R., Hardy, M. R., Lonngren, J., Arnarp, J., Haraldsson, M., and Lonn, H. (1983). Binding of synthetic oligosaccharides to the hepatic Gal/GalNAc lectin. Dependence on fine structural features. J. Biol. Chem. 258, 199–202. Magalhaes, J. G., Philpott, D. J., Nahori, M. A., Jehanno, M., Fritz, J., Le Bourhis, L., Viala, J., Hugot, J. P., Giovannini, M., Bertin, J., Lepoivre, M., Mengin-Lecreulx, D., et al. (2005). Murine Nod1 but not its human orthologue mediates innate immune detection of tracheal cytotoxin. EMBO Rep. 6, 1201–1207. Morell, A. G., Irvine, R. A., Sternlieb, I., Scheinberg, I. H., and Ashwell, G. (1968). Physical and chemical studies on ceruloplasmin V. Metabolic studies on sialic acid-free ceruloplasmin in vivo. J. Biol. Chem. 243, 155–159. Nakagawa, T., Uozumi, N., Nakano, M., Mizuno-Horikawa, Y., Okuyama, N., Taguchi, T., Gu, J., Kondo, A., Taniguchi, N., and Miyoshi, E. (2006). Fucosylation of N-glycans regulates the secretion of hepatic glycoproteins into bile ducts. J. Biol. Chem. 281, 29797–29806. Ogawa, T., Asai, Y., Sakai, Y., Oikawa, M., Fukase, K., Suda, Y., Kusumoto, S., and Tamura, T. (2003). Endotoxic and immunobiological activities of a chemically synthesized lipid A of Helicobacter pylori strain 206-1. FEMS Immunol. Med. Microbiol. 36, 1–7. Park, E. I., Mi, Y., Unverzagt, C., Gabius, H. J., and Baenziger, J. U. (2005). The asialoglycoprotein receptor clears glycoconjugates terminating with sialic acid alpha 2, 6GalNAc. Proc. Natl. Acad. Sci. USA 102, 17125–17129. Raetz, C. R., and Whitfield, C. (2002). Lipopolysaccharide endotoxins. Annu. Rev. Biochem. 71, 635–700. Sakai, Y., Oikawa, M., Yoshizaki, H., Ogawa, T., Suda, Y., Fukase, K., and Kusumoto, S. (2000). Synthesis of Helicobacter pylori lipid A and its analogue using p-(trifluoromethyl) benzyl protecting group. Tetrahedron Lett. 41, 6843–6847. Sato, M., Furuike, T., Sadamoto, R., Fujitani, N., Nakahara, T., Niikura, K., Monde, K., Kondo, H., and Nishimura, S. (2004). Glycoinsulins: Dendritic sialyloligosaccharidedisplaying insulins showing a prolonged blood-sugar-lowering activity. J. Am. Chem. Soc. 126, 14013–14022. Schleifer, K. H., and Kandler, O. (1972). Peptidoglycan types of bacterial cell walls and their taxonomic implications. Bacteriol. Rev. 36, 407–477. Stenbak, C. R., Ryu, J. H., Leulier, F., Pili-Floury, S., Parquet, C., Herve, M., Chaput, C., Boneca, I. G., Lee, W. J., Lemaitre, B., and Mengin-Lecreulx, D. (2004). Peptidoglycan molecular requirements allowing detection by the Drosophila immune deficiency pathway. J. Immunol. 173, 7339–7348. Suda, Y., Tochio, H., Kawano, K., Takada, H., Yoshida, T., Kotani, S., and Kusumoto, S. (1995). Cytokine-inducing glycolipids in the lipoteichoic acid fraction from Enterococcus hirae ATCC 9790. FEMS Immunol. Med. Microbiol. 12, 97–112. Takada, H., and Kotani, S. (1992). Vol: I Molecular Biochemistry and Cellular Biology. CRC Press, Boca Raton. Tanaka, K., and Fukase, K. (2008). PET (positron emission tomography) imaging of biomolecules using metal-DOTA complexes: A new collaborative challenge by chemists, biologists, and physicians for future diagnostics and exploration of in vivo dynamics. Org. Biomol. Chem. 6, 815–828.

342


Tanaka, K., Masuyama, T., Hasegawa, K., Tahara, T., Mizuma, H., Wada, Y., Watanabe, Y., and Fukase, K. (2008). A submicrogram-scale protocol for biomoleculebased PET imaging by rapid 6P-azaelectrocyclization: visualization of sialic acid dependent circulatory residence of glycoproteins. Angew. Chem. Int. Ed. 47, 102–105. Tanaka, K., Minami, K., Tahara, T., Fujii, Y., Siwu, E. R. O., Nozaki, S., Onoe, H., Yokoi, S., Koyama, K., Watanabe, Y., Fukase, K. (2010). Electrocyclization-based labeling allows efficient in vivo imaging of cellular trafficking. ChemMedChem. 5, 841–845. Varki, A., and Angata, T. (2006). Siglecs—The major subfamily of I-type lectins. Glycobiology 16, 1R–27R. Vyas, S. P., Singh, A., and Sihorkar, V. (2001). Ligand-receptor-mediated drug delivery: An emerging paradigm in cellular drug targeting. Crit. Rev. Ther. Drug Carrier Syst. 18, 1–76. Yano, T., Mita, S., Ohmori, H., Oshima, Y., Fujimoto, Y., Ueda, R., Takada, H., Goldman, W. E., Fukase, K., Silverman, N., Yoshimori, T., and Kurata, S. (2008). Autophagic control of listeria through intracellular innate immune recognition in Drosophila. Nat. Immunol. 9, 908–916. Yoshizaki, H., Fukuda, N., Sato, K., Oikawa, M., Fukase, K., Suda, Y., and Kusumoto, S. (2001). First total synthesis of the Re-type lipopolysaccharide. Angew. Chem. Int. Ed. Engl. 40, 1475–1480.

C H A P T E R

S E V E N T E E N

Multivalent Ligands for Siglecs Mary K. O’Reilly and James C. Paulson Contents 1. Introduction 1.1. Sialic acid-binding immunoglobulin-like lectins (Siglecs) 1.2. Glycan-binding specificity and cell-type expression 1.3. Cis- and trans-ligand binding 1.4. Multivalent scaffolds for siglec ligands 2. Materials and Methods 2.1. Reagents and cells 2.2. Preparation of siglec-expressing cells 2.3. PAA probe to siglec-expressing cells 2.4. PAA probes to siglec-Fc beads 2.5. Siglec-Fc to PAA probe beads 2.6. Siglec-Fc to biotinylated free saccharide-coated beads 2.7. CHO-Siglec cells to PAA probe beads 3. Conclusions and Future Directions Acknowledgments References

344 344 347 348 349 350 350 351 352 355 356 358 359 359 360 361

Abstract Siglecs have emerged as an important family of immunomodulatory glycanbinding proteins that can bind sialoside ligands both on the same cell surface, in cis, and on other cells, in trans. Expression of siglecs varies among a variety of immune cells, and tools to probe siglecs on these cells are crucial to understanding their function. In designing synthetic ligands, competition by cis ligands requires the use of multivalency to achieve sufficient avidity to stably bind siglecs on native cells. This chapter describes the use of multivalent ligands to probe cell surfaces, as well as to investigate ligand binding to recombinant siglecs.

Departments of Chemical Physiology and Molecular Biology, The Scripps Research Institute, La Jolla, California, USA Methods in Enzymology, Volume 478 ISSN 0076-6879, DOI: 10.1016/S0076-6879(10)78017-4

#


343

344

Mary K. O’Reilly and James C. Paulson

Abbreviations NeuAc BPC BPA Gal GlcNAc HBSS BSA FBS CHO PAA FITC Ig PBS DTAF

N-Acetyl neuraminic acid Biphenylcarbonyl Biphenylacetyl Galactose N-Acetylglucosamine Hank’s balanced salt solution Bovine serum albumin Fetal bovine serum Chinese hamster ovary Polyacrylamide Fluorescein isothiocyanate Immunoglobulin Phosphate-buffered saline Dichlorotriazinylaminofluorescein

1. Introduction 1.1. Sialic acid-binding immunoglobulin-like lectins (Siglecs) Many glycan-binding proteins are involved in the regulation of the immune system, through both activating and inhibitory mechanisms, as well as cell– cell adhesion, homing of immune cells, and pathogen recognition. The sialic acid-binding immunoglobulin-like lectin, or Siglec, family comprises glycan-binding proteins believed to be involved in all of these functions (Crocker et al., 2007). Sharing a common sialic acid-binding function via the terminal V-set Ig domain and variable numbers of C2-set Ig-like domains, these receptors nevertheless have overlapping but distinct celltype distribution and specificity for the underlying glycan (Table 17.1). Many contain intracellular signaling motifs, such as the immunoreceptor tyrosine inhibitory motif (ITIM), immunoreceptor tyrosine activatory motif (ITAM), or Grb-binding domain (Crocker et al., 2007). Four of the siglecs are highly conserved among species, including Sialoadhesin (Siglec1), CD22 (Siglec-2), CD33 (Siglec-3), and MAG (Siglec-4), while the remaining, known as CD33-like siglecs, are rapidly evolving, presumably due to adaptive pressures from viruses and microorganisms that have gained the ability to incorporate sialic acid (Angata, 2006; Severi et al., 2007). The expression of siglecs predominantly on immune cells and the presence of

Table 17.1 Glycan-binding specificity and cellular distribution of siglecs

Siglec (other names)

Murine ortholog or paralog

Sialoadhesin (SAD, Sn, Siglec-1)

Sialoadhesin (mSiglec-1, mSn)

CD22 (Siglec-2)

mCD22 (mSiglec-2)

CD33 (Siglec-3)

mCD33 (mSiglec-3)

MAG (Siglec-4)

mMAG (mSiglec-4)

Siglec-5

–

Sialoside preferencea

Cell-type expressionb

a3 b4

a6 a4

Tissue macrophages (activated monocytes) 6S

a6

b4

a3

b3

Monocytes, basophils, CD34þ cells, dendritic cells, macrophages, mast cells, neutrophils (granulocytes, myeloid progenitors) Oligodendrocytes, Schwann cells

a8

Neutrophils, monocytes, basophils, CD34þ cells, macrophages, mast cells (B cells)

a6

Siglec-6

–

a6

Siglec-7

–

a8

Siglec-8

Siglec-9

Siglec-F

–

Basophils, mast cells, placental trophoblasts (B cells) a3

b4

6S a3 b4 a3

a3

B cells

b4

NK cells, dendritic cells, monocytes, CD8þ T cells (monocytes) Eosinophils, mast cells (basophils)

6S a3

Monocytes, neutrophils, dendritic cells, CD34þ cells, CD8þ T cells (NK cells) (continued)

Table 17.1 (continued) Siglec (other names)

Murine ortholog or paralog

Siglec-10

Siglec-G

Sialoside preferencea a6 a6

Siglec-11

–

a8

Siglec-14

–

a8

Cell-type expressionb

b4

B cells, CD34þ cells, dendritic cells, monocytes, NK cells (eosinophils)

b4

Monocytes, macrophages, brain microglia Not determined, but expected to be similar to Siglec-5 based on sequence homology

a6

Siglec-15

mSiglec-15

a6

Macrophages, monocytes, dendritic cells

Siglec-16

–

a8

Macrophages (brain microglia)

–

Siglec-E

–

Siglec-H

a3

Key: a

b

NeuAc

NeuGc

Gal

GalNAc

b3 a6

– GlcNAc

Neutrophils, monocytes, dendritic cells Plasmatoid dendritic cells (macrophages)

Fuc

S

Sulfate

Sialoside preferences are taken from recent reviews (Crocker et al., 2007; O’Reilly and Paulson, 2009; von Gunten and Bochner, 2008), data from the Consortium for Functional Glycomics http://www.functionalglycomics.org, or inferred from the binding preferences of highly homologous siglecs. Carbohydrate sequences shown refer to preferences of the human counterparts, with the exception of Siglec-E. Compiled from recent reviews (Crocker et al., 2007; O’Reilly and Paulson, 2009; von Gunten and Bochner, 2008).

Multivalent Ligands for Siglecs

347

intracellular signaling motifs suggest a role in immunomodulation for siglecs, which has been validated for many, though the role of glycan binding is still poorly understood (Crocker and Redelinghuys, 2008).

1.2. Glycan-binding specificity and cell-type expression Synthetic sialoside ligands of siglecs have been developed to probe their function and glycan-binding specificity, and to detect siglecs on different cell types. This chapter will address the detection of siglecs on cells using ligand-based probes, which requires consideration of both cell-type expression and glycan-binding specificity. As shown in Table 17.1, siglecs are expressed on a variety of cells, most of which are immune cells (Crocker et al., 2007; O’Reilly and Paulson, 2009; von Gunten and Bochner, 2008). Certain siglecs, such as CD22 and Siglec-8, are expressed predominantly on one cell type, B cells and eosinophils, respectively. Others can be expressed on several cell types, such as Siglec-9 on monocytes, dendritic cells, and neutrophils. Table 17.1 also shows the preferred glycan(s) for each siglec that has been shown to bind sialic acid. Similar to cellular distribution, some siglecs have strict specificity, while others can bind several different glycan structures. Specificity can be considered from the perspective of the siglec and of the carbohydrate ligand, which may also have one or more cognate binding partners. CD22 is highly specific for sialosides with the a-2,6 linkage, but other more promiscuous siglecs can bind this sialoside as well, precluding specific targeting of this sequence to CD22. The discovery that the preferred ligand of human CD22 includes a sulfate group on the 6-position of GlcNAc may improve the ability to achieve more selective binding (Blixt et al., 2004; Kimura et al., 2007). Siglec-7 exhibits a clear preference for glycans with the NeuAca2,8-NeuAca2,3-Galb1,4-GlcNAc sequence, but also bind NeuAca2,3-Galb1,4-GlcNAc and NeuAca2,6Galb1,4-GlcNAc (O’Reilly and Paulson, unpublished results). Siglec-8, expressed on eosinophils, binds preferentially to 60 -sulfo-sialyl LewisX. As an example of specificity from the perspective of the ligand, a polyacrylamide (PAA) polymer of 60 -sulfo-sialyl LewisX binds selectively to only eosinophils among leukocytes in a sample of whole blood (Hudson et al., 2009). Several labs have explored the use of sialic acid analogs to achieve enhanced binding and selectivity for one siglec over others (Blixt et al., 2008; Chokhawala et al., 2008). A biphenyl substitution at the 9-position of sialic acid was able to enhance the affinity of CD22 for the ligand, NeuAca2,6-Galb1,4-GlcNAc, by 100-fold, for example (Kelm et al., 2002). The use of glycan arrays is greatly accelerating the structure–activity relationship for siglec ligands, although more work is needed before the goal of a specific ligand for each siglec can be achieved.

348


1.3. Cis- and trans-ligand binding In nature, siglecs can bind glycans terminating in sialic acid both on the same cell (in cis) or on other cells, glycoproteins, viruses, etc. (in trans). The masking effect of cis ligands on siglecs (Fig. 17.1) has been known since the demonstration that binding of a synthetic multivalent CD22 ligand to CD22 on B cells could be enabled or enhanced by removal of sialic acids from the cell surface or destruction of the sialic acid glycerol side chain, a key binding determinant (Razi and Varki, 1998). While the highest affinities exhibited by siglecs for their preferred ligands is micromolar (Kd) (Bakker et al., 2002), the concentration of sialic acids on the cell is estimated to be in the millimolar range (e.g., 25 mM in the glycocalyx of B-cells (Collins et al., 2004). The endogenous ligands have not been identified for all siglecs, but CD22 has been shown to predominantly bind to the glycans of other molecules of CD22 in cis (Han et al., 2005), and to the B cell receptor, IgM, in trans with other B cells (Ramya et al., 2010). The ability of CD22 to bind glycans on other cells in trans was demonstrated by using fluorescence microscopy to visualize the colocalization of CD22 at the site of cell–cell contact between two B cells (Collins et al., 2004). Importantly, this localization was dependent on the expression of a2,6 sialosides on the trans cell. Binding to glycans in trans on pathogenic organisms has been documented for several siglecs, including HIV-1 to sialoadhesin, Campylobacter jejuni to Siglec-7, Group B Streptococcus to Siglec-9, and Neisseria meningitidis to Siglec-F (Avril et al., 2006; Carlin et al., 2009a,b; Jones

Sialic acid (NeuAc) Man or Gal GlcNAc or GalNAc

PAA probe

Figure 17.1 Schematic of competition between trans ligands and cis ligands of siglecs. Using CD22 as an example, cis ligand binding leads to masking of the ligand-binding site. Only with sufficient avidity (or removal of sialic acids) can trans ligands compete with cis ligands to achieve stable binding at the cell surface.


349

et al., 2003; Rempel et al., 2008; Severi et al., 2007; van der Kuyl et al., 2007). Presumably, these interactions are influenced by the degree of cis ligand masking, making the composition of cis ligands on the cell surface a possible determinant for recognition of pathogens and the immune response. This scenario highlights the need for ligand-based methods of siglec detection. While specific antibodies can be used to probe cell types for siglec expression, only multivalent ligand-based probes can define the functional availability of siglecs. Many factors are involved in modulation of masking, including expression levels of sialyltransferases and sialidases, as well as enzymes regulating the biosynthesis of underlying glycan structures. An additional level of regulation is achieved by postglycosylational modifications, including sulfation, acetylation, and sialic acid cyclization, which are regulated by other enzymes (Cariappa et al., 2009; Yu and Chen, 2007).

1.4. Multivalent scaffolds for siglec ligands Due to the low affinity of siglec–ligand interactions and competition from cis ligands, multivalency is needed to achieve the avidity required of synthetic ligands (Fig. 17.1). Polymers have provided a convenient scaffold for siglec ligands with defined lengths and substitution densities. Ruthenium-catalyzed olefin metathesis polymerization (ROMP) has been used to prepare polymers of the CD22 ligand to study CD22 function (Courtney et al., 2009; Yang et al., 2002). The study of siglecs and other glycan-binding proteins has drawn heavily on the use of PAA constructed with pendant carbohydrate ligands and biotin groups (Chinarev et al., 2010; Rapoport et al., 2006). This chapter will focus primarily on PAA polymer-based siglec ligands with a brief section on univalent biotinylated ligands because the reagents needed for these methods are readily available, and no further synthesis is required. Other multivalent scaffolds for siglec ligands have been developed more recently, which have the benefit of being more rigid and structurally defined. Viral capsids (e.g., cowpea mosaic virus and bacteriophage Qb) have been chemoenzymatically decorated with a high-affinity CD22 ligand with remarkable control over spacing and valency (Kaltgrad et al., 2008). These were able to bind to CD22 on native B cells. Another system proved that with the proper spacing and geometry, valency becomes less important. A heterobifunctional CD22 ligand bearing a hapten is able to drive the self-assembly of CD22–IgM complexes at the surface of native B cells (O’Reilly et al., 2008). The maximum valency of this complex is 10, and in fact the same ligand was able to mediate stable complex formation between CD22 and the lower valency antibodies, IgA and IgG. For purposes of in vivo drug delivery to specific cell types, liposomes decorated with CD22 ligand and loaded with doxorubicin have been shown to bind and kill native B cells, to prolong life in a murine model of disseminated B cell lymphoma, and to kill malignant B cells in samples taken from lymphoma patients (Chen et al., 2010). Liposomes, viral capsids, and heterobifunctional ligands may also be used to

350


probe immune cells as described herein, but given the labor-intensive preparation of these alternate platforms, it is more practical to use the available PAA or biotinylated probes. Finally, glycan arrays are commonly used to probe the binding specificity of glycan-binding proteins such as siglecs (Blixt et al., 2004, 2008; Bochner et al., 2005). The analogous experiment described in this chapter would be soluble siglec-Fc binding to PAA probes immobilized on magnetic beads. The glycan array may be preferred for a broad screening due to the highthroughput nature and the relatively miniscule amounts of glycan required. However, the array has not yet been optimized for screening of siglec-expressing cells, and it is often desirable to investigate siglecs in a more native-like environment, considering such effects as lateral mobility and masking by cis ligands.

2. Materials and Methods 2.1. Reagents and cells Siglec-expressing cells can be primary cells from human or murine origin, cell lines that natively express siglecs, or cells transfected with siglecs, most commonly CHO cells. Some commonly used B cell lines used to probe CD22 include BJAB, Daudi, and Raji, all of which are maintained in RPMI media containing 10% fetal bovine serum (FBS). The BJAB (K20) cell line is of particular interest due to a mutation in an epimerase that is required to synthesize sialic acid (Hinderlich et al., 2001). Growing BJAB (K20) cells in serum-free media results in asialo cells that thus lack cis ligands of CD22. To wean BJAB (K20) cells off of serum, cells are initially grown in RPMI containing 10% FBS and 50 mM 2-mercaptoethanol, and then switched to half of the previous media and half serum-free HYQ-SFM medium for 2 days, replacing with fresh media every day, and then replacing with 100% serumfree HYQ-SFM. Cells become semiadherent, and 2 mM EDTA can be used to dislodge cells. At this point media is changed daily. Many of the siglecs have also been cloned and stably transfected into CHO cells as a convenient and comparable model for studying in situ siglec function (Khatua et al., 2010; Munday et al., 2001; Tateno et al., 2007). These are grown in 1:1 DMEM:F12, 10% FBS, and, if cloned as previously described, 250 mg/mL Hygromycin B (Roche Diagnostics). As this is an adherent cell line, Trypsin/EDTA should be used to dislodge the cells for passaging, but only 2 mM EDTA should be used to harvest cells for experiments, as trypsin will degrade cell-surface proteins. It should be noted that CHO cells do not express appreciable amounts of a2-6 sialosides. CD22 on CHO cells is therefore unmasked and removal of cis ligands is unnecessary. Siglec-Fc chimerae are also a common source of siglecs, and represent truncated fusions of the siglecs that comprise part of the extracellular portion of the siglec, including the entire N-terminal V-set carbohydrate-binding domain, and an IgG constant fragment (Fc) at the C-terminus.


351

The Fc domain dimerizes, which provides bivalency to the siglec, while it also enables detection, complexation, and/or immobilization of the siglec by using anti-IgG antibodies. These constructs can be expressed in COS cells as previously described (Nath et al., 1995; van der Merwe et al., 1996; Vinson et al., 1996). Alternatively, many siglec-Fc fusions are commercially available from R&D Systems. Both biotinylated PAA probes and biotinylated univalent ligands are commercially available from Glycotech (Gaithersburg, MD) or can be requested by participating investigators from the Consortium for Functional Glycomics (http://www.functionalglycomics.edu). PAA probes are water soluble and can be stored in solution for months to years at 4 C or below. Extended storage can lead to precipitation, in which case the probe can be coaxed back into solution with gentle mixing and/or warming. PAA probes obtained from the Consortium for Functional Glycomics are substituted with 20 mol% carbohydrate ligands and 5 mol% biotin. Many are available as either a low molecular weight version (30 kDa) or a high molecular weight version (1500 kDa). Glycotech probes are provided as 30-kDa polymers in which typically every 5th amide is substituted with biotin in a 4:1 ratio. In principle, PAA probes could be precomplexed with streptavidin prior to the binding. On one hand, this may greatly increase valency if multiple polymer chains are cross-linked by the tetravalent streptavidin. On the other hand, precomplexation may greatly restrict the degrees of freedom available to the polymer backbone. The diminished flexibility of the chain may dampen the ability to bind its cognate siglec. This effect may be particularly important if the siglec does not have lateral mobility, such as recombinant siglec immobilized on beads, or even on cells at 4 C. Another consideration of precomplexation of the probe is that it is likely that if PAA probes can be sufficiently cross-linked with streptavidin, the complex may be too large to be internalized by CD22 via clathrin-dependent endocytosis, enabling binding assays at 37 C without the complication of internalization.

2.2. Preparation of siglec-expressing cells Due to the high concentration of sialosides at the cell surface, siglecs are constitutively bound by neighboring ligands residing on the same cell surface (cis ligands). One notable exception to this is sialoadhesin, which, because of 17 extracellular Ig-domains, extends beyond the cellular glycocalyx and is thought to be the only unmasked siglec. Synthetic multivalent ligands have been designed that can compete with these cis ligands to achieve stable binding to siglecs on the cell surface, but for purposes of siglec detection, removal of these ligands may in some cases be desirable or necessary. Two simple methods are available to achieve this purpose, and include removal of sialic acid with neuraminidases, and destruction of the glycerol side chain of sialic acid by periodate oxidation.

352


A commonly used neuraminidase for the purpose of removing cis ligands is the Arthrobacter ureafaciens sialidase (AUS), which is available from Roche. Cells are suspended in Hank’s Balanced Salt Solution (Gibco) containing 5 mg/mL bovine serum albumin (Sigma) (HBSS/BSA) at a density of 0.2–1 107 cells/mL. AUS is then added to a final concentration of 200 mU/mL and the cells are incubated at 37 C for 30 min. After washing twice with cold HBSS/BSA, cells are ready for probing with the exogenous ligand. Periodate oxidation is another convenient method to destroy cis ligands. Cells are resuspended in 1 106/mL, and 1 mM sodium periodate (SigmaAldrich, cat. no. 311448) is added from a 200 mM stock solution in water. Cells are incubated at 4 C for 10 min before quenching the reaction by adding equimolar glycerol. After washing the cells twice in cold HBSS/ BSA, they are ready for analysis. An important consideration for subsequent use of periodate-oxidized cells is that this treatment only addresses glycans that are located on the cell surface at the time of treatment, since periodate is cell-impermeable at 4 C. Due to the rapid turnover of the cell surface, there is a constant replenishment of glycoproteins and glycolipids from de novo biosynthesis and recycling from intracellular compartments. Even a brief warming of the cells to 37 C could lead to a significant repopulation of cis ligands at the cell surface. While this effect may be less of a concern after sialidase treatment, which is done at 37 C and could address rapidly recycling factors as they reach the cell surface, it is still a consideration for newly emerging glycans.

2.3. PAA probe to siglec-expressing cells When staining siglec-expressing cells with ligand-conjugated PAA probes, there are several considerations. If the lower molecular weight polymer is being used, then cells will most likely need to be treated with sialidase or periodate, unless the cells are devoid of cis ligands, as is the case of CD22 expressed on CHO cells. For example, eosinophils and Siglec-F expressing CHO cells had to be pretreated with sialidase to achieve binding of a 30kDa polymer of 60 -sulfo-sialyl LewisX, while the high-molecular-weight polymer could bind both treated and untreated cells (Fig. 17.2). Another consideration is the intrinsic affinity of the ligand. For example, NeuGcLacNAc-PAA that is not substituted with the affinity-enhancing BPA group at the 9 position does not bind to CD22 on native B cells even if it is incorporated into a high molecular weight polymer (Fig. 17.3). These data also show that the binding of the native ligands is much more sensitive to the masking effect of cis ligands compared to the high-affinity versions. While modest binding of the NeuGc probe to native murine B cells has since been observed, the BPA-substituted version is a much more robust binding partner (Duong et al., 2010). In the latter example, a lower

353


300

SigF-CHO 30 kDa

0

0

30 kDa

102

103

104

100

101

300

101

300

100

1000 kDa

102

103

104

1000 kDa

0

0

Cell number

300

Eosinophils

100 Key:

101

102 Control

103

104

100 101 x 6′-Sulfo-sLe -PAA Native cells

102

103

104

Asialo cells

Figure 17.2 PAA probe binding to siglec-expressing cells reveals importance of valency. 60 -sulfo-sialyl-Lewisx-PAA of low molecular weight (30 kDa, approximately 15-mer) or high molecular weight (1000 kDa, approximately 500-mer) was incubated with eosinophils, which express Siglec-F, or CHO cells transfected with Siglec-F. Asialo cells were prepared by pretreatment of native cells with sialidase (Tateno et al., 2007). Reproduced with permission from the American Society of Microbiology.

concentration of probe was used (1.25 mg/mL). Thus, some optimization may be advantageous for each siglec-probe combination. 1. Check that no precipitation of the polymer has occurred, and if it has, pipet the solution up and down to redissolve the polymer. 2. Wash and resuspend cells at a density of 2 106 mL 1 in 100 mL of HBSS/BSA. 3. Add 0.1–1 mg of polymer for a final probe concentration of 1–10 mg/mL. 4. Incubate at 4 C for 1 h with end-over-end rotation or occasional mixing. 5. Pellet cells by centrifugation in an Eppendorf centrifuge (5415D or similar model) at 2000 rpm for 5 min at 4 C. 6. Wash cells with 1 mL ice-cold HBSS/BSA twice, pelleting after each wash as in step 5. 7. Resuspend in 100 mL of HBSS/BSA. 8. Add 1 mL of DTAF-Streptavidin ( Jackson Laboratories, catalog # 016010-084) and incubate at 4 C for 30 min. 9. Wash away unbound streptavidin with HBSS/BSA as above.

354


A

B 80

Murine B cells

100

103

0

70 140 210 280

101 102 NeuAc-PAA

100

101

102

103

Cell number 0 10 20 30 40 50 60 70 80 0 20

40

60

Cell number 0 70 140 210 280

BJAB

101 102 BPA-NeuGc-PAA

BPC-NeuAc-PAA Key:

Background

101 102 NeuGc-PAA

Native

103

103

Asialo

Figure 17.3 High-affinity PAA probe binding to siglec-expressing cells shows the importance of intrinsic affinity in overcoming cis ligands. A human BJAB B cell line (A) or murine B cells (B) were probed with native (NeuAc(Gc)a2,6-LacNAc-) or high-affinity BPC(BPA) NeuAc(Gc)a2,6-LacNAc- ligands appended to PAA (Collins et al., 2006). Reproduced with permission from the Journal of Immunology.

10. Resuspend cells in 200 mL HBSS/BSA. 11. Analyze by flow cytometry (10,000 cells/sample). It is important that these experiments be carried out at 4 C to prevent endocytosis, which would cause internalization of the probe and thus an underestimation of binding. Alternatively, if performing the experiment at 37 C is desired, maintaining the cells in hypertonic media will prevent endocytosis by preventing the formation of clathrin-coated pits (Heuser and Anderson, 1989). We should note, however, that this method only affects siglecs that undergo clathrin-dependent internalization. Siglec-F, for example, is endocytosed by a clathrin-independent mechanism (Tateno et al., 2007), and the mechanism for many others is not yet known. Cells are first incubated for 45 min at 37 C in media or buffer containing 0.45 M sucrose, then washed twice with the same sucrose-containing media at 4 C. Since this inhibitory effect is reversible, cells must be kept in this high concentration of sucrose throughout the experiment. We have shown that this


355

environment does not affect CD22 binding to synthetic multivalent ligands (O’Reilly and Paulson, unpublished observations).

2.4. PAA probes to siglec-Fc beads While there are certain advantages to testing the binding of glycans to siglecs in their native environment, it can also be useful to examine specificity by using recombinant siglec-Fc fusion proteins, which enables a quantitative measure without the complexity of the cellular environment. This section will discuss immobilized siglec-Fc, while Section 2.5 will describe its use in the soluble form. Commercially available magnetic beads conjugated to Protein A (DynabeadsÒ Protein A for Immunoprecipitation, Cat. No. 100-01, Invitrogen) are used for immobilization through the Fc portion. These beads have a capacity of approximately 8 mg human IgG per milligram of beads, and are supplied as 30 mg/mL. Either purified siglec-Fc or culture supernatant from siglec-Fc expressing COS cells is applied to Protein A beads for immobilization as previously described (O’Reilly et al., 2008). One advantage of this method is that a siglec-Fc purification step is built in. In fact, it may be preferable to use culture media from siglec-Fc expressing cells to load the beads if these are available because the conditions used to purify the fusion protein can lead to loss of binding activity. This direct method negates the harsh elution step. These beads can then be probed with the carbohydrate-conjugated PAA probes as described for siglec-expressing cells in Section 2.3, except that instead of using centrifugation for the washing steps, beads are isolated using a magnetic tube rack (MagneSphereÒ Technology Magnetic Separation Stand (12-position), catalog # Z5342, Promega). Using beads as a model for siglec-expressing cells has the benefit of removing complications due to both cis ligands and siglec endocytosis at elevated temperatures. Therefore, binding can be examined at a variety of temperatures. This was done using hCD22-coated beads and BPCNeuAcLacNAc-PAA, revealing that at least for these interacting partners, binding may be improved at elevated temperatures (Fig. 17.4). Using beads therefore may be a way to achieve binding where binding to cells at 4 C is not observed. Because PAA probes are not very structurally well defined, increased temperatures may help the polymer to overcome any secondary structure and increase the opportunity for favorable binding. Another advantage of using beads is that very long incubations are possible to ensure that binding has reached equilibrium. Long incubations may compromise the integrity of cells or cause other unknown changes. Disadvantages include lack of lateral mobility of siglecs. When analyzing the binding of probes to beads by flow cytometry, we have noticed that the predominant population by dot plot (forward scatter vs. side scatter) is accompanied by a series of less abundant populations with

356


100

% of max

80 60 40 20 0 100

101 102 103 Streptavidin (DTAF)

104

LacNAc-PAA, 4 ⬚C BPCNeuAc-LacNAc-PAA,

4 ⬚C

LacNAc-PAA, 37 ⬚C BPCNeuAc-LacNAc-PAA,

37 ⬚C

Figure 17.4 Siglec-Fc beads enable binding at elevated temperature without endocytosis to reveal improved binding. Human CD22–Fc beads were probed with high molecular weight LacNAc-PAA or BPCNeuAca2,6-LacNAc-PAA at 4 or 37 C for 16 h.

increasing forward and side scatter. The abundance of each population decreases with increasing scattering properties, and the degree to which this effect occurs seems to correlate with the amount of ligand bound. It is possible that the multivalent ligands can cross-link beads to change their light scattering properties, though this effect has not been further investigated.

2.5. Siglec-Fc to PAA probe beads As opposed to immobilizing the siglec, the reverse experiment can be carried out by immobilizing the PAA probe. With this method, it is important to realize that despite the bivalency of the siglec-Fc, we do not observe any binding unless the siglec-Fc is precomplexed with the secondary antibody, FITC-anti-human IgG. Figure 17.5 gives the example of Siglec-7 binding to beads coated with NeuAca2,8-NeuAca2,3-LacNAcPAA, but this principle applies to each siglec-Fc that we have tested. Precomplexation leads to greater valency due to crosslinking of at least two, but possibly many more, siglec chimera dimers, depending on the isotype of the secondary antibody. In practice, precomplexation is not performed as a separate step, but rather by adding the secondary antibody and siglec-Fc simultaneously to PAA beads.

357


Control PAA beads

100

Ligand-PAA beads

80 60 40

2-Step, no precomplexed % of max

20 0 100 80 60 1-Step, precomplexed

40 20 0 100

101

102 103 Siglec-7-Fc

104

Figure 17.5 Precomplexation of Siglec-Fc is key to engaging with ligand–PAA-coated beads. Siglec-7-Fc from culture supernatant was incubated with streptavidin beads coated with unsubstituted PAA or NeuAca2,8-NeuAca2,3-Galb1,4-GlcNAc-PAA, either in the absence (top, 2-step) or presence (botton, 1-step) of the detection antibody, FITC-antihuman IgG. For the 2-step method (top), unbound probe was washed away prior to adding the detection antibody.

1. Remove a 5-mL aliquot (3–3.5 106 beads) of streptavidin-coated Dynal microbeads (DynabeadsÒ M-280 Streptavidin Cat. No. 112-05D, Invitrogen) and wash with HBSS/BSA. 2. Add 6 mg of PAA probe in 0.5 mL HBSS/BSA and incubate at 25 C for 1 h. 3. Wash beads twice using magnetic stand with 1 mL HBSS/BSA and resuspend in 50 mL. 4. Combine 2 mL of beads with 50 mL of culture supernatant from siglec-Fc expressing cells and 1 mL of FITC-antihuman IgG (Jackson Immunoresearch, cat. no. 109-095-098). 5. Incubate at 4 C with end-over-end rotation for 30 min. 6. Isolate beads using a magnetic stand and wash twice with 1 mL HBSS/BSA. 7. Resuspend beads with 200 mL HBSS/BSA and analyze bead fluorescence by flow cytometry (10,000 beads/sample).

358


2.6. Siglec-Fc to biotinylated free saccharide-coated beads Similar to immobilizing the PAA probe to streptavidin beads, free ligands conjugated to biotin can also be used to load streptavidin beads, and these reagents are available both commercially and from the Consortium for Functional Glycomics. Like the PAA-loaded beads, this platform provides high valency. It differs from PAA-loaded beads in certain ways that may be desirable for some applications. For instance, titrating in free biotin can govern the loading of the bead, and thus the avidity (Fig. 17.6). PAA probes, on the other hand, are not as easy to control in terms of valency. Because of the effectively irreversible binding of biotin to streptavidin, the polymers may become trapped in conformations with unknown numbers of ligands being accessible for siglec binding. For competition assays with free inhibitor, lower avidity may be desired, whereas higher avidity may be needed in other circumstances to achieve measurable binding. Also, biotinylated free ligands are easier to access synthetically, so candidate ligands could be tested without having to synthesize the entire polymer. The procedure used to analyze siglec binding in the experiment shown in Fig. 17.6 is as follows, and can be modified to use different ligands, siglecs, and loading densities. 1. Retrieve a 20-mL aliquot of Dynamax beads (Invitrogen). 2. Wash twice with 1 mL of PBS, using the magnetic eppendorf rack to collect the beads. 100

% of max

80 60 40 20 0 100

101

102 103 Siglec-E-Fc

104

Control

15 pmol biotin

No biotin

50 pmol biotin

Figure 17.6 Tuning the density of biotinylated ligand-coated beads to modulate avidity of siglec-Fc binding. Streptavidin beads were coated with biotinylated NeuAca2,8NeuAca2,3-Galb1,4-GlcNAc after pretreatment of beads with varying concentrations of free biotin to adjust the density of carbohydrate ligands. Beads were then probed with siglec-E-Fc from culture supernatant in the presence of the detection antibody, FITCantihuman IgG. Control beads had neither biotin nor ligand.


359

3. Resuspend in 0.4 mL of PBS and remove four 100-mL aliquots. 4. Add 0, 5, 15, or 50 pmol, respectively, of 5-(biotinamido)pentylamine from a 50 mM stock in PBS. 5. Incubate for 1 h at 25 C with end-over-end rotation. 6. Wash each sample twice with 1 mL HBSS/BSA. 7. Resuspend each in 50 mL HBSS/BSA. 8. Add 1 mL of biotinylated ligand (from 55 mM stock) and incubate for 1 h at 25 C with end-over-end rotation. 9. Wash each sample twice with 1 mL HBSS/BSA. 10. Resuspend each in 50 mL HBSS/BSA. 11. Add 5-mL aliquots of beads to 200-mL aliquots of culture supernatant from siglec-Fc-expressing cells. 12. Immediately add 4 mL of FITC-antihuman-IgG and incubate at 4 C for 30 min with end-over-end rotation. 13. Wash each sample twice with HBSS/BSA. 14. Resuspend beads in 200-mL aliquots of HBSS/BSA and analyze by flow cytometry (10,000 beads/sample).

2.7. CHO-Siglec cells to PAA probe beads Higher ligand valency may be achieved by coating synthetic beads with the PAA probe of interest. Combining these beads with cells may prove a better model for cell–cell adhesion than soluble probes. This method has the advantage of providing very high valency for both ligand and siglec. Microscopy can then be used to visualize the cell–bead interactions. This method may also be of interest for experiments done at elevated temperatures. At 37 C, most siglecs will undergo clathrin-dependent or independent endocytosis, transporting the ligand inside the cell. We have observed that with a sufficiently large ligand scaffold, such as a highly cross-linked anti-NP IgM, internalization is precluded (O’Reilly and Paulson, unpublished observations). The procedure for PAA bead–cell adhesion has been published previously, and was used to demonstrate bead-bound BPANeuGc-PAA binding to murine B cells (Fig. 17.7) (Collins et al., 2006). These results are consistent with the free probe-binding results shown in Fig. 17.3. Therefore, as an alternative, the PAA probe can be bound to the cells first, followed by washing and then exposure to the streptavidin-coated magnetic beads. This procedure has also been described (Collins et al., 2006).

3. Conclusions and Future Directions Discovery of ligand analogs with enhanced affinity and selectivity will improve the precision with which siglecs can be studied. Substituents at various positions, most commonly the 5 and 9 positions of sialic acid,

360


30 kDa

1000 kDa

NeuGc-PAA

BPA-NeuGc-PAA

Figure 17.7 High-affinity ligand is required for adhesion between PAA-coated beads and CD22-expressing cells. Streptavidin beads coated with 30 or 1000 kDa versions of NeuGca2,6-LacNAc-PAA or BPANeuGca2,6-LacNAc-PAA were exposed to primary murine B cells (Collins et al., 2006). Reproduced with permission from the Journal of Immunology.

have been shown to enhance or diminish siglec affinity, thus setting the stage to develop ligands with a high degree of selectivity for specific siglecs (Blixt et al., 2008; Chokhawala et al., 2008). These tools will enable improved detection and targeting of siglecs in complex biological systems. Depending on the application, different permutations of carbohydrate probe and siglec immobilization will serve different needs for the detection and analysis of siglecs in their native environment. Siglecs are already considered targets for immunotherapy of cancers, autoimmunity, and other inflammatory disorders (O’Reilly and Paulson, 2009). With the continued development of selective probes will come improved methods for assessing changes in the availability of siglec binding sites on different cells and under different conditions (i.e., transformed cells, microenvironments of inflammation, etc.). This can in turn lead to enhanced targeting strategies using glycan-based drug delivery vehicles. These probes will also be useful for investigating the innate functions of siglecs and the contribution of glycan binding.

ACKNOWLEDGMENTS The authors wish to thank Anna Tran-Crie for her assistance in preparation of the manuscript, and Cory Rillahan, Dr. Hua Tian, and Dr. Christoph Rademacher for careful reading of the manuscript and helpful suggestions. This work was supported by NIH R01AI050143 & R01GM60938 to J. C. P., and an American Cancer Society postdoctoral fellowship to M. K. O.


361

REFERENCES Angata, T. (2006). Molecular diversity and evolution of the Siglec family of cell-surface lectins. Mol. Divers. 10, 555–566. Avril, T., Wagner, E. R., Willison, H. J., and Crocker, P. R. (2006). Sialic acid-binding immunoglobulin-like lectin 7 mediates selective recognition of sialylated glycans expressed on Campylobacter jejuni lipooligosaccharides. Infect. Immun. 74, 4133–4141. Bakker, T. R., Piperi, C., Davies, E. A., and Merwe, P. A. (2002). Comparison of CD22 binding to native CD45 and synthetic oligosaccharide. Eur. J. Immunol. 32, 1924–1932. Blixt, O., Head, S., Mondala, T., Scanlan, C., Huflejt, M. E., Alvarez, R., Bryan, M. C., Fazio, F., Calarese, D., Stevens, J., Razi, N., Stevens, D. J., et al. (2004). Printed covalent glycan array for ligand profiling of diverse glycan binding proteins. Proc. Natl. Acad. Sci. USA 101, 17033–17038. Blixt, O., Han, S., Liao, L., Zeng, Y., Hoffmann, J., Futakawa, S., and Paulson, J. C. (2008). Sialoside analogue arrays for rapid identification of high affinity siglec ligands. J. Am. Chem. Soc. 130, 6680–6681. Bochner, B. S., Alvarez, R. A., Mehta, P., Bovin, N. V., Blixt, O., White, J. R., and Schnaar, R. L. (2005). Glycan array screening reveals a candidate ligand for Siglec-8. J. Biol. Chem. 280, 4307–4312. Cariappa, A., Takematsu, H., Liu, H., Diaz, S., Haider, K., Boboila, C., Kalloo, G., Connole, M., Shi, H. N., Varki, N., Varki, A., and Pillai, S. (2009). B Cell antigen receptor signal strength and peripheral B cell development are regulated by a 9-O-acetyl sialic acid esterase. J. Exp. Med. 206, 125–138. Carlin, A. F., Chang, Y. C., Areschoug, T., Lindahl, G., Hurtado-Ziola, N., King, C. C., Varki, A., and Nizet, V. (2009a). Group B Streptococcus suppression of phagocyte functions by protein-mediated engagement of human Siglec-5. J. Exp. Med. 206, 1691–1699. Carlin, A. F., Uchiyama, S., Chang, Y. C., Lewis, A. L., Nizet, V., and Varki, A. (2009b). Molecular mimicry of host sialylated glycans allows a bacterial pathogen to engage neutrophil Siglec-9 and dampen the innate immune response. Blood 113, 3333–3336. Chen, W. C., Completo, G. C., Sigal, D. S., Crocker, P. R., Saven, A., and Paulson, J. C. (2010). In vivo targeting of B-cell lymphoma with glycan ligands of CD22. Blood 115, 4778–4786. Chinarev, A. A., Galanina, O. E., and Bovin, N. V. (2010). Biotinylated multivalent glycoconjugates for surface coating. Methods Mol. Biol. 600, 67–78. Chokhawala, H. A., Huang, S., Lau, K., Yu, H., Cheng, J., Thon, V., Hurtado-Ziola, N., Guerrero, J. A., Varki, A., and Chen, X. (2008). Combinatorial chemoenzymatic synthesis and high-throughput screening of sialosides. ACS Chem. Biol. 3, 567–576. Collins, B. E., Blixt, O., DeSieno, A. R., Bovin, N., Marth, J. D., and Paulson, J. C. (2004). Masking of CD22 by cis ligands does not prevent redistribution of CD22 to sites of cell contact. Proc. Natl. Acad. Sci. USA 101, 6104–6109. Collins, B. E., Blixt, O., Han, S., Duong, B., Li, H., Nathan, J. K., Bovin, N., and Paulson, J. C. (2006). High-affinity ligand probes of CD22 overcome the threshold set by cis ligands to allow for binding, endocytosis, and killing of B cells. J. Immunol. 177, 2994–3003. Courtney, A. H., Puffer, E. B., Pontrello, J. K., Yang, Z. Q., and Kiessling, L. L. (2009). Sialylated multivalent antigens engage CD22 in trans and inhibit B cell activation. Proc. Natl. Acad. Sci. USA 106, 2500–2505. Crocker, P. R., and Redelinghuys, P. (2008). Siglecs as positive and negative regulators of the immune system. Biochem. Soc. Trans. 36, 1467–1471. Crocker, P. R., Paulson, J. C., and Varki, A. (2007). Siglecs their roles in the immune system. Nat. Rev. Immunol. 7, 255–266.

362


Duong, B. H., Tian, H., Ota, T., Completo, G., Han, S., Vela, J. L., Ota, M., Kubitz, M., Bovin, N., Paulson, J., and Nemazee, D. (2010). Decoration of T-independent antigen with ligands for CD22 and Siglec-G can suppress immunity and induce B cell tolerance in vivo. J. Exp. Med. 207, 173–187S1–S4. Han, S., Collins, B. E., Bengtson, P., and Paulson, J. C. (2005). Homomultimeric complexes of CD22 in B cells revealed by protein-glycan cross-linking. Nat. Chem. Biol. 1, 93–97. Heuser, J. E., and Anderson, R. G. (1989). Hypertonic media inhibit receptor-mediated endocytosis by blocking clathrin-coated pit formation. J. Cell Biol. 108, 389–400. Hinderlich, S., Berger, M., Keppler, O. T., Pawlita, M., and Reutter, W. (2001). Biosynthesis of N-acetylneuraminic acid in cells lacking UDP-N-acetylglucosamine 2-epimerase/N-acetylmannosamine kinase. Biol. Chem. 382, 291–297. Hudson, S. A., Bovin, N. V., Schnaar, R. L., Crocker, P. R., and Bochner, B. S. (2009). Eosinophil-selective binding and proapoptotic effect in vitro of a synthetic Siglec-8 ligand, polymeric 6’-sulfated sialyl Lewis x. J. Pharmacol. Exp. Ther. 330, 608–612. Jones, C., Virji, M., and Crocker, P. R. (2003). Recognition of sialylated meningococcal lipopolysaccharide by siglecs expressed on myeloid cells leads to enhanced bacterial uptake. Mol. Microbiol. 49, 1213–1225. Kaltgrad, E., O’Reilly, M. K., Liao, L., Han, S., Paulson, J. C., and Finn, M. G. (2008). On-virus construction of polyvalent glycan ligands for cell-surface receptors. J. Am. Chem. Soc. 130, 4578–4579. Kelm, S., Gerlach, J., Brossmer, R., Danzer, C. P., and Nitschke, L. (2002). The ligandbinding domain of CD22 is needed for inhibition of the B cell receptor signal, as demonstrated by a novel human CD22-specific inhibitor compound. J. Exp. Med. 195, 1207–1213. Khatua, B., Ghoshal, A., Bhattacharya, K., Mandal, C., Saha, B., and Crocker, P. R. (2010). Sialic acids acquired by Pseudomonas aeruginosa are involved in reduced complement deposition and siglec mediated host-cell recognition. FEBS Lett. 584, 555–561. Kimura, N., Ohmori, K., Miyazaki, K., Izawa, M., Matsuzaki, Y., Yasuda, Y., Takematsu, H., Kozutsumi, Y., Moriyama, A., and Kannagi, R. (2007). Human B-lymphocytes express alpha2–6-sialylated 6-sulfo-N-acetyllactosamine serving as a preferred ligand for CD22/ Siglec-2. J. Biol. Chem. 282, 32200–32207. Munday, J., Kerr, S., Ni, J., Cornish, A. L., Zhang, J. Q., Nicoll, G., Floyd, H., Mattei, M. G., Moore, P., Liu, D., and Crocker, P. R. (2001). Identification, characterization and leucocyte expression of Siglec-10, a novel human sialic acid-binding receptor. Biochem. J. 355, 489–497. Nath, D., van der Merwe, P. A., Kelm, S., Bradfield, P., and Crocker, P. R. (1995). The amino-terminal immunoglobulin-like domain of sialoadhesin contains the sialic acid binding site. Comparison with CD22. J. Biol. Chem. 270, 26184–26191. O’Reilly, M. K., Collins, B. E., Han, S., Liao, L., Rillahan, C., Kitov, P. I., Bundle, D. R., and Paulson, J. C. (2008). Bifunctional CD22 ligands use multimeric immunoglobulins as protein scaffolds in assembly of immune complexes on B cells. J. Am. Chem. Soc. 130, 7736–7745. O’Reilly, M. K., and Paulson, J. C. (2009). Siglecs as targets for therapy in immune-cellmediated disease. Trends Pharmacol. Sci. 30, 240–248. Ramya, T. N., Weerapana, E., Liao, L., Zeng, Y., Tateno, H., Yates, J. R., III, Cravatt, B. F., and Paulson, J. C. (2010). In situ trans ligands of CD22 identified by glycan-protein photocross-linking-enabled proteomics. Mol. Cell Proteomics 9, 1339–1351. Rapoport, E. M., Pazynina, G. V., Sablina, M. A., Crocker, P. R., and Bovin, N. V. (2006). Probing sialic acid binding Ig-like lectins (siglecs) with sulfated oligosaccharides. Biochemistry (Mosc) 71, 496–504. Razi, N., and Varki, A. (1998). Masking and unmasking of the sialic acid-binding lectin activity of CD22 (Siglec-2) on B lymphocytes. Proc. Natl. Acad. Sci. USA 95, 7469–7474.


363

Rempel, H., Calosing, C., Sun, B., and Pulliam, L. (2008). Sialoadhesin expressed on IFNinduced monocytes binds HIV-1 and enhances infectivity. PLoS One 3, e1967. Severi, E., Hood, D. W., and Thomas, G. H. (2007). Sialic acid utilization by bacterial pathogens. Microbiology 153, 2817–2822. Tateno, H., Li, H., Schur, M. J., Bovin, N., Crocker, P. R., Wakarchuk, W. W., and Paulson, J. C. (2007). Distinct endocytic mechanisms of CD22 (Siglec-2) and Siglec-F reflect roles in cell signaling and innate immunity. Mol. Cell Biol. 27, 5699–5710. van der Kuyl, A. C., van den Burg, R., Zorgdrager, F., Groot, F., Berkhout, B., and Cornelissen, M. (2007). Sialoadhesin (CD169) expression in CD14+ cells is upregulated early after HIV-1 infection and increases during disease progression. PLoS One 2, e257. van der Merwe, P. A., Crocker, P. R., Vinson, M., Barclay, A. N., Schauer, R., and Kelm, S. (1996). Localization of the putative sialic acid-binding site on the immunoglobulin superfamily cell-surface molecule CD22. J. Biol. Chem. 271, 9273–9280. Vinson, M., van der Merwe, P. A., Kelm, S., May, A., Jones, E. Y., and Crocker, P. R. (1996). Characterization of the sialic acid-binding site in sialoadhesin by site-directed mutagenesis. J. Biol. Chem. 271, 9267–9272. von Gunten, S., and Bochner, B. S. (2008). Basic and clinical immunology of Siglecs. Ann. NY. Acad. Sci. 1143, 61–82. Yang, Z. Q., Puffer, E. B., Pontrello, J. K., and Kiessling, L. L. (2002). Synthesis of a multivalent display of a CD22-binding trisaccharide. Carbohydr. Res. 337, 1605–1613. Yu, H., and Chen, X. (2007). Carbohydrate post-glycosylational modifications. Org. Biomol. Chem. 5, 865–872.

C H A P T E R

E I G H T E E N

Intramolecular Glycan–Protein Interactions in Glycoproteins Adam W. Barb,* Andrew J. Borgert,† Mian Liu,* George Barany,‡ and David Live* Contents 365 367 374 382 382

1. Introduction 2. O-Linked Glycoproteins 3. N-Linked Glycoproteins Acknowledgments References

Abstract Glycoproteins are a major class of glycoconjugates displaying a variety of mutual interactions between glycan and protein moieties that ultimately affect molecular organization. Modulation of the pendant glycan structures is important in tuning the functions of glycoproteins. Here we discuss structural aspects and some of the challenges to studying intramolecular interactions between carbohydrate and protein elements in several forms of O-linked as well as Nlinked glycoproteins. These illustrate the importance of the relationship of context to function in protein glycosylation.

1. Introduction The challenges of glycomics are formidable, considering the diversity of oligosaccharides that arise from the large number of constituent determinants (Cummings, 2009). In addition, carbohydrates are often found in combination with other molecular structures, further elaborating the complexity (van Kooyk and Rabinovich, 2008). Glycoproteins may represent the most diverse category of glycoconjugates, particularly in light of the estimate that over half of all mammalian proteins carry glycans (Apweiler * Complex Carbohydrate Research Center, University of Georgia, Athens, Georgia, USA Center for Magnetic Resonance Research, University of Minnesota, Minneapolis, Minnesota, USA Department of Chemistry, University of Minnesota, Minneapolis, Minnesota, USA

{ {


#


365

366

Adam W. Barb et al.

et al., 1999). Further, the glycans associated with a specific protein may vary over time with physiological states of cells and tissues (Ohtsubo and Marth, 2006). The intricate processes of posttranslational glycosylation in both assembly and remodeling of the glycans also introduce a degree of microheterogeneity in natural mature glycoproteins even when these are isolated from a single source (Rich and Withers, 2009). The complex mixtures obtained from natural sources have presented difficulties in studies of glycoproteins and in understanding the intramolecular relationships between carbohydrate and protein moieties in glycoproteins. Such associations do, however, have an impact on the distinct properties and functions of a glycoprotein, particularly their recognition, and therefore on functional glycomics. In this chapter, we address aspects of these intramolecular relationships. The increasing interest in conformational aspects of glycopeptides and glycoproteins is reflected in a recent review on the subject (Meyer and Moller, 2007). For the most part, the glycosylation modifications of proteins fall into two classes, N-linked, where the glycan is joined to the protein through the side chain amide of an Asn residue, or O-linked, using the hydroxyl of a Ser or Thr residue (Varki et al., 2009). The N-linked glycosylation occurs cotranslationally through the transfer of a preassembled common oligosaccharide core to the side chain Asn nitrogen, in the consensus sequence Asn-Xaa-Ser/Thr (where Xaa is any residue except Pro) on the growing polypeptide chain, connected through a b-GlcNAc residue. Thus, the residues closest to the linkage site are constant, even though the more distal residues can vary. O-linked glycans are assembled in a stepwise manner on the protein, and are more varied in protein sequence context, in linking glycan, and in glycan composition. Mucin glycosylation is initiated with an a-O-GalNAc (Ten Hagen et al., 2003), although the importance of Oglycans based on other linkages, such as a-O-Man (Barresi and Campbell, 2006; Chai et al., 1999) and modifications of a single b-O-GlcNAc (Whelan and Hart, 2006), has been recognized recently. Proteoglycans are also a significant class of O-linked glycoproteins, characterized by long carbohydrate chains linked to the protein backbone through a xylose, with the proportion of long carbohydrate polymers overwhelming the protein. Due to the absence of significant three-dimensional structural data for this latter class, this will not be discussed here. The significant chemical structural differences of the two major types of linkages impact the intramolecular interactions between glycan and protein components. For N-linked glycans, the point of linkage is three bonds removed from the polypeptide backbone. Additionally, the b stereochemistry of the glycosidic linkage directs the glycan away from the backbone, decreasing the contact between glycan and protein components in the immediate vicinity of the modification. Sites of N-linked glycosylation tend to be widely dispersed. In contrast, the linkage point for O-linked

Intramolecular Interactions

367

glycans is closer to the backbone, only two bonds removed, and for those with a-glycosidic linkages, the initiating residue lies proximal to the peptide backbone, facilitating intimate interactions. Additionally, for mucin-like Olinked glycosylation there are often numerous and neighboring sites of modification, which can amplify the effects of the glycan–polypeptide intramolecular backbone interactions. The recognized high-resolution structure determination techniques for identifying intramolecular interactions are crystallography and nuclear magnetic resonance (NMR). Generally, glycoyslation is considered detrimental to protein crystallization (Lee et al., 2009), and often efforts are made to remove or remodel glycans to either eliminate this concern, or, particularly in the case of N-linked structures, to minimize the heterogeneity by trimming back the glycans (Lee et al., 2009; Rich and Withers, 2009). For mucins, the high density of glycosylation would only further compound this, and may explain the lack of crystallographic data on these molecules. Solution state NMR, though, has proven to be an effective tool in examining carbohydrate structures on glycoproteins (Meyer and Moller, 2007). The method also offers a distinct advantage in directly accessing the dynamics of the structures which reflect both the intramolecular and intermolecular interactions of glycoproteins.

2. O-Linked Glycoproteins The significance of intramolecular interactions in affecting the properties of native mucin O-linked glycoproteins at a global scale was realized early on from a variety of biophysical studies. Extended structures were visualized in electron micrographs of glycosylated mucin domains (Rose et al., 1984). Comparison of glycosylated and deglycosylated mucins using NMR (Gerken and Jentoft, 1987; Gerken et al., 1989) and light scattering techniques (Shogren et al., 1989) demonstrated the organizational consequences of the a-O-GalNAc modification, insofar as an extended and more rigid organization was observed. While the structural properties of mucin domains differ from those of globular proteins, they do present ordered structures that dictate the dispositions of their glycans. Intramolecular interactions are responsible for this organization, with significant implications for recognition of their glycans. This affects cellular signaling through cellsurface glycoproteins, with CD43 and CD45 being two examples (Garner and Baum, 2008). Studies on natural material have provided insights into the larger scale conformational aspects of mucins. However, the intrinsic natural high molecular weight and microheterogeneity render such material problematic for high-resolution structural analysis that would enable elucidation of the detailed intramolecular interactions giving rise to the conformational features. An exception to this is the highly regular fish antifreeze

368

Adam W. Barb et al.

mucin glycoprotein (AFGP), which has a repeating triad of amino acids, predominantly AAT with some PAT, and with the T residues glycosylated. It has been possible to isolate a fraction of modest molecular weight from natural AFGP preparations, making its study tractable (Bush and Feeney, 1986; Lane et al., 1998). Given the general complications associated with accessing a broader range of structures from natural material, peptide synthesis methodology has emerged as an attractive and important alternative (Buskas et al., 2006). This has permitted the preparation of glycopeptides, particularly those bearing short glycans, with a wide variety of defined amino acid sequences and patterns of glycosylation. The extension of conventional peptide synthesis methodology to O-linked glycopeptides requires additional considerations in preparing glycosylated building blocks before assembly (Buskas et al., 2006), and in the deprotection of the sugar hydroxyls at the final stages of synthesis. These considerations have been successfully addressed (e.g., Liu et al., 2005, 2008), and recent advances employing microwave-assisted solid-phase peptide synthesis techniques are further enhancing glycopeptide synthesis efficiency (Matsushita et al., 2006). With the findings that the initial S- or T-linked GalNAc residues dominate the intramolecular interactions organizing mucin glycopeptides (Coltart et al., 2002), as discussed below, this synthetically most accessible and simplest mucin form is an effective model for investigating the intramolecular interactions and the core glycopeptide scaffold. While the minimal S/T-a-O-GalNAc element, or Tn antigen, is not normally revealed in humans, it is found on cell surfaces of tumor cells associated with aberrantly glycosylated mucins, and is correlated with a poor clinical prognosis. This has generated interest in the Tn glycopeptides (Springer, 1997). Mucin structures with more complex glycans have been successfully prepared using chemical and/or enzymatic approaches for elaborating the carbohydrates (Matsushita et al., 2006; Tarp et al., 2007). Although many conformational questions regarding mucin motifs can be addressed with modest sized fragments, application of native chemical ligation (NCL) methods has offered significant advances for exploring larger segments (Kan and Danishefsky, 2009; Payne and Wong, 2010). In considering a glycopeptide-based strategy for studying mucin glycoproteins, the potential absence of native tertiary interactions in studying short segments is a concern. Mucins, however, adopt extended conformations which preclude long-range tertiary interactions with sequentially remote regions in the native glycoprotein. Therefore, the same local interactions dominate the conformations of the motifs, both as parts of the native structures and as isolated short segments of these glycoproteins. Thus, mucin glycopeptides are expected to provide realistic structural models for components of the native glycoprotein. As elaborated below, these short segments display the features that are consistent with


369

those of the global arrangement of glycosylated mucins. Thus, glycopeptides are valuable tools for studying O-linked structures. NMR studies on AFGP provided interatomic distance information from a number of nuclear Overhauser effect (NOE) interactions and bond angles based on proton J couplings. These, along with conformational energy calculations, provided a model for the glycosylated AT*AA sequence in the AFGP, revealing an extended and stabilized structure for the motif consistent with the larger organization of mucins (Lane et al., 1998). The AFGP studies also provided evidence for the existence of a hydrogen bond between the GalNAc amide and the carbonyl of the Thr residue to which it is attached (Mimura et al., 1992). Extension to examples with clusters of immediately adjacent sites of glycosylation, based on synthetic segments of mucins from glycophorin (Schuster et al., 1999) and MUC7 (Naganagowda et al., 1999), have also been reported. In these latter cases, a twisted extended structure of the backbone for the glycosylated forms was found, with an indication that the peptide backbone structure showed characteristics of a polyproline II helix. Combined synthetic and NMR efforts facilitated a systematic analysis of a series of glycopeptide constructs based on an Nterminal motif from the cell surface glycoprotein CD43, S*T*T*AV, where the asterisks denote glycosylation (Coltart et al., 2002). Three constructs were prepared with conventional mucin a-linked glycans of increasing complexity, GalNAca (Tn), Galb1-3GalNAca (T), and Galb1-3 (Neu5Aca2,6)GalNAca (ST). The ST glycan has been associated with CD43 in acute myelogenous leukemia (Fukuda et al., 1986). For the glycopeptide core, consisting of the GalNAc residues and the peptide, numerous NMR NOE contacts, including between the proximal sugar and peptide, were observed. These provided internuclear distance relationships (Coltart et al., 2002). The distances, along with J coupling parameters that relate to bond torsion angle, provided an extensive set of constraints for structure calculations. Key features, largely invariant with the size of the attached glycan, are immediately evident from these parameters. The HN to Ha 3J couplings are large, supporting an extended backbone structure. The numerous NOE interactions between the proximal GalNAc and the peptide backbone show interactions for this sugar, while there are a lack of NOE interactions between the peripheral sugar residues and the core glycopeptide, indicating that they are not in intimate contact with the peptide. Using the NMR constraints, structural refinement calculations were carried out with the Xplor program (Schwieters et al., 2006), and resulting coordinates for structures fitting the experimental constraints are available from the Protein Data Bank database (http://www.rcsb.org/pdb) entry 1kyj. The tightly clustered family of accepted structures is consistent with an extended, stable, and well-ordered structure. An important feature of Thr residues with a-O-GalNAc attached is the small 3J coupling between the Thr Ha and Hb. This limits the range of allowed torsion angles between

370

Adam W. Barb et al.

these C–H bonds to values either close to 90 or 270 , with the former confirmed by NOEs. The structures showed a spacial relationship for the Nacetyl amide of the three glycosylated residues with both the glycosidic and carbonyl oxygens of their respective amino acid residues that indicate hydrogen bonding interactions. The existence of such an interaction is independently supported by the slow exchange rates for the N-acetyl amide protons (Coltart et al., 2002), also seen by others (Lane et al., 1998). The structure also reveals hydrophobic interactions between the N-acetyl methyl groups and side chain methyl groups on the iþ2 reside that would aid in promoting and propagating the extended structure down the polypeptide chain. In the absence of glycosylation, NMR results for the naked peptide itself suggest considerable conformational flexibility, typical of short peptides in general. NMR relaxation parameters, NOE, T1, T1r, and T2, are linked directly to molecular dynamics (Cavanaugh et al., 2007; Ishima and Torchia, 2000), and these have been used to verify the conclusions above by 13C relaxation measurements of the tri-Tn-S*T*T*AV glycopeptide listed in Table 18.1 (D. Live, unpublished results). The S2-order parameters extracted from these measurements were derived with the extended model free analysis using the program Modelfree (Mandel et al., 1995). S2 has a maximum value of 1 when there is no local motion, and values of 0.6 can be related to angular variations of only 30 (Ishima and Torchia, 2000), indicating that at most there is only a limited range of segmental motion for this glycopeptide core, in keeping with a well-defined conformation. Such effects have been observed in mucin glycopeptides by others (Grinstead et al., 2002). These data provide direct evidence that the glycosylated core, comprising the three glycosylated amino acids and their attached GalNAcs, largely behave as a single conformational entity. The existence of well-defined structure has important implications, since this establishes the relative orientations of the glycans: a feature which is relevant to recognition of mucin motifs on specific glycoproteins. Table 18.1 13C NMR relaxation parameters, T1 (s), T2 (s), and NOE measured at natural abundance using proton detected 2D 1H-13C experiments (Cavanaugh et al., 2007), and the derived order parameter S2 for sites in the (Tn)3 S*T*T*AV moleculea

T1 T2 NOE S2 a

S1a

T2a

T3a

A4a

T2b

T3b

S1GC1

T2GC1

T3GC1

0.41 0.25 1.53 0.69

0.36 0.20 1.35 0.83

0.37 0.21 1.36 0.81

0.43 0.27 1.56 0.61

0.37 0.16 1.29 0.88

0.36 0.19 1.32 0.90

0.41 0.26 1.54 0.67

0.40 0.21 1.39 0.76

0.41 0.24 1.43 0.71

S1GC1, T2GC1, and T3GC1 refer to the anomeric carbon sites on the GalNAc residues associated with the respective amino acid residues. Measured at 18.8 T, 800 MHz 1H frequency.

371


Additionally, NMR residual dipolar couplings (RDCs) (de Alba and Tjandra, 2002) for this construct have been determined (D. Live and A. Borgert, unpublished results) in a weakly aligning cetylpyridinium bromide (CPBr)/hexanol/sodium bromide liquid crystalline medium (Barrientos et al., 2000). These provided additional structural constraints, independent of the earlier data, and were consistent with the original structure. Further refinement of the structure with these values resulted in only minor structural adjustments (Fig. 18.1). The consistency of these data with the earlier structure, and the sensitivity of the RDCs to motion over a broad range of frequencies, further support the contention that this glycopeptide does not undergo major conformational fluctuations. The effects of variation in local density of glycosylation on conformation have been explored in a MUC2-derived sequence, PTTTPLK, which is known to be a substrate for polypeptide GalNAc transferases (Takeuchi et al., 2002). Constructs were synthesized (Liu et al., 2005) with all permutations of the pattern of glycosylation by a-O-GalNAc on the threonine residues, and NMR studies were carried out. The structures of these molecules were computed from NOE and J coupling NMR experimental restraints (A. Borgert, M. Liu, G. Barany, and D. Live, unpublished results). For those with two or three substitutions, RDC values in didodecyl/ dihexyl-phosphatidylcholine media (Ottiger and Bax, 1999) were determined and used as well. In this triplet motif, the organization around the respective individually glycosylated Thr residues are largely unchanged relative to the peptide backbone, as neighboring sites are glycosylated (Fig. 18.2). With this increasing density of glycosylation, the extent of segmental motion becomes more restricted, reflecting the increased carbohydrate–peptide interactions. Interestingly, the core structure with all three Thr residues glycosylated overlays the analogous glycosylated S*T*T* segment of the earlier construct quite well, with preservation of the associated interactions, suggesting a consistent triplet structural motif and the relative lack of sensitivity to Ser versus Thr residue at the N-terminal position. T3

S1 T2

Figure 18.1 Overlay of the closest to the average of the families of structures of the S*T*T* segment of (Tn)3 S*T*T*AV determined without RDC constraints, sticks, and with RDC constraints included, lines.

372

Adam W. Barb et al.

PT*TTPLK

PT*TPLK

PTTT*PLK

PT*T*T*PLK

Figure 18.2 The closest to the average for the family of structures for each of the three PTTTPLK constructs with a single site of a-O-GalNAc glycosylation, and the one with all three sites glycosylated.

There have been several structural studies on peptides and mucin glycopeptides from the tandem repeat of MUC1 (Dziadek et al., 2006; Grinstead et al., 2002; Kirnarsky et al., 2000). This glycoprotein has been the focus of attention since it is overexpressed with aberrant glycosylation, particularly the Tn epitope, in tumor cells (Dziadek et al., 2006). The most detailed study is an NMR analysis of a glycosylated form of the full tandem repeat element GSTAPPAHGVTSAPDTRPAP with a single ST epitope at Thr11, as recently reported (Dziadek et al., 2006). The NOE contacts and 3J coupling parameters, as well as structural features in the vicinity of the single glycoyslation site, are quite consistent with those previously published findings for glycosylated amino acids in the clustered S*T*T*AV structure (Coltart et al., 2002). These include contacts between the N-acetyl methyl group and the iþ2 Ala Me, and a 2.5 A˚ distance between the N-acetyl N and the Thr11 carbonyl oxygen, supporting a direct intramolecular interaction through a hydrogen bond. A number of mucin structures reported are consistent with direct hydrogen bonding between a GalNAc amide and the peptide backbone, for example, AFGP (Mimura et al., 1992), MUC1 glycopeptide (Dziadek et al., 2006), MUC2 glycopeptides, and S*T*T*AV (Coltart et al., 2002) glycopeptide. However, on the basis of NMR and molecular dynamics simulations of isolated C- and N-capped a-O-GalNAc-Ser and Thr amino acids and dipeptides composed of Ser and Thr residues, questions have been raised about their existence in other contexts (Corzana et al., 2006b, 2007,


373

2009). In these instances, solvent-mediated hydrogen bonds have been proposed. The orientations and interatomic distances of the hydrogen bonding functional groups found in the larger glycopeptides would seem to preclude insertion of a water molecule, as do the reduced amide exchange rates observed in the larger systems. While direct hydrogen bonding may not be the case in all instances of larger glycopeptides, it would appear that the amino acid residues immediately adjacent to the glycosylation sites can play important roles both in restricting the local conformation and on local solvation. Thus, very small models may be of too limited size to accurately reflect all of the interactions between carbohydrate and peptide components in more native-like mucin systems. The differential response to a versus b stereochemistry at GalNAc glycosidic linkages to the amino acid can be deduced directly from comparison of the (a-T)3 S*T*T*AV and the (b-T)3 S*T*T*AV constructs (Coltart et al., 2002), and provides further evidence for the role of specific interactions on the mechanism of structural stabilization. The reduced number of glycan to peptide NOEs, the 3J coupling values, and poorer dispersion of the peptide amide chemical shifts in the case of the b-linkage relative to the a-linkage, as well as reduced amide proton exchange lifetimes, suggest diminished interactions between the carbohydrate and peptide components. These parameters are further consistent with dynamic averaging of multiple conformational states. The increased conformational lability is consistent with the fact that the b-linkage redirects the GalNAc residue away from the peptide, largely disrupting the interactions among functional groups of the two moieties found in the a-linked forms. While only the a-linkage occurs in conventional mucins, an analogous b-O-linked GlcNAc modification occurs naturally as a sparsely distributed transient regulatory modification on some cytosolic proteins (Whelan and Hart, 2006). Investigation of N- and C-terminal capped Ser/Thr-b-O-GlcNAc models are consistent with the sugar residue being oriented away from the backbone, and with increased flexibility (Corzana et al., 2006a). This modification has also been studied on a synthetic b-O-GlcNAc glycopeptide from RNA polymerase II, where a site of b-O-GlcNAc attachment has been identified (Simanek et al., 1998). NOE interactions with the peptide backbone are lacking here, consistent with the GlcNAc projecting away from the peptide portion. However, it was found that glycosylation did induce a propensity for a turn in the backbone. Analysis showed clustering into one of two major families. Together, these are consistent with greater molecular flexibility and a more exposed carbohydrate. This GlcNAc modification on two related peptide sequences has been examined in complex with an MHC molecule by crystallography (Glithero et al., 1999). In support of glycopeptide flexibility, the crystal structure for one of the glycopeptides in the complex shows evidence that two different rotamers in the GlcNAc glycosidic linkage can occur. Thus, whereas the

374

Adam W. Barb et al.

a-linkage positions functional groups to promote conformational and chemical stability of mucin glycopeptides, nature has employed the less constrained b-linkage in a more labile regulatory application that can modulate phosphorylation. Another O-linked modification for which glycan–polypeptide interactions have ramifications is that of a-O-mannose-linked glycans. Recently, this has been identified in mammalian glycoproteins, with a-dystroglycan being the only well-characterized example (Barresi and Campbell, 2006). There is evidence for other glycoproteins modified in this way as well (Chai et al., 1999). This modification has taken on considerable importance, since aberrations in the O-Man-linked glycan have been related to forms of muscular dystrophy. Electron micrographs show the central mucin-like region of this glycoprotein is extended (Brancaccio et al., 1995), as in conventional mucins with a-O-GalNAcs. This posed the question of whether an a-O-Man modification can support an extended arrangement, and if so, through what mechanism. To investigate this, a-O-Man glycopeptides based on a-dystroglycan sequences were synthesized and studied by NMR (Liu et al., 2008). The resulting structures show considerable disorder, relative to the same sequence with a-O-GalNAc, particularly in the arrangement of the pendant sugars (A. Borgert, M. Liu, G. Barany, and D. Live, unpublished results). This can be rationalized in the context of the earlier findings that specific sugar functional groups, notably the N-acetyl modification, and their location are important for interactions that stabilize extended structures. In their absence, glycosylation should not lead to the extended arrangement. While emphasis has been placed on the presence of the a-O-Man-linked glycans because of their identification with biological function, investigations of the central mucin-like region of a-dystroglycan (Sasaki et al., 1998) have noted that there are also a number of coexisting a-OGalNAc-linked glycans in this region. From the inability of the a-O-Man modifications to stabilize the observed extended structure, it appears that the intramolecular interactions involving these a-O-GalNAc-linked glycans are important in imparting the structural stability to the extended arrangement. This would be important for its mechanical function in tissue organization, as well as for the appropriate presentation of the O-Man glycans to receptors.

3. N-Linked Glycoproteins Implicit in one of the functions of N-linked glycans, the participation in protein folding quality control (Varki et al., 2009), is intramolecular interaction between the glycan component and the polypeptide chain. As with O-linked species, N-linked glycoproteins display microheterogeneity, a feature that similarly complicates their study. In view of the significance of


375

tertiary interactions in globular glycoproteins with N-linked glycosylation, the rationale for studying isolated glycopeptide segments as models in these latter cases is less well founded. Rather, to accurately describe conformational properties, there is the more demanding requirement for the full glycoprotein, and particularly, for homogeneous glycoprotein samples. This has been approached by both chemical and biological strategies. From a purely chemical perspective, NCL strategies have been extended to the assembly of defined synthetic fragments, both peptide and glycopeptide, into full glycoproteins. This has shown impressive results and considerable promise (Kan and Danishefsky, 2009; Payne and Wong, 2010; Yamamoto et al., 2008). A variation on this theme involves bacterial expression of a protein segment that is then linked, using NCL, to a synthetic glycopeptide, completing the glycoprotein (Piontek et al., 2009). Chemoenzymatic synthesis has been used as well (Wang, 2008). Biological approaches have used wild-type cells, or those with modifications in the glycosylation machinery, often combined with in vitro enzymatic remodeling of the product glycoproteins (Chang et al., 2007; Lee et al., 2009; Li et al., 2001; Lustbader et al., 1996; Rich and Withers, 2009; Schwarz et al., 2010; Slynko et al., 2009), to achieve the final preparations. In the biosynthetic approach, it is necessary to consider that the pendant glycans installed can be organism specific, which can affect the detailed characteristics of the product, as well as choice of expression system. An additional factor in the prospects for success of the various approaches is that, in the absence of some or all of the attached glycoforms, the underlying protein, or even the partially glycosylated form, may not properly fold or may aggregate (Lee et al., 2009). For Chinese hamster ovary (CHO) cells, there are a variety of available glycosylation mutants (North et al., 2010) that can aid in producing more homogeneous glycoproteins on their own, or provide material that is more readily remodeled. Human embryonic kidney (HEK) cells offer promise in expressing glycoproteins bearing human glycoforms (Lee et al., 2009). The relative ease in working with yeast cultures has made them, and in particular Pichia pastoris, popular organisms for producing glycoproteins (Rich and Withers, 2009). The high mannose glycans it installs can be trimmed to the level of the first N-acetylglucosamine residue by treatment with endoglycosidase H; the GlcNAc-Asn site then serves as a site for enzymatically reattaching a desired glycoform using methods recently reported (Rich and Withers, 2009; Wang, 2008). For glycoproteins isolated directly from their natural sources, complex-type glycans may also be trimmed in a similar manner with endoglycosidase D or S (Allhorn et al., 2008; Yamaguchi et al., 2006). Exposed complex-type glycans may sometimes be completely removed from native glycoproteins using PNGase F (Plummer and Tarentino, 1991). Heterogeneity of mature glycoproteins may be dramatically reduced by sequentially trimming the glycan termini using a host of

376

Adam W. Barb et al.

obligate exoglycosidases, including a neuraminidase (Clostridium perfringens), a b1-4-galactosidase (Bacteroides fragilis), an N-acetylglucosaminidase (Xanthomonas manihotis), and an a1-3-mannosidase (X. manihotis; New England Biolabs). Following digestion with a1-3 mannosidase, an a1-6 mannosidase (X. manihotis) will remove the a1-6-linked mannose residue. It is important to note that this enzyme does not act on branched substrates; therefore, the a1-3-linked mannose residue must be removed first (New England Biolabs; A.W. Barb and J.H. Prestegard, unpublished data). Glycosidases are commonly inhibited by steric interactions with the polypeptide. This is particularly so with endoglycosidases which are often ineffective towards glycans buried or confined by protein tertiary structure (Blanchard et al., 2008). Steric effects appear to be less of an obstacle for exoglycosidase-catalyzed digestions. Incomplete or heterogeneous glycan termini may be remodeled using glycosyltransferases and their corresponding sugar nucleotide substrates. This strategy has been employed to incorporate NMR-active isotopes into mammalian-expressed glycoproteins (Barb et al., 2009; Macnaughtan et al., 2008; Yamaguchi et al., 1998). Commonly used enzymes are the human a2-3 and a2-6 sialyl-transferases and the bovine b1-4 galactosyltransferase (Chung et al., 2006; Krapp et al., 2003; Raju et al., 2001; Scallon et al., 2007). In most cases, these reactions may be run to completion with each glycan terminus modified. There have been reports of incomplete sialylation of the occluded N-glycans of the IgG Fc fragment discussed below (Barb et al., 2009; Kobata, 2008; Raju et al., 2001); some of this effect is attributable to the slow sialylation of the a1-6Man-linked branch by the human a2-6 sialyl-transferase (Barb et al., 2009). Glycans on intact glycoproteins (Fig. 18.3A), whether as originally isolated or after remodeling, can be characterized by matrix-assisted laser desorption ionization (MALDI)-MS or electrospray ionization (ESI)-MS methods (Gong et al., 2009). However, minor glycoforms may be poorly resolved by MS. For proteins with multiple glycosylation sites, deconvoluting the contributions from different glycans and peak overlap can be challenging. As an alternative approach, liberated, permethylated glycans can be analyzed with MALDI-MS (Anumula and Taylor, 1992). These techniques do not perform well in distinguishing configurational isomers, which generally require releasing glycans followed by derivatization and HPLC-based coelution (Holland et al., 2002; Rice, 2000; Takahashi et al., 1995). A 1D NMR-based approach has been described that can quickly differentiate isomers, but this is limited with heterogeneous samples and is fundamentally insensitive (Vliegenthart et al., 1983). Because of the dynamic features of N-glycans, even when crystallography of glycoproteins with extended oligosaccharides is successful, the electron density for portions of the carbohydrate may be unresolved. Furthermore, even if their electron density is observed, identification of

377


A

a1-2 a1-2

a1-2

a1-3

B

a1-3

b1-4

a1-6

a1-2

a1-6 b1-4

a1-6

a2-3 b1-4

b1-2

a1-3

a1-3 b1-4 a1-6

High mannose

a1-6 b1-4

b – GlcNAc

Hybrid

a1-6

a2-3 b1-4

OH H4 H6 O HO

a1-3

NAc

Asn H1

a1-6

b1-4

a1-6

b–GlcNAc

N H3

b1-2

H6′ O H2 H5

b1-4

Complex

Figure 18.3 Panel (A): Types of carbohydrate structures and common linkages that are N-linked to glycoproteins. Note the a1-6 and a1-3-linked mannose residues which also specify the terminal branches of the complex-type biantennary glycan. Gray squares represent N-acetylglucosamine (GlcNAc), dark gray circles mannose, open triangles fucose, open circles galactose, and black diamonds N-acetylneuraminic acid residues. Panel (B): Structure of a core b-GlcNAc residue depicted as the first residue of an N-glycan with a linkage to the asparagine residue and a linkage to the second bGlcNAc residue of the chitobiose core. Note the hydrophobic surface of this residue formed by the H1, H3, H5, H6, and H60 protons.

the actual sugars represented by a particular region of electron density may be uncertain (Chen et al., 2005). Unlike for amino acids, information from other sources may be insufficient to unequivocally know the glycan sequence. It has been noted that, historically, there are a number of errors in the carbohydrate portion of X-ray determined glycoprotein structures deposited in the PDB, although efforts are being made to correct these (Lutteke, 2009). The dynamics of the N-linked glycans puts NMR at a significant advantage in monitoring the characteristics of oligosaccharide chain conformation and interactions, not only because it is better suited for describing dynamics, but also because the experiments are done in solution. Hence, effects of crystal packing on the glycans, which largely extend from the protein surface, do not bias the results. NMR also provides explicit assignment of each sugar residue for which shift assignments can be made. While the addition of carbohydrate adds complexity to the molecules, crosspeaks from the carbohydrate components in 2D or 3D 1H–13C maps fall in regions where they can readily be resolved from those of the protein component (de Beer et al., 1994). NMR studies of these generally large glycoprotein systems can benefit from the incorporation of 13C and 15N labeling that has proven so valuable in protein structural studies. The labeling can be done either uniformly, segmentally, or on a residue (either amino acid or sugar) specific basis, using P. pastoris, CHO (Lustbader et al., 1996), HEK (Liu et al., 2007), and bacterial expression systems.

378

Adam W. Barb et al.

A survey of glycoprotein structures has found that N-linked glycans occur in a wide variety of protein secondary structural elements and surface geometries (Petrescu et al., 2006). Linked Asn residues have been observed in several topologies, ranging from the bottoms of deeply recessed concave surfaces to steeply convex surfaces (Petrescu et al., 2004). These observations argue against the presence of a dominant single strong and specific interaction between proximal residues and the proteins, and are consistent with the initial sugar residues being oriented largely out of the way of the protein backbone and interacting comparatively weakly with it. The most frequently observed contacts of the N-glycan with the polypeptide are to the first N-acetylglucosamine residue, and are variable. This residue is often found near a hydrophobic surface or pocket with the H1– H3–H5 face of the residue (Fig. 18.3B) lying against it. The N-acetyl methyl can also make hydrophobic contacts with the surface by occupying a shallow pocket. Indeed, aromatic and proline residues are enriched around N-X-T/S sequons (Petrescu et al., 2004), and are most likely to interact with the GlcNAc, as observed by X-ray crystallography. The N78 glycan of the a-subunit of human chorionic gonadotropin (pdb 1hd4) typifies this arrangement, where the H1–H3–H5 face is against proline and valine residues, and the methyl is in a pocket formed by valine, isoleucine, and alanine residues (Erbel et al., 2000). Hydrogen bonds to the carbonyl oxygen of the N-acetyl are also observed, as in the Epstein-Barr virus major envelope glycoprotein gp350 (pdb 2h6o) at N195 where the oxygen atom is in position to form a hydrogen bond with a backbone NH of G289 (Szakonyi et al., 2006). In many cases, however, there are no clear contacts between the first residue and the polypeptide. Similar interactions are observed, though less frequently, for the second N-acetylglucosamine residue in the glycan. In theory, polar contacts with any sugar hydrogen bond donor or acceptor may occur, although most observed interactions involve the C6 OH and the N-acetyl carbonyl oxygen atom. For example, the O6 from the second N-acetylglucosamine residue of the IgA Fc N263 glycan (pdb 1ow0) is in position to form a hydrogen bond to the side chain oxygens from D255 (Herr et al., 2003). Sugar ring proton hydrophobic contacts are likewise observed, including N-acetyl methyl–hydrophobic interactions as exemplified by the human FcaRI structure (pdb 1ow0) with the methyl from the second N-acetylglucosamine on the N58 glycan packed between side chains from Glu and Tyr residues (Herr et al., 2003). The dearth of coordinates beyond the first two GlcNAc residues indicates that there are few stable interactions between the distal carbohydrate residues and the polypeptide. There are notable exceptions with extensive glycan coordinates in the Protein Data Bank, although it is important to note that even when glycans are observed in X-ray crystallography, structures may be influenced by crystal packing (Chen et al., 2005), and may not


379

represent the ensemble of solution conformations available to the N-glycan. Where coordinates are available, glycans do not typically interact tightly with the polypeptide. Unlike high affinity interactions, N-glycan–polypeptide interactions lack a high degree of surface complementarity, buried hydrophobic surfaces, and extensive hydrogen bonding. NMR studies on some N-linked glycosylation sites show NOE contacts between the proximal sugars and the Asn side chain (Erbel et al., 2000; Slynko et al., 2009; Wyss et al., 1995), as well as those of neighboring side chains, indicating proximity of this sugar residue to the amino acid side chains. This can vary even among sites in a single glycoprotein. For human chorionic gonadotropin, glycan–protein NOEs for the glycan at the N52 site are largely lacking, but are present at the N78 site (Erbel et al., 2000; Weller et al., 1996). It is also noted in these glycoproteins that the mobility of the GlcNAc is more restricted than the peripheral sugar residues, implying interactions of some nature. In CD2, even the first GlcNAc has several NOE contacts with the K61 side chain, four residues removed from the glycosylation site, and has a stabilizing effect on protein attributed to defusing the locally high concentration of positive charges on the protein surface (Hanson et al., 2009; Wyss et al., 1995) particularly associated with K61. Nonetheless, using wild-type and K61A mutants of CD2, it has been shown that, independent of surface charge, the presence of the first sugar is an important contributor to protein folding and stability (Hanson et al., 2009). Indeed, trimming the N-glycan to the first N-acetylglucosamine residue is a strategy used for structural biology applications that often maintains protein stability, while substantively reducing conformational and configurational heterogeneity (Chang et al., 2007). Although both hydrophobic and polar distal residue–polypeptide interactions are observed, hydrophobic interactions probably dominate the stabilizing and enhanced folding benefits of N-glycans. It is notable that many carbohydrate residues have defined hydrophobic faces, as mentioned above. b-Galactose and b-mannose residues are examples where the H1– H3–H4–H5–H6–H60 (Fig. 18.3) and H1–H2–H3–H5–H6–H60 faces, respectively, are markedly hydrophobic in nature. It is likely that these faces cover hydrophobic polypeptide surfaces, although these may not always be obvious due to the highly dynamic nature of the distal portion of N-glycans. This is demonstrated in erythropoietin (EPO), a glycosylated peptidic growth factor, where glycosylation shields hydrophobic patches that are otherwise accessible to promote aggregation (Cheetham et al., 1998; Narhi et al., 1991; Toyoda et al., 2000, 2002). In a structure of the human Zn-a2-glycoprotein (pdb 1zag), the hydrophobic face of the b-mannose residue is proximal to a Tyr side chain (Sanchez et al., 1999). The H3ax–H4– H6 face of N-acetylneuraminic acid has also been observed in contact with hydrophobic residues, and was observed packed against a Trp side chain in the structure of fibrinogen (pdb 3ghg). These results are unusual for

380

Adam W. Barb et al.

resolving coordinates for a large glycan with seemingly few conformationstabilizing contacts (Kollman et al., 2009). The Fc fragment of immunoglobulin G (IgG) has extensive distal glycan–polypeptide interactions which affect Fc receptor binding and directly impact human disease. IgG is a 150 kDa protein with two antigen binding Fab fragments and one Fc fragment (Fig. 18.4). The Fc fragment mediates interactions between IgG and immune system components, including the Fcg receptors, and the C1q component of complement, among others (Roitt et al., 2001). Even separated from the Fab fragments, the Fc fragment maintains its full strength in these interactions. The heavy chain of IgG has one conserved N-glycosylation site at N297, with a complex-type, biantennary glycan that generally varies in the amount of terminal galactose when purified from the serum of healthy individuals (Arnold et al., 2007). In rheumatoid arthritis patients, the amount of terminal galactose is inversely proportional to the severity of the disease (Parekh et al., 1985). X-ray diffraction-derived structural models of the Fc fragment are unusual in that nearly the entirety of the glycan is resolved and occupies the space between the two Cg2 domains (pdb 1fc1; Deisenhofer, 1981). The position of these glycans in the polypeptide interstitial region likely accounts for the observed regular structure. The residues on the a1-6Manlinked branch of the glycan are stabilized through hydrophobic and polar interactions with the surface of the Cg2 domain. The H1–H3–H5 face of the first GlcNAc residue makes some hydrophobic contacts with V264, and the N-acetyl carbonyl oxygen atom hydrogen bonds to the D265 side chain. H4, H6, and H60 of the second residue make hydrophobic contacts, and the

Fab

Fab Light chain

Heavy chain Asn 297

Hinge Cg2

Cg2

Fc Cg3

Cg3

Figure 18.4 Schematic of IgG showing the Fc fragment, the position of the conserved N-glycan, and the relative orientations of the Cg2 and Cg3 domains.


381

N-acetyl carbonyl oxygen atom is in position to hydrogen bond with the R301 side chain NH. As is commonly observed for N-glycans with a core fucose residue, no contacts between the fucose and the protein surface are observed in these crystal structures. The b-mannose residue makes extensive contacts with its H1–H2–H3–H5 face to F241. From this point, the a1-3Man branch of the glycan extends into the dimer interface, away from the polypeptide surface, and the a1-6Man branch follows the contour of the Cg2 domain. The a1-6Man N-acetylglucosamine residue makes extensive contacts through its H1–H3–H5 face to F243. The K246 amine is in position to bind O4 of GlcNAc and the anomeric oxygen of galactose on the a1-6Man branch. An NMR study of the Fc-conjugated glycan revealed a broader 13C line width for the anomeric carbon of galactose on the a1-6Man-linked branch, when compared to the same carbon on the a1-3Man-linked branch (Yamaguchi et al., 1998). This offers evidence in support of the hypothesis that the line widths from the a1-6Man-linked branch are broadened due to a more intimate interaction with the protein surface, restricting its motion relative to the a1-3Man-linked branch. Mutating residues along this binding interface, including F241, F243, V264, D265, and R301, dramatically increased the amount of galactose and sialic acid containing termini (Lund et al., 1996) and indicated a greater accessibility of the glycan termini to the galactosyl- and sialyl-transferases in the Golgi. The a1-3Man-linked branch residues interact with each other across the dimer interface primarily through hydrophobic interactions (Deisenhofer, 1981). The composition of the N-glycans is intimately linked to Fc structure and function. The distance between the two Cg2 domains decreases upon truncation of the glycan (Krapp et al., 2003), and backbone amide chemical shift changes at the Cg2/Cg3 interface indicate a structural rearrangement consistent with this type of movement (Yamaguchi et al., 2006). Based on the position of the glycans, it is likely that the size of the Fc glycan is a principle determinant in the spacing of the two Cg2 domains. The IgG Fc glycan composition also affects binding to the C1q component of complement (Leatherbarrow et al., 1985) and the low affinity (low mM) Fc receptors, FcgRII and FcgRIII, in contrast to the Fca receptor–IgA Fc interaction which is insensitive to glycan composition or absence of sugar (Gomes et al., 2008). Deglycosylated IgG Fc shows no measurable affinity towards either FcgRIIb or FcgRIII (Mimura et al., 2001; Yamaguchi et al., 2006). Fc with the glycans truncated to the mannose residues shows restored binding affinity for both receptors, when compared to deglycosylated forms. The GlcNAc and galactose residues slightly affect these interactions, though differently for FcgRII and FcgRIII. It is interesting to note that sialylation decreased the strength of the Fc–FcgRIIb/FcgRIII interaction roughly 10fold (Kaneko et al., 2006). One explanation is that the Fc glycans, residing between the heavy chain monomers, tune the interaction with the FcgRs

382

Adam W. Barb et al.

through bulk effects alone. Based on the structure of the Fc–FcgRIII complex, the glycans do not bind to the receptor directly (Radaev et al., 2001). However, one can speculate that specific a1-6Man-branch–Fc polypeptide interactions, rather than bulk effects, are critical for proper IgG Fc structure, dynamics, and function.

ACKNOWLEDGMENTS This work was supported by grants from the National Institutes of General Medical Science GM066148 (D.L.) and the National Institutes of Arthritis and Musculoskeletal Diseases AR056055 (D.L.), and a fellowship from the National Institutes of Arthritis and Musculoskeletal Diseases F32AR058084 (A.B.).

REFERENCES Allhorn, M., Olin, A. I., Nimmerjahn, F., and Collin, M. (2008). Human IgG/Fc gamma R interactions are modulated by streptococcal IgG glycan hydrolysis. PLoS ONE 3, e1413. Anumula, K. R., and Taylor, P. B. (1992). A comprehensive procedure for preparation of partially methylated alditol acetates from glycoprotein carbohydrates. Anal. Biochem. 203, 101–108. Apweiler, R., Hermjakob, H., and Sharon, N. (1999). On the frequency of protein glycosylation, as deduced from analysis of the SWISS-PROT database. Biochim. Biophys. Acta Gen. Subj. 1473, 4–8. Arnold, J. N., Wormald, M. R., Sim, R. B., Rudd, P. M., and Dwek, R. A. (2007). The impact of glycosylation on the biological function and structure of human immunoglobulins. Annu. Rev. Immunol. 25, 21–50. Barb, A. W., Brady, E. K., and Prestegard, J. H. (2009). Branch-specific sialylation of IgG-Fc glycans by ST6Gal 1. Biochemistry 48, 9705–9707. Barresi, R., and Campbell, K. P. (2006). Dystroglycan: From biosynthesis to pathogenesis of human disease. J. Cell Sci. 119, 199–207. Barrientos, L. G., Dolan, C., and Gronenborn, A. M. (2000). Characterization of surfactant liquid crystal phases suitable for molecular alignment and measurement of dipolar couplings. J. Biomol. NMR 16, 329–337. Blanchard, V., Frank, M., Leeflang, B. R., Boelens, R., and Kamerling, J. P. (2008). The structural basis of the difference in sensitivity for PNGase F in the de-N-glycosylation of the native bovine pancreatic ribonucleases B and BS. Biochemistry 47, 3435–3446. Brancaccio, A., Schulthess, T., Gesemann, M., and Engel, J. (1995). Electron-microscopic evidence for a mucin-like region in chick muscle alpha-dystroglycan. FEBS Lett. 368, 139–142. Bush, C. A., and Feeney, R. E. (1986). Conformation of the glycotripeptide repeating unit of antifreeze glycoprotein of polar fish as determined from the fully assigned proton NMR-spectrum. Int. J. Pept. Protein Res. 28, 386–397. Buskas, T., Ingale, S., and Boons, G. J. (2006). Glycopeptides as versatile tools for glycobiology. Glycobiology 16, 113R–136R. Cavanaugh, J., Fairbrother, W. J., Palmer, A. G., and Skelton, N. J. (2007). Protein NMR Spectroscopy. Elsevier Academic Press, Amsterdam. Chai, W. G., Yuen, C. T., Kogelberg, H., Carruthers, R. A., Margolis, R. U., Feizi, T., and Lawson, A. M. (1999). High prevalence of 2-mono- and 2, 6-di-substituted manol-


383

terminating sequences among O-glycans released from brain glycopeptides by reductive alkaline hydrolysis. Eur. J. Biochem. 263, 879–888. Chang, V. T., Crispin, M., Aricescu, A. R., Harvey, D. J., Nettleship, J. E., Fennelly, J. A., Yu, C., Boles, K. S., Evans, E. J., Stuart, D. I., Dwek, R. A., Jones, E. Y., et al. (2007). Glycoprotein structural genomics: Solving the glycosylation problem. Structure 15, 267–273. Cheetham, J. C., Smith, D. M., Aoki, K. H., Stevenson, J. L., Hoeffel, T. J., Syed, R. S., Egrie, J., and Harvey, T. S. (1998). NMR structure of human erythropoietin and a comparison with its receptor bound conformation. Nat. Struct. Biol. 5, 861–866. Chen, B., Vogan, E. M., Gong, H. Y., Skehel, J. J., Wiley, D. C., and Harrison, S. C. (2005). Determining the structure of an unliganded and fully glycosylated SIV gp120 envelope glycoprotein. Structure 13, 197–211. Chung, S., Joo, H., Jang, K., Lee, H., Lee, S., and Kim, B. (2006). Galactosylation and sialylation of terminal glycan residues of human immunoglobulin G using bacterial glycosyltransferases with in situ regeneration of sugar-nucleotides. Enzyme Microb. Technol. 39, 60–66. Coltart, D. M., Royyuru, A. K., Williams, L. J., Glunz, P. W., Sames, D., Kuduk, S. D., Schwarz, J. B., Chen, X. T., Danishefsky, S. J., and Live, D. H. (2002). Principles of mucin architecture: Structural studies on synthetic glycopeptides bearing clustered mono-, di-, tri-, and hexasaccharide glycodomains. J. Am. Chem. Soc. 124, 9833–9844. Corzana, F., Busto, J. H., Engelsen, S. B., Jimenez-Barbero, J., Asensio, J. L., Peregrina, J. M., and Avenoza, A. (2006a). Effect of beta-O-glucosylation on L-ser and L-thr diamides: A bias toward alpha-helical conformations. Chem. Eur. J. 12, 7864–7871. Corzana, F., Busto, J. H., Jimenez-Oses, G., Asensio, J. L., Jimenez-Barbero, J., Peregrina, J. M., and Avenoza, A. (2006b). New insights into alpha-GalNAc-ser motif: Influence of hydrogen bonding versus solvent interactions on the preferred conformation. J. Am. Chem. Soc. 128, 14640–14648. Corzana, F., Busto, J. H., Jimenez-Oses, G., de Luis, M. G., Asensio, J. L., JimenezBarbero, J., Peregrina, J. M., and Avenoza, A. (2007). Serine versus threonine glycosylation: The methyl group causes a drastic alteration on the carbohydrate orientation and on the surrounding water shell. J. Am. Chem. Soc. 129, 9458–9467. Corzana, F., Busto, J. H., de Luis, M. G., Jimenez-Barbero, J., Avenoza, A., and Peregrina, J. M. (2009). The nature and sequence of the amino acid aglycone strongly modulates the conformation and dynamics effects of Tn antigen’s clusters. Chem. Eur. J. 15, 3863–3874. Cummings, R. D. (2009). The repertoire of glycan determinants in the human glycome. Mol. Biosyst. 5, 1087–1104. de Alba, E., and Tjandra, N. (2002). NMR dipolar couplings for the structure determination of biopolymers in solution. Prog. Nucl. Magn. Reson. Spectrosc. 40, 175–197. de Beer, T., van Zuylen, C. E. M., Hard, K., Boelens, R., Kaptein, R., Kamerling, J. P., and Vliegenthart, J. F. G. (1994). Rapid and simple approach for the NMR resonance assignment of the carbohydrate chains of an intact glycoprotein—application of gradientenhanced natural-abundance H-1-C-13 HSQC and HSQC-TOCSY to the alpha-subunit of human chorionic-gonadotropin. FEBS Lett. 348, 1–6. Deisenhofer, J. (1981). Crystallographic refinement and atomic models of a human fc fragment and its complex with fragment B of protein A from Staphylococcus aureus at 2.9- and 2.8-A˚ resolution. Biochemistry 20, 2361–2370. Dziadek, S., Griesinger, C., Kunz, H., and Reinscheid, U. M. (2006). Synthesis and structural model of an alpha(2, 6)-sialyl-T glycosylated MUC1 eicosapeptide under physiological conditions. Chem. Eur. J. 12, 4981–4993. Erbel, P. J. A., Karimi-Nejad, Y., van Kuik, J. A., Boelens, R., Kamerling, J. P., and Vliegenthart, J. F. G. (2000). Effects of the N-linked glycans on the 3D structure of the free alpha-subunit of human chorionic gonadotropin. Biochemistry 39, 6012–6021.

384

Adam W. Barb et al.

Fukuda, M., Carlsson, S. R., Klock, J. C., and Dell, A. (1986). Structures of O-linked oligosaccharides isolated from normal granulocytes, chronic myelogenous leukemiacells, and acute myelogenous leukemia-cells. J. Biol. Chem. 261, 2796–2806. Garner, O. B., and Baum, L. G. (2008). Galectin-glycan lattices regulate cell-surface glycoprotein organization and signaling. Biochem. Soc. Trans. 36, 1472–1477. Gerken, T. A., and Jentoft, N. (1987). Structure and dynamics of porcine submaxillary mucin as determined by natural abundance C-13 NMR-spectroscopy. Biochemistry 26, 4689–4699. Gerken, T. A., Butenhof, K. J., and Shogren, R. (1989). Effects of glycosylation on the conformation and dynamics of O-linked glycoproteins—C-13 NMR-studies of ovine submaxillary mucin. Biochemistry 28, 5536–5543. Glithero, A., Tormo, J., Haurum, J. S., Arsequell, G., Valencia, G., Edwards, J., Springer, S., Townsend, A., Pao, Y. L., Wormald, M., Dwek, R. A., Jones, E. Y., et al. (1999). Crystal structures of two H-2D(b)/glycopeptide complexes suggest a molecular basis for CTL cross-reactivity. Immunity 10, 63–74. Gomes, M. M., Wall, S. B., Takahashi, K., Novak, J., Renfrow, M. B., and Herr, A. B. (2008). Analysis of IgA1 N-glycosylation and its contribution to Fc alpha RI binding. Biochemistry 47, 11285–11299. Gong, B., Cukan, M., Fisher, R., Li, H., Stadheim, T. A., and Gerngross, T. (2009). Characterization of N-linked glycosylation on recombinant glycoproteins produced in Pichia pastoris using ESI-MS and MALDI-TOF. Meth. Mol. Biol. 534, 213–223. Grinstead, J. S., Koganty, R. R., Krantz, M. J., Longenecker, B. M., and Campbell, A. P. (2002). Effect of glycosylation on muc1 humoral immune recognition: NMR studies of MUC1 glycopeptide-antibody interactions. Biochemistry 41, 9946–9961. Hanson, S. R., Culyba, E. K., Hsu, T. L., Wong, C. H., Kelly, J. W., and Powers, E. T. (2009). The core trisaccharide of an N-linked glycoprotein intrinsically accelerates folding and enhances stability. Proc. Natl. Acad. Sci. USA 106, 3131–3136. Herr, A. B., Ballister, E. R., and Bjorkman, P. J. (2003). Insights into IgA-mediated immune responses from the crystal structures of human Fc alpha RI and its complex with IgA1-Fc. Nature 423, 614–620. Holland, M., Takada, K., Okumoto, T., Takahashi, N., Kato, K., Adu, D., Ben-Smith, A., Harper, L., Savage, C. O., and Jefferis, R. (2002). Hypogalactosylation of serum IgG in patients with ANCA-associated systemic vasculitis. Clin. Exp. Immunol. 129, 183–190. Ishima, R., and Torchia, D. A. (2000). Protein dynamics from NMR. Nat. Struct. Biol. 7, 740–743. Kan, C., and Danishefsky, S. J. (2009). Recent departures in the synthesis of peptides and glycopeptides. Tetrahedron 65, 9047–9065. Kaneko, Y., Nimmerjahn, F., and Ravetch, J. V. (2006). Anti-inflammatory activity of immunoglobulin G resulting from Fc sialylation. Science 313, 670–673. Kirnarsky, L., Prakash, O., Vogen, S. M., Nomoto, M., Hollingsworth, M. A., and Sherman, S. (2000). Structural effects of O-glycosylation on a 15-residue peptide from the mucin (MUC1) core protein. Biochemistry 39, 12076–12082. Kobata, A. (2008). The N-linked sugar chains of human immunoglobulin G: Their unique pattern, and their functional roles. Biochim. Biophys. Acta Gen. Subj. 1780, 472–478. Kollman, J. M., Pandi, L., Sawaya, M. R., Riley, M., and Doolittle, R. F. (2009). Crystal structure of human fibrinogen. Biochemistry 48, 3877–3886. Krapp, S., Mimura, Y., Jefferis, R., Huber, R., and Sondermann, P. (2003). Structural analysis of human IgG-Fc glycoforms reveals a correlation between glycosylation and structural integrity. J. Mol. Biol. 325, 979–989. Lane, A. N., Hays, L. M., Feeney, R. E., Crowe, L. M., and Crowe, J. H. (1998). Conformational and dynamic properties of a 14 residue antifreeze glycopeptide from antarctic cod. Protein Sci. 7, 1555–1563.


385

Leatherbarrow, R. J., Rademacher, T. W., Dwek, R. A., Woof, J. M., Clark, A., Burton, D. R., Richardson, N., and Feinstein, A. (1985). Effector functions of a monoclonal aglycosylated mouse IgG2a: Binding and activation of complement component C1 and interaction with human monocyte Fc receptor. Mol. Immunol. 22, 407–415. Lee, J. E., Fusco, M. L., and Saphire, E. O. (2009). An efficient platform for screening expression and crystallization of glycoproteins produced in human cells. Nat. Protoc. 4, 592–604. Li, P. Z., Go, X. G., Arellano, R. O., and Renugopalakrishnan, V. (2001). Glycosylated and phosphorylated proteins—expression in yeast and oocytes of xenopus: Prospects and challenges—relevance to expression of thermostable proteins. Protein Expr. Purif. 22, 369–380. Liu, M., Barany, G., and Live, D. (2005). Parallel solid-phase synthesis of mucin-like glycopeptides. Carbohydr. Res. 340, 2111–2122. Liu, S., Venot, A., Meng, L., Tian, F., Moremen, K. W., Boons, G. J., and Prestegard, J. H. (2007). Spin-labeled analogs of CMP-NeuAc as NMR probes of the alpha-2, 6-sialyltransferase ST6Gal 1. Chem. Biol. 14, 409–418. Liu, M., Borgert, A., Barany, G., and Live, D. (2008). Conformational consequences of protein glycosylation: Preparation of O-mannosyl serine and threonine building blocks, and their incorporation into glycopeptide sequences derived from alpha-dystroglycan. Biopolymers 90, 358–368. Lund, J., Takahashi, N., Pound, J. D., Goodall, M., and Jefferis, R. (1996). Multiple interactions of IgG with its core oligosaccharide can modulate recognition by complement and human Fc gamma receptor i and influence the synthesis of its oligosaccharide chains. J. Immunol. 157, 4963–4969. Lustbader, J. W., Birken, S., Pollak, S., Pound, A., Chait, B. T., Mirza, U. A., Ramnarain, S., Canfield, R. E., and Brown, J. M. (1996). Expression of human chorionic gonadotropin uniformly labeled with NMR isotopes in Chinese hamster ovary cells: An advance toward rapid determination of glycoprotein structures. J. Biomol. NMR 7, 295–304. Lutteke, T. (2009). Analysis and validation of carbohydrate three-dimensional structures. Acta Crystallogr. D Biol. Crystallogr. 65, 156–168. Macnaughtan, M. A., Tian, F., Liu, S., Meng, L., Park, S., Azadi, P., Moremen, K. W., and Prestegard, J. H. (2008). C-13-sialic acid labeling of glycans on glycoproteins using ST6Gal 1. J. Am. Chem. Soc. 130, 11864–11865. Mandel, A. M., Akke, M., and Palmer, A. G. (1995). Backbone dynamics of Escherichia-coli ribonuclease H1—correlations with structure and function in an active enzyme. J. Mol. Biol. 246, 144–163. Matsushita, T., Hinou, H., Fumoto, M., Kurogochi, M., Fujitani, N., Shimizu, H., and Nishimura, S. I. (2006). Construction of highly glycosylated mucin-type glycopeptides based on microwave-assisted solid-phase syntheses and enzymatic modifications. J. Org. Chem. 71, 3051–3063. Meyer, B., and Moller, H. (2007). Conformation of glycopeptides and glycoproteins. Top. Curr. Chem. 267, 187–251. Mimura, Y., Yamamoto, Y., Inoue, Y., and Chujo, R. (1992). NMR-study of interaction between sugar and peptide moieties in mucin-type model glycopeptides. Int. J. Biol. Macromol. 14, 242–248. Mimura, Y., Sondermann, P., Ghirlando, R., Lund, J., Young, S. P., Goodall, M., and Jefferis, R. (2001). Role of oligosaccharide residues of IgG1-Fc in Fc gamma RIIb binding. J. Biol. Chem. 276, 45539–45547. Naganagowda, G. A., Gururaja, T. L., Satyanarayana, J., and Levine, M. J. (1999). NMR analysis of human salivary mucin (MUC7) derived O-linked model glycopeptides: Comparison of structural features and carbohydrate-peptide interactions. J. Pept. Res. 54, 290–310.

386

Adam W. Barb et al.

Narhi, L. O., Arakawa, T., Aoki, K. H., Elmore, R., Rohde, M. F., Boone, T., and Strickland, T. W. (1991). The effect of carbohydrate on the structure and stability of erythropoietin. J. Biol. Chem. 266, 23022–23026. North, S. J., Huang, H. H., Sundaram, S., Jang-Lee, J., Etienne, A. T., Trollope, A., Chalabi, S., Dell, A., Stanley, P., and Haslam, S. M. (2010). Glycomics profiling of Chinese hamster ovary cell glycosylation mutants reveals N-glycans of a novel size and complexity. J. Biol. Chem. 285, 5759–5775. Ohtsubo, K., and Marth, J. D. (2006). Glycosylation in cellular mechanisms of health and disease. Cell 126, 855–867. Ottiger, M., and Bax, A. (1999). Bicelle-based liquid crystals for NMR-measurement of dipolar couplings at acidic and basic pH values. J. Biomol. NMR 13, 187–191. Parekh, R. B., Dwek, R. A., Sutton, B. J., Fernandes, D. L., Leung, A., Stanworth, D., Rademacher, T. W., Mizuochi, T., Taniguchi, T., Matsuta, K., et al. (1985). Association of rheumatoid arthritis and primary osteoarthritis with changes in the glycosylation pattern of total serum IgG. Nature 316, 452–457. Payne, R. J., and Wong, C. H. (2010). Advances in chemical ligation strategies for the synthesis of glycopeptides and glycoproteins. Chem. Commun. 46, 21–43. Petrescu, A. J., Milac, A. L., Petrescu, S. M., Dwek, R. A., and Wormald, M. R. (2004). Statistical analysis of the protein environment of N-glycosylation sites: Implications for occupancy, structure, and folding. Glycobiology 14, 103–114. Petrescu, A. J., Wormald, M. R., and Dwek, R. A. (2006). Structural aspects of glycomes with a focus on N-glycosylation and glycoprotein folding. Curr. Opin. Struct. Biol. 16, 600–607. Piontek, C., Silva, D. V., Heinlein, C., Pohner, C., Mezzato, S., Ring, P., Martin, A., Schmid, F. X., and Unverzagt, C. (2009). Semisynthesis of a homogeneous glycoprotein enzyme: Ribonuclease C: Part 2. Angew. Chem. Int. Ed. 48, 1941–1945. Plummer, T. H., Jr., and Tarentino, A. L. (1991). Purification of the oligosaccharidecleaving enzymes of Flavobacterium meningosepticum. Glycobiology 1, 257–263. Radaev, S., Motyka, S., Fridman, W. H., Sautes-Fridman, C., and Sun, P. D. (2001). The structure of a human type III Fc gamma receptor in complex with Fc. J. Biol. Chem. 276, 16469–16477. Raju, T. S., Briggs, J. B., Chamow, S. M., Winkler, M. E., and Jones, A. J. (2001). Glycoengineering of therapeutic glycoproteins: In vitro galactosylation and sialylation of glycoproteins with terminal N-acetylglucosamine and galactose residues. Biochemistry 40, 8868–8876. Rice, K. G. (2000). Derivatization strategies for preparing N-glycan probes. Anal. Biochem. 283, 10–16. Rich, J. R., and Withers, S. G. (2009). Emerging methods for the production of homogeneous human glycoproteins. Nat. Chem. Biol. 5, 206–215. Roitt, I. M., Brostoff, J., and Male, D. K. (2001). Immunology Mosby, Edinburgh. Rose, M. C., Voter, W. A., Sage, H., Brown, C. F., and Kaufman, B. (1984). Effects of deglycosylation on the architecture of ovine submaxillary mucin glycoprotein. J. Biol. Chem. 259, 3167–3172. Sanchez, L. M., Chirino, A. J., and Bjorkman, P. (1999). Crystal structure of human ZAG, a fat-depleting factor related to MHC molecules. Science 283, 1914–1919. Sasaki, T., Yamada, H., Matsumura, K., Shimizu, T., Kobata, A., and Endo, T. (1998). Detection of O-mannosyl glycans in rabbit skeletal muscle alpha-dystroglycan. Biochim. Biophys. Acta Gen. Subj. 1425, 599–606. Scallon, B. J., Tam, S. H., McCarthy, S. G., Cai, A. N., and Raju, T. S. (2007). Higher levels of sialylated Fc glycans in immunoglobulin G molecules can adversely impact functionality. Mol. Immunol. 44, 1524–1534.


387

Schuster, O., Klich, G., Sinnwell, V., Kranz, H., Paulsen, H., and Meyer, B. (1999). ’Wavetype’ structure of a synthetic hexaglycosylated decapeptide: A part of the extracellular domain of human glycophorin A. J. Biomol. NMR 14, 33–45. Schwarz, F., Huang, W., Li, C. S., Schulz, B. L., Lizak, C., Palumbo, A., Numao, S., Neri, D., Aebi, M., and Wang, L. X. (2010). A combined method for producing homogeneous glycoproteins with eukaryotic N-glycosylation. Nat. Chem. Biol. 6, 264–266. Schwieters, C. D., Kuszewski, J. J., and Clore, G. M. (2006). Using Xplor-NIH for NMR molecular structure determination. Prog. Nucl. Magn. Reson. Spectrosc. 48, 47–62. Shogren, R., Gerken, T. A., and Jentoft, N. (1989). Role of glycosylation on the conformation and chain dimensions of O-linked glycoproteins—light-scattering-studies of ovine submaxillary mucin. Biochemistry 28, 5525–5536. Simanek, E. E., Huang, D. H., Pasternack, L., Machajewski, T. D., Seitz, O., Millar, D. S., Dyson, H. J., and Wong, C. H. (1998). Glycosylation of threonine of the repeating unit of RNA polymerase II with beta-linked N-acetylglucosame leads to a turnlike structure. J. Am. Chem. Soc. 120, 11567–11575. Slynko, V., Schubert, M., Numao, S., Kowarik, M., Aebi, M., and Allain, F. H. T. (2009). NMR structure determination of a segmentally labeled glycoprotein using in vitro glycosylation. J. Am. Chem. Soc. 131, 1274–1281. Springer, G. F. (1997). Immunoreactive T and Tn epitopes in cancer diagnosis, prognosis, and immunotherapy. J. Mol. Med. 75, 594–602. Szakonyi, G., Klein, M. G., Hannan, J. P., Young, K. A., Ma, R. L. Z., Asokan, R., Holers, V. M., and Chen, X. J. S. (2006). Structure of the Epstein-Barr virus major envelope glycoprotein. Nat. Struct. Mol. Biol. 13, 996–1001. Takahashi, N., Nakagawa, H., Fujikawa, K., Kawamura, Y., and Tomiya, N. (1995). Threedimensional elution mapping of pyridylaminated N-linked neutral and sialyl oligosaccharides. Anal. Biochem. 226, 139–146. Takeuchi, H., Kato, K., Hassan, H., Clausen, H., and Irimura, T. (2002). O-GalNAc incorporation into a cluster acceptor site of three consecutive threonines—distinct specificity of GalNAc-transferase isoforms. Eur. J. Biochem. 269, 6173–6183. Tarp, M. A., Sorensen, A. L., Mandel, U., Paulsen, H., Burchell, J., TaylorPapadimitriou, J., and Clausen, H. (2007). Identification of a novel cancer-specific immunodominant glycopeptide epitope in the MUC1 tandem repeat. Glycobiology 17, 197–209. Ten Hagen, K. G., Fritz, T. A., and Tabak, L. A. (2003). All in the family: The UDPGalNAc: Polypeptide N-acetylgalactosaminyltransferases. Glycobiology 13, 1R–16R. Toyoda, T., Itai, T., Arakawa, T., Aoki, K. H., and Yamaguchi, H. (2000). Stabilization of human recombinant erythropoietin through interactions with the highly branched Nglycans. J. Biochem. 128, 731–737. Toyoda, T., Arakawa, T., and Yamaguchi, H. (2002). N-glycans stabilize human erythropoietin through hydrophobic interactions with the hydrophobic protein surface: Studies by surface plasmon resonance analysis. J. Biochem. 131, 511–515. van Kooyk, Y., and Rabinovich, G. A. (2008). Protein-glycan interactions in the control of innate and adaptive immune responses. Nat. Immunol. 9, 593–601. Varki, A., Cummings, R. D., Esko, J. D., Freeze, H. H., Stanley, P., Bertozzi, C. R., Hart, G. W., and Mee, E. (2009). Essentials of Glycobiology Cold Spring Harbor Press, Cold Spring Harbor, New York. Vliegenthart, J. F. G., Dorland, L., and Van Halbeek, H. (1983). 1H-nuclear magnetic resonance spectroscopy as a tool in the structural analysis of carbohydrates related to glycoproteins. Adv. Carbohydr. Chem. Biochem. 41, 209–374. Wang, L. X. (2008). Chemoenzymatic synthesis of glycopeptides and glycoproteins through endoglycosidase-catalyzed transglycosylation. Carbohydr. Res. 343, 1509–1522.

388

Adam W. Barb et al.

Weller, C. T., Lustbader, J., Seshadri, K., Brown, J. M., Chadwick, C. A., Kolthoff, C. E., Ramnarain, S., Pollak, S., Canfield, R., and Homans, S. W. (1996). Structural and conformational analysis of glycan moieties in situ on isotopically C-13, N-15-enriched recombinant human chorionic gonadotropin. Biochemistry 35, 8815–8823. Whelan, S. A., and Hart, G. W. (2006). Identification of O-GlcNAc sites on proteins. In ‘‘Methods in Enzymology,’’ (M. Fukuda, ed.)., Vol. 415, pp.113–133. Wyss, D. F., Choi, J. S., Li, J., Knoppers, M. H., Willis, K. J., Arulanandam, A. R. N., Smolyar, A., Reinherz, E. L., and Wagner, G. (1995). Conformation and function of the N-linked glycan in the adhesion domain of human CD2. Science 269, 1273–1278. Yamaguchi, Y., Kato, K., Shindo, M., Aoki, S., Furusho, K., Koga, K., Takahashi, N., Arata, Y., and Shimada, I. (1998). Dynamics of the carbohydrate chains attached to the Fc portion of immunoglobulin G as studied by NMR spectroscopy assisted by selective C-13 labeling of the glycans. J. Biomol. NMR 12, 385–394. Yamaguchi, Y., Nishimura, M., Nagano, M., Yagi, H., Sasakawa, H., Uchida, K., Shitara, K., and Kato, K. (2006). Glycoform-dependent conformational alteration of the Fc region of human immunoglobulin G1 as revealed by NMR spectroscopy. Biochim. Biophys. Acta Gen. Subj. 1760, 693–700. Yamamoto, N., Tanabe, Y., Okamoto, R., Dawson, P. E., and Kajihara, Y. (2008). Chemical synthesis of a glycoprotein having an intact human complex-type sialyloligosaccharide under the Boc and Fmoc synthetic strategies. J. Am. Chem. Soc. 130, 501–510.

C H A P T E R

N I N E T E E N

Methods to Study the Biosynthesis of Bacterial Furanosides Myles B. Poulin and Todd L. Lowary Contents 1. Introduction 2. Chemoenzymatic Preparation of Furanose Nucleotides 2.1. Discussion 2.2. Procedure 3. Pyranose–Furanose Mutases Involved in Furanose Nucleotide Biosynthesis 3.1. Discussion 3.2. HPLC assays for UGM activity 4. Galactofuranosyltransferases Involved in Galactofuranoside Biosynthesis 4.1. Discussion 4.2. Enzyme-coupled spectrophotometric assay for GlfT2 activity Acknowledgments References

390 393 393 394 398 398 399 402 402 403 407 407

Abstract Carbohydrates in the thermodynamically disfavored furanose ring conformation are not present in mammalian glycoconjugates, but are widespread in the glycans produced by many bacterial pathogens. In bacteria, these furanose sugars are often found in cell surface glycoconjugates, and are essential for the viability or virulence of the organisms. As a result, the enzymes involved in the biosynthesis of bacterial furanosides are attractive targets as potential selective antimicrobial chemotherapeutics. However, before such chemotherapeutics can be designed, synthesized, and evaluated, more information about the activity and specificity of these enzymes is required. This chapter describes assays that have been used to study enzymes involved in the biosynthesis of one of the most abundant naturally occurring furanose residues, galactofuranose (Galf ). In particular, the focus is on UDP-galactopyranose mutase and galactofuranosyltransferases. The assays described in this chapter require The Alberta Ingenuity Centre for Carbohydrate Science, Department of Chemistry, University of Alberta, Edmonton, Alberta, Canada Methods in Enzymology, Volume 478 ISSN 0076-6879, DOI: 10.1016/S0076-6879(10)78019-8

#


389

390

Myles B. Poulin and Todd L. Lowary

UDP-galactofuranose (UDP-Galf); therefore, a procedure for the preparation of UDP-Galf, as well as various UDP-Galf derivatives, using a three-enzyme chemoenzymatic procedure, is also described.

1. Introduction In addition to thermodynamically favored pyranose sugars, bacteria also produce polysaccharides and glycoconjugates containing saccharides in the five-membered furanose ring conformation. These thermodynamically disfavored furanose sugars are absent in mammalian glycoconjugates, but are widespread in other domains of life ranging from bacteria and archaebacteria, to protozoa, fungi, and plants (Peltier et al., 2008a; Richards and Lowary, 2009). In many pathogenic bacteria, these furanose sugars are found in key cell surface glycoconjugates, including the lipopolysaccharide (LPS) O antigens of Escherichia coli (Stevenson et al., 1994), Klebsiella pneumoniae (Ko¨plin et al., 1997), and Samonella typhimurium (Berst et al., 1969), the capsular polysaccharide (CPS) of Campylobacter jejuni (Hanniffy et al., 1999; McNally et al., 2005; St. Michael et al., 2002), and the mycolyl arabinogalactan (mAG) complex and lipoarabinomannan (LAM) of mycobacteria (Bhamidi et al., 2008; Brennan and Nikaido, 1995), among others. In many of these organisms, the furanose residues in these glycans have been demonstrated to be essential for cell viability, or play a critical role in cell physiology (Lee et al., 1997; Pan et al., 2001). Because these furanosides are absent from mammalian glycoconjugates, there has been a surge of interest in developing inhibitors of furanose biosynthesis as potential selective chemotherapeutics to treat these pathogenic microorganisms (Lowary, 2003; Pedersen and Turco, 2003; Umesiri et al., 2010). The hexofuranose D-galactofuranose (Galf ), the pentofuranose D-arabinofuranose (Araf ), and the hexulose D-fructofuranose (Fruf ), represent three of the most common furanose sugars found in bacterial glycans, and each of these three furanose sugars employ a different type of activated furanoside donor in their biosynthesis. Microbial fructans are polymers of 20–10,000 Fruf units containing repeating b-(2!1) (inulins), b-(2!6) (levans), or a combination of b-(2!1) and b–(2!6) (mixed levans) glycosidic linkages, which are attached to the Fruf residue of D-sucrose (Velazquez-Hernandez et al., 2009). The fructosyltransferases (FruT) involved in the biosynthesis of microbial fructans are members of the glycoside hydrolase family 68, and catalyze the transglycosylation of Fruf using D-sucrose as the furanose donor (van Hijum et al., 2006) as shown (Fig. 19.1). Fruf residues have also been identified in other microbial glycoconjugates, such as the CPS of C. jejuni HS:1 (McNally et al., 2005); however, it is unclear whether these Fucf residues are also incorporated from D-sucrose.

391

Biosynthesis of Bacterial Furanosides

HO OH HO HO

O

O

O

O

O

HO

OH

HO HO

OH

HO

OH O

HO

O

O O

OH 6-Ketose

OH OH

Levan

OH

OH

HO +

HO HO

OH

HO HO

OH

HO

OH

HO OH O O

FruT O

HO O

OH

OH HO D-sucrose

OH

O OH

OH HO HO

O HO

OH HO HO

O

OH OH OH O

OH O

Inulin OH

OH HO 1-Ketose

Figure 19.1 Biosynthetic scheme for microbial fructan biosynthesis. Fructosyltransferase (FruT) enzymes catalyze the transglycosylation of Fruf from D-sucrose in the synthesis of levan and inulin polysaccharides.

The only known pathway for the biosynthesis of Araf residues present in the mAG complex and LAM of mycobacterial species utilize the lipid donor decaprenyl-phospho-arabinose (DPA) as the sole Araf source (Wolucka, 2008; Wolucka and Dehoffmann, 1995). A water soluble uridine diphosphate (UDP) derivative of Araf has also been reported in Mycobacterium smegmatis (Singh and Hogan, 1994); however, there have been no subsequent reports to demonstrate a possible biosynthetic role for UDP-Araf. Isotope labeling experiments where M. smegmatis cells were incubated with 14C- and 13C-labeled glucose elegantly show that neither the pentose phosphate pathway, which would result in the loss of C1 of glucose, or the uronic acid pathway, which would result in the loss of C6 of glucose, are involved in the biosynthesis of Araf (Klutts et al., 2002). Instead, a nonoxidative pentose phosphate pathway is utilized. Other isotope labeling experiments with 14C-labeled 5-phosphoribose-1-pyrophosphate (5-PRib-1-PP) implicate this species as an intermediate in the biosynthesis of DPA as well as decaprenyl-phospho-ribose (DPR) (Scherman et al., 1996). The subsequent identification of the mycobacterial DPR 50 -phosphate synthetase (Huang et al., 2005), and decaprenyl-phospho-ribose 20 -epimerase (DPRE) (Mikusˇova´ et al., 2005) led to the proposed pathway (Fig. 19.2) for the biosynthesis of Araf in mycobacteria. Six arabinofuranosyltransferase (Aft) enzymes involved in the biosynthesis of mycobacterial cell wall arabinan have been identified (Berg et al., 2007; Tam and Lowary, 2009); however, the exact role of many of these proteins in mAG and LAM assembly remain to be determined, and it is likely that other arabinofuranosyltransferases exist. It should be noted that while (myco)bacterial species use lipid linked donors to incorporate D-arabinofuranose residues into their glycoconjugates, plants, which produce large amounts of L-arabinofuranose-containing glycans, do

392

Glc-6-P


Non-oxidative pentose phosphate Pathway

Rib-5-P

i.

Rib-1-PP-5-P

ii.

Decaprenyl-P-Rib-5-P iii.

Araf containing glycocongugates

v.

Decaprenyl-PAraf

iv.

Decaprenyl-P-Ara-5-P

Figure 19.2 Biosynthetic scheme for Araf-containing glycocongugates in mycobacteria using decaprenyl-P-Araf donor. (i) Phosphoribosylpyrophosphate synthetase (PRPP); (ii) decaprenyl-phospho-ribose 50 -phosphate synthetase; (iii) phosphatase; (iv) decaprenyl-phospho-ribose 20 -epimerase; (v) arabinofuranosyltransferase (Aft).

Glc-1-P

GalU

UDP-Glc

GalE

Galactose salvage pathway b –Galp

GMR

a–Galp

GK

UDP-Galp

UGM

UDP-Galf

GIfT

Galf-containing glycocongugates

GalPUT Galp-1-P

Figure 19.3 Biosynthetic scheme for galactofuranose-containing glycoconjugates. The UDP-Galf donor utilized for Galf incorporation by galactofuranosyltransferases (GlfTs) is synthesized from UDP-Galp by UDP-galactopyranose mutases (UGMs). Two pathways for the biosynthesis of UDP-Galp are shown, the de novo pathway for UDP-Galp biosynthesis as well as galactose salvage pathway. GalU ¼ UDP-glucosepyrophosphatase; GalE ¼ glucose-4-epimerase; GMR ¼ galactose mutarotase; GK ¼ galactokinase; GalPUT ¼ galactose-1-phosphate–uridylyltransferase.

so via enzymes that recognize the sugar nucleotide UDP-L-Araf as the donor species (Konishi et al., 2007; Reiter and Vanzin, 2001). Galf residues are incorporated into bacterial glycans using nucleotideactivated sugar donors (Fig. 19.3). Early studies established that the Galf in S. typhimurium is synthesized from a derivative of galactopyranose and that the ring contraction does not occur at the level of either free galactose or galactose-1-phosphate (Nikaido and Sarvas, 1971; Sarvas and Nikaido, 1971). The enzyme uridine 50 -diphospho-galactopyranose mutase (UGM), which was first identified in E. coli (Nassau et al., 1996) and subsequently in K. pneumonia (Ko¨plin et al., 1997), Mycobacterium tuberculosis (Weston et al., 1998), and Deinococcus radiodurans (Partha et al., 2009b), was found to carry out the ring contraction of uridine 50 -diphospho-D-galactopyranose (UDP-Galp) to uridine 50 -diphospho-D-galactofuranose (UDPGalf), the biosynthetic precursor of bacterial Galf residues. The UDP-Galp originates either from glucopyranose-1-phosphate by de novo biosynthesis, or from free galactose via the galactose salvage pathway (Thibodeaux et al., 2008). Enzymes known as galactofuranosyltransferases (GlfTs) catalyze the final coupling of UDP-Galf to the appropriate Galf-containing glycans.


393

In this chapter, we describe assays used to study the enzymes involved in the biosynthesis of Galf-containing bacterial glycoconjugates; specifically, the UGM and GlfT proteins. We also provide a procedure for the chemoenzymatic preparation of UDP-Galf (and derivatives of UDP-Galf ) that takes advantage of the reduced substrate specificity of enzymes from the galactose salvage pathway.

2. Chemoenzymatic Preparation of Furanose Nucleotides 2.1. Discussion The assays described in this chapter are used to study the biosynthesis of Galf-containing glycoconjugates; as a result, these assays require the use of the nucleotide-activated galactofuranose donor, UDP-Galf. Multiple chemical syntheses of this activated Galf donor have been reported using standard pyrophosphate bond formation reactions between galactofuranose-1-phosphate (Galf-1-P) and UMP-morpholidate (Zhang and Liu, 2000), UMP-Nmethylimidazolide (Marlow and Kiessling, 2001), or N,N-carbonyldiimidazole (CDI) activated UMP (Tsvetkov and Nikolaev, 2000). Direct glycosylation of 1-thio-a-D-galactofuranosides with UDP (Peltier et al., 2007) has also been employed for the chemical synthesis of UDP-Galf. In addition, an enzymatic method for the production of UDP-Galf from UDP-Galp utilizing E. coli UGM was reported (Lee et al., 1996); however, at equilibrium only 10% of the desired UDP-Galf is produced. The low yields of these chemical and enzymatic syntheses make them impractical for the production of sufficient quantities of the UDP-Galf donor required to carry out detailed studies of galactofuranose biosynthesis. This procedure, originally developed by Field and coworkers (Errey et al., 2004), details the chemoenzymatic preparation of UDP-Galf (2) and UDP-Galf derivatives from the corresponding Galf-1-P (1) or Galf-1-P derivatives, UTP, and uridine-50 -diphospho-D-glucose (UDP-Glc) using three enzymes (Scheme 19.1). Galf-1-P (or derivatives thereof) and UDPGlc react to form UDP-Galf and glucose-1-phosphate (Glc-1-P) catalyzed by immobilized galactose-1-phosphate uridyltransferase (GalPUT, EC 2.7.7.12) (Liu et al., 2002). The UTP reacts with Glc-1-P to regenerate UDP-Glc catalyzed by UDP-glucose pyrophosphorylase (GalU, EC 2.7.7.9) and the inorganic pyrophosphate (PPi) produced is cleaved to inorganic phosphate by inorganic pyrophosphatase (IPP, EC 3.6.1.1). This procedure takes advantage of the reduced substrate specificity of the GalPUT enzyme of the galactose salvage pathway, which allows it to tolerate various Galp-1-P and Galf-1-P derivatives as substrates (Errey et al., 2004; Peltier et al., 2008b).

394

A


HO O HO HO

1

B O O P O– – OH O

HO O HO

3

UDP-Glc

PPi

IPP

HO O

2Pi HO

GalPUT

GalU Glc-1-P

HO

2

O O P O– – OH O

HO O

UTP HO

HO O HO

5

OUDP OH

F

7 HO O

HO

9

O O P O– – OH O

O O P O– – OH O

HO O HO

4 HO O HO

6

OUDP OH

HO O HO F

O O P O– – OH O

OUDP OH

8

OUDP OH

HO O

HO

OUDP OH

10

Scheme 19.1 (A) Generation of UDP-Galf (2) from Galf-1-P using three-enzyme chemoenzymatic reaction. (B) Other Galf-1-P derivitive used to synthesize UDP-Galf derivatives using three-enzyme chemoenzymatic reaction. UDP-Fucf (4), UDP-Araf (6), UDP-6F-Galf (8), and UDP-5d-Galf (10).

2.2. Procedure 2.2.1. Materials and methods Soluble GalU (EC 2.7.7.9, E. coli) was prepared by the method of Wang and coworkers (Liu et al., 2002) with minor modifications. Briefly, the galU gene of E. coli K-12 substrain MG 1655 was cloned and inserted into a pET15b vector using XhoI and BamHI restriction sites. Soluble GalU was expressed in E. coli BL-21 Origami cells and purified using a nickel– nitrilotriacetic (Ni–NTA) column. Alternatively GalU may be purchased from Sigma. Uridine 50 -diphosphoglucose disodium salt (UDP-Glc), uridine 50 -triphosphate trisodium salt (UTP), and IPP were obtained from Sigma-Aldrich (St. Louis, MO). The chemical synthesis of galactose-1phosphate di-triethylammonium salt (1) was achieved as previously reported (de Lederkremer et al., 1994; Tsvetkov and Nikolaev, 2000) and will not be discussed here. All other chemicals and biochemicals were reagent grade and utilized without further purification. 2.2.2. Immobilization of GalPUT The procedures of Wang and coworkers (Li et al., 2004; Liu et al., 2002) and Field and coworkers (Errey et al., 2004) for the cloning and immobilization GalPUT were followed (with minor modifications) as described below. The galPUT gene was cloned from E. coli K-12 sub-strain MG 1655 (ATCC #47076) as described by Wang and coworkers (Liu et al., 2002), and inserted into pET15b expression vector (Novagen) using Nde1 and BamH1


395

restriction sites. E. coli DH5-a cloning strain then E. coli BL-21 Origami (Novagen) expression strain were then transformed with the pET15b-GalPUT vector. The expression strain BL-21 Origami was grown in 100 mL of Luria–Bertani (LB) broth containing 100 mg/mL of ampicillin overnight at 37 C in a shaker–incubator, then 50 mL was transferred into fresh LB broth (1 L) with ampicillin and grown for an additional 2 h at 37 C until and optical density at 600 nm of 0.6–0.8 was observed. Expression of GalPUT containing a C-terminal His-tag was induced using 1 mM isopropyl 1-thio-b-D-galactopyranoside (IPTG) and the culture was incubated for an additional 5 h in a shaker–incubator at 30 C. Cells were harvested by centrifugation (Beckman centrifuge, 11,325g, 15 min, 4 C) and the cell paste was stored at –20 C. It is important to note that storage of the cell paste longer than 3 months results in a significant decrease in enzyme activity. Cell paste from 2 L of culture was resuspended in 100 mL (1/20th the culture volume) of resuspension buffer (300 mM NaCl, 10 mM imidazole, and 50 mM sodium phosphate, pH 8.0, supplemented with protease inhibitors) and lysed using a benchtop cell disruptor (Constant Systems Inc., NC) set to 20 Kpsi. The insoluble cellular debris was removed by centrifugation (Beckman Ultracentrifuge, 105,000g, 1H, 4 C). The supernatant was applied to a 10 mL column of Ni–NTA agarose resin (Qiagen), equilibrated with 3 column volumes of resuspension buffer at 4 C, with a flow rate of 1 mL/min and the absorbance of the flow through was monitored at 280 nm. The resin was washed with 4–5 column volumes of wash buffer (300 mM NaCl, 20 mM imidazole, 50 mM sodium phosphate, pH 8.0) until the elution profile returned to baseline. At this stage GalPUT remained immobilized on the resin. The immobilized GalPUT was equilibrated with 4 column volumes of GalPUT reaction buffer (5 mM KCl, 10 mM MgCl2, 50 mM HEPES, pH 8.0). The resin containing immobilized GalPUT was used directly for the generation of UDP-Galf. The immobilized enzyme should be used within 3–4 days of generation or, for best results, immediately after production. 2.2.3. Generation of UDP-galactofuranose (2) The chemoenzymatic reaction was carried out in 1 dram plastic vials. UTP (24.4 mg, 40 mmol) was added to a solution of Galf-1-P (20.4 mg, 44 mmol) prepared in 204 mL of MilliQ water. To this solution was added 2 GalPUT reaction buffer (234 mL, equal to the volume of water for a final concentration of 1 GalPUT reaction buffer). The enzymes GalU (15 mL, 20 U/ 15 mL), IPP (15 mL, 0.25 U/mL), and immobilized GalPUT (0.8 mL of resin containing 20 U of enzyme) were added to the reaction solution. The reaction was initiated by the addition of UDP-Glc (3 mL, 50 mM ) solution, flushed with nitrogen or argon gas, and incubated at room temperature with gentle rotation for 16 h. The reaction was monitored by HPLC (Varian Microsorb 100-5 C18, 4.6 250 mm; Fig. 19.4). A small aliquot from the reaction mixture (5 mL)

396


2.0 UDP-Galf

1.8

Absorbance at 262 nm

1.6 1.4 1.2 1.0 0.8 0.6 UDP 0.4

20 H

0.2 0.0 0.0

18 H

UTP Uridine UMP 5.0

10.0

16 H 15.0

20.0 25.0 30.0 Time (min)

35.0

40.0

45.0

50.0

Figure 19.4 Analytical HPLC profile for three-enzyme chemoenzymatic generation of UDP-Galf. Peaks corresponding to the UDP-Galf product (14.5 min), the UTP starting material (26.0 min), and the degradation products uridine (6.0 min), UMP (9.7 min), and UDP (16.5 min) are labeled. Depletion of the UTP peak is observed between 16 and 20 h.

was diluted with 30 mL of water and centrifuged (Eppendorf centrifuge, 16,000g, 10 min) at room temperature using AmiconÒ Ultra-0.5 Ultracel10 centrifugal filters (Millipore) to remove the protein and resin before HPLC injection. The sample was separated using the elution conditions shown (Table 19.1) and detected by the absorbance of uridine at 262 nm. Under these conditions the retention time for uridine, UMP, UDP-Galf, UDP, and UTP were 6.0, 9.7, 14.5, 16.5, and 26.0 min, respectively. Completion of the reaction was indicated by the presence of the peak for UDP-Galf and depletion of the UTP peak. If the reaction did not reach completion after 16 h, additional Galf-1-P (10 mL, 100 mg/mL), GalU (10 mL, 19 mg/mL), IPP (10 mL, 0.25 U/m L), and immobilized GalPUT (1 mL) were added and the reaction monitored hourly until complete consumption of the UTP was observed. Purification of the resulting UDP-Galf was achieved following the procedure of Rose et al. (2008) with few modifications. The immobilized GalPUT was removed by transferring the reaction mixture to a 10 mL BD column cartridge and washed with 3–5 mL of cold water (4 C). The flow through was transferred to an AmiconÒ Ultra-15 Ultracel-10 centrifugal filter (Millipore) and centrifuged to remove the IPP and GalU proteins. An additional

397


Table 19.1 Analytical HPLC elution conditions

a b

Time (min)

Flow rate (mL/min)

% Buffer Aa

% Buffer Bb

0 10 25 35 36 45

0.8 0.8 0.8 0.8 0.8 0.8

96 96 0 0 96 96

4 4 100 100 4 4

200 mM triethylammonium acetate, pH 6.6. 200 mM triethylammonium acetate, pH 6.6, with 1.5% acetonitrile.

Table 19.2 Semipreparative HPLC elution conditions

a

Time (min)

Flow rate (mL/min)

% Buffera

% Water

0 10 18 19 25

6.0 6.0 6.0 6.0 6.0

100 0 0 100 100

0 100 100 0 0

5 mM sodium phosphate buffer, pH 6.80.

3 mL of cold water was added and the mixture was centrifuged for an additional 15 min. This can be repeated a second time as long as the flow through volume does not exceed 15 mL. Salts were removed by gel filtration (Sephadex G-15, 25 1100 mm, 0.5 mL/min) eluting with cold water (4 C). Elution of the product was detected by absorbance at 262 nm and typically elutes after 180–240 mL in a volume of approximately 50–60 mL. The volume was reduced by evaporation of the water under reduced pressure, ensuring the temperature remained under 25 C, to a final volume of 5–10 mL. The sample was purified by semipreparative HPLC (Varian Microsorb 300-5 C18, 21.4 250 mm) using the elution conditions shown (Table 19.2) and detected by the nucleotide absorbance at 262 nm. Injection volumes of 3 mL were used and under these conditions the desired fraction typically elutes at 17.5 min. Salts were again removed by gel filtration (see above). This step was essential to prevent the decomposition of UDP-Galf observed when stored under high salt concentration. The samples were then lyophilized and the resulting solid was stored under desiccation at –20 C. Under these conditions, UDP-Galf was stable for >4 months. This procedure typically yields between 18 and 22 mg (67–82%) of UDPGalf disodium salt (610.27 g/mol). The purity of these samples can be assessed by 1H NMR spectroscopy and ESI mass spectrometry (Rose et al., 2008).

398


2.2.4. Generation of UDP-galactofuranose derivatives The three-enzyme chemoenzymatic reaction described above has also been used for the preparation of UDP-Galf derivatives such as uridine-50 -diphospho6-deoxy-D-galactofuranose (UDP-fucofuranose, UDP-Fucf ) (4), uridine-50 diphospho-L-arabinofuranose (UDP-Araf ) (6), and uridine-50 -diphospho-5deoxy-D-galactofuranose (UDP-5d-Galf ) (10) (Scheme 19.1B). The preparation of these derivatives typically required longer reaction times to reach completion (30–40 h); however, reactions should not be left for greater than 48 h. Otherwise, significant decomposition of the reactants and products to form UMP and uridine was observed. The retention times for (4), (6), and (10) were found to be 21.5, 15.5, and 19.0 min, respectively, using the analytical HPLC conditions (Table 19.1). Typical yields for these compounds vary from 35% to 50%. A three-enzyme chemoenzymatic reaction employing soluble GalPUT has also been described and used to synthesize (2), (4), (6), and additionally uridine-50 -diphospho-6-deoxy-6-fluoro-D-galactofuranose (UDP-6F-Galf ) (8) but will not be discussed here. For more information on this method see the work of Ferrie`res and coworkers (Peltier et al., 2008b).

3. Pyranose–Furanose Mutases Involved in Furanose Nucleotide Biosynthesis 3.1. Discussion Pyranose–furanose mutase enzymes are flavoproteins that catalyze the ring contraction involved in the biosynthesis of furanose sugar nucleotides from the corresponding pyranose sugar nucleotides (Scheme 19.2) using a unique catalytic mechanism (Sanders et al., 2001; Soltero-Higgin et al., 2004a). The pyranose–furanose interconversion favors the pyranose ring conformation in a ratio of approximately 9:1 (Richards and Lowary, 2009); as a result, assays are typically run in the reverse direction using a NDP-furanose as the substrate. In addition, pyranose–furanose mutases

HO ~90%

HO

OH O HO OUDP 11

UGM

HO O HO HO

OUDP ~10% OH 2

Scheme 19.2 UGM catalyzed ring contraction of UDP-Galp (11) to UDP-Galf (2). The reaction favors the pyranose ring conformation in a ratio of 9:1.


399

are only active when the flavin adenine dinucleotide (FAD) cofactor is in the reduced state (Sanders et al., 2001); therefore, freshly prepared sodium dithionite (20 mM ) is included in the reaction mixture. The most commonly used assay for the enzyme is the HPLC method described below; however, other assays involving radiochemical detection and fluorescence polarization have also been reported (Carlson et al., 2006; Scherman et al., 2003; Soltero-Higgin et al., 2004b).

3.2. HPLC assays for UGM activity 3.2.1. Principle The assays described here are used to measure the activity and kinetics of UGM proteins (as well as other pyranose–furanose mutase enzymes) based on the formation of NDP-pyranose from the corresponding NDP-furanose catalyzed by the enzyme (Scheme 19.2). The two ring forms can be separated using reversed phase C18 (Zhang and Liu, 2000) or anion exchange (Lee et al., 1996) HPLC, and detected using the absorbance of the nucleotide at 262 nm. The procedure described below is based on the C18 HPLC method (with minor modifications), which is applied to the measurement of the activity of E. coli UGM (EC 5.4.99.9) using UDP-Galf as the substrate. 3.2.2. Materials and methods Uridine 50 -diphosphate (UDP) disodium salt and sodium dithionite were obtained from Sigma-Aldrich and J. T. Baker Chemicals, respectively. UDP-Galf disodium salt was prepared using the three-enzyme chemoenzymatic reaction described above. All other chemicals and biochemicals were reagent grade and used without further purification. E. coli UGM was prepared as described previously (Poulin et al., 2010) and will not be discussed in detail here. In brief, the glf gene was cloned from E. coli K-1 strain VW187 (Marolda et al., 1990) and inserted into a pET22b expression vector (Novagen) using NdeI and XhoI restriction sites. Expression of soluble UGM containing C-terminal His-Tag in E. coli BL21 was observed after induction with 0.5 mM IPTG and growth from 2 h at room temperature (22 C). Best results were obtained when expression was induced at an OD600 of 0.6 and the broth was cooled on ice (20 min) prior to the addition of IPTG. The protein was purified using Ni–NTA agarose as previously published (Poulin et al., 2010) and typically yields 5–7 mg of protein per liter of culture in greater than 90% purity. Assay reagents UGM reaction buffer, 100 mM potassium phosphate, pH 7.4 UDP-Galf, 2.0 mM in UGM reaction buffer Sodium dithionite, 60 mM in UGM reaction buffer E. coli UGM, 1.1 mM in UGM reaction buffer

400


The 2.0 mM stock solution of UDP-Galf was prepared by dissolving UDP-Galf (1.22 mg/mL) in UGM reaction buffer. The exact concentration was calibrated by HPLC (Varian Microsorb 100-5 C18, 4.6 250 mm) by coinjection with a known concentration of UDP. Separation was achieved using an isocratic elution of 50 mM triethylammonium acetate (TEAA), pH 6.6, containing 1.5% acetonitrile (0.6 mL/min) and the products were detected by the absorbance of UDP at 262 nm. All other UDP-Galf stock solutions (1000, 500, 200, 100, 50, 25, and 20 mM) were prepared by serial dilution of the 2.0 mM stock solution. 3.2.3. Assay procedure An enzymatic reaction containing 183 nM UGM (10 mL of 1.1 mM solution), 20 mM freshly prepared sodium dithionite (20 mL of 60 mM solution), and 1.0 mM UDP-Galf (30 mL of 2.0 mM solution) was prepared to a final volume of 60 mL UGM reaction buffer. The reaction was incubated at 37 C for 5 min then quenched by heating at 90 C for 5 min. The denatured protein was removed and the sample was filtered by centrifugation (Eppendorf centrifuge, 16,000g, 30 min, 4 C) using AmiconÒ Ultra-0.5 Ultracel-10-centrifugal filters (Millipore) before being injected into the HPLC (Varian Microsorb 100-5 C18, 4.6 250 mm). HPLC separation of the reaction mixture was achieved using an isocratic elution of 50 mM TEAA, pH 6.6, containing 1.5% acetonitrile (0.6 mL/min) and the products were detected by the absorbance of uridine at 262 nm (Fig. 19.5). The appearance of a new product peak corresponding to UDP-Galp and depletion of the UDP-Galf peak signifies UGM activity. Standard samples of UDP-Galp and UDP-Galf were also run under identical HPLC conditions to confirm the identity of the observed peaks. Under these conditions, UDP-Galp and UDP-Galf eluted at approximately 10.5 and 13.2 min, respectively. The exact retention times may vary from day to day depending on the exact concentration and pH of the TEAA buffer and amount of acetonitrile used in the mobile phase; therefore, UDP-Galf and UDP-Galp standards were also run. 3.2.4. Kinetic assay Reactions were prepared containing 4.8 nM UGM (5 mL of 28.2 nM solution), 20 mM sodium dithionite (10 mL of 60 mM solution), and UDP-Galf (500, 250, 100, 50, 25, 12.5, and 10 mM; 15 mL of 1000, 500, 200, 100, 50, 25, and 20 mM solutions, respectively) in a final volume of 30 mL UGM reaction buffer. The reactions were incubated for 2.0 min at 37 C, and then quenched by heating at 90 C for 5 min. The denatured protein was removed by centrifugation (Eppendorf centrifuge, 16,000g, 30 min, 4 C) using AmiconÒ Ultra-0.5 Ultracel-10-centrifugal filters (Millipore). Each reaction was analyzed by HPLC as described above. The amount of UDP-Galp produced per minute was calculated from the

401


0.8 UDP-Galf 0.7

UDP-Galp

Absorbance at 262 nm

0.6 0.5 0.4 0.3 0.2 i. 0.1

ii.

0.0 iii. 0.0

2.0

4.0

6.0

8.0

10.0 12.0 14.0 Time (min)

16.

0

18.0 20.0 22.0

Figure 19.5 UGM activity assay HPLC profile for E. coli UGM reacting with UDPGalf. The peaks for UDP-Galp (10.5 min) and UDP-Galf (13.2 min) are labeled. (i) Control containing 1.0 mM UDP-Galf; (ii) reaction of E. coli UGM with 1.0 mM UDPGalf, depletion of the UDP-Galf peak and formation of a UDP-Galp peak can be clearly seen; (iii) control containing 1.0 mM UDP-Galp.

ratio of product to substrate peaks, and the kinetic parameters were deduced by fitting the data to the Michaelis–Menten equation using the GraphPad Prism software. The exact concentration of UGM used in the kinetics assay can vary; however, the concentration was controlled so that 30% conversion of UDP-Galf was observed. The assays described here have also been used to study bacterial pyranose–furanose mutase proteins from other sources. Assays with preparations of UGM from K. pneumoniae, M. tuberculosis, and D. radiodurans (Beis et al., 2005; Partha et al., 2009a), as well as E. coli K-12 (Sanders et al., 2001; Zhang and Liu, 2000), have been previously reported. These assays have also been adapted to study pyranose–furanose mutase enzymes involved in the biosynthesis of thymidine 50 -diphospho-D-fucofuranose (TDP-Fucf ) (Wang et al., 2008) and uridine 50 -diphopho-2-acetamido-2deoxy-D-galactofuranose (UDP-GalfNAc) (Poulin et al., 2010). Also, studies of UGM activity with various fluorinated UDP-galactose derivatives have been useful for understanding the mechanism and substrate binding of these bacterial enzymes (Burton et al., 1997; Eppe et al., 2009; Errey et al., 2009; Zhang and Liu, 2001).

402


4. Galactofuranosyltransferases Involved in Galactofuranoside Biosynthesis 4.1. Discussion GlfTs are the final enzymes involved in the biosynthesis of Galf-containing glycoconjugates. These enzymes catalyze glycosidic bond formation between the activated UDP-Galf donor and an acceptor species, typically a glycan. Two bifunctional GlfTs, encoded by the Rv3808c and Rv3782 genes, from M. tuberculosis are reported to be required for the biosynthesis of the galactan portion of the mycobacterial mAG complex (Belanova et al., 2008; Mikusˇova´ et al., 2000, 2006). The first, GlfT1, encoded by Rv3782, add the first two Galf residues during the biosynthesis of the mAG complex, indicating that this enzyme has both b-(1!5)and b-(1!6)-transferase activity (Alderwick et al., 2008). The second bifunctional transferase, GlfT2, which is encoded by Rv3808c, adds the remaining Galf residues to the mAG complex with repeating b-(1!5)and b-(1!6)-glycosidic linkages (Rose et al., 2006). The latter enzyme has been shown to possess a single active site that is responsible for both b-(1!5)- and b-(1!6)-transferase activity (Szczepina et al., 2009). Milligram scale reactions with trisaccharides bearing either linkage have yielded oligosaccharides approaching physiological length (Lowary, unpublished observations). In addition, work from the Kiessling lab has also reported processive polymerization by GlfT2 using alternate acceptor substrates (May et al., 2009). Few other GlfT enzymes are reported in the literature, possibly due to the difficulties associated with purifying and studying these GlfT proteins, and problems in accessing substrates for these enzymes. Examples, include the WbbO protein from K. pneumoniae, which has been demonstrated to catalyze the transfer of both Galp and Galf residues to synthesize the galactan I portion of the K. pneumoniae O-antigen (Guan et al., 2001). In addition, in vitro assays with purified Wbbl from E. coli K-12 demonstrate that this enzyme catalyzes the transfer of Galf to an octyl a-glucopyranoside acceptor (Wing et al., 2006). This section describes the procedure for a continuous enzyme-coupled spectrophotometric assay developed to characterize the activity of the GlfT2 enzyme of M. tuberculosis (Rose et al., 2008). The use of this assay to determine both the donor and acceptor kinetics and screen for potential inhibitors is also discussed. Although this assay was developed to study the activity of GlfT2 from M. tuberculosis, it could be modified to study other GlfTs by changing the nature of the carbohydrate acceptor; however, no reports on the use of this assay to study the activity of other GlfT proteins have been reported to date.

403


4.2. Enzyme-coupled spectrophotometric assay for GlfT2 activity 4.2.1. Principle The assay described here uses a known coupled spectrophotometric glycosyltransferase assay (Gosselin et al., 1994), which has been modified to measure the activity of GlfT2. One molecule of UDP is produced as a side product for every molecule of UDP-Galf consumed in the GlfT2 enzymatic reaction (Scheme 19.3). The UDP then reacts with phosphoenolpyruvic acid (PEP) to produce pyruvate and UTP, a reaction catalyzed by pyruvate kinase (PK, EC 2.7.1.40). Pyruvate then serves as a substrate for lactate dehydrogenase (LDH, EC 1.1.1.27) to produce lactate. In the process, one molecule of reduced b-nicotinamide adenine dinucleotide (NADH) is oxidized for every molecule of pyruvate reduced. The assay measures the decrease in the absorbance of NADH at 340 nm. Because PK and LDH are present in large excess, this measure is directly proportional to GlfT2 activity. 4.2.2. Materials and methods BioUltraPure grade 3-(N-morpholino)propanesulfonic acid (MOPS) was obtained from BioShop (Burlington, ON) and magnesium chloride hexahydrate(MgCl2) was obtained from EMD Biosciences (La Jolla, CA). PEP monocyclohexylamine salt, potassium chloride (KCl), PK (type III, GlfT2

(11) + UDP-Galf

(12) +

PK

UDP

PEP HO

O

UTP

NADH HO

O(CH2)7CH3

HO HO HO

O

HO

HO

O HO

O OH OH

HO

O

O(CH2)7CH3

HO

O

O

HO HO

OH O

O

O OH OH

HO

OH

11

O

NAD

Lactate +

HO OH

O

LDH

Pyruvate

OH

12

OH

Scheme 19.3 Reaction scheme for an enzyme-coupled spectrophotometric assay of GlfT2. The activity of GlfT2 is measured based on the decrease in the Abs340 of NADH as it is converted to NADþ. A b-D-Galf-(1!5)-b-D-Galf-(1!6)-b-D-GalfOOct acceptor (11) is shown as well as the resulting b-D-Galf-(1!6)-b-D-Galf-(1!5)b-D-Galf-(1!6)-b-D-Galf-OOct product (12).

404


lyophilized powder, rabbit muscle), LDH (type XI, salt free, rabbit muscle), and reduced b-NADH disodium salt were obtained from Sigma-Aldrich. All other chemicals and biochemicals were of reagent grade and were used without further purification. The acceptor substrate (11) was chemically synthesized as previously described (Completo and Lowary, 2008). 4.2.3. Preparation of GlfT2 protein Various methods have been developed to clone and purify recombinant GlfT2 (Alderwick et al., 2008; Rose et al., 2006, 2008) and they will not be discussed here in detail. Briefly, the Rv3808c gene of M. tuberculosis was cloned and inserted into a pET-15b vector (Rose et al., 2006). Optimal yields of soluble, full-length N-terminal His-tagged GlfT2 were obtained when the pET-15b/Rv3808c vector was expressed in E. coli RosettaTM (DE3) QuartersTM competent cells and purified using Ni–NTA agarose as described previously (Rose et al., 2008), followed by concentration and buffer exchange by ultrafiltration using AmiconÒ Ultra-15 Ultracel-10 centrifugal filters (Millipore) (Sephacryl S-100-HR gel filtration, which was reported in the initial paper describing the assay (Rose et al., 2008) is no longer used). This typically yields 40 mg of GlfT2 per liter of culture volume with a specific activity of 4.3 U/mg GlfT2. The protein can be stored in 10% glycerol at –80 C for 6 months with no significant decrease in activity. Reagent stock solutions MOPS buffer, 1 M, pH 7.6 KCl, 2 M in MilliQ water MgCl2, 1 M in MilliQ water NADH, 15 mM in 100 mM MOPS buffer, pH 7.6 PEP, 100 mM in 250 mM MOPS buffer, pH 7.6 PK, 5.0 U/mL in 100 mM MOPS buffer, pH 7.6 LDH, 16.8 U/mL in 100 mM MOPS buffer, pH 7.6 UDP-Galf, 40 mM in 100 mM MOPS buffer, pH 7.6 Acceptor (11), 40 mM in MilliQ water Inhibitor, 32 mM in MilliQ water GlfT2, 0.75 mg/10 mL in 100 mM MOPS, pH 7.6 4.2.4. General assay procedure The procedures of Rose et al. (2008) were used with minor modifications. The spectrophotometric assays were performed in flat-bottomed 384-well microtiter plates (Corning, NY) that contain 100 mM MOPS, pH 7.6, 50 mM KCl, 20 mM MgCl2, 1.1 mM NADH, 3.5 mM PEP, 3.75 U PK, and 8.4 U LDH per well in a 40 mL final volume. The absorbance values at 340 nm (Abs340) were continuously monitored every 15 s for 5 min at

405


37 C using a Spectra Max 340PC microplate reader employing SOFTmaxÒ PRO software (Molecular Devices, Sunnyvale, CA). The optimum amount of GlfT2 per well was previously determined to be 0.75 mg when using the acceptor trisaccharide (11); however, this amount will vary depending on the nature of the acceptor used. The amount of GlfT2 used in the assay should be such that the initial velocity is linear for at least 2–3 min. Buffer, KCl, MgCl2, and acceptor stocks were stored at 4 C for up to 3 months and other stock solutions listed above were prepared fresh on the day the assay was run. The exact concentration of UDP-Galf and acceptor (11) used for enzyme activity, donor kinetics, acceptor kinetics, and inhibitor screening assays will vary. Table 19.3 details the reagent volumes per well for each type of assay. 4.2.5. Measuring GlfT2 activity The activity of GlfT2 was measured under conditions where the concentrations of both substrates were saturating. Two wells (one test and one no acceptor blank) were prepared in triplicate as described in Table 19.3, and the temperature was raised to 37 C. The reaction was initiated by adding GlfT2 (0.75 mg in 10 mL of 100 mM MOPS, pH 7.6) to all wells and then measured as described above. The blank reactions were essential to correct for any GlfT2-independent absorbance decrease and should be performed in parallel with all GlfT2 assays described here. Table 19.3 Reagent volumes used for GlfT2 assays Reagent volume used per well (mL)

a

Reagent stock concentration

GlfT2 activity

Donor kinetics

Acceptor kinetics

Inhibitor screen

MOPS buffer, 1 M, pH 7.6 KCl, 2 M MgCl2, 1 M NADH, 15 mM PEP, 100 mM PK, 5.0 U/mL LDH, 16.8 U/mL UDP-Galf, 40 mM Acceptor (X), 40 mM Inhibitor, 32 mM MilliQ water Final volume

1.48

1.48

1.48

1.48

1.00 0.80 2.94 1.40 1.50 1.00 3.00 2.00 – 15.88 30

1.00 0.80 2.94 1.40 1.50 1.00 –a 2.00 – 12.88 25

1.00 0.80 2.94 1.40 1.50 1.00 3.00 –a – 11.88 25

1.00 0.80 2.94 1.40 1.50 1.00 –a –a 5.00 –a 30

See text.

406


4.2.6. Donor kinetics assay Ten wells were used to measure donor kinetics. First, 5 mL serial dilutions of the UDP-Galf stock solution were prepared in MilliQ water so that the final concentrations of UDP-Galf in each well (based on 40 mL final volume) was 4000, 2000, 1000, 750, 500, 375, 250, 125, 62.5, and 0 mM, respectively. Twenty-five microliters of the reaction mixture described in Table 19.3 was added, and the temperature was raised to 37 C. Reactions were initiated by the addition of GlfT2 (0.75 mg in 10 mL of 100 mM MOPS, pH 7.6) and measured as described above. Reaction velocities in the 0 mM donor wells were used to correct for GlfT2-independent background rates. 4.2.7. Acceptor kinetics assay Nine wells were used to measure acceptor kinetics. First, 5 mL serial dilutions of the acceptor (11) stock were prepared in MilliQ water so that the final concentration of acceptor in each well (based on 40 mL final volume) was 8000, 6000, 4000, 2000, 1000, 500, 250, 125, and 0 mM, respectively. Twenty-five microliters of the reaction mixture described in Table 19.3 was added, and the temperature was raised to 37 C. The reactions were again initiated by the addition of GlfT2 enzyme (0.75 mg in 10 mL of 100 mM MOPS, pH 7.6) and measured as described above. Reaction velocities in the 0 mM acceptor wells were used to correct for GlfT2-independent background rates. 4.2.8. Acceptor inhibitor screen One well (in triplicate) was required for each inhibitor being screened and was prepared as outlined in Table 19.3 for a final inhibitor concentration of 4.0 mM. A positive control well was also prepared (in triplicate), as described for acceptor kinetics assay in Table 19.3, which contains 2.0 mM acceptor (11). All wells contain 3.0 mM UDP-Galf. The reactions were initiated by the addition of GlfT2 (0.5 mg in 10 mL of 100 mM MOPS, pH 7.6) and measured as described above. The percentage of inhibition was assessed by comparing the reaction velocities of the inhibitor wells to that of the positive controls. A slower rate indicates inhibition, whereas a faster rate indicates that the inhibitor may function as a substrate for the GlfT2 enzyme. 4.2.9. Donor inhibitor screen These assays were performed as above but UDP-Galf was present at 0.375 mM and acceptor (11) was at 2.0 mM. It should be appreciated that donor analogs can contain free UDP, as a result of hydrolysis during synthesis or storage. It was therefore helpful to perform donor analog checks in the absence of both UDP-Galf and GlfT2. Contaminating UDP will be seen as a rapid drop of absorbance at 340 nm, or as a much lower initial absorbance value.


407

ACKNOWLEDGMENTS This work was supported by the Alberta Ingenuity Centre for Carbohydrate Science and the Natural Sciences and Engineering Research Council of Canada. M.B.P. is supported by a Ph.D. studentship from Alberta Innovates–Technology Futures.

REFERENCES Alderwick, L. J., Dover, L. G., Veerapen, N., Gurcha, S. S., Kremer, L., Roper, D. L., Pathak, A. K., Reynolds, R. C., and Besra, G. S. (2008). Expression, purification and characterisation of soluble GlfT and the identification of a novel galactofuranosyltransferase Rv3782 involved in priming GlfT-mediated galactan polymerisation in Mycobacterium tuberculosis. Protein Expr. Purif. 58, 332–341. Beis, K., Srikannathasan, V., Liu, H., Fullerton, S. W. B., Bamford, V. A., Sanders, D. A. R., Whitfield, C., McNeil, M. R., and Naismith, J. H. (2005). Crystal structures of Mycobacteria tuberculosis and Klebsiella pneumoniae UDP-galactopyranose mutase in the oxidised state and Klebsiella pneumoniae UDP-galactopyranose mutase in the (Active) reduced state. J. Mol. Biol. 348, 971–982. Belanova, M., Dianiskova, P., Brennan, P. J., Completo, G. C., Rose, N. L., Lowary, T. L., and Mikusova, K. (2008). Galactosyl transferases in mycobacterial cell wall synthesis. J. Bacteriol. 190, 1141–1145. Berg, S., Kaur, D., Jackson, M., and Brennan, P. J. (2007). The glycosyltransferases of Mycobacterium tuberculosis—Roles in the synthesis of arabinogalactan, lipoarabinomannan, and other glycoconjugates. Glycobiology 17, 35R–56R. Berst, M., Hellerqvist, C. G., Lindberg, B., Lu¨deritz, O., Svensson, S., and Westphal, O. (1969). Structural investigations on T1 lipopolysaccharides. Eur. J. Biochem. 11, 353–359. Bhamidi, S., Scherman, M. S., Rithner, C. D., Prenni, J. E., Chatterjee, D., Khoo, K. H., and McNeil, M. R. (2008). The identification and location of succinyl residues and the characterization of the interior arabinan region allow for a model of the complete primary structure of Mycobacterium tuberculosis mycolyl arabinogalactan. J. Biol. Chem. 283, 12992–13000. Brennan, P. J., and Nikaido, H. (1995). The envelope of mycobacteria. Annu. Rev. Biochem. 64, 29–63. Burton, A., Wyatt, P., and Boons, G. J. (1997). Preparation of fluorinated galactosyl nucleoside diphosphates to study the mechanism of the enzyme galactopyranose mutase. J. Chem. Soc. Perkin Trans. 1, 2375–2382. Carlson, E. E., May, J. F., and Kiessling, L. L. (2006). Chemical probes of UDP-galactopyranose mutase. Chem. Biol. 13, 825–837. Completo, G. C., and Lowary, T. L. (2008). Synthesis of galactofuranose-containing acceptor substrates for mycobacterial galactofuranosyltransferases. J. Org. Chem. 73, 4513–4525. de Lederkremer, R. M., Nahmad, V. B., and Varela, O. (1994). Synthesis of a-D-galactofuranosyl phosphate. J. Org. Chem. 59, 690–692. Eppe, G., Peltier, P., Daniellou, R., Nugier-Chauvin, C., Ferrie`res, V., and Vincent, S. P. (2009). Probing UDP-galactopyranose mutase binding pocket: A dramatic effect on substitution of the 6-position of UDP-galactofuranose. Bioorg. Med. Chem. Lett. 19, 814–816. Errey, J. C., Mukhopadhyay, B., Kartha, K. P. R., and Field, R. A. (2004). Flexible enzymatic and chemo-enzymatic approaches to a broad range of uridine-diphosphosugars. Chem. Commun. 7, 2706–2707.

408


Errey, J. C., Mann, M. C., Fairhurst, S. A., Hill, L., McNeil, M. R., Naismith, J. H., Percy, J. M., Whitfield, C., and Field, R. A. (2009). Sugar nucleotide recognition by Klebsiella pneumoniae UDP-D-galactopyranose mutase: Fluorinated substrates, kinetics and equilibria. Org. Biomol. Chem. 7, 1009–1016. Gosselin, S., Alhussaini, M., Streiff, M. B., Takabayashi, K., and Palcic, M. M. (1994). A continuous spectrophotometric assay for glycosyltransferases. Anal. Biochem. 220, 92–97. Guan, S., Clarke, A. J., and Whitfield, C. (2001). Functional analysis of the galactosyltransferases required for biosynthesis of D-galactan I, a component of the lipopolysaccharide O1 antigen of Klebsiella pneumoniae. J. Bacteriol. 183, 3318–3327. Hanniffy, O. M., Shashkov, A. S., Moran, A. P., Prendergast, M. M., Senchenkova, S. N., Knirel, Y. A., and Savage, A. V. (1999). Chemical structure of a polysaccharide from Campylobacter jejuni 176.83 (serotype O:41) containing only furanose sugars. Carbohydr. Res. 319, 124–132. Huang, H. R., Scherman, M. S., D’Haeze, W., Vereecke, D., Holsters, M., Crick, D. C., and McNeil, M. R. (2005). Identification and active expression of the Mycobacterium tuberculosis gene encoding 5-phospho-a-D-ribose-1-diphosphate: Decaprenyl-phosphate 5-phosphoribosyltransferase, the first enzyme committed to decaprenylphosphoryl-Darabinose synthesis. J. Biol. Chem. 280, 24539–24543. Klutts, J. S., Hatanaka, K., Pan, Y. T., and Elbein, A. D. (2002). Biosynthesis of D-arabinose in Mycobacterium smegmatis: Specific labeling from D-glucose. Arch. Biochem. Biophys. 398, 229–239. Konishi, T., Takeda, T., Miyazaki, Y., Ohnishi-Kameyama, M., Hayashi, T., O’Neill, M. A., and Ishii, T. (2007). A plant mutase that interconverts UDP-arabinofuranose and UDP-arabinopyranose. Glycobiology 17, 345–354. Ko¨plin, R., Brisson, J. R., and Whitfield, C. (1997). UDP-galactofuranose precursor required for formation of the lipopolysaccharide O antigen of Klebsiella pneumoniae serotype O1 is synthesized by the product of the rfbD(KP01) gene. J. Biol. Chem. 272, 4121–4128. Lee, R., Monsey, D., Weston, A., Duncan, J. K., Rithner, C., and McNeil, M. (1996). Enzymatic synthesis of UDP-galactofuranose and an assay for UDP-galactopyranose mutase based on high-performance liquid chromatography. Anal. Biochem. 242, 1–7. Lee, R. E., Smith, M. D., Nash, R. J., Griffiths, R. C., McNeil, M., Grewal, R. K., Yan, W. X., Besra, G. S., Brennan, P. J., and Fleet, G. W. J. (1997). Inhibition of UDPGal mutase and mycobacterial galactan biosynthesis by pyrrolidine analogues of galactofuranose. Tetrahedron Lett. 38, 6733–6736. Li, M., Shao, J., Xiao, M., and Wang, P. G. (2004). Synthesis of glycoconjugates through biosynthesis pathway engineering. Synthesis of Carbohydrates Through Biotechnology American Chemical Society, Washington, DC. pp. 165–181. Liu, Z. Y., Zhang, J. B., Chen, X., and Wang, P. G. (2002). Combined biosynthetic pathway for de novo production of UDP-galactose: Catalysis with multiple enzymes immobilized on agarose beads. Chembiochem 3, 348–355. Lowary, T. L. (2003). Recent progress towards the identification of inhibitors of mycobacterial cell wall polysaccharide biosynthesis. Mini Rev. Med. Chem. 3, 694–707. Marlow, A. L., and Kiessling, L. L. (2001). Improved chemical synthesis of UDP-galactofuranose. Org. Lett. 3, 2517–2519. Marolda, C. L., Welsh, J., Dafoe, L., and Valvano, M. A. (1990). Genetic analysis of the O7polysaccharide biosynthesis region from the Escherichia coli O7:K1 strain VW187. J. Bacteriol. 172, 3590–3599. May, J. F., Splain, R. A., Brotschi, C., and Kiessling, L. L. (2009). A tethering mechanism for length control in a processive carbohydrate polymerization. Proc. Natl. Acad. Sci. USA 106, 11851–11856. McNally, D. J., Jarrell, H. C., Li, J., Khieu, N. H., Vinogradov, E., Szymanski, C. M., and Brisson, J. R. (2005). The HS:1 serostrain of Campylobacter jejuni has a complex teichoic


409

acid-like capsular polysaccharide with nonstoichiometric fructofuranose branches and Omethyl phosphoramidate groups. FEBS J. 272, 4407–4422. Mikusˇova´, K., Yagi, T., Stern, R., McNeil, M. R., Besra, G. S., Crick, D. C., and Brennan, P. J. (2000). Biosynthesis of the galactan component of the mycobacterial cell wall. J. Biol. Chem. 275, 33890–33897. Mikusˇova´, K., Huang, H. R., Yagi, T., Holsters, M., Vereecke, D., D’Haeze, W., Scherman, M. S., Brennan, P. J., McNeil, M. R., and Crick, D. C. (2005). Decaprenylphosphoryl arabinofuranose, the donor of the D-arabinofuranosyl residues of mycobacterial arabinan, is formed via a two-step epimerization of decaprenylphosphoryl ribose. J. Bacteriol. 187, 8020–8025. Mikusˇova´, K., Belanova, M., Kordulakova, J., Honda, K., McNeil, M. R., Mahapatra, S., Crick, D. C., and Brennan, P. J. (2006). Identification of a novel galactosyl transferase involved in biosynthesis of the mycobacterial cell wall. J. Bacteriol. 188, 6592–6598. Nassau, P. M., Martin, S. L., Brown, R. E., Weston, A., Monsey, D., McNeil, M. R., and Duncan, K. (1996). Galactofuranose biosynthesis in Escherichia coli K-12: Identification and cloning of UDP-galactopyranose mutase. J. Bacteriol. 178, 1047–1052. Nikaido, H., and Sarvas, M. (1971). Biosynthesis of T1 antigen in Salmonella; biosynthesis in a cell-free system. J. Bacteriol. 105, 1073–1082. Pan, F., Jackson, M., Ma, Y. F., and McNeil, M. (2001). Cell wall core galactofuran synthesis is essential for growth of mycobacteria. J. Bacteriol. 183, 3991–3998. Partha, S. K., Bonderoff, S. A., van Straaten, K. E., and Sanders, D. A. R. (2009a). Expression, purification and preliminary X-ray crystallographic analysis of UDP-galactopyranose mutase from Deinococcus radiodurans. Acta Crystallogr. F Struct. Biol. Cryst. Commun. 65, 843–845. Partha, S. K., van Straaten, K. E., and Sanders, D. A. R. (2009b). Structural basis of substrate binding to UDP-galactopyranose mutase: Crystal structures in the reduced and oxidized state complexed with UDP-galactopyranose and UDP. J. Mol. Biol. 394, 864–877. Pedersen, L. L., and Turco, S. J. (2003). Galactofuranose metabolism: A potential target for antimicrobial chemotherapy. Cell. Mol. Life Sci. 60, 259–266. Peltier, P., Daniellou, R., Nugier-Chauvin, C., and Ferrie`res, V. (2007). Versatile synthesis of rare nucleotide furanoses. Org. Lett. 9, 5227–5230. Peltier, P., Euzen, R., Daniellou, R., Nugier-Chauvin, C., and Ferrie`res, V. (2008a). Recent knowledge and innovations related to hexofuranosides: Structure, synthesis and applications. Carbohydr. Res. 343, 1897–1923. Peltier, P., Guegan, J. P., Daniellou, R., Nugier-Chauvin, C., and Ferrie`res, V. (2008b). Stereoselective chemoenzymatic synthesis of UDP-1,2-cis-furanoses from a, b-furanosyl 1-phosphates. Eur. J. Org. Chem. 2008, 5988–5994. Poulin, M. B., Nothaft, H., Hug, I., Feldman, M. F., Szymanski, C. M., and Lowary, T. L. (2010). Characterization of a bifunctional pyranose-furanose mutase from Campylobacter jejuni 11168. J. Biol. Chem. 285, 493–501. Reiter, W. D., and Vanzin, G. F. (2001). Molecular genetics of nucleotide sugar interconversion pathways in plants. Plant Mol. Biol. 47, 95–113. Richards, M. R., and Lowary, T. L. (2009). Chemistry and biology of galactofuranosecontaining polysaccharides. Chembiochem 10, 1920–1938. Rose, N. L., Completo, G. C., Lin, S. J., McNeil, M., Palcic, M. M., and Lowary, T. L. (2006). Expression, purification, and characterization of a galactofuranosyltransferase involved in Mycobacterium tuberculosis arabinogalactan biosynthesis. J. Am. Chem. Soc. 128, 6721–6729. Rose, N. L., Zheng, R. B., Pearcey, J., Zhou, R., Completo, G. C., and Lowary, T. L. (2008). Development of a coupled spectrophotometric assay for GlfT2, a bifunctional mycobacterial galactofuranosyltransferase. Carbohydr. Res. 343, 2130–2139.

410


Sanders, D. A. R., Staines, A. G., McMahon, S. A., McNeil, M. R., Whitfield, C., and Naismith, J. H. (2001). UDP-galactopyranose mutase has a novel structure and mechanism. Nat. Struct. Biol. 8, 858–863. Sarvas, M., and Nikaido, H. (1971). Biosynthesis of T1 antigen in Salmonella; origin of D-galactofuranose and D-ribofuranose residues. J. Bacteriol. 105, 1063–1072. Scherman, M. S., KalbeBournonville, L., Bush, D., Deng, L. Y., and McNeil, M. (1996). Polyprenylphosphate-pentoses in mycobacteria are synthesized from 5-phosphoribose pyrophosphate. J. Biol. Chem. 271, 29652–29658. Scherman, M. S., Winans, K. A., Stern, R. J., Jones, V., Bertozzi, C. R., and McNeil, M. R. (2003). Drug targeting Mycobacterium tuberculosis cell wall synthesis: Development of a microtiter plate-based screen for UDP-galactopyranose mutase and identification of an inhibitor from a uridine-based library. Antimicrob. Agents Chemother. 47, 378–382. Singh, S., and Hogan, S. E. (1994). Isolation and characterization of sugar nucleotides from Mycobacterium smegmatis. Microbios 77, 217–222. Soltero-Higgin, M., Carlson, E. E., Gruber, T. D., and Kiessling, L. L. (2004a). A unique catalytic mechanism for UDP-galactopyranose mutase. Nat. Struct. Mol. Biol. 11, 539–543. Soltero-Higgin, M., Carlson, E. E., Phillips, J. H., and Kiessling, L. L. (2004b). Identification of inhibitors for UDP-galactopyranose mutase. J. Am. Chem. Soc. 126, 10532–10533. St. Michael, F., Szymanski, C. M., Li, J. J., Chan, K. H., Khieu, N. H., Larocque, S., Wakarchuk, W. W., Brisson, J. R., and Monteiro, M. A. (2002). The structures of the lipooligosaccharide and capsule polysaccharide of Campylobacter jejuni genome sequenced strain NCTC 11168. Eur. J. Biochem. 269, 5119–5136. Stevenson, G., Neal, B., Liu, D., Hobbs, M., Packer, N. H., Batley, M., Redmond, J. W., Lindquist, L., and Reeves, P. (1994). Structure of the O antigen of Escherichia coli K-12 and the sequence of its rfb gene cluster. J. Bacteriol. 176, 4144–4156. Szczepina, M. G., Zheng, R. B., Completo, G. C., Lowary, T. L., and Pinto, B. M. (2009). STD-NMR studies suggest that two acceptor substrates for GlfT2, a bifunctional galactofuranosyltransferase required for the biosynthesis of Mycobacterium tuberculosis arabinogalactan, compete for the same binding site. Chembiochem 10, 2052–2059. Tam, P. H., and Lowary, T. L. (2009). Recent advances in mycobacterial cell wall glycan biosynthesis. Curr. Opin. Chem. Biol. 13, 618–625. Thibodeaux, C. J., Melancon, C. E., and Liu, H. W. (2008). Natural-product sugar biosynthesis and enzymatic glycodiversification. Angew. Chem. Int. Ed. 47, 9814–9859. Tsvetkov, Y. E., and Nikolaev, A. V. (2000). The first chemical synthesis of UDP-a-Dgalactofuranose. J. Chem. Soc. Perkin Trans. 1, 889–891. Umesiri, F. E., Sanki, A. K., Boucau, J., Ronning, D. R., and Sucheck, S. J. (2010). Recent advances toward the inhibition of mAG and LAM synthesis in Mycobacterium tuberculosis. Med. Res. Rev. 30, 290–326. van Hijum, S., Kralj, S., Ozimek, L. K., Dijkhuizen, L., and van Geel-Schutten, I. G. H. (2006). Structure–function relationships of glucansucrase and fructansucrase enzymes from lactic acid bacteria. Microbiol. Mol. Biol. Rev. 70, 157–176. Velazquez-Hernandez, M. L., Baizabal-Aguirre, V. M., Bravo-Patino, A., CajeroJuarez, M., Chavez-Moctezuma, M. P., and Valdez-Alarcon, J. J. (2009). Microbial fructosyltransferases and the role of fructans. J. Appl. Microbiol. 106, 1763–1778. Wang, Q., Ding, P., Perepelov, A. V., Xu, Y. L., Wang, Y., Knirel, Y. A., Wang, L., and Feng, L. (2008). Characterization of the dTDP-D-fucofuranose biosynthetic pathway in Escherichia coli O52. Mol. Microbiol. 70, 1358–1367. Weston, A., Stern, R. J., Lee, R. E., Nassau, P. M., Monsey, D., Martin, S. L., Scherman, M. S., Besra, G. S., Duncan, K., and McNeil, M. R. (1998). Biosynthetic origin of mycobacterial cell wall galactofuranosyl residues. Tuber. Lung Dis. 78, 123–131. Wing, C., Errey, J. C., Mukhopadhyay, B., Blanchard, J. S., and Field, R. A. (2006). Expression and initial characterization of WbbI, a putative D-Galf: a-D-Glc b-1, 6-galactofuranosyltransferase from Escherichia coli K-12. Org. Biomol. Chem. 4, 3945–3950.


411

Wolucka, B. A. (2008). Biosynthesis of D-arabinose in mycobacteria—a novel bacterial pathway with implications for antimycobacterial therapy. FEBS J. 275, 2691–2711. Wolucka, B. A., and Dehoffmann, E. (1995). The presence of b-D-ribosyl-1-monphosphodecaprenol in mycobacteria. J. Biol. Chem. 270, 20151–20155. Zhang, Q. B., and Liu, H. W. (2000). Studies of UDP-galactopyranose mutase from Escherichia coli: An unusual role of reduced FAD in its catalysis. J. Am. Chem. Soc. 122, 9065–9070. Zhang, Q. B., and Liu, H. W. (2001). Mechanistic investigation of UDP-galactopyranose mutase from Escherichia coli using 2-and 3-fluorinated UDP-galactofuranose as probes. J. Am. Chem. Soc. 123, 6756–6766.

C H A P T E R

T W E N T Y

The Synthesis of 1,2-cis-Amino Containing Oligosaccharides Toward Biological Investigation Shino Manabe Abstract 1,2-cis-Aminoglycoside structure is frequently found in bioactive oligosaccharide. In this chapter, the recent development of 1,2-cis-aminoglycoside preparation is described. The interface investigations between biology and chemistry with using synthetic oligosaccharides are also described.

A major obstacle in glycobiology and glycomedicine is the lack of pure and structurally defined glycoconjugates. These compounds are difficult to isolate from natural resources because they are typically found in low concentrations and in microheterogeneous forms. At the present time, oligosaccharides and glycoconjugates with rigorously defined structures can be obtained only through synthetic chemistry. It is clear that increasing the supply of oligosaccharides will make a significant impact on glycoscience (Boltje et al., 2009; Bongat and Demchenko, 2007; Stallforth et al., 2009; Zhu and Schmidt, 2009). The 1,2-cis-aminoglycoside structure is frequently found in bioactive oligosaccharides. Representative examples of oligosaccharides are glycosylphosphatidylinositol anchor 1, the aminoglycoside antibiotic neomycin 2, and glycosaminoglycan, including hemostat heparin 3. Recently, a multiple cis-aminoglycoside having an oligosaccharide 4 was found in Campylobacter jejuni as an N-linked oligosaccharide (Fig. 20.1). The stereoselective introduction of the glycosidic linkage is the key issue in oligosaccharide synthesis. 1,2-trans-Glycosides can be easily prepared with the anchimeric assistance of a neighboring participant (Schemes 20.1 and 20.2). Once a C-2 acyl or carbamate group carrying donor 5 is activated, the acyl or carbamate group participates in the resulting oxocarbenium ion to form cis-fused five-membered ring 7. Next, the acceptor RIKEN Advanced Science Institute, Saitama, Japan Methods in Enzymology, Volume 478 ISSN 0076-6879, DOI: 10.1016/S0076-6879(10)78020-4

#


413

414

Shino Manabe

H3N

HO HO O HO

OH O

O P O O HO HO HO HO HO

O O

NH2 O

HO HO

H2N HO

O O

O

H2N O

OH

O O

OH O

O HO HO

O HO

OH O

H2N

O

H2N O O O P O O O O O

O

OH O

NH2

NH2 OH

OH R1 OH OH OH

2 Neomycin B

R2 R3

O 1 Glycosylphosphatidylinositol (GPI) anchor OSO3– O O – HO O2C AcHN O HO

O

HO OH O HO AcHN O HO AcHN O

OSO3– O

O OH HO AcHN O O O HO – O3SO 3 CO2– Heparin anticoagulant drug

HO HO HO

OH

O

OH O O OH AcHN O HO AcHN O

O

O

OH

OH O

HO Me AcHN AcHN O

O

O

H N NHAc

ONH

4 N-linked oligosaccharide of campyrobacter jejuni

Figure 20.1 1,2-cis-aminoglycosides containing oligosaccharides.

attacks the cation from the opposite side of the five-membered ring to give the 1,2-trans-glycoside 8. Although 1,2-cis-glycosides are thermodynamically more stable due to the anomeric effect, a reliable methodology for 1,2-cis-glycoside preparation is still lacking. Furthermore, recent rapid oligosaccharide synthesis based on solid-phase and polymer-supported technology requires a more strictly stereocontrolled glycosylation reaction, since

415

1,2-cis-aminoglycosides

R O

X

PO

O

O

O

O

O

O

O

5

OR

PO

O

O

R

O

PO

PO

O

O H

R

R

R

6

7

8 1, 2-trans glycoside

Scheme 20.1 The 1,2-trans-glycosylation reaction with C-2 acyl donor. The 2-acyl group is represented.

O

O

PO O PO

PO

X

9

O PO

OR

10 1, 2-trans glycoside

X

O X

PO OR 11 1, 2-cis glycoside

Scheme 20.2 The 1,2-cis-glycosylation reaction.

purification of stereoisomers after cleavage is rather difficult (Ito and Manabe, 2007; Seeberger, 2008). Preparation of 1,2-cis-amino sugars has remained essentially unchanged since the 1970s. Lemieux and Paulsen reported that 2-azido-2-deoxy glycosides 12 are good glycosyl donors for 1,2-cis-glycosylation to afford aminoglycosides and these are still the first choice for 1,2-cis-aminoglycoside preparation (Lemieux and Ratcliffe, 1979; Paulsen et al., 1978). Unfortunately, a major problem of the 2-azido donor 12 is its moderate cisselectivity in the glycosylation reactions, which, in some cases, can be improved with the aid of solvent effects (Demchenko et al., 1997; Hashimoto et al., 1984). Furthermore, the preparation of such 2-azido glycosyl donors remains problematic because the reaction from glucal 13 using NaN3 and ceric ammonium nitrate in CH3CN lacks stereoselectivity while generating many by-products, hence purification by silica gel column chromatography is necessary (Fig. 20.2). An alternative preparation of 2-azido sugars from 2-amino sugar 14 requires the potentially explosive TfN3 (Alper et al., 1996; Titz et al., 2006; Vasella et al., 1991). Furthermore, a multistep synthetic sequence is required for the preparation of 2-azido sugars from mannose 15 via 2-triflate (Pozsgay, 2001).

416

Shino Manabe

O

PO

X

NH2 14

PO

TfN3, DMAP, CH3CN or, TfN3, K2CO3, CuSO4, H2O

CAN NaN3 O CH3CN PO 13

O N3 12

X

1) Tf2O, base, CH2Cl2 PO 2) NaN3, DMF

OH O

X

15

X = leaving group SR, SAr, imidate, F, Cl etc.

Figure 20.2

Preparation of 2-azide carrying glycosyl donor.

In order to achieve a high yielding and 1,2-cis-selective glycosylation reaction, several demands must be fulfilled: (1) stable under various reaction conditions, including basic conditions, that are necessary for preparation of the sugar unit; (2) stable under Lewis acidic glycosylation reaction conditions; and (3) readily and selectively removable after glycosylation reaction. Additionally, interaction from 2-position with the oxocarbenium ion intermediate should be suppressed in order to enhance 1,2-cis-stereoselective glycosylation. Despite the progress made in synthetic organic chemistry, the lack of any candidates that can fulfill all of the above criteria, other than the azido group, has prevented the development of alternate protecting groups for 1,2-cis-glycosylation reactions. In this chapter, recent progress in the 1,2-cis-glycosylation reaction of amino sugars and the preparation of 1,2-cis-amino sugars containing oligosaccharides and their biological importance is surveyed. The imine-protecting group has not been a major glycosyl donor because it is not stable under many conditions, especially under acidic conditions. The p-methoxybenzylidene protected glycosyl trichloroacetimidate donor 16 exhibits high a-selectivity by using Ni(4-FPhCN)4(OTf)2 as a catalyst to give disaccharide 18 in favor of the a-product (Scheme 20.3) (Mensah and Nguyen, 2009). The choice of the mediator is important. Hence, the use of AgOTf as a mediator gave b-selective glycosylation, while HgCN did not exhibit any selectivity. FeCl3 mediates the a-glycosylation of N-acetyl glucosamine pentaacetate in a 100 g-scale reaction in refluxing ClCH2CH2Cl (Wei et al., 2008). During the reaction, anomerization from the b-direction to the a-direction may occur under acidic conditions (vide infra). The cis-GalNAc-Ser/Thr motif is found in O-linked glycoproteins. Gin constructed the a-selective GalNAc-Ser/Thr motif under basic conditions (Scheme 20.4) (Ryan and Gin, 2008). The p-nosyl protected aziridine 20

417


OAc O

AcO AcO

N O

+ CCl3

Ni(4-F-PhCN)4(OTf)2 CH2Cl2, 25 °C, 12 h

HO O BnO BnO BnO OMe

NH MeO

17

77% a:b = 20:1

O BnO

18

OBn O

+ p-Ns N OH NHAc

19

NHBn

BnO

OBn O

O NHAc

O

20 Base KH NaH

OMe

The a-selective glycosylation reaction. BnO

BnO

N O BnO BnO

MeO

16

Scheme 20.3

BnO

OAc O

AcO AcO

Solvent DMF THF

21 21a:22b 10:1 1:20

HN p-Ns

NHBn O

Yield(%) 73 77

Scheme 20.4 Preparation of the a-Gal-Ser/Thr motif via aziridine opening under basic conditions.

reacts with 2-acetamide GalNAc hemiacetal 19 under basic conditions. Interestingly, the selectivity was influenced by the choice of base and solvent. Namely, by using NaH in THF, b-selectivity was observed, whereas a-selectivity was observed when KH and DMF were used. Schmidt prepared the cis-glycosyl linkage from 2-nitroglycal by conjugate addition (Scheme 20.5; Winterfeld and Schmidt, 2001). The nitroglycal 23 was prepared from galactal 21 by addition of acetyl nitrate and subsequent elimination of acetic acid in a one-pot operation. The Michael addition between Ser or Thr derivative and nitroglycal was carried out in the presence of 0.1 equivalent of t-BuOK in toluene. The bulky protecting TBDPS group at the 6-position of compound 22 enhances the reactivity and steroselectivity. After removal of the TBDPS group, sialic acid derivative 26 was introduced in EtCN. The nitro group was reduced by Raney Ni, then protecting group manipulation gave the building block 28 suitable for glycopeptide synthesis. Hashimoto reported that 2-azido carrying glycosyl diphenyl phosphate 29 showed high a-selectivity with HClO4 in dioxane-Et2O. By using this methodology, protected galactosyl serine and threonine were prepared at room temperature with high a-selectivity (Scheme 20.6; Koshiba et al., 2008).

418

Shino Manabe

NHBoc BnO

OTBDPS O

BnO

(1) HNO3, Ac2O (2) Et3N, CH2Cl2

BnO

OTBDPS O

BnO

O2N

O 25

CO2Me

AcHN AcO AcO BnO

O

O

CO2Me

O2N

OAc 26

AcO

O

BnO

O

AcHN AcO

OAc

AcO

(2) TESOTf, EtCN 47% OAc AcO OP(OEt)2

O

KO Bu Toluene 97%

NO2

(1) TBAF, AcOH THF NHBoc OtBu

O

t

23

OTBDPS O

24

BnO

84% 22

BnO

OtBu

HO

O

NHBoc OtBu

27

O

OAc CO2Me

O (1) Raney-Ni AcHN AcO AcO H2, EtOH BnO (2) Ac2O, pyridine

O

O

75%

BnO

NHBoc

AcHN

OtBu

O

28

O

Scheme 20.5 Preparation of O-linked oligosaccharide via nitroglycal methodology.

AcO AcO

OAc O N3 29

O

P(OPh)2 O

HClO4 5AMS dioxane, Et2O 82% a :b 91:9

AcO AcO

OAc O N3

O

FmocHN

CO2Bn

30

Scheme 20.6 a-Selective glycosylation with phosphate donor.

Demchenko demonstrated that 2-azido-2-deoxy glucopyranosyl thiocyanate 31 was a useful a-glycosylation donor, although the number of examples was limited (Scheme 20.7; Kochetkov et al., 1993). Ando and Kiso reported the 4,6-silylene group is effective for a-selective glycosylation due to steric hindrance of the protecting group. Even with the

419


Troc group, which can be expected for neighboring group assistance at the C2 position, high a-selectivity was observed (Scheme 20.8; Imamura et al., 2003, 2008). Boons reported sulfides are good external additives for cis-glycosylation involving azido donor glycosylation (Park et al., 2007). During the activation of imidate donor 37 in the presence of various sulfides, the a-selectivity was increased (Scheme 20.9). They speculated that sulfides coordinate to the oxocarbenium ion 38 from the b-side to give glycosyl sulfonium ion 39

AcO AcO AcO

TrO S C N + AcO AcO

O N3

O

Si

O O

AcO

+ O

AcO AcO

SPh NHTroc

34

N3

O O AcO AcO AcO 33 OMe

OMe

32

Scheme 20.7

O

81%

AcO

31

AcO AcO AcO

TrClO4 CH2Cl2

The a-glycosylation with thiocyanate donor. NIS TfOH MS4A CH2Cl2

OH O

O

96%

OAc

35

Si

O O

O AcO TrocHN O AcO AcO 36

O

O

OAc

Scheme 20.8 The a-glycosylation with 4,6-silylene protecting group.

AcO AcO

OAc O N3

O

CCl3

TMSOTf

AcO AcO AcO

O

NH

37

38

40

OMe

OAc O

AcO AcO

N3

N3

OBz O

HO BzO BzO

Sulfide

AcO AcO

39 OAc O

O N3BzO BzO

OBz O

41

OMe

Sulfide

Yield (%)

41a / b ratio

Non PhSEt

85

10 / 1 14 / 1

Thiophene

92 95

18 / 1

Scheme 20.9 The glycosylation reaction in the presence of sulfide.

SR2

420

Shino Manabe

based on 1H NMR analyses and density functional theory calculations, and the alcohol attacks the sulfonium ion from the a-side. The electrochemistry is very useful methodology for observation of the reactive species involved in the glycosylation reaction (Nokami et al., 2007, 2009). The thioglycoside 42 was quantitatively converted to the corresponding triflate 43 by electrolysis. The a- and b-glycosyl sulfonium ions 44 were also quantitatively prepared by addition of sulfide to the resulting glycosyl triflate 43. When Me2S was employed as an additive sulfide, both a- and b-glycosyl sulfonium ions 44 were observed at low temperature. After addition of MeOH, it was clearly demonstrating that the reactivity of the a-glycosyl sulfonium ion was higher than that of the b-glycosyl sulfonium ion (Scheme 20.10). In 2001, a novel trans-carbamate glycosyl donor 46 that exhibits high 1,2-cis-selectivity in glycosylation reactions (Scheme 20.11) was reported (Benakli et al., 2001; Boysen et al., 2005). Subsequent reports, however, described the drawbacks of such trans-carbamate donors, including significant side reactions such as sulfenylation from PhSOTf and glycosylation at the nitrogen atom (Kerns et al., 2003; Wei and Kerns, 2005a,b). To avoid such side reactions, the nitrogen atom was protected as an acetimide. Unfortunately, the stereoselectivity was significantly reduced and, in some cases, the reaction generated undesired b-glycoside as a major isomer. In addition, selective cleavage of an acetyl group from an imide is rather difficult (Scheme 20.12). Manabe showed that N-benzyl-2,3-trans-carbamate glycosyl donors 53 possess high cis-selectivity in glycosylation reactions (Manabe et al., 2006). Employing the secondary hydroxy group as a glycosyl acceptor, high

AcO AcO AcO

O

AcO AcO AcO

S

N3

O

Me2S

AcO AcO AcO

N3 OTf 43

O 44

SMe2

AcO MeOH AcO AcO

O

OMe

N3

N3

45

OTf

Scheme 20.10 Preparation of glycosyl sulfonium ion from thioglycoside via triflate by electrolysis.

AcO

OAc O O

NH

SPh +

OAll O

HO BnO BnO

90% OMe

O 46

PhSOTf CH2Cl2

47

AcO

OAc O O O

NH O BnO BnO 48

OAll O OMe

Scheme 20.11 The a-glycosylation reaction using 2,3-trans-carbamate having donor.

421


MeO2C AcO AcO

O

O

OAc

MeO2C AcO AcO

O

X +

OAc

HO O

OBn O

OBn O O

STol

NH

O 51 4 0 – 60% STol

NH

O

X = Br or imidate 49

50

MeO2C AcO AcO

O OAc

O

OBn O

STol N AcO

O O

O

52 2 0–30%

OAc OAc CO2Me

Scheme 20.12 Disadvantage of 2,3-trans-carbamate donor.

BnO ClAcO O

O

O 53

N Bn

SPh

+

MeO2C

OBn O

OH

O 54

O

AgOTf PhSCl DTBMP toluene dioxane 71%

MeO2C

BnO

O

OBn O

O NBn

O

O

O O OAcCl 55

Scheme 20.13 The cis-glycosylation reaction with benzylated 2,3-trans-carbamate donor.

a-selectivity was observed near room temperature (Scheme 20.13). When a primary acceptor was employed as a glycosyl acceptor, a-selectivity was achieved with the aid of a solvent effect (toluene–dioxane). As shown in Scheme 20.14, the a component of the immune system stimulating O-polysaccharide from Proteus mirabilis O48 was synthesized by successive 1,2-cis-glycosylations using a one-pot methodology. The bromide 56 was selectively activated in the presence of thioglycoside 57, then the disaccharide 58 was activated by addition of a further amount of AgOTf and PhSCl to give the trisaccharide 60. This was the first example showing the preparation of two cis-glycoside bonds in a one-pot operation. Ye reported the N-acetyl 2,3-oxazolidinone-protected thioglycoside can be activated by a benzensulfinyl morpholine–Tf2O combination at 73 C (Yiqun et al., 2008a,b). Interestingly, in the presence of 2,4,6-tritert-butylpyrimidine as a base, complete b-selectivity was observed, while in the absence of 2,4,6-tri-tert-butylpyrimidine, complete a-selectivity was

422

Shino Manabe

observed. This a-selectivity may be due to anomerization via endocyclic cleavage under acidic conditions. Since the glycosides are symmetric acetals, there are two possibilities for cleavage of the acetal. The common mode of cleavage of glycosides is exocyclic cleavage where the bond between the anomeric carbon and the exocyclic oxygen breaks to give the cyclic oxacarbonium ion (Scheme 20.15, path A). This type of cleavage is a well-accepted mechanism whose importance has been clearly seen in the great success of oligosaccharide synthesis. The second is endocyclic cleavage, where the bond between the anomeric carbon and the pyranose ring oxygen breaks (Scheme 20.15, path B). Post and Karplus hypothesized a mechanism for the endo-mode hydrolysis of oligosaccharides based on molecular dynamics calculations on lysozyme where the conformation of the N-acetylglucopyranoside was restricted to the chair form (Post and Karplus, 1986). Since then, several studies have reported on endocyclic cleavage of pyranosides.

HO

OBn O

ClAcO O

+ N Br Bn

O

OBn O

O

ClAcO O

SPh

O

N Bn

O

56

AgOTf DTBMP toluene dioxane

OBn O N O Bn

OBn O

O

57

SPh

N Bn

O 58

ClAcO O O

AgOTf PhSCl OBn 81% O N O Bn O O

OBn O

AcO HO

OMP NPhth

59

OBn O N AcO O Bn

OBn O

OMP NPhth

60

Scheme 20.14 The two cis-glycosyl bond formation in a one-pot reaction.

O

Exocyclic cleavage H+ Path A

62

Path B O OR Path A

61

Endcyclic cleavage

OH

H+ Path B

63

Scheme 20.15 Endocyclic and exocyclic cleavages.

OR

423


Crich (Crich and Jayalath, 2005), Oscarson (Boysen et al., 2005; Olsson et al., 2008), and Manabe (Manabe et al., 2006) independently showed that pyranosides with 2,3-trans-carbamate carrying pyranosides are easily anomerized from the b- to the a-direction under weak acidic conditions. Manabe showed clear evidence that this anomerization is caused via endocyclic cleavage and recyclization (Manabe et al., 2009). In the presence of reducing reagent Et3SiH, the reduced alcohol 68 was obtained as well as the starting material b-glycoside 64 and anomerized a-glycoside 67 (Scheme 20.16). The cations generated by endocyclic cleavage of 65 and 66 were reduced by Et3SiH to give the reduced product 68. In addition, the cation 65 generated via endocyclic cleavage could be trapped by intra- and intermolecular Friedel–Crafts reactions. The feasibility of endocyclic cleavage reaction with 2,3-trans-carbamate group was supported by density functional theory calculations (Satoh et al., 2009). Synthesis of bioactive oligosaccharides using the 2,3-trans-carbamate donor would be interesting. It was reported that oligosaccharide 71 (Fig. 20.3) exhibits anti-Helicobacter pylori activity. The gram-negative bacterium H. pylori infects the stomachs of nearly half the human population. Since the first demonstration of the potential pathogenic character of H. pylori, accumulated evidence strongly suggests that H. pylori causes gastric ulcers, carcinoma, and cancer (Warren and Marshall, 1983). The antiH. Pylori oligosaccharide 71 has a core-2 branched-type oligosaccharide with a characteristic a-N-acetyl glucosamine at the nonreducing end. It is

PO

PO PO O

OP SPh

PO

SR O

O

O O

OH

N 69

70 Toluene

PO O

OP O YR

H+

PO PO O

OH X

X O

64 b -anomer X = O, or NBn Y = O, S

PO PO O

YR

O

OH X

O

65

PO PO O

66

YR

PO PO O

O

X YR 67 a -anomer

O

Et3SiH OH YR X

O 68 Reduced product

Scheme 20.16 The capture of cations generated by endocyclic cleavage reaction.

424

Shino Manabe

believed that the nonreducing terminal a-1,4-GlcNAc is essential for growth inhibition of H. pylori and that the antibiotic activity is due to biosynthetic inhibition of cholesteryl-a-D-glucoside, an important component of the H. pylori cell membrane (Fig. 20.4). At that time, an antiH. Pylori oligosaccharide was available only as a recombinant glycoprotein (CD43) form, synthesized using a-1,4-N-acetyl glucosamine transferases in Chinese hamster ovary cells.

HO HO HO

O AcHN O HO O HO OH HO

OH O

HO O HO

OH O NHAc

O

HO

OH O

HO

O

AcNH O O

OH

AcHN

OMe

71

Figure 20.3 Anti-Helicobacter pylori oligosaccharide. Cholesterol

HO

HO HO HO

O AcHN

O

OH O

HO O OH HO

HO O

H. pylori

O NHAc

HO HO HO OH

HO HO HO

O O AcNH OH HO O O O AcHN O OH

Protein

O OH

Cholesteryl-a-D-glucoside

Figure 20.4 Mechanism of Helicobacer pylori growth inhibition

425


The synthesis of the anti-H. pylori oligosaccharide was achieved using 2,3-trans-carbamate carrying glycosyl donor (Scheme 20.17) (Manabe et al., 2007). The donor 72 reacts with the acceptor 73 with complete cis-stereoselectivity in quantitative yield. The upper trisaccharide and lower trisaccharide portions were prepared from the common disaccharide 74. The thio-disaccharide 74 was directly activated to give the upper and lower trisaccharides 75 and 76. After transformation to the trisaccharides to acceptor and donor, the two trisaccharides were coupled to give the hexasaccharide in high yield. After sequential deprotection and selective N-acetylation, the hexasaccharide 71 was afforded in an efficient manner. Seeberger’s group also synthesized a pentasaccharide having a 1,2-cisGlcNAc moiety by using a 2-azido group (Wang et al., 2007). The cholesterol a-glucosyltransferase was expressed and enzyme activity inhibition was tested by using the chemically synthesized 1,2-a-GlcNAc having a pentasaccharide (IC50 0.47 mM) (Lee et al., 2006). Furthermore, the a1,4-GlcNAc capped mucin-type O-glycan more efficiently inhibits a-glucosyltransferase than a-glucosyl cholesterol (Kobayashi et al., 2009; Lee et al., 2008). Recently, the gram-negative bacterium C. jejuni was found to contain a unique N-linked protein glycosylation system (Szymanski et al., 1999; Wildt and Gerngross, 2005). The N-glycans 4 of C. jejuni plays an important role in host adherence, invasion, and colonization, although the

HO

OBn O

ClAcO O

+ N

O

AgOTf, DTBMP MS4A dioxane / toluene

Me

OBn O

S

AcO

N

92%

OAc

Br

O

Me

Bn

OBn O

ClAcO O

73

72

74 HO HO HO

OBn O

ClAcO O

N

O

O Bn AcO

Glycosylation and OR deprotetion NPhth

OBn O

O BnO

75

N O

O Bn

O O

AcO OAc

Me

OH O OH

HO HO OH

O O

S

OAc

HO

Ph OBn O

Me

O

O

OBn O

OBn O

AcHN O

OBn O OAc

ClAcO O

O

Bn AcO

O NHAc

HO O HO

O O AcNH

OH O

HO O O

HO OH

AcHN

OMe

71

N3 OMe

76

Scheme 20.17 The synthesis of anti-Helicobacter pylori oligosaccharide using 2,3-transcarbamate donor.

426

Shino Manabe

mechanism is not known (Karlyshev et al., 2004). The key structural feature of the oligosaccharide is a five cis-linked galactosamine motif with bacillosamine (Bac; 2,4-diacetamido-2,4,6-trideoxyglucopyranose) at the end, that is, GalNAc-a1,4-GalNAc-a1,4-[Glc-b1,3-]GalNAc-a1,4-GalNAc-a 1,4-GlaNAc-a1,3-Bac-b1,N-Asn (Young et al., 2002). The glycan is attached to the asparagine amido side chain at the Asn-X-Ser/Thr motif where X can be any amino acid except proline, similar to the eukaryotes system. The N-glycans 4 is encoded by 12 genes that are organized into a protein glycosylation locus, pgl. The heptasaccharide glycan is synthesized in a stepwise manner on an Und-PP carrier (Scheme 20.18). First, PglA adds a GalNAc residue to undecaprenyl phosphate bacillosamine 78 Bac, to yield a disaccharide undecaprenyl phosphate 80. PglA shows relaxed substrate specificity. 6-Hydroxybacillosamine is also transferred to PglA, as well as bacillosamine, but not GlcNAc, in vitro. Next, PglJ adds a second GalNAc residue to form the trisaccharide undecaprenyl phosphate 81. PglH then transfers three more GalNAc residues. Finally, PglI attaches the sidebranched glucose to give the complete heptasaccharide undecaprenyl phosphate 83. Imperiali’s group overexpressed the enzymes involving C. jejuni N-glycan biosynthesis, PglA, PglJ, PglH, and PlgI (Glover et al., 2005a,b). By using the overexpressed enzymes, the C. jejuni N-glycan was synthesized in vitro. N-linked oligosaccharides are usually transferred to amido residues of proteins by oligosaccharyl transferases. The biosynthesis of the N-linked oligosaccharide shows homology to the eukaryotic pathway (Szymanski et al., 2003; Weerapana and Imperiali, 2006). The oligosaccharyl transferase complex of mammals, yeast, and bacteria is a supermolecular machinery system and it transfers the dolicholyl-phosphate-linked oligosaccharide to the nascent polypeptide. Supermolecular complex formation is crucial for oligosaccharyl transferring activity in the case of yeast and mammalian transferases. In contrast to the mammalian and yeast transferases, the oligosaccharyl transferase in C. jejuni was reported to be a single unit, PglB, an 82 kDa integral membrane protein that catalyzes the transfer reaction. Imperiali’s group prepared a membrane fraction from E. coli in which PglB has been overexpressed. Using the PglB together with chemoenzymatically synthesized undecaprenyl-linked GalNAc-bacillosamine disaccharide (Weerapana et al., 2005), several glycopeptides were prepared (Glover, 2005; Chen et al., 2007). While PglB accepts the unnatural 6hydroxybacillosamine and GlcNAc analogs, the bacillosamine substrate is the most effective substrate. Even though the substrate specificity of PglB is relaxed, chitobiose was not transferred to the peptide. Aebi showed that PglB required an acetamido group of the donor for efficient glycan transfer to the protein (Wacker et al., 2006). Using an in vitro N-glycosylation strategy using PglB, a suitable quantity of the 13C/15N-labeled glycoprotein was obtained for NMR structural

427


HO UDP-GlcNAc 77

HO

PgIFED

UDP-Bac 2, 4diNAc UMP Me 78 O AcHN Und-P HO PglA PglC AcHN O–PP-Und Und-PP-Bac2,4-diNAc 79

O

NHAc Me

O

OH O

UDPHO GalNAc UDP AcHN

AcHN O

O

NHAc Me

AcHN

PglJ

O

O

O

PP-Und

PP-Und Und-PP-Bac2,4-diNAc-GalNAc Und-PP-Bac2,4-diNAc-(GalNAc)2 80 81

HO OH O HO AcHN O

3UDPGalNAc

HO AcHN

OH O

HO UDPGalNAc UDP AcHN

OH O

OH O HO AcHN O

OH O HO AcHN O OH UDP O HO AcHN O

3UDP

PglH

OH O HO AcHN ONHAc Me AcHN O O

HO OH O HO AcHN O

OH O HO AcHN O

HO HO HO UDP-Glc UDP

PgII

PP-Und Und-PP-Bac2,4-diNAc-(GalNAc)5 82

OH O O OH AcHN O

O

OH O HO AcHN O

OH O

HO AcHN

O

3

NHAc

Und-P

7

OPO3–

Me AcHN O O PP-Und Und-PP-Bac2,4-diNAc-(GalNAc)2-(GalNAc)Glc-(GalNAc)2 83

Scheme 20.18

Biosynthetic pathway of N-glycan in Campylobacter jejuni.

investigations (Slynko et al., 2009). NMR analysis revealed that the heptasaccharide forms a well-defined rod-like structure, in contrast to the general belief that glycans adapt flexible conformations. Recently, homogeneous eukaryotic N-glycoprotein was prepared by using C. jejuni glycosylation machinery (Schwarz et al., 2010). The glycoprotein with C. jejuni glycan

428

Shino Manabe

was prepared in E. coli. After trimming of the glycan by a-N-acetylgalactosamidase, eukaryotic glycan was transferred with the key GlcNAc-Asn structure by endo-b-N-acetylglucosiminidase using Man3GlcNAc oxazoline as a donor substrate (Li et al., 2005). The chemical synthesis of the oligosaccharide portion of C. jejuni Nglycan was achieved by Ito using an azido donor (Scheme 20.19) (Amin et al., 2007; Ishiwata et al., 2006). From the common precursor 84, the galactosamine donor 85 and Bac derivative 86 were prepared. A pentafluoropropionyl ester as a protecting group was introduced at the 4-position in order to enhance the a-stereoselectivity in glycosylation reaction (Scheme 20.20). The expected a-selectivity was obtained due to the strong

HO

HO AcHN O HO AcHN O

PFPO

85 O

HO

OBn O

BnO

Ph O O

F HO HO N3 HO

OTBDPS

N3 84

O O

O

HO

OH

OH O

O AcHN HO AcHN O

OH

O N3 Me

OH O

O

N3

F

N3

OH

OH O

4

86

OBn O

87

HO Me AcHN AcHN O

OTBDPS

OBn PFPO O O OBn

BnO BnO

O

O

H N NHAc

O

NH

PFP = pentafluoropropionyl

Scheme 20.19 Synthetic analysis of N-glycan in Campylobacter jejuni.

PFPO BnO

OBn O N3

RO BnO

OBn O N3

85 R = PFP 88 R = Ac

RO F

BnO

OBn O

R = PFP

RO BnO

N3 R = Ac R = PFP

BnO

O O OBn O N3 R = Ac

Scheme 20.20 The mechanism of a-selective glycosylation.

OBn O N3 OR'

89 R = PFP 90 R = Ac


429

electron-withdrawing nature of the pentafluoropropionyl group in the axial orientation to neutralize the dipole moment in the transition state. In the case of galactoside glycosylation, Boons reported the axial 4-position protecting group affects the stereoselectivity by a remote neighboring effect (Demchenko et al., 1999). Glycosaminoglycans including heparin and heparin sulfate are heavily Nand O-sulfated linear oligosaccharides. Heparin and heparin sulfate consist of 1,4-disaccharide units of a-L-iduronic or b-D-glucuronic acid and either Nacetyl- or N-sulfo-a-D-glucosamine (Esko and Selleck, 2002). To date, more than 100 heparin sulfate-binding proteins have been identified (Ori et al., 2008). Heparin has been used as an anticoagulant drug isolated from porcine mucosal tissue. Heparin binds to plasma protein antithrombin III causing a conformational change. The activated antithrombin III then inactivates thrombin and other proteases involved in blood clotting. The pentasaccharide, Fondaparinax, is now on the market (Petitou and van Boeckel, 2004). It has been proposed that the spatial organization of negative charge clusters in heparin sulfate is important for binding and biological activity. To investigate the biological activities related to the heparin sulfate-binding protein, the structures of well-defined oligosaccharides are required. However, the availability of glycosaminoglycans is rather scarce due to their complex structure and diversity. Synthetic and chemoenzymatic approaches have the potential to provide sufficient amounts of well-defined oligosaccharides. For the above reason, several excellent synthetic approaches for heparin are reported (Petitou et al., 2001; Lubineau et al., 2004; de Paz and Martin-Lomas, 2005; Zhou et al., 2006; Chen et al., 2008; Noti et al., 2006). Hung reported the efficient synthesis of heparin oligosaccharides (Lee et al., 2004). The disaccharide-repeating unit was prepared from a 2-azido donor 91 and a properly protected L-iduronic acid 92 (Scheme 20.21). The 1,6-idose anhydride 92 was prepared from diacetoneglucose in a few steps. The glycosylation reaction using 2-azido donor 91 gave both a- and b-disaccharides 93a and 93b. After acetolysis of the 1,6-anhydride, 93a was achieved in the presence of a catalytic amount of Cu(OTf)2, the disaccharide 94 was converted to the trichloroacetimidate donor. After elongation of the glycosyl chain, the 6-hydroxy group was oxidized under TEMPO oxidation conditions. After deprotection and sulfation, several heparins of different lengths were obtained. Boons prepared a library of 12 oligosaccharides using properly protected disaccharides (Arungundram et al., 2009). The structure–activity relationship for inhibition of aspartyl protease b-site amyloid precursor proteincleaving enzyme I (BACE-1) was determined using chemically synthesized oligosaccharides. BACE-1 generates a membrane-bound protein, which is further processed by the g-secretase enzyme complex to generate the neurotoxic amyloid b-peptide. The inhibition of BACE-1 would be potentially useful for treatment of Alzheimer’s disease (Asai et al., 2006).

430

Shino Manabe

The last example of interesting activity from a-1,2-cis-amino oligosaccharide is lipoteichoic acid (LTA) of Streptococcus pneumoniae (Fig. 20.5). Recently, Schmidt reported the synthesis of LTA of S. pneumoniae 97 (Pedersen et al., 2010). The pneumococcal LTA is recognized by the innate immune system (Hoebe et al., 2005). Activation is supposed to occur through the Toll-like receptor 2 with CD14 as a coreceptor. (Schlo¨der et al., 2003) However, it is known that Toll-like receptor 2 recognizes a

BzO NAPO BnO

O

CCl3

O

N3

NH

91

BzO NAPO BnO

AcO

BzO NAPO BnO TMSOTf

O

O

O O BnO BzO

92

93a a 61% 93b b 11% BzO NAPO BnO

O N3

O

OAc

AcO

94

O OBn

–

O2C

O

O

OBz

n O BnO

O N3

95 –O SO 3 O HO HO –O SHN 3

O

(1) Deprotection (2) Oxidation (3) Sulfation

OBz

OBz O OBn

Cu(OTf)2 Ac2O

N3

O

(1) Deprotection (2) Donor transformation (3) Elongation

O N3

+

HO BnO BzO

OMe

OSO3– O OH

96

n

OSO3–

O O HO – O3SHN OMe

Scheme 20.21 The heparin synthesis.

H3N

H3N HO HO HO

O OH

H3N Me O

O

O

O P

AcHN O HO

O

O

O HO

O NHAc

O

O P

O

O

OH OH

O O NHAc

OH

97

O O P O O HO HO

O OH

H3N Me O AcHN

O HO O

OH O

O O

HO

O

C13H27 O C13H27 O

Figure 20.5 The structure of lipoteichoic acid of Streptococcus pneumoniae.


431

broad range of structurally different bacterial compounds, and specific stimulation of the immune system though a TLR-2-LTA interaction has been questioned (Hashimoto et al., 2006; Tawaratsumida et al., 2009). The oligosaccharide 97 was synthesized in 88 steps assembled from nine building blocks. The 1,2-cis-aminoglycosides were introduced by using a 2-azide carrying a glycosyl donor. By using the synthetic oligosaccharide 97, it is clearly revealed that neither TLR2 nor TLR4 were the receptors for the oligosaccharide, although 97 shows immunological activity resulting in cytokine release by an unknown mechanism. As described in this chapter, 1,2-cis-aminoglycoside preparation methodology is developing rapidly and complex oligosaccharide synthesis is now possible. Further progress in investigating biological events using the chemically synthesized oligosaccharides is promised.

REFERENCES Alper, P. B., Hung, S.-C., and Wong, C.-H. (1996). Metal catalyzed diazo transfer for the synthesis of azides from amines. Tetrahedron Lett. 37, 6029–6032. Amin, M. N., Ishiwata, A., and Ito, Y. (2007). Synthesis of N-linked glycan derived from Gram-negative bacterium, Campylobacter jejuni. Tetrahedron 63, 8181–8198. Arungundram, S., Al-Mafraji, K., Asong, J., Leach, F. E., 3rd, Amster, I. J., Venot, A., Turnbull, J. E., and Boons, G.-J. (2009). Modular synthesis of heparan sulfate oligosaccharides for structure-activity relationship studies. J. Am. Chem. Soc. 131, 17394–17405. Asai, M., Hattori, C., Iwata, N., Saido, T. C., Sasagawa, N., Szabo´, B., Hashimoto, Y., Maruyama, K., Tanuma, S., Kiso, Y., and Ishiura, S. (2006). The novel b-secretase inhibitor KMI-429 reduces amyloid b peptide production in amyloid precursor protein transgenic and wild-type mice. J. Neurochem. 96, 533–540. Benakli, K., Zha, C., and Kerns, R. J. (2001). Oxazolidinone protected 2-amino-2-deoxyD-glucose derivatives as versatile intermediates in stereoselective oligosaccharide synthesis and the formation of a-linked glycosides. J. Am. Chem. Soc. 123, 9461–9462. Boltje, T. J., Buskus, T., and Boons, G.-J. (2009). Opportunities and challenges in synthetic oligosaccharide and glycoconjugate research. Nat. Chem. 1, 611–622. Bongat, A. F. G., and Demchenko, A. V. (2007). Recent trends in the synthesis of O-glycosides of 2-amino-2-deoxysugars. Carbohydr. Res. 342, 374–406. Boysen, M., Gemma, E., Lahmann, M., and Oscarson, S. (2005). Ethyl 2-acetamido-4, 6-diO-benzyl-2, 3-N, O-carbonyl-2-deoxy-1-thio-a-D-glycopyranoside as a versatile GlcNAc donor. Chem. Commun. 3044–3046. Chen, M. M., Glover, K. J., and Imperiali, B. (2007). From peptide to protein: Comparative analysis of the substrate specificity of N-linked glycosylation in C. jejuni. Biochemistry 46, 5579–5585. Chen, J. F., Zhou, Y., Chen, C., Xu, W. C., and Yu, B. (2008). Synthesis of a tetrasaccharide substrate of heparanase. Carbohydr. Res. 343, 2853–2862. Crich, D., and Jayalath, P. (2005). Stereocontrolled formation of b-glucosides and related linkages in the absence of neighboring group participation: Influence of a trans-fused 2, 3-O-carbonate group. J. Org. Chem. 70, 7252–7259. de Paz, J. L., and Martin-Lomas, M. (2005). Synthesis and biological evaluation of a heparinlike hexasaccharide with the structural motifs for binding to FGF and FGFR. Eur. J. Org. Chem. 1849–1858.

432

Shino Manabe

Demchenko, A., Stauch, T., and Boons, G.-J. (1997). Solvent and other effects on the stereoselectivity of thioglycoside glycosidations. Synlett 818–820. Demchenko, A. V., Rousson, E., and Boons, G.-J. (1999). Stereoselective 1, 2-cis-galactosylation assisted by remote neighboring group participation and solvent effects. Tetrahedron Lett. 40, 6523–6526. Esko, J. D., and Selleck, S. B. (2002). Order out of chaos: Assembly of ligand binding sites in heparan sulfate. Annu. Rev. Biochem. 71, 435–471. Glover, K. J., Weerapana, E., and Imperiali, B. (2005a). In vitro assembly of the undecaprenylpyrophosphate linked heptasaccharide for prokaryotic N-linked glycosylarion. Proc. Natl. Acad. Sci. USA 102, 14255–14259. Glover, K. J., Weerapana, E., Numao, S., and Imperiali, B. (2005b). Chemoenzymatic synthesis of glycopeptides with PglB, a bacterial oligosaccharyl transferase from Campylobacter jejuni. Chem. Biol. 12, 1311–1316. Hashimoto, S., Hayashi, M., and Noyori, R. (1984). Glycosylation using glucopyranosyl fluorides and silicon-based catalysis. Solvent dependency of the stereoselection. Tetrahedron Lett. 25, 1379–1382. Hashimoto, M., Tawaratsumida, K., Kariya, H., Kiyohara, A., Suda, Y., Kirikae, F., Kirikae, T., and Go¨tz, F. (2006). Not lipoteichoic acid but lipoproteins appear to be the dominant immunobiologically active compounds in Staphylococcus aureus. J. Immunol. 177, 3162–3169. Hoebe, K., Georgel, P., Rutshchmann, S., Du, X., Mudd, S., Crozat, K., Sovath, S., Shamel, S., Hartung, T., Za¨hringer, U., and Beutle, B. (2005). CD36 is a sensor of diacylglycerides. Nature 433, 523–527. Imamura, A., Ando, H., Korogi, S., Tanabe, G., Muraoka, O., Ishida, H., and Kiso, M. (2003). Di-tert-butylsilylene (DTBS) group-directed a-selective galactosylation unaffected by C-2 participating functionalities. Tetrahedron Lett. 44, 6725–6728. Imamura, A., Ando, H., Ishida, H., and Kiso, M. (2008). DTBS EFFECT: The unique sterically driven director for a-galactosylation. Heterocycles 76, 883–908. Ishiwata, A., Ohta, S., and Ito, Y. (2006). A stereoselective 1, 2-cis glycosylation toward the synthesis of a novel N-linked glycan from the Gram-negative bacterium, Campylobacter jejuni. Carbohydr. Res. 341, 1557–1573. Ito, Y., and Manabe, S. (2007). Polymer supported oligosaccharide synthesis and related technologies. ( J. P. Karmerling, G.-J. Boons, Y. C. Lee, A. Suzuki, N. Taniguchi, and A. G. J. Voragen, eds.), Vol. I, pp. 335–378. Elsevier, Amsterdam. Karlyshev, A. V., Everest, P., Linton, D., Cawthrw, S., Newell, D. G., and Wren, B. W. (2004). The Campylobacter jejuni general glycosylation system is important for attachment to human epithelial cells and in the colonization of chicks. Microbiology 150, 1957–1964. Kerns, R. J., Zha, C., Benakli, K., and Liang, Y.-Z. (2003). Extended applications and potential limitations of ring-fused 2, 3-oxazolidinone thioglycosides in glycoconjugate synthesis. Tetrahedron Lett. 44, 8069–8072. Kobayashi, M., Lee, H., Nakayama, J., and Fukuda, M. (2009). Roles of gastric mucin-type O-glycans in the pathogenesis of Helicobacter pylori infection. Glycobiology 10, 453–461. Kochetkov, N. K., Klimov, E. M., Malysheva, N. N., and Demchenko, A. V. (1993). Stereospecific synthesis of 1, 2-cis glycosides of 2-amino sugars. Carbohydr. Res. 242, C7–C10. Koshiba, M., Suzuki, N., Arihara, R., Tsuda, T., Nambu, H., Nakamura, S., and Hashimoto, S. (2008). Catalytic stereoselective glycosidation with glycosyl diphenyl phosphates: Rapid construction of 1, 2-cis-a-glycosidic linkages. Chem. Asian J. 3, 1664–1677. Lee, J.-C., Lu, X.-A., Kulkan, S. S., Wen, Y.-S., and Hung, S.-C. (2004). Synthesis of heparin oligosaccharide. J. Am. Chem. Soc. 126, 476–477.


433

Lee, H., Kobayashi, M., Wang, P., Nakayama, J., Seeberger, P. H., and Fukuda, M. (2006). Expression cloning of cholesterol a-glucosyltransferase, a unique enzyme that can be inhibited by natural antibiotic gastric mucin O-glycans, from Helicobacter pylori. Biochem. Biophys. Res. Commun. 349, 1235–1241. Lee, H., Wang, P., Hoshino, H., Ito, Y., Kobayashi, M., Nakayama, J., Seberger, P. H., and Fukuda, M. (2008). a1, 4GlcNAc-capped mucin-type O-glycan inhibits cholesterol aglucosyltransferase from Helicobacter pylori and suppresses H. pylori growth. Glycobiology 18, 549–558. Lemieux, R. U., and Ratcliffe, R. M. (1979). The azidonitration of tri-O-acetyl-D-galactal. Can. J. Chem. 57, 1244–1251. Li, B., Zeng, Y., Hauser, S., Song, H., and Wang, L.-X. (2005). Highly efficient endoglycosidase-catalyzed synthesis of glycopeptides using oligosaccharide oxazolines as donor substrates. J. Am. Chem. Soc. 127, 9692–9693. Lubineau, A., Lortat-Jacob, H., Gavaed, O., Sarrazin, S., and Bonnaffe, D. (2004). Synthesis of tailor-made glycoconjugate mimetics of heparan sulfate that bind IFN-g in the nanomolar range. Chem. Eur. J. 10, 4265–4282. Manabe, S., Ishii, K., and Ito, Y. (2006). N-Benzyl-2, 3-oxazolidinone as a glycosyl donor for Selective a-glycosylation and one-pot oligosaccharide synthesis involving 1, 2-cisglycosylation. J. Am. Chem. Soc. 128, 10666–10667. Manabe, S., Ishii, K., and Ito, Y. (2007). Synthesis of a natural oligosaccharide antibiotic active against Helicobacter pylori. J. Org. Chem. 72, 6107–6115. Manabe, S., Ishii, K., Hashizume, D., Koshino, H., and Ito, Y. (2009). Evidence for endocyclic cleavage of conformationally restricted glycopyranosides. Chem. Eur. J. 15, 6894–6901. Mensah, E. A., and Nguyen, H. M. (2009). Nickel-catalyzed stereoselective formation of a2-deoxy-2-amino glycosides. J. Am. Chem. Soc. 131, 8778–8780. Nokami, T., Shibuya, A., Tsuyama, H., Suga, S., Bowers, A. A., Crich, D., and Yoshida, J.-I. (2007). Electrochemical generation of glycosyl triflate pools. J. Am. Chem. Soc. 129, 10922–10928. Nokami, T., Shibuya, A., Manabe, S., Ito, Y., and Yoshida, J.-I. (2009). a- and b-glycosyl sulfonium ions: Generation and reactivity. Chem. Eur. J. 15, 2252–2255. Noti, C., dePaz, J. L., Polito, L., and Seeberger, P. H. (2006b). Preparation and use of microarrays containing synthetic heparin oligosaccharides for the rapid analysis of heparin-protein interactions. Chem. Eur. J. 12, 8664–8686. Olsson, J. D. M., Eriksson, L., Lahmann, M., and Oscarson, S. (2008). Investigations of glycosylation reactions with 2-N-acetyl-2N, 3O-oxazolidinone-protected glucosamine donors. J. Org. Chem. 73, 7181–7188. Ori, A., Wilkinson, M. C., and Femrnig, D. G. (2008). The heparanome and regulation of cell function: Structures, functions and challenges. Front. Biosci. 13, 4309–4338. Park, J., Kawatkar, S., Kim, J.-H., and Boons, G.-J. (2007). Stereoselective glycosylations of 2azido-2-deoxy-glucosides using intermediate sulfonium ions. Org. Lett. 9, 1959–1962. Paulsen, H., Kalar, C., and Stenzel, W. (1978). Building units for oligosaccharides, XI. Synthesis of a-glycosidically linked di- and oligosaccharides of 2-amino-2-deoxy-Dgalactopyranose. Chem. Ber. 111, 2358–2369. Pedersen, C. M., Figueroa-Perez, I., Lindner, B., Ulmer, A. J., Zahringer, U., and Schmidt, R. R. (2010). Total synthesis of lipoteichoic acid of Streptococcus pneumoniae. Angew. Chem. Int. Ed. 49, 2585–2590. Petitou, M., Imberty, A., Duchaussoy, P., Driguez, P. A., Ceccato, M. L., Gourvenec, F., Sizun, P., Herault, J. P., Perez, S., and Herbert, J. M. (2001). Experimental proof for the structure of a thrombin-inhibiting heparin molecule. Chem. Eur. J. 7, 858–873. Petitou, M., and van Boeckel, C. A. A. (2004). A synthetic antithrombin III binding pentasaccharide is now a drug! What comes next? Angew. Chem. Int. Ed. 32, 3118–3133.

434

Shino Manabe

Post, C. B., and Karplus, M. (1986). Does lysozyme follow the lysozyme pathway? An alternative based on dynamic, structural, and stereoelectronic considerations. J. Am. Chem. Soc. 108, 1317–1319. Pozsgay, V. (2001). Large scale synthesis of 2-azidodeoxy glucosyl donors. J. Carbohydr. Chem. 21, 659–665. Ryan, D. A., and Gin, D. Y. (2008). Ring-Opening of azididine-2-carboxamides with carbohydrate C1-O-nucleophiles. Stereoselective preparation of a- and b-O-glycosyl serine conjugates. J. Am. Chem. Soc. 130, 15228–15229. Satoh, H., Hutter, J., Luthi, H. P., Manabe, S., Ishii, K., and Ito, Y. (2009). Low-barrier pathway for endo-cleavage induced anomerization of pyranosides with N-benzyl-2, 3-trans-oxazolidinone groups. Eur. J. Org. Chem. 2009, 1127–1131. Schlo¨der, N. W., Morath, S., Alexander, C., Hamann, L., Hartung, T., Zahringer, U., Gobel, U. B., Weber, J. R., and Schumann, R. R. (2003). Lipoteichoic acid (LTA) of Streptococcus pneumoniae and Staphylococcus aureus activates immune cells via Toll-like receptor (TLR)-2, lipopolysaccharide-binding protein (LBP), and CD14, whereas TLR-4 and MD-2 are not involved. J. Biol. Chem. 278, 15587–15594. Schwarz, F., Huang, W., Li, C., Schulz, B. L., Lizak, C., Palumbo, A., Numao, S., Neri, D., Aebi, M., and Wang, L.-X. (2010). A combined method for producing homogeneous glycoproteins with eukaryotic N-glycosylation. Nat. Chem. Biol. 5, 264–266. Seeberger, P. H. (2008). Automated oligosaccharide synthesis. Chem. Soc. Rev. 37, 19–28. Slynko, V., Schubert, M., Numao, S., Kowarik, M., Aebi, M., and Allain, F. H. T. (2009). NMR structure determination of a segmentally labeled glycoprotein using in vitro glycosylation. J. Am. Chem. Soc. 13, 1274–1281. Stallforth, P., Lepenies, B., Adibekian, A., and Seeberger, P. H. (2009). Carbohydrates: A frontier in medicinal chemistry. J. Med. Chem. 52, 5561–5577. Szymanski, C. M., Yao, R., Ewing, C. P., Trust, T. J., and Guerry, P. (1999). Evidence for a system of general protein glycosylation in Campylobacter jejuni. Mol. Microbiol. 32, 1022–1030. Szymanski, C. M., Logan, S. M., Linton, D., and Wren, B. W. (2003). Campylobacter—A tale of two protein glycosylation systems. Trends Microbiol. 11, 233–238. Tawaratsumida, K., Furuyashiki, Katsuimoto, M., Fujimoto, Y., Fukase, K., Suda, Y., and Hashimoto, M. (2009). Characterization of N-terminal structure of TLR2-activating lipoprotein in Staphylococcus aureus. J. Biol. Chem. 284, 9147–9152. Titz, A., Radic, Z., Schwardt, O., and Ernst, B. (2006). A safe and convenient method for the preparation of triflyl azide, and its use in diazo transfer reactions to primary amines. Tetrahedron Lett. 47, 2383–2385. Vasella, A., Witzig, C., Chiara, J. L., and Martin-Lomas, M. (1991). Convenient synthesis of 2-azido-2-deoxy-aldoses by diazo transfer. Helv. Chem. Acta 74, 2073–2077. Wacker, M., Feldman, M. F., Callewaert, N., Kowarik, M., Clarke, B. R., Phol, N. L., Hernandez, M., Vines, E. D., Valvano, M. A., Whitfield, C., and Aebi, M. (2006). Substrate specificity of bacterial oligosaccharyltransferase suggests a common transfer mechanism for the bacterial and eukaryotic systems. Proc. Natl. Acad. Sci. USA 103, 7088–7093. Wang, P., Lee, H., Fukuda, M., and Seeberger, P. H. (2007). One-pot synthesis of a pentasaccharide with antibiotic activity against Helicobacter pylori. Chem. Commun. 19, 1963–1965. Warren, J. R., and Marshall, B. J. (1983). Unidentified curved bacilli on gastric epithelium in active chronic gastritis. Lancet 321, 1273–1275. Weerapana, E., and Imperiali, B. (2006). Aspargin-linked protein glycosylation: From eukaryotic to prokaryotic systems. Glycobiology 16, 91R–101R. Weerapana, E., Glover, K. J., Chen, M. M., and Imperiali, B. (2005). Investigating bacterial N-linked glycosylation: Synthesis and glycosyl acceptor activity of the undecaprenyl pyrophosphate-linked bacillosamine. J. Am. Chem. Soc. 127, 13766–13767.


435

Wei, P., and Kerns, R. J. (2005a). Factors affecting stereocontrol during glycosidation of 2, 3-oxazolidinone-protected 1-tolylthio-N-acetyl-D-glucosamine. J. Org. Chem. 70, 4195–4198. Wei, P., and Kerns, R. J. (2005b). Chemoselective deprotection and functional group interconversion of ring-fused 2N, 3O-oxazolidinones of N-acetyl-D-glucosamine. Tetrahedron Lett. 46, 6901–6905. Wei, G., Lv, X., and Du, Y. (2008). FeCl3-catalyzed a-glycosidation of glucosamine pentaacetates. Carbohydr. Res. 343, 3096–3099. Wildt, S., and Gerngross, T. U. (2005). The humanization of N-glycosylation pathways in yeast. Nat. Rev. Microbiol. 3, 119–128. Winterfeld, G. A., and Schmidt, R. R. (2001). Nitroglycal concatenation: A broadly applicable and efficient approach to the synthesis of complex O-glycans. Angew. Chem. Int. Ed. 40, 2654–2657. Yiqun, G., Zhang, L.-H., and Ye, X. S. (2008a). Pre-activation protocol leading to highly seteroselectivity-conrallable glycosylations of oxazolidinone protected glucosamines. Chem. Commun. 597–599. Yiqun, G., Zhang, L.-H., and Ye, X.-S. (2008b). Stereoselectivity investigation on glycosylation of oxazolidinone protected 2-amino-2-deoxy-D-glucose donors based on preactivation protocol. Tetrahedron 64, 4949–4958. Young, N. M., Brisson, J.-R., Kelly, J., Watson, D. C., Tessier, L., Lanthier, P. H., Jarrell, H. C., Cadotte, N., St. Michael, F., Aberg, E., and Szmanski, C. M. (2002). Structure of the N-linked glycan present on multiple glycoproteins in the gram-negative bacterium, Campylobacter jejuni. J. Biol. Chem. 277, 42530–42539. Zhou, Y., Lin, F., Chen, J. F., and Yu, B. (2006). Toward synthesis of the regular sequence of heparin: Synthesis of two tetrasaccharide precursors. Carbohydr. Res. 341, 1619–1629. Zhu, X., and Schmidt, R. R. (2009). New principles for glycoside-bond formation. Angew. Chem. Int. Ed. 48, 1900–1934.

C H A P T E R

T W E N T Y- O N E

Aminoglycosides: Redesign Strategies for Improved Antibiotics and Compounds for Treatment of Human Genetic Diseases Varvara Pokrovskaya, Igor Nudelman, Jeyakumar Kandasamy, and Timor Baasov Contents 1. Introduction 2. Aminoglycoside Antibiotics: Their Mode of Action and Major Drawbacks 3. Strategies Toward Development of Improved Antibiotics 3.1. Alteration of neomycin B at C500 position to prevent APH(30 ) resistance 3.2. Dual activity of C500 -modified neomycin B derivatives against Bacillus anthracis 3.3. 30 ,40 -Methylidene protected aminoglycosides: A strategy to reduce toxicity and overcome resistance enzymes 3.4. Neomycin B-based hybrid antibiotics: A strategy to delay resistance development 4. Aminoglycosides as Readthrough Inducers for the Treatment of Genetic Diseases 4.1. Development of new variants of aminoglycosides with improved readthrough activity and reduced toxicity 5. Concluding Remarks and Future Perspectives Acknowledgments References

438 440 443 444 447 449 450 454 455 458 458 459

Abstract Aminoglycosides are highly potent, broad-spectrum antibiotics that kill bacteria by binding to the ribosomal decoding site and reducing the fidelity of protein synthesis. The emergence of bacterial strains resistant to these drugs, as well The Edith and Joseph Fischer Enzyme Inhibitors Laboratory, Schulich Faculty of Chemistry, Technion—Israel Institute of Technology, Haifa, Israel Methods in Enzymology, Volume 478 ISSN 0076-6879, DOI: 10.1016/S0076-6879(10)78021-6

#


437

438

Varvara Pokrovskaya et al.

as their relative toxicity, have inspired extensive searches toward the goal of obtaining novel molecular designs with improved antibacterial activity and reduced toxicity. In recent years, a new therapeutic approach that employs the ability of certain aminoglycosides to induce mammalian ribosomes to readthrough premature stop codon mutations has emerged. This new and challenging task has introduced fresh research avenues in the field of aminoglycosides research. In this chapter, our recent observations and current challenges in the design of aminoglycosides with improved antibacterial activity and the treatment of human genetic diseases are discussed.

1. Introduction The discovery of streptomycin by Selman Waksman in 1944 was a landmark not just in antibiotic history but also in a revolutionary recognition of complex carbohydrates as an important class of natural products (Schatz et al., 1944). Streptomycin was the first aminoglycoside to be isolated from a bacterial source and the first effective antibiotic against Mycobacterium tuberculosis. In the following decades, several milestone aminoglycoside drugs (Fig. 21.1), such as neomycin, kanamycin, tobramycin, and others, were isolated from soil bacteria by intense search for natural products with antibacterial activity (Umezawa and Hooper, 1982). However, the rapid spread of antibiotic resistance to this family of antibacterial agents in pathogenic bacteria (Vakulenko and Mobashery, 2003), along with their relative toxicity to mammals, have stimulated medicinal chemistry approaches toward the development of improved antibiotics. Earlier studies on direct chemical modification of existing aminoglycoside drugs, with the aim of circumventing the resistance mechanisms, have yielded several semisynthetic drugs such as amikacin, dibekacin, netilmicin, and isepamicin that were introduced into clinical use in the 1970s and 1980s (Kondo and Hotta, 1999). More recent advancements in studies of resistance mechanisms (Wright, 2008), high-resolution structures of aminoglycosides in complex with their ribosomal RNA (rRNA) target (Francois et al., 2005; Ogle and Ramakrishnan, 2005), along with the identification and characterization of biosynthetic enzymes for certain aminoglycosides (Llewellyn and Spencer, 2006), have stimulated the development of innovative chemical (Li and Chang, 2006; Silva and Carvalho, 2007) and chemoenzymatic strategies (Llewellyn and Spencer, 2008; Nudelman et al., 2008) toward improved aminoglycoside derivatives and mimetics (Chittapragada et al., 2009; Hermann, 2007). Although the prokaryotic selectivity of action is critical to the therapeutic utility of aminoglycosides as antibiotics, they are not entirely selective to bacterial ribosome; they also bind to the eukaryotic rRNA (Bottger et al., 2001)

4, 5-Disubstituted

4, 6-Disubstituted

6' R1 I R1 6' R2 O R3 HO O 1' HO R4 1' II NH 4 2 H2N H2N O 4 NH2 O 1 NHR2 NH2 O HO HO 5 OH 6O 5'' O Neamine O Me paromamine III OH NH2 HO NH OH O OH O Me HN Ribostamycin OH IV butirosin B R1 Neamine Ribostamycin Neomycin B Paromomycin Paromamine Butirosin B

NH2 NH2 NH2 OH OH NH2

R1

R2 H H H H H AHB

Gentamicin C1 Gentamicin C1a Gentamicin C2 Gentamicin C2a Gentamicin C2b Geneticin (G418)

NHCH3 H NH2 CH3 NHCH3 CH3

R2 CH3 NH2 CH3 NH2 H OH

R1 R2

6' NH2 O 1' R3 O 4 HO

NH2 1

6O

NHR4 OH OH

O HO H2N

Kanamycin A Kanamycin B Amikacin Tobramycin Debekacin Arbekacin

R 1 R2

R3

R4

OH OH OH OH OH OH OH H H H H H OH

OH NH2 OH NH2 NH2 N2H

H H AHB H H AHB

AHB = NH2

Neomycin family

Gentamicin family

O Kanamycin family

Figure 21.1 Chemical structures of 4,5- and 4,6-disubtituted 2-DOS containing natural and synthetic aminoglycosides.

440


and promote mistranslation (Eustice and Wilhelm, 1984). The use of this disadvantage of aminoglycoside antibiotics for the possible treatment of human genetic diseases caused by premature nonsense mutations is extremely challenging (Zingman et al., 2007). There are more than 1800 inherited human diseases caused by nonsense mutations, that is, alterations in the genetic code that prematurely stop the translation of proteins. Aminoglycosides have emerged as vanguard pharmacogenetic agents in treating such genetic disorders due to their unique ability to induce mammalian ribosomes to readthrough premature stop codon mutations. In numerous preclinical and pilot clinical studies, this new therapeutic approach shows promise in phenotype correction by promoting otherwise defective protein synthesis (Kellermayer et al., 2006). However, severe side effects of existing aminoglycoside drugs, including high toxicity to mammals and the reduced readthrough efficiency at subtoxic doses, have inspired extensive searches toward the goal of obtaining novel molecular designs with improved readthrough activity and reduced toxicity (Hainrichson et al., 2008; Hermann, 2007). During the past few years, several comprehensive review articles that cover traditional areas of aminoglycoside use as antibiotics, including development of novel semisynthetic aminoglycoside derivatives (Chittapragada et al., 2009; Zhou et al., 2007), molecular mechanism of action, mechanisms of resistance (Wright, 2008), and toxicity (Guthrie, 2008), have been published. In addition, the book titled ‘‘Aminoglycoside Antibiotics: From Chemical Biology to Drug Discovery’’ was published in 2007 by John Wiley & Sons, Inc., providing an excellent overview of recent advances in the field (Arya, 2007). Therefore, in this chapter we focus mostly on our recent efforts to develop new designs with improved activity against resistant and pathogenic bacteria, new designs that are active against drug-resistant bacteria and exhibit reduced potential for generating new bacterial resistance, and new designs with improved premature stop codon suppression activity and reduced toxicity.

2. Aminoglycoside Antibiotics: Their Mode of Action and Major Drawbacks The majority of aminoglycosides consist of a central aminocyclitol ring, usually 2-deoxystreptamine (2-DOS) linked to one or more amino sugars by glycosidic bonds. Depending on the substitution pattern on the 2-DOS ring, aminoglycosides can be divided into two major classes: 4,5and 4,6-disubstituted 2-DOS aminoglycosides (Fig. 21.1). The nomenclature usually refers to ring I as primed and corresponds to the amino sugar

Designer Aminoglycosides

441

linked at position 4. Ring II is unprimed and corresponds to the central 2-DOS ring. Ring III is referred to as the doubly primed and corresponds to the substituent in position 5 or 6 of the 2-DOS. Rings with sequential numbers (IV, V) are usually attached to ring III (Fig. 21.1). Neomycin B, a representative of 4,5-disubstituted 2-DOS subclass, is used topically in the form of creams and lotions for the treatment of bacterial infections caused from skin burns, wounds, and dermatitis ( Jana and Deb, 2006). Paromomycin, another representative of 4,5-disubstituted 2-DOS, is used therapeutically against intestinal parasites ( Jana and Deb, 2006) and in the treatment of a variety of tropical diseases, including leishmaniasis and certain types of fungal infection (Sundar and Chakravarty, 2008). On the other hand, the 4,6-disubstituted 2-DOS subclass antibiotics, including gentamicin, amikacin, and tobramycin, have important clinical applications in the treatment of serious Gram-negative bacterial infections, especially in cases of opportunistic bacteria accompanying cystic fibrosis (CF), AIDS, and cancer. Both the 4,5- and 4,6-disubstituted subclasses of aminoglycosides exert their antibacterial activity by targeting the phylogenetically conserved decoding site (A-site) of bacterial 16S rRNA in the 30S ribosomal subunit (Moazed and Noller, 1987) and interfering with decoding and global translation processes. During the recent years, several studies on NMR and crystal structures of aminoglycosides bound to bacterial A-site oligonucleotide models (Fourmy et al., 1998; Francois et al., 2005), along with crystal structures of the bacterial 30S and 70S ribosomal particles (Carter et al., 2000; Selmer et al., 2006) with and without the bound aminoglycoside, have provided fascinating insights into our understanding of the decoding mechanism in prokaryote cells and of how 2-DOS aminoglycosides induce the deleterious misreading of the genetic code. These structures revealed that upon binding to the 30S subunit, aminoglycosides displace the two noncomplementary adenines, A1492 and A1493, at the A-site of the ribosome and lock them into so-called ‘‘on’’ state orientation, similar to that observed during mRNA decoding. As a result, during the codon–anticodon interaction and the proofreading process not only the cognate tRNA is stabilized, but near-cognate tRNA is stabilized as well which causes the misreading process, the accumulation of truncated and nonfunctional proteins and eventually leading to bacterial cell death. While this mechanism of action is now well accepted for the majority of 2-DOS aminoglycosides, the recent crystallographic investigation of a series of aminoglycosides bound to the A-site oligonucleotide models (Francois et al., 2005; Vicens and Westhof, 2003) suggest that the actual molecular mechanism of this ‘‘molecular switch’’ system is more complex and that additional thermodynamic and kinetic factors are likely to govern the impact of aminoglycosides on prokaryotic translation (Ogle and Ramakrishnan, 2005). Indeed, recent

442


study to characterize the energetics and dynamics associated with the aminoglycoside–rRNA interaction demonstrated that the aminoglycosideinduced reduction in the mobility of the A1492 residue is an important determinant of antibacterial activity (Kaul et al., 2006). Resistance to the aminoglycosides occurs by three methods: (1) decrease of intracellular drug concentration (import and efflux), (2) modification of the target rRNA and ribosomal proteins, and (3) enzymatic drug modification (Wright, 2008). The latter is the most prevalent mechanism in clinical isolates of resistant bacteria. Three distinct classes of aminoglycoside-modifying enzymes are known: the aminoglycoside phosphotransferases (APHs), the aminoglycoside-acetyltransferases (AACs), and the aminoglycoside-adenyltransferases (ANTs; Fig. 21.2). These modifications reduce the binding affinity of aminoglycosides to the A-site and thus significantly decrease their antibacterial potential. Members of each of these classes of enzymes are widespread in both Gram-negative and Gram-positive bacteria, and 3D crystal structures of representative proteins from each class have recently been solved (Magnet and Blanchard, 2005; Wright, 2008). Among these three enzyme classes, aminoglycoside 30 -phosphotransferases [APH (30 )s], of which seven isozymes are known, are most widely represented. These enzymes catalyze transfer of g-phosphoryl group of ATP to the 30 -hydroxyl of many aminoglycosides, rendering them inactive. Although the enzymes of all three classes are typically monofunctional enzymes, the recent emergence of genes encoding bifunctional aminoglycoside-modifying enzymes, for example, the bifunctional AAC(60 )/APH(200 ) enzyme, is another level of sophistication relevant to the clinical use of aminoglycosides (Zhang et al., 2009). AAC(6′)–APH(2′′)

ANT(4′) 4′

APH(3′)

H2N AAC(6′)–APH(2′′)

AAC(3)

6′ NH2

O HO HO 3′ H2N O O HO 5′′ O OH NH2 3′′′ O O OH

AAC(1) 3 NH2 1 NH 2 OH

OH

Neomycin B

Figure 21.2 Aminoglycoside-modifying enzymes and their modification sites are highlighted on the structure of neomycin B.

443


Another main drawback of aminoglycoside antibiotics is their relatively high toxicity to humans. The aminoglycosides-induced toxic effects include nephrotoxicity, ototoxicity (vestibular and auditory), and, rarely, neuromuscular blockade and hypersensitivity reactions (Nagai and Takano, 2004; Talaska and Schacht, 2007). The first two types of toxicities were suggested to result from a combination of several mechanisms, such as inhibition of phospholipases, interaction with phospholipids, as well as formation of free radicals (Forge and Schacht, 2000). Nephrotoxicity receives the most attention, perhaps because of easier documentation of reduced renal function, but it is usually reversible. Ototoxicity is usually irreversible and it is believed to result from aminoglycoside binding to phospholipids and from disrupting mitochondrial protein synthesis due to accumulation of drug in the inner ear (Hobbie et al., 2008a,b). Numerous studies suggested that the interference between aminoglycosides and some steps of calciummediated acetylcholine release at the level of presynaptic structures is the main cause of the neuromuscular blockage induced by aminoglycosides (Albiero et al., 1978).

3. Strategies Toward Development of Improved Antibiotics The problems of bacterial resistance and inherent toxicity of aminoglycosides have inspired continuous attempts for the development of improved aminoglycoside variants by using series of diverse approaches (Fig. 21.3). In general, these approaches can be divided into two different categories: (1) the ‘‘modifications of existing drugs’’ and (2) the design of ‘‘aminoglycosides mimetics.’’ The first category includes either the direct

Design strategies for aminoglycoside antibiotics improvement

Modifications of the existing drugs

Direct modifications

Glycosylation strategies

Aminoglycoside mimetics

Total-synthetic mimetics

Semisynthetic mimetics

Figure 21.3 Design strategies employed for development of improved aminoglycoside antibiotics.

444


modification of the intact aminoglycoside by attachment of various appendages at different locations, or glycosylation of the selected aminoglycoside scaffolds, usually the pseudodisaccharides-like neamine and paromamine, with various natural or unnatural sugars, or a combination of both. The second category considers the generation of aminoglycoside mimetics either by further minimization of the original aminoglycoside entity to one ring, usually rings I or II, to which different appendages are attached at different locations, or rational design of the completely new entities. Using these strategies, many semisynthetic analogs of natural aminoglycosides (Chittapragada et al., 2009; Li and Chang, 2006; Wang and Chang, 2007; Zhou et al., 2007) and aminoglycoside mimetics (Chittapragada et al., 2009; Hermann, 2007) have been reported during recent years. Some of these designs were found to be effective against aminoglycoside-resistant bacterial strains. Little progress, however, has been made toward the discovery of new aminoglycoside derivatives with diminished toxicity (Shitara et al., 1995), which indeed is one of the remaining and perhaps the most challenging task. In addition, the latest semisynthetic aminoglycoside introduced into human therapy was two decades ago (arbekacin, Fig. 21.1, a kanamycin B derivative used in Japan since 1990; Kondo and Hotta, 1999), while the resistance to all the currently available aminoglycosides is increasing in prevalence. Clearly, a novel aminoglycoside derivative(s) with reduced toxicity demonstrated efficiency against the current generation of resistant pathogens, and preferably with reduced potential for generating bacterial resistance should be an important addition for the treatment of infectious diseases. To address the need of such designs, the following sections summarize our recent efforts to develop new aminoglycoside antibiotics by using four different strategies (Fig. 21.4): (1) the designs which in addition to targeting rRNA also resist to aminoglycoside-modifying enzyme(s); (2) the designs that target both the toxigenic bacterium and its lethal toxin; (3) the designs which in addition to resisting to aminoglycoside-modifying enzyme(s) and targeting rRNA, also exhibit reduced toxicity; (4) the designs of hybrid antibiotics which in addition to resisting to existing aminoglycoside-modifying enzymes and targeting rRNA, also delay the development of new resistance.

3.1. Alteration of neomycin B at C500 position to prevent APH(30 ) resistance In the presence of APH(30 ) enzymes the majority of aminoglycoside drugs undergo phosphorylation at C30 -OH, but neomycin B (NeoB) undergoes phosphorylation at two distinct positions: C30 -OH and C500 -OH (Fig. 21.2). We hypothesized that by attaching an extra rigid sugar ring at C500 -OH of NeoB, in addition to blocking this position from initial phosphorylation may also inhibit the formation of a precise ternary complex

4⬘

NH2

NH2 O

O O O 3⬘H2N O HO HO

HN

HO HO H2N O O HO 5⬘⬘ O

Toxicity reduction

NH2 NH2 O

Overcoming APH(3⬘)

O OH

H2N

3⬘, 4⬘-Methylidene protected pseudotrisaccharide NH2

HO

OH NH2 O O OH

Overcoming APH(3⬘) resistance

NH2 NH2

HO

NH2 O NH2 O

O

OH

O

Neomycin B Delay in resistance development

H2N

NH2

OH

5′′ O HO O n(NH2) H 2N O OH HO

OH

B. anthracis inhibition

HO HO O H2N O

OH

Pseudopentasaccharides

OH O HO HO

NH2 NH2 O HO O HO H2N O O 5⬘⬘S HO S O O 5⬘⬘ O H 2N O NH2 H 2N OH OH O O OH H2N H2N OH O HO NeoB dimer H2N

F

NH2 NH2 OH

HO HO

O N

N X

HOOC N H2N

N N N

NH2 O H2N

NH2 OO

Y 5⬘⬘ O

OH NH2 O O OH

NH2

OH

OH

NeoB–Ciprofloxacin hybrids

Figure 21.4 Redesign of neomycin B (NeoB) for improved antibacterial performance by using four different strategies.

446


1, R = O

HO

11, R =

O

O

O OH

H2N

OH

O

O

HO HO

R 5''

HO

O

OH O

O

OH

OH

4, R = HO HO

NH2 O

H2N

NH2

O

OH

O

O NH2

O

MIC = 40 (kcat / Km) = 12

OH

OH

5, R =

R = OH; neomycin B

MIC = 40 (kcat / Km) = 3.0

O

OH

MIC = 128 (kcat / Km) = 9.7

NH2 O

O

HO H2N

NH2

OH

H2N

OH

3, R =

MIC = 512 (kcat / Km) = 13 9, R =

O

MIC = 40 (kcat / Km)=5.1

MIC = >512 (kcat / Km) = 8.3

H2N HO

O OH

S

O

OH

H2N HO

OH

MIC = 64 (kcat / Km) = 6.7 10, R =

2, R = O

NH2

OH H2N

OH HO HO

MIC = 64 (kcat / Km) = 28

NH2

H2N HO

O

O

OH 8, R =

HO

O

O

7, R = H2N

HO

OH

MIC = 64 (kcat / Km) = 5.4

6, R =

OH O NH2 OH

MIC = 128 (kcat / Km) = 12

O

NH2

HO HO

O

O

MIC = 32 (kcat / Km) = 2.5

NH2 MIC = 32 (kcat / Km) = 6.7

Figure 21.5 Comparative antibacterial activity against P. aeruinosa and specificity constants (kcat/Km values in 104 M 1s 1) with APH(30 )-IIIa enzyme of NeoB and its 500 -modified derivatives 1–11. Minimal inhibitory concentration (MIC) values are given in mg/mL.

required for the phosphorylation of the C30 -OH by APH(30 )s, while the affinity of the resulting derivative to the target rRNA would not change or even increase. Using this strategy, a series of branched derivatives of NeoB, compounds 1–11 (Fig. 21.5), were synthesized. All these compounds were assembled by employing the general synthetic approach shown in Scheme 21.1. Initially, the NeoB was converted into the common acceptors, to which various donors were attached, followed by a two-step deprotection to yield the target C500 -branched derivatives. All new structures keep the whole antibiotic constitution intact as a recognition element to the rRNA, while the extended sugar rings (ring V in structure 1–10, and rings V and VI in 11) of each structure was designed in a manner that incorporates the potential functionalities directed for the recognition of the phosphodiester bond of RNA. The designed structures (compounds 1–11) exhibited similar or better antibacterial activities to that of the parent NeoB against selected bacterial

447


NH2

N3 O

AcO AcO

N3 O HX O 5⬘⬘ O

Neomycin B

N3

n(NH2)

N3 OAc

O

Donor Coupling step

O Deprotection steps

OH

N3

O

OAc H2N

OAc X = O, S

Common NeoB acceptor

H2N X 5⬘⬘

n(NH2)

OAc N3 O

O

HO HO

LG

X = O, S

NH2 O

NH2

O

OH

O

NH2 O

O

OH

OH Pseudopentasaccharides

Scheme 21.1 General synthetic strategy for the assembly of C500 -derivatives of NeoB (LG ¼ leaving group).

strains, including pathogenic and resistant strains, and especially good activities were observed against Pseudomonas aeruginosa (Fridman et al., 2003; Hainrichson et al., 2005; Fig. 21.5). The specificity constant values (kcat/ Km) of 1–11 with APH(30 )-IIIa enzyme were lower than that of NeoB, implying that these derivatives are poorer substrates of the enzyme than the parent NeoB. Since the strains of P. aeruginosa harbor a chromosomal APH (30 )-IIb-encoding gene (Hainrichson et al., 2007), the observed superior activity of new derivatives to that of NeoB in this bacterium could be ascribed because their inferior substrate activity for the APH(30 )-IIb resistance enzyme.

3.2. Dual activity of C500 -modified neomycin B derivatives against Bacillus anthracis Anthrax is an infectious disease caused by toxigenic strains of the Grampositive bacterium B. anthracis. It has been well established that the anthrax toxins (protective antigen, PA, edema factor, EF, and lethal factor, LF) play a major role from the very beginning of infection to death of the host. Among them, LF is considered the dominant virulence factor of anthrax and therefore an intensive search for specific inhibitors of LF has been performed during the last decade (Forino et al., 2005; Shoop et al., 2005). Unlike this strategy, we anticipated that it would be highly beneficial if the developed material were bifunctional, with the ability to inactivate the released LF toxin and, in parallel, to function as an antibiotic. Indeed, in the earlier experiments, Wong and coworkers (Numa et al., 2005) tested a library of approximately 3000 compounds, over 60 of which were synthetic and commercial aminoglycosides, and have demonstrated that NeoB is the most potent inhibitor of LF with the apparent Ki value in the low nanomolar concentration range. To improve the inhibitory effect of NeoB derivatives and in parallel to address the need of such dual activity designs, in collaboration with Wong’s

448


laboratory, initially we tested the C500 -modified NeoB derivatives 1, 2, 4–11 (Fig. 21.5) as inhibitors of LF along with against B. anthracis (Sterne strain) (Fridman et al., 2005). Most of the compounds exhibited significant antibacterial activity against B. anthracis and displayed activity levels comparable to that of NeoB. At low ionic strength assay conditions, the compounds 5, 6, and 9 exhibited the Ki values in the range of 0.2–1.3 nM, indicating them as predominantly better inhibitors of LF than the NeoB (Ki ¼ 37 nM). However, an increase in the ionic strength from 0 to 150 mM KCl (best resembles the physiological ionic strength in many cell types) drastically shifted the measured Ki values of all the tested compounds by a factor of 1500–53,000 towards higher concentrations, indicating that the predominant interaction between LF protein and the aminoglycosides is electrostatic in origin. Since at these conditions several C500 -derivatives were only two- to fivefold better inhibitors than NeoB, we attempted to further increase this gap and developed the disulfide dimer than NeoB, compound 12 (Fig. 21.6). At both low and high salt concentrations (150 mM KCl), compound 12 showed 53-fold higher affinity to LF relative to that of NeoB, indicating that twice the number of charged groups in 12 is probably responsible for the increased affinity. Compound 12 also displayed significant antibacterial activity against B. anthracis. Thus, the design strategy employed in this particular study provided a new direction for the development of novel antibiotics that target both the toxigenic bacterium and its released lethal toxin. NH2 OH O HO

NH2

HO NH2 O

HO HO

H2N HO

O

O

O

H2N

NH2 NH2 OH

OH NH2 O OH O OH NeoB Ki = 37 nM Ki (150 mM KCl) = 59 mM

HO H2N H2N

O

O

HO HO H2N

O

5''

O

S NH2

O H2N

S

NH2 O O O O

5''

OH OH O H2N HO

12

O

NH2 NH2 OH

OH OH

H2N

Ki = 0.7 nM Ki (150 mM KCl) = 1.1 mM

Figure 21.6 Comparative apparent inhibition constants (Ki values) of NeoB and its disulfide dimer 12 against anthrax lethal factor (LF) toxin activity at low and high salt conditions.

449


3.3. 30 ,40 -Methylidene protected aminoglycosides: A strategy to reduce toxicity and overcome resistance enzymes It is highly noteworthy that, although a new synthetic variant may exhibit favorable antibacterial activity, high toxicity can prevent its clinical application. Therefore, the design of novel variants of aminoglycosides, which in addition to resisting to aminoglycoside-modifying enzymes and targeting rRNA, will also exhibit reduced toxicity is of high urgency. To address the need of such designs, a new pseudodisaccharide 13 with a 30 ,40 -protection was designed and its properties were evaluated in comparison to other two structurally related pseudodisaccharides, compounds 14 and 15 (Fig. 21.7; Chen et al., 2008). We anticipated that (1) the preservation of 30 ,40 -oxygens in 13 should keep the lower basicity of the 20 -NH2 group and subsequently the lower toxicity than those of the 30 ,40 -dideoxy analog 15; (2) the methylidene group should also protect 13 from the action of various APH(30 ) and ANT(40 ) resistance enzymes; (3) the 30 ,40 -methylidene protection is supposed to be substantially stable under both acid and base conditions usually used in carbohydrate chemistry, and should easily be constructed from the corresponding 30 ,40 -diol. The observed data in this study demonstrated a relationship between the basicity of the 20 -amine group and the estimated LD50 values in mice: the increase in the basicity of the 20 -amino functionality is associated with the acute toxicity increase of an aminoglycoside (Fig. 21.7), with 13 being least toxic. Similar results have been obtained by replacement of the 5-OH with 5-fluorine in kanamycin B and its several clinical derivatives (Shitara et al., 1995). The toxicities of the resulting fluoro analogs were significantly lower than the parent compounds and this phenomenon again was attributed to basicity reduction of the 3-NH2 group induced by the strongly electronwithdrawing 5-fluorine. Thus, significantly high acute toxicity of the clinical drugs such as tobramycin (30 -deoxy), gentamicin (30 ,40 -dideoxy), dibekacin (30 ,40 -dideoxy), and arbekacin (30 ,40 -dideoxy) could be ascribed to the increased basicity of 20 -NH2 group (ring I) in these drugs caused

O O

NH2 O NH2

O HO

NH2

NH2 OH

13 LD50 = 303 2'-N pKa = 6.0

HO HO

Methods in Enzymology 478 - Glycomics

Methods in Enzymology 479 - Functional Glycomics

Methods in Enzymology Vol 416: Glycomics

Methods in Enzymology Vol 417: Functional Glycomics

Methods in Enzymology Vol 416: Glycomics

Glycomics. Methods and Protocols

Methods in Enzymology. Liposomes

Glycomics: Methods and Protocols

Glycomics: Methods and Protocols

Methods in Enzymology. Liposomes

Methods in Enzymology (Liposomes )